PAPERS VIEWPOINTS NEWS EDITORIAL appropriate casemanagement ofchildhoodfeverinMozambique Contribution ofcommunity health workerstoimprovingaccesstimelyand Jeanne Koepsell,IbadulHaque Khan,AgbessiAmouzou Tanya Guenther,SalimSadruddin,KarenFinnegan,EricaWetzler , FatimaIbo,PauloRapaz, year estimates Economics in“GlobalHealth2035”:asensitivityanalysis ofthevaluealife Angela YChang,LisaARobinson,JamesKHammitt,Stephen C Resch Reducing maternalmortalityinsub–SaharanAfrica:therole ofethicalconsumerism Dileep Wijeratne,AndrewDavidWeeks Empowering peopleforsustainabledevelopment:theOttawa Charterandbeyond Sagar Dugani,ZulfiqarA.Bhutta,NiranjanKissoon development inThailand Surveillance ofantimicrobialconsumption:methodologicalreviewforsystems on behalfoftheThaiSACWorkingGroup Nithima Sumpradit,RungpetchSakulbumrungsil,SasiJaroenpoj,VaravootSermsinsiri; Viroj Tangcharoensathien,AngkanaSommanustweechai,BoonratChanthong, Men’s health:timeforanewapproachtopolicyandpractice? Peter Baker,TimShand Zika: informationinthenickoftime Jessica LWalker,JamesHConway,ESvenson ultrasound restriction Stopping femalefeticideinIndia:thefailureandunintendedconsequenceof Sheida Tabaie Lessons fromBrazil:onthedifficultiesofbuildingauniversalhealthcaresystem Valbona Muzaka performance? Does SDG3haveanadequatetheoryofchangeforimprovinghealthsystems Gabriel Seidman four parameterstoachieveindividualandcollectiveaccountability A newparadigmonhealthcareaccountabilitytoimprovethequalityofsystem: Michelangelo Casali Umberto Genovese,SaraDelSordo,GabriellaPravettoni,IgorMAkulin,RiccardoZoja, communicable diseaseinterventionsinKenya Research forActionablePolicies:implementationscienceprioritiestoscaleupnon– von Rège,DavidWata,PamWilliams,GeraldYonga;ParticipantsfromtheNCDSymposiuminKenya Elijah Ogola,CarolOlale,DeborahOlwal–Modi,RoseRao,SarasRosin,OnyangoSangoro,Daniel Walter Mwanda,DanielMwai,JuliusMwangi,EstherMunyoro,ZacharyMuriuki,JamesNjoroge, James Kayima,AlfredKaragu,DorcasKiptui,AnneKorir,NkathaMeme,BredaMunoz, Robai Gakunga,GladwellGathecha,RainerHilscher,MuhammadJamiHusain,LydiaKaduka, Amuyunzu–Nyamongo, GiselaAbbam,NaftaliBusakhala,AbigailChakava,JonathanDick, Sujha Subramanian,JosephKibachio,SonjaHoover,PatrickEdwards,EvansAmukoye,Mary Resources Agencies Regions sustainable developmentgoalsforhealth Community–based primaryhealthcare:acorestrategyforachieving Zulfiqar ABhutta Journal ofGlobalHealth:TheMissionStatement (continued ontheinside)
88 80 76 72 66 61 58 55 50 43 39 30 21 12 4 2 1
journal of global health June, 2017 Vol. 7 No. 1 PART TWO ISSN 2047-2978 Vol. 7No.1/2 June 2017 global journal of health journal of global health www.jogh.org
Journal of Global Health (JoGH) is a peer-review general medical journal focusing on issues relevant with the Global Health Academy of the University of Edinburgh, while also maintaining collaboration with for global health. The mission of the journal is to serve the community of researchers, funding agencies, academic partners throughout the world. international organizations, policy-makers and other stakeholders by providing an independent critical The views expressed by the authors do not reflect the views of JoGH or its publisher. assessment of the key issues that dominated the global health community. Journal design: Snježana Engelman Džafić for LASERplus, Zagreb, Croatia JoGH publishes original articles, viewpoints, research protocols and other publication items, making Realisation: LASERplus, Zagreb, Croatia, www.laser-plus.hr them available in full text online as soon as they are ready for publication. The articles are collected in Creative Commons Licence: CC BY 4.0. two annual print issues, at the end of June and December. All submissions and editorial inquiries should Print ISSN: 2047-2978 be directed to editor[at]jogh.org. Online ISSN: 2047-2986 Publisher: Edinburgh University Global Health Society (EUGHS). The aim of the EUGHS is to inspire and JoGH editors and editorial board members are independent researchers based at academic institutions, educate individuals, groups, organizations and communities about global health issues. The society is thus being well placed to provide professional evaluation of the key topics and ongoing activities and affiliated to the Centre for Population Health Sciences of the University of Edinburgh, and it works closely programs.
Neonatal Health; Deputy Director of the Intensive Care Unit and of Epidemiology; Croatian Centre for Global Health; Department EDITORS-IN-CHIEF Clinical Director of the General Intensive Care Unit; The University of Public Health; The University of Split School of Medicine, Split, of Melbourne and Royal Children’s Hospital, Melbourne, Australia Croatia • Assistant Professor Tomislav Lauc – Assistant Professor Ana Marušić – Chair, Department of Research in • Professor Wei Wang – Professor of Postgraduate Medicine; Professor of Orthodontics and Dental Anthropology; Editor-in-Chief, Biomedicine and Health, The University of Split School of Medicine, School of Medical Sciences, Edith Cowan University, Perth, Australia South European Journal of Orthodontics and Dentofacial Research; Split, Croatia Department of Anthropology, School of Social Sciences and Professor Igor Rudan – Chair in International Health and Molecular EDITORIAL COUNCIL Humanities; University of Zagreb, Zagreb, Croatia • Professor Joy Medicine; Co-Director, WHO Collaborating Centre for Population E. Lawn – Director, Global Evidence and Policy, Saving Newborn Health Research and Training; Co-Director, Centre for Global Health Professor James Smith – Vice-Principal International; Professor Lives programme, Save the Children; Director, MARCH Centre; The Research, College of Medicine and Veterinary Medicine, The of African and Development Studies; School of Social and Political London School of Hygiene and Tropical Medicine, London, UK • University of Edinburgh, Edinburgh, Scotland, UK Science; The University of Edinburgh, Edinburgh, Scotland, UK • Clinical Assistant Professor James A. Litch – Director, Perinatal Professor Harry Campbell – Professor of Genetic Epidemiology Jake Broadhurst – Head, Global Academies Directorate; The Interventions Program; Global Alliance to Prevent Prematurity and and Public Health; Co-Director, WHO Collaborating Centre for University of Edinburgh, Edinburgh, Scotland, UK • Professor Stillbirth (GAPPS), Seattle Children's; Department of Global Health Population Health Research and Training; Co-Director, Centre Mickey Chopra – Honorary Professor of Global Health; Centre and Department of Epidemiology; University of Washington, Seattle, for Global Health Research, College of Medicine and Veterinary for Global Health Research, College of Medicine and Veterinary WA, USA • Dr Li Liu – Assistant Scientist; Division of Global Medicine, The University of Edinburgh, Edinburgh, Scotland, UK Medicine, The University of Edinburgh, Edinburgh, Scotland, UK; Disease Epidemiology and Control; The Johns Hopkins Center for Lead health Specialist, The World Bank, Washington, DC, USA • Global Health; Johns Hopkins University, Baltimore, MD, USA • Dr Elizabeth Grant – Director, Global Health Academy; Senior REGIONAL EDITORS Professor Azeem Majeed – Head, Department of Primary Care Lecturer in Global Health and Development; Programme Director, & Public Health; Imperial College London, UK • Professor Julie North America: Professor Robert E. Black – Director for the Distance Learning Masters Programmes; Centre for Population Meeks Gardner – Head, Caribbean Child Development Centre; Institute of International Programs; Johns Hopkins Bloomberg School Health Sciences; College of Medicine and Veterinary Medicine; Consortium for Social Research and Development; The University The University of Edinburgh, Edinburgh, Scotland, UK • Professor of Public Health; Johns Hopkins University, Baltimore, MD, USA • of the West Indies, Mona, Jamaica • Professor Katherine L. Devi Sridhar – Chair in Global Public Health; Centre for Global Professor Rifat Atun - Professor of Global Health Systems; Director O'Brien – Executive Director, The International Vaccine Access Health Research, College of Medicine and Veterinary Medicine, The Center (IVAC); Global Disease Epidemiology and Control, Center for of Global Health Systems Cluster; Harvard T. H. Chan School of Public University of Edinburgh, Edinburgh, Scotland, UK and Co-Director, Health, Harvard University, Cambridge, MA, USA Global Health; Johns Hopkins University Bloomberg School of Public Centre for AIDS Interdisciplinary Research (CAIRO), Blavatnik School Health, Johns Hopkins University, Baltimore, MD, USA • Associate South America: Professor Claudio F. Lanata – Nutritional of Government, Oxford University, Oxford, UK • Professor Sarah Professor Ozren Polašek – Associate Professor of Biostatistics; Research Institute; Lima, Peru • Professor Cesar G. Victora – Cunningham-Burley – Dean of Molecular, Genetic and Population Director, Croatian Centre for Global Health; Department of Public Professor of Epidemiology; Federal University of Pelotas; Pelotas, Health Sciences, Professor of Medical and Family Sociology, The Health; The University of Split School of Medicine, Split, Croatia Brazil University of Edinburgh, Edinburgh, Scotland, UK • Professor Aziz • Professor Carine Ronsmans – Professor of Epidemiology; Sheikh – Co-Director of the Centre of Medical Informatics and Europe: Professor Sue C. Welburn – Vice-Principal, Global Head, Department of Infectious Diseases Epidemiology, The London Professor of Primary Care Research and Development, The University School of Hygiene and Tropical Medicine, London, UK • Professor Journal of Global Health is published biannually by the Edinburgh University Global Health Society (EUGHS). Access; Professor, Medical and Veterinary Molecular Epidemiology; of Edinburgh, Edinburgh, Scotland, UK The University of Edinburgh, Edinburgh, Scotland, UK • Dr Evropi Craig E. Rubens – Affiliate Professor of Global Health and Affiliate The aim of the EUGHS is to inspire and educate individuals, groups, organizations and communities about global Theodoratou – Centre for Global Health Research, College of Professor of Pediatrics; Executive Director, Global Alliance to Prevent EDITORIAL BOARD health issues. Its key objectives are to provide a platform for University of Edinburgh students to share ideas and/ Medicine and Veterinary Medicine, The University of Edinburgh, Prematurity and Stillbirth (GAPPS); Seattle Children’s and University of Washington School of Medicine, Seattle, WA, USA Professor Samir or experiences in Global Health with each other; organize meetings and other events to raise awareness of Global Edinburgh, Scotland, UK Dr Davies Adeloye – Covenant University Ota Ogun State, K. Saha – Head, Department of Microbiology; Executive Director, Nigeria; and Centre for Global Health Research, The University of Health issues within the University and more widely; and to give students opportunities to present their work. The Africa: Professor Shabir A. Madhi – Executive Director, National The Child Health Research Foundation; Bangladesh Institute of Child Edinburgh, Edinburgh, Scotland, UK • Professor Abdullah H. Institute for Communicable Diseases; DST/NRF South African Health, Dhaka Shishu Hospital, Dhaka, Bangladesh • Professor society is affiliated to the Centre for Population Health Sciences of the University of Edinburgh, and it works close- Baqui – Director, International Center for Maternal and Newborn Research Chair in Vaccine Preventable Diseases; University of the Joshua A. Salomon – Professor of Global Health; Department of Witwatersrand, Johannesburg, South Africa • Professor Mark R. Health; The Johns Hopkins Center for Global Health; Johns Hopkins ly with the Global Health Academy of the University of Edinburgh. University, Baltimore, MD, USA • Dr Luciana Brondi – Centre Global Health and Population; Harvard School of Public Health and Tomlinson – Department of Psychology; Stellenbosch University, Harvard Global Health Institute; Harvard University, Boston, MA, USA Matieland, South Africa for Global Health Research, College of Medicine and Veterinary Medicine, The University of Edinburgh, Edinburgh, Scotland, UK • • Dr Giorgio Tamburlini – Research Director, Institute of Child Asia: Professor Zulfiqar A. Bhutta – The Noordin Noormahomed Dr W. Abdullah Brooks – Director, Kamalapur Field Site; Head, Health IRCCS Burlo Garofolo; Scientific Director, European School for Sheriff Endowed Professor and Founding Chair; Department of Infectious Diseases Unit; The Johns Hopkins Center for Global Health; Maternal, Newborn, Child and Adolescent Health; Director, WHO Paediatrics and Child Health; The Aga Khan University, Karachi, Johns Hopkins University, Baltimore, MD, USA Professor Graham Collaborating Centre on Maternal and Child Health; The Institute Pakistan • Professor Shams El Arifeen – Professor of Brown – Foundation Director; Nossal Institute for Global Health; Burlo Garofolo, Trieste, Italy • Assistant Professor Alexander C. Subscription (including postage): GBP 30 per any single issue, GBP 50 for annual subscription, Epidemiology; Director, Centre for Child and Adolescent Health; The University of Melbourne, Melbourne, Australia • Ruth M. Tsai – Assistant Professor of Psychiatry; Department of Psychiatry; International Centre for Diarrhoeal Disease Research, Bangladesh Campbell – Honorary Associate/Researcher; Harvard Global Equity Massachusetts General Hospital and Harvard Medical School; Boston, GBP 25 for students, GBP 20 for EUGHS members. Please contact JoGH office for details. (ICDDR, B); James P. Grant School of Public Health, BRAC University, Initiative, Cambridge, MA, USA • Associate Professor Josip Car USA • Professor Peter Tugwell – Professor of Medicine and Dhaka, Bangladesh – Director of Health Services and Outcomes Research Programme; Epidemiology & Community Medicine; co-Director, WHO Collaborating Contact for subscription information: Lee Kong Chian School of Medicine; Imperial College &Nanyang Centre for Knowledge Translation and Health Technology Assessment China: Professor Yan Guo – Professor of Public Health; Vice Technological University, Singapore and Director, Global eHealth in Health Equity; Director, Centre for Global Health; Institute of Journal of Global Health President, Peking University Health Science Center; Vice Chairman Unit, School of Public Health, The Imperial College London, London, Population Health and Department of Medicine; Faculty of Medicine, Rachael Atherton of the Chinese Rural Health Association; Vice Director of the China UK • Professor Simon Cousens – Professor of Epidemiology The University of Ottawa, Ottawa, Canada • Associate Professor Academy of Health Policy (PKU); School of Public Health, Peking and Medical Statistics; Director of Research Degrees; Department Anita K. Zaidi – Director of Research, Department of Paediatrics Edinburgh University Global Health Society University,Beijing, China • Dr Kit Yee Chan – Visiting Scholar, of Infectious Disease Epidemiology; London School of Hygiene and and Microbiology; The Aga Khan University, Karachi, Pakistan and School of Public Health, Peking University Health Science Center; Tropical Medicine, London, UK • Professor Karen Edmond – Director, Enteric and Diarrheal Diseases, Bill and Melinda Gates Centre for Global Health Research Peking University, Beijing, China and Reader, Center for Population Winthrop Professor; School of Paediatrics and Child Health; The Foundation, Seattle, WA, USA • Associate Professor Lina Zgaga College of Medicine and Veterinary Medicine Health Sciences; College of Medicine and Veterinary Medicine; The University of Western Australia, Perth, Australia • Dr Christa L. – Associate Professor of Epidemiology; Department of Public Health University of Edinburgh, Scotland, UK Fischer Walker – Associate Scientist; Division of Global Disease and Primary Care; Trinity College Centre for Health Sciences, Dublin, The University of Edinburgh India: Professor K. Srinath Reddy – President, Public Health Epidemiology and Control; The Johns Hopkins Center for Global Ireland • Professor Stanley H. Zlotkin – Professor, Departments Teviot Place Foundation of India; President, World Heart Federation; Chair, Core Health; Johns Hopkins University, Baltimore, MD, USA • Professor of Paediatrics, Nutritional Sciences and Dalla Lana School of Public Edinburgh EH8 9AG, Scotland, UK Advisory Group on Health and Human Rights, National Human Rights Wendy J. Graham – Chair in Obstetrics and Gynaecology; The HealthVice President, Medical and Academic Affairs, Department Commission, India; President, National Board of Examinations, Institute of Applied Sciences; The University of Aberdeen, Aberdeen, of Pediatrics and Surgery, Division of Clinical Medicine; The Hospital Phone +44 (0) 131 651 5142 Fax +44 (0) 131 650 6909 India; Bernard Lown Visiting Professor of Cardiovascular Health, UK • Dr David B. Hipgrave – Honorary Senior Researcher; Nossal for Sick Children, The University of Toronto, Toronto, Canada • Dr Henry Kalter – Institute of International Programs; Johns Hopkins [email protected] Harvard School of Public Health, Boston, MA, USA • Dr Harish Institute of Global Health; University of Melbourne, Melbourne, Bloomberg School of Public Health; Johns Hopkins University, Nair – Public Health Foundation of India; Indian Institute of Public Australia and UNICEF, New York, NY, USA • Dr Hope L. Johnson – Center for Global Health; Johns Hopkins University Bloomberg Baltimore, MD, USA Health, New Delhi, India; Centre for Global Health Research, College School of Public Health, Johns Hopkins University, Baltimore, MD, of Medicine and Veterinary Medicine, The University of Edinburgh, USA and Head, Programme Outcomes and Impact, GAVI Alliance, MANAGING EDITOR/NEWS SECTION EDITOR Edinburgh, Scotland, UK Geneva, Switzerland • Professor Keith P. Klugman – William H. Australia and Western Pacific: Professor Trevor Duke Foege Chair of Global Health; Hubert Department of Global Health, Rachael Atherton – Centre for Global Health Research; College Journal design: Snježana Engelman Džafic´ for LASERplus, Zagreb, Croatia – Director, Centre for International Child Health; Director, WHO Rollins School of Public Health; The Emory Universtiy, Atlanta, GA, of Medicine and Veterinary Medicine, The University of Edinburgh, Collaborating Centre for Research and Training in Child and USA • Assistant Professor Ivana Kolčić – Assistant Professor Edinburgh, Scotland, UK Realisation: LASERplus, Zagreb, Croatia, www.laser-plus.hr Journal of
Part two Global Health: The Mission Statement
The Journal of Global Health is a peer-reviewed journal pub- lished by the Edinburgh University Global Health Society, a not-for-profit organization registered in the UK. TheJour - nal publishes editorials, news, viewpoints, original research and review articles in two issues per year. The Journal’s mission is to serve the community of research- ers, funding agencies, international organizations, policy- makers and other stakeholders in the field of international health by: • presenting important news from all world regions, key organizations and resources for global health and devel- opment; • providing an independent assessment of the key issues that dominated the previous semester in the field of glob- al health and development; • publishing high-quality peer-reviewed original research and providing objective reviews of global health and de- velopment issues; • allowing independent authors and stakeholders to voice their personal opinions on issues in global health. Each issue is dedicated to a specific theme, which is intro- duced in the editorial and in one or more viewpoints and related articles. The news section brings up to five news items, selected by the Journal’s editorial team, relevant to seven regions of the world, seven international agencies and seven key resources important to human population health and development. We particularly welcome submissions addressing persist- ing inequities in human health and development globally and within regions. We encourage content that could assist international organizations to align their investments in China managed to reduce its child mortality by two health research and development with objective measure- thirds between 1990 and 2006, thus achieving Mil- ments or estimates the disease burden or health problems lennium development Goal 4 nearly a decade be- fore the deadline in 2015 that was suggested by the that they aim to address. Finally, we promote submissions United Nations. This stunning success, given coun- that highlight or analyse particularly successful or harmful try's population, was a combined result of econom- practices in management of the key resources important ic and social development, clear and progressive for human population health and development. health policies, improved nutrition and micronutri- All editors and editorial board members of the Journal are ent supplementation, development of China's independent health professionals based at academic institu- health system, increased coverage and quality of tions or international public organisations and so are well implementation of life–saving interventions such as placed to provide objective professional evaluation of key vaccines and antibiotics. topics and ongoing activities and programs. We aim to stay true to principles of not-for-profit work, open knowledge and free publishing, and independence of academic thought from commercial or political constraints and influences. Join us in this publishing effort to provide evidence base for global health!
March 7, 2011 The Editors, Journal of Global Health
www.jogh.org June 2017 • Vol. 7 No. 1 PaPErSVIEWPOINTS June 2017 •Vol. 7No. 1•010501 rei Goto atun Shingo Fukuma and resilienthealthsystems Earthquake: lessonsfordevelopingresponsive Fukushima aftertheGreatEastJapan [email protected] USa Boston Ma Harvard University Harvard THChanSchoolofPublicHealth Department ofGlobalHealthandPopulation Prof. r Correspondence to: * 6 5 4 3 2 1 Jointlastauthorship. Department ofGlobalHealthandPopulation, r Graduate SchoolofBusinessadministration, Boston UniversitySchoolofPublic Center forInnovativeresearch Department ofHealthcareEpidemiology, USa Harvard University,Boston,Massachusetts, Harvard T.H.ChanSchoolofPublicHealth, USa School ofMedicine,Indianapolis,Indiana, Keio University,Yokohama,Japan Massachusetts, USa 801 Massachusettsavenue,Boston, Health, DepartmentofGlobal Fukushima, Japan (CIrC(2)LE), FukushimaMedicalUniversity, Communities andClinicalExcellence 606–8501, Japan. Kyoto University, Yoshidakonoe, Sakyo, Kyoto egenstrief Institute,IndianaUniversity 1,6,* ifat a , ShunichiFukuhara 02115 4 tun , ThomasSInui
1,2 , Shahiraahmed 2,5
, rifat 1,2,*
3 , increases indemandgenerated bythedisaster. unexpected sudden meet effectively to capacity its limited Fukushima results suggest that poorer reserves in the health care delivery system in The trends. national the to compared observed were individuals older in mortality rates for selected non–infectious conditions common among saster, and explored health systems’ strengths and vulnerabilities. Spikes tems infrastructure in Fukushima before and five years following the di Conclusions We analyze changes in indicators of health and health sys gency ambulanceservices. ness to acute demand because of pre–existing restricted capacity in emer responsive limited indicate may which finding a Fukushima: in rise a a sharp rise in ambulance calls in Iwate and Miyagi, we did not see such in all three prefectures to those in Japan as a whole. Although we found ter. The cause–specific mortality rate from cancer followed similar trends cause mortality (Miyagi and Fukushima) were also observed post–disas to lung disease (all three prefectures), stroke (Iwate and Miyagi), and all– creased nationwide. Compared to Japan average, spikes in mortality due myocardialinfarction increased de it while Fukushima in substantially to due rate mortality the that found we disaster triple the After rates. al all–cause mortality in Fukushima was in general higher than the nation residents than the Japan average. Even before the disaster, age–adjusted 100 per clinics and calls ambulance nurses, physicians, fewer with underserved’, ‘medically were prefectures three All average. Japan the than industry higher–level less and capita, per product domestic gross socio–economic indicators, an older population, lower productivity and Findings All three prefectures, and in particular Fukushima, had lower prefecture. changes over time in age–adjusted cause–specific mortality rates in each describe to plots line time–trendWe made ‘disaster’. the (2011–2014) affected by the disaster, to compare trends before (2005–2010) and after Japan for Fukushima, Miyagi, and Iwate, the prefectures that were most of Statistics Official of site portal the from available (2005–2014) data Methods to enhance its responsiveness and resilience to current and future shocks. re–designing Fukushima’s health system, and by extension that of Japan, indicator changes that accompanied the disaster, and discuss options for health population–level examine we study this In substantial. was tem this ‘triple disaster’ on the health of local populations and the health sys of impact adverse The Japan. of history the in severedisasters most the of one produced meltdowns, nuclear–reactor and tsunami a by lowed Background On 11 March 2011, the Great East Japan Earthquake, fol We used country–level (Japan–average) or prefecture–level or (Japan–average) country–level Weused 278 www.jogh.org • doi:10.7189/jogh.07.010501 global journal of health
000 000 ------www.jogh.org • doi:10.7189/jogh.07.010501 siveness andresilience tocurrent andfuture shocks. surrounding prefectures, and discusses options for re–designing the health system to enhance its respon and Fukushima in system health the and health population on disaster triple the of effect the ascertain to disaster triple the after beforeand indicators health population–level in changes examines study This triple disasteronthehealthsysteminFukushimaandhealthresponse. the of effect the examine to undertaken been has study no date, However,to exceeded. been have may health care, the capacity of health systems of Fukushima and the other most–affected nearby prefectures for demand and need of surge this with Faced homes. original their in not but prefecture Fukushima the history of Japan [ Japan of history the in disasters severe most the of one produced meltdowns nuclear–reactor damage–related tsunami and tsunami, a by followed Earthquake, Japan East Great the earthquake, massive a 2011, 11th March On ]. The causes of death of greatest interest were: all– were:interestgreatest of death of causes The [7]. Japan in statistics vital of survey throughthe ly Our main health outcome indicators were cause–specific mortality rates, which are measured regular- Health outcomeindicators Those indicatorswere measured ofHealth,LabourandWelfare byJapaneseMinistry [ surveys patient visits number of hospitalizations, number of ambulance calls, and health expenditure per capita. cators: number of hospitals, number of clinics, number of physicians, number of nurses, number of out To assess health system factors that might affect mortality rates, we used the following supply–side indi Health systemindicators Affairswere ofInternal andCommunications[ measuredoftheJapaneseMinistry bysurveys disaster. indicators the Those to due evacuees category,of job proportion,number ment and rate crime years, percentage of productive population, fertility rate, real gross domestic product (GDP), unemploy 65 over density,people population of percentagesize, population including: indicators, cio–economic To analyze contextual characteristics in Fukushima and other prefectures, we used demographic and so Population indicators before (2005–2010)andafter(2011–2014)thedisaster. time the span to and indicators most for wereavailable data which for 2005–2014 period time to study our Weconfined basis. annual or quarterly a on website the at available made and government anese data are officially compiled and aggregated from national surveys and administrative registers by the Jap [4 Center Statistics National the by managed Japan, of Official Statistics of website the was data of source below.main delineated Our further aggregates, prefectural–level and (average) overall Japan for collected were indicators outcome and systems, health population, of list predetermined a for Data known forbeingsocio–economicallylesswelldevelopedcompared tootherregions ofJapan. is Tohokuwhich region,the in are Iwate and Miyagi Fukushima, of prefectures three All Earthquake. all and for Fukushima, Miyagi and Iwate – the prefectures that were most affected by the Great East Japan We used publicly available data, data from sources, government and published literature for Japan over Setting anddatasources METHODS 108 by to evacuate their homes. The Japanese government decided to restrict access to nearby areas and about triggered core meltdowns. Radioactive materials leaking from the plant forced people who had lived near that events of series a to led damage induced tsunami The tsunami. huge a by hit was station power ar to US$ 5.2 billion in 2017 exchange rates) [ destroyed. The total cost of the damage to public facilities was estimated at 599.4 billion yen (equivalent [3 tsunami the by caused were deaths those of many and “disaster”, the in died people 3770 nearly Fukushima, In infrastructure (includinghospitals,clinicsandemergency transportation),homesandlives. of destruction with substantial, was system health the on and populations local of health the on disaster [2 meltdown)” nuclear and tsunami, (earthquake, disaster “triple this 000 people were still considered to be displaced evacuees as of July 2015, including 63 including 2015, July of as evacuees displaced be to considered still were people 000 ]. More than 18 than More ]. 1 ]. Among all of Japan’s prefectures, Fukushima was the most severely affected by by affectedseverely most Japan’sof the all was Among Fukushima prefectures,].
030 housing facilities were completely destroyed, and 75 and destroyed, completely were facilities housing 030 279 4 ]. At the time of the disaster, the Fukushima Daiichi nucle
]. The adverse impact of the triple triple the of impact adverse The ]. Lesson fromFukushima forhealthsystems June 2017 •Vol. 7No. 1•010501
159 were partially partially were 159
5 000 inside inside 000 ]. 6 ]. ]. These These ]. ------
VIEWPOINTSPaPErS PaPErSVIEWPOINTS June 2017 •Vol. 7No. 1•010501 er indicators).We extracteddatafrom theportal siteofOfficial StatisticsofJapan[ *Data before the disaster were measured in 2010. Data after the disaster were measured in 2012 (real GDP), 2015 (number of evacuees) or 2013 (oth- Fukuma etal. Table1. Population Number ofevacueestotheotherprefectures (per1000people) Number of evacuees to the other areas in the same prefecture (per 1000 people) Crime rate(per100 Tertiary (%) industry Secondary industry (%) industry Secondary (%) industry Primary Percentage ofjobcategory: Unemployment proportion (%) Real GDPpercapita(millionyen) Real GDP(trillionyen) Total fertilityrate Fertility rate(per1000people) Percentage ofproductive populationaged15–64(%) Population density(/km Percentage ofelderlyover65(%) I Population (100 ndIcators 000 people) after the triple disaster, but the industrial production index, which is used to track the production of of production the track to used is which index, production industrial the disaster,but triple the after Fukushima’s economic indicators as measured by average real GDP and income per person remained flat in 2010. after the disaster reaching 26.9% of the total population in the prefecture in 2013, compared with 25.0% years 65 of age the over people of proportion increasingrapidly, morean changing with is Fukushima country,the restof the in trendsobserved the Comparedto overall. Japan structurein in age 0.53% the and Iwate, in 2.9% Miyagi, in 1.29% to compared 2010 and 2015 between 3.9% prefectures: other in In Fukushima, the relative decline in the population level after the disaster was greater than that observed (Tablejobs in“high–levelindustry” 1). in participated prefecturesthree the of each in population the of proportion smaller A overall. Japan in 23.0% to compared Iwate and Miyagi Fukushima, in respectively population the of 27.2% and 22.3%, respectively, compared to 4.0 million yen for Japan overall. Elderly over 65 years of age represented 25.0%, GDP per capita in 2010, before the disaster, was 3.8, 3.5, 3.3 million yen in Fukushima, Miyagi and Iwate, Table1 Changes inpopulationindicators r Japan)–average, Fukushima,MiyagiandIwate. 2014) and to compare the trends before (2005–2010) and after (2011–2014) the disaster in Japan (using produce descriptive statistics and to establish a time–trend line plot to examine changes over time (2005– to disaster the after LP,and StataCorp before TX: indicators WeStation, all USA). College analyzed 13. v.13Release Software:Stata Statistical using Stata analyzed were2013. data (StataCorp. quantitative All Analyses cause–specific mortalityratesbasedonthemodelpopulationofJapanin1985[ age–adjusted Weused phase. premorbid long a have might cancers since change, acute an show not might also respond to the altered circumstances imposed on the disease. Cancer mortality, however, might es suffered. Similarly, marginally compensated chronic pulmonary disease and/or reactive airway disease disaster, social and physical dislocations, as might suicide in the face of great personal and physical loss We reasoned a priori that cardiovascular mortality and stroke might be acutely reactive to the stress of the suicide. cause mortality, and that from myocardial infarction, cerebrovascular disease, cancer, lung disease, and 000 people) ESULTS and healthsystemindicatorsofFukushima,Iwate,MiyagiJapanbefore andaftertheearthquake* 2 ) displays demographic and socio–economic indicators before and after the triple disaster. Real disaster.Real triple the after and before indicators socio–economic and demographic displays
280 eoeAtrBfr fe eoeAtrBfr After Before After Before After Before After Before 4. 4. 2. 1. 718. 4. 341.3 343.4 84.8 87.1 319.5 322.3 141.2 147.2 60.0 29.2 .915 .513 .714 .71.43 1.37 1.46 1.37 1.34 1.25 1.53 1.49 256. 606. 145. 5862.1 65.8 59.0 61.4 63.4 66.0 60.4 62.5 502. 232. 722. 3025.1 23.0 28.7 27.2 23.8 22.3 26.9 25.0 051. 332. 351. 2061263.9 1270.6 13.1 13.5 23.0 23.3 19.7 20.5 . . 001. . . 1011.1 11.0 6.0 5.9 10.1 10.0 5.3 6.7 7.6 ...... 4.0 5.1 3.3 5.1 4.1 5.7 3.6 5.1 ...... 4.1 4.0 3.6 3.3 3.8 3.5 3.9 3.8 ...... 1. 517.5 512.5 4.7 4.4 9.1 8.2 7.6 7.6 ...... 8.2 8.5 7.2 7.4 8.2 8.2 7.5 8.0 – F 8 ukushIma ]. 44.1 60.6 – – – 70.5 22.1 5.0 – m Iyag I 53.9 6.7 – – – www.jogh.org 62.3 24.3 12.0 – 8 • doi:10.7189/jogh.07.010501 I wate ]. 24.7 1.5 – – – 66.5 23.7 4.0 – J apan 190.5 – – – – - expenditure per capita. Number ofhospitalizations.2G: Numberofambulancecall.2H:Health cians. 2D:Numberofregistered nurses.2E:Numberofoutpatients. 2F: 2A: Numberofhospitals.2B: ofclinics.2C:Numberphysi www.jogh.org Figure 1. Time–trend inhealthsystemindicators5yearsafterthedisaster. • doi:10.7189/jogh.07.010501 highest numberofevacueeslocatedtootherprefectures ofJapan(44 Iwate, Fukushima had the highest number of evacuees residing in the same prefecture (60 was worse for those from Fukushima, due to the nuclear power plant accident. Compared to Miyagi and all over Japan), and were unable to return to the coastal areas most affected by the disaster. This situation By 2015, four years after the disaster,the 190 after years four 2015, By compared toJapanasawhole,remained lowandactuallyimproved afterthedisaster. unemployment rates declined, between 2010 and 2012. Fukushima crime rates, which were already low [ achieved levels pre–disaster the to recovered not had 2014 by and declined, industries, manufacturing Figure 1 Figure nurses, ambulancecallsandclinicsper100 physicians, fewer disaster,with the before underserved” “medically were studied prefectures three All Changes inhealthsystemindicators and clinics declined in Fukushima immediately after the disaster – a reflection of the physical destruc physical reflection the a of – disaster the after immediately Fukushima in declined clinics and 9 ]. By contrast, in Miyagi, Iwate and Japan overall, average real GDP and income per person rose, while
shows time–trends for indicators related to health system capacity. The number of hospitals hospitals capacity.of system number health The to related indicators for time–trends shows
000 people had remained as evacuees (located to prefectures to (located evacuees as remained had people 000 281 000 residents compared withJapanaverages.
- age (a deficit of 30 physicians/100 er inFukushimacompared withtheJapan–aver nurses in hospitals pre–disaster already was low and physicians of number The facilities. of tion workers in Fukushima at present (see present at Fukushima in workers care health of shortage structural the address to needed still however,are efforts, Long–term areas. afflicted the to dispatched were fectures ter, teams of health professionals from other pre 120 nurses/100 spond toneed. damaged emergency transportation system to re the of inability the reflect might base population older an of spite in rates call ambulance in ing’ Iwate, Miyage and the rest of Japan. This ‘flatten to prefecture’scompared the population in sons spite of it having a higher proportion of older per yagi, we did not see such a rise in Fukushima, in Mi and Iwate in calls ambulance in rise sharp a in Japan’s aging society [ mand from increasing numbers of elderly patients de rising the reflect may which trend a whole, a as Japan in gradually increased calls lance the rise expenditures [ expenditures rise the of elements main the constituted workers) care health recruitmentof for as well areas,as coastal in clinics and hospitals of reconstruction and (repair care medical of provision for penditures ex and exposure) radiation with deal to needed facilities decontamination of construction (the reconstructionprojects environmentalfor health to related investments capital major that reveals expenditures health rising the of breakdown A dramatically in Fukushima just after the disaster. however,expenditures, roseHealth clinics). and may reflect the loss of medical facilities (hospitals especially in Fukushima and Miyagi. This decline talizations declined immediately after the disaster, hospi and visits outpatient of number the tem, sys health the in utilization service of terms In panels 2Cand2D).
100). Lesson fromFukushima forhealthsystems
000 people). Soon after the disas June 2017 •Vol. 7No. 1•010501 9 ]. The number of ambu of number The ]. 10]. Although we found
000 people and
600), and the Figure 1, Figure ------
VIEWPOINTSPaPErS PaPErSVIEWPOINTS June 2017 •Vol. 7No. 1•010501 Fukuma etal. justed deathrateduetosuicide. disease. 1D:Age–adjusteddeath rateduetocancer. 1E: Age–adjusteddeathrateduetolungdisease.1F:Age–ad 1B: Age–adjusteddeathratedue tocardiovascular disease.1C: Age–adjusteddeathrateduetocerebrovascular Figure 2. persons and property. To our knowledge, this is the first paper to describe changes over time in multidi quake and its sequelae of a tsunami and nuclear reactor meltdown were responsible for major damage to Earth Japan East Great the Japan, in Iwate and Miyake Fukushima, of prefecturesaffected three the In Discussion in 2009,butnot2011,whenthere mayhavebeenasharperrateofdecline inthethree prefectures. suicides of excess an been have may there inspection, simple prefectures.By three the in disaster the to ter as well as in Japan as a whole. There appears to have been no ‘epidemic’ of suicides temporally related eases in 2011 ( dis lung from mortality of rates the in increase sharp also were there Iwate and Miyagi Fukushima, In from MIwere evident(seeFigure 2,panel1B). Fukushima also differed from the rates and trends in Iwate and Miyagi where reductions in mortality rates in MI to due rate Mortality nationwide. decreased rate this while Fukushima, in substantially increased shima to those in Japan as a whole, after the disaster the mortality rate due to myocardial infarction (MI) Fuku in trends similar cancer, show to due deaths as such rates, mortality cause–specific other While yagi from 386deathsper100 100 per fromdeaths Iwate 418 in mortality in rise er in Fukushima that rate was 415, rising to 480 in 2011, but decreasing to 403 in 2012. We found a high national average for Japan. The mortality in Japan in 2010 was 390 deaths per 100 the than higher general in was Fukushima in mortality disaster,all–cause the beforeage–adjusted Even ter inFukushima,Miyagi,IwateandtheaverageindicatorsforJapanasawhole. Figure 2 displays time–trends in health indicators before (2005–2010) and after (2011–2015) the disas Changes inhealthoutcomeindicators The suicide rates ( study prefectures themortalityratesformlungdiseasere–established theirdownward trend. served from 47 deaths per 100 per deaths from47 served ob rise sharp a was there Fukushima in 2010, to 2005 period the in diseases fromlung rates mortality Time–trend inhealthoutcome indicators5yearsafterthedisaster. 1A:Age–adjustedall–causedeathrate. Figure 2, panel 1E). While, nationally and in Fukushima there were steep declines in the
Figure 2, panel 1F) in the three prefectures appeared to be declining before the disas 000 populationto713in2011afterthedisaster.
000 in 2010 to 54 deaths per 100 per deaths 54 to 2010 in 000 282
000 population in 2010 to 699 in 2011and in Mi in 2011and in 699 to 2010 in population 000
000 in 2011. From 2012, in the three the in From2012, 2011. in 000 www.jogh.org • doi:10.7189/jogh.07.010501
000 population while ------www.jogh.org • doi:10.7189/jogh.07.010501 fected areas[12, fected af- Welfarethe and to Labor Health, of Ministry the by dispatched were teams’ response care health tal nition by the government and local authorities of the psychological consequences of the events and ‘men - disaster the suicide rates did not spike. It was reported that within days of the disaster there was a recog Mental health problems typically emerge after major disasters, but in the three prefectures affected by the proportion of elderly residents as a result of younger and healthier people migrating out of the prefecture. high rates of MI could reflect the changing demographic profiles in Fukushima, leading a relatively higher the prefectures.Fourth, other than longer be to likely more was MIs for time’) onset–to–balloon ‘pain by all prefectures in Japan, so it is likely that the time lapse for an effective response (for example as measured sudden rise in demand generated by the disaster. Third, Fukushima has the third largest land–area among and unexpected meet to response effective an hindering Fukushima, in whole a as system delivery care neighboring prefectures. Second, and related to the first explanation, there were poorer in health reserves its in than Fukushima in severe more was physicians of shortage the First, Fukushima. in system health the for challenges of set unique a present may Fukushima in MI to due rate mortality high observed The age theirchronic conditions,anadversitywhichcouldhaveresulted inexcesspremature deaths[ Many people in the afflicted areas struggled with access to medications and treatments to effectively man magnified the adverse effects of the disaster, with the tsunami leading to widespread destruction of homes. have might which milieu), socio–economic the as (such them in factors contextual unique the as well as health systems factors, such as poor quality of care and inadequate supply to of attributed resourcesbe could in difference these This prefectures, rates. national average the than higher was disaster the by fected af prefectures other two the and Fukushima in mortality all–cause disaster,age–adjusted the Following generatedbythedisaster.to meetthesuddenandunexpectedriseindemandforhealthservices siveness and resilience of the health system in Fukushima than Iwate and Miyage. Hence, it was unable ambulance services. These results suggest that there were poorer reserves, and weaker emergency respon emergency with need/demand to respond to capacity limited a to due Fukushima, in system health the of responsiveness poor indicate may which Fukushima, in rise a such see not did we Miyagi, and Iwate trends in Fukushima to those in Japan as a whole. Although we found a sharp rise in ambulance calls in decreased nationwide, but other cause–specific mortality rates such as deaths due to cancer show similar We found that mortality rate due to MI increased substantially in Fukushima, for example, while this rate ic conditionscommoninagingpopulations. triple disaster. The health outcomes examined appear to show manifestations of the high burden of chron the by brought destruction the to 2011 in vulnerable demographically and socio–economically already By our observations, Fukushima, its residential populations, and those of its neighboring prefectures were riod before (2005–2010)andfiveyearsafter(2011–2014)thedisaster. pe the in prefecturesaffected other and Fukushima for indicators system health and health mensional after the disaster [ disaster the after resilience community increase to strengthened and maintained was cohesion social disaster the after and cohesion before the disaster was shown to be associated with lower risk of post–traumatic stress disorder, stronger fabric to protect its members, especially the elderly and vulnerable [ a weave to able been have to appears society ‘tear,’the not fabric social the did only not exemplary: was responsivenessresilienceJapan’sand of responsedisaster resiliencesocietal the and The to system. health The aftermath of the triple disaster revealed social cohesion, as well as the strengths and deficiencies in the an important issue affecting Japan’s health system not only in Fukushima, but also elsewhere in rural Japan. tices generally. The undersupply of health care workers, largely as a result of geographic maldistribution, is fromJapan’sothers and ways, resourcesremainingunusual prac its in cultural mobilized and society civil There were also some striking and encouraging responses to the disaster, some from the health system, as it the disasterdespitetheirparticipationincommunity–basedprograms [ self–reported mental health found that respondents felt a sense of “isolation” for at least 18 months after rates inFukushimadeclined from 6.7per100 crime Conversely,that increase, report not we did Crimes Japan. of areas affected the in emerge not did and its resilience came under pressure, as the health system tried to meet the ongoing needs of vulnerable resources. human challenged, additional Yet,was needed system much health the responsivenessof the nity as alternative information source and communication platform. Voluntarism was evident – providing commu local in role important an played (ICT) technology communications information in Advances ]. Risk factors for mental illness need continued attention, however [ however attention, continued need illness mental for factors Risk 13]. ]. Social violence, witnessed in other countries in the aftermath of natural disasters, disasters, natural of aftermath the in countries other in witnessed violence, Social 17]. 283 000 peoplein2010to5.3per 100 Lesson fromFukushima forhealthsystems 15]. June 2017 •Vol. 7No. 1•010501 16]. Community–level social 000 in2012. ]. A survey of of survey A 14]. 11]. ------
VIEWPOINTSPaPErS PaPErSVIEWPOINTS June 2017 •Vol. 7No. 1•010501 Fukuma etal. in Japan in 2015 elaborated these lessons as follows [ follows as lessons these elaborated 2015 in Japan in Meeting WorldRegional the Summit of Health auspices the under Fukushima in meeting group expert resilience.An disaster of aspects community and determinants social wider managing relationto in also inform health system transformations elsewhere to enhance responsiveness and resilience to shocks, but In retrospect, several lessons emerge from the response of Fukushima to the triple disaster, ones that may shocks. systems toenhancetheirresponsiveness andexternal andresilience tomajorinternal health re–designing for options discussing when important be should Fukushima from learned lessons study,the the by on faced reflecting challenges methodological the for allowing Even Japan. of fectures major natural disaster revealed common problems for health systems that may be applicable to other pre not be generalizable to other disasters. However, notwithstanding contextual differences, this unexpected may prefecturesaffected the of system health the on disaster triple the of effect the Third, populations. of Official Statistics of Japan and were not able to secure individual level data on the affected and control site portal the from data available used only we Second, children. their for exposure radiation of effects term long the about concerned be might who families example, for Fukushima; leaving families young many of result the be may This people. elderly of proportion increasing an with rapidly more aging is those changes would affect both numerators and denominators in our analysis. For example, Fukushima aggregated data. Socio–economic and demographics changed in the three study prefectures over time and There are two significant limitations of this study. First, this is an ecological study using prefecture–level for there–design ofsustainablehealthsystems[20]. challenge major a be will population careresourceselderly long–term the careand for medical for need Increased system. health to shock internal an itself, in is, aging future.population the fact, in In priority formative change is needed to create health systems in Japan and globally that are responsive and resil pact on health systems and health are globally relevant. Given the uncertainties, nothing less than trans- im their and anywhereworldwide, and anytime happen could disasters major and shocks Contextual coverage [22, and be resilient to the emerging challenges and shocks, while continuing to provide effective universal health respond to have and – milieu socio–cultural changing and disasters natural from shocks ecological crises, economic disability), and illness chronic of burden the in of rise rapid a and aging population to (leading transitions epidemiological and demographic swift of result a as – contexts changing rapidly to subject are globally,countries other and which Japan in systems health by faced challenges the illustrates Fukushima of thehealthsystemtoimprove communication andinclusivedecisionmaking. social networks and building social capital; and (iv) developing and strengthening leadership at all levels tion plans involving public agencies and the private sector; (iii) enabling community mobilization through groups, especially the vulnerable, to inform current and future policies; (ii) establishing multi–sector ac population different in mortality,welfare including social disability,disasters, and of destitution, fects ef long–term the monitoring better (i) by enhanced and groupexpert developed the be can concluded, grating health system and social actions for a more comprehensive response. Resilience, on the other hand, management systemstodeploymobileheathteamsandhealth workersinhealthsystems;and(vi)inte just–in–time creating (v) needs; security health population and individual assure to food and clothing, shelter, temporary communication, transportation, to access immediate providing (iv) disaster; a after resourcesresourcescritical for of increased demand supply and limited immediately emerge to that due dissemination to the public; (iii) creating sufficient reserves to rapidly mobilize and fill healthsystem ‘gaps’ tion; (ii) gathering timely information across for targeted and industry sectors of government action and lishing a local, regional and national framework for rapid information–sharing, decision–making and ac brought by an aging of the population – such as disability and multimorbidity [ multimorbidity and disability as such – population the of aging an by brought system health the to Challenges shift. unfavorable appropriatethis responsean to system quiring health reapparent, particularly are Iwate, and Miyagi disaster–affectedFukushima, prefecturesthe of in shifts Demographic countries. other of ahead is expectancy life average of terms in and aging rapidly is Japan ency inrelation tothecourseofevents,timelycommunicationandeffective informationdissemination. transpar of levels high for demand public strong was Thereresources. human health of shortage the and – supplies critical for management chain supply on effect adverse consequent with – transport media, the hospital responses, speed of communication, managing varied messages emerging from official sources and Responsiveness challenges in the disaster were related to effective and timely integration of community and layered process [18],wasevidentinFukushima. populations, in particular the elderly. Community resilience, which depends on local context and multi 23].
284 ]: Responsiveness can be enhanced by (i) estab (i) by enhanced be can Responsiveness 21]: www.jogh.org • doi:10.7189/jogh.07.010501 ] – should be given given be should – 19] ------www.jogh.org rEFEr ENCES • doi:10.7189/jogh.07.010501 23 20 22 21 2 1 3 4 7 5 8 6 9 10 11 13 12 17 15 14 19 16 18 Welcome to Ministry of Health. Labour and Welfare. Available: M Matsumoto - aquickreport. MitigAdaptStrategiesGlobChange.2011;16:803-18. Tsunamiand Earthquake Japan East Greatfrom YasuharaYokokithe Damage N, S, S. Mimura Kawagoe Kazama K, H, K Hayashi 2011; Asia PacJPublicHealth.2012; Government Government of Japan. Portal Site of Official Statistics of Japan. Available: http://www.e-stat.go.jp/SG1/estat/eStatTopPor Outline of Vital Statistics in Japan. [cited 2016 Mar 30]. Available: Government of Japan. Ministry of Internal Affairs of Internal of Japan. Ministry and Communications. Government Available: talE.doc. Accessed:10November2015. Atun equity. go.jp/toukei/saikin/hw/jinkou/other/00sibou/1.html. Accessed:10November2015 Ministry of Health, Labour and Welfare, Government of Japan. Age adjusted mortality rate. Available: line/index.html. Accessed:30March 2016. dex.html. Accessed:30March 2016. Prefecture F. Steps for Revitalization in Fukushima. 2015. Available:2015. Fukushima. in F.Revitalization Prefecture for Steps coverage inLatinAmerica. fukkoukeikaku1151.html. Accessed:12November2015. Atun line:27648258 Atun R 12 April2016. World Health Summit. Impact. Available: S0140-6736(14)62254-6 to primary care in Estonian health system: analysis of national panel data. panel national of analysis system: health Estonian in care primary to nese]. Available: http://www.fdma.go.jp/neuter/topics/fieldList9_3.html. Accessed:15November2015 Japa in [website operations. Rescue on Report Annual Japan. Agency,of Management Government Disaster and Fire N Muramatsu response for the 2011 Japan Disaster. Japan 2011 the for response and NuclearIncident. Shultz Medline:22003986 Kim Y line:21804114 tudinal evidencefrom the2011 BrisbaneFloodDisaster. Hikichi H October 2016. tal and health. Springer; 2013. Available: S Nagata R Goodwin S1049023X11006364 Atun R Aida J doi:10.1111/jjns.12051 der in the aftermath of a disaster? A natural experiment from the 2011 Tohoku earthquake and tsunami. Tohokuand 2011 earthquake the from experiment natural A disaster? a of aftermath the in der miol. 2016; WickesR housing at 10 and 20 months after the Great East Japan Earthquake. order oftheday. first, and will not be the last such challenge we face globally. Learning from our experience must be the [ brings transition this disability and multimorbidity the ient to future shocks and emerging contextual challenges, including the rapid aging of our societies and R R 11: , Kawachi I . Great East Japan Earthquake and early mental-health-care response. closure.pdf (availableuponrequest from thecorresponding author),anddeclare noconflictofinterest. www.icmje.org/coi_dis at form Interest Competing Unified the completed - authors The interest: of Conflict and SA. RA and Shu F guided the writing of the manuscript drafted by SF and SA with input from all authors. SF by led was which analysis, the guided and study the conceived F Shu and RA contribution: Authorship Funding: None. , de Andrade , Aydın, S , Gurol-Urganci I F Kelly JM, Lancet. 2013; . Transitioning health systems for multimorbidity.for systems Transitioninghealth . A Matsunaga , 1759. R Zahnow , , Aida , J , Tomita, N , Takahashi, M 183: H Akiyama , K Inoue , Medline:21413825 doi:10.7189/jogh.06.020701 doi:10.1093/geront/gnr067 , Chakraborty , S 902- , Subramanian , Tsuboya, T D Forbes ,
doi:10.1111/j.1440-1819.2011.02270.x LO, Almeida 382: 10. , Taylor, M . Lessons learned from the Great East Japan Earthquake: impact on child and adolescent health. adolescent and child on impact Earthquake: Japan East Great the from learned Lessons . PLoS One.2012;7: , Teramoto, C . Earthquake, tsunami, radiation leak, and crisis in rural health in Japan. in health rural in crisis and leak, radiation tsunami, Earthquake, . , Hone T Medline:27026337 65- Pell L S Sun , . Japan: super-aging society preparing for the future. the for preparing super-agingsociety Japan: . Lancet. 2015; , Verdeli, H , Kondo , K 99. 24: , Piquero , , Sümer , S Medline:23810020 SV, Kondo K G , Gaines , , 681-8. , Cotlear . Follow-up study of the general and mental health of people living in temporary temporary in living people of health mental and general the of study Follow-up . , Kawachi , I , Leon , , Stokes J http://link.springer.com/chapter/10.1007/978-1-4614-7464-7_7 http://www.worldhealthsummit.org/about-whs/impact.html#c25906
Medline:22815307 385: AR. Neighborhood structure, social capital, and community resilience: longi resilience: community and capital, social structure, Neighborhood AR. , Aran , M SO Prehosp Disaster Med. 2011 Med. Disaster Prehosp e37690. D doi:10.1093/aje/kwv335 . Disaster, social capital, and health. In: Global perspectives on social capi Jr. Modelling psychological responses to the Great East Japan Earthquake Earthquake Japan East Great the Jr.to responses psychological Modelling , Dmytraczenko 1230- A Rosen GR, , Habicht T Medline:22666380 , Gürol, I . Can community social cohesion preventstresscohesion posttraumatic social disor community Can . 285 47. doi:10.1016/S0140-6736(13)61051-X Medline:25458725 Soc SciQ.2015;
, et al. Universal health coverage in Turkey:in coverage health Universal al. et , of enhancement Lancet. 2015 Lancet. , et al. Triple health al. Trauma:Threatmental et evidence-based , , etal.Shiftingchronic disease management from hospitals T http://www.mhlw.go.jp/english/. Accessed: 30 March . 2016 doi:10.1177/1010539512453255 , Frenz http://www.mhlw.go.jp/english/database/db-hw/out- ]. The Fukushima triple disaster is not the the not is disaster triple Fukushima The 18].
Jpn J Nurs Sci. 2015; P doi:10.1371/journal.pone.0037690 , et al. Health system reform and universal health https://www.pref.fukushima.lg.jp/sec/11015b/ doi:10.1007/s11027-011-9297-7 ; ; 26: 386 96: doi:10.1016/S0140-6736(14)61646-9 Psychiatry Clin Psychiatry Neurosci. 2011; 141 : 330- 721
J Glob Health. 2016 Health. Glob J
-5. Gerontologist. 2011; Gerontologist. Lesson fromFukushima forhealthsystems -2. 53. doi:10.1111/ssqu.12144 June 2017 •Vol. 7No. 1•010501 http://www.soumu.go.jp/english/in- Medline:22107762 Medline:26063473 12: 162-5.
Rural Remote Health Remote Rural Medline:25243491 http://www.mhlw. ;6: 51: 020701
425-
. Accessed: 14 doi:10.1017/ doi:10.1016/
Am J Epide- J Am 65: . Accessed: 32. 539- . Med Med- 48. ------.
VIEWPOINTSPaPErS PaPErSVIEWPOINTS June 2017 •Vol. 7No. 1•010502 Till Bärnighausen Günther Fink Zhihui Li financial burden:evidencefromJamaica health careutilizationandreducesfamilies’ User–fee–removal improvesequityofchildren’s material. The onlineversionofthisarticlecontainssupplementary Electronic material: supplementary [email protected] Boston, Ma 677 Huntingtona Harvard SchoolofPublicHealth Department ofGlobalHealthandPopulation r [email protected] USa Boston, Ma 677 Huntingtona Harvard SchoolofPublicHealth Department ofGlobalHealthandPopulation Zhihui Li Correspondence to: 2 1 Jamaica Northern CaribbeanUniversity,Mandeville, Massachusetts, USa Population, HarvardUniversity,Boston, Department ofGlobalHealthand ifat a tun
1 02115 02115 , MingqiangLi 1 ve ve
, PaulBourne
1 , rifatatun
1
,
2 , 1
nya, Tanzania, Burkina Faso, Niger, Democratic Republic of Congo, Leso and arguably increased households’ health expenditures. Studies from Ke services, health to access equitable on effects negative to points evidence the International Monetary Fund (IMF) have promoted user fees [ health services more judiciously [ health care financing burden on governments and encourage clients to use fees have been used to generate revenues for health care providers, reduce User fees refer to charges related to health services at the point of use. Such while theequitystatusimproved fastinthemedium– tolong–term. children, for care health to access of equity the deteriorated policy the short–term, the In indications: different have results long–term to medium– the and short–term The Jamaica. in expenditures health household their reducing and children among utilization care health Conclusions erty shrunk rapidly after 2007 and remained small in subsequent years. pov in not children and poverty in childrenburden between nancial fi differencein The level. low a at kept and implemented was policy –1, to –11 CI: (95% points percentage 6 by expenditure as a share of household’s non–food expenditures) reduced household’sThe long–term. to medium– the burdenfinancial (health in groups wealth across utilization care health in difference minimal was There 2008. after narrowed gap this while poverty; in children utilization increased at a faster rate among children not in poverty than (CI) 1.1 to 3.5, interval confidence 95% 2.0, ratio (odds 97% by increased weeks 4 al, the odds of seeking for health care if the children fell ill in the past Results financial burdens, as well as the impact in the medium– to long–term. fee–removal policy on children’s health care utilization and households’ (ITS) analysis was used to examine the immediate impact of the user– series time Interrupted planning. for purposes other among and tion a national household survey, which collects data on health care utiliza vey of Living Conditions (JSLC) for the periods 1996 to 2012. JSLC is Sur Jamaica fromthe data roundsof 14 utilizes study This Methods long–term impacts of Jamaica’s user–fee–removal for children in 2007. and short– the assesses paper This countries. per–middle–income up in as well as Caribbean, the and American Latin unexploredin ly care utilization and households’ financial burdens has remained large Background Immediately following the implementation of user–fee–remov- 286 The impact of user–fee policies on the equity of health health of equity the on policies user–fee of impact The User–fee–removal had a positive impact on promoting on impact positive a had User–fee–removal P = 0.018). In the short–term (2007–2008), health care 1 ]. Historically, both the World Bank and www.jogh.org P = • doi:10.7189/jogh.07.010502 0.020) right after the the after right 0.020) global journal of 2 , health 3 ]. Yet ------www.jogh.org • doi:10.7189/jogh.07.010502 Peru [ user fees over the past three decades. In the 1980s, user fees were introduced in Honduras, Jamaica, and waveredhave criticizing countries or (LAC) advocating Caribbean between the and American Latin The children andpregnant women[13]. Universal Health Coverage (UHC) [ passed resolutions 58.31 and 58.33 in 2005, urging member states to remove user fees in order to achieve of the high financial burden [ health service utilization, with the poor and those in rural areas disproportionally disadvantaged because tho, and Papua New Guinea have found that the introduction or increase of user fees significantly reduced Box 1. of the West Indies (see Hospital sector,University public the the for in except years 0–18 aged children all for fees user moved re that policy new a implemented Jamaica of Government the 2007, May In Jamaica’ssystem. health region. LAC the in country upper–middle–income an Jamaica, on focuses study Our penditures intheLACregion [16- ex household and utilization care health on policy user–fee–removal of impact the regarding evidence providedconcrete have studies few services, health of delivery the professionals,and health ronmentof envi work the care, patient of quality the on user–fees of effects the assessed have studies of handful a icy change, ITS analysis assumes no changes in other factors that have a potential impact on the outcomes for the policy’s causal effects. By comparing the changes in outcomes right before and right after the pol burden. provide financial strong to households’ evidence analysis (ITS) series time interrupted used We causal relationship between user–fee–removal policy and the changes in health care utilization, as well as a identify not availability,could data and by constrained largely were studies earlier in used Methods and inequalitiesinaccesstohealthcare [23, groupscountry income are and social segregation expanding universalhealthcoverage,with substantial bean, with different characteristics from Africa: Most countries in LAC belong to upper–middle– or high– have been mostly limited to Africa [ Earlier studies on the impact of user–fee–removal on health care utilization and household expenditures children from better–off familiesandcouldalsobedifferent inthemedium– tolong–term. households. Third, the immediate impact of the policy may vary between children from poor families and poor the for especially children, sick with families in expenditures health household reduce will moval user–fee–re Second, care. health access to barrier important an eliminates it because children, among utilization care health increase will user–fee–removal First, hypotheses: three study,tested our we In health care utilizationandhouseholdhealthexpenditure bothonaverageandacross incomegroups. ambulatory services andtheprovision services ambulatory ofpharmaceuticals[ ita). The private sector consists of approximately 200 beds (around 0.1 hospital beds per capita) and dominates proximately 5000 hospital beds across secondary and primary care facilities (around 1.8 hospital beds per cap Jamaica’sap comprises and services providerhospital and health public primary of the is sector health public ance, accountedfor18%ofthetotal[20]. diture and out–of–pocket payments contributed 25%, while other private sources, such as private health insur tional PPP–adjusted US$). In 2013, expenditurethe government accounted for 57% of the total health expen interna 2011 (constant 512 US$ was 2013 in capita per Totalexpenditure health. health on GDP the of 6% Jamaica’s health system is financed through a mix of public and private sources. The government spends around tality ratefellby34%from 25.4 per1000livebirthsin1990to16.72006[ under–5 mortalityratedecreased by36%from 30.6per1000livebirthsin1990to19.52006.Infantmor maternal maternal mortality increased from 79 per 100 reached the MDG4 and MDG5 targets. Before the implementation of user–fee–removal policy in 2007, Jamaica’s expectancy,improvementslife moderate in Despite mortality,infant mortality,under–five and not has Jamaica employment rateofthetotallaborforce was13.2%[19]. (constant 2011 international PPP adjusted US$) and a total population of 2.7 million in 2014. In 2014, the un 8467 US$ of capital per (GDP) Product Domestic Gross a with country upper–middle–income an is Jamaica 14]. In the 2000s, Jamaica and Ecuador removed user fees in the public health sector [ Background informationonJamaica’s healthsystem Box 2). Our study aims to evaluate the impact of user–fee–removal on children's 4 - 11]. Recognizing user fee as a barrier to access health services, the WHO 18]. 12]. UNICEF also committed to support the removal of user fees for 4 – 10, 22]. Our study is of in Latin a America country and the Carib
287 000 live births in 1990 to 91 per 100 24].
20]. User feeremovalandequityofchildren’s healthcare June 2017 •Vol. 7No. 1•010502 19].
000 live births in 2006. The introduces 1 Box 15]. Although ------
VIEWPOINTSPaPErS PaPErSVIEWPOINTS June 2017 •Vol. 7No. 1•010502 Li etal. data from 1996–2012inTables Document. S2,S3andS4inOnlineSupplementary quality.data high with are variables Weusing key resultsregression the ensureITS to the providedtext main the in 2004–2012 from data using results regression the presented only we analysis, ITS ducting able in the regression analysis, has 26.9% missing values before 2004. To solve this problem, when con incomplete: for example, the education level of the household head, which is an important control vari- study.fromthe cluded study.Wethis in surveys roundsof 14 used are totally waves earlier the of Some ex are waves these thus and surveys, 2011 and 2005 2003, in collected not were data module Health 1996–2012. from data use paper,we this For protection. social and housing, education, health, tion, household survey, which consists of six core modules: demographic characteristics, household consump representative nationally a – (JSLC) Conditions Living of Survey Jamaica the from data uses study This Data sources METHODS would shedlightfortheothercountriesonhowtoachievehealthequityinSDGera. an equity dimension, which is a prioritized by the Sustainable Development Goals (SDGs). Our findings fee removal to reduce child and infant mortality. We assessed the impact of user–fee–removal policy with health [ services to access to linked closely is mortality child that Given [19]. world the in countries income per–middle up other in 4% and LAC in countries for 5% to compared 2%, of rate annual an by declined rate tality child and infant mortality has been among the slowest. Between 1990 and 2006, Jamaica’s under–5 mor reducing for target (MDGs) Goals Development Millennium the reaching Jamaica’sin region, progress Evaluating user–fee–removal policy for children has strong policy significance. Of all countries in the LAC as itsmedium–tolong–termimpactofapolicy. that coincide with the policy of interest. Furthermore, ITS analysis can inform us the immediate, as well Box 2. midwives, healers,andother healthprofessionals [30]. to the survey [27– prior 4–weeks the in problem health a experienced she/he if professional health a fromcare sought ual health care As with services. earlier studies, our measure of health care utilization is whether an individ to due burden financial households’ (ii) and utilization care health (i) outcomes: of types Wetwo have Outcome variables time ofpolicychange. the by years 18 under or 18 over we they whether ascertain difficult to was it as fromanalysis excluded also were place, took exemption fee user the when policy.years the 18 of aged Moreover,tion subjects implementa- the after or before happened illness their whether identify to possible not was it as 2007), We excluded the observations interviewed within 4–weeks after the policy implementation date (28 May aged 0–18yearsold,andon1April2008,Jamaicaremoved userfeesforadults. children for fee user removedWest Jamaica the 2007, of May Hospital 28 University on the Indies: for except facilities health public all in fees user abolished Jamaica facilities, health public in user–fees of 23–years After began revising its public health sector fee structure. User fees were removed in 1975 and reintroduced in 1984. past five decades, user fees have been abolished and/or altered eight times: In 1968, Jamaica’s health authorities In Jamaica, adjustments to user fees is nothing new, as this practice dates back to the 1960s ( tor, exceptattheUniversityHospitalofWest. Elections, the JLP administration fulfilled its campaign promise by removing user fees in the public health sec General 2007 the won party JLP the sector.After health public the in patients all for fees user remove to ise the General Election in Jamaica in September, 2007, the Jamaica Labour Party (JLP) made the campaign prom During adults. to consideredno–user–fees and extending years children 0–18 for aged policy no–user–fee the introduced has it government, in was (PNP) People’sParty the National When promise: campaign the of tice paigns to seek for votes. The removal of user fees between 2007 and 2008 in public health facilities was a prac cam in services care health cost lower Historically,and better of promise used Jamaica’shave parties political Background informationonJamaica’s user–fee–removal policy 25,
29]. According to the JSLC, health professionals include doctors, nurses, pharmacists, 26], Jamaica’s experience can provide evidence for countries aimed at applying user– 288 www.jogh.org • doi:10.7189/jogh.07.010502 Table 5). Over the ------www.jogh.org • doi:10.7189/jogh.07.010502 of 28May2007.Thistestcouldexcludethepossibilitythat theresults are drivenbyseasonalfactors. instead 2006, May 28 in implemented was policy user–fee–removal the that assumes check robustness ities, opportunity cost of visiting health facilities, etc., it should also largely reflect on adults. The second driven by variables, unobserved such as changes in health system capacities, distance to the health facil the first, we assume the removal of user fees in 2007 was targeted at adults over 18. If our estimates were We conduct two robustness checks to ensure the results are not driven by unobservable confounders. In Logit regression foritsanalysis. and (OLS) squares least ordinary both use we variable, dependent binary a is utilization care health As lows ustoidentifytheeffects ofpolicychangeonchildren from different wealth levels. poverty.in al not childrenstratification vs Such poverty in children by analysis the stratify we burden, household’s and financial utilization care health of equity the policy’son the impact Tocapture further trend segment. inthepost–intervention and subscript year 2004. year the represent to 4 use we example, For 2000. from years of number the indicating variable, time the is expenditure betweenpatientsfrom householdsinpovertyandthosenotpoverty. health catastrophic encountering of likelihood the differencein as defined was burden financial in Gap poverty.in not children and poverty in children between rates visiting professional health in difference “People in poverty” was defined as those in the lowest wealth quintile. Utilization gap was defined as the sumption [22]. the household’s out–of–pocket health expenditure was larger than 40% of the household’s non–food con of share the when catastrophic be to considered was expenditure Healthcare insurance. by covered not ters, public/private hospitals, and costs of medicines purchased from public/private sources, which were survey [31]. Out–of–pocket health expenditure was defined as expenditures at public/private health cen hold’s non–food consumption if the individual experienced a health problem in the 4–weeks prior to the house the of a share as expenditures health out–of–pocket as burden financial households’ Wedefine and 2012. These characteristics remain stable over years. For example, the mean age of the respondents remainthe characteristics of These age 2012. mean and the example, For years. over stable Table 1 summarizes the individual–level and household–level characteristics in 2004, 2006, 2008, 2010 r sis aswellthedecisiontosubmitforpublication. data in the study and takes responsibility for the integrity of the data and the accuracy of the data analy Derekthe of year1996–2012via Gordon Databank. Thecorresponding the all author hadfullaccessto study. this data sourcefor JSLC funding the no to was Thereaccess obtained author corresponding The Role ofthefundingsource where to longer–termimpactofthepolicy. TheITSmodel usedinouranalysisisrepresented as: right before and right after the implementation of the user–fee–removal policy, but also analyze medium– We2007. May 28 on burdencarefinancial health and in utilization directly changes thus the can assess implemented was policy the when September,date to exact fromMay the is covering 2007 in data Our policy. this feature, ITS regressions enable examination of any significant changes after the introduction of a new ing that no other relevant changes that might impact outcomes coincide with the policy of interest. With and medium– to long–term changes in outcomes between the pre– and post–treatment segments, assum den, and equity. With a clear intervention time point, ITS regressions are able to identify both immediate bur financial utilization, care health on user–fee–removal of impact the assess to analysis WeITS used Statistical analysis change of outcome when the policy occurred. The interaction term, ESULTS Y it is the dependent variable for an individual subscript observation, Post t refers to the time, t is the time dummy for being in the post–treatment period, estimating the immediate immediate the estimating period, post–treatment the in being for dummy time the is x it are the individual–level and household–level variables at time 289 User feeremovalandequityofchildren’s healthcare trend June 2017 •Vol. 7No. 1•010502 t × i post refers to the individual case t , measures the change in t.
Trend ------t
VIEWPOINTSPaPErS PaPErSVIEWPOINTS June 2017 •Vol. 7No. 1•010502 Li etal. Table 1. education ofthedwellinginstead. imum education level of the household member instead; if still not available, we used the maximum ucation level of the spouse of the household head instead; if still not available, we used the max- tion levelofthehouseholdheadisavailable,weuseditdirectly; ifnotavailable,weusedtheed *The education level of the household is obtained through the following approach: If the educa Higher education(Grade13+) education(Grade7–13) Secondary education(Grade1–6) Primary No education Education levelofthehouseholdhead: Live intowns Live inruralareas 43.0% 20.0% Live inurbanareas 41.0% 19.0% 43.0% Number ofhouseholdmembers 23.0% 42.0% 21.0% Household’s characteristics: 42.0% 23.0% Covered byprivateorpublichealthinsurance Respondent istheheadofhousehold Male Age Individual’s characteristics: Description ofkeyvariables* of health care utilization by 97% (OR 97% by utilization care health of The implementation of user–fee–removal policy in 2007 immediately and significantly increased the odds the regression results forchildren inpovertyandchildren notinpovertyrespectively. of that age group. Columns 2 and 3, as well as columns 5 and 6, stratify the children by wealth and show and children aged less than 5 years old (columns 4–6). Column 1 and 4 shows the results for all children Table 2 presents the ITS regression results among individuals aged less than 18–years old (columns 1–3) shrank bynearlytwo–thirds between2008and2012,reached 8.7%in2012. in poverty increased their utilization at a higher rate than the children not in poverty. The utilization gap children the as However,2008) 2008. (after in long–term 21.7% to medium– the trendreversedin this between 1996 and 2006. In the short–term (2007–2008), the utilization gap further increased and reached 2000 and2001,20012002, 2009and2010.3.Sampleweightisappliedtoallavailableyears. the generationofeachdatapoint inthefigure above,wecombineddatafrom 1996and1997,1998 1999, tion numberisabovefifteen thousand,suchas2008,and 2012). To increase numbersinvolvedin theobservation numbersbetween five thousandandeightthousand.Forseveralyears,theobserva (Most yearshaveobservation byyear vary numbersintheJSLCsurveys fourweeksafterit.2.Theobservations and thesampleinterviewed beforesplit the2007sampleintotwoparts–theinterviewed theimplementationofuser-fee-removal policy Figure 1. Healthcare utilizationamongunder-18 children fellillinthepast4 weeks. 1.To generatethisfigure, we
30 70 10 80 57.0% 58.0% 26.0% 61.0% 27.0% 57.0% 24.0% 53.0% 32.0% 37.0% 18.0% 19.0% 59.0% 19.0% 52.0% 23.0% 19.0% 51.0% 29.0% 16.0% 59.0% 31.0% 60.0% 22.0% 24.0% 41.0% 40.0% 39.0% 39.0% 37.0% .%70 30 10 15.0% 11.0% 13.0% 2.0% 7.0% 4.0% 7.0% 2.0% 4.0% 3.0% 0420 0821 2012 2010 2008 2006 2004 074. 913. 40.5 39.0 39.1 41.7 40.7 . . . . 4.0 4.0 4.2 4.2 4.2 = 2.0, 95% CI 1.1 to 3.5, to 1.1 CI 95% 2.0, 290 - - P gap gradually rose from 2.6% to 16.4% utilization the 2007, in change policy the Before (CI). intervals confidence and children not in poverty – with 95% poverty in children between rates iting the difference in health professional vis – gap utilization the presents 2 Figure to 69.9%. from 2007 to 2012, ranging from 68.5% years the in high remained rates The 2007. in change policy the after 69.4% mentation of user–fee–removal policy to imple the before 54.2% from creased in largely which time, over utilization 1 Figure Healthcare utilization range from 3.0to3.2. quintiles wealth households’ average accounts for 39% to 41% of the sample, ranges from 39.0 to 41.7 years old; male = 0.018) among all children aged less less aged children all among 0.018) www.jogh.org shows children’s health care children’shealth shows • doi:10.7189/jogh.07.010502 - - - - www.jogh.org • doi:10.7189/jogh.07.010502 2007 are combinedtoyear2006prevent3.Sampleweightisappliedallavailableyears. losingobservations. beforeand 2001,20012002,20092010.2.Subjectsunder18yearsoldin2007interviewed 28May generation ofeachdatapointinthefigure above,wecombineddatafrom 1996and1997,19981999,2000 number isabovefifteen thousand,suchas2008,and2012). To increasenumbersinvolvedinthe theobservation numbersbetweenfivethousandandeightthousand.Forseveralyears,theobservation years haveobservation under-18 byyear(most vary childrennumbersintheJSLCsurveys fellillinthepast4weeks.1.Theobservations is also observed inyears2010and 2012. is alsoobserved ly to encounter catastrophic health expenditures than the households not in poverty. Such a phenomenon long–term. In 2008, households in poverty, for the first time in the year analyzed, became no more like to medium– the in low remained and (2007–2008) short–term the in rapidly reduced gap burden cial finan The expenditures. catastrophichealth encounter to children sick with poverty in not households and poverty in households difference between the is which burden gap, financial the presents 4 Figure encountered catastrophic healthexpenditure. childrenwith households of 0.6% only 2012, 18–years In than 2012. less and aged 2007 between cline centage of households with sick children suffering from catastrophic health expenditure continued to de per The 2007. in change policy the after 2.0% to 3.1% from reduced immediately expenditure health sufferingfromchildrencatastrophic sick with households Wependiture.proportionof the that observe Figure 3 presents the percentage of households with sick children suffering from catastrophic health ex Healthcare expenditures care utilization increased by 354% (OR health of odds the 4, column in shown As years. 5 under childrenresultsareaged for the 4–6 Columns F–test in columns 2 and 3 rejected the null hypothesis that two models are the same (F icy change. There is no significant change to the health care utilization among children in poverty. A joint of seeking for health care when fell ill by 82% (OR than 18 years. The stratified regressions show that children not in poverty significantly increased the odds Figure 2. data from 1996to2012(seeTable Document). S2andS3inOnlineSupplementary Document Supplementary Online The results from the OLS regressions are consistent very with the OLS regression results (see Table S1 in in columns5and6rejected thenullhypothesisthattwomodelsare thesame(F “post” are large for both wealth groups, yet the effects are insignificant. A joint F–test on the results shown of magnitudes the that show 6 and 5 column stratified in The results years. 5 than childrenless all aged The difference inhealthcare utilizationbetweenchildren inpovertyandchildren notinpoverty, among ). These results are also consistent with the regression results with with results regression the with consistent also are results These ). = 4.5, 95% CI 1.0 to 21.2, 291
= 1.8, 95% CI 1.1 to 3.0, User feeremovalandequityofchildren’s healthcare P = 0.054) after the policy change among June 2017 •Vol. 7No. 1•010502 P = 0.005) following the pol- = 586, P = 135, < 0.001). P < 0.001). - - - - -
VIEWPOINTSPaPErS PaPErSVIEWPOINTS Table 2. children agedlessthan5years(Logitregression, presented inoddsratioand95%CI)* June 2017 •Vol. 7No. 1•010502 level ofthehouseholdmemberinstead. education maximum the use we available, not still if instead; head household the of spouse the of level education the use we available, not if directly; ‖ §Significance atthe10%level. ‡Significance atthe5%level. †Significance atthe1%level. one samplingregion. Therobust standard errors are reported inparentheses. TwoPSUs. the above level one groupedinto wereis PSUs which level, region sampling at clusteredStandarderrorsare dwellings. of selection a stage *The design of JSLC is a two–stage stratified random sampling design, with the first stage a selection of Primary Sampling Units (PSUs), and the second OR –oddsratio,CIconfidenceinterval Li etal. otted09 08–.2 .8(.314)09 08–.2 .5(.710)08 05–.2 0.88(0.73–1.05) 0.87(0.54–1.42) 0.85(0.67–1.06) 0.94(0.87–1.02) 1.08(0.83–1.40) 0.95(0.89–1.02) Post×trend Post Trend Age Male health insurance Enrolled inprivate health insurance Enrolled inpublic members only Household size, Richest Richer Middle Poorer Wealth (thepoorest wealthquintileisthereference group): Town Urban Place ofresidence (“rural”isthereference group): (Grade 1–6) education Primary Education leveloftheheadhousehold(“noeducation”isreference group)||: (Grade 7–13) education Secondary (Grade 13+) Higher education N Cons. The education level of the household is obtained through the following approach: If the education level of the household head is available, we use it ITS regression ontheimpactofuser–fee–removal policyonhealthcare utilizationamongchildren lessthan18–yearsand .9(.011) . 07–.2 .3(.112) .6(.514)1(.714)1.16(0.96–1.42) 1(0.67–1.48) 1.16(0.95–1.42) 1.82(1.10–3.00)† 1.13(1.01–1.27)† 1.47(0.23–9.45) 0.9(0.73–1.12) 1.97 (1.12–3.46)† 1.09 (1.00–1.18)† .5(.409) .8(.010)09 09–.7‡08 07–.0‡08 07–.5‡0.84(0.79–0.89)‡ 0.84(0.73–0.95)‡ 0.84(0.79–0.90)‡ 0.95(0.93–0.97)‡ 0.98(0.90–1.06) 0.95 (0.94–0.97)§ .0(.824) .5(.943)18 11–.3†11 07–.2 .1(.205) 1.48(0.81–2.69) 0.11(0.02–0.56)‡ 1.11(0.76–1.62) 1.82(1.13–2.93)† 1.45(0.49–4.31) 1.70 (1.18–2.44)§ .7(.310) .5(.510)09 08–.1‡09 09–.5 .6(.911)0.95(0.87–1.03) 0.96(0.79–1.17) 0.97(0.90–1.05) 0.95(0.89–1.01)‡ 0.95(0.85–1.07) 0.97 (0.93–1.01)‡ 1.72 (1.17–2.55)§ 1.90 (1.34–2.69)§ 1.55 (1.27–1.90)§ .1(.807) .5(.404) 06–.4 .5 04–.6 .7(.704) 1.19(0.75–1.88) 0.17(0.07–0.42)§ 0.55§ (0.40–0.76) 1(0.65–1.54) 0.25(0.14–0.45)§ 0.61 (0.48–0.76)§ .6(.509) .4(.205) .5(.117)07 05–.2 .6(.907) 1.12(0.63–1.98) 0.36(0.19–0.70)§ 0.76(0.51–1.12) 0.95 (0.51–1.75) 0.34(0.22–0.53)§ 0.66 (0.45–0.97)§ .6(.608) .4(.910) .6(.114)06 04–.0 .2(.008) 0.97(0.51–1.84) 0.42(0.20–0.87)† 0.67(0.40–1.10) 0.76 (0.41–1.40) 0.44(0.19–1.02)‡ 0.56 (0.36–0.89)† .5(.210)09(.215)09 07–.1 .5(.412)07 03–.7 1.04(0.78–1.39) 0.78(0.39–1.57) 0.95(0.74–1.23) 0.98(0.79–1.21) 0.9(0.52–1.55) 0.95 (0.82–1.09) .1(.344)40 17–.4‡16 06–.6 .3(.865) 31 37–59) 1.65(0.50–5.45) 13.16(3.77–45.93)‡ 2.53(0.98–6.52)‡ 1.67(0.63–4.46) 4.01(1.71–9.44)‡ 1.91 (0.83–4.43) 1.18 (0.90–1.55) .6(.517)09 05–.0 .3(.217)10 05–.7 .2(.821)1.29(0.75–2.21) 1.14(0.70–1.83) 0.62(0.18–2.12) 1.4(0.49–3.99) 1.04(0.58–1.87) 1.13(0.72–1.78) 1.11(0.70–1.76) 0.92(0.50–1.70) 1.29(0.79–2.11) 1.01(0.44–2.32) 1.06 (0.65–1.71) 1.18 (0.77–1.82) (OR, 95%CI) 1 (0.56–1.77) Overall Overall 1931 (1)
5.78 (1.18–28.39)† u (OR, 95%CI) nder In poverty In poverty 18 441 (2) years
old
.2(.817)08 01–.7 .4(.28.8 0.95(0.23–3.87) 5.94(0.42–84.18) 0.83(0.19–3.57) 0.92 (0.48–1.74) Not in poverty Not inpoverty (OR, 95%CI) 1488 (3) 292 4.54 (0.98–21.16)‡ 2.33 (1.10–4.96)† 2.11 (1.16–3.81)† 1.75 (1.10–2.78)† 1.32 (0.75–2.33) (OR, 95%CI) Overall Overall 959 (4) available years. and 2010.3.Sampleweightisapplied toall and 1999,20002001,2001 and2002,2009 we combineddatafrom 1996and1997,1998 generation ofeachdatapointin thefigure above, increase numbers involvedinthe theobservation fifteen thousand,suchas2008, and2012). To numberisabove several years,theobservation between fivethousandandeightthousand.For numbers by year(mostyearshaveobservation vary numbersintheJSLCsurveys observations fourweeksafterit.2.The the sampleinterviewed implementation ofuser-fee-removal policyand beforetwo parts–thesampleinterviewed the generate thisfigure, wesplitthe2007sample into if thechildren fellillinthepast4weeks.1.To expenditure inthe4weekspreceding thesurvey under-18 children suffered catastrophic health Figure 3.
The proportion ofhouseholdswith .7(.4178)2.93(0.70–12.20) 7.17 (0.44–117.88) (OR, 95%CI) u www.jogh.org In poverty In poverty nder 237 (5) 5 years • doi:10.7189/jogh.07.010502
old
Not in poverty Not inpoverty (OR, 95%CI) 722 (6) www.jogh.org • doi:10.7189/jogh.07.010502 mentary Document). mentary sistent with the regression results using data from year 1996 to 2012 (see Table S4 in hypothesis that the models are the same (F = 194, null the rejected 5, and 4 columns as well as 3, and 2 columns in shown results the on tests F Joint “post”. of magnitudes insignificant yet negative, show 6 and 5 columns stratified in The results years. years. May 2007are combinedtoyear2006prevent3.Sampleweightisappliedallavailable losingobservations. before2000 and2001,20012002,20092010.2.Subjectsunder18yearsoldin2007interviewed 28 the generationofeachdatapointinfigure above,wecombineddatafrom 1996and1997,19981999, tion numberisabovefifteen thousand,suchas2008,and2012). To increasenumbersinvolvedin theobservation numbersbetweenfivethousandandeightthousand.Forseveralyears,theobserva (Most yearshaveobservation byyear vary and householdsnotinpovertywithsickchildren.numberstheJSLCsurveys 1.Theobservations Tablein S5 checks: First, we assume the removal of user fees in 2007 was targeted at adults robustness aged two more conducted than we 18–years. results, our drive not do confounders unobservable that sureTo make Robustness check 1, to –15 CI (95% points age out–of–pocket health care expenditure in household’s non–food consumption reduced by 7.1 percent- of share the 4, column in shown As years. 5 than less aged children for results the are 4–6 Columns children notinpoverty. to –2, that the policy change reduced the financial burden significantly by 12.1 percentage points (95% CI –22 age points (95% CI –11 to –1, The results show that the user–fee–removal policy significantly reduced financial burden by 6.2 percent cover children aged less than18–years and the last three columns refer to children aged less than 5 years. Table3 Figure 4. aged more than18–years intermsofhealthcare utilizationandfinancial burden. cally significant, indicating that the policy change in 2007 did not have any notable impact on the adults we can see that the coefficients on “post” and “post×trend” are neither with large magnitudes nor statisti P = shows the ITS regression results on the household’s financial burden. The first three columns columns three first The household’s burden. the financial on results regression ITS the shows Difference inprobability ofexperiencingcatastrophic healthexpenditures betweenhouseholdsinpoverty 0.02) among children in poverty and 5 percentage points (95% CI –12 to 2, Online Supplementary Document Supplementary Online P = 0.075) after the policy change among all children aged less than 5 than less aged children all P among change policy = the after 0.075) = 0.02) among children under 18–years. The stratified regressions show 293 presents the regression results of the test. As expected, expected, As test. the of resultsregression the presents P < 0.001; F = 167, User feeremovalandequityofchildren’s healthcare P < 0.001). These results are con- June 2017 •Vol. 7No. 1•010502 P Online Supple- = 0.133) among 0.133) among - - -
VIEWPOINTSPaPErS PaPErSVIEWPOINTS June 2017 •Vol. 7No. 1•010502 Table 3. Li etal. non–food consumption* ¶r level ofthehouseholdmemberinstead;ifstillnotavailable,weusemaximumeducationdwellinginstead. education maximum the use we available, not still if instead; head household the of spouse the of level education the use we available, not directly;if ‖ §Represents significanceatthe1%level. ‡Represents significanceatthe10%level. †Represents significanceatthe5%level. sampling region. The robust standard errors are reported in parentheses. We excluded the top 1% of individuals with the highest health care cost (outliers). TwoPSUs. the above level one is which Standardregionlevel, dwellings. errorssampling of areclusteredat selection a stage weregroupedone PSUs into second the and (PSUs), Units Sampling Primary of a selection stage first the with design, sampling random stratified two–stage a is JSLC of design *The Richer ihreuain(rd 3)–.0 004‡006(.1)–.1 008 008(.0)–.1 005 –0.009(0.009) –0.011(0.015) –0.008(0.006) 0.006(0.009) –0.045(0.020)† –0.013(0.008) –0.01(0.013) 0.006(0.012) r 0.000(0.008) –0.008(0.004)‡ cons –0.023(0.014) Higher education(Grade13+) –0.005(0.006) education Secondary (Grade 7–13) education(Grade1–6) Primary –0.008 Education leveloftheheadhousehold(“noeducation”isreference group)||: –0.002(0.002) Town –0.006§(0.001) –0.007(0.001)§ Urban –0.002(0.002) Place ofresidence (“rural”isthereference group): –0.006(0.001)§ Household size,membersonly Richest N Middle Poorer Wealth (thepoorest wealthquintileisthereference group)†: insurance Enrolled inpublichealth insurance Enrolled inprivatehealth Head ofthehousehold Male Age (Post×trend) Trend changeafteruser–fee–removal policy: moval policy(post) Level changeafteruser–fee–re- Trend 2 2 ¶ The education level of the household is obtained through the following approach: If the education level of the household head is available, we use it represents theadjustedRsquare. ITS regressions onimpactofuser–fee–removal policyonout–of–pockethealth care expenditure asashare ofhousehold’s [ of user fees could effectively promote utilization because it removes financial barrier to access health care elimination that elsewhere studies earlier with consistent is finding This long–term. to medium– the in health care utilization immediately after the introduction of the policy and the utilization remained high Figure 1 shows that the implementation of user–fee–removal policy in Jamaica led to increased children’s DISCUSSION 2007. Tables S6, S7 and S8 in May 28 of instead 2006, May 28 on implemented was policy user–fee–removal the assume we Second, tistically significant,suggestingthe robustness of ourfindings. native starting date. None of the coefficients on “post” and “post*trend” are with large magnitudes or sta 1 – 10]. 001(001§000(.0)–.0 <.0) 003(.0) –0.004(0.002)‡ –0.003(0.001)† –0.001(<0.001)§ 0.000(0.001) –0.001 (<0.001)§ 000(.0) 001(.1) 007(.0)–.0 005 004(.1)–0.005 (0.009) –0.024(0.015) –0.009(0.005) –0.007(0.007) –0.008 –0.021(0.011)‡ –0.010 (0.004)† 0.005(0.007) –0.004(0.003) –0.008(0.003)§ –0.004 (0.004) –0.007 (0.002)§ –0.011 (0.006)‡ 0.032 (0.026) –0.021 (0.006)§ –0.017 (0.009)‡ –0.025 (0.010)† –0.022 –0.022 –0.025(0.010)† –0.017 (0.009)‡ –0.021(0.006)§ 0.032(0.026) –0.011 (0.006)‡ –0.057(0.040) –0.091(0.085) –0.071(0.036)‡ –0.051(0.031) –0.121(0.045)† –0.062 (0.023)† .2 002§013 006 .3 004§012 000 .9 005 0.135(0.049)† 0.099(0.085) 0.132§(0.040) 0.132(0.034)§ 0.133†(0.056) 0.128 (0.022)§ –0.006(0.004) –0.016(0.020) –0.007(0.005) –0.003(0.004) –0.015(0.012) –0.005 (0.004) 001(.0)–.1 008†–.1 001 007(.1)–.0 004 –0.015(0.014) 0.008 (0.004)‡ –0.005(0.014) –0.007 (0.011) –0.014(0.011) –0.015(0.008)† –0.008(0.008) –0.011 (0.008) –0.001(0.016) –0.006(0.007) –0.004(0.003) –0.006(0.009) –0.004 (0.003) 0(0.008) 0.001(0.015) –0.001(0.007) –0.002(0.006) –0.007(0.007) –0.003 (0.004) 0.003 (0.004) 0.002(0.008) 0.008(0.015) 0.100(0.008)§ 0.005(0.007) –0.073(0.034)‡ 0.015 (0.066) 0.004(0.006) 0.013(0.011) 0.006 (0.004)
vrl npvryNti oet vrl npvryNotinpoverty Inpoverty Overall Notinpoverty Inpoverty Overall (0.009) (0.006) –0.008 0.076 0.004 1921 1) a ged Online Supplementary Online Document Supplementary
less 0.062
439 than 2) 18 years 294
1482 0.1 3) 0.004 (0.005) 0.01 (0.007) (0.012) (0.006) –0.009 .9 .7 0.132 0.076 0.094 0.003 951 presents the results of using the alter 4) www.jogh.org a ged
less • doi:10.7189/jogh.07.010502 234
than 5) 5 years –0.003 (0.001)† (0.003)† (0.002)§ (0.009)† 717 6) - - www.jogh.org • doi:10.7189/jogh.07.010502 The OLS regressions in Tableregressionsin OLS The in S1 Table 5. 2 Figure Better healthcare accessisanessentialfactortosavetheselives[ [ antibiotics life–saving potentially with treated not were pneumonia suspected with boys Jamaican able and curable, for example, the 2005 MICS survey showed that 35% of Jamaican girls and 60% of the points among children aged less than 5 years. In fact, a large proportion of children’s deaths are prevent percentage 32.5 and 18–years than less aged children among points percentage 15.8 by increased tion take bothshort–termandthelong–termeffects into considerationwhendesigninguser–feepolicies. should policymakers Hence, time. over change may policies of effects the that suggest also results Our UHC. achieve to strategy potential a of part as considered be should and feasible is fees user removing nancial burden which households may confront [ confront may households which burden nancial fi the reducing and services health of utilization equitable the promoting by children for care health to where and strongly confirm the effectiveness of user–fee–removal policies in improving the equal access else undertaken studies earlier with line in however,are resultslimitations, our these Notwithstanding out thepossibilityofconfounders. sets of robustness checks were conducted, this study is and still could observational not completely rule two although Fourth, wereconducted. surveys the months the in sick be to likely less drenaremoreor may generate biased estimates if the non–food consumption is not evenly divided over months or if chil recallWeperiod. reflect method to This consumption period. non–food 4–week yearly the the adjusted 4–week yearly,a collected with not arebut data expenditure care health Third, possible. be would ysis health outcomes. If more comprehensive data with larger sample size were available, more detailed anal utilization among infants and can neither draw any conclusion on the link between the policy change and care health the on analysis an conduct to able not are we size, sample limited the to due clear.Second, Whether this happened in the case of Jamaica and the extent to which it changed people’sresources. health behavior of is un- wastage to leading them, overuse to tend may people inexpensive, or free become services health When hazard. moral or demand unmet of release the to due was utilization care health The study has four potential limitations. First, we cannot conclusively determine whether the increase in improve [34– accesstohealthservices help and patients, for barriers financial of part reduce exemptions user–fee that demonstrating where, to relieve the financial burden of the poor. Our results are consistent with earlier studies undertaken else appear to benefit more than the children not in poverty, which indicates that the policy had a larger effect than 18–years and 7.1 percentage points among children aged less than 5 years. The children in poverty less aged children among points percentage 6.2 by consumption non–food households’ in expenditure care health out–of–pocket of share the reduced significantly policy user–fee–removal the that Wefind sults. producedifferentre could studies of lengths various because period, study the to attention special pay should one analysis, equity conducting while that suggests finding This 2012. and 2008 between poor non– than pace faster a at increased poverty in children by utilization the as rapidly decreased gap tion short–term. While in the medium– to long–term (after 2008), the in behavior their changing in quicker be to tend and policies new receivingabout at information ter to children in poverty. One potential explanation for this observation is that wealthier households are bet gap enlarged due to the faster increase in health care utilization among children not in poverty compared long–term results appear to have different equity impact: In the short–term (2007–2008), the utilization JPL –JamaicaLabourParty, PNP–People’s NationalParty Revised userfees d Removal ofuserfees–allpublicpatients Re–introduction ofuserfees Removal ofuserfees Removal ofuserfees(children aged0–18yearsold) Adjustment ofuserfees(upwards) Adjustment ofuserfees(upwards) Adjustment ofuserfees(upwards) eta Ils User feeschangesinJamaica,1968–2008[21] , combined with the ITS results in results ITS the with combined , Online Supplementary Document Supplementary Online 37]. 295 Table2 1 – April 2008–Present ]. An important implication of our results is that that is results our of implication important An 10]. , implies that the short–term and the medium– to to medium– the and short–term the that implies , May 2007 1968 1984 1975 2005 1999 1993 y ear Figure 2 further indicates that the utiliza User feeremovalandequityofchildren’s healthcare 33]. June 2017 •Vol. 7No. 1•010502 suggest that health care utiliza care health that suggest I n
government PNP PNP PNP PNP PNP JLP JLP JLP (Jlp or pnp) ]. 32]. ------
VIEWPOINTSPaPErS PaPErSVIEWPOINTS June 2017 •Vol. 7No. 1•010502 Li etal. rEFEr ENCES 4 3 9 8 7 6 5 2 1 19 18 17 16 15 14 13 12 11 10 20 from Uganda.HealthPolicyPlan.2005;20:100-8.Medline:15746218 Nabyonga J, Desmet M, Karamagi H, Kadama PY, Omaswa FG, Walker O. Abolition of cost-sharing is pro-poor: evidence dence. AmJPublicHealthRes.2013;1:196-202.doi:10.12691/ajphr-1-8-1 evi the of analysis criticical a countries: African sub-Saharan in services health on fees user of impacts The EU. John S0140-6736(09)60258-0 miol Community Health. J Epidemiol Community Health. 2013;67:751-7. Ridde V, Haddad S, Heinmüller R. Improving equity by removing healthcare fees for children in Burkina Faso. J Epide doi:10.1016/j.socscimed.2005.07.004 2006;62:866-76. Med. Sci Soc Uganda. in expenditures health catastrophic and utilization fees: Xu K, Evans DB, Kadama P, Nabyonga J, Ogwal PO, Nabukhonzo P, et al. Understanding the impact of eliminating user Zambia. SocSciMed.2010;71:743-50.Medline:20542363 F,Masiye from care:experience rural health in abolition fee Fromuser McIntyreD. targetedto BM, exemptions Chitah Organ. 2004;82:187-95.Medline:15112007 Burnham GM, Pariyo G, Galiwango E, Wabwire-Mangen F. Discontinuation of cost sharing in Uganda. Bull World Health line:19909514 vices among poor and rural residents in Uganda: are reforms benefitting the poor? Int J Equity Health. 2009;8:39. Pariyo GW, Ekirapa-Kiracho E, Okui O, Rahman MH, Peterson S, Bishai DM, et al. Changes in utilization of health ser Yates R. Universal health care and the removal of user fees. Lancet. 2009;373:2078-81. Database SystRev. 2011;(4):CD009094.Medline:21491414 Lagarde M, Palmer N. The impact of user fees on access to health services in low-and middle-income countries. Cochrane 202080 World Bank.Indicators.Available: http://data.worldbank.org/indicator work oftheprofessional nurse[thesis].Wellington: Victoria University ofWellington; 2013. the and provided care access, on impact system: health public Jamaican the in fees user of abolition The A. Campbell heapol/17.3.281 2002;17:281-7. Plan. Policy Health Ecuador. in services health reproductive Bratt JH, Weaver MA, Foreit J, De Vargas T, Janowitz B. The impact of price changes on demand for family planning and Indian MedJ.2012;61:168-73.Medline:23155965 De La Haye W, Alexis S. The impact of a no-user-fee policy on the quality of patient care/service delivery in Jamaica. Westhttp://www.capricaribbean.com/documents/no-user-fee-policy-public-hospitals-jamaica Available:2013. Jamaica. in hospitals public in policy fee no-user the of survey A survey? a free or Fee TCPR. Institute 2017. Available:2000. world. changing a in protection social and security Income 2000: Report Office. Labour Labour World International support theprocess. HealthPolicyPlan.2011;26Suppl2:ii104-117. to experience international from learning fees: user Removing E. Araujo A, Schmidt G, Cometto N, Brikci B, McPake en/. Accessed:14March 2017. World Health Organization. World Health Assembly. Available: care—Jamaica. SocEconStud.2010;59:123-52. W,preventivehealth Gordon-StrachanBailey for G, fees user of impact The D. Alleyne S, Lalta J, Barnett A, Henry-Lee Plan. 2011;26Suppl2:ii41-51.Medline:22027918 Policy Health paradox. Uganda the F,fees: Mugisha user JN, Oremof Nabyonga Abolition B. Criel J, Macq C, Kirunga Accessed: 12July2016. bank.org/curated/en/408381468044133381/Jamaicas-effort-in-improving-universal-access-within-fiscal-constraints/ Chao S. Jamaica’s effort in improving universal access within fiscal constraints. 2013. Available: coi_disclosure.pdf (available upon request from the corresponding author), and declare no conflict of interest. interests: Competing thors haveread thepaperandagreed tothesubmission. study.this au of processAll all in supervision performed has Atun Rifat manuscript. the of revisions critical to contributed Bärnighausen Till and Bourne, Paul Fink, Guenther Li. Mingqiang with together manuscript Databank and had full access to all the data in the study. Zhihui Li conducted all data analysis and drafted the GordonDerek the via 1996–2012 year of data JSLC the to access obtained Li Zhihui contribution: Authors’ Funding: None. of PublicHealth(IRB15–3750). T.H.Harvard the of Chan (IRB) Board Review Institutional the by approved was study The approval: Ethics per, andtoDrMargaret E.Krukforhelpfulcommentsonanearlierdraftofthemanuscript. Acknowledgements: The authors are thankful to Dr Jessica Cohen for advice on the methodology of this pa doi:10.1186/1475-9276-8-39 http://www.ilo.org/public/english/standards/relm/gb/docs/gb279/pdf/esp-7.pdf
The authors have completed the Unified Competing Interest form at www.icmje.org/ at form Interest Competing Unified the completed have authors The 296 doi:10.1093/heapol/czr065 doi:10.1016/j.socscimed.2010.04.029 http://www.who.int/mediacentre/events/governance/wha/ Medline:22027915 . Accessed:12July2016. doi:10.1093/heapol/czi012 Medline:23776054 Medline:12135994 www.jogh.org Medline:19362359 . Accessed: 11 November 2015. doi:10.1093/heapol/czr064 • doi:10.7189/jogh.07.010502 http://documents.world- doi:10.1136/jech-2012- . Accessed: 13 January January 13 Accessed: . Medline:16139936
doi:10.1093/ doi:10.1016/ Med------. www.jogh.org rEFEr ENCES • doi:10.7189/jogh.07.010502 23 22 21 24 31 30 33 32 29 28 34 26 25 38 37 36 27 35 coverage inLatinAmerica.Lancet.2015;385:1230-47.Medline:25458725 Atun R, De Andrade LOM, Almeida G, Cotlear D, Dmytraczenko T, Frenz P, et al. Health-system reform and universal health able: http://www.who.int/health_financing/pb_2.pdf 2017. . Accessed:13January World Health Organization. Designing health financing systems to reduce catastrophic health expenditure. 2005. Avail docman&task=doc_view&gid=19100&Itemid=270. Accessed:4May2017. Available: 2013. Jamaica. in Coverage Universal M. Coombs care inLatinAmerica.Lancet.2015;385:1248-59.Medline:25458715 F,Knaul O, Gómez-Dantés D, Cotlear Overcominghealth segregational. in social et O, Cetrángolo BarretoIC, R, Atun Nouna District, Burkina Faso. Bull World Health Organ. 2006;84:21-7. from study a society: low-income a in care health for expenditure household Catastrophic S. TT,Flessa Su B, Kouyaté ofLivingConditions.Kingston:PIOJ&STATIN:Planning InstituteofJamaica.JamaicaSurvey 2013. ity from childhood pneumonia by 2015. PLoS Med. 2011;8:e1001099. Med. PLoS 2015. by pneumonia childhood from ity Rudan I, El Arifeen S, Bhutta ZA, Black RE, Brooks A, Chan KY, et al. Setting research priorities to reduce global mortal org/americas/jamaica/child_and_maternal_health_in_jamaica/. Accessed:27March 2016. Available:2011. Jamaica. in health maternal and Child Health. of Ministry Jamaica systems andpopulations.IntJEquityHealth.2013;12:18. Levesque J-F, Harris MF, Russell G. Patient-centred access to health care: conceptualising access at the interface of health domized trial. 2014. Available: Tanzania:in transfers cash conditional Community-based N. Reese K, Kosec S, Hausladen D, Evans results from ran a S1020-49892013000200002 enue. HealthPolicyPlan.1995;10:164-70.Medline:10143454 Mwabu G, Mwanzia J, Liambila W. User charges health facilities in Kenya: effect in government on attendance and rev pmed.1001099 systematic review. Trop MedIntHealth.2010;15:508-19.Medline:20345556 Rutherford ME, Mulholland K, Hill PC. How access to health care relates to under-five mortality in sub-Saharan Africa: 2005;331:747-9. Medline:16195292 BMJ. model. simulation fees: user removing of mortality child on Impact A. Taylor R, Keith SS, Morris C, James Econ. 2005;14:55-91.doi:10.1093/jae/ejh034 Afr J Uganda. from evidence fees: user health of abolition the P.of Mpuga impact K, welfare Deininger and Economic 1998. HealthEcon.2003;12:921-34.Medline:14601155 Wagstaff A, Doorslaer EV. Catastrophe and impoverishment in paying for health care: with applications to Vietnam 1993- Medline:12112492 2002;11:431-46. Econ. Health Indonesia. in care medical for options management risk Social N. Prescott M, Pradhan 2013;33:83-9. Publica. Salud Panam Rev countries. Caribbean and American Latin Almeida G, Sarti FM. Measuring evolution of income-related inequalities in health and health care utilization in selected Plan. 2002;17:304-13.Medline:12135997 Meuwissen LE. Problems of cost recovery implementation in district health care: a case study from Niger. Health Policy doi:10.1002/hec.689 https://openknowledge.worldbank.org/handle/10986/17220 doi:10.1136/bmj.331.7519.747 doi:10.1093/heapol/17.3.304 297
doi:10.1002/hec.776 Medline:23496984 doi:10.1093/heapol/10.2.164 http://www2.paho.org/hq/index.php?option=com_ Medline:16501711 doi:10.1016/S0140-6736(14)61647-0 User feeremovalandequityofchildren’s healthcare doi:10.1016/S0140-6736(14)61646-9 Medline:21980266 doi:10.1111/j.1365-3156.2010.02497.x doi:10.1186/1475-9276-12-18 June 2017 •Vol. 7No. 1•010502 http://www.commonwealthhealth. Medline:23525337
doi:10.2471/BLT.05.023739 . Accessed. 13 January 2017.
doi:10.1371/journal. doi:10.1590/ - - - -
VIEWPOINTSPaPErS PaPErSVIEWPOINTS June 2017 •Vol. 7No. 1•010601 social autopsystudy Eastern r environmental factorsofchildmortalityin Sociodemographic, behavioral,and material. The onlineversionofthisarticlecontainssupplementary Electronic material: supplementary Félicitée Nguefack Paul–roger Libite [email protected] USa Baltimore, MD21205 E8610 615 N.WolfeSt. Johns Hopkins Bloomberg School of Public Health Department ofInternationalHealth a Correspondence to: 4 3 2 1 alain KKoffi MinistryofHealth,Yaoundé,Cameroon University ofYaoundéI,Department National InstituteofStatistics,Yaoundé, Department ofInternationalHealth,Johns Pediatrics, Yaoundé,Cameroon Cameroon Baltimore, USa Hopkins SchoolofHygieneandPublicHealth, lain Koffi
1 , romainSWounang
2 , HenryDKalter 3 egion of Cameroon: results from a egion ofCameroon:resultsfroma , SeidouMoluh
2 1 4 , ,
households (96.1%) used firewood for cooking, and 79.2% (n the of all respectively.Almost sanitation, and water drinking of just 26.8% and 11.2% lived in households with an improved source Findings Of the 635 deceased children with a completed interview, children model. alongthePathwaytoSurvival the for care–seeking and recognition illness and care, child normal of indicators preventive key of coverage the including deaths, old month 1–59 the of characteristics household and ciodemographic so important identify to sought study present The Cameroon. of mendouka, and Abong–Mbang health districts in the Eastern Region of under–five years old deaths from 2007 to 2010 in Doume, Nguele ed in 2012 to measure the biological causes and social determinants Methods A retrospective verbal/social autopsy survey was conduct work againstchildrenfrom surviving thesediseases. factors system health and cultural sociodemographic, several laria, ma- and diarrhea, pneumonia, as such diseases treatable and able Background While most child deaths are caused by highly prevent (n = many result, a As care. formal seek to decision the to onset illness from days 2 of delay median a with but care, health formal seek to became fatally ill, the majority (83.7%) of caregivers sought or tried children beside them when they cooked. When 614 of the children of the 236 mothers who cooked inside their home usually had their the illchildren inthestudy area. of survival the in difference a make could that measures are cles, obsta- economic overcomethe can caregivers or parents that such living of standards improving and care–seeking, health propriate ap and timely facilitate to and illnesses childhood of signs danger decision to seek care. Increasing caregivers’ ability to recognize the to indoor smoke, and health–related behaviors such as delaying the exposure to led that customs prevailing conditions, living poor ed the deaths of 1–59–month old children in the study setting includ to contributed that factors social common most The Conclusions tion, healthcare andotherrelated costs. transporta for expenses the were system health formal the cessing ac to barriers main The severe. to moderate or mild from gressed 111) children were taken for care only after their illness pro illness their after only care for taken were children 111) 298 www.jogh.org • doi:10.7189/jogh.07.010601 global journal of = 187) 187) health ------www.jogh.org • doi:10.7189/jogh.07.010601 regions of Cameroon[ of regions the country experiencing the second highest under–five mortality at 187 deaths per 1000 live births [ [ 2015 in 1000 per deaths 88 creasedto from 138 deaths per 1000 live births to 150 deaths per 1000, has now reversed, with the rate having de- 1990s, the during rate mortality under–five the in trend upward the that estimated is it Cameroon, In [ to timelyandqualityhealthcareonchildsurvival interventions mortality.child to access contribute poor rolealso of the factors on cultural literature extensive is There Burstrom[ and Macassa Mozambique, In affordablecare. or health ity access to safe water and sanitation, cannot afford nutritious diets, and often have no access to good–qual [ outcomes health other most than deprivation material or poverty effectsof the to sensitive more being as described is survival Child of progress orachievementmisleadinglessmeaningful. assessment MDGs country–level any render which countries, within rates mortality under–5 the in ties thereexample, For agenda. development arepost–2015 transformative and dispari universal persistent However, some critics posit that the MDGs failed to address some gaps that need to be tackled in the new 75% [2 by ratio mortality maternal the reduce to 5) (MDG five goal and two–thirds by mortality child reduce to aimed that 4) (MDG four goal namely health, child and reproductive, maternal, to dedicated MDGs ers to prioritize and design evidence–based child survival interventions. ers toprioritizeanddesignevidence–basedchildsurvival [ mortality child to contribute that factors modifiable and complex the of some unveil to was study this of purpose the perspective, policy Froma region ofCameroon.in Doume,Nguelemendouka,andAbong–MbangdistrictstheEastern 2010 occurredfromto that 2007 age of months children1–59 of deaths the of data (SA) autopsy social tal and child mortality in selected, high–mortality countries such as Cameroon. This paper focuses on the demiology Reference Group’s (CHERG) effort to directly measure the causes and determinants of neona Epi- Health Child fromWHO/UNICEF–supported drawn the series a of part is presentmanuscript The appropriate choices for its situation and accelerate the achievement of improved child survival outcomes. make Cameroon help could survival, child affect that factors contextual and behavioral the with along [1 2030 by births live 1000 per deaths fewer or 25 of rate mortality under–five an achieve to order in progress There has been a recent call from the international community for Cameroon to accelerate the pace of its [ 1990 since halved than more has The United Nations recently released its 2015 report that found that the global under–five mortality rate Pathway Analysis SA questionnaire [ questionnaire SA Analysis Pathway CHERG the with death, of cause biomedical the determine to questionnaire autopsy verbal (PHMRC) Consortium Research Metrics Health Population the blended chronologically VASAquestionnaire The The description of the data collection tools and the fieldwork is available in a paper published earlier [ perioduntilthedesiredthe survey samplesizeofdeaths of660childdeathswasachieved. over time in back moving census, the to prior years four the in death such one least at with household each in death old years under–five recent most one the taking by period recall the minimize to was egy to 2010 identified by the census birth histories conducted in the last quarter of 2010. The sampling strat [ The sampling methodology of the verbal/social autopsy (VASA) study has been fully described elsewhere project. (PSI) fromInternational Services October to December 2010 for a Home–Based Management of Malaria 16 a all of census baseline a in mothers interviewed all for recorded histories birth complete the from came deaths and births on Information Abong–Mbang districtsofCameroon. that contributed to the child deaths that occurred from 2007 to 2010 in Doume, Nguelemendouka, and community,household, the identify to was study the of aim fundamental The factors system health and METHODS 11]. In brief, the study universe included 930 deaths of young children (1–59 months of age) from 2007 ]. ]. To reach that goal, an understanding of the most important biological causes of child deaths, deaths, child of causes biological important most the of understanding Toan ]. goal, that reach ], that led to its being dubbed “the forgotten region”, for use by health manag health by use forgottenfor region”,“the dubbed being its to led that 10], 3 ]. Poor households are more likely to be exposed to diseases, often lack lack often diseases, to exposed be to likely more are households Poor ]. 1 ]. This encouraging progress may be attributable to at least the two two the least at to attributable be progressmay encouraging This ]. ] to inquire about well–child and illness events leading up to a a to up leading events illness and well–child about inquire to 12] 954 households in the three districts undertaken by Population Population by undertaken districts three the in households 954 1 ]. Yet,]. regionof Eastern the remain,with disparities important 9 299 ] in the study area, known as one of the most impoverished impoverished most the of one as known area, study the in ] 5 – Factors ofchildmortalityineasternCameroon 7 ]. 4 ] concluded that behavioral and and behavioral that concluded ] June 2017 •Vol. 7No. 1•010601 11 8 ]. ]. ]. - - - - -
VIEWPOINTSPaPErS PaPErSVIEWPOINTS June 2017 •Vol. 7No. 1•010601 Koffi etal. ventions examined by the Lives Saved (LiST) tool [ shown to be efficacious and effective in promoting child survival and thus are included among the inter formed consentbeforewasconducted. theinterview by the Johns Hopkins School of Public Health’s Institutional Review Board. All respondents provided in then Research Committee, Cameroon National the VASA approved by The first Cameroonwas in study Document. of thesymptomsandtheirseverityscoringare givenin OnlineSupplementary (requiring referral to higher level formal care) or possibly severe (requiring formal health care). The listing the VA instrument but not in the IMCI, two physician authors (HDK, AKK) assigned symptoms as severe Illnesses (IMCI) severity grading for the first symptoms as observed. For the illness symptoms that were in served symptoms by using the World Health organizations’ (WHO) Integrated Management of Childhood Separate to that scoring system, we derived a symptom severity scoring system based on the caregivers ob according toNunnalycriteria[16]. scores summated the of reliability the justifying responses, consistent highly elicited scores the in items 0.90 at the onset of the fatal illness and when the decision to seek care was made. This suggested that the [ paper prior a in ness severity were constructed: normal/mild, moderate, and severe. Details of the method were provided on caregivers’ attempts at care–seeking for their child’s illness. Hence, three independent categories of ill onset illness at severity illness perceived of impact the assess to order in status mental and level activity child’sthe caregivers’reportsof on based developed was system scoring a behavior,addition, feeding In ables usedthroughout thispaperare provided inKoffi etal.[ tion (WHO), and so should be accessible to all children. The list and definitions of some operational vari curring either in the post–neonatal (1–11 months of age) period (41.1%) or second–year (26.3%) of life. at illness onset was 12 months (interquartile range IQR:7–24 months), with two–thirds of the deaths oc The sociodemographic characteristics of the deceased children are presented in fathers ofthedeceasedchildren, 8.7%grand–mothers,and6%othersrelatives. were 10.6% while mothers, were (74.7%) respondents the of two–thirds than More sample. study the in included deaths age) of months (1–59 child 660 the of (96.2%) 635 for completed were Interviews r terns encompassed by the Pathway to Survival model [ care–seekingpat recognitionand illness (c) and home; the outside and careprovidedinside child both determinants of the deceased children; (b) coverage of key interventions along the continuum of normal [ paper or The analysis of data on preventive and curative care followed the same procedures as described in a pri tion occurred from 5March to15April2012. respondent’scollec main Data the considered. respondents,were among answers responsesdiscordant with cases necessary.In if permitted were respondents eligible additional delivery.Hence, and cy terview covered the fatal illness from onset to death, including for neonatal deaths, the mother’s pregnan in The illness. the child’sfor the child of the providedknowledgeable careto most and son illness fatal during the forty days of data collection. The interviewers were trained to select as the respondent the per office supervisors by visits field two received addition in and (NIS) Institute Statistics National fromthe supervisor field one by led was team Each 2010. in PSI by conducted survey mortality the during ment involve prior their and languages, local and districts the of knowledge their on based district) per (one groupsthree into split were interviewers The languages. local the and French in conducted all practice, field of days 3 by followed netbook, the on interview the of conduct and standards ethical procedures, at least a high school education, received 10 days of in–classroom training in the VASA study background, had and languages local the of speakers native were who interviewers female twenty fieldwork, the For (CAPI)captureputer AidedPersonalInterview oftheVASA dataonanetbookcomputer. interview software application (Serpro S.A, Santiago, Chile) that was developed to enable direct, field–based Com CSProX a into inserted then were translations The Abakoum. and Onveng moon, Mpoong Baka, Maka, Mongo–Ewondo, languages– local major six to transliterated phonetically were providers, care health formal and traditional local of names the and symptoms and signs illness as such items, questionnaire to French, which is understood by the majority of persons in the study area. Only the local terms for key death. The VASA questionnaire was developed in English and, for the study in Cameroon was translated ESULTS ], and was guided by the following: (a) review of several sociodemographic and household household and sociodemographic several of review (a) following: the by guided was and 13],
]. The Cronbach’s [ The coefficients 11]. alpha 300 12] or recommended by the World Health Organiza 9 , 12 ] of the summated scores showed values of of values showed scores summated the of 15] , 14 13]. ]. All the examined interventions have been www.jogh.org Table 1. The median age • doi:10.7189/jogh.07.010601 ------Table 1. *Q1–Q3: First and third quartiles of the interquartile range (IQR). IQR –interquartile range www.jogh.org Other healthprovider orfacility Hospital Place ofbirth: Masculinity ratio Female Male Sex: Don’t know 24–59 12–23 1–11 Median ageatdeath(inmonths): Median illness 12(IQR:7–24) Median ageatillnessonset(inmonths) c nruet elhpoie rfclt 03.2 Other 20 Home En route toahealthprovider orfacility Other healthprovider orfacility Hospital Place ofdeath: Other Home On route toahealthprovider orfacility haracterIstIcs Characteristics of635deceasedchildren • doi:10.7189/jogh.07.010601 (ndy)7(IQR:3–14) duration (indays) started at0-23monthsandsatisfiedoneoftheaboveconditions . replacement feedseachday(includingmilkand solid,semi-solidandsoftfoods).****Children whosefatalillness children whosefatalillnessstartedat6-8months oldand9-23monthsold,respectively, andreceived atleastfour received, respectively, atleasttwoand threenon-liquidfeedingseachday. complementary ***Never-breastfed breast milkasfood.**Breastfed children whosefatalillnessstartedat6-8monthsoldand9-23old, months ofage(1-5months),he/shewasbeingbreastfed atthetimeoffatalillnessandwasnotgivenanythingbut months of age, only 32.1% (n 32.1% only age, of months 6–23 at began illness whose children 330 the Among breastfed. exclusively were age of months 0–5 at Figure 1. non–liquid feeds each day, while 11.5% (n always sleptunderaninsecticide–treated bednetbefore theirfatalillnessbegan. usually away from the mother when she cooked inside the house. Less than half (46.5%) of the children was he/she ie, smoke, to exposed be to not were likely children (20.8%) five in one care.About of uum Figure 2 shows some preventive home care received by the 1–59 months old children along the contin ment feedseachday. Appropriate feedingforchildren whoseillnessstartedatage0–23months.*Child’ began before 6 12 (IQR:8–24) 1 18.4 117 51.0 49.0 324 311 32.4 26.3 205 41.1 167 261 6 58.0 368 21.6 137 68.8 437 411.7 74 510.2 65 7.1 45 *p n* 1 4 3 96 ercentage 0.2 0.6 0.5 = 106) of breastfed children received the recommended complementary complementary recommended the received children breastfed of 106) only 15.5% (n (see age their for fed appropriately ing tween 0–23 months of age. Overall, just 36.3% (n be- started illnesses fatal whose children 446 the of began illness the before intake nutritional the of summary a presents 1 Figure nity wasabout10years. commu same the in continuously living been had families time caregiver’s usual health care center was 30.0 minutes. The median the to time travel median The cooking. for fuel as firewood used households the of (96.1%) majority vast The toilet). pit proved im or (flush sanitation water,and drinking of source improved electricity,as such facilities modern had households the of less an or quarter a About (68.3%). worker farmer/agricultural was ner size was 7.4 persons. The occupation most cited for the breadwin level of education (1–6 years of schooling). The average household 20 years of age. More than two–thirds (71.2%) had some primary third of the mothers (29.3%) lost their index child before reaching a About age. of years 20 before union in entered (67.7%) thirds two– interview; the of time the at man a with living or married were mothers the of ner,Approximately80% household. the and Table 2 jority (58.0%)alsodiedathome. (68.8%) of the 635 deceased children were born at home; the ma Most 96.0. of ratio male a with males, than females of deaths more slightly were There days. 7 was duration illness Median = 301 38) of non–breastfed children received at least four replace shows the characteristics of the mother, her domestic part
= 18) of the 116 children whose fatal illnesses began Factors ofchildmortalityineasternCameroon June 2017 •Vol. 7No. 1•010601 ). In more detail, detail, more In 1). Figure = 162) were be------
VIEWPOINTSPaPErS PaPErSVIEWPOINTS June 2017 •Vol. 7No. 1•010601 Table 2. (excluding poliovaccinegivenatbirth). vaccination givenatbirth;FullyImmunizedchildren received BCG,measles,andthree doseseachofDPTandpoliovaccine either from thevaccinationcard orwhenthere wasnowrittenrecord, from therespondent (mainlythemother).Polio0isPolio their motherwhenshecookedinsidethehome.**Insecticide-treated bednet.***Informationonimmunizationswasobtained RegionofCameroon,districts, inEastern from 2007-2010.*Proportion of children whowere NOTusuallybesideorcarriedby Koffi etal. Figure 2. *IQR: firstandthird quartilesoftheinterquartile range(Q1–Q3) IQR –interquartile range eintae iet ers elhfclt mn 30.0(IQR:25–60) 11.2 Median traveltimetonearest healthfacility(min) 71 Floor ofthehousemadecement 68.3 Household usesfirewood forcooking Use ofimproved sanitation(improved pitfortoilet) 10(IQR:5–20) Use ofpipedwater–in–housewatersupply 434 Household haselectricity Household size(mean) Median yearscontinuouslylivingincommunity Main breadwinner isfarmer/agricultural worker Main breadwinner –other Main breadwinner –mother Main breadwinner –father Household characteristics: Don’t know >6 1–6 0 Mean yearsofschooling(inyears): characteristics: Paternal Don’t know >6 1–6 0 Mean yearsofschooling: Don’t know 30 ormore 25–29 20–24 17–19 13–16 Mother’s medianageattimeofchilddeath(inyears): Don’t know 20–44 16–19 12–15 Median agewhenfirstmarried(years): Married orlivingwithaman m aternal
Characteristics ofthemotherandherhousehold,635childdeaths characterIstIcs Coverage alongthecontinuumofcare for1-59-montholdchilddeathsinDoume,NguelemendoukaandAbong-Mbang
314.7 96.1 93 26.8 610 21.4 170 137 7.4 (IQR:5–10) 24.4 7.9 67.7 155 50 32.0 430 29.3 37.6 203 186 239 7 1.7 6.6 (IQR:5–8) 23.0 71.2 11 4.1 146 452 1.7 26 27.1 5.3 (IQR:4–6) 18.0 11 26.9 172 17.5 114 8.8 171 111 2.4 56 29.9 24 (IQR:19–31) 42.0 12 25.7 152 214 80.2 131 18 (IQR:15–20) 509 n 1.1 p ercentage 302 Figure 3 Figure tributed to the deaths of the children are presented in The breakdowns in the Pathway to Survival that con fatal illnessbegan. dren received at least one dose of vitamin A before the Almost all (92.3%) of the 539 6–59–months–old chil doses). against polio by age one (just 52.7% had had all immunized three fully be to likely least were children Measles, ranging from 86.0% to 94.4%. The deceased The highest coverage was for Polio 0, BCG, DPT1, and these diseases before they reached their first birthday. against immunized fully were children the of 36.8% just Overall, age. of year one by diseases childhood vaccinations against each of the six major preventable (1.8%). 11 other the for sought or given was care no and (2.4%) children were said to have “died immediately,” n (95.8%, all almost = that the child was ill. Care was provided or sought for sibly severe sign or symptom when they first noticed pos- or severe a had child their that recognizeddren chil 614 the of caregivers the of (96.1%) all Nearly types ofactionstakenfortheillness. the on information provided caretakers whose dren children 12–59 months of age (n 2 Figure . This analysis included only the 614 chil 614 the only included analysis This . further shows the percentage of deceased deceased of percentage the shows further www.jogh.org 588) of the children; while 15 15 while children; the of 588) • doi:10.7189/jogh.07.010601 = 372) who received - - - - not allowtheirillnessseverityranking. formal care (N = among children forwhomcaregivers triedtoseekorsought some Don’t know. decided toseekformalcare. N/M=normal/mild,Mod=moderate,Svr=severe *CHWs–Trained CommunityHealthworker. **DK illness durationduetotheskewedvaluesforthesevariables.§Illnessseverityatonset.§§Illnessonsetandwhencaregiver Region ofCameroon, from 2007-2010.¶Medianvaluesare reported fortheageatillnessonset,delaytoformalcare, andthe www.jogh.org Figure 4. Figure 3. Illness severityrankingatonset and atdecisiontoseekcare for614YoungThe “PathwaytoSurvival” ChilddeathsinDoume,NguelemendoukaandAbong-Mbangdistricts,Eastern • doi:10.7189/jogh.07.010601 506). 8children hadmissinginformationthatdid home, and 242 of these 283 later sought or tried to seek care outside the home. In total then, 547 the (89.1%) inside care received first (48.1%) children 283 other the home; the outside care seek to was sought The first action taken for about half (51.9%, n (51.9%, half about for taken action first The 303 = 305) of the 588 children for whom care was given or or given was care whom for children 588 the of 305) ( layed by 2 days (median time) after the onset of the illness de was care formal seek to decision the well, as groups who soughtortriedtoseekonlyformalcare. Forboth those and care formal and informal both seek to tried or sought who those for both days) 1–3 (IQR: days 2 lay) from the illness onset until seeking formal care was only.care informal received(de duration median The pharmacist/drug seller, and formal care, and 33 (6.0%) a from or healer traditional a from care as such mal, pital, 106 (19.4%) received or tried to seek both infor private doctor or NGO/Government center/post or hos followings: a trained community health worker (CHW), the of one either at or by provide care ie, care, formal majority (74.6%, n the home. When care was sought outside the home, the available received, sought, or tried to seek care outside was data care–seeking whom for children 614 the of or other health facility ( facility health other or from a health worker in the community or at a hospital compared to the median of 2.5 days among those who ill, moderately or normal be to perceived were who addition, the median delay was 2 days among children Figure 4 ), regardless of whether formal care was sought Factors ofchildmortalityineasternCameroon = 408) received or tried to seek only June 2017 •Vol. 7No. 1•010601 2 (2) = 1.261; P = 0.2614). In In 0.2614). - - - -
VIEWPOINTSPaPErS PaPErSVIEWPOINTS June 2017 •Vol. 7No. 1•010601 Koffi etal. ferred(n = re only either were 224 remaining The recommendations. care home any given nor referred not were In addition, about 39% (n Among those who sought or tried to seek some formal care (N (51.5% vs38.6%)wasstatisticallysignificant( groups two the differencebetween The care. formal seek to made was decision the time the at ill verely third (38.6% or 142 out of 367) of those children who left the first provider alive being judged to be se care.morecomparesa formal slightly This than seek to to made was decision the time the severelyat ill n or (51.5%, = Out of 469 children that reached a first provider, 101 (21.5%) died at that provider. Approximately half able data. NGO or government hospital, and 2 – for which the name or type could not be identified with the avail- 10 to a health post, 22 to a private doctor or clinic, 178 to an NGO or clinic, government and 226 to an utes median travel time, IQR:15–60–minute. Thirty–one (31) went to a community health worker (CHW), provider. The remaining 469 (91.2%) children reached the first health care provider after about 30 min carehealth beforedied they because facility routeen died out, setting reachnot could or care health the the reach not did (8.8%) 45 care, formal some seek to tried caregivers whom for children 514 the Of was statisticallysignificant(t (±SD = er words, the mean illness severity score increased from 1.66 (±SD time of illness onset, became severely ill by the time their caregivers decided to seek formal care. In oth ( care formal seek to made 24.1% (122 out of 506) at thetime of illness onset to 43.5%(or 220 out of 506) whenthedecisionwas from increased ill severely be to perceivedchildren of percent the severity), illness perceived on mation a concern or problem.a concern had they carereportingthat seek did who those vider,carethan seek not did morecaregiverswho with pro health a at care–seeking for constraints primary the were (22.8%) distance and (24.3%) transport of lack (82.3%), child’sCost their illness. for providerfatal care health a fromcare seeking in problems or concerns some had they that reported care seek to tried or sought received, children whose (68.0%) Figure 5 explores the care–seeking constraints for fatal child illnesses. In total, 400 of the 588 caregivers dren ( differencesignificant severelystatistically the chil no and ill ill normal/moderately the between delay in was there that showed test median A care. formal seek to made was decision the when ill severelywere Figure 5. ferral andwenttoasecondhealthcare provider. re the accepted or recommendations the all followed (77%–82%) caregivers the of provided,most rals refer or received, were however, referred; were recommendations when alive provider first the left that were referred and received home care recommendations (n 0.923) when the decision was made to seek formal care, and the difference of –0.49 (±SD 2 Main care–seeking constraintsforchildillness (N = (1) = 20) to a second health carehealth second provider,a to 20) receivedonly care home recommendations (n 0.831, P 52) of those children who died at the first provider were judged by their caregivers to be be to caregivers their by provider judged were first the at died who children those of 52)
= 0.3619). = ). Many (n Many 4). Figure 144) of the 368 that reached a health care provider and left the provider alive = –12.35; P < 0.0001). 304
= 2 111) children who were mildly or moderately ill at the the at ill moderately or mildly were who children 111) 1 = 9.325; P 400 caregivers). = < 47). In summary, just 67 (18.2%) of the 368 0.010. = 514, including 8 who had missing infor = 1.043) at onset of the illness to 2.15 www.jogh.org • doi:10.7189/jogh.07.010601 = = 157), or or 157), 0.893) 0.893) ------www.jogh.org • doi:10.7189/jogh.07.010601 from pneumonia was 15% in Cameroon [ Cameroon in 15% was pneumonia from [ pollution air effectsof the to vulnerable more be may they because children among particularly (ARLI), infections respiratory lower acute other and pneumonia of According to the World Health Organization, exposure to indoor air pollution more than doubles the risk air pollutionintheworld[22]. instead of outdoors, leading to 80% of the children being exposed to some of the highest levels of indoor home the into vented is smoke Hence, lighting. for lamps kerosene and cooking, for firewood namely sources, energyaffordable and available domestically on relied households of majority overwhelming roomsleeping the a addition, as In and night. daytime at the during cooking roomfor single serves that fia palm fronds or metal roofing. Besides, ita was locally and culturally accepted for households to have typical of the traditional Bantu dwellings made of sun–dried bricks placed in a wooden frame, with Raf was zones, rural in those especially area, study the in households of majority the of characteristic The ing thepracticeofchildmarriage[21]. cation among girls and eliminating gender gaps in education are known to be important strategies in end edu to improvingaccess Besides, area. this in needed Progressis uneven. remains application their but conventions, protection child major the all to signatories are Cameroon, ity.including countries, Many because it compromises the development of girls, and often results in early pregnancy and infant mortal [ rights human actuality,of In violation a concern. is of marriage is child age) of years 18 than (less age young a at childrenmarriage deceased or enteredthe union of into mothers the of half that fact The health statusofchildren from poorfamiliesiscompromised bytheirfamilies’circumstances [ general the that demonstrated that studies previous several with concurs households deprived in lived care system factors related to the children’s deaths. Finding that the majority of children in the study area health and community households, several assess to opportunity offerunique data a autopsy social The D [ care of quality poor in resultingmedicines essential of supplies and personnel medical adequate taining cially during the rainy season. In addition, public health facilities in the region often face difficulty main roadsquality,rural poor and of arepopulation, often study our for facility health usual the to utes espe care: access to public health care throughout this region is limited, with a median travel time of 30 min health seek to decision timely hindereda conditions following the suspect we But areunknown. setting a 'waiting game' to see whether the illness subsides on its own [ ment practices [ factors, such as religious and cultural norms, the cost of seeking health care, economic and and the social acceptability by of also treatbut services, care health of availability the by only not determined largely ditional healer [ prescribetreatment fromthat seekingfirst traditions tra cultural a and etiologies disease of derstanding un of lack a or condition, illness the of gravity the recognize to inability an from result to shown been careseek previousto in deciding described in been delay has The 1) delay so–called has (or and studies (n = many result, a As care. formal seek to made was decision the when to illness the of onset from days n = sibly severe illnesses. And, unlike the deceased newborn cases [ childrenthe When caregiversthe of all almost ill, fatally became recognized severeof symptoms pos or under–five–years–of–age children’s [ deathsare attributabletoundernutrition their poor nutritional status prior to the illness onset. Indeed, it is estimated that more than one–third of 64% the deceased children whose illness started at 0–23 months old may have also been endangered by ment feeding that, in turn, could lead to malnutrition, illness, and mortality [ complementary/replace inappropriate and breastfeeding interrupted of levels increased have to known also died—are they before in living were children the ones the as such – households poor or Deprived a chimneycouldsignificantlyimprove families’andchildren's health[ the traditional 3–rock cook stove with an improved stove and venting the smoke out of the house through to document the amount and composition of pollutant exposures among children. Until then, replacing pollutant exposure research and intervention. Future study considerations should include direct measures yses among 1–59 months old children in the study setting. This finding sets the stage for more in–depth pneumonia was responsible for 17%–20% of the deaths by expert algorithm and physician–coded anal ]. Likewise, the current study revealed that unaffordable costs for transportation and health care are care health and transportation for unaffordablecosts that revealed study current the Likewise, 38]. ISCUSSION 514) sought or tried to seek some formal care. The only problem with that was the long delay of 2 2 of delay long the was that with problem only The care. formal some seek to tried or sought 514) 111) children were taken for care only after their illness progressed from mild or moderate to severe. 36 32– ]. Other argued that past experience with similar illnesses can motivate mothers to play 35]. Some authors have posited that the decision to seek care for childhood illness is ]. The unpublished VASA study report [ report VASAstudy unpublished The 24]. 305 ]. Besides, the proportion of child deaths deaths child proportionof the Besides, 23]. 11], a greater number of caregivers (83.7%, 37]. The reasons for delay 1 in this study Factors ofchildmortalityineasternCameroon 26, June 2017 •Vol. 7No. 1•010601 27]. 28– 31]. 30]. The health of about ] revealed that that revealed 25] 17– 19]. ], 20], ------
VIEWPOINTSPaPErS PaPErSVIEWPOINTS June 2017 •Vol. 7No. 1•010601 Koffi etal. seeking basicchildhealthcare mightbeefficacious inthelongterm. to obstacles overcomeeconomic caretakerscan the parentsor that improvedsuch An standardliving of the publichealthagendaiscrucialtoconsolidatinggainsmadeduringpasttwodecades. on high feeding child young and infant improved keeping addition, In coverage. management case ing treatment from an appropriate provider is an important component of any intervention aimed at improv prompt seek children’sto among caregivers demand informed Building care–seeking. health propriate givers’ ability to recognize danger signs of child illnesses and facilitate behavior change for timely and ap- Short–term interventions could include the introduction of the C–IMCI program that could increase care smoke, andhealth–related behaviorssuchasdelayingthedecisiontoseekcare. indoor to exposure to lead that practices cultural or customs prevailing status, nutritional poor tions, contributed to the deaths of children under five years of age. Among these factors are poor living condi of Cameroon sheds light on the most common household, community and health care system factors that study conducted in Doume, Nguelemendouka, and Abong–Mbang health districts in the Eastern Region globally. and nationally both autopsy support social and recent emphasis The find decade past the over ignored been have that aspects that important is it era, post–MDG the about and accomplished be not In conclusion, as the global health community deliberates the strong likelihood that MDG–4 targets will CONCLUSIONS production likelytoaffect childhealthindependentlyofhouseholdorneighborhood economicstatus. levels of economic development and access to health care, as well as distinct climatic conditions and food disparate variation, cultural and socio–demographic marked exhibit Cameroon try.in regions Different addition, the findings in this study in the Eastern region of Cameroon cannot be applied to the whole coun along with the quality of may interviewers/supervisor/trainers have led to better overall recall of events. In interviews, face–to–face the during used modes prompting and conversational However,the questions. sensitive to answers desirable socially providing of likelihood the and events past of bias recall cluding it is possible that the data may have been the affecteddeceased newborns, by different types of biases, in of caregivers main the wererespondents the that fact the to added and that Given 2–3years). (IQR: years 2 was deaths months 1–59 the for period recall median the period: recall the to refers limitation second [ children all to accessible be should that interventions studying are we since necessary so not and children) and controls (alive children). However, the lack of a comparison group in SA studies is common (deceased cases among interventions of coverage the between differences significant were there whether The major limitation of this study was the absence of a comparison group that would allow analysis to test ing andcasemanagement[39– prompttreatment–seekfacilitates week per days 7 night, or day the of time any at worker health a sult health care system, and addresses family and community practices. The opportunity with C–IMCI to con hood Illness (C–IMCI) strategy that improves case management skills of health workers, strengthens the Child of Management Integrated community–based the reinforce to need the warrants onset illness at groups perceivedseverity within nor levels, facility the at or community the at care fromseeking cantly signifi after.providerdays vary mal fewnot a did died days and 2 of 1 delay long this that fact the And This delay may have played a major role in the death of the severely ill children who reached the first for of care–seeking andtoprovide aneffective meansoftransportation[ key barriers to seeking care in this region, and confirmed findings that suggest a need to mitigate the costs coi_disclosure.pdf (availableonrequest from theAKK).We declare thatwehavenoconflictsofinterest. interests: Competing thors havegivenfinalapproval oftheversiontobepublished. input from HDK. AKK drafted the manuscript with significant input from HDK, FN, RPL, SM and RW. All au significant with data, the interpreted and analyzed AKK data. of acquisition the for responsibleRW were RL, Authorship declaration: HDK, AKK contributed towards the conception and design of the study. AKK, HDK, through agranttotheU.S.Fund forUNICEFtheChildHealthEpidemiologyReference Group. Foundation Gates Melinda VASAand the Bill for the providedCameroonFunding by was in Funding: study astheplatformtoVASA datathatserved materials ofthefullbirthhistory study. Acknowledgments
: We thank Population Services International (PSI) for help in providing all the datasets and All authors have completed the Unified Competing Interest form at www.icmje. at org/ form Interest Competing Unified the completed have authors All 41]. 306 11]. www.jogh.org • doi:10.7189/jogh.07.010601 ]. A A 12]. ------www.jogh.org rEFEr ENCES • doi:10.7189/jogh.07.010601 5 4 3 2 1 14 9 8 7 6 17 16 NunnalyJ.Psychometrictheory. 1998.NewYork: McGraw-Hill;1978. 15 13 12 11 10 26 25 24 23 22 21 20 19 18 epidemics in Niger, Nigeria, and Chad. PLoS Med. 2007;4:e16. RF,Grais JP,Guthmann S, Gerstl C, Dubray NargayeA, Djibo relatedmortality high Unacceptably al. et KD, measles to could bedone?AfrJHealthSci.2006;13:139-43.Medline:17348755 Macassa G, Burström B. Determinants of social inequalities in child mortality in Mozambique: What do we know? What S0140-6736(05)74234-3 2005;365:1099-104. Lancet. inequalities. health of determinants Social M. Marmot sources/Static/Products/Progress2015/English2015.pdf. Accessed:10October2015. York.New 2015. Available:Report Goals Development Millennium Nations. United port_2015_Web_8_Sept_15.pdf. Accessed:2October2015. York.Available:Children’sNew Nations Fund, United 2015. Report mortality: child in trends & Levels Estimation. Mortality Child for GroupInter-Agency Nations United erature review to examine the concept and the development of the method. Popul Health Metr. 2011;9:45. Metr. Health Popul method. the of development the and concept the examine to review erature Kalter HD, Salgado R, Babille M, Koffi AK, Black RE. Social autopsy for maternal and child deaths: a comprehensive lit USA:INSetORCMacro;ton, Maryland, 2004. Institut National de la Statistique (INS) et ORC Macro.Enquête Démographique et de Santé du Cameroun 2004. Calver Medline:10159677 Frankenberg E. The effects of access to health care on infant mortality in Indonesia. Health Transit Rev. 1995;5:143-63. mortality inruralTanzania, 2005–2007.GlobHealthAction.2012;5:1-9. WaiswaA, Mathew H, Masanja B, Sartorius D, Kadobera P. childhood on facility health formal effectto The distance of line:21819605 vival (BASICS)Project. 1998.Available: http://pdf.usaid.gov/pdf_docs/PNACF082.pdf.Accessed:14October2015. tigating and Identifying the Causes of Death for Children Under Five. The Basic Support for Institutionalizing Child Sur Aguilar AM, Alvarado R, Cordero D, Kelly P, Zamora A, Salgado R. Mortality Survey in Bolivia: The Final Report. Inves jogh.06.010603 veloping countries?Demography. 2009;46:303-24.Medline:21305395 de in children older for stronger health and status socioeconomic between relationship the Is WilliamsJ. L, Cameron BF02310555 1951;16:297-334. Psychometrika. tests. of structure internal the and alpha Coefficient LJ. Cronbach from the 2012 National Verbal and Social Autopsy Study. Results J Niger: Glob in Health. mortality child 2016;6:010603. of determinants Social HD. Kalter K, Bensaïd O, Habi AG, A, Yaroh Maina AK, Koffi doi:10.1093/ije/dyq018 terventions to reduce maternal, neonatal and child mortality. Int J Epidemiol. 2010;39 suppl 1:i3-6. Boschi-Pinto C, Young M, Black RE. The Child Health Epidemiology Reference Group reviews of the effectiveness of in in Doume, Nguelemendouka and Abong–Mbang health districts, Eastern Region of Cameroon. J Glob Health. Glob 2015;5:010413. Medline:26171142 J Cameroon. of Region Eastern districts, health Abong–Mbang and Nguelemendouka Doume, in Libite Koffi AK, mortality neonatal of determinants identifies study S, autopsy Moluh Social P, HD. Kalter R, Wounang OMD_Rapport_2015_Cameroun.pdf. Accessed:17July2016. Available:2015. Septembre Cameroun, 2015. en ment Institut National de la Statistique (INS) du Cameroun. Rapport national sur les Objectifs du Millénaire pour le dévelope Medline:24112419 proved cookstoves in three resource-limited settings: study protocol for a randomized controlled trial. Trials. 2013;14:327. Klasen E, Miranda JJ, S, Khatry Menya D, Gilman RH, Tielsch JM, et al. Feasibility trial intervention of two types of im 2017. Nov2013.pdf. Accessed:9February quête Autopsie Verbale/ Autopsie Sociale (AVAS). Cameroun 2013. Available: En (CHERG)- Group. Reference Epidemiology Health Child et (INS-CAMEROUN) Statistique la de National Institut vival. 2015.Geneva:World HealthOrganization. World Health Organization. Countdown to 2015. A Decade of Tracking Progress for maternal, Newborn and Child Sur 2008;86:390-398C. Medline:18545742 and pneumonia risk in children aged under five years: a systematic review and meta-analysis. Bull World Health Organ. Dherani M, Pope D, Mascarenhas M, Smith KR, Weber M, Bruce N. Indoor air pollution from unprocessed solid fuel use 2012;7:265-70. gheart.2012.06.016 Heart. Glob environment. indoor the in fuel biomass ing Diette GB, Accinelli RA, Balmes JR, Buist AS, Checkley W, Garbe P, et al. Obstructive lung disease and exposure to burn NATIONAL_STRATEGY_ON_CHILD_MARRIAGE-PRINT_READY.pdf. Accessed: 13June2015. pregnancy.teenage and marriage child fromfree Available:society 2015. A http://www.unicef.org/uganda/2019/2020. – 2014/2015 pregnancy teenage and marriage child end to strategy national The Children’sFund. Nations United un.org/womenwatch/daw/cedaw/text/econvention.htm 2017. . Accessed:9February Available:women. against discrimination of forms all of elimination the on Convention Nations. United mortality inRufijiDSS, Tanzania. GlobHealthAction.2013;6:19278. under-five and status socio-economic household between Relationship K. Klipstein-Grobusch H, Masanja C, Nattey review of the international evidence.BrMedBull.2010;93:7-26. review oftheinternational Houweling TA, Kunst AE. Socio-economic inequalities in childhood, mortality in low- and middle-income countries: a doi:10.1186/1478-7954-9-45 doi:10.1186/1745-6215-14-327 doi:10.7189/jogh.05.010413 doi:10.2471/BLT.07.044529 307
http://slmp-550-104.slc.westdc.net/~stat54/downloads/2015/ http://www.unicef.org/publications/files/Child_Mortality_Re Medline:17199407 Medline:20007188 doi:10.3402/gha.v6i0.19278 doi:10.1353/dem.0.0054 Medline:23151364 http://www.stat.cm/downloads/2016/VASA_ Factors ofchildmortalityineasternCameroon Medline:23139916
doi:10.1371/journal.pmed.0040016 June 2017 •Vol. 7No. 1• 010601 Medline:15781105 http://mdgs.un.org/unsd/mdg/Re- doi:10.1093/bmb/ldp048 Medline:26955473 Medline:20348123
doi:10.1016/j.
doi:10.1007/ doi:10.1016/ doi:10.7189/ http://www. Med------
VIEWPOINTSPaPErS PaPErSVIEWPOINTS June 2017 •Vol. 7No. 1•010601 Koffi etal. rEFEr ENCES 41 40 39 38 37 36 35 34 33 32 31 30 29 28 27 pl):99-115. Medline:20168036 Sup 2009;20(4 Underserved. Poor Care Health J areas. rural underserved in systems health transforming Nicaragua: GeorgeEP,Menotti A, in illness childhood of management case Community DR. Marsh CM, Reyes I, Montes D, Rivera lar J.2011;10:159.Medline:21651828 plicated malaria in two rural districts in Zambia: implications for nationwide scale-up of home-based management. Ma P,Chanda V,Chalwe HB, Moonga B, Hamainza P, Banda F.Pagnoni uncom treating of effectiveness and costs Relative line:23144228 outcomes of a community case management program in Cameroon (2013). Health Policy Plan. 2013;28:705-16. mid-term access: equitable with gap treatment LV,the NarrowingMoukam G. M, Baugh Littrell YoumbaJC, R, Libite Medline:23769757 Midwifery. services. midwifery 2013;29:1211-21. in task-shifting of implementation the to facilitators and barriers on Colvin CJ, de Heer J, Winterton L, Mellenkamp M, Glenton C, Noyes J, et al. A systematic review of qualitative evidence D’Souza RM.Care seekingbehavior. ClinInfectDis.1999;28:234. doi:10.1186/1475-2875-8-144 2009;8:144. J. Tanzania.Malar south-eastern in degedege for facilities health of use appropriate and timely explaining factors Socio-cultural al. MW,F,et Hetzel Kessy I, A, D, Mayumana Dillip Gosoniu C, Lengeler 55. Medline:10747275 2000;5:145- TropHealth. Guatemala. Int in Med P.illness Heuveline child N, for Goldman behaviour Health-seeking Medline:17591337 ness and its implications for safe motherhood in interventions rural Bangladesh. J Health Popul Nutr. 2006;24:403-12. ill serious of episodes for healthcare-seeking in delay Perceived J. YunusChakraborty I, M, Bashir I, Anwar J, Killewo 6736(07)61690-0 2008;371:243-60. Lancet. consequences. health and exposures regional Black RE, Allen LH, Bhutta ZA, Caulfield LE, de Onis M, Ezzati M, et al. and Maternal child global undernutrition: and doi:10.1016/S0140-6736(12)60560-1 2012;379:2151-61. Lancet. 2000. since trends time with 2010 for analysis systematic updated an mortality: child of causes national and regional, Global, al. et JE, Lawn S, Scott J, Perin S, Cousens HL, Johnson L, Liu 2017. nutrition/topics/global_strategy/en/. Accessed:9February WorldYoungand Available:Infant 2003. for Feeding. Strategy Organization.Child Global Health study.pital-Zimbabwe: anobservation CentAfrJMed.2007;53:30-4. Chimhuya S, Kambarami RA, Mujuru H. The levels of malnutrition and risk factors for mortality at Harare Central Hos 6736(07)61690-0 2008;371:243-60. Lancet. consequences. health and exposures regional Black RE, Allen LH, Bhutta ZA, Caulfield LE, de Onis M, Ezzati M, et al. and Maternal child global undernutrition: and nutritional statusinCameroon. BMCPublicHealth.2006;6:98. child of determinants environmental and socioeconomic community and Household JA. Salomon E, Majid R, Pongou ijtld.13.0485 trial of a cookstove intervention in rural Malawi. Int J Tuberc Lung Dis. 2014;18:240-7. Medline:24429320 HR, Jary Kachidiku J, Banda H, Kapanga M, Doyle JV, Banda E, et al. Feasibility of conducting a randomised controlled doi:10.1093/heapol/czs110 doi:10.1016/j.midw.2013.05.001
doi:10.1046/j.1365-3156.2000.00527.x doi:10.1353/hpu.0.0205 doi:10.1186/1475-2875-10-159 308 Medline:16618370 Medline:10064233 Medline:20355679 Medline:18207566 Medline:18207566 www.jogh.org doi:10.1186/1471-2458-6-98 doi:10.1086/515120 • doi:10.7189/jogh.07.010601
doi:10.1016/S0140- doi:10.1016/S0140- http://www.who.int/ Medline:19563640 Medline:22579125 doi:10.5588/ Med------
www.jogh.org Program Prevention ofMother–to–ChildTransmission to monitorearlyinfantinfectionswithinthe transmission ofHIVinSoutha Toward eliminationofmother–to–child ameena EbrahimGoga Bhardwaj adrian Puren Mazanderani 6 5 4 3 2 1 Gayle GSherman [email protected] South a Sandringham 2031 1 Modderfonteinr Centre forHIVandSTIs a Correspondence to: 12 11 10 9 8 7 Department ofPaediatrics,UniversityPretoria, Health SystemsresearchUnit,SouthafricanMedical United NationsChildren’sFund,NewYork,USa School ofPublicHealth,UniversitytheWesternCape, Division ofVirologyandCommunicableDiseases,School Department ofBiology,MassachusettsInstitute Nelson MandelaUniversity,PortElizabeth;andUniversityof United NationsChildren’sFund,Pretoria,Southafrica School ofPublicHealth,FacultyHealthSciences, Department ofMedicalVirology,UniversityPretoria, Department ofPaediatricsandChildHealth,Faculty Centre forHIVandSTI,NationalInstituteCommunicable hmad HaeriMazanderani South africa research Council,CapeTown,Southafrica Cape Town,Southafrica Johannesburg, Southafrica Pathology, UniversityoftheWitwatersrandMedicalSchool, Technology, Cambridge,Massachusetts,USa South africa,Pretoria,africa University oftheWitwatersrand,Johannesburg,Southafrica Pretoria, Southafrica Johannesburg, Southafrica of HealthSciences,UniversitytheWitwatersrand, Services, Johannesburg,Southafrica Diseases, adivisionoftheNationalHealthLaboratory frica • doi:10.7189/jogh.07.010701
5 , ronelleNiit 1,3 1,8
oad , PeterBarron , DebraJJackson
1,2 , ahmadHaeri
6 11,12 , MargaretOkobi 4 , Sanjana 9,10 , 7 , 309 Background South Africa has utilized three indepen- in SouthAfrica. eMTCT track to sources data routine of usefulness dence intervals for 2012. This analysis validates the confi- survey SAPMTCTE the within fall that mates esti- rate MTCT early similar very providesources, Conclusion native estimate of the HIV–exposed infant population. alter correctedan be employing can by this but age, cover EID over–estimates data DHIS 2014. in 87% to 2010 in 52% around from increases data NHLS from coverage EID National 2012. by rates MTCT NHLS age–adjusted with converge to thereafter timated MTCT rates in 2010, the MTCT rate declines sponding time periods. Although DHIS data over–es corre in surveys SAPMTCTE three by measured as interval confidence 95% the within fall sistently 4.1% in 2010, 2.6% in 2011 and 2.3% in 2012 con Results The age–adjusted NHLS early MTCT rates of infant diagnosis(EID)coverage. the NHLS and DHIS were also used to compare early transmission (MTCT) rates in South Africa. Data from mother–to–child early compare to used were veys ber 2014, from the NHLS, DHIS and SAPMTCTE sur data, collected between 1 January 2010 to 31 Decem Methods sion (eMTCT). transmis mother–to–child of elimination achieving toward works Africa South as rates transmission ing limitations, and make recommendations for monitor compare the results of each, outlining advantages and African PMTCT Evaluation (SAPMTCTE) surveys. We South and (DHIS), System Information Health trict National Health Laboratory Service (NHLS), the Dis sion (PMTCT) of HIV. These include the South African gram for the prevention of mother–to–child transmis pro its of impact the measure to sources data dent HIV polymerase chain reaction (PCR) test test (PCR) reaction chain polymerase HIV frica: how best frica: howbest HS n DI, w ruie data routine two DHIS, and NHLS June 2017 •Vol. 7No. 1•010701 global journal of health ------
VIEWPOINTSPaPErS PaPErSVIEWPOINTS June 2017 •Vol. 7No. 1•010701 Sherman etal. Table 1. PCR –polymerasechainreaction, cps/ml–copiespermilliliter, wks–weeks,VLviral load,NVP–Nevirapine If HIVPCRnegative: If HIVPCRpositive: If HIV–exposed & asymptomatic: If HIV–exposed&symptomatic: y ear
oF g uIdelIne South AfricanNationalGuidelinesforEarlyInfantDiagnosisofHIVExposedInfants[ sequently beapproved bynational,regional andglobalvalidationcommittees. sub must reportthat validation national a targets,inform eMTCT to appropriateneed tools monitoring achieve orderto In units. administrative sub–national lowest–performing the of one least at in achieved impact targets have successfully been met for one year, process targets for two years, and eMTCT has been <2% in non–breastfeeding populations. tion ofmother–to–childtransmission(eMTCT)status[4 elimina its of pre–validation for prepares Africa South as recommendations make and limitations, and advantages respective their outlining method, each of results the Weprogram.compare PMTCT tional na the of effectiveness the assess to 2010 since conducted been have surveys (SAPMTCTE) Evaluation PMTCT African South facility–based national, three Furthermore, population. infant HIV–exposed the mother–to–child transmission (MTCT) rate and the coverage of early infant diagnosis (EID) testing among early the monitor to independently used been have (DHIS) System Information Health District the and (NHLS) Service Laboratory Health National fromthe program.Data PMTCT the effectivenessof the tor infections per 100 per infections HIV pediatric new ≤50 are targets eMTCT. impact of required The certification to prior met be to need nization (WHO) [ Worldthe by Orga suggested Health been have HIV of eMTCT of validation for processes and Criteria Elimination ofmother–to–childtransmission decreaseto continued [1 has infants infected vertically of number the 2004, since 31% and 29% between remaining consistently the national antenatal HIV sero–prevalence, as measured among women attending public health facilities, HIV,of (PMTCT) Despite infection. HIV infant of incidence the curbing in made been have strides huge Since 2004, when South Africa officially launched its program to prevent mother–to–child transmission ing in HIV–exposed infants at six weeks of age [ In 2004, the South African National Department of Health (NDOH) recommended routine HIV PCR test South AfricanGuidelinesforEarlyInfantDiagnosis test negative at birth and the falling–away of the standard six–week test ( [ guidelines national into introduced was infants HIV–exposed all for testing birth routine infants, fected the country [6 of parts some in implemented was infants “high–risk” but asymptomatic of testing birth targeted tional addi the 2013 in decade, a over for care of standard been has age of weeks six at infants HIV–exposed all testing and age of weeks six to prior infants symptomatic testing Whereas guidelines. EID to made 5 ]. Additional changes to EID guidelines include a second HIV PCR test at 10 weeks of age for those who ]. Subsequently, in June 2015, as a means of ensuring earlier detection of intra–uterine in- Repeat HIVPCRtestifinfantsymptomatic,andrepeat 6wksaftercessationofbreastfeeding asymptomatic only ifchildis Repeat HIVPCR baseline HIV VLtestat at ≥6wks HIV PCR
4
000 live births and a transmission rate of either <5% in breastfeeding populations or or populations breastfeeding in <5% either of rate transmission a and births live 000 ]. These minimum global standards refer to specific impact and process targets which 2004 – 3 ]. Various]. differentutilizing methods moni- to employed been have sets data asymptomatic only ifchildis Repeat HIVPCR baseline HIV VLtestat at 6wks HIV PCR 2008 4 Countries are encouraged to apply for validation of eMTCT once 310
5 HIV PCRtestatpresentation status confirms HIVpositive VL >10 HIVVL: Confirmatory at 6wks HIV PCR ]. Since then, a number of important changes have been 000 cps/ml 000 cps/ml 2010 ]. status confirms HIVpositive Any quantifiedVL HIVVL: Confirmatory at 6wks HIV PCR 3,5–10 www.jogh.org Table 1 2013 ] ). Furthermore, the test • doi:10.7189/jogh.07.010701 cps/ml VL>1000 maternal breastfeeding and Repeat HIVPCRif 12 wksNVP) 18 wks(ifreceived Repeat HIVPCRat at 10wks Repeat HIVPCR PCR HIV Confirmatory at birth HIV PCR 2015
------www.jogh.org • doi:10.7189/jogh.07.010701 NHLS andDHISdata. from calculated both been has testing infant early of uptake addition, In surveys. threeSAPMTCTE and DHIS the NHLS, the namely sources,different data threefrom different methodologies using measured typically monitored between the ages of four to eight weeks of age. In South Africa, early MTCT has been is and intrapartum, or intra–uterine either acquired is that transmission vertical to refers MTCT Early Data sources usedtomonitor MTCTinthePMTCTprogram HIV PCR test [ confirmatory a to load viral HIV an from changed has status infection HIV confirming for method ing as adenominator. number of HIV PCR positive infants as the numerator and DBS HIV ELISA positive with HIV PCR result, the using calculated was respectively.MTCT positive, early PCR The HIV DBS and ELISA HIV DBS ed (DBS) samples from them. HIV–exposed infants and HIV–infected infants were defined as those thatspot test blood dried collected and status mother’sinfection their HIV of regardless age, of weeks 8 and 4 The SAPMTCTE surveys enrolled infants attending their 6–week immunisation visit, if they were between infant thatisfirsttestedbetweentheagesof4and12weeks. and a denominator of “infant first PCR test conducted around 6 weeks”. Around 6 weeks is defined asan weeks” 6 around positive PCR first “infant of numerator a with rate” weeks 6 around positive test PCR The DHIS indicator used to monitor positivity in HIV–exposed infants around 6 weeks is the “infant first performed inthisagegroup. MTCT rates by calculating the proportion of HIV PCR positive tests to the total number of HIV PCR tests early proxyfor a as age of months <2 children in rates positivity PCR HIV the reports CDW NHLS The Early MTCTrates NHLS andDHISwere usedtocompare EIDcoverage. fromthe data rates, MTCT early to addition In initiated. was survey each which in year the to allocated years (2010–2014). For the SAPMTCTE surveys, the MTCT rates and their 95% confidencefive across intervalsrates were MTCT early compare to used were surveys, SAPMTCTE and DHIS the NHLS, the es, HIV PCR test data collected between 1 2010 January and 31 December 2014 from three different sourc METHODS 2012 andOctobertoMay2013. partum. Three surveys have been conducted to date from June to December 2010, August 2011 to March post months 18 and 15 12, 9, 6, 3, at recentlymore and age of weeks 6 at HIV of MTCT determine to at least 130 first DTP immunisations per year [ tion–based representative sample. The sampling unit of these surveys was primary level clinics reporting popula a using program PMTCT the of impact the determining of aim the with Council, Research ical The SAPMTCTE were national surveys conducted over three consecutive years by the South African Med PMTCT indicators. national level to track health service delivery. The data elements collected include those that make up the vincial and national level for collation. There is monthly reporting at sub–district, district, provincial and scribed registers at each facility. They are prethen capturedin summated electronically and in collected the are DHIS data and These transmitted to NHLS. prothe from results PCR HIV includes and Africa, South in districts health 52 the of each in carefacilities fromhealth aggregatedata all gathers DHIS The across thecountry. performed and the number of HIV PCR positive test results for approximately 4000 health care facilities corporate data warehouse (CDW). Monthly reports are generated detailing the number of HIV PCR tests NHLS the in centrally stored and results test the with together (LIS) system information laboratory the into entered is information This requested. test of type the and performed was testing which at facility the collection, specimen of date the patient, each for details identifying stipulates that form requisition test NHLS an by country.accompanied the is of test population laboratory total the Every of 80% at ed The NHLS provides diagnostic services for the whole of the public health sector in South Africa, estimat using routine data. laboratory 5 , 8 ]. These changes in EID guidelines have important implications for monitoring MTCT 1 311 11]. The primary objective of the SAPMTCTE surveys was
June 2017 •Vol. 7No. 1•010701 Towards eMTCT ofHIVinSA ------
VIEWPOINTSPaPErS PaPErSVIEWPOINTS June 2017 •Vol. 7No. 1•010701 Sherman etal. Table 2. MTCT rateswere higherthantheDHISratesby0.2%and0.5%,respectively ( ( increased years all for rates MTCT early the age, of weeks 4–12 infants on performed tests PCR HIV include to adjusted was data NHLS the When for2012. confidence interval survey’sSAPMTCTE 95% the within lie to years two next the over dramatically falls rate However,this MTCT rate was reported as more than double the NHLS and SAPMTCTE MTCT rates in 2010 ( 2 The number of HIV PCR tests as recorded by the DHIS is consistently higher than for NHLS data ( as measured bythreeincorresponding SAPMTCTEsurveys timeperiods( interval confidence 95% the within fall consistently rates MTCT early NHLS the data, SAPMTCTE and ( 0.1% by formly fromyear per rate MTCT early NHLS the age), of weeks 4–8 infants (ie, veys decreases2010–2014 uni- are re–adjusted to only include HIV PCR tests performed for the same age ranges as the SAPMTCTE sur ( 2012 in –0,2% and 2011, in 0.0% 2010, in differing0.7% by surveys, The early MTCT rates as determined from NHLS data compare closely with results from the SAPMTCTE Early MTCTrates r [ percentage a as expressed infants, HIV–exposed of number expected the by divided age of months ≤2 aged infants in tests PCR HIV registered of number the as coverage EID defines CDW NHLS The Early InfantDiagnosiscoverage 4–8weeksofage. the SAPMTCTEsurveys, in and age, of weeks 4–12 data, DHIS the in testing PCR HIV of age the match re–extractedto was data threediffered,the sets in data MTCT early calculate to used infants CDW the NHLS of ages the Because sero–prevalence, reported bytheNDOH[3 HIV antenatal maternal national with DHIS, by captured women’, all to births live ‘total multiplying by mothers in DHIS, an denominator alternative is also used that estimates HIV–exposed infant population births to HIV positive women”. To account for potential under–reporting of infants born to HIV–infected “live of denominator a and weeks” 6 around conducted test PCR first “infant of numerator a with rate” uptake weeks 6 around test PCR first “infant the is coverage EID monitor to used indicator DHIS The alence ofHIVreported bytheNDOH[3 al registered live births published by STATS SA multiplied by the national antenatal maternal sero–prev The denominator (ie, the HIV–exposed population requiring HIV testing) is calculated using the nation mate theeligibleagegroup andreported intheSAPMTCTE surveys agegroup intheDHISdatarespectively. approxiclosely more to weeks, 4–12 5) and weeks 4–8 4) ages at presentedare data NHLS age, of <2months 1) at reported data NHLS routine the to addition In 1)–3). sets data in tabulated are SAPMTCTE and DHIS NHLS, for rates MTCT of *Comparison DHIS –DistrictHealthInformation System Evaluation, PMTCT African South – SAPMTCTE reaction, chain polymerase – PCR Service, Laboratory Health National – NHLS )NHLSHIVPCRtests 1) )DHISHIVPCRtests 2) 3) SAPMTCTE MTCT rate SAPMTCTEMTCTrate 3) )NHLSHIVPCRtests 4) )NHLSHIVPCRtests 5) ). The same is true for the calculated early MTCT rate, except for the 2014 estimate. The DHIS early early DHIS The estimate. 2014 the for except rate, MTCT early calculated the for true is same The ). ESULTS d NHLS HIVPCR+tests HS%pstv I C et 2ots42 .%24 .%1.8% 2.0% 2.4% 2.7% 4.2% <2months NHLS %positiveHIVPCRtests DHIS HIVPCR+tests HS%pstv I C et 6wes90 .%27 .%1.6% 2.1% 2.7% 4.3% 9.0% ±6weeks DHIS %positiveHIVPCRtests (95% confidenceintervals) NHLS HIVPCR+tests HS%pstv I C et – ek .%26 .%19 1.7% 1.9% 2.3% 2.6% 4.1% 4–8weeks NHLS %positiveHIVPCRtests NHLS HIVPCR+tests HS%pstv I C et –2wes49 .%27 .%2.1% 2.3% 2.7% 3.2% 4.9% 4–12weeks NHLS %positiveHIVPCRtests ata Early HIVtransmissionratesinSouthAfrica2010–2014*
sets
TableImportantly,4). set Data 2, NHLS the in infants of ranges age the matching after
–2wes75 1552 045106 5064 5823 6125 7158 139 4–12 weeks 4–12 weeks 4–8 weeks 3.5% 3.5% 4–8 weeks – ek 113 4–8 weeks – ek 8947 1037 3624 3579 4110 4271 4849 4–8 weeks 2ots119 <2months 2ots58 6944 924054 3912 4440 4609 5282 <2months 6wes178 ±6 weeks 6wes17 ±6 weeks a ge , 12]. 312 ]. (2.9%–4.1%)
Table2 0021 0221 2014 2013 2012 2011 2010 2 5661 144089 5184 6611 9556 528 0 164 808 4 211 241 2 157 722 1 187 517 , Data set 5). In 2013 and 2014 the NHLS early early NHLS the 2014 and 2013 In 5). set Data , (2.1–3.2%) 2.7% 2.7% 8 184 181 4 237 942 1 176 411 2 206 020 (2.0–3.2%) www.jogh.org Table data NHLS the When 2). 2.6% 2.6% Table 2,Dataset3). 0 195 400 6 243 869 8 186 787 9 216 990 Table 2). • doi:10.7189/jogh.07.010701 8 222 188 8 247 786 6 208 969 1 236 410 Table 2). Table 559 037 364 708 1 ]. ]. - - - - www.jogh.org • doi:10.7189/jogh.07.010701 labor intensive. the urgent need for continued, regular parallel surveillance activities that are expensive and resource and routinesources resultsand data obviates survey between effectiveness.similarities PMTCT The national measure to needed targetsinitially PMTCT was same the monitor to methodologies multiple of use The a convergence inboth theabsolutenumberstestedaswelloverallcoverage rates. Both NHLS and DHIS document an increase in infant HIV testing coverage between 2010 and 2014 with MTCT ratesin2013and2014theNHLSvsDHIS data. confirmatory HIV PCR testing, resulting in double counting, may be the explanation for the higher early in increase an that postulated is it Furthermore, testing. birth on report to workers care health for place in yet was mechanism no because data DHIS the from excluded but data NHLS the in included being testing neonatal to attributable likely is NHLS and DHIS to according performed tests PCR HIV of ber decreasingdifferencethe num- guidelines, the African in South introducedthe was into neonates of ing Fromperformed. was grouptesting age targetedPCR week when HIV 2013, little wheretest- birth very 0–4 the included and group age this excluded that data NHLS the with compared as infants old week 8–12 of inclusion recordedreflects an tests likely PCR DHIS HIV the of by number higher the 2012, By weeks) [13]. 6 around aged infants on performed test PCR HIV first the reporting of instead ages all of children on performed tests PCR HIV all reporting (ie, incorrectly indicators DHIS the reporting were workers care DHIS data are likely attributable to quality improvement training in the field after it was found that health The differences between DHIS and NHLS data in 2010–2012, and the remarkable reduction in MTCT in ofthesurveys. confidence intervals veys conducted in the same year, demonstrating that results from routine data sources lie within the 95% sur SAPMTCTE the to them comparing by verified was sets data routine distinct these from calculated for monitoring the effectiveness of South Africa’s PMTCT program. The accuracy of the early MTCT rates This analysis shows that data from routine sources, namely the NHLS and DHIS, produce similar results D for allnineprovinces (datanotshown). Early MTCT and EID coverage rates between 2010 and 2014 showed similar patterns to the national one ence in2014(Figure 1). differ2% a only with time convergesover coverage NHLS EID reportedthe used, that by is with inator pears to be consistently higher and exceeds 100% by 2013. However, when the alternative DHIS denom ( 2014 in 87% and 2012 in 68% National EID coverage as determined from NHLS data are seen to increase steadily from 52% in 2010 to EID coverage ISCUSSION ). Alternatively, EID coverage as reported from DHIS data ap data Alternatively,DHIS 1). fromFigurereported as coverage EID 313
June 2017 •Vol. 7No. 1•010701 coverage ofEIDforeachdataset. the linegraphsrepresent the DHIScalculation and alternative by NHLS,DHISandfrom an HIV-exposed infantsasestimated represents thepopulationof 2010–2014. Thebargraph Coverage ratesinSouthAfrica Figure 1. Towards eMTCT ofHIVinSA Early Infant Diagnosis Early InfantDiagnosis - - - -
VIEWPOINTSPaPErS PaPErSVIEWPOINTS June 2017 •Vol. 7No. 1•010701 Sherman etal. lection of age data [ data age of lection 100% testing coverage, close to zero missed diagnostic and opportunities accurate in col the laboratory to close be must there country’s rate the MTCT of reflection true a be to Importantly,data LIS. this for providedrequisition reliableand forms laboratory on fromcapturing data requisitionthe the into forms be to information accurate requiresprogram PMTCT the monitor to data laboratory routine of use The tocalculateMTCTrates. rates andEIDcoverage,aswelllimitationstotheSAMTCTEsurveys There are clear requirements and limitations to using routine NHLS and DHIS data to monitor early MTCT Limitations point–estimate islikelyanunder–estimationoftrueearly infantHIVinfectionprevalence [ those who had failed to present for immunisation, and those who had died by 6 weeks of age. Hence, the visit, immunisation first the at only,ill immunisation were for who presenting infants infants excluding primarilyrelate provideLimitations oftheSAPMTCTEsurvey to sampling. Thesurveys dataforhealthy their status,forreasons whichincludeseroconversion duringpregnancy, orchosenot todisclose[ found that 3–4% of HIV positive women did not report being positive, either because they did not know survey’sSAPMTCTE The coverage. EID over–estimate will taking frompopulation fant history maternal in HIV–exposed the determining that likelihood the support further survey SAPMTCTE fromthe Data population. infant HIV–exposed the estimate to denominator alternative an using by addressed be can [ resulting2013 low by 100% exceeds that coverage of over–estimation an in positive women” is likely under reported. Hence, the denominator used to calculate EID coverage is too HIV to births “live the ward, labor the in workers care health to status positive HIV their disclose not do or testing HIV recent or any without deliver women of number unknown an because instance, For [ challenges of number great a remain there Africa, South in services PMTCT monitor to used data tical data improvement interventions found to significantly increase the completeness and accuracy of the manner.consistent a in information improvedhas undoubtedly this While recent prac over with years, Limitations to the use of DHIS data primarily relate to training health care workers to capture the correct be overestimated accountingforalowerEIDcoverage. may sector,careinfants health HIV–exposed registeredprivate fromareof births the number live the of the ANC Maternal Sero–prevalence data from 2013 and, hence, may not be accurate. Since a proportion data. As these are published after a lag of 2–3 years, EID coverage for 2014 has been calculated based on sero–prevalence fromSTATSdenominator) maternal and antenatal data national registered SA birth live noted above, limitations include calculating the number of HIV exposed infants requiring testing (ie, the Regarding calculating EID coverage from NHLS data, in addition to challenges in de–duplicating data as 15,2016). Strategic Planning,NDOH.Personalcommunication,February Director, (M Chief year Wolmarans, financial 2016/17 the in facilities sector public all in implemented of unique patient identifiers. The NDOH has communicated that unique patient identifiers will likelybe introduction the without data NHLS routinefrom calculated be longer no can transmission, post–natal positive and repeat testing for those infants who are symptomatic. Hence, reliable MTCT rates, including test who infants those for testing PCR HIV confirmatory recommend guidelines current age, of months Whereas in the past it was assumed that very few infants would access more than one HIV PCR test by 2 fore, infants with multiple HIV PCR tests cannot be distinguished from infants with a single HIV PCR test. unique identifier for individual patients and no accurate means of de–duplicating test result data. There no currently is there that is limitation important further A results). invalid or indeterminate (eg, rors er analytical processing) and for insufficient sample (eg, pre–analytical to relatedresult negative or tive SAPMTCTE surveys exclude certain groups of infants who are possibly at high risk of HIV–infection [ the whereas tested and exposed HIV be to known infants all for data PCR recordHIV NHLS and DHIS for. Both controlled be cannot that sources data the between differences certain are there that preciate Whereas each methodology has its own advantages, disadvantages and challenges, it is important to ap- Advantages anddisadvantages EID coverage and comes at very little additional cost. In contrast, DHIS data takes time to collate and and collate to time takes data DHIS contrast, In cost. additional little very at comes and coverage EID Clear advantages of using NHLS data are that it allows for near real–time monitoring of early MTCT and are included in both the DHIS and NHLS data, the SAPMTCTE surveys do not include duplicate testing. sive of mothers who do not report being HIV positive. Whereas repeat HIV PCR tests on the same patient arethereforeHIV–exposureand for inclu infants all tested surveys SAPMTCTE the hand, other the On
]. Missed diagnostic opportunities are defined as samples yielding neither a posi a neither yielding samples as defined are opportunities diagnostic Missed 14]. 314 www.jogh.org ]. As demonstrated, this this demonstrated, As 16]. • doi:10.7189/jogh.07.010701 17]. 11]. ]. 15]. 18 ]. ]. ------www.jogh.org • doi:10.7189/jogh.07.010701 fant tocare,betweeninfantdiagnosisandinitiationoftreatment. andtheinterval where no other data are available. Examples include determining the rate of linking HIV PCR positive in questions detailed more specific, answering and sources; data routine validating for reserved odically; Ideally,consuming. time peri and conducted expensive be are should they data, surveys national–level the LIS does not hold clinical data. While the SAPMTCTE surveys are likely to provide the most accurate that is DHIS to CDW NHLS the Conversely,of expense. disadvantage greater distinct much a at comes rates andEIDcoverage. inform South Africa’s pediatric HIV surveillance systems as well as other countries monitoring early MTCT eMTCT targets. By outlining the value of routine laboratory data, it is anticipated that these findings will will not only preclude unnecessary duplication of data capturing within the DHIS but also reliably inform itoring the effectiveness of the national PMTCT program using routine laboratory data are envisaged. This patient identifiers and consolidating clinical information within the LIS, a more efficient method of mon value of SAPMTCTE surveys will be in periodically validating routine data methods. By introducing unique continued the data, DHIS and NHLS both of accuracy the to challenges new pose guidelines EID tional na in changes recent As activities. surveillance parallel regular of value continued the about questions intervals. These surveys validate the accuracy, and therefore usefulness, of routine data sources, and raise confidence survey the within falling estimates DHIS and laboratory the with rates, MTCT early similar evaluating the effectiveness of the PMTCT program on early MTCT. All three methodologies provide very district in the country. Additionally, three national surveys have been conducted between 2010 and 2012, erage, has been monitored in parallel using routine laboratory data and operational data collected by each The effectiveness of South Africa’s PMTCT program, as determined by the early MTCT rate and EID cov tive from theNDOH. direcclear a and personnel overcome to effortconsolidated laboratory therea and if is clinical between in order to accurately calculate MTCT rates using routine laboratory data. These challenges will be easier immunisation bookletswithuniquebarcodes thatcanbecaptured withintheLIS)willbeaprerequisite MTCT rates. Furthermore, a unique patient identifier, employed from birth, (eg, printing patient–retained validation particularly as South Africa recommends breastfeeding but has no system for monitoring final Access to accurate clinical information will be important for documenting the process targets for eMTCT a prescribed minimumclinicaldatasetwillneedtobestrictlyadhered to. ing tool will, nevertheless, pose certain challenges. In particular, the accurate and consistent capturing of itor the effectiveness of the PMTCT program at all levels of health care delivery. A consolidated monitor ing among health care workers of laboratory data and provide a more robust data set from which to mon treatment initiation and retention in care. This will preclude the duplication unnecessary of data captur infant care, into linkage infant regimens, treatment maternal on data including captured, be to cascade PMTCT CDW.the the acrossfrom within data populated clinical data enable clinical would priate This data, there are opportunities of addressing this by incorporating NHLS requisition forms with the appro clinical hold not does LIS the Although activities. monitoring routine current from forged be could od quired to achieve elimination. By merging DHIS and NHLS CDW data, a streamlined and efficient meth targetsreportingPMTCT improvementtimely of for and scope accurate Therereundoubtedly the is in The wayforward closure.pdf (available uponrequest from thecorresponding author),and declare noconflictofinterest. Competing interests: The authors completed the Unified Competing Interest form at www.icmje.org/coi_dis wrote thearticle.GGS,AHM, PB,SB,RN,MO,AP, DJJ,AEGprovided criticalrevision andfinalapproval. AP,MO, SB, PB, AHM, GGS, data. acquiredthe RN and AHM and GGS data. analysed/interpretedthe AEG DJJ, Authorship declaration: GGS, PB, SB, AEG conceived of study design and selected indicators for analysis. GGS Agency for International Development. A.H.M. acknowledges support from the Discovery Foundation [034203]. Funding: This analysis was supported by the President’s Emergency Plan for AIDS Relief through United States of HealthforDistrictInformationSystem(DHIS)data. al Health for Services Laboratory provision of test laboratory data and the South African National Department Acknowledgments: The authors gratefully acknowledge the CDW manager, Mrs Sue Candy, and the Nation 315
June 2017 •Vol. 7No. 1•010701 Towards eMTCT ofHIVinSA ------
VIEWPOINTSPaPErS PaPErSVIEWPOINTS June 2017 •Vol. 7No. 1•010701 Sherman etal. rEFEr ENCES 2 1 16 18 17 11 9 8 7 6 4 3 12 10 5 13 15 14 O Shisana Suppl 1):235-8.Medline:24893499 2014;104(3 J. Med Afr S Africa. South in programme mother-to-childtransmission of prevention the of plementation Sherman GG, Lilian RR, Bhardwaj S, Candy S, Barron P. Laboratory information system data demonstrate successful im National Department of Health. Policy and Guidelines for the Implementation of the PMTCT Programme. Pretoria: Na- 2016. pdf. Accessed:21February Health; 24 December 2014. Available: http://www.sahivsoc.org/upload/documents/HIV%20guidelines%20_Jan%202015. of Department National Pretoria: Adults. and Adolescents Children, in HIV of Management the and (PMTCT) HIV of Mother-to-Childof Prevention the Transmissionfor Guidelines Consolidated National Health. of Department National files/file/45/1335/877/pmtctguidelines_march2013_doh.pdf. 2016. Accessed:21February Available:2013. http://www.up.ac.za/media/shared/Legacy/siteHealth; of Department National Pretoria: 2013. March- National Department of Health. The South African Antiretroviral Treatment Guidelines 2013 PMTCT Guidelines: revised 2004. National Department of Health. National Antiretroviral Treatment Guidelines. Pretoria: National Department of Health; 2016. 21 February 2016. am/10665/112858/1/9789241505888_eng.pdf?ua=1&ua=1. Accessed:21February WHO; Geneva: syphilis. and HIV of (EMTCT) Transmission Validation:Mother-to-Childfor Processesof and Elimination Criteria World on Guidance Global Organization. Health 2016. category/176-reports-2015. Accessed:15February Available:http://www.health.gov.za/index.php/2014-03-17-09-09-8/reports/2015. Health; of Department National Pretoria: Africa. South Survey Prevalence HIV Sentinel Antenatal National 2013 The Health. of Department National 2016. uploads/pageContent/4565/SABSSM%20IV%20LEO%20final.pdf. Accessed:15February Available:http://www.hsrc.ac.za/2014. Council; Research Sciences Town:Survey,Human Cape Behaviour 2012. and . Accessed: 21 February 2016. aids/documents/legaldocument/wcms_125633.pdf. Accessed:21February tional Department of Health; 2008. Available: http://www.ilo.org/wcmsp5/groups/public/—ed_protect/—protrav/—ilo_ Available:2015. Health; http://www.sahivsoc.org/upload/documents/ART%20Guidelines%2015052015.pdf. Accessed: of Department National Pretoria: Adults. and Adolescents Children, in HIV of Management the and (PMTCT) HIV of Mother-to-Childof Prevention the Transmissionfor Guidelines Consolidated National Health. of Department National Technau doi:10.2471/BLT.11.092759 Africa. South KwaZulu-Natal, in intervention ity Child Health,UniversityoftheWitwatersrand; 2009. & Paediatrics of Department Johannesburg: Sciences. Health of Faculty children? of treatment and testing early tion, Woldesenbet able: http://www.mrc.ac.za/healthsystems/SAPMTCTE2010.pdf. 2016. Accessed: 15February Avail2010. Africa, - South Weeksin Six Mother-to-ChildPostpartum Transmissionat Measured Programme (PMTCT) Goga Goga dren_2010.pdf. 2016. Accessed:21February Statistics South Africa. Recorded live births 2014. Pretoria: Statistics South Africa; 2015. Available:2015. http://www.statssa.Africa; South Statistics Pretoria: 2014. births live Recorded Africa. South Statistics TEReport2012.pdf. Accessed:08March 2016. Health; 2010. Available:http://www.sahivsoc.org/upload/documents/Guidelines_for_Management_of_HIV_in_Chil2010. Health; of Department National Pretoria: Children. in HIV of Management the for Guidelines Health. of Department National mother-to-child transmission of HIV in South Africa. Barker P, Barron P, Bhardwaj S, Pillay Y. gov.za/publications/P0305/P03052014.pdf. 2016. Accessed:15February ness of WHO PMTCT Option A, South Africa, 2012-2013. Available:http://www.mrc.ac.za/healthsystems/SAPMTC2012-2013. Africa, South A, Option PMTCT WHO of ness Mphatswe W 2012. Available: http://apps.who.int/iris/bitstream/10665/75478/1/9789241504362_eng.pdf. Accessed: 8 March 2016. WHO; Geneva: Alive. Mothers Their Keeping and 2015 by Children Among Infections HIV New of Elimination the World Health Organization. A Short Guide on Methods Measuring the Impact of National PMTCT Programmes Towards QAD.0000000000000718 line:25469521 Africa. South in study national a of results diagnosis: HIV AE, Jackson AE, Dinh AE, K-G. Can a routine peri-partum HIV counselling and testing service for women improve access to HIV preven T Rehle , , Mate SA, Jackson D doi:10.1097/QAI.0000000000000460 TH, Jackson TH, DJ, Lombard , Simbayi , KS, Bennett B
, Goga DJ; SAPMTCTE study group. Evaluation of the Effectiveness of the National Prevention of PreventionEffectivenessof National group.the the study of of SAPMTCTE Evaluation DJ; K Zuma LC, CJ; SAPMTCTE study group. Early (4-8 weeks post-delivery) Population-level Effective , Ngidi H AE, Crowley S doi:10.7196/SAMJ.7598 The role of quality improvement in achieving effective large-scale prevention of S Jooste , , Reddy J , Doherty T 316 N Zungu , , Barker
Bull World Health Organ. 2012; Organ. WorldHealth Bull AIDS. 2015;
PM, et al. Improving public health information: a data qual , Mogashoa
, et al. South African National HIV Prevalence, Incidence Incidence Prevalence, HIV National African South al. et , J Acquir Immune Defic Syndr. 2015 Syndr. Defic Immune Acquir J 29 Suppl 2: 04 Aalbe http://apps.who.int/iris/bitstre Available: 2014. MM, et al. Missed opportunities for early infant S137- 43. www.jogh.org Medline:26102624 90: 176- • doi:10.7189/jogh.07.010701 82. Medline:22461712 ; 68: e26 doi:10.1097/ - 32. Med------www.jogh.org health: gapsandopportunities reproductive, maternal,neonatalandchild Improving coveragemeasurementfor material. The onlineversionofthisarticlecontainssupplementary Electronic material: supplementary Measurement forMNCH Group forImprovingCoverage Stanton * 2 1 Melinda KMunos USa Baltimore MD 615 NorthWolfeStreet Health Johns HopkinsBloombergSchoolofPublic Institute forInternationalPrograms Melinda Munos Correspondence to: [email protected] Individual membersarelistedin Stanton–Hill research,LLC, Institute forInternationalPrograms,Johns acknowledgements Moultonborough, NH,USa Baltimore, MD,USa Hopkins BloombergSchoolofPublicHealth, 21205
• doi:10.7189/jogh.07.010801 2
, JenniferBryce
1 , CynthiaK 1 *
; theCore
tral to assessing progress toward national and international health goals health international and national toward progress assessing to tral Regular monitoring of population–based coverage levels for RMNCH is cen and program decisions. timely,of policy set their mum high–quality,inform to data representative ernments and development partners will generate or have access to a mini (RMNCH) and for accountability rest on gov the assumption that country health child and newborn, maternal, reproductive, for agendas global on accountability in women’s and children’s health [ emphasis new a is welcome Also coverage. of levels equitable and tained, sus high, at effectiveness proven of interventions delivering on focused [ welcome are children and newborns women, among mortality reducing in investment for calls New sible foruseatscaleinLMICs. age measures and development of new measurement approaches fea cover of validation on focused RMNCH, for measurement coverage improve to actions of sets three propose we findings, our on Based birth. of time aroundthe particularly interventions, provenRMNCH Conclusions We found important gaps in coverage measurement for linking provider assessmentstohouseholdsurveys. by measure to feasible highly be would these of 13 and ity–based, one of the interventions not measured in household surveys are facil but All represented. poorly were periods intrapartum and antenatal, periconceptional, The surveys. household standard in measured are Results toprovidesurveys coverage. estimatesofintervention household to assessments provider linking of feasibility the sessed coverage, readiness, or quality. For facility–based interventions, we as intervention of measures generate tools these whether determine to vider assessments used in low– and middle–income countries (LMICs) cific mortality and stillbirth. We reviewed household surveys and pro of care for which there is evidence of effectiveness against cause–spe Methods with provider assessments. surveys approachhousehold measurementnovel linking coverage to spectrum of RMNCH interventions, and prioritize interventions for a tent to which current approaches to coverage measurement cover the the current state of coverage measurement for RMNCH, assess the ex ress toward health goals. The objectives of this review were to describe nal, neonatal, and child health (RMNCH) is central to assessing prog Background Regular monitoring of coverage for reproductive, mater Fewer than half (24 of 58) of included RMNCH interventions We included 58 interventions along the RMNCH continuum 317
1 ], especially to the extent that they are tightly tightly are they that extent the to especially ], June 2017 •Vol. 7No. 1•010801 2 ]. Taken together, the global journal of
health ------
VIEWPOINTSPaPErS PaPErSVIEWPOINTS June 2017 •Vol. 7No. 1•010801 Munos etal. RMNCH interventions. A second objective is to determine the RMNCH interventions for which linking linking which for interventions RMNCH the determine to is objective second A interventions. RMNCH ment and assess the extent to which current approaches to coverage measurement cover the spectrum of measure coverage in gaps identify to is analysis this of objective One care. of continuum RMNCH the In this article we present an analysis of the current state of coverage measurement for interventions across ( easy not is coverage ulation–based sensitive to changes in program effort, and be reliable across settings and over time. But measuring pop be population, a in coverage of measure valid a provide must decisions guide that estimates Coverage population coverage(hereafter referred toasa“linkingapproach” tocoveragemeasurement) [ of estimates support to order in providers service by provided interventions the of assessments to linked be surveys, household through collected occurred, careseeking where of reports that recommendations to led has consideration This prescribed). was drug which (eg, intervention the of details clinical specific the about whether they or their child received some interventions, especially when the caregiver does not know There is increasing recognition that caregivers cannot report accurately during a household survey interview nominator for a coverage indicator is representative of all women or children who need an intervention. Box 2. Box 1. 1 (Box data toa“goldstandard” ortruevalue,where available. measurementcompares errorsassesses and systematically biases which and study analytic an of means by ally gener is This purpose. intended its fulfils measure a which to Validationextent the study.of assessment An erage. Validity. A measure of whether an indicator provides an unbiased measure of true population intervention cov age across samples,mosttypicallythoughtofastheprecision ofapointestimate. Reliability. A measure of whether an indicator provides a consistent measure of population intervention cover provider,a service includingappropriate diagnosisandtreatment. Quality of care. tion toindividualsinneed. the presence of the drugs, necessary commodities, and trained and staffsupervised to administer the interven Readiness. A measure of whether a service provider is prepared to provide an intervention, taking into account ventions deliveredproviders. byservice the responses. of more or one of desirability social the and heaping, date or age or error recall to leading periods recall long Many factors can contribute to information bias, including poor question wording (eg, non–neutral questions), estimate. point the of over–estimation or under– in result can denominator.and or non–random merator is It bias. Information bias Information occurs when there is systematic error in providing information on the nu tion error. recall period, question wording, and type of information the respondent is asked for can all contribute to informa random and increases the variance of a coverage estimate but does not affect the point estimate. The length of the error is Information misclassification. answer,potential the in know resulting not do or question the understand Information error. Information error occurs when survey respondents provide a response even when they do not numerator. the in individuals of identification the affect may bias and error Information intervention. that received who and intervention an of need in are who individuals include should numerator numerator.The the Defining in (ornotin)thedenominator. being as individuals of misclassification in result can bias erroror Information disease. of diagnosis a or birth, pregnancychild as or such event an sex, and/or age on based identified be may individuals These tervention. Defining the denominator. The denominator should include only those individuals who are in need of an in sively discussedbyEiseleandcolleagues[4]. These key issues, and other sources of error in survey measurement of intervention coverage, have been exten inter the of assessments with sought was care where of reports caregivers’ link that Studies studies. Linking it (usuallymeasured inaprobability sampleofthepopulation). Intervention coverage. Key issues in measuring intervention coverage Key issuesinmeasuringintervention Definition ofterms provides definitions for “intervention coverage” and other terminology used in this paper) [ paper) this in used terminology other and coverage” “intervention for definitions provides
A measure of whether an individual in need of an intervention received that intervention from The proportion of a defined population in need of an intervention that actually receive 5 [ 2) Box 318
]. A particularly challenging issue is ensuring that the de the that ensuring is issue challenging particularly A ]. www.jogh.org • doi:10.7189/jogh.07.010801 3 ]. ------3 ]. ]. - - - www.jogh.org • doi:10.7189/jogh.07.010801 feeding, under–five,andcross–cutting environmental. postnatal, intrapartum, antenatal, (reproductive), periconceptional interventions: of groups following the for separately issues Wemeasurement unethical. consider considered being evaluations such to led tice and an evaluation of effectiveness has not been conducted, or where the lack of clinical equipoise has sionally it is based on consensus among experts, for example where interventions are established in prac base. Typically, this evidence is a systematic review of the published literature on effectiveness, but occa evidence underlying the describe that articles peer–reviewed published of references the lists appendix added based on evidence of their effectiveness in reducing under–five morbidity and mortality [ mortality and morbidity under–five reducing effectivenessin their of evidence on based added were interventions sanitation Waterand cleansing. cord umbilical for chlorhexidine and breastfeeding, newborn care” intervention was broken into its component practices, including thermal care, immediate or sleeping under a bednet (often treated as interventions for global monitoring purposes). The “essential breastfeeding exclusive of practice the as such behaviors, and interventions biomedical consideredboth METHODS action stepstoimprove themeasurement ofcoverageforMNCHinterventions. and propose gaps defining for basis We a care”). as of findings (“quality our delivery synthesize vention inter of quality the and/or intervention an deliver providerto readinesshealth of measurement the and coverage intervention of measurement direct Weboth feasible. assess and needed most are approaches internationally agreed–upon indicatorsformonitoringprogressinternationally inRMNCH. programmesBoth provideinterventions. sanitation for and estimates water of coverage the as well as tices, biomedical interventions, the survey programmes measure the prevalence of behaviors such as feeding prac review.this addressedby categories the all including RMNCH, of of coverage the measuring to addition In continuum the along indicators coverage of number increasing an included have programmes survey two revised through consultative processes that include stakeholders at global and country level. Over time, the questionnaires in 109 countries and since MICS 1995. has Survey carried out are279 surveys defined and 1985, since countries 91 in surveys representative nationally review.325 than morecoordinated has DHS survey–based coverage estimates used in global databases [ for sources main two the and health, and population on programmes survey household international est graphic and Health Surveys (DHS) and Multiple Indicator Cluster Surveys (MICS) [ with a representative sampling design that provide data at national scale and at regular intervals, the Demo- surveys only includes review This calculated. be can estimates uncertainty which for and population tire resentative sample of the population, and thus provide measures of coverage that take into account the en dle–income countries. These surveys are particularly valuable because they typically seek to interview a rep Household surveys are the major source for population–based intervention coverage data in low– and mid- Population–based coveragedata design suchassamplingstrategiesanddetailedsamplesizeissues. survey of details practical the address not does review The goals. RMNCH toward progress their track to seeking countries most for source data useful a not are and country) a within area sub–national a ly studies) are not included, as these surveys typically provide data for only one country (or more common efficacy or effectiveness for conducted surveys special example, (for surveys bespoke specialized, More administered regularly on a large scale (generally at national level) in low– and aremiddle–income countries. that sources other and surveys from data population–based considers only review vention).This who receivedviduals in need of an intervention it) and denominator (all individuals in need of an inter (indi numerator the both on providerepresentativeinformation sourcesmust data included surement, mea coverage For estimates. coverage produce to data careseeking with linked be could that data care of quality or readiness interventions, facility–based for and, data coverage population–based of sources for the intervention (facility–based, community–based, outreach, and/or behavioral), and identify current delivery of review,mode(s) this possible by addressedthe indicate intervention we life–saving each For Types ofdatathatare thefocusofreview [ Women’s interventions for these identifying for point Children’sstarting and a provided Health is clear evidence of effectiveness. The list of interventions included in the Global Investment Framework against major causes of maternal, newborn, and under–five mortality and stillbirths, and for which there careareacrossof thatdirected continuum RMNCH interventions the life–saving reviewon This focuses includedinthereview Interventions 319
9 ]. We consider both survey programmes in this Improving coveragemeasureforreproductiveandMNCH June 2017 •Vol. 7No. 1•010801 7 , 8 ]. These are the larg- 6 1 ]. The The ]. ]. We]. ------
VIEWPOINTSPaPErS PaPErSVIEWPOINTS June 2017 •Vol. 7No. 1•010801 Munos etal. whether they assessed readiness, observation–based quality of care, or neither.care,or of to quality able not readiness,observation–based assessed Interventions they whether determine to assessments provider these from questionnaires the reviewed we intervention, each For (TableFacility Survey) 1). gation of Quality (QIQ), and WHO’s IMCI quality of care assessments (previously the IMCI–MCE Health (SPA),sessment Evaluation’sMEASURE (R–HFA)Investi- Assessments Quick Facility and Health Rapid (WHO)’s Service Availability and Readiness Assessment (SARA), the DHS Program’s Service Provision As Organization the Health criteria: World inclusion our met that assessments provider five Weidentified andprovidersurveys assessments[11]. [ methods survey ity To identify provider assessments meeting these criteria, we hand–searched a 2009 review of health facil in this review. drugsandcommoditiestodelivertheinterventions necessary lected by readiness assessments; we included any assessment that collected data on the availability of the special assessments conducted for a specific study. There was substantial variation in the type of data col as well as assessments, single–country or one–time Weexcluded scale. national at and countries, tiple to include assessments of the provision of RMNCH interventions that are administered regularly, in mul- or without an independent assessment of the client’s need for the intervention. For this review, we sought cord information on readiness components, and may also include observations of service provision with the intervention, but readiness variables are often used as proxies for quality. Health provider surveys re of need in individuals by received actually was intervention an whether of assessment servation–based requireMeasurementsquality need. of in ob individuals an to staffvised intervention the administer to ers. We define readiness as the presence drugs, commodities, of and/or the trained necessary and super work health community and facilities, referral centers, health include may which providers– health of census or throughsurvey collected carea aretypically of provider quality readinessand service on Data Readiness andqualityofcare data considered asasource coveragedataforthepurposesofthisreview. forintervention areand indicators, not coverage and outcome key tracking for recommendedsettings been many not in routine data may be out of date, or may only be updated irregularly. For these reasons, routine data have poor data quality and completeness, and do not include important variables needed to assess equity. Some by characterized also are countries middle–income and low– most in systems health Routine collected. terest are simply not available through routine data, because the numerator, denominator, or both are not in of indicators RMNCH Many days. health child or activities community–based through and facilities in both delivered be may that immunizations or A vitamin like services for especially over–counted, be may Numerators whole. a as population representthe not thereforedo and system, health the with tact However, routine data also have important limitations. Denominators are limited to those who are in con from respondent recall inhouseholdsurveys. ascertained be can than detail greater in services on information provide to potential the have data tine availability at a relatively low cost, on a continuous basis, and at facility or district level. In addition, rou tial for use in estimating RMNCH intervention coverage. Potential advantages of routine data include their Routine data collected via the health system or by implementing programmes may also have some poten- Routine healthsystemandprogram data dicator is measured and calculated, generally expressed as an interval of time preceding the survey interview. tor. We alsonotedthereference periodforthecoverageindicator, thatis,thetimeperiodoverwhichin termine whether providedthe surveys measures of the numerator and denominator for the coverage indica We reviewed the questionnaires from MICS Round 5 and DHS Phase 6 for each RMNCH intervention to de possible toestablishadenominatoronlyforeasilyrecognizable symptomssuchasdiarrhea. be would it symptoms, recognitionspecific or requiringof diagnosis whereasa treatment for interventions survey,household a appropriatein an denominator establish to possible be generally would pregnancy),it (eg, conditions other or age on based targeted interventions preventive Wefor diagnosis). that considered particular a with children to only or range age particular a within children all to given be to is tervention sessment wasbasedontheindicationsforreceiving(forexample,whetherin aparticularintervention as Our intervention. the of need in population denominator,the coverage ie, the establish to survey h
] as well as the presentations from a technical consultation on linking household presentationshousehold the fromlinking as on well consultation as technical 10] a 320 www.jogh.org • doi:10.7189/jogh.07.010801 ------†Interviewer asksaboutavailabilityofguidelines/toolsbutdoesnotasktoseethem. †Interviewer *One healthworkerinfacilityisaskedtoreportforallworkersfacility. ontraining/supervision Assessment Rapid HealthFacilityAssessments,QIQ–QuickInvestigationofQuality, IMCIQoC–IntegratedManagementofChildhoodIllnessQualityCare SARA – Service Availability and Readiness Assessment, SPA – the DHS Program’s Service Provision Assessment (SPA), R–HFA – MEASURE Evaluation’s www.jogh.org Table 1. Case scenarios/vignettes Competency: Exit interview withpatient/caregiverExit interview Re–exam Observation of service provision ofservice Observation Quality ofcare: Availability ofdrugs/commodities Availability ofguidelines/tools Supervision Training Readiness: Geographic scope Data collectedthrough selectedprovider assessments • doi:10.7189/jogh.07.010801 data sources forcoverage, readiness, andqualityofcare foreachintervention. Table 2 presents the 58 included interventions, organized across the continuum of care, and the current r questionnaire.signs ofsepsis)usingasurvey measured in household surveys due to the difficulty in establishing a valid denominator (newborns with sis with antibiotics was categorized as infeasible, because careseeking for neonatal sepsis is not currently trained staff) is currently collected in provider assessments. On the other hand, treatment of neonatal sep and pressure apparatus, blood heating, for flame and protein/aceticacid urine for dipstick sulfate, sium is measuredery) and via readinesshousehold surveys, to (availability deliver of the magne intervention deliv- facility and consultations (ANC careseeking because feasible highly as categorized was eclampsia ly feasible candidates for a linking study. For example, magnesium sulfate for treatment of pre–eclampsia/ study. Interventions for which both careseeking and readiness could be measured were considered high quality of care might be assessed through wereobservation considered potential candidates for a linking but measured be not could readiness which study.for linking Interventions a for infeasible considered ment. Interventions for which either careseeking or readiness/quality of care could not be measured were providerassess througha measured be could intervention that for care of quality or readiness whether survey,household througha obtained be could intervention the for and data careseeking whether ering consid vey.by feasible/potential/infeasible, highly as categorized and assessed was linking of Feasibility provider was “ready” to provide the intervention, information that is not available from a household sur the whether determine to assessments provision service from data and provider) particular a from care numerator makes use of both population–based data (to estimate the number of individuals who sought The denominator). (the intervention the of need in those representativeof a estimate generate to survey level or aggregate community level. This approach makes use of population–based data from a household through an approach linking household survey data to provider assessment data, at either the individual coverage intervention population–based measure to potential the assessed we intervention, each For Feasibility forlinkingstudy to measurement inprovider assessments. ventions seeking to influence the behavior (eg, counselling on breastfeeding practices) may be amenable inter although assessments, providerthroughmeasurement to themselves lend not do behaviors many are limited to use or ownership of a commodity, such as insecticide–treated bednets (ITNs). In addition, tions, such as the availability of improved water sources and improved sanitation, and interventions that interven sector non–health include These excluded. were assessments provider through measured be ESULTS apeo essSml rcnu Sample Sampleorcensus Sample orcensus rmnch, hIv, t SARA X † * * uberculosIs dIseases , n on – SPA communIcable X X X X X X 321
–F IMCI–QoC R–HFA Improving coveragemeasureforreproductiveandMNCH X X X X X c hIld
health ( curatIve June 2017 •Vol. 7No. 1•010801 Sample ) X X X X X X X * F amIly Sample QIQ
X X X X plannIng * ------
VIEWPOINTSPaPErS PaPErSVIEWPOINTS June 2017 •Vol. 7No. 1•010801 Munos etal. Table 2. RMNCH interventions and data sources InterventIon mode oF delIvery could currently mea- InterventIons FacIlIty–based InterventIons household sured In mIcs/ measured In survey dhs? dhs/mIcs establIsh Reference Currently measured in Source of provider data, R Feasible for populatIon period standard provider assess- (readiness), linking study In need? ments?, R (readiness), O (observation) O (observation), N (no) Periconceptional: Contraception Facility, community, Yes Yes 3 y/5 y R, O SARA (R), SPA (R,O), QIQ (R,O) Highly feasible outreach Periconceptional folic acid supplementation Facility–based Yes No N Infeasible Safe abortion services Facility–based No No N Infeasible
Post abortion case management Facility–based No No R Infeasible Ectopic pregnancy case management Facility–based No No N Infeasible Antenatal: Tetanus toxoid vaccine for pregnant women Facility and outreach Yes Yes 2–5 y R, O SARA (R), SPA (R,O), R–HFA (R) Highly feasible Intermittent preventive treatment of malaria in Facility–based Yes Yes 2–5 y R, O SARA (R), SPA (R,O), R–HFA (R) Highly feasible pregnancy Syphilis detection and treatment in pregnancy Facility–based Yes No R, O SARA (R), SPA (R,O), R–HFA (R) Highly feasible Calcium supplementation for prevention and treatment Facility–based Yes No N Potential of eclampsia and pre–eclampsia Multiple micronutrient supplementation Facility–based Yes No N Potential
322 Balanced energy supplementation Facility and outreach Yes No N Infeasible Detection and management of diabetes in pregnancy Facility–based No No R SARA (R), SPA (R) Highly feasible Pregnant women sleeping under an insecticide–treated Behavior Yes Yes Last night bednet Treatment of malaria in pregnant women Facility–based No No R SARA (R), SPA (R) Infeasible Management of pre–eclampsia with magnesium sulfate Facility–based No No R SARA (R), SPA (R) Highly feasible Detection and management of fetal growth restriction Facility–based No No N Infeasible Anti–retroviral therapy for pregnant women Facility–based No No R SARA (R), SPA (R) Highly feasible Prevention of mother to child transmission of HIV Facility–based No No R SARA (R), SPA (R) Highly feasible Intrapartum: Skilled birth attendant Facility, community Yes Yes 2–5 y R SARA (R), SPA (R) Highly feasible (Service contact) www.jogh.org Clean birth practices Facility, community Yes No R SARA (R), SPA (R) Highly feasible Immediate assessment and stimulation for newborns Facility, community Partial No R SARA (R), SPA (R) Potential Neonatal resuscitation Facility–based No No R SARA (R), SPA (R), R–HFA (R) Highly feasible
• doi:10.7189/jogh.07.010801 Antibiotics for preterm premature rupture of membranes Facility–based No No R SARA (R), SPA (R) Highly feasible Antenatal corticosteroids for preterm labor Facility–based No No R SARA (R), SPA (R) Highly feasible* Magnesium sulfate for eclampsia Facility–based No No R SARA (R), SPA (R) Highly feasible Active management of the third stage of labor Facility–based Yes No R SARA (R), SPA (R), R–HFA (R) Potential Induction of labor for 41+ weeks Facility–based No No N Potential Postnatal: Postnatal visit for moms and for babies Facility, community Yes Yes 2–5 y N Highly feasible (service contact) Immediate initiation of breastfeeding Behavior occurring in Yes Yes 2–5 y N Potential facility or community Thermal care Facility, community Yes Not currently; R SPA (R) Potential likely in future www.jogh.org Table 2. Continued InterventIon mode oF delIvery could currently mea- InterventIons FacIlIty–based InterventIons household sured In mIcs/ measured In • doi:10.7189/jogh.07.010801 survey dhs? dhs/mIcs establIsh Reference Currently measured in Source of provider data, R Feasible for populatIon period standard provider assess- (readiness), linking study In need? ments?, R (readiness), O (observation) O (observation), N (no) Clorhexidine for umbilical cord cleansing Facility, community Yes No R SARA (R), SPA (R) Highly feasible Kangaroo mother care Facility, community No No R SPA (R) Potential Feeding: Breastfeeding Behavior Yes Yes 24 h Complementary feeding Behavior Yes Yes 24 h Under–five: Vitamin A supplementation Facility, outreach Yes Yes 6 mo R SARA (R), SPA (R) Infeasible Polio vaccine Facility, outreach Yes Yes 5 y R SARA (R), SPA (R) Infeasible BCG vaccine Facility–based Yes Yes 5 y R SARA (R), SPA (R) Infeasible Meningitis vaccine Facility, outreach Yes No N Infeasible Pentavalent3/DPT3 vaccine Facility–based Yes Yes 5 y R SARA (R), SPA (R), R–HFA (R) Infeasible Pneumococcal vaccine Facility–based Yes Yes 5 y R SARA (R) Infeasible Rotavirus vaccine Facility–based Yes Yes 5 y R SARA (R) Infeasible Measles vaccine Facility, outreach Yes Yes 5 y R SARA (R), SPA (R), R–HFA (R) Infeasible Antibiotics for neonatal sepsis Facility, community No No R SARA (R), SPA (R), R–HFA (R) Infeasible 323 Oral rehydration solution for diarrhea Facility, community Yes Yes 2 weeks R, O SARA (R), SPA (R,O), IMCI (R,O), Highly feasible† R–HFA (R,O) Zinc for diarrhea Facility, community Yes Yes 2 weeks R, O SARA (R), SPA (R,O), IMCI (R,O) Highly feasible Antibiotics for dysentery Facility–based No No R, O SARA (R), SPA (R,O), IMCI (R,O), Highly feasible R–HFA (R,O) Antibiotics for suspected pneumonia Facility, community No Yes 2 weeks R, O SARA (R), SPA (R,O), IMCI (R,O), Highly feasible R–HFA (R,O) Artemisinin combination therapies for malaria Facility, community No Yes 2 weeks R, O SARA (R), SPA (R,O), IMCI (R,O), Highly feasible Improving coveragemeasureforreproductiveandMNCH R–HFA (R,O) Vitamin A treatment for measles Facility–based No No R, O SARA (R), SPA (R,O), IMCI (R,O), Infeasible R–HFA (R) Management of severe malnutrition Facility–based No No N Infeasible
June 2017 •Vol. 7No. 1•010801 Cotrimoxazole for HIV Facility–based No No R SARA (R), SPA (R) Infeasible Paediatric anti–retroviral therapy for HIV Facility–based No No R SARA (R), SPA (R) Infeasible Environmental: Use of improved water source Behavior Yes Yes NA Use of improved sanitation Behavior Yes Yes NA Hygienic disposal of children's stool Behavior Yes Yes Last stool Handwashing Behavior Yes No Insecticide–treated bednet ownership Outreach Yes Yes NA Insecticide–treated bednet use Behavior Yes Yes Last night SARA – Service Availability and Readiness Assessment, SPA – the DHS Program’s Service Provision Assessment (SPA), R–HFA – MEASURE Evaluation’s Rapid Health Facility Assessments, QIQ – Quick Investiga - tion of Quality, IMCI – WHO integrated management of childhood illness (previously the IMCI–MCE Health Facility Survey) *A recent study has called into question the benefits of antenatal corticosteroids in low– and middle–income countries [ 12]. †In settings where ORS is primarily distributed through health facilities and community health workers.
VIEWPOINTSPaPErS PaPErSVIEWPOINTS June 2017 •Vol. 7No. 1•010801 Munos etal. require caregiver or respondent knowledge of specific clinical details such as a diagnosis. The exception exception The diagnosis. a require as such details clinical caregiverspecific respondent of or knowledge In general, we found that household surveys are not good sources of coverage data for interventions that Gaps incoveragemeasurement for measurement through anapproach toprovider linkinghouseholdsurveys assessments. candidates good be would and throughmeasuredprovidercurrentlyassessments are interventions tum have issues of data quality and completeness. However, we also found that many antenatal and intrapar may be measured through routine or program data, such data often lack an appropriate denominator and interventions these of some Although surveys. household population–based through measured rently lifesaving interventions in the periconceptional, antenatal, intrapartum, and postnatal periods are not cur many that found however,we interventions, environmental and health child Beyond interventions. tal currently measured through large, nationally representative household surveys, as are many environmen are months 1–59 aged children to targeted interventions many that is review this of finding positive A andprovidering householdsurveys assessmentsto provide coverage. estimatesofintervention current approaches to coverage measurement, and to assess the potential for using a new approach link in gaps identify to order in means, what by and measured, currently are interventions which out map erage of life–saving RMNCH interventions at national scale and on a regular basis. This review sought to intervention coverage to achieve mortality reductions, there is a critical need to measure population cov of importance the of awareness and RMNCH for accountability to attention global increasing the Given DISCUSSION providerbe feasibleifobservation–based assessmentswere used. might approach a linking indicators, postnatal and intrapartum five another For intervention. der–five un one and postnatal, one intrapartum, six antenatal, five – surveys household in measured currently not are which of 13 interventions, 22 for feasible highly be would approach linking a Wethat estimate cility,areacatchment a within facility.particular a to survey household the in everyone associating by or fa particular a to survey household the in individual each matching by either linked, be must then tion informa sourcesof two These providerassessment. health through a intervention) the of delivery of ity through a population–based household survey and provider readiness to deliver the intervention (or qual careseeking measure to ability the requires approach linked a using coverage intervention Estimating household surveys Feasibility ofmeasuringcoveragethrough linkedprovider assessmentsand aswellthequalityofservices. complete assessmentofhealthworkertrainingandsupervision, of these data. These two assessments provide data for most of the same interventions. SPAs provide a more agement of fetal growth restriction). The WHO’s SARA and the DHS Program’s SPA are the main sources antenatal in nature (for example, safe abortion services, calcium supplementation, and detection and man Those interventions not currently addressed by provider assessments are primarily periconceptional and readiness for 27 interventions, and readiness and observation–based quality of care for 10 interventions. that Of can the be 49 deliveredinterventions at a health facility, provider assessments currently measure Measurement ofreadiness andqualityofcare areconceptional (oneoffive)interventions measured through MICSandDHS. Similarly,contacts. service beyond surveys peri and 13) of proportionrelatively(two low a antenatal of in measured is period this for interventions included the of none yet and babies, and women for riod pe risk highest the as out stands period intrapartum the care, of continuum the Along infections. natal der–five category, however, there are gaps with respect to measuring treatment of malnutrition and neo un the Within DHS. and MICS in measured interventions environmental six of five and interventions the measured interventions fall in the under–five and environmental categories, with 11 of 18 under–five measure careseeking, ie, skilled birth attendance and postnatal visits, rather Many than of interventions. actually and coverage intervention for proxies are Two interventions outreach. measured via the two of but one (handwashing) are delivered at health facilities; five can also be delivered at community level and household surveys (DHS or MICS). Of those interventions not measured through household surveys, all regular through measured currently are interventions included the of Twenty–four,half, than fewer or cover
324 www.jogh.org • doi:10.7189/jogh.07.010801 ------www.jogh.org • doi:10.7189/jogh.07.010801 postnatal period with mixed results [ results mixed with period postnatal immediate and intrapartum the for measures coverage of range a of validity the explored have studies sures, and for which interventions alternative measurement approaches should be explored. A few recent ventions can household providesurveys accurate, precise, and reliable population–based coverage mea- inter health which understand researchbetter for to urgentneed an is There interventions. RMNCH of coverage progressin track to used be to are data survey if importance central of is countries across and sured through such surveys [ ment of fever with an ACT [15], other interventions such as antibiotics for pneumonia are not well mea- though household surveys can provide accurate coverage measures for some interventions, such as treat al that suggest they exist, validity indicator on data Where surveys. household through collected data [3 accuracy the on data of lack the is gap important Another critical toensuringprogress inRMNCH,andisnotpossibleatpresent. the coverage of interventions that protect against common causes of maternal and neonatal deaths is thus [ birth ing follow day the in and childbirth during occur deaths newborn and maternal most babies: and mothers of health the for birth after immediately aroundand period the of importance the given concern ticular par of is gap This interventions. these for estimates, coverage precise provide to sizes sample adequate antibiotics for preterm premature rupture of membranes), as household surveys typically cannot achieve not suited to measuring coverage of needed interventions by small very numbers of individuals (such as branes, which cannot be readily established through a questionnaire.survey Household aresurveys also surveys, because they require a diagnosis, such as pre–eclampsia or preterm premature rupture of mem ventions. Moreover, many of these interventions are not appropriate for measurement through inter household neonatal and maternal, reproductive, most of coverage the in progress track to used be currently therefore and cannot periods, these measurecareseekingfor primarily surveys Household birth. of time mothers. There is a clear measurement gap for interventions delivered during pregnancy and around the by recalled and recognized be easily can that symptoms on based careseeking assessing and terventions be complicated and expensive. Household surveys are generally well–suited for measuring preventive in their use in large–scale surveys is restricted to a few indicators and, where they are available, their use can is conditions for which biomarkers are available. Although new biomarker tests are increasingly available, be measured through a survey questionnaire. Efforts to validate survey–based measures of RMNCH in RMNCH of measures survey–based validate to Effortsquestionnaire. survey a throughmeasured be should continue Household to surveys be used as a source of coverage data for those indicators that can Action stream 1:householdsurveys streams ofactiontoimprove coverageforRMNCH. theavailability andqualityofdataonintervention researchRMNCH community.the practice for and priority a be must and We recommendthreeparallel national levels is essential to achieving reductions in maternal, newborn, and child deaths and stillbirths, Providing valid, population–based estimates of coverage for at RMNCH national interventions and sub– Research andpracticeagendas (see estimates effectiveness vestment Framework for Women’s and Children’s Health, In and Global included the in only those interventions on based with was published interventions of list Our limitations. of number a has review This Limitations attendance) hasnotyetemerged fortheintrapartumperiod. health informationsystemsimprove. routine and revised are instruments collection data as time over filled be will here identified gaps the of Wesome time. that in expect point particular a at opportunities and gaps the of snapshot a provides review This evolve. to continue data routine and surveys provider and household Finally,that note we ies willprovide additionalinformationaboutthefeasibilityoflinkingforvariousinterventions. going efforts to implement the linking approach using existing and new data. When complete, these stud feasibility, it is possible that another group might come to somewhat different conclusions. There are on of level each for criteria clear establish to attempted we Although subjective. somewhat was tervention in each of coverage the estimate approachto linking a using of feasibility the assessing processfor Our sider whetherandhowtomeasure theircoverage. con to need ongoing an be will there time, over emerge interventions new As omitted. been have that developing, still is evidence of body the which for emerginginterventions particularly interventions, be ], and neonatal deaths represent a growing proportion of under–five deaths [ deaths under–five of proportiongrowing a represent deaths neonatal and 13], in S1 Appendix 16]. The question of whether coverage measurements are reliable over time ]. A clear alternative to measuring careseeking (ie, skilled birth birth skilled (ie, careseeking measuring to alternative clear A 17]. 325 Online Supplementary Document Supplementary Online Improving coveragemeasureforreproductiveandMNCH ], precision, and reliability of the coverage coverage the of reliability and precision, ], June 2017 •Vol. 7No. 1•010801 ). However, there may may However,there ). ]. Tracking14]. ------
VIEWPOINTSPaPErS PaPErSVIEWPOINTS June 2017 •Vol. 7No. 1•010801 Munos etal. explored promoting forinterventions thebehavior, suchascounsellingonbreastfeeding practices. or modeling, should be explored. This is true for behaviors as well, although a linking approach should be measurement approaches, such as the use of specialized surveys, biomarkers or proxies for the intervention, which are delivered primarily or entirely outside a facility, a linking a pproach is not feasible and alternative For those interventions for which household surveys do not provide accurate or reliable measurements, and measurement approaches fornon–facility–basedinterventions Action stream 3:alternative measures fortheseinterventions. develop guidelines for their implementation and a program to ensure the regular production of coverage rate and reliable measures of intervention coverage, the RMNCH research and practice community should accu provideboth to and cost reasonable at implement to feasible be to found are approachesthat For es (feasibility, cost,accuracy, andreliability). data, may also hold promise and should be assessed using the same considerations as linking approach routine of use the including interventions, facility–based for coverage measuring to approaches Other vider assessmentsshouldbetestedandcompared. produced through this approach. In addition, different approaches to linking household surveys and pro reliabilitymeasures and coverage accuracy of the as well as countries, middle–income and low– in scale national at cost and feasibility factors: following the address should approaches linking of Assessments which the validity of the survey–based indicator is questionable, including treatment of childhood illness. urgently. Linking approaches could also be valuable for indicators currently measured in surveys, but for pursued be providercareseekingmust throughvice on collected data to surveys assessments household vider, and are currently measured in provider assessments. Measurement approaches that link these ser pro service health a aredeliveredby requirediagnosis, they a because survey household a measuredin be cannot that those including surveys, household in measured not interventions RMNCH the of Many measurement approaches forfacility–basedinterventions Action stream 2:alternative measurement approachesmeasures coverage,alternative ofintervention shouldbeexplored. reliable or precise, accurate, provide not do surveys that indicates evidence the Where questionnaires. survey DHS and MICS of revisions future inform should efforts these of results The time. over ments measurecoverage reliabilityof the of assessments include must and continue must coverage tervention sure.pdf (available uponrequest from thecorresponding author),and declare noconflictofinterest. Competing interests: The authors completed the Unified Competing Interest form at www.icmje.org/coi_disclo and comments. review critical provided Group Core ICM the of members all and paper; the of draft first the wrote and analysis the conducted MM analysis; the of conceived Group Core ICM the of members The declarations: Authorship org/) Programs totheInstituteforInternational attheJohnsHopkinsBloomberg SchoolofPublicHealth. Measurement for MNCH Interventions” from the Bill & Melinda Gates Foundation (http://www.gatesfoundation. “Improvingon OPP1084442 Number Grant Development Global by funded was Coverage work This Funding: University); CindyStanton(Stanton–HillResearch, LLC). Hopkins (Johns Munos Melinda Tropical Medicine); & Hygiene of School (London MarchantTanya CEF); (UNI- (TulaneHancioglu Eisele Attila Thomas Edinburgh);University); of (University Campbell Harry versity); CEF); Fred Arnold (ICF International); Ann Blanc (The Population Council); Jennifer Bryce (Johns Hopkins Uni membership: MNCH for Measurement Coverage Improving for Group Core The Ethics: Thisstudydidnotinvolvecollectionoranalysisofanypersonaldataandrequire ethicalapproval. sponsibility forthedecisiontosubmitpublication. sis or decision to publish. The corresponding author had full access to all the data in the review and had final re Disclaimer: The funder reviewed and provided comments on the draft manuscript but had no role in the analy- his reviewhelpfultechnicalinputs. ofthemanuscript,andforhisvery GroveJohn thank to like would authors The Acknowledgments: from for Foundation Gates Melinda & Bill the
326 www.jogh.org Agbessi Amouzou (UNI- Amouzou Agbessi • doi:10.7189/jogh.07.010801 ------www.jogh.org rEFEr ENCES • doi:10.7189/jogh.07.010801 17 2 1 16 3 5 6 4 7 9 8 12 11 10 13 15 14 able: http://www.who.int/woman_child_accountability/about/coia/en/ Accessed:4Apr2017. World Health Organization. Commission on Information and Accountability for Women's and Children's Health. Avail 54. Medline:24263249 development by investing in women’s and children’s health: a new Global Investment Framework. Lancet. 2014;383:1333- Stenberg K, Axelson H, Sheehan P, Anderson I, Gülmezoglu AM, Temmerman M, et al. Advancing social and economic J Bryce lections.org/measuringcoverageinmnch. 2015. Accessed13January Available:2013. http://www.ploscolHealth. Child and Newborn, - Maternal, in Coverage Measuring Collections. PLoS line:23667331 Cairncross S Eisele pmed.1001423 action. for recommendations and strategies, the interpretation of intervention coverage estimates from household surveys. household from estimates coverage intervention of interpretation the zambique. validity of women’s self–report of key and maternal health newborn during interventions the peripartum period in Mo Stanton 2013; pneumonia. childhood of treatment correct measuring on Bangladesh and Pakistan in study validation diarrhoea. International suring diagnosis and treatment of childhood malaria from household surveys in Zambia. and MICS household surveys. Hancioglu A dex_24302.html. Accessed:3May2014. Survey. Cluster http://www.unicef.org/statistics/inAvailable: Indicator Multiple UNICEF. Monitoring: and Statistics Hazir Medline:23667337 Althabe F tion onStudyDesign.Baltimore, MD:2014. TechnicalAssessments: Service Consulta Health and Surveys Household Programs.Linking International for Institute Applied inLowandMiddleIncomeCountries[workingpaperWP–09–11].ChapelHill,NC: 2009. A Edward 2015; Lawn trial. cluster–randomised ACT the countries: middle–income and low–income in birth preterm to due ity eted strategy to implement antenatal corticosteroid treatment versus standard care for the reduction of neonatal mortal Eisele line:25280870 13, with projections to inform post–2015 priorities: an updated systematic analysis. systematic updated an priorities: post–2015 inform to projections with 13, survival. Liu L , Oza S T 10: 385: JE, Blencowe H TP, Silumbe K TP,Rhoda F Arnold , , Begum , B Rawlins CK, e1001422. Lancet. 2014; , Belizán T Matsubiyashi , 629- Int JEpidemiol.2010; PLoS One.2013;8: , Hogan D , Hunt C F Arnold , ICF. Available: Demographic andHealthSurveys. http://www.dhsprogram.com Accessed:3May2014. 39. doi:10.1371/journal.pmed.1001386 doi:10.1016/S0140-6736(14)61698-6 K A Blanc , DA, Cutts DA, , El Arifeen El , JM, McClure Medline:25458726 ukich J doi:10.1371/journal.pmed.1001417 Medline:23667339 , Y , Boisson S , Oza S , Perin J . Measuring coverage in MNCH: tracking progress in health for women and children using DHS DHS childrenusing and women progressfor tracking health MNCH: in in coverage Measuring . M Drake , 384: doi:10.1016/S0140-6736(13)62231-X A Hancioglu , B Fapohunda , 189- FT,J Keating , S You , Hamainza B
, Rudan I e60694. PLoS Med. 2013; , Khan , EM, Hemingway–Foday J , Bostoen K 205. M Anjos Dos , 39 D , Lee Suppl 1: Medline:24853593
Medline:23667427 doi:10.1016/S0140-6736(14)61651-2 AM, Huque AM, , Lawn H Newby , doi:10.1371/journal.pmed.1001422 R Ren , AC, Waiswa P S Becker , , Keating J , Curtis V
i193- PLoS Med. 2013 Med. PLoS JE, et al. Global, regional, and national causes of child mortality in 2000– D Cantor , 10: , Barros, 205. J Requejo ,
e1001391. 327 . A Comparative Analysis of Select Health Facility Survey Methods Methods Survey Facility Health Select of Analysis Comparative A . MH, Kazmi MH, , Fung , Bennett A , et al. Every Newborn: progress, priorities, and potential beyond Medline:20348121
AJ, et al. Measuring coverage in MNCH: total survey error and errorand survey total MNCH: in coverage Measuring al. et AJ, doi:10.1016/S0140-6736(14)60496-7 , Berrueta M L Chongo , doi:10.1371/journal.pone.0060694 IC, et al. Water, sanitation and hygiene for the prevention of , et al. Measuring coverage in MNCH: new findings, new new findings, new MNCH: in coverage Measuring al. et , Medline:23667333 ; 10 N , et al : e1001423 , et al. Measuring coverage in MNCH: a prospective a MNCH: in coverage Measuring al. et , Improving coveragemeasureforreproductiveandMNCH . Measuring coverage in MNCH: accuracy of mea , et al. Measuring coverage in MNCH: testing the the testing MNCH: in coverage Measuring al. et , , Mazzoni A doi:10.1093/ije/dyq035 . Medline:23667340
PLoS Med. 2013 Med. PLoS , etal.Apopulation–based,multifac-
doi:10.1371/journal.pmed.1001391 June 2017 •Vol. 7No. 1•010801
Lancet. 2015 Lancet.
PLoS Med. 2013;
; doi:10.1371/journal. 10: ; 385 e1001386 : 10: 430
PLoS Med PLoS e1001417. - 40
Lancet . . Med Med ------. .
VIEWPOINTSPaPErS PaPErSVIEWPOINTS June 2017 •Vol. 7No. 1•010901 Sundeep Gupta Bahie Mrassekh Paul aFreeman Jess Wilhelm Henry BPerry 1. rationale,methodsanddatabasedescription care inimprovingmaternal,neonatalandchildhealth: effectiveness ofcommunity–basedprimaryhealth Comprehensive reviewoftheevidenceregarding material. The onlineversionofthisarticlecontainssupplementary Electronic material: supplementary [email protected] USa Baltimore, MD21205 615 NorthWolfeSt. School ofPublicHealth Johns HopkinsBloomberg r Henry Perry Correspondence to: 5 4 3 2 1 Department ofGlobal Independent consultant, Medical Epidemiologist, The WorldBank, Department of International Washington, USa Washington, Seattle, Health, Universityof Seattle, Washington,USa Lusaka, Zambia Washington DC,USa Maryland, USa Public Health,Baltimore, Bloomberg Schoolof Health, JohnsHopkins oom E8537
1 , 1
,
4,5 3 , 2
, constrained settings. high–mortality,in resource– interventions evidence–based delivering for platforms based community– stronger building for rationale the provide to help will series, this in paper the Journal of Global Health. These findings, guide this review, that are included as the last effectivenessimprovingthe in on CBPHC published one of MNCH this follow that cles growingrapidly. arti of series a for basis the as createdreviewdatabase this The for serves is age of years 5 than younger children and neonates, mothers, of health the improve to regarding evidence The Conclusions effectivenessthe interventions community–based of The reviewers judgedthemethodologyfor90%of assessmentstobeadequate. ed most of the assessments, led by the United States Agency for International Development. of reports has grown markedly during the past decade. A small number of funders support projects were from three countries in South Asia: India, Bangladesh and Nepal. The number the of (26.6%) One–quarter years. 3 than less lasting projects for were assessments the of 62.9% and children, or women 5000 than fewer involved assessments the of half periods; implementing three or fewer interventions in relatively small populations for relatively brief projects Two–thirdsfor hygiene. were and (63.0%) water,sanitation assessments safe of of promotion and/or provision and illness; child and neonatal acute of treatment and agnosis promotion of healthy household behaviors and appropriate utilization of health services, di that were assessed concerned promotion or provision of good nutrition and immunizations; interventions health child and neonatal The treatment. and prevention HIV and; planning, family infections; maternal other and infection HIV for treatment and screening tendant, provision of antenatal care; promotion and/or provision of safe delivery by a trained birth at and/or promotion delivery; safe and pregnancy of signs warning about education cerned data extraction forms and resolved any differences. The maternal interventions assessed con two the compared reviewer third A assessment. each for form extraction data a completed review. the in inclusion reviewers for independent qualified Two assessments 700 Results tional status,orinmortality. morbidity,nutri serious in in changes or interventions evidence–based of coverage tion population–based indicators that defined some aspect of health status: changes in popula were review.review the the in in inclusion inclusion for qualified that measures Outcome for qualified out carried projectwas the effectivenessof the of assessment an which in ect the implementation of one or more community–based interventions or an integrated proj describe that Reports reviewed. also were journal) peer–reviewed a in published not but of Medicine database (PubMed). In addition, reports in the gray literature (available online Methods 12 What are theimplicationsofthesefindings? (4) used? were strategies implementation of kinds What (3) projects? these of outcomes the following questions: (1) What kinds of projects were implemented? (2) What were the addresses review The countries. middle–income and low– in MNCH improve to CBPHC used that projects) as collectively to (referred studies research and projects, programs, of uity effects. This is the first article in a series that summarizes and analyzes the assessments eq their with along level community the at strategies delivery their also but interventions specific of effectiveness the only not to attention particular gives gives review our but ers, oth- by summarized been has (MNCH) health child and neonatal improvingmaternal, in CBPHC effectivenessof the of Evidence households. to down even and communities into facilities health beyond services health curative and preventive extend programsto health by used approach an is (CBPHC) care health primary Community–based Background
166 reports were identified through a search of articles in the National Library 328 www.jogh.org • doi:10.7189/jogh.07.010901 global journal of health ------www.jogh.org • doi:10.7189/jogh.07.010901 and middle–incomecountries(LMICs). low– in evidence this of analysis and review comprehensive a of findings the highlights that series a of zakhstan in 1978, sponsored by the World Health Organization and UNICEF [ of the Declaration of Alma–Ata at the International Conference on Primary Health Care at Alma–Ata, Ka- “ Article Vofthe1978DeclarationAlma–Atastates following [ (MNCH) has been steadily growing over the past several decades [ health child and neonatal maternal, improve can interventions community–based that evidence The as close to the community as possible by members of a health team, including community health work- services care health rehabilitative and curative, promotive, preventive, providing involves Alma–Ata, of of ery medical at services health primary care centers. health Primary care, as defined by the Declaration The broad healthcare concept of primary articulatedinthisDeclarationwas muchmore thanthedeliv the keytoattainingthistarget aspartofdevelopmentin thespiritofsocialjustice.” carehealth productiveis Primary economically life. and socially a lead to them permit will that health of level a world community in the coming decades should be the attainment by all peoples of the world by the year 2000 of equate health and social measures. A main social target of governments, international organizations and the whole of Alma–Ata,whichcalledforHealthAllbytheYear 2000through healthcare primary [ tional Conference Health on CarePrimary at Alma–Ata, Kazakhstan and the now renowned Declaration successful pioneering CBPHC projects around the world and laid the groundwork for the 1978 Interna publication, WHO seminal the was cussions Ken Newell, Director of Strengthening of Health Services at WHO [ ficials at the World Health Organization (WHO), including Halfdan Mahler, then Director–General, and Jack Carl Bryant, Taylor, and William Foege, all of whom were members of the CMC, and high–level of Barrow,Nita Dame including time, their of visionaries health global involved discussions these 1970s, the high–mortality,In in settings. resource–constrainedpeople of health the improve best can grams World Council of Churches, which provided a framework and a forum for new thinking about how pro the of (CMC) Commission Medical Christian the of formation the to led This [19]. not did who people than healthier no were regularly hospital the used and to access easy had who people the that strated by hospital–oriented Christian medical mission programspopulations served around the world demon of surveys of series a time, that At serving. were they populations the of health the improving not were hospitals that recognition the following 1960s the in began movement care health primary global The The context healthcareal primary more generally, especiallyinlightoftheupcoming40 maries of portions of this evidence do exist [ sum excellent although MNCH, improving in CBPHC of effectiveness the regarding evidence of range broad the analyzing and accumulating systematically to given attention limited been has Tothere date, the failure toimplementandscaleupevidence–basedcommunity–basedinterventions. [ mortality) reductionmaternal ters of three–quara for called (which 5 MDG and levels) 1990 comparedto 2015 year the by mortality der–5 un in reduction two–thirds a for called (which 4 (MDG) Goal Development Millennium both achieve to able were deaths child and neonatal perinatal, world’s the maternal, of 97% with countries 75 the months ofageoccureach year, mostlyfrom readily preventable ortreatable conditions[ mortality (MNCH), 8.8 million maternal deaths, stillbirths, neonatal deaths, and deaths of children 1–59 child and neonatal, maternal, reducing in world the around made been have gains major though Even approaches inimproving MNCHandtheapproaches thathavebeenusedtoachieveeffectiveness. now isanopportunetimetotakestockoftheevidenceregarding theeffectiveness ofcommunity–based Thus, (2015–2030). begun has Goals Development Sustainable the of era the and (2000–2015) ended has Goals Development Millennium Nations’ United the of era the which in time of moment a in stand grow.to Wecontinues now level community the at provided interventions individual of effectiveness world’s neonatal maternal, and child deaths remains around 50% or less [ tion, population coverage levels of evidence–based MNCH interventions in the countries with 97% of the supplementa A vitamin and immunizations for Except settings. resource–constrained most in systems tions to communities and even down to each household remains an underdeveloped component of health interven health delivering and approachcommunities an engaging as for care(CBPHC) health primary Governments have a responsibilityGovernments for the health of their people that can be fulfilled only by the provision of ad 6 ]. One of the important reasons for this disappointing result was resultdisappointing reasonswas this important for the of One ]. 329 1 Health by the People the by Health – 3 , 7 – 17]. In addition, there appears to be a rebirth of glob CBPHC, rationale,methodsanddatabasedescription [ ]. This book described a number of of number a described book This 22]. 1 20, – 24]: 3 ]. Nonetheless, community–based 21]. One of the fruits of these dis June 2017 •Vol. 7No. 1•010901 4 ]. The evidence regarding the th anniversary of the signing ofthesigning anniversary 18]. This article is the first 5 ]. Only four of ]. Onlyfourof 21, 23]. ------
VIEWPOINTSPaPErS PaPErSVIEWPOINTS June 2017 •Vol. 7No. 1•010901 Perry etal. later met face to face at UNICEF Headquarters in 2008 to discuss preliminary findings of the review. Af Johns Hopkins University ( the at Health International of Taylor,Emeritus ProfessorCarl then Dr of chairmanship the under ated cre was Panel Expert an World2006, the fromin Organization grant Health small initial an Following ployed attheoutsetofreview), andmostrecently theGatesFoundation. em was Perry Dr where NGO (the Generations Future Development, International for Agency US the over 150 people and not only APHA but also the World Health Organization, UNICEF, the World Bank, as a small volunteer effort by and Perry Freeman and others has now, more than a decade later, involved PHC in Improving Child Health, with Henry Perry and Paul Freeman serving as Co–Chairs. What began Thus, beginning in 2005, the Working Group created a Task Force for the Review of the Evidence of CB of CBPHCinimproving health,theWorking Group decidedthatacomprehensive review wasneeded. the publication of a book on CBPHC [31]. As the evidence continued to mount regarding the effectiveness to led workshops these of One CBPHC. to related themes on workshops annual day–long holding been now,decades two WorkingFor the 1997. has Groupin established was Section Health International the from the International Health Section at APHA and from APHA staff, a Working Group on CBPHC within encouragement and support of result a As populations. geographically–defined of health the improving to CBPHC of contributions the highlight to (APHA) Association Health Public American the of meetings annual the at panels organized Perry Henry Dr and deceased) (now Wyon John Dr 1990s, early the In Background ofthereview age has declined from 18.9 million in 1960 [ of years beforechildren5 dying of number throughoutThe mortality world. the maternal and child ing Over the past three decades since the Declaration of Alma–Ata, major progress has been made in reduc ticipation, andreduction ofinequities. care to also address the primary causes of ill–health through inter–sectoral collaboration, community par ers and traditional practitioners, and it broadened the concept even further by calling for primary health ber of births each year has increased from 96 million in 1960 [ ity have also been important but more gradual. The number of maternal deaths declined from 532 from declined deaths maternal of number The gradual. more but important been also have ity mortal required67% the maternal in Reductions 2015. reachfor to Goal Development Millennium the [ 53% by fallen has globally rate mortality under–5 global the years, 25 past the Over [ 1970 in births live 1000 per 148 from declined has rate mortality under–5 al practices andgreater sharingofnewinformation”[29].AsaneditorialinTheLancet good of sharing “systematic for and works what of evaluation for now,need ever,is a than There more needs andexpectationsoflocalpeopleformedicalcare. the meet effectively more time, same the at and, health population improveeffectively more to systems a more comprehensive primary health care system in resource–constrained settings that can enable health child and maternal deaths, and (2) CBPHC has the potential for providing an entry point for establishing preventable ending progressin accelerate to potential the has interventions evidence–based of coverage Our review begins with the premises that (1) further strengthening CBPHC by expanding the population the heartofwhatourreview isabout. tions), and the conditions that appear to be important for achieving success are less summarized. This is interven other with combination in (particularly practice in delivered actually are interventions these documented, evidence about the total range of CBPHC interventions for MNCH, their effectiveness, how well– generally is interventions community–based effectivenessspecific the of about evidence Although [ 303 to 1990 in the findings from this review, and to suggest next steps in research, policy and program implementation. tiveness of CBPHC in improving the health of mothers and their children, to draw conclusions regarding effecregardingthe - evidence the analyze and review summarize, to opportunity providesan series This Evaluation matters.isscience.” not. do others and work programmes some why understanding for capacity efficiency, build enhance and ing, evaluation…. [Evaluation] will not only sustain interest in global health. It will improve quality of decision mak to commitment equal an by matched been not has decade past the during investments health global in scale–up massive A afterthought. an Currently,only health. is global it in priority top the become now must “Evaluation 28], farlessthanthe75%required toachievetheMillenniumDevelopmentGoal.
000 in 2015 [ 2015 in 000
Table 1). This group participated in the initial design of the review and then ], and the global maternal mortality ratio fell by 44% during this period period this during 44% by fell ratio mortality maternal global the and 28], 330 25] to 5.9 million in 2015 [ 25] to 139 million in 2015 [ 26] despite the fact that the num www.jogh.org ] to 43 in 2015 [ 2015 in 43 to 25] • doi:10.7189/jogh.07.010901 30] observed: ], far less than than less far 26], 27]. The glob
]. 26]. 000 000 ------www.jogh.org Table 1. Maternal, Neonataland ChildHealth Maternal, †Chair of thePanel,2006–2010. *Chair ofthePanel, 2010topresent. a rl JamkhedComprehensive Rural Raj Arole n hbaAoeJamkhedComprehensive Rural Shobha Arole ai alWrdHat raiainMedicalOfficer, ChildandAdoles- World HealthOrganization Rajiv Bahl Abhay Bang Society for Education, Action and SocietyforEducation,Actionand Abhay Bang - UnitedStatesAgencyforInterna Al Bartlett Bhutta Zulfiqar oetBak Bloomberg SchoolofPublicHealth, Robert Black* Chowdhury Mushtaque Costello Anthony a aeeTropical InstituteofCommunity Dan Kaseje Kirkwood Betty Knippenberg Rudolph Kureshy Nazo Lanata Claudio Lucas Adetokunbo James Phillips Mailman School of Public Health, MailmanSchool ofPublicHealth, James Phillips Pang Ruyan School of Public Health, Peking SchoolofPublicHealth,Peking Pang Ruyan Sanders David Agnes Soucat World Health Organization Formerly Lead Economist, Human FormerlyLeadEconomist,Human World HealthOrganization Agnes Soucat alTyo†Bloomberg SchoolofPublicHealth, Carl Taylor† ayTyo Independentconsultant Taylor Mary ea itr eea nvriyo eoa Professor ofEpidemiology FederalUniversityofPelotas Cesar Victora Zonghan Zhu Capital Institute of Pediatrics and CapitalInstituteofPediatrics and Zonghan Zhu ame Members oftheExpertPanelforReviewEffectiveness HealthCare ofCommunity–BasedPrimary inImproving • doi:10.7189/jogh.07.010901 o Health Project Health Project (SEARCH) Research inCommunityHealth tional Development Khan University, Karachi,Pakistan Women andChildHealth,theAga Canada andCenterofExcellencein Hospital forSickChildren, Toronto, Centre forGlobalChildHealth, Johns HopkinsUniversity BRAC ol elhOgnzto FormerlyProfessor, International World HealthOrganization Health andDevelopment Tropical Medicine London SchoolofHygieneand UNICEF tional Development - United StatesAgencyforInterna nal Instituto deInvestigationNutricio- Harvard University Columbia University University of Western Cape School ofPublicHealth,University Johns HopkinsUniversity Medicine Association Health, Beijing;ChinesePreventive CenterforChild China Advisory rganIzat Ional
aFFIlIat Ion t Director (nowdeceased) Director cent HealthandDevelopmentUnit Director Survival, USAID;nowretiredSurvival, Formerly SeniorAdvisorforChild Professor al Health Professor,- Department ofInternation Deputy Director School ofPublicHealth;currently Formerly DeanoftheJamesGrant Adolescent Health Childand Newborn, Maternal, currently Director, Departmentof Health, UniversityCollege,London; Perinatal Care Unit,InstituteofChild Director International Health International Unit, Professor ofEpidemiologyand ResearchPublic HealthIntervention Senior AdvisorforHealth Global Health Health GrantsProgram, Bureau for Team Leader, and ChildSurvival Senior Researcher International Health International Adjunct Professor of Professor and PerinatalHealth onMaternal WHO GlobalSurvey National Coordinator forChina, Visiting Professor andformerly rfso n eneeiu CapeTown, Professor andDeanemeritus Organization Financing oftheWorld Health and Health Systems,Governance World Bankandcurrently Director of Development, AfricaRegionofthe International Health(nowdeceased) International Professor Emeritus,Departmentof Independent SeniorTechnical Expert Gates Foundationandcurrently Community HealthSolutions,the Formerly SeniorProgram Officer, Preventive Medicine Association Chairman ofChild Health,Chinese for ChildHealth,Beijing,and Center Pediatrics andChinaAdvisory Professor, CapitalInstituteof Itle 331
l India Jamkhed, India Jamkhed, Switzerland Geneva, India Gadchiroli, USA ton, DC, Washing- Pakistan Karachi, Canada and Toronto, MD, USA Baltimore, Bangladesh Dhaka, Switzerland Geneva, Kenya Kisumu, England London, NY, USA New York, USA ton, DC, Washing- Lima, Peru Nigeria Ibidan, NY, USA New York, China Beijing, Africa South Switzerland Geneva, MD, USA Baltimore, USA Vermont, Royalton, South Brazil Pelotas, China Beijing, ocatIon CBPHC, rationale,methodsanddatabasedescription FormalIzat p guIdelInes revIew June 2017 •Vol. 7No. 1•010901 artIcIpated X X X X X X X X X X X X X X X X X X X 2006 Ion
For
In oF
p p Face anel artIcIpated meetIng –
In to X X X X X X X X X X X X X X X X X X X X X 2008 –
Face oF
In
FInal p In IcIart (2016)
I rev
FInd X X X X X ew pated I
ngs oF
VIEWPOINTSPaPErS PaPErSVIEWPOINTS June 2017 •Vol. 7No. 1•010901 Perry etal. at staticfacilities(includinghealth centersandhospitals)andbecloselyintegratedwith them. can (and of course should) connect to existing health services, health programs, and health care provided the provision of health care and health services outside of health facilities at the community level. as CBPHC well as level household the at behaviors key of promotion the includes CBPHC disease. control and improve to health together work processprogramscommunities a throughhealthand is which CBPHC The Task Force andtheExpertPanelagreed onthefollowingdefinitionofCBPHC: METHODS • • • • • • • • • • • The questionswhichthereview seekstoanswerare: the methodologicalqualityofassessment. as well as – project outcomes the influenced have might factors contextual additional what project,and the by served was population of size what long, how for from, came funding the where whom, by and implemented were they where projects– the of context the understand to seeks review the addition, In tions and the role of the community and community health workers in implementing these interventions. prove MNCH. In addition, the review describes the strategies used to deliver community–based interven together, the findings comprise a comprehensive overview of the global evidence in using CBPHC to im (collectively referred to in this series as projects) that have assessed the impact of CBPHC on MNCH. Al The review consists of an analysis of documents describing research studies, field projects, and programs areas. health facilities that serve populations of mothers, neonates and children living in geographically defined of walls the beyond out carried are that approaches and interventions on focuses review The ulation. or coverage of proven interventions for mothers, neonates and children in a geographically defined pop mortality, of level morbidity, the serious as status, purposes nutritional our for here defined is outcome nity–based approaches to improve MNCH. The health of mothers, neonates and children as a measurable commu through achieved be can what regarding evidence the summarize to is review this of goal The Goals ofthereview overall effort areview oftheeffectiveness ofCBPHCinimproving MNCH. it became possible to expand the scope of the review to maternal health. Thus, we have now renamed the in their first 5 years of life). With support from USAID and the Gates Foundation between 2013 and 2016, When the review began in 2006, the focus was exclusively on child health (that is, the health of children stitute thefinalarticleinthisseries[32]. International Health at Johns Hopkins, and has participated in the final set of recommendations that con ter Dr Taylor’s death in 2010, the Panel reconvened under the leadership of Dr Robert Black, Professor of Whatadditionalresearch isneeded? Whatgenerallessonscanbedrawnfrom thefindingsofthis review? Howstrong istheevidencethatCBPHCcanpromote equity? Whatprogram elementsare correlated withimprovements health? inchildandmaternal Whatcharacteristicsdoeffective CBPHCactivitiesshare? WhatspecificCBPHCactivitiesimprove MNCH? for donors? What are the implications for local, national and global health policy, for program implementation, and national levelswithinthecontextofseriousfinancialand humanresource constraints? How can successful community–based approaches for improving MNCH be scaled up to regional and order toimprove health? childandmaternal How strong is the evidence that partnerships between communities and health systems are required in and whatcommunity–basedapproaches appeartobemosteffective? effectivenessthe facilitate system) health local the within those CBPHC (including of conditions What sustain thatimprovement? How strong is the evidence that CBPHC can improve MNCH in geographically defined populations and
332 www.jogh.org • doi:10.7189/jogh.07.010901 ------www.jogh.org • doi:10.7189/jogh.07.010901 • • • CBPHC includesthefollowingthree different typesofinterventions: part ofaprogram whichincludestheprovision athealthfacilities. ofservices may be comprehensive in scope, highly selective, or somewhere in between; and they may or may not be programs;they care health private or governmental with collaborate not may or programsmay CBPHC trition, livingstandards, andempowerment. nu income, education, indirectly) or (directlyimprove to seek that programs including se, per services health of provision the beyond improvement health to approaches multi–sectoral includes also CBPHC beyondthefacility.munity involvementandassociatedservices of health facilities. CBPHC does not include health care provided at a health facility unless there is com through population outreachoutside defined geographically a of improvinghealth involves the CBPHC gram interventions on the health of all mothers and/or children in a geographically defined area, although effectiveness pro the of on - was focus The above. defined as health child or neonatal effectmaternal, on for a defined geographic population and (2) a description of the findings of an assessment of the project’s The principal inclusion criteria for the literature review were: (1) a report describing the CBPHC program Document retrieval improvements inchildhealthobtainedbyCBPHC approaches were equitable. degreethe which to reviewthe addition, In concerning documentation available of analysis an included • • • • • Neonatal andchildhealth • • • • health Maternal come indicatorswasrequired tobepresent inorder fortheassessmenttobeincludedinreview. served could reasonably be attributed to CBPHC program interventions. At least one of the following out ob changes any that way a such in indicators MNCH in changes of assessments out carried that studies research and projects programs, community–based described that documents sought TaskForce The qualifying forreview Types neonatalandchildhealthinterventions ofassessmentsmaternal, Healthcommunicationwithindividuals,familiesandcommunities; Changeinmortality. Changeinnutritionalstatus tive services (eg,community–basedtreatmenttive services ofpneumonia). community,the in care health Provisionof cura or immunizations) (eg, services preventive including and services; planning, delivering,evaluatingandusinghealth for mobilization andcommunityinvolvement Social Change inmortality(perinatal,neonatal,infant,1–4–year, andunder–5mortality); diarrhea, malaria,andlow–birthweight);or, Change in the incidence or in the outcome of serious, life–threatening morbidity (such as pneumonia, choma, whichcanresult inblindness); Change in the incidence or in the outcome of serious but non–life–threatening morbidity (such as tra deficiency); Change in nutritional status (as measured by anthropometry, anemia, or assessment of micro–nutrient diarrhea andmalaria; pneumonia, water,of drinking appropriatetreatment of appropriatesanitation; appropriatetreatment washing; hand appropriate therapy; preventive intermittent and nets bed insecticide–treated with ia propriate breastfeeding; immunizations;vitaminAsupplementation;appropriate prevention ofmalar ap- including feeding, child young and infant appropriate period; neonatal the during care propriate ap delivery; (clean interventions evidence–based more or one of coverage population the in Change sia, eclampsia,sepsis,hemorrhage); or, Change in the incidence or in the outcome of serious, life–threatening morbidity (such as pre–eclamp tal care, delivery by a trained attendant, delivery in a health facility, clean delivery, and postpartum care) Change in the population coverage of one or more evidence–based interventions (utilization of antena 333 CBPHC, rationale,methodsanddatabasedescription June 2017 •Vol. 7No. 1•010901 ------
VIEWPOINTSPaPErS PaPErSVIEWPOINTS June 2017 •Vol. 7No. 1•010901 Perry etal. Figure 1. nity-based primary healthcarenity-based primary (CBPHC). Selection process ofassessmentstheeffectiveness ofcommu - tions from othersources, books,orbookchapters. evalua were PubMed through identified not were that assessments Other Document. Supplementary mented by US–based non–governmental organizations. These are listed separately in Table S4 in imple and Program Grants Health and Survival Child USAID the by funded projects survival child of project evaluations were respectively) 80, and (16 most through PubMed, identified not were that view Of the 33 maternal health assessments and the 115 neonatal/child health assessments included in the re counted thisasonlyoneassessment. and assessment single articles/reportsa various with associated the combined have we which in cases of phy.in when Thus, bibliogra the referencesin these areincluded of All document. fromone derived morethan is database and child health), and the equity review. There are a number of cases in which a single assessment in our health neonatal for analyses the in included were these review,of which (and review health child the in with these 700 assessments. The bibliography also indicates which references were in the maternal health in TableS3 healthreview and548fortheneonatal/childhealthreviewternal ( ma the for criteria the met assessments 152 of total A evaluation. of form satisfactory a had they if and intervention, the of description adequate an provided review,they for if criteria the met they if cluded in were journals scientific peer–reviewed in published not Documents Panel. Expert the of members including included, be to documents for suggestions their for field the in persons knowledgeable tacted Online Supplementary Supplementary in Online S6 and S5 Appendices arein forms contained These out. carried terventions in the for except identical were assessments health maternal for form the and assessments health child Two data extraction forms were prepared through an iterative process. The extraction form to be used for The documentreview process contains a bibliography with the references associated associated references the with bibliography a contains Document Supplementary Online
above we refer to the number of articles/reports, there are a small number number small a aretherearticles/reports, of number the to refer we above 1 Figure 334
qualify for the review.the Finally,for con Task Forcequalify the might which articles published and reports, ments, of Public Health asking for information about docu Health Associations, and the Association of Schools Public of Federation World the Group), CORE (the Health Child for Group Resources and tion American Public Health Association, the Collabora the Council, Health Global the of those cluding in listservs, health global used widely on out sent were broadcastssearch, PubMed the to addition In mentioned above. measures outcome health child or neonatal other the of more or one included also reports the unless included not were children of development logical improvements in neurological, emotional or psycho were outcomes the which in CBPHC of fectiveness ef the of Assessments team. study the of members 1 ure ( health child or neonatal on articles 4276 and health maternal on articles 7890 yielding queries, two these using 2015 December 31 until up cally periodi searched was Medicine’sdatabase PubMed Document in S2 and TablesS1 (see related terms were identified to create a search query “community health,” and “developing countries” and health,” “child health,” “maternal for terms Key fined area. de geographically a in children their and mothers of subset a on was focus the transmission), HIV maternal–to–child of studies in (eg, cases some in ). The ). articles were screened separately by two ). The United States National Library of of Library National States United The ). Figure 1). www.jogh.org Online Supplementary Supplementary Online • doi:10.7189/jogh.07.010901 Online Online Fig------www.jogh.org • doi:10.7189/jogh.07.010901 scribes the health impact of an integrated primary health care project in South Africa led by Sidney Kark Our review includes 16 assessments of projects that were completed before 1980. The earliest report de ed fiveormore interventions. health assessments On three–quartersthe other hand, among the terventions. maternal (75.7%) includ and child health assessments, 51.6% were of only one intervention, and 87.4% were of four or fewer in neonatal the Among years. 3 than less for implemented were (62.9%) two–thirds almost and duration in years 2 than less were projects of (46.3%) half Almost 2000. since begun had projects the of 61.9% and professional titlesare containedinTable Document. S7inOnlineSupplementary ers, many of whom worked on a volunteer basis, are shown in the acknowledgment section; their names review- the of names The USA). Georgia, Atlanta, Prevention, and Control Disease for Centers US Info, (Epi 3.5.4) (version database INFO EPI an to transferred was review summative final the and reviews, for the review. A third reviewer provided quality control and resolved any difference observed in the two qualified that assessment each for Form Extraction Data a completed Twoeach reviewers independent munity wasinimplementation. regarding how the interventions were implemented at the community level and what the role of the com Document. Both forms were developed with the purpose of extracting all possible information available assessments are based on data derived from projects reaching more than 25 than more reaching projects from derived data on based are assessments Half (49.3%) of the assessments are of projects serving 5000 or fewer women and children. 18.2% of the bers, research healthstaff. workers,andgovernment implemented projects at the local level were community health workers (CHWs), local community mem only 4.3% of assessments were local communities the only identified implementers. Those who actually ( level local provincial,or national, the at governments with working researchorganizations),often and universities The implementing and facilitating organizations for these projects were primarily private entities (NGOs, tional level,and3.2%atamultinationallevel. na a at provincial/state3.7% the level, at sub–province)7.5% (or level, district the province),at 37.5% in a single community, 38.1% in a set of communities not encompassing an entire health district (or sub– icas (Table 2 and Table S8 in ments came from Africa WHO Region, 28.5% from the South–East Asia Region, and 9.7% from the Amer assess- country–specific the of 49.0% respectively). 47, and 77, (86, assessments of number largest the try. Thus, altogether, 786 country–specific assessments were identified. India, Bangladesh, and Nepal had Among the 700 assessments in our data set, a small proportion contained data from more than one coun respectively. setting, peri–urban or urban an in exclusively out carried were 11.1% and 16.9% while part, in least at settings rural in out reviewwerecarried our in included assessments the of three–fourths(78.4%) Over orunpublishedprojectternet) evaluations. in the on (available literature gray fromthe either are 12.7% and internet), the on available not reports or books (mostly publication of type other aresome 4.0% peer–reviewedjournals, in published articles scientific are assessments of 78.8% Overall, analysis. our in assessments separate as counted been have USA). Forthepurposeofthisreview, andchildhealthoutcomes the39assessmentswithbothmaternal was queried using EPI INFO version 3.5.4 and STATA version 14 (StatCorp LLC, College Station, Texas, MNCH improving in CBPHC of effectiveness the of assessments 700 describing database electronic An Database description ects,” andtheevaluationsoftheireffectiveness as“assessments.” projects/programs” but for practicality’s sake we will refer to them throughout this series simply as “proj studies/field “research as to referred be properly should they whole, a as activities community–level of group this to referring When setting. field typical more in time of period long a over programs opment devel and health of array comprehensive provided programs a which of assessments (2) to setting field the efficacy of single interventions over a short period of time in a highly supervised and well–supported is a heterogeneous group of assessments in the sense that they range from (1) research reports describing employed one or more CBPHC interventions for improving maternal, neonatal and/or child health. This that programs and projects, studies, field for out carried were review our in included assessments The used Comment onterminology ). While communities were — by definition — involved in all of these projects, in in projects, these of all in involved — definition by — were communities While Table3). Online Supplementary Document 335 CBPHC, rationale,methodsanddatabasedescription ). 8.6% of reports assessed interventions June 2017 •Vol. 7No. 1•010901
000 women and children. and women 000 ------
VIEWPOINTSPaPErS PaPErSVIEWPOINTS June 2017 •Vol. 7No. 1•010901 (in 5-yearintervals). Perry etal. Figure 2. Number of assessments in data set by year of publication Number ofassessmentsindata set byyearofpublication tiple countries. *The total number of countries listed here exceeds the number of assessments because some assessments were conducted in mul toxoid immunizationinColumbia,SouthAmerica,published1966[ [ 1952 in published and 1940s the in Table 3. maternal, neonataland childhealthbyregion andthecountrieswithgreatestmaternal, numberofassessments *Percentages adduptomore than100%becauseprojects oftenutilizedmore thanoneImplementer. Table 2. Total Europe Western Pacific Expatriates iityo elhwre rohrgvrmn–adhat okr/rfsinl 304 Local communitymembers(nottrainedasaCHW) healthworkers/professionals ofhealthworkerorother government–paid Ministry Research workersonlyfortheproject Community healthworkers(eitherpaidorvolunteer) Implementers atthecommunitylevel: Faith–based organization National NGO Local NGO Local government Private organization/university/research organization International NGO International Eastern Mediterranean Eastern State ornationalgovernment Facilitating and/orstakeholderorganization: Americas South–East Asia Africa who r egIon Number ofassessmentstheeffectiveness healthcare ofcommunity–basedprimary inimproving Implementers ofprojects forimproving MNCH
n 786* 224 385 umber 37 61 76 4 ]. The next earliest report concerns the effectiveness of tetanus tetanus of effectiveness the concerns report earliest next The 33]. % ( 336 0.%Tanzania 100.0% 85 Bangladesh 28.5% 49.0% n .%Pakistan 0.5% 4.7% 7.8% 9.7% =76*c = 786)* S7 and S8 in S8 and S7 outcomes and 183 neonatal/child outcomes (see Tables maternal 56 review: the in included assessments 700 the in measured outcomes 239 of We total a identified Types ofoutcomesassessed (MDGs) ( Goals Development Millennium the of establishment larly in the period 2001–2011, the decade following the ments published between 1980 and 2015, but particu There has been a rapid growth in the number of assess time Number ofassessmentscompletedover was a slight decline in the number of publications. was aslightdeclineinthenumberofpublications. end of 2015 when the assessment retrieval ended, there the from until years 2011 five the In shown). not (data studies health child/neonatal for and maternal for both Figure 2 Ethiopia Uganda Malawi Ghana Kenya Nepal India ountry Online Supplementary Document Supplementary Online ). The surge in publications is present www.jogh.org 34]. n n 200 238 519 125 243 254 281 424 umber 33 27 85 umber 19 27 28 30 34 35 36 47 86 77 • doi:10.7189/jogh.07.010901 % ( % ( n n 10.9 28.6 43.4 34.0 74.1 12.1 17.9 34.7 36.3 40.1 60.6 = 786)* 2.4 3.4 3.6 3.8 4.3 4.5 4.6 6.0 9.8 = 700) 4.7 3.9 ). ). - - - www.jogh.org • doi:10.7189/jogh.07.010901 as welltheassessment. project the funded donor the cases, all) not (but most In assessments. of number smaller a funded that in healthsystemcapacity. munity case management of acute childhood illness provided by CHWs, and measures of improvements com of quality caretakers, and parents among knowledge health–related in changes development, tor were included with other indicators that did qualify for the review. These include progress in psychomo vices. In addition, some assessments contained outcome measures that did not qualify for the review but status, population coverage of healthy behaviors, and changes in the appropriate utilization of health ser mortality,in changes were: outcomes morbidity,health serious child and neonatal Common nutritional pregnancy.during infections transmitted sexually other and HIV for screening and practices, birth safe of knowledge planning, family post–partum of receipt vaccination, toxoid tetanus of receipt plements, provider,skilled a delivery,sup by facility nutritional delivery of emergencies,receipt obstetric for care at attendance care,mortality, antenatal in: of changes receiptwere outcomes health maternal Common with anyone who is interested (contact Henry Perry [email protected] anyonewhoisinterestedPerry ). (contactHenry is reported in this series. Any of the project assessments included in this review are available to be shared [ series this in ticles ar subsequent the for basis the as serves It existence. in database similar Weother any of aware not are development ofthedatabase Availability ofthedatabaseforfurtheranalysesandpotential ( World(6.2%) the Bank and (7.7%), agencies UN other assessments (15.8%), followed by the World Health Organization (14.2%), the Gates Foundation (10.7%), review.the in included assessments the of thirdof largest(33.4%) number next the supported UNICEF financial support for the assessments included in our review, contributing to the financial support of one– The United States Agency for International Development (USAID) was far and away the largest source of Source offinancialsupport forassessments of theassessments. of the assessments, and they considered the assessment quality to be good, high, or exceptional for 88.4% trolled before–after assessment designs. Reviewers considered the methodology to be adequate in 89.8% uncon were assessments the of 27.4% assessments. the of 33.7% in present were designs assessment from a comparison group as well as pre– and post–intervention data were present. Randomized controlled fourths (72.5%), pre– and post–intervention data were collected. In 44.6% of the assessments, both data In the majority (61.0%) of the assessments, a control or comparison group was present. In almost three– Types ofresearch methodologies usedtoassesseffectiveness *Multiple fundersmayhavesupported asingleproject/assessment. Table 4. diseases. chronicHIV, and as malaria, such tuberculosis, priorities health global other to approaches nity–based commu include to MNCH beyond database this expanding for exists also potential the And, available. publicly and updated regularly is that database dynamic a as this maintaining for exists potential The Wellcome Trust Development Agency(CIDA) Canadian International Development(UK) Department forInternational World Bank Other UNagency(eg,UNDP, UNFPA, UNHCR,WFP) The BillandMelindaGatesFoundation World HealthOrganization (includingthePan AmericanHealthOrganization) UNICEF US Agency for International Development US AgencyforInternational d onor Leading sources offinancial supportforprojects whoseassessmentswere includedinthedatabase 32, 35– ]. However, there is an opportunity for more analyses of the database than than database the However,of 40]. moreanalyses for opportunity an thereis 337
). There were numerous other donors donors other numerous were There Table4). CBPHC, rationale,methodsanddatabasedescription n sessments umber June 2017 •Vol. 7No. 1•010901
oF 110 233 18 23 28 43 54 75 99
proJects
supported / as - % ( n 10.7 14.1 15.7 33.3 = 700)* 2.6 3.3 4.0 6.1 7.7 ------
VIEWPOINTSPaPErS PaPErSVIEWPOINTS June 2017 •Vol. 7No. 1•010901 Perry etal. projects with long–term evidence of effectiveness in improving MNCH [ integrated of characteristics common the of improvementsanalysis achieving an for (vi) [38], MNCH in ness of CBPHC in improving child health [ analysis of the effectiveness of CBPHC in improving neonatal health [ [ health maternal improving in CBPHC of effectiveness the of analysis an (i) include: These This topicisanimportantonebuttimeandresources were notsufficient thisouteither. tocarry Seven subsequent articles are being published in this series that answer the questions posed by the review. Finally, ourreview hasnotincludedtheeffectiveness ofCBPHCinreducing miscarriages andstillbirths. Subsequent articlesinthisseries would have made our review more complete, but time and resources were not sufficient to carry this out. PHC in increasing the coverage of family planning is services extensive. Thus, inclusion of this literature ily planning all have favorable benefits for MNCH. Furthermore, the evidence on the effectiveness of CB It is well–known that the use of family planning, birth spacing, and the reduction of unmet need for fam and thepotentialforfurtherstrengthening ofCBPHCtoimprove MNCHglobally. clusions about the overall effectiveness of CBPHC, the most common strategies used in implementation, the broad scope of evidence related to the effectiveness of CBPHC in improving MNCH and to draw con approachspecific one compared implementation to another. to Rather, to is study our of aim the review effectiveness the about or of effect intervention, of specific magnitude any the for about evidence, the of outcome measures used to assess outcomes, it is difficult to make definitive statements about the the(3) strength and implemented, were they which in manner the (2) implemented, interventions of types the (1) of heterogeneity the Given well. as important is employed indicator(s) outcome of type the course, of And, factors. contextual myriad and implementation, of quality the implemented, interventions of The degree to which these projects improved MNCH depended on many factors: the type(s) and number under thecircumstances. possible implementation of quality highest assurethe to made effortsbeen presumablyextra had which in and effectivenessundertaken of was documentation circumstanceswhich special in in achieved been has what of rather but CBPHC of practice day–to–day current of effectiveness the of representative not is database the Thus, assistance. technical or support donor international of type some without mented points. But it is the case that very few of the assessments in our database were of projects that were imple fact, it is often difficult to determine exactly where a project might lie on a continuum between these end in and, issue this on information collect not did form extraction data Our conditions. world” “real der routinein cur efficacyEffectivenessto un settings. contrast areout in assessments, carried assessments, oc not do often that conditions are These established. been has engagement community optimal when and support, logistical and resources, supervision, training, optimal have members staff field when es, assessments, of course, are carried out for projects that have been implemented under ideal circumstanc Efficacy explore. adequately to able not are we which issue important an is assessments effectiveness to The degree to which the assessments included in our review represent efficacy assessments as compared fectiveness ofCBPHCinimproving MNCH. fromreviewef our numerousdemonstrate in the that included of been value have the that assessments ent and should be recognized. Nonetheless, the inability to document these experiences does not detract pres is bias publication serious a that means literature. This scientific the in or documents open–access program failures and serious challenges encountered in program implementation are rarely described in This review is limited to documents that describe the impact of project interventions. As is well–known, these wouldhavechangedtheoverallfindingsofour review. the review. for qualified have would that articles additional several identified have we However,of none began, study this of review.portion our up in write included and be analysis could data these the Since of (80) one–fifth only that such were constraints time and resource but available, publicly are and sion inclu for projectcriteria the meet archivethat survival evaluations child unpublished more400 of than that met the inclusion criteria. In addition, the USAID Child Survival and Health Grants program has an sources and time constraints prevented screening other electronic databases beyond PubMed for reports Our review is a comprehensive one, but we make no claim that it is a complete or systematic review. Re ommendations oftheExpert Panel[32]. ing equitable improvements in child health [
338 37], (iv) an analysis of the effectiveness of CBPHC in promot 40], (v) the strategies employed by effective CBPHC programs 36], (iii) an analysis of the effective www.jogh.org 39], and (vii) summary and rec • doi:10.7189/jogh.07.010901 35 ], (ii) an an (ii) ], ------www.jogh.org rEFEr ENCES • doi:10.7189/jogh.07.010901 4 3 1 2 Bhutta 2017. 2017.Accessed:27February pdf?sequence=3&isAllowed=y26 February https://openknowledge.worldbank.org/bitstream/handle/10986/23833/9781464803482. Available: Bank. WorldWashington,DC: Edition. Third Health, Child and Newborn, Maternal, Reproductive, Priorities: Control ease 35. Lassi a review oftheevidence.Geneva:World HealthOrganization; 2004. 2003;362:65-71. Medline:12853204 Lancet. year? this prevent we can deaths child many How SS. Morris ZA, RW,Bhutta Steketee RE, G, Black Jones Hill Z Medline:24304052 strengthening CBPHC. by 2030 year the by children their and women for care to access universal achieving for and deaths mulation of policies and programs that will be useful for ending preventable maternal, neonatal and child for the to contribute to is series this of aim The Panel. Expert an of recommendations and conclusions with along database this of analyses of findings the describe series this in follow that articles The level. community the at implemented wereprojects how to given been has attention Special assembled. been has MNCH improving in CBPHC of effectiveness the regarding evidence the of database extensive An CONCLUSIONS closure.pdf (availableuponrequest from thecorresponding author),anddeclare no conflictofinterest. Conflict of interest: included inourreport. Alloftheauthorsparticipated intherevision ofearlierdraftsandapproved thefinaldraft. review. overall the data to relatedquantitative decisions the all of in analysis participated the and performed end JS wroteHP declaration: HP,Authorship draft. first the PF, project this guided SG and BR, from the to beginning the ecution ofthereview. ex the in role no had support financial provided that organizations The Foundation. Gates the and velopment, WorldThe ticle: UNICEF,Organization, Health De International for Agency WorldStates the United the Bank, ar this in described work the conduct to used were that funds providedorganizations following The Funding: the executionofreview. in no role had support providedfinancial search.Weorganizationsthat literature The puterized all. them thank database and inputting it into an electronic format. Claire Twose provided expert technical assistance for the com Carol Mary Perry,Mirlene Jennings, SchleiffMeike Randolph, Elizabeth provided the assembling with assistance of the Acknowledgments: Jean Yuan.Jean in contained is them about information Further volunteers. unpaid as helped them of Most Tamarro, Henry Taylor, Muyiwa Tegbe, Angeline Ti, Charles Teller, Yetsa Meghan Tuakli,Suchdev,Sylla, Parminder Steinhardt,Mariame Laura Stake, Jess Stephen Wilhelm, Sillan, Olga Donna Sharan, Wollinka,kar,Mona and ca, Jeeva Rima, Jon Rohde, Evan Russell, Emma Sacks, Kwame Sakyi, Juan Sanchez, Nirali Shah, Manjunath Shan Laura Podewils, Jon Poehlman, Ramaswamy Premkumar, Braveen Ragunanthan, Bahie Mary Rassekh, James Ric Perry, Henry Perez, Erika Patel, Zohra Passeri, Carlo Parekh, Vikash Paredes, Pat Ogundalu, Oluwatosin Neat, Lenna Morgan, Mary Mirchandani, Gita Metangmo, Pierre–Marie WilliamMenson, nager,Mendoza–Sassi, Raul mar, Stacy Laswell, Ramiro Llanque, Amanda Long, Ron Mataya, Colin McCord, Meredith McMorrow, Henri Me WoonKu Kim, Ajoy Dennis Hoar,Kim, Hershberger,Jani, Sandy Cho Ann Asim Hill, Hashemi, Zelee Paymon Paul Freeman, Asha George, Juliana Grant, Stacy Grau, Sundeep Gupta, Nancy Habarta, Nowreen Haq, Runa Haq, Fort, Meredith Fan, Qi Enoh, Sheila Emami, Ashkan Dowell, Duane RichardDortonne, Jean Davachi, Christine Crouse, Deanna Christie, Len Chitnis, Ketan Cheatham, Cesar,Elizabeth raci Chang, Stephanie Chan, Elizabeth Ju Bryant, Jack Brown, Amberle Bowen, Lisa Boswell, WarrenBerggren,Claire Berggren, Gretchen Hill, worth Adhikari, Asma Aftab, Azal Ahmadi, Iain Aitken, Laura Altobelli, Chidinma Anakwenze, Ramin Asgary, Ann Ash Binita assessments: of reviewers as serving for people following Team.the WeStudy thank the of member a as phase ofthereview. TheWorld Bankmadeitpossibleforoneofitsconsultants,DrBahieRassekh,toparticipate initial the during Perry Dr to support salary and support, administrative space, office providing for Generations and particularly Health its Section International staff, which administered some of these funds. We thank Future Association Health Public American the to grateful Wealso Foundation. are Gates the and Generations, Future ment of the World Health Organization, the CORE Group (Collaboration and Resources for Child Health)/USAID, Develop and Health Adolescent UNICEF,and review: Child this of of Department Worldpenses the the Bank, ZS, Kumar ZS, R , Kirkwood B ZA, Black online supplementary document. Binita Adhikari, Omar Balsara, David Exe, Pam Flynn, Jennifer Hutain, RE. Global maternal, newborn, and child health–so near and yet so far. , Bhutta , , Edmond K doi:10.1056/NEJMra1111853 We are grateful to the following organizations that provided small grants to cover the ex the cover to grants small provided that organizations following the to grateful Weare All authors have completed the Unified Competing Interest Form at www.icmje.org/coi_dis- ZA. Community-based care to improve maternal, newborn, and child health. 2016. In: Dis- In: 2016. health. child and careCommunity-based improveZA. newborn, to maternal, . Family and community practices that promote child survival, growth and development: doi:10.1016/S0140-6736(03)13811-1 339 CBPHC, rationale,methodsanddatabasedescription June 2017 •Vol. 7No. 1•010901 N Engl J Med. 2013; TableS5 369: 2226------
-
VIEWPOINTSPaPErS PaPErSVIEWPOINTS June 2017 •Vol. 7No. 1•010901 Perry etal. rEFEr ENCES 13 12 14 10 11 9 15 8 16 7 27 22 21 20 18 23 6 29 28 26 25 24 19 17 5 www.who.int/workforcealliance/knowledge/publications/alliance/Global_CHW_web.pdf. 2017. Accessed:26February tional health systems. Geneva: World Health Organization and the Global Health Workforcena into integration for Alliance; recommendation 2010. and Available:studies, review, case http:// systematic country a Goals: Development Millennium Bhutta children through community-based approaches. Glob Public Health. Public Glob approaches. community-based through children line:19890758 ress for maternal, newborn, and child survival. child and newborn, maternal, for ress Freeman line:15866863 2005 Pediatrics. evidence. the of review a countries: developing in outcomes health ity and mortality and improving neonatal outcomes. S Lewin Lassi doi:10.1186/1471-2393-9-2 mortality. maternal reduce to interventions level Bhutta 6736(15)00519-X primary primary and community health care for maternal and child health and the management of infectious diseases. Salam Database SystRev. 2010;3: Victora key messagesfrom DiseaseControl Priorities3rd Edition.Lancet. 2016;388:2811-24. Gogia S systematic review. Black Kidney E titudes, andtransmission. meta-analysis of community-based trials. community-based of meta-analysis Sazawal S care approach. A Casebook.Franklin,WV, USA:Future Generations;1995. ment. In: Taylor C, Desai A, Knutsson K, O’Dea-Knutsson P, Taylor-Ide D, editors. Partnerships for Social Development: soon: a pathway towards evidence-based, respectful maternity care worldwide. Lancet. 2016 Lancet. worldwide. care maternity respectful evidence-based, towards pathway a soon: Schiffman 3099(03)00737-0 CEF; 2014.Available: http://sowc2015.unicef.org/. 2017. Accessed:27February York:New UNI Child. Every for UNICEF.Innovation Future. the World'sReimagine the 2015: of Children State The 6736(15)00120-8 Newell S Litsios line:12456122 S Litsios Chan M line:27642019 tember 1978 Geneva, Switzerland: World Health Organization; 1978. Available:http://www.who.int/publications/al1978. Organization;World Health Switzerland: Geneva, 1978 tember WHO. UNICEF. Declaration of Alma-Ata. International Conference on Primary Health Care, Alma-Ata, USSR, 6-12 Sep evidence. the of review a countries: developing in health able: http://www.who.int/pmnch/Countdownto2015FINALREPORT-apr7.pdf. 2017. Accessed:27February UNICEF. Tracking Progress in Maternal, Newborn and Child Survival. The 2008 Report New York: Geneva; 2008. Avail- 2016. Available: WorldWHO, UNICEF,Bank, UNFPA. WorldTrends2015. 2015. to 1990 Organization.Mortality: Health Maternal in You 91. Ahmad ganization, Geneva. Care,Health WorldUSSR: Conference1978; Ata. Alma Primary on International (editors). UNICEF WHO. Or Health 2017. maata_declaration_en.pdf?ua=1. Accessed:27February M Arole y ru fr hl Mraiy Estimation. Mortality Child for Group cy mortality between 1990 and 2015, with scenario-based projections to 2030: a systematic analysis by the UN Inter-agen Miller S doi:10.1053/j.semperi.2010.09.008 Medline:11100613 D ZS, Haider ZS, RE, Levin C , Hug L RA, Haroon S ZA, Lassi ZA, ZA, Darmstadt ZA, KW, editor. HealthbythePeople.Geneva,Switzerland:World HealthOrganization; 1975. CG, Requejo , Sachdev , , Abalos E OB, Lopez S Munabi-Babigumira , . Return toAlma-Ata. . Return D Kasaje , . The Christian Medical Commission and the development of the Worldthe Organization’sof Health development the health and primary Commission Medical Christian The . . The long and difficult road to Alma-Ata: a personal reflection. personal a Alma-Ata: to road difficult and long The . P , Winter , Black J , Perry , http://apps.who.int/iris/bitstream/10665/194254/1/9789241565141_eng.pdf?ua=1 , Darmstadt , , Ejdemyr S Am JPublicHealth.2004; doi:10.1080/17441690903330305 doi:10.1542/peds.2004-1441 doi:10.2190/RP8C-L5UB-4RAF-NRH2 doi:10.1016/S0140-6736(16)31472-6 G Pariyo ZS, BA, Bhutta BA, RE. Effect of pneumonia case management on mortality in neonates, infants, and preschool children: a
, Chamillard M , Walker N HB, Gupta HB, HS. Home visits by community health workers to prevent neonatal deaths in developing countries: a a countries: developing in preventdeaths to neonatal workers health community by visits Home HS. HR, Khan , Taylor, C Bull World HealthOrgan. 2010; AD, Inoue M , Ahmed JH, Barros
GL, Hasan GL, wal
GL, Agar GL, , Idele P Infect DisPoverty. 2014;3: CD004015. ZA. Community-based intervention packages for reducing maternal and neonatal morbid neonatal and reducingmaternal for packages intervention Community-based ZA. . The Christian Medical Commission’s Role in the Worldwide Primary Health Care Move Care WorldwideHealth the Commission’s Primary in Medical Role Christian The . KS, Gulmezoglu , Chou D
L Huicho , HH, Das SK, Rassekh SK, Lancet. 2008; C Glenton , AJ, Berman P . The decline in child mortality: a reappraisal. , Hogan D , Ciapponi A BS, Haws BS, S , Liu L JK, Bhutta Medline:20238326 , Baqui , . Global experience of Community Health Workers for delivery of health related related health of Workers delivery Health for Community of experience Global . 94:
B Lancet Infect Dis. 2003 Dis. Infect Lancet , Mathers C K Daniels , 372: . Accelerating progress in achieving the millennium development goal for for goal development millennium the achieving in progress Accelerating . , Temmerman M RA. Community-based interventions for improving perinatal and neonatal neonatal and improvingperinatal for interventions Community-based RA. , Bhutta Z 1884-
, Colaci D AM, Meads act 2015 Lancet.
AH. Community-based intervention packages for improving perinatal perinatal improving for packages intervention Community-based AH.
865-6. Lancet. 2016; Lancet. ZA. Impact of community-based interventions on HIV knowledge, at
340 88: Cochrane Database Syst Rev. 2010 93. 26.
658- , Boerma T Medline:15514223 , et al. Global, regional, and national levels and trends in under-5 X Bosch-Capblanch ,
M Penny hlbrh 2009 Childbirth. Pregnancy BMC , Comande D Medline:18790292 Medline:25126420 CA, Deeks 66B. ; , et al. Reproductive, maternal, newborn, and child health: 386
Semin Perinatol Semin 387: Medline:20865070 , et al. Countdown to 2015: a decade of tracking prog : 2275 ;3: 2049- , et al. Beyond too little, too late and too much, too JJ, et al. Systematic review of effect of community- 547 - 86
- 59. Glob Public Health. 2012 Health. Public Glob 56 . doi:10.2105/AJPH.94.11.1884 Medline:26361942
doi:10.1016/S0140-6736(08)61372-0 , van Wykvan , Int J Health Serv.2002; Health J Int . doi:10.1186/2049-9957-3-26 Medline:26477328 Medline:12954560 Bull World Health Organ. 2000; . 2010; . www.jogh.org doi:10.2471/BLT.09.069369 ; 11 34: BE, et al. Lay health workers in in workers health Lay al. et BE, : CD007754 ; 115 462- • doi:10.7189/jogh.07.010901 ;9:2. (2 (2 76. Suppl):519 . Accessed: 7 October 7 Accessed: .
Medline:19154588
doi:10.1016/S0140- ; Medline:21094420 doi:10.1016/S0140- 388 doi:10.1016/S1473- . Medline:21069697 32: ;7: : 2176 400- 709- - 617 -
19. Cochrane Cochrane 92 32. 78: . . Med Med Med Med- 1175------
www.jogh.org rEFEr ENCES • doi:10.7189/jogh.07.010901 33 38 32 30 31 34 39 37 36 35 40 pert Panel. Ex an of recommendations and conclusions 8. health: child and neonatal maternal, improving in care health primary Kark H Perry findings. health child 4. health: child and neonatal maternal, improving in care health primary community-based of fectiveness Black R ment SciencesforHealth(incollaborationwiththeHarvard SchoolofPublicHealth); 2002. health. global for priority top the Evaluation: Rohde 6736(10)60056-6 Newell H Perry J GlobHealth.2017;7: nity-based primary health care in improving maternal, neonatal and child health: 6. strategies used by effective projects health findings. the effectiveness of community-based primary health care in improving maternal, neonatal and child health: 2. maternal P Freeman health findings. neonatal 3. health: child and neonatal maternal, improving in care health primary community-based of effectiveness Sacks E Jennings M term impactonmortalityinchildren youngerthanfiveyearsofage. long- of evidence with Programs 7. health: child and neonatal maternal, improving in care health primary nity-based Schleiff M torum. Final report of a double-blind controlled field trial. fects. ef equity 5. health: child and neonatal maternal, improving in care health primary community-based of effectiveness SL, Cassel J J GlobHealth.2017;7: JE, WyonJE, J . Taylor. C, Expert Panel. Comprehensive , Freeman P KW,Lehmann A Duenas , Rassekh , B , Rassekh , B J GlobHealth.2017;7: , Kumapley , R M Schleiff , , Pradhan S J GlobHealth.2017;7: J GlobHealth.2017;7: J GlobHealth.2017;7: . ThePholelaHealthCentre; aprogress report. , editors. Community-Based Health Care: Lessons from Bangladesh to Boston. Boston, MA: Manage MA: Boston, Boston. to fromBangladesh Lessons Care: Health Community-Based editors. , , Sakyi K , Gupta , S , Gupta , S E Sacks , , Schleiff M 010906. , Freeman, P , Jennings M 010905. , Freeman, P , Freeman, P B Rassekh , 010904. , LeBlanc , , Sacks E 010908. , Gupta , S 010902. 010903. , Rassekh B . Comprehensive. . Comprehensive. DR, GarcesDR, Osorio N S Gupta , , Freeman P , Rassekh , B review of the evidence regarding the effectiveness of community-based
act 2010; Lancet. , Gupta S 341 H Perry , , Gupta S
H Perry ,
Bull World Health Organ. 1966 review of the evidence regardingevidence reviewthe of effectivenessthe commu of review of the evidence regardingevidence reviewthe of effectivenessthe commu of , et al. Comprehensive . Comprehensive . . The use of toxoid for the prevention of tetanus neona preventiontetanus the of for toxoid of use The . , et al. Comprehensive S AfrMedJ.1952; . Comprehensive. 375:
J GlobHealth.2017;7: 526. CBPHC, rationale,methodsanddatabasedescription Medline:20079530 review of the evidence regarding the ef the regarding evidence the of review reviewregardingevidence the of the 26: review of the evidence regarding the June 2017 •Vol. 7No. 1•010901 101-4. review of the evidence regarding ; 35 010907. : 863 Medline:14913265 - 71 doi:10.1016/S0140- . Medline:5338377 ------.
VIEWPOINTSPaPErS PaPErSVIEWPOINTS June 2017 •Vol. 7No. 1•010902 child health:2.maternalhealthfindings health careinimprovingmaternal,neonataland the effectivenessofcommunity–basedprimary a material. The onlineversionofthisarticlecontainssupplementary Electronic material: supplementary Perry Pradhan Sacks [email protected] USa Baltimore, MD21205 615 NorthWolfeSt. Health Johns HopkinsBloombergSchoolofPublic r Henry Perry Correspondence to: 6 5 4 3 2 1 Mary CarolJennings Gupta The WorldBank,Washington,Districtof Medical Epidemiologist,Lusaka,Zambia Department ofGlobalHealth,University Independent Consultant,Seattle, Institute forGlobalHealth,DukeUniversity, Department ofInternationalHealth,Johns Columbia, USa Washington, Seattle,USa Washington, USa Durham, NorthCarolina,USa Baltimore, MD,USa Hopkins BloombergSchoolofPublicHealth, oom E8537 comprehensive review of the evidence regarding
1 1 5 , Paul a Freeman , BahieMrassekh 2
, MeikeSchleiff
1 , Subarna 3,4 1
6 , Sundeep , Emma , HenryB
maternal deaths by2030. maternal to accelerate improvements in maternal health and to end preventable approach comprehensive a of component important an be will tions interven approaches. Community–based holistic from such benefits added potential emphasizing approaches, facility–based and nity– commu mortality.combined projectsand Most morbidity maternal of indicators key improving in CBPHC of effectiveness the of idence Conclusions This comprehensive and systematic review provides ev teams from peripheralhealthfacilities. outreach by services women’sof provisionparticipatory and groups, of formation visits, home collaboration, community involved often results these achieve to used strategies community–based The pact. tended delivery and facility–based deliveries all showed a positive im at- attendance, care antenatal on interventions community–based of Assessments effect. significant a find not did one and decreases cant tum hemorrhage, malaria or eclampsia. Of those, six reported signifi Seven assessments measured changes in maternal morbidity: postpar decreases. significant reported six testing, statistical performed that seven the of and outcome, primary a as mortality maternal reported assessment 15 indicators. multiple on results reporting assessments (50%); attended delivery (66%) and facility delivery (69%), with many attendance care antenatal (21%); morbidity maternal assessments); of (19% mortality maternal categories: main five into grouped were total of 1298 discrete interventions were assessed. Outcome measures ed 1–10 specific interventions aimed at improving maternal health.A sessments were set in rural communities. 72% of assessments includ Results health outcomes,andstrategiesusedinimplementation. Data were analyzed to identify themes in interventions implemented, extraction. data underwent that assessments included and 2015, ber Methods ported elsewhere inthisseries. re are health child and neonatal on Findings (MNCH). health child and neonatal maternal, improving in (CBPHC) care health primary community–based of effectiveness the of review larger a in included projects) as to referred (collectively studies research and programs, projects, of assessments of findings the WeBackgroundsummarize 152 assessments met inclusion criteria. The majority of as of majority The criteria. inclusion met assessments 152 We searched PUBMED and other databases through Decem 342 www.jogh.org • doi:10.7189/jogh.07.010902 global journal of health ------www.jogh.org • doi:10.7189/jogh.07.010902 likely insufficient for further substantial reductions in maternal mortality and morbidity [ morbidity and mortality maternal in substantial reductions insufficient further likely for However,facilities. providedhealth is in alone services clinical of delivery increasingquality facility and Traditionally, health maternal programs in low–income settings have focused on improving the access to in parentheses are followed by the letter “S” and a number indicating the order of the reference in Ap in reference the of order the indicating number a and “S” letter the by followed are parentheses in text main the in cited year and assessments the assessments; 152 referencesthese the for contains ment in S1 Appendix analyses. WA,descriptive additional for used was USA) Seattle (Microsoft, Excel Microsoft USA). Georgia, Atlanta, Prevention, and Control Disease for Centers were transferred to an electronic database and initially analyzed in EPI INFO version 3.5.4 (Epi Info, US resolved any discrepancies between the initial two reviews to provide a final summative review. The data reviewer thirdindependent a form; extraction standardizeddata a using assessments fromthese mation A total of 152 assessments met the final inclusion criteria. Two reviewers independently abstracted infor inclusion. for criteria met databases colleague and personal of searchesthroughliterature gray the from identified wereAdditionally, that inclusion. documents for 33 criteria met 120 these, Of 2015. cember Two(HP,authors the of throughPUBMED on De published articles 7890 of abstracts reviewedthe MJ) tion withacommunitycomponentbasedoutsideofphysical healthfacility. interven We health health. a as maternal CBPHC, in defined change targetinga activity an assessed (4) delivery,facility–based attendance, care delivery); antenatal attended (eg, service) maternal key a of age (mortality,health maternal in morbidity,cover change population a or measured status, (3) nutritional tended to improve maternal health; (2) included interventions that took place outside of a health facility; Documents were eligible for inclusion in the present in assessment if they: (1) involved an intervention fect, weusedbroad inclusioncriteria. mental designs, pre–post comparisons, program evaluations, and general descriptions of intervention ef quasi–experi also but trials clinical included only not that documents of set comprehensive a provide order to In inclusion. for considered be to documents of listings further for health public global of field the in organizations and professionals knowledgeable to requests made articles. We review from tified and CABI Publishing Database Subsets to identify additional documents. We included assessments iden system, Review Cochrane the POPLINE, including databases additional searched review maternal our [ series this in elsewhere described are series this for methods reviewshared The comes. maternal health irrespective of inclusion of assessment of outcome on fetal, newborn or child health out effectivenessof sessments broadly, on effectthe intervention assessed CBPHC that a document of any as We conducted a search on PUBMED for assessments of CBPHC on maternal health. We defined such as METHODS [3 complications pregnancy manage to essential is hospitals referral and centers care has received much less attention. Although ready access to and appropriate utilization of health primary health maternal improving to make can (CBPHC) care health primary community–based that tribution to 216 maternal deathsper100 to 216maternal and 2015) was not met: only a 44% decline has been achieved globally – representing a decline from 385 South Asia [ of parts and Africa sub–Saharan slow,in discouragingly particularly been has countries low–income in services has been acknowledged as a global health priority, recent progress in improving maternal health care health maternal of quality the and to access the increasing by health improvingmaternal Although health. in maternal improvements accelerate effortto an in designed been have interventions community–based of number child health also reported in this journal [9 child healthalsoreported inthisjournal effectivenessand the improvingneonatal on in series CBPHC a of of maternal, part is paper Our health. maternal on impact their measured and interventions community–based implemented that projects) as This review is derived from assessments of projects, programs and research studies (hereafter referred to nal health. by providing a broader and more in–depth review of community–based approaches to improving mater [ (2016) al. et Lassi by review published recently a of findings mortality.the extends maternal also It reducing in interventions community–level of effectiveness the of studies controlled to limited was that [ al. et Kidney by review previous a of focus the extends It defined. broadly health maternal proving The purpose of this paper is to review the available evidence regarding the effectiveness of CBPHC in im 5 ]. The Millennium Development Goal 5 (reducing maternal mortality by 75% between 1990 000 livebirthsbetween1990and2015[6 343 – 14].
June 2017 •Vol. 7No. 1• 010902 Online Supplementary Docu Supplementary Online ]. CBPHC andmaternalhealthfindings , 4 ], an increasing an ], 9 1 ]. In addition, addition, In ]. , 2 ]. The con The ]. 8 7 ------] ] -
VIEWPOINTSPaPErS PaPErSVIEWPOINTS June 2017 •Vol. 7No. 1•010902 Jennings etal. primary mortalityandmorbidityoutcomeswillbereservedforasubsequentarticle. primary for conducted delivery.have facility–based we as assessment an detailed as However,in these including delivery,attended care, antenatal of coverage population the and in changes of description brief a make we outcomes, health maternal on interventions community–based of impact the explore Tofully more porting positiveresults. re only avoid to and field the of representation fair a provide to effects insignificant statistically ported re or significance report to failed that documents of Weeffectiveness.descriptions of include evidence of strength the as well as project each by employed interventions the of characteristics key the describe morbidity.and Wemortality maternal of outcomes primary the on interventions of effects the sured ternal articles and present a more detailed descriptive analysis of documents from this database that mea ma of database full the of characteristics powered.basic weresimilarly the comes Herenot describe we power to detect a statistically significant difference in that outcome, while assessments of secondary out sufficient provided that designs study had outcomes primary general, In goals. its and project of type the on depending secondary and primary as indicators outcome defined data extracted who Reviewers Document. ence inAppendixS1OnlineSupplementary author and year followed in parentheses by the letter “S” and a number indicating the order of the refer other paper in this series [ series this in paper other from 1 to 27. (A copy of the data extraction form is contained in Online Supplementary Document of an number in ranging interventions, discrete more or one of effectiveness the described assessment Each Interventions in asinglecommunity. 3% at the national level. 30% of took interventions place in a group of communities, and 9% took place intervention that took place at the district or sub–province level; 8% took place at the province level; and and 10% in urban locations. The largest percentage (48%) of the 152 assessments were performed for an peri–urban in 11% with (83%), communities rural in performed were assessments 152 the of majority ic regions, with the majority of them in South–East Asia (41%) and West Sub–Saharan Africa (22%). The sessments included data from multiple countries in multiple regions. Countries were from six geograph and 11, respectively). Data from a total of 169 countries were included in these 152 assessments. Six as Bangladesh, India, Pakistan and Nepal were the location of the largest number of assessments (16, 15, 14 Community settings r in S1 pendix immunizations, attendance of a skilled attendant at delivery,at attendant skilled a of attendance immunizations, care. of levels higher referralsto making or visits, postpartum and antenatal included activities These care. health maternal routine provided or ed In total, the 152 assessments described 1298 discrete 57% interventions. of these promotinterventions prised ofbetween1and10interventions. com- packages described documents the of (72%) majority a intervention; one only of implementation ESULTS . In the tables, these assessments are cited by the first first the by cited are assessments these tables, the In Document. Supplementary Online
9 ]). As shown in shown As ]). 344 , a small number of assessments (2%) described the the described (2%) assessments of number small a 1, Figure
improving health. maternal healthcarecommunity-based primary in in individualassessmentsofthe effectiveness of Figure 1. Number of interventions implemented implemented Number ofinterventions www.jogh.org • doi:10.7189/jogh.07.010902 ------www.jogh.org • doi:10.7189/jogh.07.010902 kinds ofinterventions. Box 1 describes three examples of intervention packages from three assessments with a larger number of erment programs, women’s andparticipatory groups. including participation in micro–credit and savings groups, conditional cash transfers, women’s empow- mother,the of conditions socio–economic targeted interventions these of 6% infections. and disorders, screening and treating medical conditions such as high–risk pregnancy, gestational diabetes, hypertensive included pregnancy.activities of These complications medical addressed interventions these of 37% Box 1. pregnancy,of cessation of the weeks to related6 cause within fromany woman a or woman pregnant a according assessments of majority the of in defined was death death the definition: WHO to maternal A mortality Impact onmaternal Documentofanotherarticleinthisseries[ assessments isincludedinOnlineSupplementary 152 these among indicators outcome the of list complete A deliveries. facility measured 69% and tendant; care attendance; 66% measured attendance of deliveries by a skilled provider or trained traditional birth at antenatal measured assessments the of 50% morbidity.addition, maternal In measured 21% dicator,and coverage of deliveries taking place at a facility. 19% of the assessments included maternal mortality as an in lation coverage of deliveries by a skilled provider or a trained traditional birth attendant, and (5) population maternal mortality, (2) maternal morbidity, (3) population coverage of antenatal care attendance, (4) popu aids in the assessment of effectiveness.intervention We extracted counts of indicators in five categories: (1) indicators outcome of Categorization indicators. outcome of multitude a described assessments 152 The Categorization ofoutcomeindicators • • • • • • • • • • • Example 3. A package of community–based interventions implemented in four districts of rural Bangladesh [ • • • • • • [16]: ing interventions follow the included India northern in states eight in implemented package community–based A 2. Example • • • • • • • [15]: terventions Example 1. A community–based package implemented in 12 villages in rural India included the following in Facilitate access to clinical services inhealthfacilities Facilitateaccesstoclinicalservices complications Referralformaternal Promotion ofvaccinationsforpregnant women Provision ofantenatalcare, care, delivery andpostnatalcare Promotion ofimmediateand exclusivebreastfeeding Promotion byatrainedhealthworker ofdelivery Promotion ofbirthplanning Identificationofpregnancies atanearlystage Promotion offamilyplanning Training andlinking traditionalbirthattendantstocommunityhealthworkers Formationofvillagehealthcommittees Promotion ofimmediatepostpartum breastfeeding Behaviorchangemessagestopromote savingmoneyforbirthplanningandchildbirth; Provision/promotion ofiron–folic acidtablets Provision oftetanusimmunizations Provision ofantenatalandpostnatalhomevisitsbyhealthworkers Distributionofiron/folate tabletsinthecommunity Provision oftreatment forsimpleillnessesbycommunityhealthworkers Provision ofhomevisitsforantenatalcare bypublichealthnurses Provision ofweeklyantenatalclinicsatoutreach sites Provision atoutreach ofservices sitesbyfacility–basedproviders attendant birth skilled a of use the promotionof planned, is delivery home a if and, facility a at delivery Promotionof level ofcare and workers health community by community the higher in a mothers to high–risk referral of Identification home economicsbycommunityhealthworkers Provision of maternal education on child birth, child care, breastfeeding, immunizations, family spacing, and Examples of community–based intervention packagesforimproving health Examples ofcommunity–basedintervention maternal 345
June 2017 •Vol. 7No. 1• 010902 CBPHC andmaternalhealthfindings 9 ]. 17]: - - - - -
VIEWPOINTSPaPErS PaPErSVIEWPOINTS June 2017 •Vol. 7No. 1•010902 Table 1. Jennings etal. [S41–42] A&B 2007 Guatemala– Curamericas– 2008 [S33] Nicaragua CARE 2008 [S19] Asha–India [S70] Koenig 1988 [S128] Seim 2014 1990 [S50] Fauveau 1991 [S51] Fauveau 1995 [S152] Zhenxuan 2004 [S83] Manandhar 2013 [S39] Colbourn d ocument Effect mortalityoutcomes* size,directionhealthcare andsignificanceofcommunity–basedprimary onmaternal ery/surveillance based impact–oriented care deliv community– and Groups Care campaign mobilization and communication facilities and through community to communities linking through services maternal of quality improve and access Increase communities slum in empowerment women's and care antenatal and primary community–based of Provision family planningservices community–based of Provision cases identify and refer protracted labor to mobilization Community the homeandcommunity in child) and (maternal care tive Provision of primary and preven nity commu and home the in events nity care and of surveillance vital Provision of antenatal and mater vices ser emergency of strengthening community–based and facility– with campaign education health mass community–based Linked cycle action–learning and month per meetings community,9 the with in groups women's Participatory provement athealthcenters im- quality and community the in groups women's Participatory I nterventIon
type
- - - - - Decrease Decrease Decrease Decrease Decrease Decrease Decrease Decrease Decrease Increase e FFect 2007 2002– from years 5 over Guatemala in municipalities rural 3 in age, tive reproduc of women mid–point) (at 11 from size in ranging population Program evaluation of intervention in population of 174 Program evaluation of intervention in years from 2002–2007 5 over Nicaragua in communities rural 173 in age) reproductive of en porting datafrom 2007–2008 ban slums in India, over 20 years, re population of 300 of population intervention of evaluation Program 1978–1981 from years 3 over Bangladesh ral 1976–1985 Matlab, Bangladesh over 9 years from in control, in 79 and intervention 187 from data surveillance demographic using areas control to intervention Quasi–experimental study compared from 2008–2011 years 3 over Niger rural in births 12 assessment, impact Pilot with 196 villages non–intervention boring neigh to compared villages vention Non–randomized evaluation of inter 1978–1981 from years 3 over Bangladesh ral with 196.000 total population, in ru villages non–intervention boring neigh to compared villages vention Non–randomized evaluation of inter 1985–1988 peri–urban China, over 3 years from in county one in year per deliveries paredcontrolto area,8000 covering Quasi–experimental pilot study com years from 2001–2003 28 in clusters community of pairs 12 of trial controlled Cluster–randomized over 29monthsfrom 2007–2010 Malawi, rural in districts 3 in tion, 20 and 14 tion, interven facility to compared nity commu of trial controlled domized ran cluster Two–by–two factorial s tudy 123 (at end evaluation) to 14 to evaluation) end (at 123 931 women in rural Nepal, over 2 523 people in 149 villages, 70 in
populatIon
576 births during interven during births 576
576 births during baseline during births 576 000 total population, in ru 346
367 (58 000 people in ur in people 000
052 wom
254 272 272 ------data) births to zero, over 4 years of live 000 100 per 51124 and zero, to births live 000 100 per (508 control as used data national to relative eas creased ar in all intervention de ratio mortality Maternal country–wide crease of42.6%) de- national a to compared 49.2% of decrease (a well as time that over births live 100 per 60 to 119 from decreased area intervention entire the for ratio mortality maternal intervention; the of years 6 the over liveries de 5500 of average annual an with births, live 000 100 per 34 to 150 from creased de rate mortality Maternal per 100 540 to compared slums Asha in deaths Zero ing age) 10 per 8.6 of rate vs 5.0 of (rate RR 1.73, 95% CI, 1.02–2.93) intervention: over control in vention (authors reported RR inter vs control in mortality maternal of rate lower 42% vs 121 deaths per 100 per deaths 121 vs (66 control vs intervention in maternal mortality rate for decrease overall Significant 100 per 170 to 630 from y, tality ratio fell by 73% over 3 mor maternal Birth–related per 100000livebirths 380 vs 140 or 0.13–0.93), CI, 95% 0.35, (OR: area control to compared vention inter in mortality maternal of odds in decrease 65% fect sizenotreported) (ef ratio mortality maternal in change significant no but age) childbearing of women 100 Maternal mortality ratio (per ( tervention areas and by 5.5% in the in 75.7% by creased live births,respectively e compared to 341 per 100 maternal mortality ratio of 69 a 0.05–0.90), CI, 95% 0.22, pared to control clusters (OR: com clusters intervention ty ternal mortality in communi ma of decreaseodds 78% in 95% CI,0.46–2.57) 1.08, (OR: control to pared ty facili in mortality ternal ma- of odds in increase 8% P FFect + >
000 women of child–bear
0.05) in the control areas omnt am com arm community 000 livebirths
000 live births in India India in births live 000 0 lv brh) de births) live 000
sIze
and www.jogh.org
conFIdence • doi:10.7189/jogh.07.010902
Interval
000 000 000 000 000 ------
verse results) di- testing; tistical sta for sufficiently powered (not N/A country ratio) country overall to pared com areas slum in ratio mortality (maternal N/A vention area) inter hospital ral refer primary the in endline to line fromcreasedbase de rate mortality (maternal N/A P P P P P reported) ( terval in confidence on Significant, based s P IgnIFIcance = < < < < < 0.854 0.05 0.001 0.001 0.05 0.001 P value not
level † ------† Significant results indicatedinboldfont. Document. mentary in S1 Appendix in shown indicator.referencesare outcome full The primary the was mortality maternal which in assessments For * OR–oddsratio N/A –notavailable;RRrateratio,CIconfidenceinterval, www.jogh.org Table 1. [S53] Findley 2015 [S117] Purdin 2009 [S47] Emond 2002 [S73] Lamb 1984 [S54] Foord 1995 d ocument Continued • doi:10.7189/jogh.07.010902 tems integration sys health and change Behavior centers ric centers linked to primary care and creation of emergency obstet Community education campaign community the careProvisionin antenatal of veillance incommunity sur statistics vital and nutrition care, medical direct of Provision lishment ofreferral linkages care in the community, and estab Provision of primary and antenatal I nterventIon described in are so and indicator outcome primary a morbidity,as maternal morbidity discreteassessed a assessed 7 tum hemorrhage (14 assessments), anemia (13), eclampsia (8) or malaria (6). Of these 29 documents that 29 of the 152 assessments measured changes in maternal morbidity, most commonly measuring postpar morbidity Impact onmaternal al ornationalpopulation–levelcomparisons. 1984 [S73]; Emond 2002 [S47]) and the remainder suggesting substantial decreases compared to region Lamb [S41–42]; 2007 A&B Guatemala Curamericas [S19]; 2008 (Asha–India post–intervention deaths maternal zero to reduction a reporting four with impact, substantial mortality.suggested reports These maternal in decrease of suggestion was there assessments these all in but testing, statistical permit not did outcome primary a as mortality maternal with assessments 8 remaining the of design The ument). interval CI 0.51–1.63) 2013, (Colbourn reference [S39] in Appendix 1 in 9% (non–significant) reduction in oddsmortality ofrate maternal (odds ratio OR 0.91, 95% confidence a with control, the against itself by considered was arm community the when direction in reversed fect was analyzed together with the community arm, in comparison to the control arm. The suggestion of ef 1 42% to 78% and 1 suggested a trend toward increased mortality but this effect was not significant ( ranging from mortality in decreases significant reported 6 testing, significance statistical performed that ( indicator outcome primary a as mortality mortality,assessed ternal 15 ma assessed that documents 32 the Of causes. accidental excluding but management, its or pregnancy none reported a worsening of maternal morbidity.none reported aworsening ofmaternal nal morbidity indicators; one assessment suggested a decrease but did not report significance testing, and ). This suggestion of an increased mortality effect was only present when the facility–based intervention
type Table 2. Six of these assessments reported a significant decrease in at least one of the mater - - - - Decrease Decrease Decrease Decrease Decrease e FFect years from 2009–2013 in 3 states in northern Nigeria over 4 en at baseline and 4628 at follow–up, fore compared to after, of 2360 wom vention compared to control and be Non–randomized evaluation of inter 2004–2007 from years 4 over Pakistan in gees community of 96 of community among intervention of evaluation non–controlled Non–randomized 1995–1997 from months 30 over Brazil in trict population of 42 a in intervention an of evaluation non–controlled Non–randomized from 1974–1984 years 10 over Gambia rural in 2000, of population total with villages 4 in impact intervention of evaluation non–controlled Non–randomized 2 yearsfrom 1989–1991 in a rural district of the Gambia over area, each with a population of 1300, control similar to compared vention Non–randomized evaluation of inter s tudy
populatIon 347
000 in an urban dis
300 Afghan refu Afghan 300 ------100 per (controls) 1262 to and 1270 to 1057 (interventions) from communities, control in the intervention than in the ratio showed a larger decrease Estimated maternal mortality over 4years births live 000 100 per 102 to 291 from decreased ratio mortality maternal Annual e tervention during the 1 year after the in tion, to zero maternal deaths interven to prior births live creased from 335 per 100 000 de ratio mortality Maternal intervention areasintervention ing rates in comparable non– us expected be would that 16 annual the to compared intervention, of years 8 the served in the community for ob were age) bearing child of women 1000 (per deaths pregnancy–related No in control area 700 to compared tervention 100 per 130 of ratio mortality maternal a giving area, trol compared to 5 deaths in con area intervention in death 1 FFect 000 livebirths 000 live births in the in
sIze
and June 2017 •Vol. 7No. 1•010902 ). Of the 7 assessments assessments 7 the Of Table1).
Online Supplementary Online Doc- Supplementary conFIdence CBPHC andmaternalhealthfindings
Interval ------mates) N/A (based on esti- separate sources) two from culated cal rates endline and (baseline N/A s tistical testing) sta for sufficiently powered (not N/A tistical testing) sta for sufficiently powered (not N/A tistical testing) sta for sufficiently powered (not N/A IgnIFIcance Online Supple- Online
level Table † ------
VIEWPOINTSPaPErS PaPErSVIEWPOINTS June 2017 •Vol. 7No. 1•010902 Table 2. Jennings etal. ‡PPH definedineachassessmentas bloodloss≥ †Significant results indicatedinboldfont. Document. mentary indicator.outcome primary a as morbidity maternal referencesanalyzed full that in The assessments S1 *For Appendix arein shown SD–standardCI –confidenceinterval, deviation,OR–oddsratio,PPHpostpartumhemorrhage, NS–not(statistically)significant,RRrate ratio §Chi–square testofdifference betweencontrol and intervention. Incidence ofpostpartumhemorrhage(PPH) r [S45] 2006 Derman [S95] 2011 Mobeen [S138] 2013 Stanton [S90] 2008–5 Mbonye Prevalence ofmalariaandanemiaintreatment interventions [S89] 2008–3 Mbonye [S105] 2009 Ndiaye [S130] 2005 suddin Sham- Eclampsia eFerence Effect morbidityoutcomes* size,directionhealthcare andsignificanceofcommunity–basedprimary onmaternal I placebo) at home births they attended (or misoprostol oral administered (ANMs) midwives nurse Auxiliary tended at they deliveries home at placebo) (or misoprostoladministered (TBAs) attendants birth Trainedtraditional home birthstheyattended prophylactic oxytocin (or placebo) at injected officers health Community pared to routine care in health clinics community,com the in pregnancy in malaria for (IPT) treatment tive ers administered intermittent preven work- health community of cadres 4 health clinics cy in the community, compared to in tive treatment for malaria in pregnan ers administered intermittent preven work- health community of cadres 4 control anemiaduringpregnancy and to distribute iron supplements, to nutrition, and health maternal mote community–based volunteers to pro using program deviance Positive not receive injections did who cases control to compared hospital, to referral to prior cases nosed eclamptic and severe eclamptic diag to sulfate magnesium of tion community, home–based administra involving study Quasi–experimental nterventIon loss greater than or equal to 1000mL (Kapungu 2013 [S65]; Fauveau 1990 [S50]; Derman 2006 [S45]). blood as hemorrhage postpartum severe defined and 500mL, to equal or than greater loss blood sured vention delivered by a community health worker. These documents used the standard definition of mea Three of the seven documents measured change in postpartum hemorrhage following a preventive inter Postpartum hemorrhage
type
------500 mL;severe PPHdefinedineachassessmentasbloodloss≥ e Decrease Decrease Decrease Decrease A non–randomized communi non–randomized A Decrease Decrease A non–randomized communi non–randomized A Decrease Decrease Decrease FFect prove- ment) (im- p 2005 2002– from years 3 over India rural in ANMs, 25 by delivery home–based after placebo to en to oral misoprostol and 808 trolled trial assigned 812 wom placebo–con randomized A from 2006–2007 months 24 over Pakistan ral 81 TBAs, in one province in ru by delivery home–based after misoprostol and 585 to placebo oral to women 534 signed as trial placebo-controlled double–blind randomized A from 2011–2012 897 in control, over 19 months and intervention in 689 na, ficer, in 4 rural districts in Gha randomized to study arm by of community health officers were 54 by conducted births signed randomized controlled trial as- cluster– community–based, A months from 2003–2005 21 over Uganda rural central, in district one of counties sub– 9 in control to munities) com (4 women 704 and tion interven to communities) (21 women 2081 assigned trial ty months from 2003–2005 21 over Uganda rural central, in district one of counties sub– 9 in control to munities) com (4 women 704 and tion interven to communities) (21 women 2081 assigned trial ty gal over9monthsin2003 Sene rural in community one in women 371 assessed ples sam cross–section dependent using pre–post evaluation of in design quasi–experimental A months in2001 tion and 132 in control, over 6 interven in 133 Bangladesh, in districts 3 from cases 256 opulatIon 348
------e phylaxis every 18 women given chemopro for prevented PPH case 1 0.91). 0.04– CI: 95% 0.16, RR: control, in 1.2% vs intervention in (0.2% 0.74); 83% decrease in severe PPH 0.39– CI: 95% 0.53, RR: control, (6.4% in intervention vs 12.6% in 47% decrease in incidence of PPH 95% CI:0.27–1.22) 0.57, (RR: PPH severe in crease de Insignificant 0.59–0.97); CI 95% 0.76, RR: control, in 21.9% vs intervention in (16.5% ery 24% reduction in PPH after deliv trol group) con in 8 intervention, in case (1 PPH severe in change significant RR: 0.49, 95% CI: 0.27–0.88) No in intervention vs 5.5% in control, (2.6% PPH in 51% of Reduction based control (P IPT compared to 39.9% in facility– received intervention nity–based commu the in women of 67.5% (both control in 13.1% to 39.1% from and intervention in 17.6% to 49.5% from creased de- episodes malaria of Prevalence (both P 13.1%) to 39.1%, from cilities, fa in decrease 66% vs 17.6%, to community,in 49.5% from(64% nity and facility distribution of IPT commu- both in episodes malaria reported of prevalence Decreased cases (6.86 ± 0.12–0.53) itive deviance) (OR: 0.25, 95% CI: compared to control area (no pos- intervention the in surements, mea- hemoglobin mean on based anemia, of risk in reduction 75% (4.7 cases intervention the in vulsions con mean of number Decreased FFect ±
sIze SD2.64) compared to control <
and 0.001)
conFIdence SD 2.97)(P www.jogh.org < 1000 mL. 0.001)
Interval P < < • doi:10.7189/jogh.07.010902 0.001) 0.001). 0.001). ------s P severe PPH PPH P PPH P PPH P was notreported reported malaria RR difference in Significance for P was notreported) reported malaria RR difference in [Significance for P P P PPH, Severe PPH‡ IgnIFIcance < < < < < Online Supple- Online 0.001 0.001; 0.001 0.001 0.003§ < < < 0.001, 0.001, 0.05; NS 0.05; NS
level † - - www.jogh.org • doi:10.7189/jogh.07.010902 hemorrhage. postpartum severe on effect significant a have not did two remaining the and hemorrhage, postpartum Malaria ( 66% to 24% from ranging creases de with significant, statistically were hemorrhage postpartum in reduction of measurements three The munities, butthedifference wasnotstatisticallysignificant. promotion of maternal ANC attendance. ANC attendance increased in both control and intervention com ods evaluation of integrated community case management for childhood illness that was combined with The third assessment that found no change in ANC coverage (Langston 2014, [S74]) was a mixed–meth rate, orneonatalmortality. delivery,institutional stillbirth as care–seeking, breastfeeding,such outcomes other or attendance ANC improvementsin no werethere 0.38–0.94), CI 95% control0.60, groupto (OR compared intervention the in morbidities poor.maternal of set a of odds reportreductionthe did a in assessment the Although urban the for program care health newborn and maternal city–wide a through strengthened been had India for the purpose of improving perinatal health, including increasing attendance at ANC clinics which controlled trial testing the impact of creating and mobilizing women’s groups in urban slums in Mumbai, The second assessment with no change in ANC attendance (More 2012 [S97]) was a cluster–randomized ing mostrecent pregnancy. iron/folicof ingestion anemia, of knowledge as such cators reportedand supplements, acid dur anemia non–recipient (control) barrios showed some increases and some decreases on numerous outcome indi and (intervention) recipient Both burden. anemia high a with communities to education health lated an evaluation of a pilot program in Mozambique that provided iron and folic acid along with anemia–re The first assessment that found no change in ANC coverage (Helen Keller International 2003, [S60]) was here insomedetail. three those describe we and coverage, in change no found assessments Three coverage. ANC in crease de a observed assessments No (ANC). care antenatal for attendance increasedreported assessments 34 indicator,outcome primary a as attendance antenatal of coverage measured that assessments 37 the Of Antenatal care Impact onpopulationcoverageofevidence–basedinterventions to caseswhodidnotreceive injectionsathome(Shamsuddin2005,[S130]). magnesium sulfate injections at home prior to hospital transfer, reporting a significant decrease compared received who cases pre–eclamptic or eclamptic in convulsions of frequency measured assessment One Eclampsia (Ndiaye 2009[S105]). prevalence of anemia, reporting a significant decrease of 75% in the interventionarea vs the control area the assessed report One [S89]). 2008–3 (Mbonye risk in difference relative the of testing significance report not did but groups, distribution facility and community both in decreases similar reported and IPT (Mbonye 2008–5, [S90]). The second assessment measured prevalence of reported malaria episodes of use and to access increased approach community–based The arm. facility the to compared arm nity commu the in doses two recommended the to adherence higher significantly described results the and community vs 76.1% in the facility, mended (during the second trimester) more frequently than the women in the facility arm (92.4% in the recom as IPT of dose first their received arm community the in women However,the [S90]). 2008–5 of malaria in pregnancy, (mean hemoglobin increased by 6.7% with 2 doses of IPT in both arms) (Mbonye decreases in anemia in both community–based and facility–based intermittent preventive treatment (IPT) but were reported in separate peer–reviewed publications. One document reported equivalent, significant (one assessment). Of note, two of these assessments pertained to different aspects of a single intervention episodes malarial maternal of prevalence the and assessments) (two areas malaria–endemic in anemia Two assessments reported measures outcomes related of primary to malaria, including the prevalence of nificantly earlier stage of pregnancy compared to those in the facility arm (21 weeks vs 23 weeks, P Table2 < 0.001). Women in the community arm also received IPT at a sig 349 ). One assessment reported a significant decrease in severe in decrease significant a reported assessment One ).
June 2017 •Vol. 7No. 1• 010902 CBPHC andmaternalhealthfindings P < 0.001), 0.001), ------
VIEWPOINTSPaPErS PaPErSVIEWPOINTS June 2017 •Vol. 7No. 1•010902 Jennings etal. to increase the coverage of insecticide–treated bed nets for pregnant women and to expand the coverage Community–based approaches, particularly through home visits provided by CHWs, are commonly used munizations. cility were also a common approach to provide prenatal care, family planning services, and maternal im fa health peripheral a at based team health mobile a by community care.the stetrical Outreachto visits emergencyand ob delivery antenatal, seek to families their and women motivate to strategy common a with problems requiring referral. The formation andstrengtheningwomen’s ofparticipatory groups was mothers postpartum identify and care maternal postpartum provide to used also ery.were visits Home education/counseling, as well as to promote healthy behaviors such as family planning and facility deliv and services health provide to women, pregnant identify to used strategy common a were visits Home • • • • • for improving neonatalandchildhealthare reported elsewhere [ strategies to improve neonatal and child health. Strategies to implement community–based interventions community–based with overlap often health maternal strengthen to used strategies Community–based • • • • • A typicalsetofimplementationstrategiesisthefollowing(Baqui2008[S24]): discussedaboveareCommon strategiesusedtoimplementtheinterventions highlightedhere. Implementation strategies non–CHW membersofthelocalcommunity(40assessments). CHWs were most frequently combined with local government health officials (69 assessments), and with ects). Multiple categories of implementers were present in three–fourths (71%) of the individual projects. proj (4/152 expatriates projects), and project (31/152 the implement to researchspecifically staffhired health professionals (78/152 projects), local community members not trained as CHWs (48/152 projects), projects included in our database. In addition to CHWs, project implementers included local government 152 the of 132 in implementation intervention in involved were (CHWs) workers health Community Implementers centers, andbirthinghuts. health hospitals, were assessments these in included facilities of types The increase. an reported seven tor. None of these assessments observed a decrease in coverage; one observed no change in coverage and Eight assessments measured the percentage of births occurring in a facility as a primary outcome indica Changes infacility–baseddeliveries calculated thestatisticalsignificanceofcoveragechangesfoundasignificantincrease. that assessments two The deliveries. unattended completely to compared as attendant birth traditional liveries attended by a trained traditional birth attendant. Two assessments measured the attendance by a de of percentage the measured assessment one while attendant,” birth “skilled a by attended deliveries of percentage measured specifically assessments Nine here. definition the standardize to attempted not The precise definition of a skilled or trained birth attendant varied among the assessments, and we have deliveries. attended of coverage the in increase an reportedindicator. assessments outcome 12 mary All pri a as delivery at attendant trained or skilled a of presence the of coverage measured assessments 12 Changes inattendeddelivery Engagedcommunitiesintheplanningand/orevaluation ofCBPHCprogramming EngagedcommunitiesintheselectionandsupportofCHWs Engagedcommunityleaderstomobilizecommunitiesforahealth–related activity Establishedcommunitycollaborationssuchastheformationofhealthcommittees Usedhomevisitsandhealthcenterstodeliverinterventions Deliveredandpromoted services through interventions bothskilledandtraditionalhealthworkers care Integratednutritionwithprimary services interventions andnewborn Combinedmaternal ofhealthinfrastructure ministry (facilitiesandpersonnel) Usedexistinggovernment cro–credit savingsgroups) women’smi (including and groupsgroups existing with collaborated or groups community Formed
350 13]. Thesecommonstrategiesinclude: www.jogh.org • doi:10.7189/jogh.07.010902 ------www.jogh.org • doi:10.7189/jogh.07.010902 reaching the 2015 Millennium Development Goals for maternal health [ health maternal for Goals Development Millennium 2015 the reaching Despite these observations, promoting facility deliveries has been a focus of many interventions aimed at nity–level workers. facilities (by improving logistics and training staff), and strengthening the supervisory system of commu- health peripheral at provided care of quality the strengthening included interventions CBPHC mented imple- also that projects by out carried often activities health–system–related Other arise). emergencies tems, and community savings or insurance schemes to cover transport and hospital costs when obstetric facilitating referrals (by forming community emergency response committees, community transport sys include health maternal improving for CBPHC with associated strategies strengthening systems Health home visitsandmobileoutreach teams. crease the coverage of anti–retroviral treatment of HIV–positive pregnant women include CHWs making in- to and infection HIV with women of detection approachesthe Community–based expand comes. to out neonatal and perinatal improving for also but outcomes maternal improving for only not effective of intermittent preventive treatment of malaria in malaria–endemic areas. These are interventions that are time to support increased contacts [ over evolved delivery.have carefacility–based antenatal and at attendance for recommendations Global focusedonthreeantenatalcaretrainedprovider,based interventions interventions: attendance,delivery Our analysis of the effectiveness of CBPHC approaches in increasing the population coverage of evidence– cant positiveeffect. signifi- a showed eclampsia on interventions CBPHC of assessment one and effects, positive significant showed all malaria–relatedanemia and malaria maternal on interventions CBPHC of Threeassessments of incidence the postpartum hemorrhage showed significant decreases. One of the three showed a significant decreasein All three assessments of the statistical significance of impact of CBPHC interventions on the incidence of and allowednon–randomizedassessmentsaswellfrom thegrayliterature. [ mortality maternal in reduction no but morbidity maternal in tions reduc reported which interventions, community–based of impact the of review Cochrane 2010 a to trast tistical significance of the impact. All eight of these reported a favorable effect on maternal mortality. In con were eight additional assessments that reported trends in maternal mortality but could not measure the sta- There decrease. significant statistically a showed mortality maternal in change the of significance statistical the measure to able were that assessments seven the of Six settings. selected in mortality maternal in tions bidity, increasing population coverage of evidence–based interventions, and possibly contributing to reduc mor maternal reducing on interventions CBPHC of impact positive a for evidence provides analysis This DISCUSSION access to emergency and critical care for prompt attention to life–threatening maternal complications [ [ attendance birth skilled increase to important are system countries for all births. Strategies that integrate both skilled and traditional birth attendants into the health many of capacity short–term the beyond often provideris skilled highly other or midwife trained mally Delivery attended by a skilled provider improves delivery outcomes [20], but delivery by a fully and for cy care inresource–constrained settings. vention may help to expand the coverage of more frequent, high–quality and woman–centered pregnan birth attendants, who can serve as doulas (birth companions for facility births) and collaborators in the the in collaborators and births) facility for companions (birth doulas as serve can who attendants, birth traditional as well as process care maternal the in them incorporate to efforts continue to important is this study were implemented by a wide variety of different types of community–based health workers. It The vast majority of community–based primary care interventions described by assessments included in delivery, andfacilitydelivery. important to strengthen community–based interventions to promote antenatal care attendance, attended [4 facility fromhealth away a hour one morethan live globally largepopulations A low–income proportionthe of byitselfisinsufficientdelivery mortalityratios[ to reduce maternal 2030 Sustainable Development Goals. However, recent literature suggests that a high rate of institutional some argue that facility delivery is not a necessary requisite for the reduction of maternal mortality [ although the literature points out deficiencies in quality that are commonly at observed facilities [ ], making utilization of health facilities and emergencyand facilities health of utilization Thereforemaking challenge. ], a care services is it postpartum hemorrhage (which was a secondary outcome for all three projects).three all for outcome secondary a was severe(which hemorrhage postpartum 19], and the provision of antenatal care as a community–based inter 351
21, ]. Delivery at a health facility improves facility health a at Delivery 22]. ], our inclusion criteria were broad were criteria inclusion our 18], 1 , 26]. June 2017 •Vol. 7No. 1•010902 ] and now for reaching the the reaching for now and 25] CBPHC andmaternalhealthfindings 23 2 ] and ] and , 24 3 ], ], ]. ]. ------
VIEWPOINTSPaPErS PaPErSVIEWPOINTS June 2017 •Vol. 7No. 1•010902 Jennings etal. es for improving neonatal and child health care has been summarized [ summarized been has care health child and neonatal improving for es ly dependent on the context. Although evidence of the cost–effectiveness of community–based approach therefore levels of maternal health are already high. Thus, the cost–effectiveness of CPBHC may be high ternal health – for example in settings where high–quality facility–based care is available and utilized and important to note that there are certain settings in which CBPHC may not be effective in improving ma Additionally,rare. is are it health maternal improving to approaches community–based of effectiveness This review did not focus on assessments of cost–effectiveness. It is worth noting that studies of the cost– statistical significance). reach not did it size sample small the to due but reality in achieved been have may impact an that ing powered(mean adequately not was change significant statistically a show not did that study the that is in maternal mortality in one setting [ the literature in which the same community–based intervention shows a statistically significant reduction in Additionally, examples are significance. there statistical reach not does decline the but impact, tality As our findings indicate, there are numerous assessments in which there is a suggestion of maternal mor demonstration of a statistically significant decline in maternal mortality is a challenge for field programs. a maternal mortality ratio of 1000, only 1% of live births are associated with a maternal death. Thus, the mortality is a rareMaternal event, even in settings wheremortality is relatively maternal high: even with Study limitations from malariaandhemorrhage followinghomedelivery. delivery are needed,aspointedoutatseveralpointsinthis paper. richness of this data set is such that only a limited analysis of the data is provided here. Further analyses Finally,conclusions. detailed draw to ability the our limits measurement indicator of standardization of However, no studies assessing mHealth interventions were identified for our review. In addition, the lack reproductive andpromote agehavethepotentialtolink clientswithservices [ utilizationofservices new For interventions. example, mhealth strategies involving community health workers and women of [33, nents other authors have summarized potential frameworks to select appropriate intervention package compo However,approaches. community–based in components intervention against or for recommendations specific make to able being from us precludes packages intervention of variety wide a of inclusion Our have beenevenmore compelling. would health maternal improving in CBPHC of effectiveness the for evidence the that view,expect we 32]. Had assessments of the effectiveness of community–based family planning been included in our re effectivelybe can planning ily provided care[ health platform through primary community–based a simply by reducing the number of women who become pregnant). There is extensive evidence that fam important for improving maternal health (by, among other things, reducing the number of maternal deaths is planning family that review.evidence our extensive in However,is included there assessments the in health becausetheireffects are indirect andnotreadily measured inspecificprogram settings,including maternal on interventions planning family community–based of effects the assess not did review This er, toadequatelyexplore theseissuesisbeyondthescopeofthispaper. important in assessing how CBPHC can most effectively contribute to improve health. maternal Howev care,maternity the availability and utilization of health facilities, and the local geographic context are all provide to personnel trained of availability The setting. which in useful most are components CBPHC The local context in which the assessments were carried out is important to more fully understand which search onthecost–effectiveness healthinterventions. ofcommunity–basedmaternal improving neonatal healthare importantaswellbutsummarizedinanother articleinthisseries[ and referral of obstetrical emergencies. Finally, the closely related contributions that CBPHC can make to childbirth, ANC, promotionfor appropriatethe utilization of to facility contributions its of because also mortality, maternal in reductions but to contribute to potential its and morbidity maternal reducing on have can it effectsdirect the of because only not – program health maternal comprehensively–designed The evidence provided here supports the recommendation that CBPHC is an important component of a CONCLUSIONS [ ]. Community–based interventions show great potential for reducing morbidity of mothers mothers of morbidity reducing for potential great show interventions Community–based 27]. ]. The nature of intervention packages evolves with technology and with the emergence of of emergence the with and technology with evolves packages intervention of nature The 34].
28] but not in another [ 352 29]. One of the explanations for this finding www.jogh.org 8 ], there is a need for more re more for need a is there ], • doi:10.7189/jogh.07.010902 10]. ]. 35]. 30– ------www.jogh.org rEFEr ENCES • doi:10.7189/jogh.07.010902 14 13 12 11 10 9 8 7 6 5 3 2 1 4 database description.JGlobHealth.2017;7:010901. and methods rationale, 1. health: child and neonatal maternal, improving in care health primary community-based of Perry H, Rassekh B, Gupta S, Wilhelm J, Freeman P. A comprehensive review of the evidence regarding the effectiveness pdf?sequence=3. Accessed:26April2017. World Bank; [263-94]. Available: ease Control Priorities: Reproductive, Maternal, Newborn, and Child Health, Third Edition [Internet]. Washington, DC: Dis In: 2016. health. child and careCommunity-based ZA. Bhutta improveR, to Kumar newborn, ZS, Lassi maternal, doi:10.1186/1471-2393-9-2 ee itretos o eue aenl otlt. M Penny hlbrh 2009;9:2. Childbirth. Pregnancy BMC mortality. maternal reduce to interventions level Kidney E, Winter HR, Khan KS, Gulmezoglu AM, Meads CA, Deeks JJ, et al. Systematic review of effect of community- 6736(15)00838-7 Maternal Mortality Estimation Inter-Agency Group. Lancet. 2016;387:462-74. UN the by analysis systematic a 2030: projectionsto scenario-based with 2015, and 1990 between mortality maternal trendsand regional,levels in Global, national al. and et A, Gemmill AB, Moller S, Zhang D, Hogan D, Chou L, Alkema int/reproductivehealth/publications/monitoring/maternal-mortality-2015/en/ WorldUNICEF,WHO, Available:Bank, 2015. 2015. to 1990 UNFPA.Mortality: Trends Maternal in es in areas of high maternal mortality in Africa. Int J Gynaecol Obstet. 2006;92:308-19. Kayongo M, Rubardt M, Butera J, Abdullah M, Mboninyibuka D, Madili M. Making EmOC a reality–CARE’s experienc India. SocSciMed.2007;64:2083-95.Medline:17374551 Hulton LA, Matthews Z, Stones RW. Applying a framework for assessing the quality of maternal health services in urban line:23826302 is there an association between institutional birth proportion and mortality? maternal PLoS One. 2013;8:e67452. Randive B, Diwan V, De Costa A. India’s Conditional Cash Transfer Programme (the JSY) to promote institutional Birth: Int JGynaecolObstet.2005;88:208-15.Medline:15694109 Pearson L, Shoo R. Availability and use of emergency Sudan, Kenya, and Rwanda, Uganda. obstetric Southern services: ijgo.2005.12.003 term impacton mortality inchildren youngerthanfiveyears ofage.JGlobHealth.2017;7:010309. long- of evidence with programs 7. health: child and neonatal maternal, improving in care health primary nity-based Perry H, Rassekh B, Gupta S, Freeman P. A comprehensive review of the evidence regarding the effectiveness of commu J GlobHealth.2017;7:010906. nity-based primary health care in improving maternal, neonatal and child health: 6. strategies used by effective projects. Perry H, Rassekh B, Gupta S, Freeman P. A comprehensive review of the evidence regarding the effectiveness of commu fects. JGlobHealth.2017;7:010905. ef equity 5. health: child and neonatal maternal, improving in care health primary community-based of effectiveness Schleiff M, Kumapley R, Freeman P, Gupta S, Rassekh B, Perry H. A comprehensive review of the evidence regarding the findings. JGlobHealth.2017;7:010904. health child 4. health: child and neonatal maternal, improving in care health primary community-based of fectiveness Freeman P, Schleiff M, Sacks E, Rassekh H. B, A Gupta comprehensiveS, Perry review of the evidence regarding the ef health findings.JGlobHealth.2017;7:010903. the effectiveness of community-based primary health care in improving maternal, neonatal and child health: 3. neonatal FreemanP,E, Sacks regardingevidence the comprehensivereview of A al. et S, Gupta B, Rassekh M, Jennings K, Sakyi Acknowledgments: disclosure.pdf (availableuponrequest from thecorresponding author),anddeclare noconflictofinterest. www.icmje.org/coi_ at Form Interest Competing Unified the completed have authors All interests: Competing guided itsanalysis.Alloftheauthorsparticipatedinrevision ofearlierdraftsandapproved thefinaldraft. contributions: Authorship ecution ofthereview. ex the in role no had support financial provided that organizations The Foundation. Gates the and velopment, WorldThe ticle: UNICEF,Organization, Health De International for Agency WorldStates the United the Bank, ar this in described work the conduct to used were that funds providedorganizations following The Funding: as amemberoftheStudyTeam. phase ofthereview. TheWorld Bankmadeitpossibleforoneofitsconsultants,DrBahieRassekh,toparticipate initial the during Perry Dr to support salary and support, administrative space, office providing for Generations and particularly Health its Section International staff, which administered some of these funds. We thank Future Association Health Public American the to grateful Wealso Foundation. are Gates the and Generations, Future ment of the World Health Organization, the CORE Group (Collaboration and Resources for Child Health)/USAID, Develop and Health Adolescent UNICEF,and review: Child this of of Department Worldpenses the the Bank, doi:10.1371/journal.pone.0067452 We are grateful to the following organizations that provided small grants to cover the ex the cover to grants small provided that organizations following the to grateful Weare https://openknowledge.worldbank.org/bitstream/handle/10986/23833/9781464803482. MJ wrote the first draft. MJ, SP and MS collected the evidence for this review and and review this for evidence the collected MS and SP MJ, draft. first the wrote MJ 353 doi:10.1016/j.socscimed.2007.01.019 doi:10.1016/j.ijgo.2004.09.027 . Accessed:26April2017. Medline:26584737 June 2017 •Vol. 7No. 1•010902 Medline:16442536 CBPHC andmaternalhealthfindings Medline:19154588 doi:10.1016/S0140- http://www.who.
doi:10.1016/j. Med------
VIEWPOINTSPaPErS PaPErSVIEWPOINTS June 2017 •Vol. 7No. 1•010902 Jennings etal. rEFEr ENCES 21 20 19 18 17 16 15 35 34 33 32 30 31 29 28 27 26 25 24 23 22 birth attendance.IntJGynaecolObstet.2011;115:127-34.Medline:21924419 skilled increase can systems health formal with attendants birth traditional of integration the How MorganA. A, Byrne midwifery. Lancet.2014;384:1146-57.Medline:24965814 up scaling of effect projected The al. et P,AM, Hoope-Bender Speciale ten J, MA, Sandall Dias IK, Friberg CS, Homer health/9241590947/en/. Accessed:26April2017. Available: 2003. differentials. and levels WHO. Antenatal care in developing countries. Promises, achievements and missed opportunities: an analysis of trends, ity ity and mortality and improving neonatal outcomes. Cochrane Database Syst Rev. 2010;11:CD007754. morbid neonatal and reducingfor maternal packages intervention Community-based ZA. Bhutta BA, Haider ZS, Lassi line:26340672 2015;10:e0136898. One. PLoS Bangladesh. Rural in Program (IMNCS) Survival Child Neonatal Maternal Improved Based Community Assessing al. et R, T,Shah Ishaque TR, Chowdhury SK, FT, Mistry Jhohura M, Rahman 2008;86:796-804. Organ.World Health Bull India. northern rural in mortality neonatal on programme health Baqui A, Williams EK, Rosecrans AM, Agrawal PK, Ahmed S, Darmstadt GL, et al. Impact of an integrated nutrition and ry. IndianPediatr. 1997;34:785-92.Medline:9492416 Dutt D, Srinivasa DK. Impact of maternal and child health strategy on child survival in a rural community of Pondicher line:18949217 GHSP-D-13-00031 common applications and a visual framework. Glob Health Sci Pract. 2013;1:160-71. Labrique AB, Vasudevan L, Kochi E, Fabricant R, Mehl G. mHealth innovations as health system strengthening tools: 12 s12884-015-0784-9 2015;15:337. PregnancyChildbirth. BMC analysis. landscape health–a newborn and maternal improving for approaches Innovative K. Dickson VeselJ, A, Kim Simen-Kapeu L, A, Higgins-Steele K, Lunze 8. Medline:26694075 improvingavailability, accessibility, low-resourcein services of quality and acceptability 2016;123:540- BJOG. settings. TenJM, Caglia for AD, P,models Hoope-Bender Kearns innovative of review a care: postnatal and Antenatal A. Langer doi:10.1016/j.contraception.2009.10.006 Medline:20159172 2010;81:181-4. Contraception. consultation. technical a from conclusions contraceptives: able Stanback J, Spieler J, Shah I, Finger WR. Community-based health workers can safely and effectively administer inject 2014. 2014. Available: Health Reproductive and Sexual in Investing of Benefits and Costs The Up: It Adding L. Ashford J, Darroch S, Singh Contraception. 2005;72:402-7.Medline:16307960 VahidniaN, Prata F, Dries-DaffnerM, Potts programs:distribution community-based Revisiting I. are needed? still they 2017. line:20207412 2010;375:1193-202. Lancet. trial. controlled cluster-randomised a Bangladesh: in districts rural three in Effectal. et AR, Rego K, Khan S, Shaha B, Banerjee S, women’sup scaling of Barnett K, Azad groups outcomes birth on Medline:15364188 2004;364:970-9. Lancet. trial. controlledcluster-randomised Nepal: in outcomes birth women’son with groups tion Manandhar DS, Osrin D, Shrestha BP, Mesko N, Morrison J, Tumbahangphe KM, et al. Effect of a participatory interven 31. Medline:27016548 of the Introduction and Utilization of Birthing Facilities by an Indigenous Population. Glob Health Sci Pract. 2016;4:114- Stollak I, Valdez M, Rivas K, Perry H. Casas Maternas in the Rural Highlands of Guatemala: A Mixed-Methods Case Study study. Lancet.2013;381:1747-55.Medline:23683641 cross-sectional a Health): Newborn and Maternal on Survey Multicountry WHO (the mortality maternal of reduction Souza JP, Gulmezoglu AM, Vogel J, Carroli G, Lumbiganon P, Qureshi Z, et al. Moving beyond essential interventions for whr/2005/whr2005_en.pdf?ua=1. Accessed:26April2017. Available:2005. count. child and mother every make - 2005 Report WorldHealth The WHO. line:12834953 87-8. discussion 2003;82:89-103, Obstet. Gynaecol J Int Republic. Dominican the of paradox the liveries: Miller S, Cordero M, Coleman AL, Figueroa J, Brito-Anderson S, Dabagh R, et al. Quality of care in institutionalized de line:17071268 mortality.2006;368:1477-9. maternal Lancet. reduce to strategy alternative An S. Barnett K, Azad A, Costello doi Lane K, Garrod J. The return of the Traditional Birth Attendant. J Glob Health. 2016;6:020302. Health. Glob J Attendant. Traditional Birth the of return The J. Garrod K, Lane :10.7189/jogh.06.020302 doi:10.1371/journal.pone.0136898 doi:10.2471/BLT.07.042226 doi:10.1016/S0140-6736(10)60142-0 doi:10.1016/S0020-7292(03)00148-6 doi:10.1016/S0140-6736(06)69388-4 doi:10.1016/S0140-6736(04)17021-9
doi:10.1111/1471-0528.13818 https://www.guttmacher.org/sites/default/files/report_pdf/addingitup2014.pdf doi:10.9745/GHSP-D-15-00266 http://www.who.int/reproductivehealth/publications/maternal_perinatal_ 354 doi:10.1016/j.contraception.2005.06.059 doi:10.1016/S0140-6736(13)60686-8 doi:10.1016/S0140-6736(14)60790-X doi:10.1016/j.ijgo.2011.06.019 www.jogh.org Medline:26679709 Medline:25276529 • doi:10.7189/jogh.07.010902 http://www.who.int/ Medline:27606054 Medline:21069697 . Accessed. 26 April
doi:10.9745/ doi:10.1186/ Med- Med Med- Med Med ------
www.jogh.org child health:3.neonatalhealthfindings health careinimprovingmaternal,neonataland the effectivenessofcommunity–basedprimary Comprehensive reviewoftheevidenceregarding material. The onlineversionofthisarticlecontainssupplementary Electronic material: supplementary Freeman Jennings Gupta rassekh Sakyi 5 4 3 2 1 Emma Sacks [email protected] USa Baltimore, MD21205 615 NorthWolfeSt. School ofPublicHealth Johns HopkinsBloomberg r Henry Perry Correspondence to: Independent Consultant, Lusaka, Zambia Medical Epidemiologist, District ofColumbia,USa The WorldBank,Washington, Seattle, Washington,USa University ofWashington, Department ofGlobalHealth, USa Health, Baltimore,Maryland, Bloomberg SchoolofPublic Health, JohnsHopkins Department ofInternational Seattle, Washington,USa oom E8537 1 , MaryCarol 5 • doi:10.7189/jogh.07.010903 , HenryBPerry 4
2,3 , Sundeep 1
, BahieM , Kwame 1 , Paula
1
packages of neonatal interventions andofprogramspackages ofneonatalinterventions atscale. integrated of needed areAdditionally, areas.assessments moreperi–urban and ban Further research on this topic is needed in Africa and Latin America, as well as in ur timely,offercare. facilities high–quality if outcomes health better produce only will latter however,the emergencies; obstetrical for referrals facilitate to linkages facility and support for pregnant and postpartum mothers and for establishing community– mortality, resource–constrained settings. CBPHC is especially important for education Conclusions many were pilotstudies,ratherthanprojects atscale. and Asia, South in conducted half than more with scope, geographic in biased quite that favored the poorest segment of the project population. However, the studies were measurable equity component showed that CBPHC produced neonatal health benefits a included that assessments the of all Almost breastfeeding). early including always, signs), and promotion of and support for exclusive breastfeeding (sometimes, but not terventions used in these projects focused on health education (recognition of danger in the of women’sSeveral participatory groups. and visits, home (CHWs), workers health community utilized database our in assessed projects neonatal the of Many outcomes. health neonatal beneficial reported three–quarters almost design, study experimental strongest the with those Of effect. negative a demonstrated study one only effect; no reported that those did as health neonatal in improvement an ported re many as twice outcomes, health on reported that projects For programs. fective ef by shared characteristics common the about information present we and health, effectiveTherestrongbe Results is can CBPHCimprovingthat in evidence neonatal natal subset. neo the in included articles the on conducted also was assessment equity An ables. mortality. Descriptive analyses were conducted based on study type and outcome vari zation of postnatal care, nutritional status of neonates, neonatal morbidity, or neonatal illness, care illness, utili changes in knowledge about newborn seeking for newborn health: neonatal to related outcomes following the of one identify not did sessment cluded if the primary project beneficiaries were more than 28 days of age, or if the as assessments specifically relating to newborn health (N health newborn to relating specifically assessments where in this series. From this larger database (N database larger this From series. this in where else described and assembled been has CBPHC through health child and neonatal and field research studies (referred to collectively as projects) in improving maternal, programs projects, of effectiveness the of evidence comprising database A Methods first 28daysoflife. (CBPHC) and common components of programs aiming to improve health during the care health primary community–based of effectiveness the regarding evidence able avail mortality.the neonatal reviewsreducing article for This challenges be to tinue care, emergency caredelivery carepostnatal con safe obstetricand to access of Lack deaths. under–5 of proportion large increasingly an becoming is mortality neonatal continues to decline globally through programs to address the health of older infants, Background As the number of deaths among children younger than 5 years of age age of years 5 than younger children among deaths of number the As CBPHC can be effectively employed to improve neonatal health in high– 355 = 548), a subset was created from created was subset a 548), June 2017 •Vol. 7No. 1•010903 = 93). Assessments were ex were Assessments 93). global journal of health ------
VIEWPOINTSPaPErS PaPErSVIEWPOINTS review. Figure 1. June 2017 •Vol. 7No. 1•010903 Sacks etal. Selection of assessments for inclusion in the neonatal health Selection ofassessmentsforinclusion intheneonatalhealth haran Africa, and 41% of births in South Asia still take place outside of health facilities [ [5 infections and neonates are complications of preterm birth, intrapartum–related complications (often birth asphyxia), settings settings where home deliveries are common and access to facility–based care for neonates is limited [ Community–based approaches to reducing neonatal mortality are of particular importance in low–income rates aswell. based efforts in education, support and referral may be important in settings with high facility delivery mortality.of risk high thereaftera soon and have birth they their when during newborns Community– prove neonatal health will be essential for the near term to promote healthy home practices and to reach im to approaches community–based facilities, in care high–quality providing with challenges tinuing under–5 deaths [2 of 45% for accounting present at age, of years 5 than younger children among mortality proportionof child health, including the search strategy, has been described elsewhere in this series [ The methodology for assembling a database of 548 assessments of the effectiveness of CBPHC in improving METHODS [ life of day first the on still die annually during their first month of life [ infants reductionslive–born marked million Despite 2.7 1990, since globally mortality child overall in projects, programs, or field research studies (described collectively as projects) on mortality, morbidity, nu grey literature for documents that described the implementation of CBPHC and assessed the effect of these cilities. The larger study conducted a search of published documents in PubMed, personal sources, and the community outside of a health facility. There could also be associated activities that took place in health fa- sidered CBPHC to be any activity in which one or more health–related interventions were carried out in the complements other reviews that have been carried out on this topic [ er review of the effectiveness of community–based primary health (CBPHC) in improving child health. It (CBPHC) inimproving neonatalhealthusingasubsetofarticlesfrom adatabaseassembledforbroad care health primary community–based of effectiveness the to related findings the analyzes paper This study design,outcomevariables,program components,andreported neonatalhealthimpact.
]. Given that 51% of births in the least developed countries, 49% of births in sub–Sa in births of 49% countries, developed least the in births of 51% that Given ]. ]. Approximately 73% of neonatal deaths occur during the first week of life [ 3 ] and 32% during the first 6 hours of life [ life of hours 6 first the during 32% and ] 356 1 ]. Neonatal mortality is becoming an increasingly large and primary healthcare.and primary This yielded380articles. ital syphilis prevention; congenital syphilis treatment; nus prevention; neonatal tetanus treatment; congen mother–to–child transmission of HIV; neonatal teta- of prevention (IMCI); Illness Childhood of agement laria prevention; malaria treatment; Integrated Man tions; diarrhea treatment; pneumonia treatment; ma immuniza weight); birth (including weight/height child breastfeeding; health; neonatal/perinatal were: form, extraction data the on defined as tions, interventions pertaining to neonates. These interven ticles were selected that had been coded with relevant ar stage, first the In process1). three–stage(Figure a in health neonatal of analysis the for wereselected assessments set, data health child the with Starting lanta, Georgia, USA). At Prevention, and Control Disease for Centers (US to an electronic database using EPI INFO version 3.5.4 resolveder differences.any weredata The transferred about the assessment and a third independent review information extracted Twoindependently reviewers review.the in included assessments health child and erature and elsewhere, yielding a total of 548 neonatal addition, 115 reports were identified from the grey lit screening via PubMed, 433 qualified for the review. In for identified articles 4276 Of intervention. based tritional status, or population coverage of an evidence– 7 4 – ]. The key causes of death among among death of causes key The ]. 9 www.jogh.org ]. Projects were assessed by their • doi:10.7189/jogh.07.010903 10]. In brief, we con 1 ], and the con 3 ], 36% 2 , 6 , 7 ]. ]. ------www.jogh.org • doi:10.7189/jogh.07.010903 reference can be obtained from Appendix S2 in author’s last name and year of publication, with the reference number in brackets with a prefix S. The full In this paper, when assessments selected for this analysis are specifically cited, we cite them with the first ment. within each category. A table of only the RCTs is presented in Table S1 of outcomes health beneficial of proportion the present to set data the on analyses descriptive conducted Wedesigns. non–experimental other and observational and trials; controlled(RCTs); non–randomized trials controlled randomized categories: three into separated were but included, were designs study All ables, andcategorizedbythetypestrength ofstudydesign. articles. Articles were coded by the primary and secondary health condition addressed, the outcome vari- related to nutritional status, morbidity or mortality). The final database for this sub–analysis included 93 illness, careborn seeking illness, for utilization newborn of postnatal care, or a neonatal health outcome new about (knowledge health neonatal to related directly outcome an have not did article the if made if the target population was not infants under age one. This yielded 108 articles. Further exclusions were wereexcluded werereviewed.articles then Articles 380 these of abstracts and titles stage, second the In As shown in Description ofdatabase r portant dimensionoftheseprojects. CHWs, as they varied widely among the included assessments, although we recognize that this is an im- of characteristics descriptive other or health, of ministry the by recognition formal any), (if pensation com of level training, of here regarding length specification project.Wefurther the provideany not do of activities the in assist programto system/health broader health the project fromor the training ceives rewho actor community–level any referto hereto used is (CHW) worker health community term The ences forallthe93assessmentsselectedanalysisinthispapercanbelocated. implemented by CHWs (n Most of the 93 assessments in our analysis were of projects that focused on a set of communities (n Africa andSouthAsia. sub–Saharan in countries four in one and countries African 10 in implemented one country: one than TwoTanzania3. Kenya, had Ghana, each sessments; Uganda, and more projectsin of were assessments as 4 had Brazil (6). Pakistan and (12) Nepal (12), Bangladesh by followed 16), (with India was ments of the effectiveness of CBPHC in improving neonatal health. The with country the most reported assess areas(n = (n urban werein some although 67), = areas(n peri–urban or 19) = Projects7). weremostly and 5 projects were implemented in one community only. Overwhelmingly, the projects were in rural or a district (n district a or = level, (n multinational studies Veryor few 42). national provincial,= the at were 10) ESULTS Figure 2, South Asia was far more represented than Africa or Latin America for assessments = 61), and many utilized ministry of health staff (n 357
Online Supplementary Online Document Supplementary where the full refer the neonataldatabase(n = were implemented whoseassessmentsare in Figure 2. Regions oftheworldwhere projects June 2017 •Vol. 7No. 1•010903 Online Supplementary Docu- CBPHC andneonatalhealthfindings = 37), local field research 93). = 36) 36) ------
VIEWPOINTSPaPErS PaPErSVIEWPOINTS June 2017 •Vol. 7No. 1•010903 Sacks etal. *The columnsumstomore than93sincemanyassessmentsdescribedmultiple interventions. Table 1. Congenital syphilistreatment Congenital syphilisprevention HIV/AIDS (prevention ofmother–to–childtransmissionHIV) Pneumonia treatment Neonatal tetanusprevention Malaria prevention Immunizations Malaria treatment Diarrhea treatment Integrated ManagementofChildhoodIllness(IMCI) healthcarePrimary Promotion of improved weight among neonates (including birth weight) Promotion ofbreastfeeding duringtheneonatalperiod General promotion ofimproved neonatalhealth I nterventIon Interventions reported healthcareInterventions inassessmentsofcommunity–basedprimary improving neonatalhealth or negative effects were all statistically significant ( significant statistically all were effects negative or ther: (1) a significant positive effect, or (2) no significant effect or (3) a significant negative effect. Positive outcome and impact indicators and also in terms of the type of outcome. Outcomes were classified as ei [ scheme Donabedian the of terms in them classify and these list Table3 Among the 93 assessments included in our analysis, 45 separate indicators were measured. Documentprovides oftheRCTassessments. in OnlineSupplementary asummary design. Other study designs less commonly used were case–control and cross–sectional studies. Table S1 projects (21%) used an uncontrolled before–after study design, and a tenth (9%) used a descriptive study the of fifth A trials. controlled) (non–randomized, quasi–experimental were 15% another (RCTs),and The assessments utilized a range of methodologies. Almost half (46%) were randomized controlled trials Outcomes five years. between one to five years. Fewer than 5% of the projects in the review were implemented for more than ments were implemented for less than one year, and another three–quarters (72%) were implemented for assess- the of (24%) One–quarter timeline. short relatively a over implemented generally were Projects ( nia treatment, and tetanus prevention; no studies addressed pneumonia prevention or tetanus treatment prevention,pneumo treatment,tetanus preventionor micronutrient malaria munizations, distribution, infants. Other common activities carried out by these projects included general health primary care, im scribed an intervention that focused on the prevention of low birth weight or the care of low–birth weight de- (24%) one–quarter and intervention, breastfeeding a described (38%) assessments the of one–third ed what were classified in the data extraction process as “neonatal/perinatal health” interventions. Almost Three–quarters (76%) of the 93 assessments identified for this review described projects that implement implemented Interventions services. health of ministry of part formal a were who CHWs volunteer or paid using projects many are there (n ers = (n members community local and 26) = and exclusive mutually werenot categories these 27); statistical testing demonstrated a difference that was not statistically significant (P testing was not performed. not was testing we haveincludedtheminTable 3. qualify,did that indicators outcome so health other with along assessments project of part as measured were they but preparedness) birth or activities, group in participation illness, neonatal for seeking care 3 by specific health outcome or process/output indicator. A few process/output indicators shown in more of the types of health indicators that were used in selecting assessments for inclusion in the review Table 1). did not meet the criteria for inclusion in the review (eg, knowledge measures, quality of care measures,
and Table2 provide an analysis of effectiveness in terms of one or or one of terms in effectiveness of analysis an provideTable 3 358 n umber P
oF ≤
assessments 0.05). No significant effects were those in which which in those were effects significant No 0.05). 11 12 12 14 15 21 33 67 1 1 5 7 7 7
In
revIew * www.jogh.org ] of input, process, output, output, process, input, of 11] • doi:10.7189/jogh.07.010903 > p ercentage 0.05), or significance 11.8 12.9 12.9 15.1 16.1 22.6 35.5 72.0 1.1 1.1 5.4 7.5 7.5 7.5 ( n = 93) Table 2 and Table - - - - www.jogh.org Table 2. Assessments of community–based primary health care projects that document improvements in neonatal health as defined by health outcome and health impact indicators* outcome measure assessment methodology wIth FIndIngs total
• doi:10.7189/jogh.07.010903 Randomized controlled assessments Non–randomized controlled Observational (mostly pre/post interven- assessments tion) assessments Positive effect (n = 31) No significant or negative Positive effect No significant Positive effect (n = 13) No significant or effect (n = 12) (n = 8) or negative negative effect effect (n = 2) (n = 7) Nutritional status: Birth weight/low birth weight Christian 2003 [S23] Larocque 2006 [S46] Ahrari 2006 [S2]; 4 Tielsch 2008 [S82] Small–for–gestational age Christian 2003 [S23] 1 Preterm birth Christian 2003 [S23] 1 Morbidity: Neonatal sepsis Gill 2014 [S34]; Soofi 2012 [S77] 2 Neonatal omphalitis Mullany 2006 [S53]; Soofi 2012 [S77] 2 HIV mother–to–child Gupta 2013 [S36]; Vogt 2015 [S86] 3 transmission/infection rate Kagaayi 2005 [S40] Diarrhea/dysentery Osendarp 2001 [S61] el–Rafie 1990 [S31] Tielsch 2008 [S82] 3 Acute respiratory infection Datta 1987 [S27] Tielsch 2008 [S82] 2 Mortality: Neonatal mortality rate Bang 2005 [S13]; Baqui 2009 [S14]; Bhutta 2008 Azad 2010 [S9]; Colbourn Bang 1999 [S12]; Singh 2014 Rana 2011 [S69] 26 [S20]; Bhandari 2013 [S19]; El Arifeen 2012 [S30]; 2013 [S24]; Gill 2014 Memon 2015 [S51]; [S74] 359 Fottrell 2013 [S33]; Kumar 2008 [S45]; Lewycka [S34]; Kirkwood 2013 Spencer 1987 [S78] 2013 [S47]; Manandhar 2014 [S50]; Perry 2006 [S44]; More 2012 [S52]; [S64]; Persson 2013 [S66]; Rahman 1982 [S68]; Sloan 2008 [S76]; Soofi Tielsch 2007 [S81]; Tripathy 2010 [S83] 2012 [S77] Early neonatal mortality rate Memon 2015 [S51] Singh 2014 2 [S74] Perinatal mortality rate Bang 2005 [S13]; Bhutta 2008 [S20]; Kumar 2008 Bang 1999 [S12]; Seim 2014 [S72] 7 [S45]; Jokhio 2005 [S39] Memon 2015 [S51] Early infant mortality rate Christian 2004 [S22] 1 Infant mortality rate Lewycka 2015 [S47]; Perry 2006 [S64]; Shankar Benn 2008 [S18]; Sloan Perry 2006 [S64] Anand 2000 [S5]; Li Becker 1993 [S17] 9 2008 [S73] 2008 [S76] 2007 [S48]; ASHA–India
June 2017 •Vol. 7No. 1•010903 2008 [S7] Sepsis–specific case fatality rate Khanal 2011 [S42] 1
Diarrhea–specific mortality el–Rafie 1990 [S31] 1 CBPHC andneonatalhealthfindings Tetanus–specific mortality rate Rahman 1982 [S68] Newell 1996 [S59] Becker 1993 [S17]; 5 ASHA–India 2008 [S7]; Anand 2000 [S5] Pneumonia–specific mortality Bang 1994 [S11] 1 rate Low birth weight–specific Sloan 2008 [S76]; Tielsch 2007 [S81] 2 mortality rate Total number of assessments 31 12 8 2 13 7 73 *See Appendix S2 in Online Supplementary Document.
VIEWPOINTSPaPErS PaPErSVIEWPOINTS June 2017 •Vol. 7No. 1•010903 Sacks etal.
Table 3. Assessments of community–based primary health care projects that document improvements in neonatal health as defined by health process/output indicators* process and output measures assessment methodology wIth FIndIngs total Randomized controlled assessments Non–randomized controlled assessments Observational (mostly pre/post intervention) assessments Positive effect (n = 36) No useful or Positive effect (n = 28) No useful or negative Positive effect No useful or
negative effect effect (n = 5) (n = 31) negative effect (n = 5) (n = 5) Newborn care practices: Thermal care Kumar 2008 [S45]; Findley 2013 Sloan 2008 [S76] Khan 2013 [S41]; Syed 5 [S32] 2006 [S79] Colostrum administration Kumar 2008 [S45] Khan 2013 [S41]; Malekafzali 2000 [S49] Vir 2013 [S85] 5 Memon 2015 [S51] Cord cleansing with chlorhexidine El Arifeen 2012 [S30]; Mullany 2006 Orabaton 2015 [S60] 5 [S53]; Mullany 2013 [S54]; Soofi 2012 [S77] Delayed bathing of the newborn within Kumar 2008 [S45]; Penfold 2014 Khan 2013 [S41] Sitrin 2015 [S75] 4 360 the first six hours after birth [S63] Clean hygiene practices for home delivery Fottrell 2013 [S33]; Kumar 2008 Memon 2015 [S51]; Parashar 2013 [S62]; 7 [S45]; Penfold 2014 [S63] Khan 2013 [S41] Sitrin 2015 [S75] Knowledge on newborn health: Knowledge of newborn danger signs Findley 2013 [S32] Khan 2013 [S41] Callaghan–Koru 2013 4 [S21]; Dongre 2009 [S29] Knowledge on early breastfeeding Malekafzali 2000 [S49] 1 Knowledge on feeding during diarrhea ep - Malekafzali 2000 [S49] 1 isodes Feeding practices and micronutrient supplementation: Breastfeeding within the first two hours Findley 2013 [S32] Memon 2015 [S51]; Malekafzali 2000 [S49] Vir 2013 [S85] Khan 2013 [S41] 7 www.jogh.org Crookston 2000 [S26]; Syed 2006 [S79] Proper feeding during diarrhea episodes Malekafzali 2000 [S49] 1
• doi:10.7189/jogh.07.010903 Exclusive breastfeeding Bashour 2008 [S16]; Coutinho 2005 Balaluka 2012 [S10]; Malekafzali 2000 [S49] Neumann 1993 [S57]; Khan 2013 17 [S25]; Haider 2000 [S37]; Qureshi Crookston 2000 [S26]; Thiam 1995 [S80] [S41]; Neumann 2011 [S67]; Rotheram–Borus 2014 Haider 2002 [S38]; Khan 1999 [S57]; [S71]; Kimani–Murage 2015 [S43]; 2013 [S41] Neutzling 1993 Lewycka 2013 [S47] [S58] Micronutrient supplementation coverage Bang 2005 [S13]; Benn 2008 [S18]; Christian 2003 Tielsch 2008 7 Daulaire 1992 [S28]; Osendarp 2001 [S23] [S82] [S61]; Shankar 2008 [S73] Referral and treatment of health conditions: Receipt of Amoxicillin within 24 h of onset Murray 2014 [S55] 1 of pneumonia symptoms www.jogh.org • doi:10.7189/jogh.07.010903
Table 3. Continued
process and output measures assessment methodology wIth FIndIngs total Randomized controlled assessments Non–randomized controlled assessments Observational (mostly pre/post intervention) assessments Positive effect (n = 36) No useful or Positive effect (n = 28) No useful or negative Positive effect No useful or negative effect effect (n = 5) (n = 31) negative effect (n = 5) (n = 5) Referral of sick newborns Ansah Manu 2014 [S6] Bhutta 2008 [S20] Baqui 2008 [S15] 3 Treatment of diarrhea with ORT Thiam 1995 [S80] 1 Accuracy of assessments and adherence to protocols: Correct determination of low birth weight Amano 2015 [S4] 1 and very low birth weight by CHWs Error free management of cases of pneu - Perry 2016 [S65] 1 monia by traditional birth attendants Correct interpretation of growth chart by Malekafzali 2000 [S49] 1 mothers Detection/identification of sick newborns Ansah Manu 2014 [S6] Baqui 2008 [S15] Rana 2011 [S69] 3 361 Adherence to protocols for management Amano 2015 [S4] 1
of LBW and VLBW Health care utilization and birth preparedness: Antenatal care attendance Persson 2013 [S66] Uzondu 2015 [S84]; Memon 2015 [S51] Wangalwa 2012 [S88]; 7 Baqui 2008 [S15] AFK– Pakistan 2014 [S1]; Rana 2011 [S69] Delivery in a health facility or by a skilled Bhutta 2008 [S20]; Colbourn 2013 Memon 2015 [S51]; AFK–Pakistan 2014 10 birth attendant [S24] Uzondu 2015 [S84]; [S1]; Awoonor– Wil- Khan 2013 [S41] liams 2004 [S8]; Gopinath 2011 [S35]; Murray 2014 [S55]; Wangalwa 2012 [S88] June 2017 •Vol. 7No. 1•010903 Receipt of postnatal care Findley 2013 [S32] Bashour 2008 [S16] AFK–Pakistan 2014 4
[S1]; Wangalwa 2012 CBPHC andneonatalhealthfindings [S88] Care seeking for neonatal illnesses Bhandari 2013 [S19]; Ansah Manu Ali 2005 [S3] Murray 2014 [S55]; 6 2014 [S6] Nalwadda 2013 [S56]; Dongre 2009 [S29] Immunization coverage Rahman 1982 [S68]; Findley 2013 Bashour 2008 [S16] Memon 2015 [S51] Becker 1993 [S17]; 6 [S32] Nalwadda 2013 [S56] Participation in group activities Gopinath 2011 [S35] 1 Birth preparedness Waiswa 2015 [S87] Perry 2016 [S65] 2 Total number of assessments 37 5 28 6 31 5 111 ORT – oral rehydration therapy, LBW – low birth weight, VLBW – very low birth weight *See Appendix S2 in Online Supplementary Document.
VIEWPOINTSPaPErS PaPErSVIEWPOINTS June 2017 •Vol. 7No. 1•010903 Sacks etal. (60) of the 93 reports, community participation was reported to have had an effect, and in all of these these of all in and effect, an had have to reported was participation community reports, 93 the of (60) fect of community participation on health outcome and whether or not the outcome was positive. In 65% The data extraction form asked reviewers to subjectively judge any whether ef the assessment observed andpromotionleadership, adaptivelearning ofequity. promoting local resource use. Less–commonly stated goals and activities were promotion of community women’s or community empowerment, forging links between the community and the health system and As shown in shown As projects). (carried out in 75% of the projects) and the formation of women’s support groups (present in 36% of the As shown in andtoraiseawareness ofdangersignsforwhichfacility–basedcareborn, shouldbesought. ferred to as support groups) to raise awareness about healthy practices during pregnancy and for the new women’s participatory and tetanus; neonatal re groupsagainst (sometimes immunization maternal and care antenatal for especially facilities, health from outreach sepsis; neonatal for CHWs by referral early and treatment community–based illnesses; serious with neonates of treatment/referral early and signs, health included: home visitation by CHWs for education in relation to prevention, recognition of danger Key intervention implementation strategies that were utilized in CBPHC projects that improved neonatal the findingsthat relate specificallytoneonatalhealthinterventions. tal and child health is contained in another article in this series [ A more detailed analysis of community–based implementation strategies for improving maternal, neona Implementation strategies Tablein S1 studies(whichwereobservational mostlypre/postassessments). intervention surements among non–randomized controlled assessments and 31 out of 36 (86%) measurements among surements among randomized assessments demonstrative positive effects, as did 28 out of 34 (82%) mea process/output measures shown in Similarly,outcome. health positive the a for demonstrated (65%) 20 of out 13 assessments), tervention pre/postin (mostly analysis our in included assessments observational and non–randomized 50 the Of a negativeeffect. trials (RCTs), 31 (72%) showed a positive outcome and 12 (28%) showed either no effect or (in one case) analysis observed a positive outcome or a favorable health impact. Among the 43 randomized controlled This analysis indicates that, for a range of indicators, between 65–90% of the assessments included in our (mostlypre/postassessments,13outof20(65%)demonstratedpositiveeffects.servational intervention) ob uncontrolled the Among effects. positive demonstrated 10 out 8 assessments), mortality were which of (all assessments controlled non–randomized among mortality.measurements for 10 34 the of Among Table 2 demonstrated positive effects: 2 out of 4 for nutritional status, 6 out of 6 for morbidity, and 24 out Overall, 31 of the 43 measurements of outcomes of randomized controlled assessments that are shown in included amongourassessments. provides details of the 43 randomized controlled trials trials controlled randomized 43 the of details provides Document Supplementary Online Figure 3, the most common associated implementation strategies were the training of CHWs , over half of the projects had stated goals and associated activities of promoting of activities associated and goals stated had projects the of half over 4, Figure
Table 3, the findings are strongly favorable. 37 out of 42 (88%) mea 362 12]. However, here we mention some of of theseactivities. since someprojects hadmore thanone (n = PHC projects toimprove neonatal health carried out in the implementationof CB- Figure 3. 93). Thesumisgreater than100% www.jogh.org Common associated activities Common associatedactivities • doi:10.7189/jogh.07.010903
------www.jogh.org Table 4. Online Supplementary Document . *See AppendixS2inOnlineSupplementary qial Mortality Equitable Pro–equitable assessment o nqial Mortality Inequitable utcome
oF Equity assessments of community–based primary healthcareEquity assessmentsofcommunity–basedprimary inimproving neonatalhealth*
• doi:10.7189/jogh.07.010903 o Tetanus neonatorummortalityrate Geography Geography Breastfeeding initiationwithinthefirst houroflife Exclusive breastfeeding from birthto6mo Breastfeeding Skilled attendantatbirth Skilled birthattendance Neonatal mortalityrate Mortality Postnatal care coverage Postnatal care Perinatal mortalityrate Neonatal mortalityrate Neonatal moralityrate xlsv ratedn rmbrht oSocio–economic status Exclusive breastfeeding from birthto6mo Breastfeeding utcome households) (Figure 4). other to compared households disadvantaged most the in newborns for favorable less were outcomes the (ie, “inequitable” were outcomes the (20%) assessment equity two only in and households), other and disadvantaged most the in favorable equally was outcome the (ie, “equitable” be to considered was for the newborns in the most disadvantaged households). For one equity assessment (10%), the outcome these 10 projects, 7 (70%) were considered to be “pro–equitable” (ie, the outcomes were more favorable for reported assessments equity 10 the geography,Of (income, etc.). equity of categories different ing us outcomes, health of equity examined review health neonatal our in assessments 93 the of 8 total, In findings specifictoneonatalhealth projects. [ elsewhere described are database our in projects health child the distance or socioeconomic challenges such as cost or other barriers. The equity effects assessed among all of because whether care, health facility–based accessing in difficulties have who those reach to able ten This arises from the fact that community–based approaches contain strong outreach elements and are of based approaches in reaching those most in need and in improving the health of the most disadvantaged. In terms of coverage, community–based efforts are generally designed to be more equitable than facility– Equity effect onneonatalhealthwaspositiveinalmostall(93%)ofthesecases. the and outcomes, health on impact an had system health the and community the between linkages the cases the effect was judged to be positive. In over half (52%) of the 93 reports, the reviewers judged that
IndIcator e Geography Awoonor–Williams 2004[S8] Socio–economic status(includingeducation) Awoonor–Williams 2004[S8] Socio–economic status(includingeducation) Geography Geography Geography Socio–economic status quIty
363 category than onestrategy. 100% sincemanyprojects employedmore health (n = of CBPHCprojects toimprove neonatal activities carriedoutintheimplementation Figure 4. ]. Here, however, we present the the present however,we Here, 13]. June 2017 •Vol. 7No. 1•010903 Common associated goals and Common associatedgoalsand 93). Thesumisgreater than CBPHC andneonatalhealthfindings r Crookston 2000[S26] Crookston 2000[S26] Newell 1966[S59] Bang 2005[S13],1999[S12] Bang 1999[S12] ASHA–India 2008[S7]) Coutinho 2005[S25] Razzaque 2007[S70] eFerence - -
VIEWPOINTSPaPErS PaPErSVIEWPOINTS June 2017 •Vol. 7No. 1•010903 Sacks etal. all study types were included (such as gray literature reports), there may have been a tendency by orgaby tendency a been have literaturereports),theremay gray as (such wereincluded types study all The large proportion of positive outcomes could be partially due to publication bias. Especially given that populations overlongperiods oftimeisnotknownatpresent. support were assured. So whether similar results can be achieved under more routine condition in larger logistical and supervision, training, high–quality which in circumstances ideal relatively under ulations the long term. Furthermore, many (but not all) of the projects were implemented in relatively small pop in successful were they if ascertain to unable are we so and timeline short relatively a projectshad Most This study had a number of limitations. The evidence is derived from projects mostly in rural South Asia. healthbenefitmore disadvantaged groupsborn toagreater degree thanothers. natal mortality. Our equity analysis shows that almost all of the CBPHC interventions for improving new neo reducing for approach this of effects favorable the and sought be should care facility–based which for signs danger about educate to and period, postpartum/postnatal pregnancythe during and practices ally, our analysis identifies the capacity of participatory women’s groups to raise awareness about healthy Addition tetanus). neonatal against immunization maternal provide to (especially facilities health from referral early of outreachimportance treatmentand and signs, dangerteams with mobile neonates by of clude home visitation by CHWs to educate mothers about healthy household practices, danger signs, the Key community–based intervention strategies that were demonstrated to be successful in our analysis in providedcan be their patients. many interventions near CHWs live if especially the home, in or close to that expected is it (CHWs), workers health community utilize projects care neonatal many that Given to identifysickneonatesinneedoffurthercare andtreatment forcertainconditions. settings where births occur at home and hygiene is poor, to improve neonatal care practices at home, and natal sepsis. Strong evidence was found for the capacity of CHWs to promote clean delivery, especially in neo of treatment for referral and diagnosis early and hypothermia, of cord,prevention umbilical the to and exclusivebreastfeeding, promotion ofcleanliness,applicationatopicalantiseptic(chlorhexidine) immediate of promotion includes care neonatal Home–based support. and training proper have they if life – especially those caused by infection – can be effectively addressed at the community level by CHWs of month first the during children among death of causes leading the of many that reveals analysis Our tion, making it more difficult to judge the effectiveness if scaled up without focused attention or resources. of the studies with the strongest designs also had the most intensive out support the in interven carrying Further,analysis. our in included assessments the of all almost in missing was topic this on mation many contacts between patients and providers, whether they take place in facilities or in homes; however, infor worth noting the importance of assessing and improving the quality of care provided at the time of health is It health. neonatal on outcomes favorable produce to appeared approaches rigor,CBPHC odological meth- of levels all for that demonstrated and design study the of rigor the by findings the summarize did ditions (to assess to what degree the assessments were of CBPHC efficacy as opposed to effectiveness), we mented interventions or the degree to which projects were implemented under ideal vs more routine con imple of quality the of assessments include not did review our While well–studied. relatively also was mortality period; neonatal the during breastfeeding exclusive and care postnatal of coverage population to related were analysis our in included assessments the in used indicators outcome common most The and awareness ofdangersignsforwhichfacility–basedcare shouldbesought. practices household healthy throughpromotionof health improvedneonatal to contribution a make to developed and facility coverage of antenatal, delivery, and postnatal care increases, CBPHC can continue ers during their and delivery immediately following birth. Even if health primary care areservices better will be essential for the near term in order to achieve universal coverage of health services for these moth [ breastfeeding of initiation early for 50% and newborns, for visits postnatal delivery,for at 28% attendant skilled for 65% low: quite is mortality neonatal improving for important are that interventions of coverage mortality,national neonatal median of burdenthe greatest [1 visits care antenatal four obtain Africa sub–Saharan in
364
]. Furthermore, among the 75 countries with the the with countries 75 the among Furthermore, ]. ]. Community–based approaches Community–based 14]. www.jogh.org • doi:10.7189/jogh.07.010903 ------www.jogh.org • doi:10.7189/jogh.07.010903 essential inorder toquicklyacceleratethedeclineofneonatalmortalityinhigh–burden countries. tant for improving neonatal health, expanding the coverage will of also community–based be services carefacility–based carethough health even Thus, [2]. hospitals) at million impor-0.59 is and centers 760 additional an them, need who neonates (170 averted be could deaths neonatal 00 primary at 000 the of 90% reach to able were infection) serious very if treatment or newborns preterm for care ive be provided at primary health care centers and at hospitals but not in the community (eg, full support- can that interventions the If [15]. worldwide stillbirth of prevalence the on effects have will vention, evidence that community–based efforts to improve antenatal care, especially nutrition and malaria pre- growingreview, is this there in included not were stillbirths of number the reducing for approaches CBPHC While period. 5–year a over 90% of coverage a achieved gradually interventions nity–based commu- the of all if 700 saved that be estimates would [15] that analysis lives separate newborn 000 Similarly,year). a each occurring currently deaths neonatal million 2.7 of total the of (27.4% nually tal mortality to reach 90% population coverage would avert an estimated 740 000 neonatal deaths an- currently available community–based interventions with evidence of effectiveness for reducing neona- According to one recently published analysis based on modeling tools [2], immediately scaling up the er varietyofgeographicareas, inurbanandperi–urbansettings,forlonger–termprograms. here at scale using rigorous operations research methodologies. Further research is also needed in a wid complications. The next step in this process is to test the types of interventions and approaches described with newborns for facilities to referral timely improving on and here, described approaches and tions settings will have to partially depend for the foreseeable future upon strengthening the types of interven high–mortality,in mortality neonatal resource–constrainedreducing in progress services, facility–based through 2030 by services newborn essential of coverage universal provide to able be not will countries many Since newborns. their reachand to mothers need all will carehealth neonatal interventions based evidence– essential and basic 2030, year the by deaths preventableneonatal end to and coverage health universal achieve to order In apparent. readily is mortality neonatal in declines accelerate to need The sive review, thefieldisvastandsomestudiesmaynothavebeenincluded. effective at scale, go beyond the scope of this analysis. Finally, while this is intended to be a comprehen be to them for required conditions specific the and out, carried were strategies intervention the how ly review were limited. The context in which projects were carried out is also wide: details regarding exact our in included projects the in deployed and trained were CHWs how regarding details the of analysis tion is wide. For example, it is known that the capacity and competence of CHWs varies widely; further varia the but definitions, and categories useful provideWe to impossible. standardizationaimed made which used, measurements and indicators definitions, of range wide the by limited further was study nizations to promote their successful work and only publish studies which had a beneficial impact. This able impactonneonatalmortalityandmorbidityworldwide. siz a have will home the in provided be can that care of quality highest the receive to newborns all for provided services provisionfor teams health mobile by care.postnatal and antenatal of ways Identifying propriately, (2) formation and support of participatory women’s groups, and (3) strengthening of outreach ap facilities health of of utilization need and in referral, neonates of identification behaviors, household healthy promote to frequently homes visit to CHWs of deployment and training (1) involving platform community–based a of strengthening and development requirethe will this evidence, current the upon Based health. neonatal improving for interventions evidence–based of coverage universal reaching for terventions will be essential for accelerating progress in reducing neonatal mortality in the near term and mentation of these interventions and strategies at scale. The scaling up of effective community–based in imple- of effectiveness the regarding evidence assemble to begin to need a is there Now strong. is tings The evidence regarding the potential of CBPHC to improve neonatal health in resource–constrained set CONCLUSIONS 365 June 2017 •Vol. 7No. 1•010903 CBPHC andneonatalhealthfindings ------
VIEWPOINTSPaPErS PaPErSVIEWPOINTS June 2017 •Vol. 7No. 1•010903 Sacks etal. rEFEr ENCES 14 13 9 12 10 8 7 11 6 5 3 2 1 4 15 health outcomes in developing countries: a review of the evidence. the of review a countries: developing in outcomes health Gogia S doi:10.1542/peds.2004-1441 doi:10.1038/jp.2016.33 review.systematic a countries: middle-income and low- Bhutta Accessed: 26April2017. https://openknowledge.worldbank.org/bitstream/handle/10986/23833/9781464803482.pdf?sequence=3&isAllowed=y Washington,Health. Child Worldand DC: Newborn, Available:Reproductive,Maternal, Bank. ControlPriorities: ease Lassi andChildHealth,Thirdborn, Edition.Washington, DC:World Bank; 2016. an updated systematic analysis for 2010 with time trends since 2000. since trends time with 2010 for analysis systematic updated an Black R doi:10.1016/S0140-6736(12)60560-1 fects. ef equity 5. health: child and neonatal maternal, improving in care health primary community-based of effectiveness Victora L Liu doi:10.2471/BLT.15.160945 tive projects. effecby used strategies 6. health: child and neonatal improvingmaternal, in care health primary community-based of Schleiff M i lw ad oe-ideicm countries. lower-middle-income and low- six 2013: a vital-registration and modelling-based study. Lancet Glob Health. 2014;2:e635-44. Health. Glob study.Lancet modelling-based and vital-registration a 2013: in countries 186 in birth, of day the including death, neonatal of risk daily of Estimation JE. Lawn SN, Cousens S, Oza key messagesfrom DiseaseControl Priorities3rd Edition.Lancet.2016;388:2811-24. Black RE, Levin C, Walker N, Chou D, Liu L, Temmerman M, et al. Reproductive, maternal, newborn, and child health: org/sowc2016/. Accessed:25April2017. U Baqui doi:10.1016/S2214-109X(14)70309-2 Perry jama.1988.03410120089033 H Perry ress for maternal, newborn, and child survival. child and newborn, maternal, for ress A Donabedian database description. and methods rationale, 1. health: child and neonatal maternal, improving in care health primary community-based of Chou 6736(15)00519-X 2017;2:020401. tool. (LiST) saved lives the using analysis An care? health primary community-based strengthening NICEF. The State of the World's Children 2016: A Fair Chance for Every Child. 2016. Available:http://www.unicef.2016. Child. Every for Chance Fair A World'sNICEF.2016: the Children of State The closure.pdf (availableuponrequest from thecorresponding author),anddeclare noconflictofinterest. Conflict of interest: of theauthorsparticipatedinrevision ofearlierdraftsandapproved the finaldraft. All data. the of analysis primary the conducted PF and KS ES, draft. wrotefirst ES the declaration: Authorship ecution ofthereview. ex the in role no had support financial provided that organizations The Foundation. Gates the and velopment, WorldThe ticle: UNICEF,Organization, Health De International for Agency WorldStates the United the Bank, ar this in described work the conduct to used were that funds providedorganizations following The Funding: ticipate asamemberoftheStudyTeam. review.the of phase initial WorldThe par to Rassekh, Bahie Dr consultants, its of one for possible it made Bank Future Generations for providing office during space,supportthe administrativeto support,Dr andPerry salary ciation and particularly its Health International Section staff, which administered some of these funds. We thank USAID, Future Generations, and the Gates Foundation. We are also grateful to the American Public Health Asso velopment oftheWorld Health Organization, theCOREGroup (CollaborationandResources forChildHealth)/ UNICEF,review: this De of and expenses Health the Adolescent and Child Worldof the Department the Bank, cover to grants small provided organizationsthat following Wethe analysis. to the grateful and are database the earlier drafts of this manuscript, and the many students and research assistants who contributed to the assembling on comments for Bhutta Zulfi and Bang Abhay Hodgins, Steve thank to wish authors The Acknowledgements: , Johnson , ZS, Kumar ZS, R J GlobHealth.2017;7: HB, Sacks E I Friberg VB, D Mitra AH, ZA, Darmstadt ZA, , Laxminarayan R , Sachdev , CG, Requejo B Rassekh , , Kumapley R J GlobHealth.2017;7: . The quality of care. How can it be assessed? be it can How care. of quality The . S Cousens HL, HP. in neonates in preventingmortality for workers health community careby neonatal Home-based , Schleiff M , Bhutta , N Begum , S Gupta , M Christian , JH, Barros
J GlobHealth.2017;7: GL, Hasan GL, , Freeman P , Temmerman M All authors have completed the Unified Competing Interest Form at www.icmje.org/coi_dis- ZA. Community-based care to improve maternal, newborn, and child health. 2016. In: Dis- In: 2016. health. child and careCommunity-based improveZA. newborn, to maternal, 010905. , Wilhelm, J , Kumapley R L Hurt , J Perin , AJ, Berman P , WalkerN , BS, Haws BS, , Gupta S 010906. S Soremekun , S Scott , , FreemanP , , Walker N , Gupta S , Rassekh B Perry , RA. Community-based interventions for improving perinatal and neonatal neonatal and improvingperinatal for interventions Community-based RA. , Bhutta Z 010901. , Lawn ,
Lancet. 2016 Lancet. ul ol Hat Ogn 2016 Organ. Health World Bull 366 HB. How many lives of mothers and children could be averted by by averted be could children and mothers of lives many How HB. . Comprehensive review of the evidence regardingeffectivenessevidence the the of reviewComprehensive . . Comprehensive review of the evidence regarding the effectiveness , editors. Disease Control Priorities: Reproductive, New Maternal, K Edmond , JE, et al. Global, regional, and national causes of child mortality: mortality: child of causes national and regional, Global, al. et JE, , Boerma , Perry
J Perinatol. 2016 Perinatol. J
JAMA. 1988; JAMA. HB. Comprehensive review of the evidence regarding the ; T 387 , et al. Countdown to 2015: a decade of tracking prog , et al. Neonatal mortality within 24 hours of birth in in birth of hours 24 within mortality Neonatal al. et , :
2049 Pediatrics. 2005; Pediatrics.
Lancet. 2012; Lancet. - 260: 59. ; Medline:26477328 36 1743-8.
Suppl 1 Suppl www.jogh.org ; 94 379: 115: Medline:3045356 : 752 : 2151- 519- S55 - • doi:10.7189/jogh.07.010903 58B - 617. 73 61. .
. Medline:27843165 Medline:25442688 Medline:27109093 doi:10.1016/S0140- Medline:22579125 Medline:15866863
J Glob Health. Health. Glob J doi:10.1001/ ------.
www.jogh.org child health:4.healthfindings health careinimprovingmaternal,neonataland the effectivenessofcommunity–basedprimary Comprehensive reviewoftheevidenceregarding material. The onlineversionofthisarticlecontainssupplementary Electronic material: supplementary Sundeep Gupta Emma Sacks 5 4 3 2 1 Paul aFreeman [email protected] USa Baltimore, MD21205 615 NorthWolfeSt. Health Johns HopkinsBloombergSchoolofPublic r Henry Perry Correspondence to: Medical Epidemiologist,Lusaka,Zambia The WorldBank,Washington,Districtof Department ofInternationalHealth,Johns University ofWashingtonSchoolPublic Independent consultant,Seattle, Columbia, USa Baltimore, Maryland,USa Hopkins BloombergSchoolofPublicHealth, Health, Seattle,Washington,USa Washington, USa oom E8537
• doi:10.7189/jogh.07.010904
3 , BahieMrassekh
1,2 5 , HenryBPerry , MeikeSchleiff
4 3 , 3 , live births, a decline from 90.4 per 1000 live births in 1990 [ 1990 in frombirths decline live a 1000 births, per live 90.4 years of age (referred to hereafter as under–5 mortality) was 42.5 per 1000 natal period. In 2015, the global mortality rate for children younger than 5 neo the childrenbeyond of improving health in carethe (CBPHC) health This paper concentrates on the effectiveness of community–based primary ering these interventions areering theseinterventions effective. tions to improve child health and that four major strategies for deliv tiveness for CBPHC implementation of an extensive range of interven Conclusions This review shows that there is strong evidence of effec ally importanteffects. that have consistently produced statistically significant and operation controlled presentedrandomized trials have care) we health primary comprehensive and diseases; childhood of management integrated immunizations; treatment; and prevention HIV malaria; and ease dis diarrheal pneumonia, controlof interventions; (nutritional tions interven of categories major all For workers. community–level ing at the community level largely by engaging communities and support addressed be can settings resource–constrained in mortality child of causes major the that evidence strong provide findings The Results tabase foranalysis. view. Data from the latter review were transferred to an electronic da re summative consolidated independent an by followed document were examined for inclusion. Two separate reviews took place of each population defined geographically a in children among mortality) or morbidity serious status, nutritional interventions, survival child health (defined as changes in population coverage of one or more key child on interventions CBPHC more or one of impact the sessing as books and reports documents, reviewedprotocol. detailed Peer a Methods We reviewed relevant documents from 1950 onwards using be effective. to appear strategies implementation and interventions specific what detail greater in exploring and is evidence the strong how amining ex by knowledge this to further contribute to aims review Our fied. for child health terventions at the community level have been identi where under–5 mortality is also decreasing more slowly. Essential in countries Asian south some in and Africa sub–Saharan of much in high remain rates mortality 2000, since mortality under–5 global in decline accelerated an been has there Although period. neonatal the based primary health care (CBPHC) in improving child health beyond Background 367 This paper assesses the effectiveness of community– of effectiveness the assesses paper This
June 2017 •Vol. 7No. 1•010904 global journal of 1 ]. Although Although ]. health ------
VIEWPOINTSPaPErS PaPErSVIEWPOINTS June 2017 •Vol. 7No. 1•010904 Freeman etal. Figure 1. Flowchart ofselectionassessments forchildhealthreview. ing childhealthbeyondtheneonatalperiod. The purpose of this paper is to summarize the evidence regarding the effectiveness of CBPHC for improv systems are required inorder toimprove health? childandmaternal PHC activities share, and how strong is the evidence that partnerships between communities and health kinds of community–based approaches appear to be most effective. What characteristics do effective CB what and effectiveness, how,facilitate and conditions them what does who but level community the at appear to be effective. Our concern is not just to strengthen the evidence about which interventions work activities specific what detail greater in exploring and (CBPHC) care health primary community–based for evidence the is strong how examining by knowledge this to further contribute to aims review Our wouldavert1.5milliondeathsofchildrenterventions 1–59monthseachyear[ has been estimated that scaling up these interventions with an essential package of community–based in ment of pneumonia, malaria and diarrhea without danger signs and referral if danger signs appear [ detection and treatment or referral of children with severe acute undernutrition; and detection and treat tribution and promotion of insecticide–treated bed nets (ITNs) or indoor residual spraying (IRS) or both; co–trimoxazole for HIV–positive children, education on the safe disposal of feces and hand washing, dis immunizations, zinc, and A vitamin with supplementation feeding, complementary and breastfeeding of promotion as: identified been have level community the at health child for interventions Essential and malaria(12%)[2 the major causes of mortality in children are pneumonia (26% of deaths in this age group), diarrhea (18%), tality (infant, 1–4 year, and under–5 mortality). Further details regarding the methodology are reported areregardingyear, methodology details 1–4 the Further (infant, mortality). tality under–5 and threatening morbidity (such as pneumonia, diarrhea, malaria, and low–birth weight); and change in mor life– serious, of outcome the in or incidence the in change micro–nutrientdeficiency); of assessment or more evidence–based interventions; change in nutritional status (as measured by anthropometry, anemia, or one of coverage population the in werechanges measuresincluded Outcome whereincluded. health Only those assessments which had clear documentation of the and intervention(s) their impact on child Georgia, USA). Atlanta, Prevention, and Control Disease for Centers US Info, (Epi 3.5.4 version INFO EPI using place view. Data from the latter review were transferred to an electronic database for analysis. Data analysis took Two independent reviews were carried out and followed by an independent consolidated summative re- population. defined geographically a in children among morbidity,mortality), serious or status, tional indicator, survival nutri child evidence–based key a of (coverage health child on interventions CBPHC more or one of impact the assessing books and reports articles, peer–reviewed Weprotocol.examined detailed a using projects) as to referred(hereafter researchstudies and programsprojects, of fectiveness providereviewto Our aims comprehensivea fromreview documents of onwards1950 ef- the assessing METHODS creasing more slowly [ de also is mortality under–5 where countries Asian south some in and Africa sub–Saharan of much in there has been an accelerated decline in global under–5 mortality since 2000, mortality rates remain high
]. Undernutrition isacauseof45%allunder–5deaths[ ]. Undernutrition 1 ]. Following the neonatal period (when 45% of under–5 deaths occur currently), 368 The remaining 489 assessments ( assessments 489 remaining The elsewhere inthisseries[5 ternal and neonatal health and not reported here [ here reported not and health neonatal and ternal reported in the other papers in this series focusing on ma is assessments these of analysis An health. maternal and neonatal on focus to found was intervention the ments assess 12 another In assessments. these of 48 in month study population was clearly documented as less than one the of age The children. month–old 1–59 and neonates There were 548 assessments included in our database for General findings r ESULTS www.jogh.org ]. 1 • doi:10.7189/jogh.07.010904 3 ) focused pre focused 1 ) Figure ]. ]. 4 6 ]. It , 7 ]. ]. ------www.jogh.org • doi:10.7189/jogh.07.010904 *The sumofthiscolumnexceeds489sincemanyassessmentsdescribedmore thanoneintervention. Table 1. text. the assessments in our review that are cited here can be identified from the number in brackets in the rentheses in the text below can be found in Appendix S2 in in listed categories the to cording Below we provide an analysis of the interventions for children beyond the neonatal period grouped ac- other quarter(21%)containedonlytwo,onethree ormore. an projectsand described assessments the of (52%) half intervention Although one vention). only with healthcare”including “primary andcountingIntegratedManagementofChildhoodIllnessasoneinter Table 2 shows the frequency of assessments according to the number of interventions implemented (not ganizations, financing,traininganduseofradios. number of assessments and so have been grouped as in Other Interventions of the other interventions shown in of the 129 projects that were classified as providing “primary health care” also implemented one or more 5 but All assessments. 489 these in described interventions child common Tablemost the lists below 1 are indicatedinparenthesis withaprefix Sthroughout thispaper. liography of these assessments in contained in Appendix S1 in dominately on children beyond the neonatal period, but many also include neonates. The complete bib Table 2. included in the “Others” group in and above included not categories intervention paper.Other this in detail in analyzed not are ventions Other HIV/AIDS treatment only HIV prevention only HIV prevention andHIV/AIDStreatment HIV prevention orHIV/AIDStreatment Pneumonia treatment only Pneumonia prevention only Pneumonia prevention andtreatment Pneumonia prevention ortreatment Integrated ManagementofChildhoodIllness healthcarePrimary Immunizations Malaria treatment only Malaria prevention only Malaria prevention andtreatment Malaria prevention ortreatment Diarrhea treatment only Total Projects categorizedas“Other” withinterventions 5 to7 3 to4 2 Diarrhea prevention only 1 n Diarrhea prevention andtreatment Diarrhea prevention ortreatment feeding promotion, orprovision ofmicronutrients) complementary promotion, breastfeeding monitoring, (growth activity nutrition–related Any I nterventIon umber
oF Number of intervention category areas category amongprojectsNumber ofintervention thatfocusedonchildren beyondtheneonatalperiod Leading categories of child health interventions includedinassessments Leading categoriesofchildhealthinterventions
InterventIons
area
per
proJect Table 1 focused on trachoma prevention, tuberculosis, community or . The full list of studies reviewed and referred to in the pa- the in to referred and reviewed studies of list full TableThe 1. Table 1 369 . Some categories of child interventions had a relatively small Online Supplementary Document F Online Supplementary Online Document, Supplementary requency 489 243 24 49 76 97 June 2017 •Vol. 7No. 1•010904 Table 1. These Other inter CBPHC andchildhealthfindings 0 22.1 22.5 26.4 27.0 108 110 129 132 5 30.3 150 8 37.4 52.2 183 255 n 24 24 13 42 40 19 46 11 27 91 30 48 98 2 o *p .* p ercentage 100.0 10.5 16.6 21.3 51.6 4.9 ercentage (%) 18.6 20.0 4.9 0.0 4.9 2.7 8.6 8.2 3.9 9.4 2.2 5.5 6.1 9.8 ( , where n = 489) and - - - - -
VIEWPOINTSPaPErS PaPErSVIEWPOINTS June 2017 •Vol. 7No. 1•010904 Freeman etal. assessments Table 3. Randomized, controlled t Uncontrolled, before–after Non–randomized, controlled Total Non–study activity Descriptive Cross–sectional Case–control, cross–sectional ype
oF
study Type ofstudymethodologyusedamongchildhealth ment). Theremaining assessmentsinourreview hadsimilareffects stated. unlessotherwise Docu Supplementary Online in S2 Appendix in presented are (these effects significant of size greatest ducing household air pollution, HIV prevention and co–trimoxazole prophylaxis for HIV–infected and and HIV–infected for prophylaxis co–trimoxazole and prevention HIV pollution, air household ducing re- sanitation, and water drinking safe soap, with washing hand vaccinations, includes component This Prevent severity and/ordurationofdiarrheal episodesinchildren [S19–24]. Further studies have demonstrated the strong efficacy of zinc supplementation in reducing the incidence, below.section nutrition the under moredetail presentedin be will supplementation zinc and A vitamin of Studies [S18]. pneumonia of incidence decreased a demonstrated supplementation zinc or A min trolled trial assessment, a community–based integrated nutrition program apparently not including vita con randomized one In S17]. [S16, diarrhea and pneumonia both of incidence the decrease nificantly promotionand supplementation providedZinc washing hand sig of to were found CHWs also by each other study, the incidence of pneumonia was decreased through vitamin A supplementation by 44% [S15]. an In [S14]. 26% by mortality child decreasedpneumonia–specific supplementation A vitamin one, In mortality. pneumonia reducing in supplementation A vitamin community–based of efficacy the strated months of life, adequate complementary feeding, and vitamin A supplementation. Several RCTs demon six first the during breastfeeding exclusive birth: from practices health good are component this Under Protect monly usedbyCHWsinthesestudies. (PHC), they will be discussed under those sections below. Co–trimoxazole was the antibiotic most com Care Health Primary or (IMCI) Illnesses Childhood of Management Integrated of part was management studies showed decreases in child pneumonia–specific incidence or mortality but as their pneumonia case other 20 Over S13]. [S12, management case good–quality implementing through level community the at infections respiratory treating by significantly pneumonia of severity clinical the decrease can CHWs Two[S6–11]. 69% to 28% fromRCTs ranging other years, that 5 demonstrated than less aged children studies – also observed significant operationally important decreases in pneumonia–specific mortalityin ument with an S prefix refers to the number of the assessment in Appendix S2 in brackets in number The article. this of end referencesthe of at list the fromreferencesin them thecited erencing assessments from our database with numbers in brackets, preceded by an S prefix, to distinguish refbe Throughoutwill - [S1–5]. we 60% article to this 13% of range reductionsthe – in antibiotics with pneumonia of treatment (CHW) worker health community of result a as mortality child in reductions randomized controlled studies (RCTs) that all showed operationally important and statistically significant This part of the framework includes diagnosis, screening, triage and treatment. Our review includes five Treat tect andPrevent framework[8 Treat,a accordingto integrated arediarrhea and Propneumonia address to actions (GAPPD), Diarrhea [ deaths globally,of mortality 18% under–5 for of accounting cause single leading the is Pneumonia Findings specifictopneumoniaanddiarrhea children younger than 5 years of age [ age of years 5 than younger children Diarrhea is a major cause of child mortality and morbidity globally and is responsible for 9% of deaths of . Many other assessments – mainly non–randomized controlled, uncontrolled and case–control and uncontrolled controlled, non–randomized mainly – assessments other Many .
F requency 177 127 8 100.0 489 74 24 27 45 15 ]. We willfollowthisframeworkinpresenting ourfindings. p ercentage 15.3 36.6 26.3 4.3 5.6 9.3 3.1 2 (%) ]. Under the Integrated Global Action Plan for Pneumonia and and Pneumonia for Plan Action Global Integrated the Under ]. 370
sessments that have the strongest study designs and designs study strongest the have that sessments line Supplementary Document). We focus on those as all 489 assessments (as presented in Appendix S1 in of analysis detailed a from us prevent limitations Space (data notshown). in listed areas categorical intervention major the across evenly fairly spread are methodologies study make up the other quarter of assessments. These various methodologies study of types Other comparisons. after studies and one–quarter (26%) are uncontrolled, before– controlled are (52%) One–half studies. 489 these for used methodologies study of Tabletypes the outlines 3 www.jogh.org Online Supplementary Online Doc- Supplementary • doi:10.7189/jogh.07.010904 Table 1 Table On- 2 ]. ]. ------www.jogh.org • doi:10.7189/jogh.07.010904 tion are presented laterintherespective sections. nutri and HIV immunizations, concerning interventions community–based of efficacy The S43]. [S42, were similarly effective in reducing the matter particulate remove that Filter Family Lifestraw and BioSand Water as [S39–41]. rhea such filters diar childhood of incidence the decreaseeffectiveapproach to an as demonstrated was water of ization steril Solar [S35–38]. studies several in diarrhea childhood reducing effectivein found was agent fying produced puri locally another or hypochlorite sodium with household the within water of Purification munity levelorbyCHWs–wasalsoeffective [S32–34]. along with education about good household sanitation practices – whether by nurses working at the com mentioned in the previous sentence [S27–31]. Teaching mothers to use oral rehydration solution at home along withtheprovisionof soapalsodecreased diarrhea childhood even greateran to degree those than rhea [S25, S26]. Randomized controlled trial assessments of education of caregivers about hand washing proper disposal of animal feces from living areas produced decreases in the incidence of childhood diar Randomized controlled trials found that community education focused specifically on the importance of groups. community with meeting or households visiting by usually CHWs, trained by out carried ity HIV–exposed children. Education of community members about diarrheal disease was a common activ munity capacity building. Other studies presented in importance of adapting interventions to local community circumstances as well as the importance of com mal training could decrease child mortality by diagnosing and treating malaria themselves, illustrated the of malaria. Kidane et al. [S55], by showing that mothers in a remote area of Ethiopia (Tigray) with mini The assessments included in sented laterinthispaper. (IMCI) or with other integrated approaches (such as Care Groups and Health Primary Care) will be pre prevention and treatment of malaria with Integrated Community Case Management of Childhood Illness include which Studies [S78–83]. individually evidence strong as show not did but ITNs of use the on tains have some effectiveness in reducing all–cause child mortality [S76, S77]. Some other studies focused cur impregnated that evidence provided studies Several [S73–75]. effective also was treatment malaria with ITNs of distribution Combining [S69–72]. sessions outreach clinic mobile of time the at cination vac measles with ITNs of distribution the combining was outcomes important operationally produced approachwhich used commonly A [S62–68]. malaria of prevention for nets bed impregnated of bution There were several other assessments that provided evidence in support of the community–based distri and malariacontrol provided bymobileteamsfrom peripheralhealthfacilities. gies, treatment of malaria within the community by CHWs and mothers, engagement of women’s groups, ventions presented include use of CHWs involved in house–to–house and group implementation strate of the effectiveness of community–based approaches to the prevention and control of malaria. The inter evidence strong demonstrate assessments These impacts. mortality marked showing some with effects, programmatic important operationally shown have that control and prevention malaria for terventions in shown As base. (RDT). test diagnostic rapid a by assisted or only signs clinical on based be may CHWs by malaria of cases of diagnosis munity–based Com medication. anti–malarial with malaria of (IPT) treatment preventive intermittent and CHWs, by residualhold treatmentantimalarial spraying, patient’sthe within community the in or (HH) household for malaria prevention and treatment include: distribution of insecticide–treated bed nets (ITNs), house [ mortality under–5 of 15% of cause the is malaria Africa, In endemic. is it where countries those in mortality child of causes commonest three the of one is Malaria Findings specifictomalaria decreases in overall child mortality as well [S93,94]. The cost-effectiveness of combining malaria and and malaria combining of cost-effectiveness The [S93,94]. well as mortality child overall in decreases ity or an improvement in CHW performance outcomes related to malaria, some demonstrated important morbid malaria–related in reduction a demonstrated treatment malaria of studies these of many While rhea orpneumonia[S84–92]. diar concurrent of treatment the with combination in or alone community the in CHWs by malaria of treatment effective demonstrated also studies other Several S58]. [S57, time same the at child same the in pneumonia of management initial the with association in community the in malaria treat and agnose , there are now a number of randomized controlled trials of community–based in- community–based of controlledtrials randomized of number a Tablenow arethere 4, Table 4 presents illustrative randomized controlled trials from our data our from trials controlled randomized illustrative presents Table4 above present important aspects of the community–based treatment E. coli concentration in water and decreasing episodes of diarrhea 371
Table 4 provide good evidence that CHWs can di 2 ]. Major community–based interventions interventions community–based Major ]. June 2017 •Vol. 7No. 1•010904 CBPHC andchildhealthfindings ------
VIEWPOINTSPaPErS PaPErSVIEWPOINTS Online Supplementary Document. *See AppendixS2inOnlineSupplementary month(s), RDT–Rapiddiagnostictest,WAZ: weight–for–ageZscore, WHZ–weight–for–heightZscore, y–year(s) – mo net, bed insecticide–treated Long–lasing LLIN– net, bed insecticide–treated ITN– Household, HH– treatment, preventive Intermittent IPT– worker, health Community CHW– communication, change behavior BCC– therapy,Artemether–lumefantrine, combination AL– Artemisinin ACT– June 2017 •Vol. 7No. 1•010904 Table 4. Freeman etal. munization) im routine of time (at mo 15 and 3,9, at methamine [Sulfadoxine–pyri IPT plus monthlyIPTfor3mo CHW by RDT) an (using malaria of treatment HH present) pneumonia of symptoms if amoxicillin with ment treat- also (and ACT with malaria of treatment CHW present) pneumonia of symptoms if amoxicillin with ment treat also (and AL with results), RDT on (based malaria of treatment CHW tion educa with Distribution malaria usinganRDT treat to CHWs Training tion educa with Distribution tion educa with Distribution tion educa with Distribution Distribution ofimpregnated bednetswithcommunityeducation: I quine bymothers chloro with Treatment Community andhouseholdmalariatreatment andprophylaxis: en’s groups level and community wom Education viaCHWatHH house) (CHW goinghouseto Distribution witheducation cation edu without Distribution given to head of household training plus given LLITN distribution work tosupportLLITN net health Community nterventIon Randomized controlled trailsofcommunity–basedmalariaprevention and treatment projects focusingonchildren and early treatment [S97, 98]. Some other studies that focused on malaria treatment or IPT at the com the at IPT or treatment malaria on focused that studies other Some 98]. [S97, treatment early and prevention malaria about mothers educate effectively can vendors drug and Trainedhealers traditional tion andtherefore the findingneedstobeinterpreted withcaution. interven the of description adequate an provide not did outcomes, important operationally significant reporting although result, this reporting assessment However,the [S96]. mortality child on impact ble munity can play in malaria prevention. School teachers, for instance, can provide IPT with a demonstra com - the of members other that role important the of evidence demonstrated have studies Several 61]. Table 4 of thecapacityCHWstoaccuratelydiagnosemalariausing RDTs isalsoanimportantfinding[S56]. demonstration The [S95]. inconclusive were findings However,the studied. was treatment pneumonia ------phylaxis chemo–pro- Coverage of morbidity phylaxis; chemo–pro- Coverage of Percentage of children receiv- 609children 4–59mo Morbidity 11 Morbidity morbidity Coverage and diagnosis Accuracy of mortality Coverage and 2260children 6moto Mortality 5000–10 Mortality t Mortality 40villages Coverage oeae1400children Coverage 219children in16 Morbidity Children in 2015 house- Morbidity oeae11villages Coverage also demonstrates the operational effectiveness of community–level IPT provided by CHWs [S59– ype
oF
outcome
age 600 children 3moof IPT) (one–half alsoreceived 500 children 1–10y <5 y Children in8villages 1457 children 0to15y Percentage of children 0 to <5 Children in160villages <6 y in eacharm p 5385 children 0to<5y villages holds opulatIon 400 children 6moto
000 children sIze
oF
study
area children 3–18mo) (among period intervention the during efficacy Protective by 3dosesofIPT children of coverage group; only HH– with compared group IPT + HH in malaria RDT–confirmed of Incidence ly appropriate treatment pneumonia who received ear with diagnosed children of percentage AL; received who children febrile of Percentage antibiotics appropriate and prompt ing s holds; A.gambiensisdensity house all to coverage ITN unnecessarily withACT Percentage of children treated child mortality ITN; an under sleeping y among children 1to<5y mortality Malaria–specific mortality y) <5 to (1 all–cause y; mo–4 y; mortality among children 1 Mortality among children 1–7 l–as hl otlt erae y4%003[S55] 0.003 Decreased by41% All–cause childmortality sleeping underanITN Percentage of total population ver Percentage of patients with fe y withmalaria Percentage of children 0 to <5 ing underanITN sleep children of Percentage month followup 6– a of time at ITN using Percentage of total population pecIFIc 372
outcome - - - - nrae y2%<.01[S61] <0.0001 Increased by22% oncreased by97% HH onlygroup); (compared with Reduced by85% increased by53% Decreased by77%; nrae y3%<.0 [S58] <0.001 Increased by34% erae y4%001[S56] 0.001 Decreased by45% [S47] decreased by12% Increased by72%; 0.05 Decreased by30% decreased by33% decreased by18%; Decreased by25%; control e decreased by99% Increased by99%; nrae y4%<.0 [S54] <0.001 Increased by49% erae y7%<.0 [S51] [S50] <0.001 Decreased by72% 0.05 Decreased by38% nrae y2%00 [S52] 0.05 Increased by27% y) (in children 0 to <5 Increased by32% FFect
compared www.jogh.org
to
• doi:10.7189/jogh.07.010904 .1 .5[S48],[S48] 0.01; 0.05 0.05; 0.01 sIgnIFIcance <0.0001; <0.0001; s <0.001 0.001, 0.001, tat 0.001 0.001 .0 [S53] 0.001 0.01; 0.01; 0.01; 0.01; IstIcal
[S59], [S60] [S57], [S57] [S45], [S46] [S49], [S49] r number [S44], [S44], eFerence *
- - - www.jogh.org • doi:10.7189/jogh.07.010904 showed markedincreases inknowledgeaboutHIVinfection. them of all virtually and interventions, other many with along project, the of part was PMTCT without or with HIV/AIDS review,about our education in included projects survival child NGO–led many In children achievingvirological suppression thanchildren inthecontrol group ( ty–based adherence support for 982 children on antiretroviral treatment was found to lead to 60% more hours of birth were found to decrease mother–to–child transmission of HIV by 60% [S117]. Communi 72 within newborns their to nevirapine give to them advised and mothers HIV–positive to nevirapine attendance at health facilities [S116]. Community household visits by midwives who gave counseling and and infants infected and HIV–exposed of identification in increase 27% significant statistically a to lead (ART)apy Similarly,[S115]. earlier also was infants infected HIV– for to found was visiting home CHW anti–retroviralcare.of postnatal Initiation and antenatal with ther also and PMTCT with compliance er one project, intensive follow–up care by CHWs at the homes of HIV–infected mothers led to much great In projects. HIV–control for important be to found often was CHWs by visiting household of role The monitoring were apartofthisproject. grated program for HIV exposed infants [S114]. Household visits by CHWs, immunizations and growth inte comprehensive a of part as medication antiretroviral received mothers HIV–positive when group, probability of survival of children to 18 months of age was 84% higher, compared to those in the control is discussed in our maternal health paper, only a few examples will be mentioned here. In one study, the sion (PMTCT) was the most commonly studied HIV intervention in the assessments reviewed. As PMTCT transmis mother–to–child preventionof The [S113]. clinic health the at only seen persons than testing undergoHIV to antiretroviraltherapy,likely taking more much were home at visited were that contacts [S112]. Several studies reported on community–based HIV testing. One study found that among persons members community providedwereby activities of monitoring the and drug the provisionof The bers. a reduction of 77% in the mortality of their originally HIV–negative, under–10 year–old household mem to led adults HIV–infected of co–trimoxazole treatmentwith community–level that demonstrated study One level. community the control at and prevention HIV/AIDS on specifically studies fewer wereThere infection Findings specifictohumanimmunodeficiencyvirus not asstrong [S99–111]. was evidence of strength the but above findings our with consistent were that results had level munity nizations with mothers. Those mothers who had increased awareness were much more likely to take their and explained by community–level workers, significantly increased community awareness about immu pletion coverage levels by 50% [S125]. Mass media using TV, radio, newspapers and leaflets, distributed ers (who in turn educated community members) was found effective, increasing full immunization com- Promotion of community participation through education of village leaders, teachers, and extension work Village–level approaches tocommunity mobilization to a150%increase inthecoverageof12–23month–old children withfullimmunization[S124]. led activities Their children. of growthmonitoring out carried also and (ORS) solution rehydration oral These CHWs promoted immunizations, use of health facilities, and household diarrhea management with Establishment of village networks of trained traditional birth attendants and female CHWs was effective. contributed togreatly increased immunization coverage ratesforchildren [S119–123]. nity health education workers, and by members of mobile health teams coming from health facilities have commu female by purpose, this for just recruited members community by households, visiting CHWs throughoutimmunizations for coverage population of providedrates education high Peer world. the by tion in immunization activities and CHWs mobilizing communities have played a key role in producing Community–based interventions involving CHWs reaching to the household level to promote participa Promotion anduptakethrough CHWsorothersinroutine systems tivity below. health. We have disaggregated the community–based assessments in our database under the areas of ac Immunizations against infectious diseases are well–established as an essential PHC intervention for child Findings specifictoimmunizations 373
June 2017 •Vol. 7No. 1•010904 P CBPHC andchildhealthfindings = 0.01) [S118]. ------
VIEWPOINTSPaPErS PaPErSVIEWPOINTS June 2017 •Vol. 7No. 1•010904 Freeman etal. moderately malnourished, mildly malnourished, or of normal weight and whose mother had not been been not had mother whose and weight normal of or malnourished, mildly malnourished, moderately ( sibling older the than higher always was score Z age for weight mean sibling younger the group), tion been severely malnourished and whose mother had been exposed to the Hearth approach (the interven age groupings of younger siblings (6–8, 9–11, 12–14, etc.). For younger siblings whose older sibling had 3–month different 10 for compared were Outcomes program. Hearth the to exposed been not had ers been overcome using the Hearth approach should have better nutrition than similar children whose moth had childrenolder of siblings undernutrition younger whose that tested was severewith undernutrition hypothesis the period, five–year a over implemented was that project controlled non–randomized a In ing sessionsusinglocallyavailablefoods[S143]. feed their children and applying this knowledge in the care of their own children through hands–on cook dren are also identified and they are guided through a process of learning how positive deviants care and fying local “positive deviant” women who have well–nourished children. Mothers of malnourished chil compared to controls who received only deworming [S143]. The Hearth approach is a process of identi for–age Z score of <–2, the Hearth approach along with de–worming significantly improved growth when weight– a with months 15 than less aged children among Vietnam, In informative. very were also level Other controlled interventions with smaller effect sizes and statistically significant results at the at results significant statistically and effectsizes smaller with controlledinterventions Other grams. to contribute to good child nutrition. These will be covered below in the final section on integrated pro demonstrated also were programs integrated other Many nutrition. child improved also children) their programs associated with small loans (that may have enabled mothers to obtain more nutritious foods for learning Group children. their of nutrition HIV,the with improved programsmothers well–organized depressed for Even growth. child on effect important an have to found was also mothers to mentation the community, and supplementation with ready–to–use therapeutic food (RUTF). Albendazole supple el through health education involving CHWs visiting households, regular monitoring of child growth in nutrition. under protein–energy with controls to compared effects large operationally and significant statistically Table 5 presents the findings from randomized controlled and non–randomized controlled studies with Protein–energy undernutrition supplementation. micronutrient and (CF), feeding complementary (BF), breastfeeding anthropometry), by assessed ally In this section our review findings will be categorized into four areas: protein–energy undernutrition (usu [ globally mortality under–5 of 45% to contributes Undernutrition Findings specifictonutrition lio vaccinationrates[S132]. compared to controls [S131]. House–to–house administration of polio vaccine significantly increased po to not only to greatly increased immunization coverage but also to a 58% decrease in under–5 mortality nity involvement, and providing immunizations, vitamin A supplementation and growth monitoring led promotingcommu households, visiting CHWs providedby education peer of case–controlledstudy A Household vaccinationstrategies with householdvisitsbyCHWsincreased vaccinationcompletionratesfrom 30%to53%[S130]. weeks vaccination Annual [S129]. paralysis flaccid acute of incidence the in decreases significant to led point service the to come not did who those for household the at immunization by up followed points National Immunization Days, in which community mobilization and immunization at peripheral service Health Days were foundtogreatly increase fullyimmunizationcoverage[S128]. A case–controlled study of community health education campaigns associated with microcredit programs Promotion ofimmunizationsthrough microcredit programs vaccination significantlyincreased allchildhoodvaccinations[S127]. for attendance promoting workshops community–based PDR, Lao In [S126]. vaccinated be to children P = 0.005 or less in all age groups). age controlall the been in For had less groupwho or sibling 0.005 older (childrenan with Table 5 demonstrates that undernutrition can be addressed successfully at the community lev
374
3 ] and therefore is a major concern. concern. major a is therefore and ] www.jogh.org • doi:10.7189/jogh.07.010904 P < 0.05 0.05 ------www.jogh.org Table 5. Online Supplementary Document. *See AppendixS2inOnlineSupplementary of mother–to–childtransmission,RUTF–ready–to–use–therapeutic food,WHZ–weight–for–heightZscore, y–year(s) BF – breastfeeding, HFA – height for age, HIV – human immunodeficiency virus, LAZ – length–for–age Z score, mo – month(s), PMTCT – prevention household supplementation tified milk–basedcereal Education plusmicronutrient–for acute, malnutrition RUTF forchildren withsevere Home–based distributionof Randomized controlled assessments: I monitoring education andmonthlygrowth health agentfacilitatorstoprovide Home visitsfrom community Non–randomized controlled interventions: PMTCT, andmentalhealth about BF, childnutrition,HIV, provision ofhealtheducation Paraprofessional homevisitswith regimen compliance nutrition, andperinatalHIV alcohol use,promote BF, child Home visitsbyCHWstoreduce provided athouseholdlevel 6mo Albendazole 600mgevery community level with growth monitoringat Nutrition andhygieneeducation small loansgiventomothers andchildhealthwith on maternal Facilitated group sessions learning and 23weeksofpregnancy households withmothersat12 Albendazole 400mgdistributedto children) tion incookingnutritiousfoodfor (mothers are trainedbyparticipa- demonstration (Hearth)program Using CHWsinanutritional nterventIon Studies of community–based interventions addressingStudies ofcommunity–basedinterventions protein energy undernutrition • doi:10.7189/jogh.07.010904 found not to result in statistically significant improved child nutrition in the short term but if the program Giving vouchers to mothers along with health education and a community household health package was height–for–age score (P water and have a latrine. Longer participation in the program was strongly associated with a better mean their water, treat to access indoor have to likely more was group intervention the years, 2 after nificant sig statistically not were study latter the in results the Although results. similar producedchildren 307 promotionlivestock and development community A [S145]. drenyears 4 to project0 aged for Nepal in ly village women, led over a 2–year period to a decrease of 10% in the prevalence of underweight in chil tion of increased household production of food through training Village Model Farmers, and subsequent demonstrated. In a population including 130 been also have nutrition childhood on programs voucher and agriculture of promotion of benefits The be repeated infurthersimilarstudies. sions need to be guarded due to the limited size of the populations studied. These results would need to conclu any However [S144]. education nutritional of effects family wider regardingthe evidence vides weight children than their older sibling ( normal and malnourished mildly for lower always was siblings younger the of score Z age for weight mean The out. carried was siblings younger program)with Hearth comparison the same to the exposed - nutritional status Change in Status nutritional Change in t nutritional status Change in nutritional status Change in nutritional status Change in status; morbidity nutritional Change in nutritional status Change in nutritional status Change in nutritional status Change in nutritional status Change in ype
oF
outcome < 0.00001) [S146]. p control) counselling only, and [Supplementation only, different groups 104 infantseachin3 children malnourished andwasted 1178 10–60–mo–old 14 24 townshipneighbor 0 to<6mo and theirchildren 644 depressed mothers two years age whowere treated for 610 children 18moof households 55 randomlyselected Children 0to<5yfrom hoods 0 hlrn0t< MeanHFA children 200 children 0to<3y children, 0to<6mo 4998 mothersandtheir 1200 children, 3–48mo opulatIon 374 children, 0to<5y
sIze
oF P
study < 375 0.05 for all but one age group, 6–8 months). This study pro
area
000 children younger than 5 years of age in Nepal, promo -
s or more weight gainof250g children withamean Percentage of ma orrelapse WHZ≥2 without ede Attainment of children 0–35mo in Undernutrition children 0to<6mo Mean LAZscores for of fecalworms stunting; prevalence Prevalence of in youngerchildren children, meanWAZ Mean WAZ inolder children Mean WHZsfor 12 to24mo first 6mooflife infants duringtheir Mortality ratein undernutrition children withsevere percentage of weight forage; children withnormal Percentage of pecIFIc o utcome - control e to control group) group compared supplemental 14% more (in [S133] 0.001 Increased by33% erae y2%00 [S139] 0.05 Decreased by27% nrae y7 .3 [S137] 0.034 Increased by7% Decreased by14% Decreased by9%; Increased by36% Increased by10%; nrae y1%001[S138] 0.001 Increased by19% nrae y4%00 [S142] 0.01 Increased by48% erae y4%00 [S140] 0.01 Decreased by41% decreased by18% Increased by10%; FFect June 2017 •Vol. 7No. 1•010904
compared CBPHC andchildhealthfindings
to
sIgnIFIcance s 0.001; 0.001; tat 0.001 0.001 0.05; 0.05; 0.02; 0.02; .1[S134] 0.01 0.02 IstIcal
r [S141], [S141], [S136] [S135] [S141] number eFerence *
------
VIEWPOINTSPaPErS PaPErSVIEWPOINTS June 2017 •Vol. 7No. 1•010904 Freeman etal. Online Supplementary Document. *See Appendix S2inOnlineSupplementary BF –breastfeeding, feeding,CHW –communityhealthworker, CF –complementary mo–month(s),WFA –weightforage,y–year(s) Table 6. promote exclusiveBF Training of1CHWpervillageto Breastfeeding: Randomized controlled interventions I children plussmallincome chickens toprovide protein for established toaidmothersbuy CHWs, nutritionrevolving fund Hearth program, CFeducationby education byCHWs clubs andprovision ofhealth Formation ofcommunityhealth Uncontrolled before–after studies: breastfeeding educated pregnant mothersin Peer counsellorsfrom community during thepostnatalperiod Home visitsbytrainedwomen CHWs Home counsellingbytrained by CHWs households alongwitheducation Provision offortifiedCFat workers nutrition bycommunityoutreach Training ofmothersinessential Non–randomized controlled trials: CF duringhomevisits CHW educationofmothersabout feeding: Complementary nterventIon Community–based projects thatpromoted breastfeeding feedinginchildren andcomplementary The data from data The ble 6. children in low–income settings [ young of lives the saving for strategy preventive effective most the of one be to estimated been has life mortality in resource–constrained settings. Promotion of exclusive breastfeeding for the first 6 months of morbidity,reduced nutrition, childhood good to improvedcontributor and important an is life of years two first the through BF continued with age of months 6 first the during (BF) breastfeeding Exclusive feeding Breastfeeding andcomplementary their children [S148–151]. ing vouchers with greater use of nutrition monitoring at the community level and improved nutrition of the same age [S147]. Other studies demonstrated a statistically significant association of mothers receiv childrenscoresthese ( Z increasedof 23% weight by for height mean the then years, 10 to 8 aged were children the until more years 2 for extended was children those for non–randomized controlled community–basedassessmentsincludedinourreview are presented in ed from 6 months of age onwards for children to sustain normal growth. Findings from randomized and Hearth approach mentioned in the section on protein energy undernutrition was also found to be effecbe to proteinfound on also approachsection was Hearth theenergy in mentioned undernutrition tary feeding was found to produce statistically significant improvements in mean height and weight. The complemen about Education education. the weredoing facilities fromhealth reachingout local sionals profes health trained highly more than rather counselors peer home and CHWs the when found were effectsstrongest the that is note Of facility,clubs. health health nearest community the mothers’ by and nity level by CHWs, by trained home peer counsellors, by community outreach health professionals from indicate that exclusive breastfeeding can be effectively promoted at the commu the at promotedeffectively be can breastfeeding exclusive that Tableindicate 6 t practice health–related Change in nutritional status Change in practice health–related Change in practice health–related Change in practice health–related Change in practice health–related Change in nutritional status Change in practice health–related Change in nutritional status Change in
ype
oF
outcome area p mo their children 0to<6 1115 mothersand y 1700 children 0to<3 y andtheirmothers 1000 children 0to<5 <6 mo and theirchildren 0to 726 pregnant women children 0to<6mo 175 mothersandtheir mo their children, 0to<6 1597 mothersand 10 healthclinics the catchmentareas of Children 9–14min mo in8districts 320 infants0to<6 118 infants opulatIon 9 ]. Complementary ]. feeding Complementary (CF) to supplement breastfeeding is need
sIze
oF 376
study
s <6 moofage exclusively breastfed to Percentage ofchildren malnutrition prevalence ofsevere WFA children; Prevalence ofnormal 0–6 mo Exclusive BFinchildren Early initiationofBF; to <6moofage Exclusive breastfeeding, <6 moofage exclusively breastfed to Percentage ofchildren <6 moofage exclusively breastfed to Percentage ofchildren for oneyear enrolled intheprogram afterbeing underweight Odds ofbeing until 6moofage exclusively breastfed Percentage ofchildren rvlneo tnigDecreased by Prevalence ofstunting pecIFIc
outcome P = www.jogh.org 0.029) comparedcontrols0.029) to of 13%; decreased e 50%; increased Decreased by Compared to FFect Increased by Increased by Increased by Increased by Increased by Increased by increased by baseline, baseline, by 17% by 60% control
38% 75% 64% 16% 63% 10% 22% compared • doi:10.7189/jogh.07.010904
to
sIgnIFIcance s 0.001; 0.001; 0.001; 0.001; tat 0.001 .0 [S157] 0.007 0.001 .0 [S153] 0.001 [S152] 0.001 00 [S155] <0.05 .0 [S156] 0.001 .5[S151] 0.05 .1[S154] 0.01 IstIcal
r [S159], [S159], [S158], [S158], [S159] [S158] number eFerence Ta * ------www.jogh.org • doi:10.7189/jogh.07.010904 children receiving community–based treatment for diarrhea and pneumonia increased significantly in the in Rwanda with complete mortality data further supports this. This assessment found that the number of a decrease in under–5 mortality. A large assessment of children younger than 5 years of age in 15 districts to lead may indeed and level community the at successfully implemented be can iCCM that show 8 ble tasks. above the all of quality the maintaining with concerned often are iCCM and IMCI of ies hold visiting, and may also be responsible for such activities as promotion and distribution of ITNs. Stud perform their tasks [ to supervision maintain their skills, and to be well–supplied with the drugs and equipment to necessary regular receive to patients, of referral for staff facility health local community,their to well–linked be to their of support and confidence the have to well–trained, be to need CHWs effective, be to iCCM For nesses of childhood (acute respiratory infection, diarrhea, malaria and in some cases acute malnutrition). Integrated Community Case Management (iCCM) enables CHWs to diagnose and treat serious acute ill cilitate outreach activitiesfrom thelocalhealth centersuchasimmunizations. fa also CHWs treatment. for facility health nearest the to patients escort even or refer and signs danger diarrhea.for arerecognizeORS CHWs to than treatmenttaught other childrenwithout illnesses of with usually groups, with meeting and door–to–door going CHWs trained by community the in performed be can that illness acute serious potentially of recognition early and activities preventive of consists ally usu C–IMCI), (or IMCI Community called component, community Its facilities. health at illness hood Integrated Management of Childhood Illness (IMCI) integrates the prevention and treatment of all child Community CaseManagement(iCCM) Integrated ManagementofChildhoodIllness(IMCI)andIntegrated summarizes the findings of assessments of C–IMCI and iCCM interventions.The studies described in been includedinourreview. projectsthe of assessments available and have level, approaches community the at developed been have services provided be integrated as much as practical for the benefit of all. To do this, a range of integrated that important is it themselves, in CHWs and CHWs in mothers of confidence increasingthe for and es have the capacity to treat any illnesses). Therefore, for the most cost–effective and efficient use of resourc if CHWs have to turn patients away because they can only deal with one disease entity (or if they do not themselves in confidence lose may CHWs and CHWs in confidence lose may Mothers infections. hood portunities to update immunization status need to be taken at every opportunity to prevent serious child our review included vitamin A, zinc, iron and multivitamins. for qualified assessments Typeswhose projects in included were that supplementation micronutrient of Micronutrient supplementation and inotherstudieswithsimilarresults [S143,S159,S160]. in cited study the in age, of months 15 than younger children undernourished in tive inate in a particular area. Undernutrition is a common risk factor for childhood infections [ infections childhood for factor risk common a is Undernutrition area. particular a in inate Children present with a variety of common diseases even when one disease such as malaria may predom Findings specifictointegratedapproaches tochildhealth endemic areas. reason for caution in providing iron supplementation to children aged 1 to <6 months of age in malaria– provides finding this anemia, treating for supplementation iron of value the confirmed areas endemic non–malaria– in studies other While [S176]. supplementation iron received who those in significantly increased morbidity severe or death of risk the area,malaria–prone a in age of months <6 to 1 children of study one in that is note particular Of S175]. [S174, studies controlled other in demonstrated been also has zinc receivingchildren in diarrhea of incidence the in decrease A A. vitamin as extent same the to not but age, of months 12–48 children in mortality all–cause decreased supplementation zinc Daily child mortality. Italsodecreases childmortality from pneumoniaandmeasles. thalmia (a condition of eye dryness and eventual scarring produced by vitamin A deficiency) and all–cause xerophof rate decreasethe a in to leads A vitamin with glutamate monosodium market fortifying Even mothers, to and newborns, especially to children 6–59 months of age leads to decreased child mortality. effects.important ally Table providedsupplementation A vitamin to that level household shows the 7 at and non–randomized controlled studies that have been included in this review and that have operation 12 ]. ]. These CHWs often also have community health education roles, perform house 377
Table 7 contains details about randomized June 2017 •Vol. 7No. 1•010904 CBPHC andchildhealthfindings [S141], [S141], Table5 10, ]. Op- 11]. Table8 Ta ------
VIEWPOINTSPaPErS PaPErSVIEWPOINTS Table 7. June 2017 •Vol. 7No. 1•010904 Freeman etal. and Eatthehouseholdlevel weekly IU 8333 A vitamin Supplemental Vitamin Asupplementation: Randomized controlled interventions: I dose 50 and children, 100 children, (200 VitaminA late supplementation fo and daily IU 3330 A vitamin Maternal Vitamin 4mo(60 Aevery mo 200 A Vitamin Vitamin Agivenatbirth(50 days 1and2afterdelivery Infants received 24 000 IU of vitamin A on later mo 6–8 again and age of mo 1–3 at dren Vitamin A 200 Vitamin A60 children 4mo every 100 and children 200 A Vitamin ic(0m)wel o n erMriiy809children, Morbidity Zinc (70mg)weeklyforoneyear mg zinc6daysaweek Vitamin A 200 Zinc supplementation: nutrition education by accompanied and year a twice infants children and 100 200 A Vitamin infants children and 100 200 A Vitamin with 10mgofzinc Daily supplementation al upeetto ih1 go icMriiy854children Morbidity Daily supplementation with 10 mg of zinc iron andBvitamins containing food) of top on sprinkle to der pow (a “Sprinkles” of households to Sale daily (10mg) zinc µg) (5 acid folic mg), (12.5 Iron, folate and zinc supplementation: iron Iron supplementation: with diarrhea) Zinc 20mg zinc daily for 15 d (for children ent powdercontainingiron for2mo micronutri with fortification home Daily controls) and daily MMP sachets 2 to (compared week a times 2 sachets 2 supplement: (MMP) powder mineral and Multivitamin visits from CHWs door–to–door immunization, mothers, to iron of Provision children, to provision A Vitamin practices, weaning on Education sold inmarketswithvitaminA gluconate monosodium of Fortification Vitamin Asupplementation: Non–randomized controlled interventions: nterventIon Studies ofmicronutrient supplementationatthecommunitylevel 0 Ueey4m otlt 28 Mortality 4mo 000 IUevery
000 IU for 1–3 mo–old chil
0 I fr –1 mo–old 6–11 for IU 000 000 IU as one dose plus 10
000 IU for 12–59 mo–old mo–old 12–59 for IU 000 000 IU for 12–59 mo–old mo–old 12–59 for IU 000 000 IU for 12–59 mo–old mo–old 12–59 for IU 000 000 IU every 6 mo for 18 18 for mo 6 every IU 000
000 IU for 12–59 mo–old mo–old 12–59 for IU 000
000 IU for 1–11 mo– old 000 IU –5m) in a single single a in –5m) IU 000 000 IU for 1–11 mo–old
000 IU for 1–11m–old 1–11m–old for IU 000 0 U otlt 28 Mortality 000 IU) 0 U otlt 93nwon l–as natmraiyDcesdb 5 .4 [S169] 0.045 Decreased by15% All–causeinfantmortality 7953newborns Mortality 000 IU)
- - - - outcome t otlt 7764children, 0 Mortality otlt 3786children, 0 Mortality otlt 3389pregnant Mortality obdt 12 Morbidity otlt 5786newborns Mortality otlt 25 Mortality otlt 9200children, 0 Mortality obdt 148children, Morbidity status nutritional Change in obdt 1405children, Morbidity otlt 21 Mortality obdt 561children, 0 Morbidity obdt Children, 1to<6 Morbidity obdt 139children Morbidity status nutritional Change in obdt 115children, 0 Morbidity otlt 6663children, Mortality 5755children 0 Morbidity ype
oF
to <5y study p to <5years children women and 6–72 mo 9–72 mo 0 to<5y 6–72 mo to <5y 6–18 mo 6–72 mo 0–36 mo 720 children 6–47 mo 485 days 12–48 mofor 6–48 mo to <5y mo 6–35 mo to <5y 1103 children, 0 the 3groups to <5yineachof 14 0–35 moand to <5y opulatIon 630 children, 109 children, 000 children, 630 children, 274 children, 551 women
area
sIze
oF
378 s of deathinboys risk girls; in death of Risk –9m otlt erae y2%00 [S14] 0.05 Decreased by26% 1–59 momortality mortality neonatal and Perinatal, ness blind night of Incidence tality rateformeasles 1–59 mo mortality; case fa 6m oflife 1st the during Mortality –9m otlt erae y3%00 [S167] 0.01 Decreased by34% 1–59 momortality –9m otlt erae y1%00 [S165] 0.05 males fe Decreased by19% in mortality mo 1–59 1–59 momortality niec fpemnaDcesdb 4 .1[S83] 0.01 Decreased by44% Incidence ofpneumonia rvlneo aai erae y3%<.0 [S172] <0.001 Decreased by32% Prevalence ofmalaria rvlneo tnigDcesdb 1 .1[S171] 0.01 Decreased by11% Prevalence ofstunting children. normal in infection tory respiraacute of Incidence - 48 mo 12– children in mortality all–cause of risk Relative children 0 in diarrhea of Incidence rvlneo nmaDcesdb 9 .0 [S177] 0.001 Decreased by19% Prevalence ofanemia ceived iron re that groups in death and malaria) severe (from morbidity severe of Risk rhea Duration of persistent diar tration Mean hemoglobin concen supplement MMP with compliance anemia, of Prevalence mortality Bitot’s of spots; Prevalence among children 6–35 mo mortality monia–specific pneu mo; 6–35 children All–cause mortality among
pecIFIc
outcome to < 2 y ------e Decreased by48% 59%; by Decreased erae y2%00 [S162] 0.01 Decreased by20% erae y5%001[S164] 0.001 Decreased by50% Decreased by 30%; de creased by76% erae y2%00 [S168] 0.02 Decreased by22% erae y9%000 [S166] 0.0001 Decreased by90% nrae y8 .5[S170] 0.05 Increased by8% erae y1%005[S173] 0.045 Decreased by18% erae y2%001[S174] 0.001 Decreased by25% nrae y1%00 [S176] 0.02 Increased by12% erae y2%00 [S175] 0.01 Decreased by28% nrae y7 .0 [S178] 0.001 Increased by7% to daily compared group week a times 2 in greater 200% MMP; daily in 32% by Decreased that for children in in in children for that than greater times 1.8 the control villages was in children pre–school among rate mortality 600%; by Decreased tervention villages tervention Decreased by 32%; de creased by53% FFect www.jogh.org
compared
to • doi:10.7189/jogh.07.010904
control - - - s 0.0001; 0.0001; 0.001; 0.001; 0.001; 0.001; tat 0.001 0.001 0.001 0.001 sIgnIFI 0.01; 0.01; 0.05; 0.05; 0.04 cance I stI cal -
r [S161], [S161], [S163], [S163], [S180], [S180], [S181], [S181], [S161] [S163] [S180] [S179] [S181] number eFerence * www.jogh.org • doi:10.7189/jogh.07.010904 months ( 1–year period after iCCM implementation, from 0.83 cases/1000 child–months to 3.80 cases/1000 child– munity–based PHCprograms thatare presented inTable 9. munity level in order to make essential services readily available. Our review includes a number of com com the providedat be to need also facilities fromhealth requireoutreach that immunizations as such morereferredServices or be severebirth. uncommon can give and illnesses wherecan ries and mothers inju illness, severe with patients which to facility health local the to linkages good have to need CHWs able approaches. For integrated CBPHC to be effective at the community level outside of health facilities, of women of childbearing age and children but also of men and older women) using practical and afford treatment actions aimed at meeting all the common health needs of community members (especially those and preventive essential of range comprehensive a of provision the includes (PHC) care health Primary project inourdatabaseBurundi[S195]. Integrated community–basedprimaryhealth care (CBPHC) Group Care another in found were results Similar [S194]. histories birth maternal on based assessment over the five year period of project implementation, confirmed by an independent retrospective morality rate mortality under–5 the in decrease 42% a and rate mortality infant the in decrease 49% a in sulted re approachGroup Care the that demonstrated assessment This system. registration events vital based community– a of quality the also but Groups Care of efficacy the only not demonstrated project this of community–based vital events registry system as part of the activities of the Care Groups. The assessment Another Care Group project in the rural part of the Chokwe District in Mozambique also incorporated a $2.78, respectively [S193]. adjusted life year (DALY) averted, and the annual cost per beneficiary were US$ 441, US$ 14.72 and US Tool (LiST), the project saved an estimated 6848 lives and the cost per life saved, the cost per disability– increased from by 42% in Area A and by 20% in Area B. Based on findings obtained with the Lives Saved Area A and 25% in Area B; the percentage of children 9–23m of age who ate three or more meals per day in breastfeeding60% increasedexclusive by of rates AreaB; in 71% by and AreaA in 45% increasedby (WAZ<–2.0 SD) decreased by 6% in Area A and by 10% in Area B; insecticide–treated bed net (ITN) use undernutrition with children of proportion overall the that were outcomes Key findings. endline with baseline comparing health child of indicators most across achieved were improvements Major begun. sub–areas (A and B) since project activities began several years later in Area B after activities in Area A had In a 5–year Care Group project in Sofala Province in Mozambique, the project area was divided into two up isreadily possible[13, scaling approach, basic and structure this With learned. just they messages new the sharing neighbors, her of 10–15 visiting regularly for responsible is volunteer Each messages. education new some learn health educators. The Care Group meets every two weeks with a project facilitator for two hours or so to community–based as act who volunteers community 10–15 of group a is Group Care A projects. eral CareGroups were reviewthe in included through resultssev the of of evaluation publication the the of Care Groups other studies cited in fined as correct treatment) showed a 27% ( were diagnosed with pneumonia, 91% of them received amoxicillin treatment. Overall performance (de therapy.combination artemisinin received RDT–positive as diagnosed were who who childrenthose Of of 94% and malaria for RDT a received fever a with presenting children of 88% ORS. received rhea ining and treating children. On follow up evaluation after training, 88% of children diagnosed with diar tients using iCCM protocols revealed a strong adherence to the iCCM protocol in terms of testing, exam pa treat to trained were sellers drug private which in study One sellers. drug by managed are illnesses In many parts of rural Uganda with limited access to trained health staff, up to 50% of cases of childhood based onbaselinetrends [S192]. health facility use declined significantly by 15%. These decreases were significantly greater than expected 191]. ment and that iCCM leads to more children receiving treatment for these common illnesses [S186, S189– treatand diagnosis iCCM in skills CHW of maintenance to lead fromcan visors facility health local the respectively. On average, total under–5 mortality rates declined significantly by 38% ( 38% by significantly declined rates mortality under–5 total respectively.average, On P < 0.001) and from 0.25 cases/1000 child–months to 5.28 cases/1000 child–months ( Table 8 demonstrate that monthly community–level supervision by trained super 14]. P 379 = 0.001) increase compared with baseline levels [S188]. The
June 2017 •Vol. 7No. 1•010904 CBPHC andchildhealthfindings P < 0.001), and and 0.001), P < 0.001), 0.001), ------
VIEWPOINTSPaPErS PaPErSVIEWPOINTS June 2017 •Vol. 7No. 1•010904 Table 8. Freeman etal. Community CaseManagement(iCCM) Online Supplementary Document. *See AppendixS2inOnlineSupplementary ARI –acuterespiratoryinfection,HEWhealthextensionworkers,ITN –insecticide–treated bednets,mo–month(s),yyear(s) tem strengthening sys health as well as IMCI, with associated tivities ac community and family CHWs trained as part of the Randomized controlled trials: I sion oftrainingtoCHWs provi- and facilities health Linkage of CHWs with local Non–randomized controlled trials: workers andteachers extension for seminars similar by later y 1 lowed for leaders of all villages fol ducted during the first year con seminars Awareness tension Workers (HEWs) Ex- Health of supervisors trained by C–IMCI on sion supervi monthly On–site Hstandi CMMraiyChildren <5 y in villages Mortality CHWs trainediniCCM health post working atacommunity C–IMCI with2HEWs Hstandi CMCoverage CHWs trainediniCCM CHWs trainediniCCM Peer support groups among iCCM protocols in trained sellers Drug Hstandi CMCvrg 38 Coverage CHWs trainediniCCM nterventIon Studies oftheeffectiveness ofCommunity–IntegratedManagementChildhoodIllnesses(C–IMCI)andIntegrated they are included in all community–based PHC programs [S200, S201]. The CBIO approach was pio was approach CBIO The S201]. [S200, programs PHC community–based all in included are they that so system information project the in included and documented are beneficiaries all that ensure to The census–based, impact–oriented (CBIO) methodology includes mapping and community registering nutrition andcontraceptiveuse, togetherwitha15%improvement incontraceptive coverage[S198]. worked with community volunteers and mobilized community members improving child mortality, child not improve child nutrition or contraceptive coverage [S197]. The best results were achieved when nurses did but mortality child decreasing effectivein were and care providedcurative nurses community–based munity volunteers did not reduce child mortality but did significantly improve child nutrition [S196]. The munity volunteers with community mobilization, and (4) a control group. The group that only had com community mobilization without community health nurses; (3) both community health nurses and com and volunteers community (2) Officers, Health Community alone–called nurses health community (1) Navrongo experiment in Ghana. In the Navrongo experiment in Ghana, there were four groups compared: recurringprograms.a successful be these to of element areS196–198 Assessments three fromstudies the under–5 mortality. Promotion of community involvement and training/deployment of CHWs is also shown decrease can components community–based strong with care health primary that demonstrates Table9 - - - - - status nutritional change in Mortality; t status nutritional change in Coverage; status nutritional change in Coverage; care Quality of care Quality of Coverage care Quality of ype
oF
outcome
p (175 facilities health 10 of areas catchment The population of160 a in y <2 to 0 Children ulation of18 villages with a total pop- age and their children in Women of child–bearing 500 HEWsassessed of 14 population total a with 87 HEWS to <5y 306 1575 children in 6 dis 6 in children 1575 drug sellers 7667 visits to 44 trained Sick children who made tricts opulatIon 009 children <5y
190 children 6 mo 6 children 190 000 persons) 000
sIze
oF 000
study 000
area - s feeding 0to<6mo breast exclusive of prevalence 0 mortality All–cause least fivemealsperday at receiving children of age mo fully immunized; percent- 12–23 children of Percentage vere undernutrition percentage of children with se- coverage; immunization full with children of Percentage ing treatment) initiat- of days two within up followed– and treated, sified, clas- correctly were that cases of (percentage years two over management case of Quality Under–5 mortality gram pro control malaria vertical a in working HEWs to parison com in medications malarial anti– of prescription Correct ed forARI,malaria,diarrhea Number of sick children treat illnesses Correctcommon treatmentof support groups) peer without iCCM in trained CHWs to (compared rhea diar and malaria, ARI, for ed Number of sick children treat ing underITNS sleep- children of Percentage 380 pecIFIc
outcome
to <5 y; y; <5 to - - - - - control e Increased by10.1% Decreased by 13.4%; increased by32% 21%; by Increased decreased by27% 50%; by Increased nrae y20 .4[S186] 0.04 Increased by200% erae y3%003[S185] 0.003 Decreased by38% nrae y1%00 [S187] 0.05 Increased by10% nrae y2%00 [S190] 0.05 Increased by23% nrae y2%001[S188] 0.001 Increased by27% nrae y17 .0 [S189] 0.001 Increased by167% nrae y3%00 [S191] 0.01 Increased by33% FFect
compared www.jogh.org
to
• doi:10.7189/jogh.07.010904 .1 .5[S182] 0.01; 0.05 .5 .5[S183] 0.05; 0.05 sIgnIFIcance s 0.001; 0.001; tat 0.05 IstIcal
r [S184] number eFerence *
- - - Online Supplementary Document. *Appendix S2inOnlineSupplementary CHW –communityhealthworker, healthcare, mo–month(s),PHCprimary TBAtraditionalbirthattendant,yyear(s) www.jogh.org Table 9. assistants health trained by outreach community with center health a at provided PHC TBAs curative treatment, TB control, support of immunizations, feeding, supplemental education, health outreach, with PHC provided byvolunteerCHWs services evaluating and implementing, planning, in involvement community of promotion and referral, education, Peer transport assistance when referral needed and immunizations, education, toring, moni growth of provision A, vitamin of by CHWs to all households, distribution visits frequent with PHC Census–based Non–randomized controlled assessments: services outreach plus CHWs by provided vices ser health child of range full with PHC Community HealthOfficers well–trained and CHWs volunteer with PHC promoting community involvement out CHWs PHC nurses posted in communities with services. outreach plus CHWs by provided vices ser health child of range full with PHC Randomized controlled assessments: I nterventIon Primary healthcarePrimary programs thathavestrong community–basedcomponents • doi:10.7189/jogh.07.010904 based on a qualitative review of data were that: one–on–one education of mothers was essential for im for essential was mothers of education one–on–one that: were data of review qualitative a on based who were treated with penicillin had a 42% reduced risk of overall mortality [S206]. Other key findings dehydration from diarrhea and for childhood pneumonia. The children 0–3 years of age with pneumonia Key CBPHC aspects of this project were that Family Health Workers provided treatment in the home for pared tocontrol cell[S205]. nutrition the in and cell nutrition the in greater significantly were age of months 17 beyond children of height–for–age and weight–for–age the addition, In cell. control the to pared nutrition the in as well as groupscell age nutrition month the 12–23 both and in study,nutrition post–neonatal, neonatal, perinatal, were the rates significantly during mortality reduced the In prevention. disease concerning education and immunizations, treatment, early and surveillance disease infectious included services care health daily.child twice The supplementation food as well as promotion and monitoring growth included services nutrition Child children. 200–300 approximately study.this of design contained the cell of Each aspect key a was participation community of Promotion care cell, and (D) a control cell (in which routine government services without outreach were provided). trolled study: (A) a nutrition–only cell, (B) a health–care–only cell, (C) a combined con nutrition–and–health– non–randomized this of aspect nutrition the in cells four were There health. child and CBPHC to rural Punjab of Northern India. The nutrition and health–care aspects of this study are of direct relevance ngwal Project, which pioneered many elements of CBPHC [S205]. It operated from 1967 to 1973 in the into incorporation to itself lend not does that database our in study important One used onlyforhealthpromotion andreferral forprovision atahealthcenter. ofhealthservices carried out at a time when there had not yet been many experiences with CHWs and when CHWs were The last assessment in ing compared tonationalruralindicators[S202,S203]. pometric demonstrated surveys a 68% reduction in under–5 mortality and reduced prevalence of stunt neered in Haiti in the 1970s. Assessment by retrospective maternal birth histories and household anthro - - - - Mortality 887 persons in 887personsin Mortality 2700children Mortality 36 Mortality otlt 15 Mortality otlt 6663children Mortality otlt 51 Mortality 2000children Mortality status nutritional Change in outcome t ype
oF Table 9 is the earliest one in our database and was reported in 1951 [S204]. It was
catchment area health center aged 0–6y <5 y area) intervention population of 14 0–35 mo, <5 y <5 y 6–23 mo 788 children study p opulatIon 000 children 406 (total 406 (total 551 women 407 children
area
sIze
oF
381 of 10yuntil1951 groups over a time period age all of mortality Crude tality; stunting mor under–5 All–cause tality mor under–5 All–cause tality mor under–5 All–cause in children 6–35mo mortality specific monia– Pneu mo. 6–35 children in mortality All–cause ne– otlt erae y5%00 [S197] 0.05 more than2y for intervention to posed Decreased by54% ex children of Mortality Under–5 mortality Weight–for–age Zscore score, Z Height–for–age s pecIFIc
outcome - - - - - erae y2%001[S204] 0.001 Decreased by24% children 48–59m in 28% by reduced 67%; by Decreased [S201] 0.0001 Decreased by58% erae y5%001[S200] 0.001 Decreased by52% Decreased by53%. 32%. by Decreased erae y6%001[S198] 0.001 Decreased by60% increased by14% 24%, by Increased control e FFect
compared June 2017 •Vol. 7No. 1•010904
to
CBPHC andchildhealthfindings + + health care cell com cell care health health carecom- health cell TableNara- the is 9 sIgnIFIcance s 0.0001, 0.0001, 0.018, 0.018, tat 0.001 .0 [S199] 0.001 0.05 IstIcal
r [S202], [S202], [S203] [S196] number eFerence *
- - - - -
VIEWPOINTSPaPErS PaPErSVIEWPOINTS June 2017 •Vol. 7No. 1•010904 Freeman etal. projects includedinourreview. the in met been have to appear conditions these and effectiveness, achieve to needed are communities here. However, it is clear that appropriately trained, supervised and supported CHWs along with engaged explored adequately be cannot conditions) field routine under (implementation studies effectiveness to opposed as conditions) field ideal under project implementation is, (that representefficacy studies here included assessments the which unfortunately.to degreequestions, The these address really sessments tions under routine conditions. Answering these questions is beyond the scope of this paper, and few as ventions, and (3) what are the conditions that would need to be met in order to - scale up these interven inter the deliver orderto in need they resourcesdo particular what (2) interventions, the implemented who workers community–level the are who (1) questions: Weimportant threehere addressed not have healthcare.”individuals, usuallyCHWs,isconsistentwith“community–based primary by implemented interventions two or one of effectiveness the Consequently,about ventions. evidence inter two or one only do may whom of each team a in working CHWs several amongst spread be may being implemented – even in a comprehensivenumber of health interventions care primary approach – effectivenessthe on improvingfor strategy a as CBPHC of total the level community the At health. child that this finding pertains to the biomedical interaction of iron on children exposed to malaria rather than effect.favorable However,than less a note to led important is it intervention CBPHC of plementation im which in identified have we evidence However,only time. the this is at this recommended not it so effects harmful have may otherwise, or approaches community–based through whether areas, endemic health are similar to those reported in other reviews [ improvingfindings for Our regarding child effectivenessinterventions the community–based specific of provide evidencesupportingourmajorfindingspresented here. also they but limitations, space to due article this in included not were health child improving in PHC significant and operationally important results. Other less rigorous assessments of the effectiveness of CB [ demonstrated has projects these of set a of review a as population, national or regional the in coverage in changes smaller much to comparison in particularly terventions. In virtually all cases, the changes in coverage over a 4–5 year period were quite pronounced, in survival child key of coverage population in changes measuring group, comparison a without signs Some assessments, mostly unpublished child survival project evaluations, relied on before/after study de changes inthemostobjectiveandmeaningfulindicator:mortality. been have observed outcomes the cases many In effects. important operationally and significant cally controllednon–randomized and domized producedconsistently have that database our in trials statisti and community–level workers. For all categories we of have interventions, presented findings from ran communities with working by facilities health of outside level community the at addressed be can tries This review provides strong evidence that overall the major causes of child mortality in developing coun DISCUSSION These projects are discussedindetailelsewhere inthissupplement[ • • • • dence ofthelong–termbenefitsCBPHCprojects onchildhealth.Theseprojects are: evi the document they since note particular of are database our in included assessments other Several community participationandbuildingtrustwiththe[S207]. careprogramhealth quality required a developing active and trust, building of element essential an was feeding programs was best accomplished at home or near the home; having a curative health care service special through children malnourished of rehabilitation effectiveness; and coverage improved possible in essary order to achieve a reduction in infant mortality; delegation of as services far to the as periphery werenec visits home diarrhea;weekly with child a feeding not overcomingabout for beliefs traditional proving practices related to breastfeeding, infant feeding, rehydration and feeding of sick infants and also SEARCH(SocietyforEducation,ActionandResearch inCommunityHealth)Gadchiroli, India. TheJamkhedComprehensive HealthProject inJamkhed,India;and, TheHôpitalAlbertSchweitzerinDeschapelles,Haiti; Health, PopulationandNutrition); for Bangladesh/Centre Research, Disease Diarrheal for Centre International the for site field research planning family and health maternal/child (a Bangladesh Matlab, in Program MCH–FP ICDDR,B The
382 4 , 17]. The provision of iron to children in malaria– ]. They have generally produced statistically statistically produced generally have They 16]. 15]. www.jogh.org • doi:10.7189/jogh.07.010904 ------www.jogh.org • doi:10.7189/jogh.07.010904 proving neonatalaswellchildhealth[18]. maternal, strategies are discussed in detail elsewhere in this series from the perspective of CBPHC strategies for im These up. follow or illness serious of management for facility health nearest their to members munity com protocols,link Throughwell–developed CHWs illnesses. following common manage and vention pre about caregivers child educate to households visit CHWs centers. health peripheral at based teams (3) use of participatory women’s groups; and (4) outreach services provided in the community by mobile gies are (1) house–to–house visitation by CHWs; (2) community case management of childhood illness, interventions are effective and commonly used in projects that have improved child health. These strate This review demonstrates that four major strategies for delivering community–based primary health care P 0.10 (HR baseline of that tenth one was area intervention the in mortality under–5 of hazard the tion, interven the of years three After programming.prevention and mobilization community development, infrastructure fee removal, user finding, case active CHW included intervention The [S93]. Mali madjo, Yiriin occurred conditions these meeting by achieved effectiveness the of example recent important ly community members are not convinced that the CHWs are well trained and competent. One particular if and households all with contact regular in not are CHWs the if develop to difficult is trust this ever munity trust in the CHWs providing it. It also facilitates referral to local health facilities as needed. How As the Narangwal project demonstrated four decades ago, the provision of some curative care builds com more than2hoursawayfrom theirhealthpost)makefrequent homevisitsvirtuallyimpossible. care for adults and family planning for women, as well as the large size of their catchment curative providing areas including (sometimes iCCM, beyond responsibilities job broader Their confidence. their earn to or care seek should they which for signs warning about mothers educating to attention sufficient give to [ a single rectal dose of artesunate (n months with suspected severe malaria who could not be treated orally were allocated randomly to receive Tanzania,72 and to Ghana 6 Bangladesh, aged in patients controlledtrial randomized a In [20]. sunate pre–referralof use screeningthe our escaped process.rectalhave studies concerns arte study such One important some criteria, our meet that studies relevant all include to effort every made have we While achieved through strong communityengagement. be otherwise might that outcomes best the produce not may programs engagement, community out munity members may see programs as imposed from the outside and not responsive to their needs. stakeholder,com Witha being Without evaluation. and implementation planning, program for sponsibility re shared a have they because staffprogram health with shared be to likely more is locally works what about have members community that knowledge the engagements, such of result a As [19]. outcomes nity can lead to cost–effective transformation and lasting behavior change that produces improved health atic review of child survival programs has found that programs working collaboratively with the commu Many assessments included in our review support the importance of community engagement. A system by community participation and door–to–door provision of support and health education [ education health and provisionsupport door–to–door of and participation community by nutrition that require a high level of trust between community members and providers that can be achieved vision of prompt curative treatment, they did not have significant impact on contraceptive use or on child munity. While the community–based nurses did have some impact on child mortality through their pro through the traditional community structure and engaging persons with a leadership role working within on the com focused mobilization community of processes particular The mobilization. community and volunteers community trained with conjunction in worked nurses community the when achieved The assessments from the Navrongo project in Ghana [S196–198] demonstrate that the best results were tive antimalarial with 24 hours of symptom onset was nearly twice that reported at baseline ( er,( endline at from23% to baseline at 38% [ outs stock drug and supervision training, to related shortcomings of mortality,because under–5 reducedpartly or interventions survival child key of coverage expanded not has scale, at implemented when iCCM, that found have 2015) December (31 review our for selected ment (iCCM) ( Several studies included in our review confirmed the effectiveness of Integrated Community Case Manage groupnent disabilitybyhalf(1.9%intheintervention compared to3.8%inthecontrol group). perma or death reduced significantly artesunate pre–referral hours, 15 within clinic a reached not had whom of half hours, 6 within clinic a reached not had who patients In given. be could injections larial 22, < 0.0001), the prevalence of febrile illness of children younger than 5 years of age was significantly low ]. Perhaps CHWs trained in iCCM are not able to make frequent home visits and therefore unable thereforeunable and visits home frequent make to able not are iCCM in trained CHWs Perhaps 23]. Table 8 ). However, several recent evaluations published since the end–point of publications = 8954) or placebo (n 383 P = 0.0009) and the percentagethe and effec0.0009) childrenan of starting = 8872) before referral to a clinic where antima 21 June 2017 •Vol. 7No. 1•010904 ] and low levels of care seeking seeking care of levels low and ] CBPHC andchildhealthfindings ]. A more A 24]. P = 0.0195). ------
VIEWPOINTSPaPErS PaPErSVIEWPOINTS June 2017 •Vol. 7No. 1•010904 Freeman etal. edge toimprove the implementationofchildhealthprograms atthe communitylevel.Theseareas are: Our review has identified several areas of further study that are needed to address gaps in current knowl we havehighlightedthemin thediscussion. portant findings in papers published after December 2015 that did not fit the timeline of our review, but have missed would alter the overall findings from our review. In addition, we are aware that there are im might we articles any of inclusion the that think not do However,procedures.we screening our by up We were picked our that not for review qualified that assessments some therebe that may acknowledge areintervention valid. specific each to respect with findings general the that convinced are we that such is interventions most to relationship in assessments many across findings of consistency The health. child improving in PHC of the numerous assessments that have been included in our review that demonstrate effectiveness of CB lication bias is present and should be recognized. Nonetheless, publication bias does not negate the value rarely literature.scientific the in or pub documents serious open–access a in that described means This are implementation program in encountered challenges serious and failures project well–known, is As consistent withandsupportiveofthosethatwere cited. are here cited specifically not assessments the of findings However,the analysis. this in cited be could Due to space limitations not all 489 assessments of the effectiveness of CBPHC in improving child health of theseassessments. firm conclusions. We worked to mitigate this limitation by, in some cases, following up with the authors outcomes themselves. This sometimes made it difficult to assess the strength of the evidence and to draw the about as well as strategies, implementation other and members community ology,of role the about method assessment the about sufficient information lacked review our in included studies the of Some Study limitations participation andmobilizationmakeastrong effectiveness. contributiontointervention community that evidence Wepresented centers. also health have at based teams mobile by community women’sthe providedin participatory of services outreachuse (4) (3) groups;and illness, childhood of CBPHC. These strategies are: (1) house–to–house visitation by CHWs; (2) community case management munity–based interventions are effective approaches for achieving implementation effectiveness through The findings from this review also provide strong evidence that the four key strategies of delivering com for children. results population–level favorable with level community the at delivered be can Interventions Essential The strong and consistent evidence that we have presented in this paper clearly demonstrates that all these • • • • • • • also demonstrated [ registering and visiting frequently all households with mothers and children, as more recent evidence has census–based, impact–oriented (CBIO) approach and Care Groups have demonstrated the importance of those households far away, and those who are members of religious or ethnic minorities. In our review, the mothers, all poorestfamilies, the including households, reachall effectivemust most it be to CBPHC For community leadersneedstobefollowedproduce lastingresults atscale[ proach with regular planning, monitoring and supervision of health workers, and close collaboration with recent evaluation of the extension of this program across Ghana indicates that an ongoing systematic ap and/or healthpostbyCHWshavebeenidentified[1 The following essential for interventions child health that can be provided at the level of the community is currently beingimplementedinmanyprioritycountries[ Detectandrefer children withsevere acutemalnutrition Distributeandpromote useofITNsorIRsboth Educatefamiliesonsafedisposalofchildren’s stoolsandhandwashing Provide co–trimoxazoleforHIV–positivechildren Provide vitaminAandzincsupplementation ral ofthosechildren withdangersignsofseriousdisease. refer early with diseases diarrheal and malaria pneumonia, of pneumonia treat and diagnose Prevent, feedingbeginningat6monthsofage priate complementary Promote breastfeeding (including exclusive breastfeeding during the first six months of life) and appro
26, 27] . The Care Group approach has achieved excellent results at low cost [ 384 ]: 13]. www.jogh.org 25]. • doi:10.7189/jogh.07.010904 ] and 14] and ------www.jogh.org • doi:10.7189/jogh.07.010904 be developedcanreach theirfullpotential. ery systems are needed in order for the evidence–based interventions currently known and those that will ly implemented at the community level to improve child health and (2) robust community–based deliv ing MNCH, our overall findings strongly support the conclusion that (1) CBPHC can in fact be effective Nonetheless, consistent with the purpose of our overall review of the effectiveness of CBPHC in improv health. more broadly to better support the implementation of effective CBPHC interventions for improving child tervention is beyond the scope of this paper, as is the important issue of how to strengthen health systems implementation do of not one override intervention the requirements for implementation of another in for demands the that so services of package balanced a into interventions integrate effectively most to proach compares with another in terms of effectiveness. Moreover, addressing the important issue of how ap- implementation or intervention given any how or implementation, to approach specific any or tion finitive statements about the strength of the evidence or the magnitude of effect for any specific interven implemented, and (3) the outcome measures used to assess outcomes, it is not possible to make any de Given the heterogeneity of (1) the types of interventions implemented, (2) the manner in which they were ed inasustainablemanner. Howtobestdothisneedsfurtherinvestigation. in the field depend on how well community members “own” and therefore use the interventions provid many interventions can be implemented successfully at the community level, the actual results produced need for more evidence from programs delivered at scale. Similarly, while we have provided evidence that ventions in large populations at scale. In the final paper of this series [ As can be readily seen from the tables in this paper there is a clear lack of assessments of studies of inter • • ing thattheseconditionsare metwillbethemajorchallengeindecadetocome. in order for to these be interventions effective at scale in routine settings in priority countries and ensur and well–supported CHW cadre in sufficient numbers. Understanding the conditions that need to be met requirewill bility strongstrong a as well as system health well–trained the a for to support commitment inhigh–mortalitysettingsareversal coverageoftheseinterventions clearlyneeded.Achievingthiscapa uni achieving of arecapable that systems Health age. of months children1–59 of improvinghealth the We for effectiveness of presentedevidence broadhave the ainterventions ofcommunity–based of range CONCLUSIONS closure.pdf (available uponrequest from thecorresponding author),and declare noconflictofinterest. Conflict of interest: and approved thefinaldraft. drafts earlier of revision the in participated authors the of All draft. first the wrote PF declaration: Authorship ecution ofthereview. ex the in role no had support financial provided that organizations The Foundation. Gates the and velopment, WorldThe ticle: UNICEF,Organization, Health De International for Agency WorldStates the United the Bank, ar this in described work the conduct to used were that funds providedorganizations following The Funding: made itpossibleforoneofitsconsultants,DrBahieRassekh,toparticipateasamembertheStudyTeam. space, administrative support, and salary support to Dr Perry during the initial phase of the review. The World Bank al Health Section staff, which administered some of these funds. We thank Future Generations for providing office Gates Foundation. We are also grateful to the American Public Health Association and particularly its Internation ganization, the CORE Group (Collaboration and Resources for Child Health)/USAID, Future Generations, and the CEF,Worldthe Worldthe of Development and Health Adolescent and Child of Department the Bank, Or Health We are grateful to the following organizations that provided small grants to cover the expenses of this review: UNI analysis. the and database the assembling the to contributed who assistants research and students many the and manuscript, this of drafts earlier on comments for Bhutta Zulfi thank to wish authors The Acknowledgements:
uation ofthesesettings. Effectiveness studies on how best to involve communities in the monitoring, implementation and eval tions inroutine settingsfor5ormore years; Effectiveness studies of the implementation of community based interventions at scale in large popula All authors have completed the Unified Competing Interest Form at www.icmje.org/coi_dis- 385 28] the Expert Panel highlights the June 2017 •Vol. 7No. 1•010904 CBPHC andchildhealthfindings ------
VIEWPOINTSPaPErS PaPErSVIEWPOINTS June 2017 •Vol. 7No. 1•010904 Freeman etal. rEFEr ENCES 22 21 20 19 18 17 12 11 10 9 8 7 6 5 4 16 15 14 13 3 2 1 2003;362:65-71. Medline:12853204 Lancet. year? this prevent we can deaths child many How SS. Morris ZA, RW,Bhutta Steketee RE, G, Black Jones Accessed: 20March 2017. Available:2013. (GAPPD). Diarrhoea and Pneumonia for Plan WHO. UNICEF. ending preventable child deaths from pneumonia and diarrhoea by 2025. The integrated Global Action health findings.JGlobHealth.2017;7:010903. the effectiveness of community-based primary health care in improving maternal, neonatal and child health: 3. neonatal FreemanP,E, Sacks regardingevidence the comprehensivereview of A al. et S, Gupta B, Rassekh M, Jennings K, Sakyi healthfindings. JGlobHealth.2017;7:010902. ternal ing the effectiveness of community-based health primary care in improving neonatal maternal, and child health: 2. ma- SchleiffS, Pradhan FreemanM, E, Jennings Sacks P,M, comprehensiveA regardal. reviewevidence et the S, of Gupta - database description.JGlobHealth.2017;07:010901. and methods rationale, 1. health: child and neonatal maternal, improving in care health primary community-based of Perry H, Rassekh B, Gupta S, Wilhelm J, Freeman P. A comprehensive review of the evidence regarding the effectiveness 6736(13)60937-X weight in low-income and middle-income countries. Lancet. 2013;382:427-51. Black RE, Victora CG, Walker SP, Bhutta ZA, Christian P, de Onis M, et and al. and child Maternal over undernutrition doi:10.1016/S0140-6736(12)60560-1 2012;379:2151-61. Lancet. 2000. since trends time with 2010 for analysis systematic updated an mortality: child of causes national and regional, Global, al. et JE, Lawn S, Scott J, Perin S, Cousens HL, Johnson L, Liu key messagesfrom DiseaseControl Priorities3rd Edition.Lancet.2016;388:2811-24. Black RE, Levin C, Walker N, Chou D, Liu L, Temmerman M, et al. Reproductive, maternal, newborn, and child health: 2016;94:596-604. Medline:26787148 Trop J Hyg. Am Med trial. randomized cluster A Ethiopia: in mortality child on strategy illness childhood of agement NP,Miller B, Shaw E, Hazel A, Amouzou Tafesse man- case Y,community Mekonnen integrated M, the Effects of al. et doi:10.4269/ajtmh.15-0585 program to reduce under-five mortality in Burkina Faso. Am J Trop Med Hyg. 2016;94:584-95. Hyg. J Med Trop Am Faso. Burkina in mortality under-five reduce to program T,Roberton G, Guiella M, Munos TiendrebeogoA, Maiga TamA, Y, scale-up rapid the of evaluation Independent al. et doi:10.1016/S0140-6736(08)61734-1 death and disability in severe malaria: a placebo-controlled trial. Lancet. 2009;373:557-66. Lancet. trial. placebo-controlled a malaria: severe in disability and death MF,Gomes prevent to artesunate rectalPre-referral WarsameT, al. JO, Agbenyega et Gyapong M, A, MA, Babiker Faiz Suppl 1:67-88.Medline:25207448 andearlydevelopmentinlow-middle-incomecountries:anevidence review.survival JHealthCommun. 2014;19 PP, Souza child O, enhance Fajobi to K, engagement Bose Community SK, al. et Farnsworth LL, Davidson A, Peniston J GlobHealth.2017;7:010906. nity-based primary health care in improving maternal, neonatal and child health: 6. strategies used by effective projects. Perry H, Rassekh B, Gupta S, Freeman P. A comprehensive review of the evidence regarding the effectiveness of commu S0140-6736(13)60996-4 ternal ternal and child nutrition: what can be done and at what cost? Lancet. 2013;382:452-77. Bhutta ZA, Das JK, Rizvi A, Gaffey MF, Walker N, Horton S, et al. Evidence-based interventions for improvement of ma heapol/czt005 agement (iCCM) programmes: a scoping review of programmatic evidence. J Glob Health. 2014;4:020403. Health. Glob J evidence. programmatic of review scoping a programmes: (iCCM) agement man case community integrated selected Training,Marceauin C. careX, of Bosch-Capblanch quality and supervision diarrhea, pneumonia,malaria,andmeasles.AmJClinNutr. 2004;80:193-8. with associated deaths child of cause underlying an as Undernutrition RE. Black M, Blossner M, Onis de LE, Caulfield countries. BullWorld HealthOrgan. 1995;73:443-8.Medline:7554015 developing in mortality child on JP.malnutrition Habicht of effectsDG, Jr,SchroederThe EA Frongillo DL, Pelletier strate plausible evidence for child mortality impact. Health Policy Plan. 2014;29:204-16. Ricca J, Kureshy N, Leban K, Prosnitz D, Ryan L. Community-based intervention packages facilitated by NGOs demon term impactonmortalityinchildren youngerthanfiveyearsofage.JGlobHealth.2017;7:010907. long- of evidence with programs 7. health: child and neonatal maternal, improving in care health primary nity-based Perry H, Rassekh B, Gupta S, Freeman P. A comprehensive review of the evidence regarding the effectiveness of commu 81. Medline:26374799 and child health outcomes achieved in high-mortality, resource-constrained settings. Glob Health Sci Pract. 2015;3:370- Perry H, Morrow M, Davis T, Borger S, Weiss J, DeCoster M, et al. Care Groups II: a summary of the maternal, neonatal 69. Medline:26374798 egy for improving maternal, neonatal, and child health in resource-constrained settings. Glob Health Sci Pract. 2015;3:358- Perry H, Morrow M, Borger S, Weiss J, DeCoster M, Davis T, et al. Care Groups I: an innovative community-based strat line:25520793 pdf?sequence=3&isAllowed=y Available:World Bank, Washington,DC: Edition. Third Health, Child and Newborn, Maternal, Reproductive, Priorities: Control Disease In: 2016. health. child and newborn, maternal, improve to care Community-based ZA. Bhutta R, Kumar ZS, Lassi doi:10.7189/jogh.04.020403
doi:10.9745/GHSP-D-15-00052 doi:10.9745/GHSP-D-15-00051 https://openknowledge.worldbank.org/bitstream/handle/10986/23833/9781464803482. . Accessed:20March 2017. doi:10.1080/10810730.2014.941519 doi:10.1016/S0140-6736(03)13811-1 doi:10.4269/ajtmh.15-0586 386 http://www.unicef.org/immunization/files/GAPPD.pdf Medline:15213048 Medline:23746772 www.jogh.org Medline:23434515 Medline:23746776 • doi:10.7189/jogh.07.010904 Medline:19059639
Medline:26787147 doi:10.1016/S0140- Medline:22579125
doi:10.1093/ doi:10.1016/ Med------.
www.jogh.org rEFEr ENCES • doi:10.7189/jogh.07.010904 24 23 26 25 28 27 services: services: the Navrongo experiment in Ghana. Bull World Health Organ. 2006;84:949-55. Phillips JF, Bawah AA, Binka FN. Accelerating reproductive and child health programme impact with community-based Trop MedHyg.2015;93:636-47.Medline:26195461 J Am Ethiopia. in Illnesses Childhood of Management Case Community Integrated of Scale-Up of Context the in ers Shaw B, Amouzou A, Miller NP, Tsui AO, Bryce J, Tafesse M, et al. Determinants of Utilization of Health Extension Work gladesh. JUrbanHealth.2016;93:6-18.Medline:26830423 slums: the BRAC Manoshi Project's experience with community engagement, social mapping, and census taking in Ban Marcil L, Afsana K, Perry HB. First steps in initiating an effective maternal, neonatal, and child health program in urban 2005;23:189-213. doi:10.2190/NGM3-FYDT-5827-ML1P 2004- Educ. Health Community Q Int appraisal. systems qualitative with up scaling to approach services and ning Nyonator FK, Jones TC, Miller RA, Phillips JF, Awoonor-Williams JK. Guiding the Ghana community-based health plan BLT.06.030064 conclusions andrecommendations ofanExpertPanel.JGlobHealth.2017;7:010908. garding the effectiveness of community-based health primary care in improving neonatal maternal, and child health: 8. re- evidence the of reviewcomprehensive A al. et AMR, Chowdhury ZA, Bhutta TaylorA, RE, Bang Black AroleS, CE, goals. In:E.B,editor. AidEffectiveness inGlobalHealth.NewYork: Springer;2015. health global addressing in approach (CBIO) impact-oriented census-based, the T.of effectivenessDavis The H, Perry 387 doi:10.4269/ajtmh.14-0660
doi:10.1007/s11524-016-0026-0 June 2017 •Vol. 7No. 1•010904 Medline:17242830 CBPHC andchildhealthfindings
doi:10.2471/ - - -
VIEWPOINTSPaPErS PaPErSVIEWPOINTS June 2017 •Vol. 7No. 1•010905 children child health:5.equityeffectsforneonatesand health careinimprovingmaternal,neonataland the effectivenessofcommunity–basedprimary Comprehensive reviewoftheevidenceregarding material. The onlineversionofthisarticlecontainssupplementary Electronic material: supplementary Sundeep Gupta rassekh Kumapley [email protected] USa Baltimore, MD21205 615 NorthWolfeSt. Health Johns HopkinsBloombergSchoolofPublic r Henry Perry Correspondence to: 6 5 4 3 2 1 Meike Schleiff The WorldBank,Washington,DC,USa Medical epidemiologist,Lusaka,Zambia Washington, Seattle,USa Department ofGlobalHealth,University USa Independent consultant,Seattle,Washington, UNICEF, NewYork,USa Baltimore, Maryland,USa Hopkins BloombergSchoolofPublicHealth, Department ofInternationalHealth,Johns oom E8537
6
, HenryBPerry
2 , PaulaFreeman 1
, richard
5 , BahieM 1
3,4 , Results work matrix. Weysis. organized around conceptually findings our frame logical a identified through a multi–step process, that included an equityprojects, anal 42 of sub–set a from equity to related evidence tracted ex- we health, child improve to CBPHC used that projects) as tively collec to (referred programs or projects interventions, 548 of ment assess the about information containing database a Using Methods population isanimportantone. a in children disadvantaged most the of status health the and vices ser health of utilization improve does actually or can (CBPHC) care health primary community–based which to degree the about tions worst—is a growing throughoutconcern the world. Therefore, ques tion—where utilization of health services and health status is often the popula the of segments disadvantaged most the of health the prove Background versal coverage of essential services forchildren.versal coverageofessentialservices strategy for reducing inequities in child health and for achieving uni a high burden of child mortality. Strengthening CBPHC is a necessary PHC projects, are important for reducing inequities in countries with CB in done be can as underserved, the to attention with mentation attention to and tracking of metrics across all phases of project imple one’sgreatlywith away.distance and services of decentralization The diminishes utilization where facilities, at only services strengthen that projects than equitable more be to likely inherently are projects such households, all to down reach often even that and population monly provide services that are readily accessible to the entire project Conclusions of thepopulation. vantaged segment of the project population than in the other segments disad- most the in more improved criterion equity the that meaning “pro–equitable,” were effects equity measured the all the of (78%) Most segments. advantaged more the in as population project the proved to the same degree or more in the disadvantaged segments of im project’sindicator(s) the equity that meaning effects, “equitable” or “pro–equitable” demonstrated projects 42 these for reported and out carried measurements equity 82 the of (87%) majority vast The effects. equitable achieved interventions, health child implemented Our analysis indicates that these CBPHC projects, all of which 388 The degree to which investments in health programs im Based on the observation that CBPHC projects com projects CBPHC that observation the on Based www.jogh.org • doi:10.7189/jogh.07.010905 global journal of health ------www.jogh.org • doi:10.7189/jogh.07.010905 “ proclaimed, Rights, Human for Committee Medical the to 1966 in speech a Jr.,in King, Luther Martin agenda from globalandnationalpolicy–makerstomajordonors[ and working to reduce inequities — with a goal ultimately to reach zero — has been on the global health [ hoped had stakeholders and countries many as much as diminishing not are inequities world, the around status health and programming health in provements they can be greatly reduced or even eliminated through stronger health programs. In spite of marked im because inequitable considered large and by are status health in inequalities countries, within and tries Recent evidence from tracking of the “Countdown to 2015” [ 2015” to “Countdown the of fromtracking evidence Recent done inorder toacceleratereductions inhealthinequities. ing equity components into health policies. Nonetheless, a great deal of learning and work remains to be health equity by building evidence, addressing the social determinants of health (SDH), and incorporat achieving for action mobilize and focus to world the around from associations health public national inequities are slowly improving, substantial challenges remain for how to accelerate this progress[ this accelerate to how for remain challenges substantial improving, slowly are inequities ie, the greatest burden of neonatal maternal, and child mortality). Even though some measures of health tiles in most Countdown countries (the 74 countries with 97% of the world’s child and maternal deaths, However, the poorest quintiles are still facing markedly lower levels of coverage than the wealthier quin quintiles. wealthiest the for that than faster rate a at populations national poorestof the quintiles for ing be achieved, shows that population coverage of key interventions provided by health services is improv Sustainable Development Goals (SDGs) [ [ (MDGs) Goals orderDevelopment in Millennium attention the achieve to nization coverage [ preventbreastfeeding,to exclusive insecticide–treatedimmu usage malaria, of and (ITN) use net the as countries (LMICs) are being increasingly studied. Some progress is being made in a number of areas such Issues of health inequities for neonatal maternal, and child health (MNCH) in low– and middle–income ]. We[14]. decades interpretationsfollowing article. the this use of will context the in terms the of The terminology around inequities, inequalities, and disparities has been the topic of debate over the past by 2014thisdifference haddisappeared [13]. example, in Cambodia and Sierra Leone in 2000 the richest had much higher coverage than the rest, but are those that effectively reached the poorest families [5 [ outreachlevel community the at strategies through delivered be can that interventions for than attendant) birth skilled a as (such provider health The gaps are wider for interventions that require access to fixed health facilities or repeat contacts witha program. One of the recent drivers for this scrutiny was the challenge of meeting the MDGs by 2015 and The equity effects of MNCH programs have undergone perhaps the greatest scrutiny of any global health ly [7 Overall improvements in the health of a population can occur without every sub–group benefiting equal sub–populations. defined geographically and socially different among outcomes health and mentation imple program health both of equity the examine to important is it perspective, health public a From of populations rather than more proximal indicators of health system inputs or health service utilization. than onthebasisofpoliticalorsocioeconomicprivilege” [ rather status] [health need demonstrated to according benefits of “distribution the, as equity of inition progress and even the flow of funding for different interventions [ determine to used measures the determine can definitions these and practice, and policy for plications the semantics, Braveman argues that how we define and use these terms has important and relevant im cial barriers that can lead to health outcomes that are different from those of other social groups. Beyond produceso persistently groupthat social a of characteristics other der,or religion, location, geographic Such differences stem from characteristics such as educational level, income (or wealth), race, child’s gen er utilizationofhealthcare services. groups within geographically defined populations have less access to health care resources and have low disadvantaged and marginalized which to degree the on focus often health in inequities of studies tice, but focuses specifically on the sub–set of differences that are ‘avoidable, unfair, and unjust” [ ities and of epidemiological inquiry [ entirety the encompassing essentially – status) health as herereferred to (collectively mortality or bidity fined groups in health service utilization, in risk factors for unfavorable health outcomes, in levels of mor Of all the forms of inequality, injustice in health care is the most shocking and inhumane and inequality,shocking of most carethe forms health is the in all injustice Of , 16, 17]. inequalities (often used interchangeably) refer to differences among socially or geographically de 7 ]. Further, approaches for reaching underserved populations are receiving increasing 14]. Inequity, however, “does not refer generically to all differences in health, 8 ]. At the global level, a recent declaration [9 389 5 ]. The countries that have made rapid progressrapid made have that countries The ]. coverage in ]. This is despite starting with great inequities. For ]. He focused on equity of the health status 15]. Hefocusedonequityofthehealthstatus 2 – 4 7 ]. Particularly since the 1990s, measuring measuring 1990s, the since Particularly ]. – ], when the MDGs were supposed to to supposed were MDGs the when 12], 14]. Alternately, Taylor suggested a def- 3 – 6 ]. June 2017 •Vol. 7No. 1•010905 4 ] and the newly established established newly the and ] ” [1 ” CBPHC andequityeffects ] brought together ]. Between coun Between ]. 14]. In prac- Dispar 3 , 4 ]. ]. ------
VIEWPOINTSPaPErS PaPErSVIEWPOINTS June 2017 •Vol. 7No. 1•010905 Schleiff etal. accelerating progress in countries that were lagging behind [ behind lagging were that countries in progress accelerating ferred to collectively as projects) that used CBPHC (defined in the initial paper in this series [ We used a recently assembled database containing assessments of 548 studies, projects or programs (re Data sources METHODS data analysis strategy,analysis data the in inclusion for assessment providedthe reviewersnotes of the the by in 4) or database: 1) the title of the article, 2) the documentation of the process of the intervention, 3) part of the 138 articles in which equity was mentioned in one or more of the following fields in the CBPHC project Using the PRISMA guidelines for systematic reviews on health equity [ Article review andinclusionprocess of theirassessmentusingtheprocess described in thefollowingsection. part as analysis equity projectsan 42 out of carried sub–set had From a that identified we database, this from thelatterreview were transferred toelectronic database. review.reviewssummative traction consolidated independent an wereby followed and out carried Data selected. were population, defined ex geographically data a independent in Two children among tings, (coverage of a key child survival indicator, nutritional status, serious morbidity, or mortality) in LMIC set documents, reports and books assessing the impact of one or more CBPHC interventions on child health The database and its assembly have been described elsewhere in this series [ disqualify theproject from inclusion. not did interventions facility–based more or one of facility.presence health additional a The of outside community the in out carried interventions more or one be to considered was CBPHC brief, In ments. improve these document to and health) child as to referred (henceforth health child or neonatal prove and child health interventions by income quintiles to assess equity in coverage [ coverage in equity assess to quintiles income by interventions health child and maternal key of coverage population the regarding Countries Countdown MDG from data (MICS) vey Analyses have been conducted using Demographic and Health Survey (DHS) and Multiple Indicator Sur mortality [19]. many countries, inequities in child mortality were widening in spite of overall downward trends in child [ immunizations of provision community–based and (EBF), breastfeeding exclusive of promotion tion, ed interventions are those that can fairly easily be implemented within communities, such as ITN utiliza trends toward increased equity in coverage of key interventions. Some of the most equitably implement at the local level as well. In one long–standing comprehensive health program in Haiti serving 148 serving Haiti in program health comprehensive long–standing one In well. as level local the at While equity issues are often considered from a national or large–population perspective, they may exist among populationgroups [7 distributed evenly less be to tend thus and effective, be to order in structures support other and cesses, pro referral equipment, advanced more personnel, skilled of support and education including system and treatment of serious childhood illness [ groups for interventions that require facility–based, higher–level personnel such as skilled birth attendants been conducted. graphic location, and gender of the child’s caregiver and identifies dimensions where limited analysis has child’sgeo education, sex, maternal assets), household (or wealth consider,including to need grams pro health child that equity of dimensions various the reviews article this Second, framework. logical a around them organizes and health child on programs CBPHC of effects equity regardingthe evidence the time first the for consolidates it First, literature. equity the to contributions two makes article This health careandaccesstoCHWsequallythroughout services theprogram area [ nearer valley communities. This reality persisted despite great efforts being made to extend both primary is those comparedto communities mountain moreisolated the in differedliving those among markedly area program the of sub–groups of outcomes health the and interventions, key of coverage population the facilities, health of utilization the system, delivery service community–based strong a with people 7 , 10, 20, ]. At the same time, widening inequities were observed among different population sub– population different among observed were inequities widening time, same the At 21].
, 10]. 22]. These interventions often require a more developed health 390 11, 12, ]. Analysts observed that, within within that, observed Analysts 18]. 25, www.jogh.org 26], we identified a sub–set of 10]. In short, peer–reviewed 23]. • doi:10.7189/jogh.07.010905 4 , 5 ]. Results showed showed Results ]. 24]) to im
000 000 ------exclusion criteria. www.jogh.org Figure 1. Overview ofsequencearticlereviewOverview andinclusion/ • doi:10.7189/jogh.07.010905 characteristics tolookforandhoworganize thefindingsfrom the reports weanalyzed. R Box 1. disadvantaged sub–groups were benefitting less, equally, or more than other sub–groups. If disadvantaged when –namely outcomes equity important several between differentiate to us helped categories These populationswerederserved categorizedasinequitable. groups were categorized as equitable. Indicators with findings that showed unfavorable outcomes for un other the for as groups underserved for similar were that findings with Indicators significance. practical having as their results described authors study the out, carried were not significance statistical of tests if significant statistically were or, and populations underserved favored findings if pro–equitable as cators ed inouranalysis(pro–equitable, equitable,andinequitable,asdefinedin We created three categories of outcomes for the various equity indicators used by the assessments includ Categorization ofequityoutcomes port referred to these groups by the acronym PROGRESS ( re USAID The projects. MNCH equitable implementing in considered be should that groups vantaged [ examples for some of the less–common categories. In our literature review, we identified a USAID report provide and below categories the summarize categories. We identified the among commonalities tified responses as we reviewed each assessment, and then categorized them into common themes as we iden- In order to identify the diverse criteria utilized among the studies to analyze equity, we created open–text Criteria forequityanalysis ] on incorporating equity into project designs for MNCH that offeredthat disad projectMNCH identifying into for on equity designs guidance incorporating on 27] eligion, E that wasanalyzedacross adimensionofhealthequity. Equity indicator: An indicator of child health—such as rates of home visitation for for newborns, example — different outcomes. lens and determine whether different sub–groups of the population receive different equity levels an throughof groups services population or compare achieve to used be can that — areas rural or urban an in lives child a equity: of Dimension taged groups bytheendofproject implementation. when inputs, processes and outcomes for disadvantaged groupsdisadvantaged processesadvan for improveinputs, for outcomes when than and effect: less Inequity as advantagedgroups bythe endofproject implementation. degree same the to improvedgroups disadvantaged for outcomes and processes, inputs, when effect: Equity advantaged groups bytheend ofproject implementation. for than more improved groups disadvantaged for outcomes and processes, inputs, when effect: Pro–equity Definitions ducation, and A characteristic — such as household income, level of maternal education, or whether whether or education, maternal of level income, household as such — characteristic A S ocioeconomic S tatus) [28].Thistypologyprovided guidanceforthekindsof 391
the qualityofstudywasnotassessed. adequate information on equity analysis in each article, of availability fromthe Aside 1). (Figure set data final the in articles 42 leaving analyzed, was equity how on information sufficient provide not did it because ysis come, impact). One article was excluded from the anal into log–frame categories (input, process, output, out metrics from each project being assessed were stratified fects, and what the outcome on equity actually was. The ef equity of assessment for utilized weresources data what assessment, each in defined was equity how on each of these 43 projects and extracted additional data reviewed separately RK) and (MS authors Twothe of were leftwith43projects toexaminefurther. we phase, focusing this After subgroups. population ments in which equity was not actually analyzed across assess- excluded and assessments equity–relevant of sub–set review.this systematic reviewedWe carefully P lace of residence, June 2017 •Vol. 7No. 1•010905 Box 1).We categorizedindi- R ace, O CBPHC andequityeffects ccupation, G ender, ------
VIEWPOINTSPaPErS PaPErSVIEWPOINTS June 2017 •Vol. 7No. 1•010905 Schleiff etal. Table 1. reports containingequityanalyses Total Western Pacific Americas Southeast Asia Africa g eographIcal Geographical location of Geographical locationof
regIon Figure 2. impact. or outcome output, process, input, project a to referring was it whether to as classified further was tor other entity. Secondary data included those obtained from DHS and MICS data sets. Finally, each indica to data collected by the project, while the term secondary data refers to data which were gathered by an- The type of data used to assess equity was defined as primary or secondary. The term primary data refers nition ofequity. pre–determined defi a into fit have not might that indicators relevantequity any missing avoid to done ments were categorized after the list of equity indicators used in the projects had been reviewed. This was ment. The definition of equity was not pre–determined, and the definitions of equity used in the assess to assess equity, and the scope of the assessment as well as the types of indicators measured in the assess Each assessment was further analyzed to determine the criteria used to define equity, the type of data used ect anddrew conclusionsfrom theavailableevidence. ulated the equity effects of each project. For each cell of the matrix, we described the content of the proj We created a matrix for each phase of the logical framework and, for each of the included indicators, tab sentation oftheconceptualflowthislog–framematrixfrom onephasetoanother. perspective. implementation and projectplanning froma ties implementation equity dimensions were assessed. This made it possible to identify gaps and opportuni and processes, and then moving to outputs, outcomes, and impacts [ inputs with beginning projectimplementation: of standpoint fromthe analysis different of a framework utilizing by analysis our in included assessments the by used equity health of indicators organizethe to opted we work, health child influence can equity proachthroughthat think ways us various helped the ap this While illness. of consequences and outcomes the and disease, to succumbing to vulnerability son’s experience with an illness, beginning with the socioeconomic context through exposures to disease, Barros et al. [19] offer a framework for analysis of health equity from the standpoint of an individual per Organization ofidentifiedmetricsforhealthequityintoalogicalframework often needtomakeadditionalprogress inorder toovercome inequities. ting equally, this was noted as a good sign, though not a fully optimal outcome since disadvantaged groups groups were benefitting less, this was an inequitable outcome. When disadvantaged groups were benefit n umber Generalized logframeforhealth projects. 42 14 19
1 8 oF
studIes
country or a smaller sub–population within that country asshownin orasmallersub–populationwithinthatcountry country and another had data from four African countries. All other studies focused on one (Tableworld the fromdata countries, included African studies 28 the of One 1). The assessments included in our analysis were for projects from various regions of Location ofincludedprojects r ESULTS 392 below provides a graphic repregraphic providesa below 2 Figure 29] to track at what point in project www.jogh.org • doi:10.7189/jogh.07.010905 Table 1. ------*The columntotalis82sincemany oftheassessmentsinourreview included more thanoneequityindicator. www.jogh.org Table 2. Country–level HumanDevelopmentIndex(HDI) Country–level Maternal age Maternal Nutritional status Child’s sex Geographic locationofresidence (urban vsrural) Other: Parent occupation Social standing(ethnicity, caste,religion, parent maritalstatus) education Maternal Household assets(production, otherassetssuchassavings) Household incomecategories Socioeconomic status(SES): e quIty
crIterIon Equity indicatorsusedintheassessmentsincludedanalysis • doi:10.7189/jogh.07.010905 not identifiedinouranalysiswas religion [27]. ly availablenationaldata. and two utilized surveys, both project–level data collected for assessment of the project and also public within the project’s geographic area. However, five analyses exclusively utilized data from DHS and MICS project the by specifically collected analyses equity including projects 42 the of 37 in utilized data The Kinds ofdatausedintheassessments equity had been achieved in all cases and a pro–equity result is observed in almost all. The findings for for findings The all. almost in observed is resultpro–equity a and cases all in achieved been had equity analysis, for selected processassessments the the result.for in equitable Thus,indicators an yielded two remaining the and result,pro–equitable a yielded measurements the of 13 of out Eleven system. health the with contact had or worker health a from visit home a received had household the whether cerned Document tary as shown in A number of the assessments included in our review measured process indicators through an equity lens, Assessments ofequityprocesses noted asanarea where furtherworkisneeded. is and concerning is evaluation and implementation, design, project in efforts input–related of dearth The perspective. equity an from inputs reported or analyzed explicitly set data our in included we that projects no that determined we colleagues, and co–authors among discussion and analysis careful After Assessments ofequityinputs PROGRESS report (shown in the right–hand column of USAID the by identified those with well aligned indicators equity These age. maternal as such teristics the family, religion, marital status of child’s parents, occupation of the parents, and demographic charac of group ethnic the included criteria SES Other system. collection refuse a water,or running toilet, ing ricultural production by heads of household and specific assets present in the household such as a work sets, household size, or Other earnings. maternal SES equity indicators included in the analysis were ag indicator for assessing equity was a measure of wealth, often based on household income, household as included income categories, maternal education, and household characteristics. By far, the most common WeThese (SES). status socioeconomic as referto we category a under indicators groupedequity several taged groups were distinguishedfrom more advantagedgroups. the identical indicators for more advantaged groups. health program characteristics or health status over time for more disadvantaged groups with changes in Document Supplementary line home visits analyzed across household income categories (Callaghan–Koru, 2013; reference [S15] in Across the 42 projects included in our analysis, 82 equity indicators — for example coverage of prenatal Criteria through whichequity effectswere assessed Table 3 (references in Tables 3–6, are prefixed with an S and appear in ). Two–thirds (10/13) of the measurements of equity involving process indicators con indicators process involving equity of measurements Two–thirdsthe ). of (10/13) ) — were identified. Equity was measured by comparing changes in in changes comparing by measured was Equity identified. were — ) n umber
oF
393 assessments
In Table 2 summarizes the criteria by which disadvan-
whIch 24 45 1 2 4 3 1 8 9 5 Table 2); the only PROGRESS category that was
the
IndIcator
was
used June 2017 •Vol. 7No. 1•010905 c * omparable usaIdprogressI Online Supplemen- CBPHC andequityeffects Geography Ethnicity Gender Gender Wealth Wealth Wealth Wealth Wealth Age ndIcators On------
VIEWPOINTSPaPErS PaPErSVIEWPOINTS June 2017 •Vol. 7No. 1•010905 Schleiff etal. Table 4. Table 3. *References whichare prefixed withanS appear inAppendixS1ofthe *References whichare prefixed withanSappearinAppendixS1ofthe odhgeesoei eaint lalns cr Household income Food hygienescore inrelation tocleanlinessscore o At leastoneANCvisitinhome Child illandCHWcalledtocomethehome Home visitstosupportbreastfeeding Two ormore homevisitsduringpregnancy At leastonehomevisitduringpregnancy Antenatal homevisit Urbanvsrural Number ofhomevisits Prenatal homevisit Caregivers report ofCHWsworkingincommunity CHW visittocaregivers withinthepastyear Azythromycin distribution to entire communities for trachoma Home visitduringpregnancy Postnatal homevisit p odhgeesoei eaint irhapeaec aenleuain urtoa ttsIeutbeAhmed 1993[S1] Inequitable education,nutritional status Maternal Food hygienescore inrelation todiarrhea prevalence Utilization of ambulatory careUtilization ofambulatory facility Number ofhospitaladmissions Child withfevertreated within24h Essential newborn practicesperformed Essential newborn rocess utput
IndIcator
IndIcator Assessments ofequityeffects ofCBPHCprojects usingprocess indicators* Assessments ofequityeffects ofCBPHCprojects usingoutputindicators* ed inthehomewithoutcomplexorexpensivesupplies knowledge. equitable behaviors such as breastfeeding and newborn and child health practices that can be implement on focused commonly programs pro–equitable and Equitable not. others and settings rural and urban results across different studies, with some having equitable results across household income categories or natal and delivery care. Some indicators — such as ITN coverage, availability, and use — showed mixed ante and coverage vaccine as such knowledge or equipment significant requiring interventions several included indicators Inequitable result. pro–equitable a yielded (74%) rest the and result, equitable an yielded (4/35) 11% result; inequitable an yielded (5/35) 14% only out, carried measurement 35 the Of edge and behavior change related to breastfeeding or to the population coverage level of an intervention. knowl to relate Many indicators. outcome using out carried assessments equity Tablethe lists below 5 Assessments ofequityoutcomes several equitydimensions. across practices hygienic the to related and study same the from both wereeffect inequitable an strated demon that Indicators home). the in or facility a in (either services health to access on focus category output the in effectpro–equity a demonstrating indicators the but from, generalizations major make to uity assessments used household income as the equity criterion. The number of assessments is too small eq these of (3/6) Half intervention. an of output immediate expected the or services specific of lization uity assessments using output indicators among the projects the selected uti for our analysis concerned in listed are indicators output using equity of assessments The Assessments ofequityoutputs benefits toMNCH. multiple with approaches community–based through equitably supported be can that intervention an literatureas the in noted been also has breastfeeding,which of promotion support included and plicitly effectiveness of CBPHC in improving MNCH [ tral feature of many CBPHC projects, as also discussed in the in this supplement that directly address the cen a practices, visiting home of nature equitable the support consistently frame log the of portion this
e Urban vsrural Urban vsrural Household income Household income quIty e Household income Household income education Household income,maternal Household income Household income Household income Urban vsrural Household income Urban vsrural Household assets Household income Household income quIty
crIterIon
crIterIon online supplementary document. online supplementary online supplementary document . online supplementary 394
30– 32]. Many of these home visits either implicitly or ex . Two–thirdseq Table4. six the of (4/6) o r–qial Nonyane2015[S32] Pro–equitable Siekmans2013[S38] Pro–equitable Coutinho2005[S17] Pro–equitable Callaghan–Koru2013[S15] Pro–equitable Callaghan–Koru2013[S15] Pro–equitable Baqui2008[S8] Pro–equitable 2006[S35] Perry Pro–equitable Baqui2008[S8] Pro–equitable Litrell 2013[S25] Pro–equitable Callaghan–Koru2013[S15] Cumberland2008[S19] Pro–equitable Callaghan–Koru2013[S15] Equitable Equitable r–qial Litrell 2013[S25] Pro–equitable o nqial Ahmed1993[S1] Inequitable Pro–equitable Perry 2006[S35] Perry Pro–equitable Pro–equitable Perry 2006[S35] Perry Pro–equitable r–qial Siekmans2013[S38] Pro–equitable r–qial Baqui2008[S8] Pro–equitable utcome utcome www.jogh.org • doi:10.7189/jogh.07.010905 r eFerence r eFerence ------*References whichare prefixed withanSappearinAppendixS1ofthe www.jogh.org Table 5. Antenatal care coverage Antenatal care Attended delivery Children sleepingunderITNs Postnatal care coverage Immediate drying ITN coverage ITN coverage ITN coverage ITN inhome Immunization coverage Measles vaccinationrate Any bednetavailable Acute respiratoryinfectiontreatment rate within 48hours Householdincome Child with fever treated with artemether–lumefantrine Recognition ofatleast3dangersignsinnewborns ratedn ntainwti is oro ieHouseholdincome Knowledge anduseoffamilyplanning Urbanvsrural Knowledge offamilyplanningmethods Breastfeeding initiationwithinfirsthouroflife Breastfeeding initiationwithinfirsthouroflife Exclusive breastfeeding from birthto6m Householdincome Exclusive breastfeeding Awareness ofsupport group incommunity Child withdiarrhea treated withORSorzinc Householdincome Nothing applied to umbilical cord by mother after birth coverage Health service Coverage ofanytypebednet(ITNorother) ITN coverage EPI immunizationcoverage careCoverage ofantenatalanddelivery Understanding ofoverallcleanliness o utcome
IndIcator Assessments ofequityeffects ofCBPHCprojects usingoutcomeindicators* • doi:10.7189/jogh.07.010905 which projects constituted a single intervention vs a package of interventions. Of the 42 projects, 11 11 projects, 42 the Of interventions. of package a vs intervention single a constituted projects which [ series this in paper other (an paper this of focus the not was interventions of packages of impact the of analysis in–depth While (11/82) yieldedaninequitable outcome. 13% only and outcomes, equitable an yielded (7/82) 9% outcome; pro–equitable a yielded out carried in effects equity Finally,summarized have we Overall summaryofallequityeffects outcomes, andonly16%(7/44)yieldedaninequitableeffect. ( rion crite equity the was income household which in above reported findings the all Wesummarized have Overall summaryofequityeffectsusinghousehold wealthastheequitycriterion table results andoneyieldedanequitableresult. Onlyfourofthe28yieldedaninequitableresult. pro–equi demonstrated assessments 28 the of 23 Overall, status. nutritional of measure a on four and morbidity of measure a on based were four mortality; of measure a on based were 20 impact, health of dicators (nutritional status, morbidity or mortality). Of the 28 projects that included an equity assessment Finally, Assessments ofequityhealthimpact ). Overall, 75% (33/44) of these effects were pro–equitable outcome, 9% were equitable equitable were 9% outcome, pro–equitable were effects these of (33/44) 75% Overall, Table7). lists the assessments of health equity that were carried out for health impact–relatedin health Tablefor out werecarried that equity health of assessments the lists 6 33 Household income standing, householdassets Maternal education, household income, social standing, householdassets Maternal education, household income, social Household income standing, householdassets Maternal education, household income, social Household income Household income Household income Urban vsrural Household income Household income Household income Household income Household income Household income Maternal education,socialstanding Maternal Urban vsrural Household income Urban vsrural Household income Household income Child’s sex Household income Household income Household income Maternal education Maternal e ] addresses this strategy in general – not limited to equity), we reviewed we equity), to limited not – general in strategy this addresses ] quIty
crIterIon online supplementary document. online supplementary 395 . Overall, 78% (64/82) of the equity assessments assessments equity the of (64/82) 78% Overall, Table8. o r–qial Baqui2008[S8] Pro–equitable Pro–equitable Pro–equitable r–qial Noor2007[S33] Pro–equitable Pro–equitable Nonyane2015[S32] Pro–equitable Noor2007[S33] Pro–equitable Grabowsky2005[S23] Pro–equitable Grabowsky2005[S23] Pro–equitable Skarbinski2007[S39] Pro–equitable Bawah2006[S10] Pro–equitable Mercer 2004[S28] Pro–equitable Skarbinski2007 [S39] Pro–equitable Mercer 2004[S28] Pro–equitable r–qial Siekmans2013 [S38] Pro–equitable r–qial Nonyane2015 [S32] Pro–equitable Pro–equitable Debpuur2002 [S20] Pro–equitable Coutinho2005 [S17] Pro–equitable Callaghan–Koru2013[S15] Crookston 2000[S18] Pro–equitable Equitable Nonyane2015[S32] 2008[S14] Bryce Equitable Equitable Webster 2005[S42] Webster 2005[S42] Inequitable 2008[S14] Bryce Inequitable Inequitable r–qial Nonyane2015 [S32] Pro–equitable Crookston 2000[S17] Pro–equitable Littrell 2013[S25] Equitable Webster 2005[S42] Inequitable nqial Ahmed1993[S1] Inequitable utcome June 2017 •Vol. 7No. 1•010905 Awooner–Williams 2004 [S5] Awooner–Williams 2004 [S5] Awooner–Williams 2004 [S5] Awooner–Williams 2004 [S5] r eFerence CBPHC andequityeffects - - - -
VIEWPOINTSPaPErS PaPErSVIEWPOINTS June 2017 •Vol. 7No. 1•010905 *References whichare prefixed withanSappearinAppendixS1ofthe Table 6. Schleiff etal. Table 7. status orhouseholdwealthquintileastheequitycriterion Impact Total Input Outcome Output Process t Under–5 mortalityrate Under–5 mortalityrate Householdincome Under–5 mortalityrate Under–5 mortalityrate Under–5 mortalityrate Infant, 1–4years,andunder–5mortalityrates Infant mortalityrate Infant mortalityrate Infant mortalityrate Infant mortalityrate Infant mortalityrate Infant mortalityrate Infant mortalityrate Neonatal mortalityrate Neonatal mortalityrate Perinatal mortalityrate Perinatal mortalityrate Child nutritionstatus(qualitativedata) Urbanvsrural prevalence Undernutrition Nutritionalstatus Diarrhea prevalence inchildren 0–36months ofage Urbanvsrural Diarrhea prevalence inchildren 0–18months ofage Diarrhea prevalence inchildren 0–36months ofage Tetanus neonatorummortalityrate Child (age6–59months)mortalityrate Under–5 mortalityrate Under–5 mortalityrate Neonatal moralityrate I mpact ype
oF
IndIcator IndIcator Assessments ofequityCBPHCprojects usingimpactindicators* Summary of assessments of equity using socio–economic ofassessments ofequityusingsocio–economic Summary nqial qial r–qial Total Pro–equitable Equitable Inequitable rural comparison. ects with inequitable findings included a number of SES analyses and also child gender and an urban vs utilized across all of these categories followed closely by comparing urban vs rural populations. The proj categories. Household income as part of SES was by far the most common dimension of equity, and was these of each in projects of groups small the among patterns obvious uity.any identify not Wedid also (21%) included two dimensions of equity, and only six (14%) included three or more dimensions of eq nine while equity of dimension one only equity.for of analysis dimensions an more included (64%) 27 Of the 42 projects that conducted an equity analysis, we also reviewed which ones analyzed more one or equity outcomes(eg,allofthefindingsforAhmed1993 were inequitable). of terms in same the be to tended project particularly any within dimensions equity and interventions all general, in that, was pattern clear only the were; findings the equitable how and interventions of ber ects had a package of three or more services. We could not identify any clear between patterns the num (26%) included while a eight single (19%) intervention and included 23 2 (55%) interventions, of proj 2 7 0 4 1 0 e 0 4 0 2 0 2 FFect
on
equIty 344 33 824 18 8 6 0 0 3 9 2 7 e ra srrl oshl noePoeutbeAsha–India2008[S4] Pro–equitable Urban vsrural Household income Urban vsrural Urban vsrural,householdincome Household income Bang2005[S7] Pro–equitable Household income Bang1999[S6] Human developmentindex Pro–equitable education Household assets,maternal Household income Social standing,parental occupation Urban vsrural education,child’sMaternal sex Urban vsrural Urban vsrural Urban vsrural Urban vsrural Nutritional status Nutritional status Urban vsrural Social standing,child’s sex Household income Urban vsrural Household income quIty
crIterIon online supplementary document. online supplementary 396 Table 8. t Total Impact Outcome Output Process Input ype
oF
IndIcator Summary ofallassessments ofequity Summary nqial qial r–qial Total Pro–equitable Equitable Inequitable 11 4 5 2 0 0 o r–qial Asha–India2008[S4] 2008[S14] Bryce Pro–equitable 2006[S35] Perry Pro–equitable Pro–equitable Sepulveda2006[S37] Pro–equitable Mercer 2004[S28] Aquino2009[S2] Pro–equitable Pro–equitable Bhuiya2002[S11] Pro–equitable Asha–India2008[S4] Fegan2007[S21] Pro–equitable Bang1999[S6] Pro–equitable ASHA–India2008[S4] Pro–equitable Bang1999[S6] Pro–equitable Bang2005[S7] Pro–equitable McNelly1998[S29] Pro–equitable Mustaphi2005[S30] Pro–equitable Pro–equitable Newell1966[S31] Bishai2005[S12] Equitable Razzaque2007[S36] Inequitable 2008[S14] Bryce Inequitable Razzaque2007[S36] Inequitable Inequitable r–qial Mercer 2004[S28] Pro–equitable Barreto 2007[S9] Ahmed1993[S1] Pro–equitable Barreto 2007[S9] Pro–equitable Pro–equitable utcome www.jogh.org e 7 1 4 0 2 0 FFect • doi:10.7189/jogh.07.010905
on
r equIty eFerence 482 64 328 35 23 13 26 11 6 0 4 0 - - - - www.jogh.org • doi:10.7189/jogh.07.010905 been necessary inorderbeen necessary toimplementthescaleanddepthofthisprogram atthenationallevel. has will political of level high A status). nutritional childhood of inequities (including particular in dren chil for and population general the for inequities health reduce effortsto its for recognition global Brazil Community Health Agents) who visited every home on a monthly basis. This national program has brought professionals,(called health CHWs oral includes as well as nurses and physicians to addition in workers, promotion of breastfeeding, prenatal care, immunizations, and management of diarrhea. The team of health program used a family–centered approach to provide a range of services at the community level, including was highest and the human development index was lowest at the beginning of the study period. The FHP FHP program was greatest in terms of decreasing infant mortality in municipalities where infant mortality ing Brazil’s Family Health Program (FHP) coverage on infant mortality. They identified that the effect of the Aquino et al., 2008 (reference [S12] in proaches withpoliticalwillandinvestment,anationalstrategy, andalong-termcommitment). ap community–based combining of effects pro–equity potential the of example an as serves database) A case example from Brazil of equity effects of CBPHC on improving child health (an article selected from required toreduce inequitiesinchildhealth. be will CBPHC of expansion strong that indicate findings our Therefore, care. facility–based accessing [ geographic major and countries low–income in constrained highly be to continue resourceswill term, near the For status. health child in or services health child of delivery the in inequities reduces strategy isolated an as services facility–based proving long term, but there is no evidence at present that we are aware of demonstrating that expanding or im very the in possible be may This health. child in inequities reduce eventually will care facility–based in improvements and facilities of number the of expansion that is assertion this to counter–argument The health facilityutilizationinlow–incomesettingsishighlyequitable,asexplainedfurtherbelow. facilities alone would be highly unlikely to reduce existing inequities since, in fact, it is well–known that health where settings low–income in health child in inequities reduce to potential the has CBPHC that child health services and in child health outcomes. Thus, these findings are consistent with the assertion of delivery the in inequities reduce to CPBHC of capacity the provide of strongfindings evidence These interventions. more or two across equity measured (74%) articles 31 and dimensions equity more or disadvantaged group benefitted more). Of the 42 articles in our review, 15 of them (36%) measured two moregroup degreegroup) the tage same advantaged as the to pro–equitablebenefitted or the which (in 87% of these measurements indicated that the equity effect was either equitable (in which the disadvan projects, 42 these in effects equity of measurements 82 the Of effects. equity measured 42 set, data our equity effects of CBPHC in improving child health. Out of the 546 assessments related to child health in We have carried out an equity analysis of the projects in our review that contained evidence regarding the DISCUSSION usage ofhealthfacilitiesinterms oftheSESandurban/ruralcharacteristicsusers[ growing literature, including but also going beyond the database used in this study, points to inequitable A well. as significant facilities—is of outside communities in deploy to difficult skills and frastructure requirein or – facilities health in available only often are that providinginterventions of challenge The vestment contributestolowutilizationoffacility–basedservices. family’sthat on in- return uncertain the and facility health a reach to expend to have families their and [ facility health a at care seek to likely less or more is someone roleswhether key residencedetermining play rural in and urban and income, level, education [ disrespectful as sub–groups certain by seen is that care of provision including care, poor,of the quality of high perspective fromlacking the expensive and often between, far and few for high–income settings in the Americas, Europe, and Asia. In LMIC settings, health facilities tend to be Research on the equity of facility utilization in low–income settings is limited; more evidence is available improvements onequitableutilizationofhealth facilities[ place—including expansion of community health centers—these efforts have resulted in only very small [ services are generally distributed more evenly throughout target populations than facility–based services how to access outreach services (if not to actually provide services including curative care), and outreach Most CBPHC projects are designed to reach every household with health education and information about Explaining thepro–equity effectsofcommunity–basedprimaryhealthcare ]. Meanwhile, some countries, such as Peru, where great investment in health facilities has taken taken has facilities health in investment great where Peru, as such countries, some Meanwhile, 35]. Online Supplementary Document 397 ], social and financial barriers will continue to exist in in exist to continue will barriers financial and social 34],
36, ]. Thus, the effort and resources that patients resourceseffortpatients and the that Thus, 37]. 36]. June 2017 •Vol. 7No. 1•010905 ) analyzed the effects of expand 36– CBPHC andequityeffects ]. Factors such as as such Factors 38]. 23, 34, 39]. ------
VIEWPOINTSPaPErS PaPErSVIEWPOINTS June 2017 •Vol. 7No. 1•010905 Schleiff etal. dressingVictoraAs them. [ note al. et ad- and understanding for points provideentry and additional equity health of issues frame to help can and empowerment frameworks, such as the CHOICE ( development community comprehensive approaches, more and tools health–specific public our to tion progressof kind this make to remainsdone addi work be In much to realitypopulations. tries, a all for uitable because those in the lower income quintiles are less likely to obtain services there [ care ineq health is facilities primary of utilization that finding observed commonly the to contrast stark in stand findings These surprising. not are findings these so interventions, key of coverage population expand rapidly can and facilities health from furthest those reach can approaches Community–based disadvantaged sub–populationswhere themortality ratesare thegreatest. most the among mortality child in decreases substantial achieve to necessary are health child for grams proin inequities of causes the to attention and design deliberate and will political enough; not is uities sustainability,al uity effects ofCBPHC inimproving childhealth. eq the summarizing literature peer–reviewed the in review comprehensive first the is this knowledge, access to health services without financial barriers [ barriers financial without services health to access as well as opportunities educational equitable provide to Japan’spolicies is national equity health of na to direct those resources in ways that decrease inequities. An example of progress and success in the are countries as well. Even where more resources are available to address such issues, political will is needed While the focus of this review is on low–income countries, inequities are also prevalent in higher–income isessentialforachievinghealthequity.and sustainabilityofthoseservices tralization of services and utilization of innovative and proven strategies to support the coverage, quality, areas.decen- rural the low–density Thus, in particularly – them use to need could reachwho easy all of within facilities operate and build to capacity the high–mortality,lack in settings resource–constrained contexts [47] cannot be overlooked even within such a strong outreach approach [ [ child’sethnicity education, sex, including factors equity–relevant other of ber inherently more likely to achieve favorable equity effects than facility–based approaches. However, a num are – homes to close and projectpopulation throughoutthe distributed evenly relativelyare that points Approaches that make it possible for health workers to reach all households – or at least to reach outreach fees andengagingtheprivatesector[3 user removing or minimizing as such options market–driven with along workers, lay other and CHWs using outreach, community strengthening on focus equity health improve to strategies promising most nity–based approaches that often include direct contact with all households [ throughequity commu of levels high achieve to able be would ITNs of promotiondistribution ing and of the better off segment. It makes sense that home–centered, low–resource interventions like breastfeed to other approaches improve the health of the poorest segment of the project population compared to hat There are several assessments that directly compare the degree to which CBPHC approaches as opposed the articles based on how strong the equity effect is. Second, some of the 42 assessments qualifying for for qualifying assessments 42 the of some Second, is. effect equity the strong how on based articles the This study has several limitations that we want to make explicit. First, we have not further disaggregated Limitations ofourstudy from facilities,whoare alsousuallymore disadvantagedintermsofSES[ key primary health care services, particularly for vulnerable populations and those who live further away of number a of coverage of terms in approaches facility–based solely than equitable more be can holds [ care seeking to riers care,to reduce resource antenatal bar as such services, specific for level community providedthe ers at Strong community–based programs can encourage facility utilization across income strata as can vouch those furtherawayfrom thosefacilities. to available careis community–based whether also account into taking status, health on facilities health from distance of effect the and facility health a from distance of terms in utilization care health of uity cilities, which is often absent [33, [ away) walk hour 1 or km more5 (particularly than facility health the increasingto with distance cilities fa health of utilization the in decline exponential an is there (i) following: the are health child for erage The need for alternative approaches beyond health facilities to achieve equity in and in fact universal cov 35], and (ii) there is a need for available and affordable public transportation in order to reach health fa
I nstitutional accountability,nstitutional ]. The available evidence suggests that CBPHC approaches that reach all house all reachapproaches that CBPHC that suggests evidence available The 40]. 39]. What is lacking from the literature are in–depth assessments of eq- ], just using the data available and recognizing patterns in ineq in recognizingand available data the using just 51], patterns , 44, 45]. 398 C ontribution, and ontribution, ]. Globally,coun 49]. low–income in particularly but C apacity–building, E nabling environment) nabling [50], framework www.jogh.org 20, H 41– 7 uman rights, ]. In addition, some of the • doi:10.7189/jogh.07.010905 46 43]. 48]. Health programs ], and urban vs rural rural vs urban and ], 52, O rganization- 53]. To our ------www.jogh.org rEFEr ENCES • doi:10.7189/jogh.07.010905 6 5 4 3 2 1 agenda for women and children. Lancet. 2015;385:466-76. Requejo JH, Bryce J, Barros AJ, Berman P, Bhutta Z, Chopra Lancet. M, et al. countries. Countdown to 54 2015 and beyond: from fulfilling the data 2012;379:1225-33. Medline:22464386 health survey of review retrospective a 2015: to Countdown in interventions health child and newborn, maternal, in GV,Equity Franca al. AD, et Bertoldi E, Loaiza H, Axelson C, Ronsmans AJ, Barros port. Baltimore, MD:JohnsHopkinsUniversity, 2015. Requejo J, Victora J. A Decade C, of Bryce Tracking Progress The 2015 and Re Child Survival: Newborn, for Maternal, Health. 2016;16:419.Medline:27207151 review. Public systematic BMC A services? workers health community of equity influence features programme which and programmes worker health community are equitable W,How TaegtmeyerGomez M. S, R, Theobald McCollum pmed.1001390 nal, newborn, and child health interventions. PLoS Med. 2013;10:e1001390. mater of coverage in inequalities interpreting and determining MNCH: in Coverage Measuring VictoraCG. BarrosAJ, com/amanda-moore/martin-luther-king-health-care_b_2506393.html. Accessed:20March 2017. TrackingA. Jr.’sMoore King, Luther Martin Down Words Available:2013. Care. Health on populations around theworld. of community–based approaches to continue to help bolster health equity for children in disadvantaged We have worked to be clear in our language, conservative in our claims, and yet optimistic about the role ly tobemore recent articlespublishedsincethattimeare relevant tothisanalysis. covers articles published over the past six decades through the end of 2015, we know that there are like review was beyond the scope of this article. Finally, although a thorough search has been conducted that our in included assessments 42 the in included data the of quality the of analysis An CBPHC. through everyday practice settings. But, that said, it still remains true that strong pro–equity effects are achievable to findings these generalizing about careful be must we Therefore, implementation. project for present were conditions ideal which in settings community within conducted studies efficacy are analysis our sessments needtobecomeastandard feature ofMNCHprogramming. reducing inequities in national programs of countries where the risk of child mortality is high. Equity as for important are workers community–level of utilization the and management provisionand service of to achieve pro–equity effects than projects that strengthen services only at facilities. The decentralization and are readily accessible throughout the project population, CBPHC projects are inherently more likely Based on the finding that the services provided by CBPHC projects generally reach most or all households CONCLUSIONS closure.pdf (availableuponrequest from thecorresponding author),anddeclare noconflictofinterest. Conflict of interest: the authorsparticipatedinrevision ofearlierdraftsandapproved thefinaldraft. of All data. the of analysis primary the conducted MS and wrote RK MS draft. declaration: first Authorship the ecution ofthereview. ex the in role no had support financial provided that organizations The Foundation. Gates the and velopment, WorldThe ticle: UNICEF,Organization, Health De International for Agency WorldStates the United the Bank, ar this in described work the conduct to used were that funds providedorganizations following The Funding: the StudyTeam. view. The World Bank made it possible for one of its consultants, Dr Bahie Rassekh, to participate as a member of re the of phase initial the during Perry Dr to support salary and support, administrative space, office providing Westaff,funds. Section these Health administeredof which some International its Futurefor thank Generations tions, and the Gates Foundation. We are also grateful to the American Public Health Association and particularly Genera Future Health)/USAID, Child for Resources and (Collaboration Group CORE the Organization, Health view: UNICEF, theWorld Bank,theDepartmentofChildandAdolescentHealthDevelopmentWorld re this of expenses the cover to grants Weprovidedsmall organizationsscript. that following the to grateful are manu this of draft earlier an of review her for Luna Winestock Jennifer to Wegrateful Acknowledgments: are All authors have completed the Unified Competing Interest Form at www.icmje.org/coi_dis- doi:10.1016/S0140-6736(12)60113-5 doi:10.1186/s12889-016-3043-8 399 Medline:24990815 Medline:23667332 doi:10.1016/S0140-6736(14)60925-9 June 2017 •Vol. 7No. 1•010905 http://www.huffingtonpost. CBPHC andequityeffects doi:10.1371/journal. ------
VIEWPOINTSPaPErS PaPErSVIEWPOINTS June 2017 •Vol. 7No. 1•010905 Schleiff etal. rEFEr ENCES 8 7 31 30 29 28 27 9 26 23 22 21 25 24 20 19 18 17 16 15 14 13 12 11 10 ress for maternal, newborn, and child survival. Lancet. 2016;387:2049-59. Lancet. survival. child and newborn, maternal, for ress Victora CG, Requejo JH, Barros AJ, Berman P, Bhutta Z, Boerma T, et al. Countdown to 2015: a decade of tracking prog line:22464386 2012;379:1225-33. Lancet. Countries. 54 from Data Survey of Review Retrospective A 2015: to Countdown in Barros AJ, Ronsmans C, Axelson H, Loaiza E, Bertoldi A. Equity in Maternal, Newborn, and Child Health Interventions tion. 2012.Available: http://www.wfpha.org/tl_files/doc/about/Addis_Declaration.pdf. Accessed:20March 2017. ToCall A Equity: Ac Health Global on Declaration Ababa Addis The Associations. Health WorldPublic of Federation 6736(15)00519-X health findings.JGlobHealth. 2017;7:010903. neonatal 3. health: child and neonatal, improvingCarein maternal, Health Primary effectivenessCommunity-based of FreemanE, Sacks P, comprehensiveA regardingH. reviewevidence the of Perry S, Gupta B, Rassekh the MC, Jennings nal healthFindings.JGlobHealth. 2017;7:010902. the effectiveness of Community-based Primary Health Care in improving maternal, neonatal, and child health: 2.Mater regarding evidence the P,of Freemanreview comprehensiveE, A Sacks S. M, SchleiffGupta S, Pradhan MC, Jennings framework. IntHealth.2011;3:139-46.Medline:24038362 Bryce J, Victora CG, Boerma T, Peters DH, Black RE. Evaluating the scale-up for maternal and child survival: a common heapol/czm028 Gwatkin DR. 10 Best resources on...health equity. Health Policy Plan. 2007;22:348-51. Neonatal,andChildHealthProjects.munity-Oriented Maternal, Washington, DC:USAID,2011. Com for Guide A Designs: Project into Equity Health Incorporating for Considerations C. Victora J, Winestock-Luna journal.pmed.1001333 lines for systematic reviews with a focus on health equity. PLoS Med. 2012;9:e1001333. WelchV, TugwellM, Petticrew P, guide reporting extension: Waters2012 J, PRISMA-Equity O’Neill al. D, et Moher E, 2007;6:7. Schweitzer-Haiti.Health. Albert Equity Hôpital J the Int of area service the in outcomes socio-geography,level: local the at issues equity health Health and J. access, Bryant A, Aftab L, King-Schultz HB, Perry Global ActionPlanforPneumoniaandDiarrhea (GAPPD).Geneva:World HealthOrganization; 2013. Integrated The 2025: by Diarrhea and Pneumonia from Deaths WorldChild Preventable Ending Organization. Health 2014;23:377-84. Medline:25164447 Afsana K, Haque MR, Sobhan S, Shahin SA. BRAC’s Experience in scaling-up MNP in Bangladesh. Asia Pac J Clin Nutr. doi:10.1186/1741-7015-11-261 in systematicreviewsCochraneDatabaseSyst Rev. ofinterventions. 2010;MR000028. Welch V, Tugwell P, Petticrew M, de Montigny J, Ueffing E, Kristjansson B, et al.database description.JGlobHealth.2017;7:010901. How effects on health equity are assessed and methods rationale, 1. health: child and neonatal maternal, improving in care health primary community-based of Perry H, Rassekh B, Gupta S, Wilhelm J, Freeman P. A comprehensive review of the evidence regarding the effectiveness doi:10.1186/1475-9276-6-7 2013;11:261. Med. BMC survival. child in equity to steps Simple K. Shibuya S, Gilmour 2010. E, Kurup AS, editors. Equity, Social Determinants, and Public Health Programmes. Geneva: World Health Organization; Barros F, Victora C, Scherpbier R, Gwatkin D. Health and Nutrition of Children: Equity and Social Determinants. In: Blas Medline:27084546 epidemiology.2016;26:238-40. for Epidemiol. Ann concern proper a health: in inequalities Social R. Bell M, Marmot Action. Health 2013;6:19668. doi:10.3402/gha.v6i0.19668 Glob Programme. INTREC the countries: African three in health of determinants social the and Hofman K, Blomstedt Y, Addei S, Kalage R, Maredza M, Sankoh O, et al. Addressing research capacity for health equity of Health.Geneva:World HealthOrganization; 2008. World Health Organization. Closing the Gap in a Generation: Health Equity Through Action on the Social Determinants miol. 1992;21:1043-9.Medline:1483808 Taylor CE. Surveillance for equity in primary health care: policy implications from international experience. Int J Epide 94. Medline:16533114 2006;27:167- Health. Public Rev Annu measurement. and concepts P.equity: Braveman health and disparities Health nal_report.pdf. Accessed:20March 2017. 2015 Report. 2015. Available: The Survival: Child and Newborn Maternal, Trackingfor UNICEF,Progressof Decade A Organization.World Health Lancet. countries. 54 from 2012;379:1225-33. Medline:22464386 data survey of review retrospective a 2015: to Countdown in interventions health child and newborn, maternal, in GV,Equity França al. AD, et Bertoldi E, Loaiza H, Axelson C, Ronsmans AJ, Barros Lancet. surveys. 2012;380:1149-56. Medline:22999433 national of analysis an countries: 2015 to Countdown 35 in interventions health child and nal affectcoverage mater in changes equity How al. et G, Franca M, VictoraChopra J, BarrosBhutta CG, H, Axelson A, Lancet. Countries. Countdown 54 in 2008;371:1259-67. Medline:18406860 services health child and newborn, maternal, of coverage in trends and equity Gap: The Mind CG. Victora Y,H, Kinfu Axelson JT,J, Boerma Group, Bryce Analysis Equity 2008 Countdown doi:10.1016/S0140-6736(12)60113-5 doi:10.1016/j.annepidem.2016.02.003
doi:10.1146/annurev.publhealth.27.021405.102103 http://www.countdown2015mnch.org/documents/2015Report/Countdown_to_2015_fi doi:10.1016/S0140-6736(12)60113-5 doi:10.1016/S0140-6736(12)61427-5 doi:10.1016/S0140-6736(08)60560-7 doi:10.1093/ije/21.6.1043
400 doi:10.1016/j.inhe.2011.04.003 Medline:26477328 www.jogh.org Medline:17698890 Medline:21154402 Medline:23222917 • doi:10.7189/jogh.07.010905
Medline:24344755 doi:10.1016/S0140- Medline:17678540
doi:10.1093/ doi:10.1371/ Med------
www.jogh.org rEFEr ENCES • doi:10.7189/jogh.07.010905 38 37 36 40 39 35 34 33 32 53 52 51 50 49 48 47 46 45 44 43 42 41 services utilizationinCuraçao.SocSciMed.1997;45:213-20. services of study a care: health in inequity Socioeconomic Wvd. JF,Heuler Alberts WJ. Heuvel den van JM, Eimers R, Sanderman facilities globally: a mixed-methods systematic review. PLoS Med. 2015;12:e1001847. Bohren MA, Vogel J, Hunter E, Lutsiv O, Makh S, Souza JP, et al. The mistreatment of women during childbirth in health Policy Plan.2002;17Suppl:12-9.Medline:12477737 Health Peru. in services care health outpatient of utilization the in equity and infrastructure health ValdiviaPublic M. Bangladesh. SocSciMed.2011;72:1704-10.Medline:21546145 in scheme voucher health maternal a from Lessons enhancing? equity financing demand-side Is MM. Khan S, Ahmed 95. Medline:26861932 Womens2016;61:185- Midwifery Health. J Afghanistan. in determinants attendance birth skills of analysis multilevel a from findngs childbirth: for facilities care primary Bypassing D. Peters WarrenN, S, Doocy M, Koblinsky TappisH, journal.pmed.1001847 Medline:23509239 2013;42:520-32. Epidemiol. J Int cluster-randomisedtrial. a of analysis secondary India: in mortality neonatal reduce TA,Houweling Tripathy P, women’s participatory of impact equity The al. et R, Gope S, Rath S, groupsRath N, to Nair doi:10.1016/j.socscimed.2006.01.015 South Africa: an exploration using a Geographical Information System. Soc Sci Med. 2006;63:691-705. Tanser F, Gijsbertsen B, Herbst K. Modelling and understanding primary health care accessibility and utilization in rural J GlobHealth.2017;7:010906. nity-based primary health care in improving maternal, neonatal and child health: 6. strategies used by effective projects. Perry H, Rassekh B, Gupta S, Freeman P. A comprehensive review of the evidence regarding the effectiveness of commu findings. JGlobHealth.2017;7:010904. fectiveness of Community-based Primary Health Care in improving maternal, neonatal, and child health: 4. child health Freeman P, Schleiff M, Sacks E, Rassekh H. B, A Gupta comprehensiveS, Perry review of the evidence regarding the ef zation incidenceanalysis.IntJEquity Health.2016;15:120. utili a Pakistan: in care health child newborn and mother for services public of use the in Equity N. Ashraf SM, Malik line:20653970 2010;10:217. Res. Serv Survey.WorldHealth Health the BMC from evidence need: care health to according Brazil in services outpatient of Szwarcwalduse the in inequalities Socioeconomic GN. Damacena PR, Souza-Junior CL, 6736(03)13917-7 2003;362:233-41. Lancet. enough. not is same the of more mortality: and health child to J-P.lens Habicht equity M, an Claeson Applying D, Gwatkin JA, WagstaffSchellenberg CG, A, Victora Nutr. 2003;21:168-80.Medline:14717563 Rifkin SB. A framework linking community empowerment and health equity: it is a matter of CHOICE. J Health Popul 2011;378:1094-105. Medline:21885105 Lancet. Healthy? Japan of population T,the Satoh made S, has Ikeda What M, al. Inoue et N, Kondo E, Saito N, Ikeda ington DC;World BankGroup: 2016. Washed. 3rd Health. Child and Newborn, Maternal, Reproductive, editors. N, Temmermanwalker R, M, minarayan Lassi Z, Kumar R, Bhutta Z. Community-based Care to Improve Maternal, Newborn, and Child Health. In: Black R, Lax DF.action?ressource.ressourceId=51297. Accessed:26April2017. for 174 countries. Extension of Social Security. 2015. Available: Scheil-Adlung X.Globalevidenceoninequitiesinruralhealthprotection: newdataonruraldeficitsinhealthcoverage Medline:26075988 to reduce health disparities among racial and ethnic minority populations. Cochrane Database Syst Rev. 2015;6:CD009905. Anderson LM, Adeney K, Shinn C, Safranek S, Buckner-Brown J, Krause LK. Community coalition-driven interventions Equity. Washington, DC:USAID/MCHIP, 2014. Winestock Luna J, Monga T, Morgan L. Equity matters–lessons from MCHIP and CSHGP in Measuring and Improving 6736(12)61423-8 2012;380:1331-40. Lancet. nutrition. and health, survival, child in gap equity the narrow and coverage health improve to Strategies N. Birkin D, Anthony N, Dalmiya A, Sharkey M, Chopra dle/10986/6896. Accessed:26April2017. Available: 2008. Introduction. Equity: Health Analyzing WorldBank. doi:10.1016/j.ijgo.2011.10.031 havioral intervention on maternal Health in Shivarg, India. Int J Gynaecol Obstet. 2012;117:48-55. be newborn-focused a of impact Community-driven G. V,Darmstadt V,Kumar A, Das Baqui A, N, Kumar Srivastava 2008;23:234-43. Medline:18562458 Plan. equity.Policy in improvementsHealth India: rural in programme health neonatal and maternal munity-based com government a of facilitation NGO G. Darmstadt S, Ahmed PK, Agarwal EK, Williams AM, Rosecrans AH, Baqui doi:10.1186/1472-6963-10-217 doi:10.1093/ije/dyt012 doi:10.1111/jmwh.12359 doi:10.1093/heapol/czn012 doi:10.1016/S0140-6736(11)61055-6 401 doi:10.1093/heapol/17.suppl_1.12
Medline:27459961 Medline:9225409 http://www.social-protection.org/gimi/gess/RessourceP- doi:10.1016/j.socscimed.2011.03.031 https://openknowledge.worldbank.org/han Medline:22999430 Medline:12885488 doi:10.1016/S0277-9536(96)00338-3 doi:10.1186/s12939-016-0405-x June 2017 •Vol. 7No. 1•010905 Medline:26126110 CBPHC andequityeffects
doi:10.1016/S0140- doi:10.1016/S0140- Medline:22281244 Medline:16574290 doi:10.1371/ Med------
VIEWPOINTSPaPErS PaPErSVIEWPOINTS June 2017 •Vol. 7No. 1•010906 Meike Schleiff Paul aFreeman Bahie Mrassekh Kumapley child health:6.strategiesusedbyeffectiveprojects health careinimprovingmaternal,neonataland the effectivenessofcommunity–basedprimary Comprehensive reviewoftheevidenceregarding [email protected] USa Baltimore, MD21205 615 NorthWolfeSt. Health Johns HopkinsBloombergSchoolofPublic r Henry Perry Correspondence to: 6 5 4 3 2 1 Henry BPerry Washington, Seattle,USa Department ofGlobalHealth,University USa Independent consultant,Seattle,Washington, Columbia, USa The WorldBank,Washington,Districtof Medical epidemiologist,Lusaka,Zambia UNICEF, NewYork,USa Baltimore, Maryland,USa Hopkins BloombergSchoolofPublicHealth, Department ofInternationalHealth,Johns oom E8537
2 , SundeepGupta 1 1
, EmmaSacks , richard
5,6 4 ,
3 , 1 , there have been calls for more direct funding for integrated maternal and and maternal integrated for funding directmore for calls been have there mothers, and children of deaths preventable readily of reduction the in panding immunization coverage. However, in order to accelerate progress ex HIV,and of tuberculosis, control and as malaria such programs, tical on technical cooperation pertaining to strengthening narrowly focused ver In recent decades, much of the funding for global health has concentrated WeMethods 12 screened this paperexplores thespecificstrategiesused. address specifically the effects of CBPHC on improving MNCH, while series this in mortality.articles child improvingOther in fectiveness studies (collectively referred to as projects) that have demonstrated ef research field and programs projects, of strategies delivery common the here summarize we (MNCH), health child and neonatal ternal, of community–based primary health care (CBPHC) in improving ma Background As part of our review of the evidence of the effectiveness for improving MNCH. points for program design in strengthening the effectiveness of CBPHC starting useful provide here identified strategies The Conclusions reach sitesbymobilehealthteams. participatory women’s groups, and (d) health service provision at out agents, (b) routine systematic visitation of all homes, (c) facilitator–led community trained and/or mothers by illness childhood serious of treatment possible) (when and referral, recognition, (a) identified: were delivery intervention for strategies delivery.specific tion Four strengthening, (e) use of community–level workers, and (f) interven systems health (d) members, community and beneficiaries teers, munity collaboration, (c) education for community–level staff, volun nities and health systems: (a) program design and evaluation, (b) com identified, all of which required working in partnership with commu Results Six categories of strategies for program implementation were in project implementation. used strategies key categorize and identify we Here form. extraction data a using reviewed were and inclusion for criteria the met books) publicly available sources (mostly unpublished evaluation reports and other from reports 148 including assessments, 700 of total A 2015. 1950– high–mortality,from in settings resource–constrainedhealth munity–based approaches to improving maternal, neonatal and child 402
166 published reports in PubMed of com of PubMed in reports published 166 www.jogh.org • doi:10.7189/jogh.07.010906 global journal of health ------www.jogh.org • doi:10.7189/jogh.07.010906 database. (hereaftercomprehensive referreda projects)in as included to effectivenessand whose assessed been has studies and projects programs, the by used approaches various the summarizes paper This used. have [ [ programs health child sure changes in intervention coverage in the program population as well as changes in childhood nutri childhood in changes as well programas the population in coverage intervention in surechanges edge, practice and coverage (KPC) population–based household surveys. These made it possible to mea Strategies for project design and evaluation shown in Strategies forprogram designandevaluation women’sparticipatory groups). (eg, categories several into fit strategies the of some fact, in and, strategies these of several least at used ery. deliv intervention (f) and CHWs), or workers, health community as referredto (hereafter workers and volunteers community–level of use strengthening,(e) systems health (d) members, community and ries evaluation, (b) community collaboration, (c) education for community–level staff, volunteers, beneficia and design program (a) database: our in projects the by used strategies of categories six Weidentified r marizing commonthemesaswellbyaddingupthenumberofresponses toclose–endedquestions. egies for project implementation was reviewed by reviewing all the open–responses individually and sum described the strategies that were used by projects. All available evidence in the database regarding strat that information all identify to searchedcarefully was database health child and neonatal maternal, The a third reviewer resolved anydifferences betweenthefirsttwo reviews. and community engagement. Data were extracted from each assessment by two independent reviews and tioned paper [ above–men the of Document Supplementary Online in contained is form extraction data the of copy A scriptions ofproject implementationwere completedbyreviewers. available information regarding strategies used for project implementation. In particular, open–ended de The data extraction forms used to collect information from the assessments were designed to capture the delivery,project. intervention the to evaluate related directlyand not activities associated in engage ect, projthe implement community), project,the effectivethe (including vention planpartners to engage – inter the make to used projects these that activities the mean we strategies By reviewers. the by or ects term “community” in the analysis of the findings since there was no uniform definition used in the proj the of definition strict any force to attempt Wenot played. did community the role what and strategies project the describes project’sthat the assessment containing document the in possible as information much as capture to designed was form extraction data a that is paper this for importance particular Of description ofthesearch strategyandcreation ofthedatabaseisavailableelsewhere [ and licly books. available A on total the of internet) 700 assessments were included in this review. A full pub- (documents literature “grey” the from identified were reports 148 additional An qualified. reports of community–based programs for improving MNCH in low– and middle–income countries. 552 of these PHC) in improving maternal, newborn and child health (MNCH) by reviewing 12 We conducted a comprehensive review of the effectiveness of community–based primary health care (CB METHODS [ health child and born new arematernal, that for practices Previousimportant community and reviewsfamily highlighted have port systemsthatare neededtoachievehighlevelsofpopulationcoverage. ratherthanonthestrategiesandsup tention hasbeenfocusedonthetechnicalaspectsofinterventions accelerating progress by achieving high levels of coverage of evidence–based interventions. Too often, at for strategies delivery service community–based strengthening of importance the emphasized ficiently [ care of tinuum 7 ESULTS – ], but none have to date focused specifically on the implementation strategies that effective that strategies implementation the on specifically focused date to projectshave none but 10], summarizes these strategies. The strategies were not mutually exclusive and most projects most and exclusive mutually not were strategies The strategies. these summarizes Table1 11]. The form allows for open–ended as well as close–ended responses related to strategies 3 , 4 ], and stronger community participation [ participation community stronger and ], 6 1 ] as well as specific technical interventions that can be provided in communities communities provided in be can that interventions technical specific as well as ] ], health systems strengthening [ strengthening systems health ], 403 Table 1 often included baseline and endline knowl 2 ], integration of key interventions via a con a via interventions key of integration ], 5 ]. However, none of these calls have suf have calls these However,of ]. none CBPHC andstrategiesusedbyeffectiveprojects June 2017 •Vol. 7No. 1•010906
186 published reports 11]. ------
VIEWPOINTSPaPErS PaPErSVIEWPOINTS June 2017 •Vol. 7No. 1•010906 Perry etal. Table 1. Intervention delivery Intervention Use ofcommunityhealthworkers Health systemsstrengthening members ingeneral volunteers, beneficiariesandcommunity Education ofcommunity–levelstaff, Community engagement Program designandevaluation c ategory
oF Summary ofstrategiesusedbyCBPHCprojectsSummary toimprove childhealth
strategy key informantsorparticipants infocusgroup discussions. circumstances, community members assist with data collection for structured and participate as surveys these In evaluation. project of time the at as well as phase planning project the during consulted been and whether or not the cause of death “structure” has changed over time [ area project the in deaths child of causes leading the assess to used been have Verbalmethods autopsy SEARCH [13, Group projects [ projects Group ects measured changes in mortality directly, either through prospective vital events registration as in Care es of polio) and for other vaccine–preventable diseases such as neonatal tetanus and measles. Some proj workers and communities. Examples are for surveillance acute flaccid paralysis (to identify possible cas community–based by provided information using component disease–surveillance a included ects oration with community members or the development of village rosters of beneficiaries. Sometimes proj nant women, and their young children) such as household censuses carried out by the project in collab Various approaches were used to determine the beneficiary population (usually mothers, including preg was usedtoguideproject planningorevaluation. projects, development of management and planning the in people local of viewpoints the incorporates or collaborators for these surveys. In some projects, Participatory Rural Appraisal (PRA), an approach that interviewers anthropometry.as by served determined members as community status Oftentimes, tional 14] or through retrospective measurements obtained from maternal birth histories [
12 ] and in the pioneering CBPHC field project at Gadchiroli, India, conducted by conducted India, Gadchiroli, at project field CBPHC pioneering the in and ] Provision atcommunityoutreach ofhealthservices pointsbymobileteams from peripheralfacilities women’sParticipatory groups Home visits Community casemanagement Use oftrainedandpaidworkerswith1yeartraining Use oftrainedandpaidworkerswith1–11monthstraining Use ofvolunteersforregular ongoingactivities Intermittent useofminimallytrainedvolunteersforhighlyspecific,targeted activities Training ofcommunity–levelhealthcare providers Training ofproviders healthcenter atprimary Strengthening logistics/drugsupplysystem system Strengthening ofsupervisory Strengthening ofqualitycare atreferral facility Strengthening referral system Identification ofcaseschildhoodillnessinneed referral Training oftrainers/cascadetraining Positive devianceinquiry Education ofgrandmothers Peer–to–peer education(volunteermothersvisitingneighborswithtargeted healthmessages) Skits, storiesandgamesforhealtheducationmessages Social marketing(mediacampaigns,posters,radio,etc.) Involvement ofolderfamilymembers(menandgrandparents/mothers–in–law) Formation and/orsupportofmicrocredit programs forwomen RuralAppraisal(PRA) Participatory Sharing locallyobtainedhealth–related datawiththecommunity Formation and/orsupportofwomen’s groups Collaboration withorformationofvillagehealthcommitteesand/orcollaborationlocalleaders Engagement ofcommunitiesinplanningandevaluation Determination ofcausedeathfrom verbalautopsies Retrospective birthhistories) mortalityassessment(basedonmaternal Prospective registration ofvitalevents(pregnancies, birthsanddeaths) Disease surveillance (based on information provided by community–based workers and communities) Census–taking Village rosters ofbeneficiaries RuralAppraisal(PRA) Participatory Knowledge, practiceandcoverage(KPC)householdsurveys s pecIFIc
strategy 404 www.jogh.org 17]. Finally, communities have • doi:10.7189/jogh.07.010906 15, ]. 16]. - - - - - www.jogh.org • doi:10.7189/jogh.07.010906 our database ( database our in included assessments the in mentioned commonly is and forms many takes engagement Community Strategies forcommunityengagement proach has been developed by Curamericas for isolated mountainous communities in Guatemala [ basis, usually during pregnancy, to cover all or most of the cost of such transport if needed. One such ap regular a on money of amounts small pay families whereby schemes insurance to linked sometimes are also ensure that the family can obtain transport at a fixed, fair, and affordable price. These referral systems sureand childrenarises and nearestcomplication mothers thea that access whenever can facility health en to systems transport emergency establish to communities with work to been has approach Another ects hasbeenpublished[24]. projthese of number a highlighting publication A interventions. survival child key of coverage graphic projects funded by the US Agency for International Development, often with marked expansions of geo survival child many in utilized (C–IMCI), Illness Childhood of Management Integrated munity–based serious conditions are present, prompt care is sought. This is core feature of the approach potentially known when as and, appropriatelyCom care seek people which in one is system facility.health stronger A serious childhood illness or about pregnancy and childbirth for which care should be sought at a health for signs warning about messages educational with families their and mothers providing was common Many CBPHC projects carried out health system strengthening activities of various sorts. One of the most Strategies forhealthsystemsstrengthening household. numbers ofpeer–to–peercounselorscanbetrainedtoconveykeymessagesevery approach,large training” “cascade Throughthis set. another train then who trainers of set another train then each trainers These messages. education health of projectset a a areain with trainers master of ber Another approach used by some projects is called Care Groups [ feed thembuthowtheycare forthemmore broadly. they how just not – children their for care they how village the in children well–nourished of mothers the from learn village a in children strategy, this undernourished With of [22]. mothers undernutrition educational strategy used in some projects is positive deviance inquiry, usually for addressing childhood community. innovative the particularly in One elders influential and respected are they since messages at meetings of small groups of neighbors. Sometimes projects targeted grandmothers for health education or visits home of time the at neighbors their to them conveyed then who mothers) are often most (who Other approaches involved teaching health education messages to volunteer or paid community workers projectsurvival inCambodia[20, Worldchild the Relief is approach this of community.example entire noteworthy the One or mothers, have conveyed health education messages through skits, puppet shows and games that engaged children, community.entire the to messages Others key convey to posters and radio as such channels marketing proaches to educating CHWs, beneficiaries, and community members as a whole. Some have used social ap innovative many adopted have effectivenessdatabase the projectsour of of in Assessments included and communitymembersingeneral Strategies foreducationofcommunity–levelstaff,volunteers,beneficiaries workers fornationalhealthweeksinSierraLeone[19]. community short–term and communities of mobilization national and [18] Nigeria and [12] zambique Mo in churches of mobilization include strategies engagement community strong with projects of ples ly, projects have engaged communities by sharing surveillance and evaluation results. Noteworthy exam engaging fathers, mothers–in–law, traditional healers and local drug sellers. Finally, though not common toward geared are activities special Finally,projects, some supervision. in their in participate and them Communities are commonly requested to participate in the selection of CHWs and to provide support to as wellformationandsupportofwomen’s microcredit andsavingsgroups. are common forms of community engagement, including education and consciousness raising of women existing community groups or formed new ones, often women’s groups. Activities that empower women with worked have projects Many behaviors. healthy key practice to or campaigns health in ticipate ers, including local religious leaders, are commonly consulted. Communities are often mobilized to par tence, and projects work with them in project design, implementation and evaluation. Community lead- ). Village health committees are often formed if they were not previously in exis in previously not were they if formed often are committees health VillageTable 1). 21]. 405 23], which involves training a small num- CBPHC andstrategiesusedbyeffectiveprojects June 2017 •Vol. 7No. 1•010906 25 , 26 ]. ------
VIEWPOINTSPaPErS PaPErSVIEWPOINTS June 2017 •Vol. 7No. 1•010906 Perry etal. projects withevidenceofeffectiveness inimproving neonatalandchildhealth Table 2. training whoare salaried of more or year 1 with workers of Use training whoreceive asalary of months 1–11 with workers of Use ongoing activities regular for workers unsalaried of Use targeted activities specific, highly for workers unsalaried trained minimally of use Intermittent c ategory
oF community–based primary healthcareSpecific examplesofcommunityhealthworkers(CHWs)utilized incommunity–basedprimary (CBPHC) chw vitamin A and zinc. When community–level workers did not have the capacity to treat children with acute iron, as such micronutrients of provision cases some in and RUTF), or foods, therapeutic ready–to–use as (known prepared commercially or locally available foods nutritious highly of provision fluids, tion administration of oral (and in a few cases intramuscular) antibiotics [ [ supplies other and medications for port these conditions). This requires, in addition to proper training, appropriate supervision and logistical sup treat to how mothers taught also CHWs these cases some in (or themselves conditions these manage to Some projects that were effective in improving neonatal and child health also trained and supported CHWs they canteachthesetomothers,othercaregivers, andfamilymembers. [ children sick among and period, newborn the to teach pregnant women and their families about danger signs during pregnancy and childbirth, during The review identified considerable evidence regarding the effectiveness of training and supervising CHWs agents treatment ofserious childhoodillnessbymothersand/ortrainedcommunity Community casemanagement:recognition, referral, and(whenpossible) at communityoutreach pointsbymobileteamsfrom peripheralfacilities. itation of all homes, (3) facilitator–led participatory women’s groups; and (4) provision of health services cumstances) treatment of serious childhood illness by mothers and/or CHWs; (2) routine systematic vis Four types of strategies for implementing interventions were: (1) recognition, referral, and (in certain cir Strategies forimplementationofinterventions Table 2provides alistinganddescriptionofthetypesCHWsdescribedbyreports inourdatabase. engaged for a specific local project or study while others were part of a national government–run program. others were formally paid government employees. Some CHWs were volunteers or workers while who service, had been their for certificate a or course training a attending for payment diem” “per a only ceived hours or days while for others CHWs had one year or more of full–time formal training. Some CHWs re few a only lasted training the projects, some For CHWs. of variety broad a engaged database our in ects trained and paid workers who can implement specific interventions aimed at improving MNCH. The proj formally more or volunteers trained – CHWs of types various on relyprograms often Community–based Strategies foruseofcommunityhealthworkers tics/drug supplysystemforCHWs. logis the strengthening or CHWs, providedto supervision the strengthening by CHWs, to training ing Other approaches include improving the quality of the community–based health system itself by provid helping thefacilitytoimprove itsownstockofdrugsandsupplies. or there work staffwho training of form the takes often This referrals. for care and accept to facilities of en the quality of care provided at primary health care centers or referral hospitals, including the capacity Many projects, while implementing community–based interventions, also engage in activities to strength
n Auxiliary nurses,communityhealthofficers,Auxiliary healthextensionworkers Group facilitators(animatorsorpromoters) assistants, surveillance health agents, planning family agents, health community agents, Health volunteers,femalecommunityhealthvolunteers surveillance community workers, health family teams, bridge–to–health mothers, selor coun nutrition workers, management case community animators, teers, Promoters, peer educators, malaria or nutrition agents, Care Group volun- Child HealthDayvolunteer ames
gIven
to chw s
In
thIs
accompagnateurs, category 30– 406 ]. The community–based treatment modalities included included modalities treatment community–based The 33]. 27– lead mothers, lead ]. CHWs can learn to recognize danger signs and and signs danger recognize to learn can CHWs 29]. soccoristas, 34], administration of oral rehydra www.jogh.org Care - c casional smallstipend oc- an or training, for payment diem per uniforms, as such incentives certain receive May ment pay diem per a receive May omment • doi:10.7189/jogh.07.010906 ------www.jogh.org • doi:10.7189/jogh.07.010906 The provision of immunization services by mobile health teams at points beyond a peripheral health fa health peripheral a beyond points at teams health provisionmobile The by services immunization of may have a vehicle or more likely a motorcycle, bicycle, horse or donkey, or they may even teams travel mobile by These foot. outreach. community–based of means common a is centers health at based teams mobile by sessions, immunization outreach holding including clinics, satellite at services of Provision immunization sessions,bymobileteamsfrom peripheralfacilities Provision atsatelliteclinics,including holdingoutreach ofservices Routine systematicvisitationofhomes is available for general use [ ed. A comprehensive manual for community–based diagnosis and treatment of serious childhood illness illness, they informed mothers and caretakers that urgent treatment at a referral health facility was need- [ volunteer) each surrounding households 10–15 the in mothers the of each to disseminate to message volunteers who meet with a facilitator (promoter/animator) once a month to learn a key health education [ cycle action–learning participatory a of use the ing The literature illustrates several effective approaches to facilitating participatory women’s groups, includ effective for assisting mothers to rehabilitate malnourished children detected through growth monitoring. women’s be Participatory can childbirth. also groupsand pregnancy during signs danger and pneumonia, childhood of signs diarrhea, of treatment and prevention washing, hand feeding, infant spacing, birth come generation activities). These groups may also provide a vehicle for counseling about breastfeeding, main that are a priority to the community and that may also have an indirect effect on health (such as in- do health the of outside issues address also groupsmay These newborn. the properof careand illness, serious of signs danger behaviors, healthy about education and empowerment further for opportunity Participatory women’s groups are led by facilitators with less than two weeks of training who provide the malaria), familieswillbemore predisposed tocontacttheCHWforearlyandprompt treatment. Participatory women’s groups dren develop signs of serious illness that can be managed by CHWs (such as for pneumonia, diarrhea or Finally, an ongoing program of home visitation provides a foundation of trust and awareness. When chil hensive approach totheentire todeliveringbasicservices population[ compremore a of part as programpopulation a in homes all of visitation monthly regular to [39] dren strategies using community–level workers, from weekly home visits for providing micronutrients to chil childhood pneumonia, and family planning services. There are a number of variations of home visitation complementary feeding, hand washing, prevention and treatment of diarrhea, detection and treatment of appropriate and breastfeeding about counseling providing of means effective an also is visitation Home forthosewithoutready accesstohealthfacilities. an alternative provides visitation Home [38]. cost and distance to due facilities health accessing in face women many weight newborns who need additional home visits. A number of studies have highlighted the difficulties low–birth– identify to newborns of weight the of measurement and hypothermia, of prevention birth, after immediately breastfeedingpromotion of childbirth, and pregnancy during signs danger about tion these visits include the following: education about proper nutrition, promotion of antenatal care, educa birth, and at least one visit as soon as possible after delivery [ women receive two home visits during the prenatal period, one home visit during the first 24 hours after pregnant all that recommend UNICEF and OrganizationWorld Health the evidence, current on Based diarrhea, pneumonia,andmalaria. treatment for certain conditions that can be identified at the time of home visits such as growth faltering, providearepresent,and signs danger referralwhen encourage illness, childhood pregnancyand during signs danger recognize mothers–in–law), and husbands (especially members family other and them to education providechildren, young of mothers and women pregnant identify to able generally are visits particularly during pregnancy and the early neonatal period. Community–level workers who make home services, key selected and education health essential with population program the in everyone provide Routine systematic visitation of homes makes it possible to identify those in need of basic services and to now beingscaledupinmanycountries[36]. 43, 44], andeducationsessionsledbycommunitymobilizers[ 45]. 35]. Integrated community case management (iCCM) for childhood illness is 407
41 , 42 ], formation of Care Groups (10–15 women women (10–15 Groups Care of formation ], 37]. Activities that should take place during CBPHC andstrategiesusedbyeffectiveprojects 40]. June 2017 •Vol. 7No. 1• 010906 ------
VIEWPOINTSPaPErS PaPErSVIEWPOINTS June 2017 •Vol. 7No. 1•010906 Perry etal. Table 3. lying communitiesonaregular basis. out in facilities health of *Outreach outside staff locations facility specified health at of sessions immunization and/or clinics mobile holding includes t Provision andpromotion offamilyplanning services Iodine supplementationiniodine–deficient areas where fortifiedsaltisnotconsumed (PMTCT) ofHIVinfection transmission mother–to–child of prevention and women pregnant in testing HIV of Promotion Detection andtreatment ofsyphilisin pregnant womeninareas ofhighprevalence endemic areas Intermittent preventive treatment of malaria during pregnancy (IPTp) and infancy (IPTi) in malaria– Detection/referral ofmalariawithorwithout provision ofcommunity–basedtreatment Indoor residual sprayinginmalaria–endemic areas Insecticide–treated bednets(ITNs)in malaria–endemicareas therapeuticfoods) dry mentation (including rehabilitation of children with severe acute undernutrition through ready–to–use Community–based rehabilitation of children with protein–calorie undernutrition through food supple natal sepsisbyCHW) neo of treatment and diagnosis and hypothermia, of prevention cleanliness, of promotion feeding, breast exclusive and immediate of promotion for visits home (frequent care neonatal Home–based Detection/referral ofpneumoniawithorwithoutprovision ofcommunity–basedtreatment Promotion ofcleandeliveries,especiallywhere mostbirthsoccurathomeandhygieneispoor Promotion oforalrehydration therapy(ORT) fordiarrhea withorwithoutzincsupplementation Promotion ofhygiene(includinghandwashing),safewater, andsanitation Promotion ofappropriate feedingbeginningat6monthsofage complementary of lifeandcontinuednon–exclusivebreastfeeding beyond6months Promotion of breastfeeding immediately after birth, exclusive breastfeeding during the first 6 months Provision ofpreventive zincsupplementstoallchildren 6–59monthsofage Provision ofsupplementalvitaminAtochildren 6–59monthsofageandtopost–partummothers women ofreproductive age (Hib), pneumococcus, rotavirus immunizations for children; tetanus immunization for mothers and Influenza Haemophilus b measles, Type tetanus, pertussis, diphtheria, polio, BCG, Immunizations: echnIcal
InterventIon Child health interventions withstrongChild healthinterventions evidenceofeffectiveness through community–basedimplementation implementation modality.implementation in shown interventions The Table 3 demonstrates which evidence–based child can interventions survival be implemented by which the basisofpreviously developedhouseholdregisters. on identified often are children These points. outreach the to come not did who children and mothers community building or even under a tree, and home visits are often carried out in addition to reach those or school a at as such center health a from separate points outreach peripheral at providedare Services detailed discussion of these four intervention delivery strategieshasbeenreported delivery elsewheredetailed discussionofthesefourintervention [ the LivesSavedTool (LiST)forinclusioninprogram plansforreducing under–5mortality[ [ tablets de–worming or tablets, water–purification packets, rehydration oral of distribution and ministration, vitamin A supplementation, nutritional monitoring (and referral of malnourished children), ad immunization of combination some include usually year,they a twice Weeks).occurring Generally One widely implemented variation of this strategy is Child Health Days (or sometimes called Child Health ications toprevent ortreat malaria,andgrowth monitoringtodetectcasesofchildhoodmalnutrition. natal care, testing for HIV and syphilis, distribution of insecticide–treated bed nets, distribution of med ante basic services, planning family of provision and of promotion include outreachthrough provided cility is now well–developed in many low–income countries [46]. Other examples of services that can be both in the maternal as well as the neonatal/child health CBPHC projects. More than three–fourths of the findings are contained in to describe the degree to which communities were involved in various aspects of the project. Some of the When program assessments that qualified for the review underwent data extraction, reviewers were asked Frequency ofselectedprogram–related processes
Table 4. These findings demonstrate a high degree of community engagement, 408 are those which have been identified by by identified been have which those Tableare 3 - - - management Community Community case case c X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X www.jogh.org ommunIty – based Home Home • doi:10.7189/jogh.07.010906 visits
InterventIon X X X X X X X X X X X X X X X X X X women’s Partici- groups patory patory
delIvery 49]. Amore
strategy 50]. services 47, reach Out- X X X X X X ]. 48]. - - - - neonatal andchildhealth.Bluetriangles represent contextual factors. healthcarecommunity–based primary programs forimproving maternal, www.jogh.org Figure 1. database Table 4. Inputs ImplementatIon s Processes Evaluation tage
oF
Community involvement in the implementation of maternal, neonatalandchildhealthCBPHCprojects Community involvementintheimplementationofmaternal, includedinthe g A conceptualframeworkforplanning, implementingandevaluating • doi:10.7189/jogh.07.010906 a omto n/rspoto omnt rus53.6 Community involvementinplanning Formation and/orsupportofcommunitygroups Training ofCHWs Promotion ofequity Promotion ofwomen’s empowerment Promotion ofleadershipinthecommunity Promotion ofcommunityempowerment Promotion oftheuselocalresources community andthehealthprogram Promotion ofpartnershipsbetweenthe Community involvementinimplementation Community involvementinevaluation ctIvIty gagement for virtually all aspects of programming. Each of these aspects of community engagement are engagement community of aspects these of programming.Each of aspects all virtually for gagement was an essential element for strategy implementation. By its very nature, CPBHC requires community en category, engagement delivery.strategy community each intervention Within (f) and CHWs, of use (e) munity–level staff, volunteers, beneficiaries and community members, (d) health systems strengthening, were identified: (a) program design and evaluation, (b) community collaboration, (c) education for com strategies of categories Six projectimplementation. in engagement community of degree high a mented This analysis of strategies used by effective community–based programs for improving MNCH has docu DISCUSSION of communityengagementordetailshowwasactuallycarriedout. tion provided intheassessmentwasrarely sufficient toprovide anydeeperunderstandingofthequality project.the of component engagement community the describing in detail of level this into go Informa- areings underestimateto likely highly largea since situation true the not did assessments the of portion promotedequity.ter find These projectevaluation. the in community the projectsinvolved the of 40% half promoted women’s empowerment, one–third promoted leadership in the community, and one–quar the health program, promoted the use of local resources, or promoted community empowerment. Almost nities in project implementation, and more than half promoted partnerships between the community and commu projectsengaged the of 81% projectactivities. of planning the in as well groupsas community of support or formation the in communities engaged one–third than more and CHWs trained projects p maternal ercentage that 409
( descrIbe n cbphc
= 152) oF 46.4 86.3 24.8 62.7 41.8 62.7 74.5 73.2 90.8 50.3
assessments
actIvIty proJects its importance for program effectiveness. and strategies implementation of dimension this of richness the into insights provide review this of findings The impact. that achieve to used egies in sufficient detail the exact implementation strat describing on than rather interventions, of set or intervention, the of impact health the on usually is focus the where literature, scientific reviewed MNCH have not been well described in the peer– improve to projects by used approaches and gies In general, the details of community–based strate- other articlesinthisseries[51– in discussed are health child and neonatal ternal, ma to specifically pertain they as strategies these of elaboration Further children. and neonates and its capacity to improve the health of mothers, program health the of benefit the for community part of the process of building capacity within the
oF
cbphc p neonatal ercentage actIvIty proJects
CBPHC andstrategiesusedbyeffectiveprojects and
oF 36.1 35.5 74.0 24.8 40.6 30.4 53.6 53.2 53.6 78.1 37.5 ( /
or n assessments = 548)
chIld that June 2017 •Vol. 7No. 1•010906
descrIbe health
oF
health that p maternal ercentage
descrIbe cbphc 53]. ,
neonatal oF
actIvIty 38.3 39.5 76.6 24.8 45.4 32.9 55.6 57.8 57.8 80.9 40.3
assessments proJects
and Figure 1 ( n /
= 700) or combIned
oF chIld
all
------
VIEWPOINTSPaPErS PaPErSVIEWPOINTS June 2017 •Vol. 7No. 1•010906 Perry etal. “ improving childhealthandnutrition[54]: programs of effectiveness the of review comprehensive 1980 their in observed al. et Gwatkin As vices. ser need who those reachorder to in facilities health of outside interventions of delivery munity–based identified in other articles in this series all highlight the importance of community engagement and com The framework in tem workingwiththecommunity. sys producedhealth outcomes the health by eventual together the in embedded is interventions, the of ing the health of mothers, neonates and children. The process,delivery along with the technical content contains a framework that attempts to capture the importance of community empowerment for improv dence of effectiveness [ evi long–term have projectsthat four of characteristics common the review we series this in Elsewhere at scaleinrelatively routine conditions[57]. [ work their do to need they commodities basic the including supervision, and support adequate receive the number of CHWs and their supervisors is sufficient for the population being served, and that CHWs that assuring include programslarge–scale of effectiveness the to bottlenecks the that show studies few However, these period. time longer a over scale at CBPHC of assessments rigorous of examples few are assessments of large–scale integrated programs at scale carried out for five or more years are needed. There rigorous independent, More interventions. of number small a or one of studies efficacy and trials, term short– projects, demonstration small fromlargely derived are database our up making assessments The constraints), andavailableimplementationstrategies. system health (including constraints contextual challenges, logistical resources, available the given ties priori epidemiological targetingthe interventions evidence–based of coverage high achieving of sibility priorities (ie, the most frequent and readily preventable or treatable serious conditions) as well as the fea- fectiveness in improving MNCH in a given geographical area requires knowing the local epidemiological to select the most appropriate combinations of interventions and implementation strategies. Program ef order in considered carefully be must context the successful, be to programs community–based for der Figure 1 emphasizes the importance of context. In fact, strategies in general are context– specific. In or tothosewhoneedthem. CBPHC involves,aboveall,gettingservices important asare whichservicesshouldbeoffered. decisionsconcerning ” as is people the to services getting for plans of development mortality.the on … impact Thus, little have viously cility–based approaches health care wouldavert farfewer(20%atprimary centersand22% athospitals). vented byallcurrently couldbeprevented availableinterventions with community–basedapproaches. Fa- concluded that almost two–thirds (59%) of maternal, prenatal, neonatal, and child [ deaths analysis that recentcould One be preinterventions. these of coverage population high achieve to order in terventions in evidence–based implement worker community–level a from having greatly benefit can systems Health asanimportantassetforthe communitytogaintrustinthehealthsystem[ serve provide it. Small, well–run first–level referral hospitals can be cost–effective in improving health and can who workers the and work community–based the to credibility lends care, surgical and hospital basic including care, referral and curative high–quality Additionally, and transport. compassionate adequate and vaccines of supplies steady minimum, a at need, facilities peripheral at based teams health mobile the training, equipment, supplies and transport needed to support community–level work. For example, programsHealth arefacilities ensurehealth appropriatelyto local need that staffed staffthe that and has • • • such ananalysisinclude: tunately it is beyond the scope of the current series of articles. Questions that might be addressed through reviewunfor our useful, in be ects includedwould analysis such an Although out. carried not been has 55, Unless services reach those in need, even the best–conceived primary health and nutrition care programs can ob
tions affect theeffectiveness ofpublic–sector programs? What is the contribution of civil society and NGOs to larger–scale projects and how do these contribu ferent projects, andhowdidthesefeatures contributetoeffectiveness? What kinds of community engagement and what kinds of community–level workers were used in dif ity andeffectiveness? effectivenessIs qual were maintain steps to projectsspecific taken as what weakened not, If up? scale 56]. Future research is needed to rigorously assess the effectiveness of community–based approaches
Figure 1 and in fact the strategies identified in this article as well as the interventions 58]. A more in–depth analysis of the strategies and effectiveness of the larger proj 410
www.jogh.org • doi:10.7189/jogh.07.010906 59, 60]. 61 ------] ------www.jogh.org • doi:10.7189/jogh.07.010906 recent assertion of Marston et al. [ fying implementationstrategiesusedbyprograms withdemonstratedeffectiveness. ond child revolutionsurvival [ mented that “We have the bullets [interventions] but not the guns [implementation strategies]” for a sec- la MNCH, improve to movement global the of leaders acknowledged widely the of one Victora, Cesar eradication [66]. points. Thus, for instance, home visits have proven to be an essential strategy for the final stages of polio riers – in addition to geographic barriers – in accessing services at facilities or even at peripheral outreach groupshigh–mortality,in bar social many to resource–constrainedleading uncommon, arenot settings all who need them but also of ensuring that no one is left out. Marginalization and discrimination of sub– Routine systematic home visitation has the unique advantage of not to only delivering key interventions the projects whoseassessmentsare includedinourdatabase. in used strategies common most the are they but health, child improve can that interventions menting imple to approaches only high–mortality,the in not are strategies These resource–constrainedsettings. outreach services provided by mobile teams represent important delivery strategies for improving MNCH women’s and groups,participatory visitation, home systematic routine management, case Community leased World HealthOrganization globalstrategyforwomen’s andchildren’s health[ er with health for services the co–production of health care) will be central for achieving the recently re describing andanalyzingthe process ofcommunityparticipationare available [ morerobustand approachesUseful outcomes. to health to made this that contribution the assess better ects so that they can more fully describe the role of the community in the process of implementation and review.our in projincluded CBPHC future by used be to need indicators and Appropriateframeworks sessing the quality and effectiveness of this critical dimension of CBPHC were rarely used in the assessments as for indicators and frameworks that note also we projects, CBPHC effective implementing in munity ment. While one of the strengths of our paper is highlighting and further describing the role of the com A final limitation of our review is the overall difficulty of assessing community participation and engage by tworesearchers andthenhavingathird resolve anydifferences –helpstomitigatethislimitation. reviewers.ments of However,procedurethe each assessmentreviewed– having used we independently judg subjective on based are here reported findings the of some that is study this of limitation Another capacity ofthecurrent seriesofarticlestoaddress. (including a rigorous examination of the quality of the assessments) would be useful but goes beyond the and rich evidence about strategies for implementation of CBPHC projects described in the gray literature extensive the of analysis and consolidation Further information. more provide can and articles viewed These documents are useful in part because they are not subject to the same space limitations as peer–re additional 116 assessments are either unpublished evaluation reports that are publicly available or books. peer–reviewed journal articles, and many of them describe their strategies in greater detail. Most of these ing these strategies. Our database has been strengthened by the inclusion of 116 assessments that are not tion strategies. Our data extraction process was set up to glean whatever information was available regard difficultit make implementa- wordarticles describe The fully peer–reviewed to on placed journal limits Study limitations help strengthen community–based delivery strategies, as described in As we have seen in this analysis, empowering the community to be a partner with the health system can arepervision required tomaintainquality. su and training what and on take can CHW given a interventions many how on needed be will search In addition to continuing research on the capability of CHWs to provide specific interventions,more re cal communities, examples exist in which communities have taken full responsibility for this process [ in this capacity. Although these decisions are normally made by program leaders in consultation with lo functioning longer no are who those replace to CHWs new training and promptlyselecting for veloped adequate and standardized compensation or incentives for CHWs [ munities must be appropriately trained and supported; a recent Cochrane Review identified the need for Of course, the community–level workers who implement these interventions in collaboration with com 67]. The analysis provided here helps to point the way forward by identi 64] that community participation (in which communities work togeth 411
CBPHC andstrategiesusedbyeffectiveprojects 62]. An effective strategy must be de- Figure 1. The finding supports the June 2017 •Vol. 7No. 1•010906 68, 69]. 65]. 63 ]. ]. ------
VIEWPOINTSPaPErS PaPErSVIEWPOINTS June 2017 •Vol. 7No. 1•010906 Perry etal. rEFEr ENCES 4 3 2 1 5 6 ternal, newborn, andchildhealth.Lancet.2008;372:962-71. newborn, ternal, ma Tripathyfor LH, lessons Grabman participation: G, P,Community Laverack al. M, et Rosato C, Mwansambo N, Nair framework foraction.Lancet.2008;372:990-1000. Medline:18790321 a children: and babies, newborn women, for interventions health Integrating J. Liljestrand I, Pathmanathan B, Ekman 2008;372:972-89. Medline:18790320 Lancet. make? strategies care health primary integrated can difference what survival: child and newborn, maternal, Bhutta ZA, Ali S, Cousens S, Ali TM, Haider BA, Rizvi A, et al. Alma-Ata: Rebirth and revision 6–interventions to address 15. Medline:19150128 Reich MR, Takemi K. G8 and strengthening of health systems: follow-up to the Toyako summit. Lancet. 2009;373:508- ter Health.Geneva:World HealthOrganization, 2008. World Health Organization, Organisation for Economic Co-operation and Development, World Bank. Effective Aid, Bet ment: areview oftheevidence.Geneva,Switzerland:World HealthOrganization; 2004. develop and growth survival, child promote that practices community and Family KM. Edmond B, Kirkwood Z, Hill tings isoneofthegreat frontiers forglobalhealthinthe21 high–mortality,in systems health effective resource–constrainedbuilding set of process the in partners terventions is a pressing challenge for national programs. Unleashing the full potential of communities as within existing local and health systems constraints, of CBPHC implementation strategies for MNCH in feasibility, and fit the Determining limited. remains settings priority in scale at CBPHC of effectiveness Nonetheless, awareness about the full potential of CBPHC is still not yet widespread, and evidence of the high–mortality, resource–constrained settingshaveagreat potentialforimproving MNCHatlowcost. in programs community–based that recognition growing the supports review This coverage. health sal [ [ tuberculosis [70], AIDS point for developing synergies with community–based approaches for the detection and treatment of HIV/ entry an establish can MNCH improve to CBPHC strengthening for here identified strategies the Using deathsby2030. one oftheprioritytasksforendingpreventable childandmaternal tions. Building the capacity of health systems to work with communities to implement these strategies is leaders and community members in order to achieve high levels of coverage of evidence–based interven community with partnerships functional establish system health the that require strategies These tries. coun priority in mortality child and neonatal maternal, in decline the accelerate can strategies these of strengthened health systems. The evidence from this review supports the proposition that the application their activities, how they collaborated with communities, how they have used CHWs, and how they have evaluated and planned have projects CBPHC which in ways the of overview an provides analysis This CONCLUSIONS 72], detection and treatment of adult non–communicable diseases [ disclosure.pdf (availableuponrequest from thecorresponding author),anddeclare noconflictofinterest. www.icmje.org/coi_ at Form Interest Competing Unified the completed have authors All interests: Competing final draft. approvedand drafts the earlier revision of the in participated authors the of report.All our in included data tative end and participated in all decisions related to the overall review. ES and RK performed the analysis of the quanti Authorship declaration: HP wrote the first draft. HP, PF, BR, and SG guided this project from the beginning to the ecution ofthereview. ex the in role no had support financial provided that organizations The Foundation. Gates the and velopment, WorldThe ticle: UNICEF,Organization, Health De International for Agency WorldStates the United the Bank, ar this in described work the conduct to used were that funds providedorganizations following The Funding: it possibleforoneofitsconsultants,DrBahieRassekh,toparticipateasamembertheStudyTeam. review.the of phase initial the during Perry Dr to support World The salary and support, ministrative made Bank ad- space, office providing for Generations WeFuture funds. thank these of some administered which staff, tion Sec Health International its particularly and Association Health Public American Wethe tion. to grateful also are CORE Group (Collaboration and Resources for Child Health)/USAID, Future Generations, and the Gates Founda the Organization,World Health the of Development and Health Adolescent and Child of Department the Bank, ful to the following organizations that provided small grants to cover the expenses of this review: UNICEF, the World Acknowledgments: We are grateful to Melanie Morrow for her review of an earlier draft of this paper. We are grate
doi:10.1016/S0140-6736(08)61899-1 ] and malaria [ malaria and 71] doi:10.1016/S0140-6736(08)61407-5 412 ] as well as for the promotion of family planning services services planning family of promotion the for as well as 31]
Medline:18790319 st century. doi:10.1016/S0140-6736(08)61408-7 73], and the achievement of univer doi:10.1016/S0140-6736(08)61406-3 www.jogh.org • doi:10.7189/jogh.07.010906 ------www.jogh.org rEFEr ENCES • doi:10.7189/jogh.07.010906 8 7 9 28 27 26 25 18 17 12 11 16 15 14 13 10 24 23 22 21 20 29 19 many newborn babies canwesave?Lancet.2005;365:977-88.Medline:15767001 many newborn how interventions: cost-effective Evidence-based, L. T, Bernis Adam Walkerde S, N, Cousens ZA, Bhutta GL, Darmstadt undernutrition and survival. Lancet.2008;371:417-40.Medline:18206226 andsurvival. undernutrition Bhutta ZA, Ahmed T, Black RE, Cousens S, Dewey K, Giugliani E, et al. What works? Interventions for maternal and child doi:10.1542/peds.2004-1441 Medline:15866863 2005;115:519-617. Pediatrics. evidence. the of review a countries: developing in outcomes health neonatal and improvingperinatal for interventions Community-based RA. Haws BS, Hasan GL, Darmstadt ZA, Bhutta randomised controlled trial.Lancet.2008;371:1936-44.Medline:18539225 cluster-a Bangladesh: district, Sylhet in strategies throughimplemented package careintervention service-delivery two newborn- community-based of Effect al. et HR, Seraji EK, Williams S, Ahmed GL, Darmstadt S, El-Arifeen AH, Baqui doi:10.1038/sj.jp.7211277 of the Gadchiroli field trial: effect of home-based neonatal care. J Perinatol. 2005;25 Suppl 1:S92-107. Bang AT, Reddy HM, Deshmukh MD, Baitule SB, Bang RA. Neonatal and infant mortality in the ten years (1993 to 2003) 31. Medline:27016548 of the introduction and utilization of birthing facilities by an indigenous population. Glob Health Sci Pract. 2016;4:114- Stollak I, Valdez M, Rivas K, Perry H. Casas Maternas in the Rural Highlands of Guatemala: a mixed-methods case study loads/2016/09/Curamericas_Guatemala_FE.pdf. Accessed:26April2017. in Oriented Huehuetenango, Child Guatemala”. Survival 2015. Available: Global C. Focused Strategic Assessment: USAID Child Survival and Health Grants Program “Community-Based, Impact- heapol/czt005 trial. LancetGlobHealth.2015;3:e692-700.Medline:26475016 randomised cluster a Shower): (Baby Nigeria in women pregnant in care to linkage and testing HIV of uptake on tion Ezeanolue EE, Obiefune MC, Ezeanolue CO, Ehiri JE, Osuji A, Ogidi AG, et al. Effect of a congregation-based interven on infantandchildmortalityinBolivia.JHealthPopulNutr. 2003;21:383-95. programme healthcare primary comprehensive community-based a of Impact DG. Schroeder DS, Shanklin HB, Perry a community-based programme in Gaza, Mozambique. Trans R Soc Trop Med Hyg. 2007;101:814-22. P,Edward Ernst A, Taylor under-fiveof evidence the Examining H. Perry E, Mazive S, mortality in Becker reduction C, database description.JGlobHealth.2017;7:010901. and methods rationale, 1. health: child and neonatal maternal, improving in care health primary community-based of Perry H, Rassekh B, Gupta S, Wilhelm J, Freeman P. A comprehensive review of the evidence regarding the effectiveness Medline:24349053 vival during a health system strengthening intervention in Mali: a repeated cross sectional survey. PLoS One. 2013;8:e81304. Johnson AD, Thomson DR, Atwood S, Alley I, Beckerman JL, Kone I, et al. Assessing early access to care and child sur 2007;97:240-6. Medline:17194853 dren under age 5 in rural Haiti: effects of a comprehensive health system in an impoverished setting. Am J Public Health. Perry H, Berggren W, Berggren G, Dowell D, Menager H, Bottex E, et al. Long-term reductions in mortality among chil roli, India(1993to2003).JPerinatol.2005;25Suppl1:S108-22. AT,Bang Gadchi rural in trial field the of applications and care:neonatal summary Home-based HM. Reddy RA, Bang line:1973770 1990;336:201-6. Lancet. India. Gadchiroli, in trial intervention community-based of means by mortality hood Bang AT, Bang RA, Tale O, Sontakke P, Solanki J, Wargantiwar R, et al. Reduction doi:10.1016/j.trstmh.2007.02.025 in pneumonia mortality and total child 2003;362:65-71. Medline:12853204 Lancet. year? this prevent we can deaths child many How SS. Morris ZA, RW,Bhutta Steketee RE, G, Black Jones strate plausible evidence for child mortality impact. Health Policy Plan. 2014;29:204-16. Ricca J, Kureshy N, Leban K, Prosnitz D, Ryan L. Community-based intervention packages facilitated by NGOs demon 69. Medline:26374798 egy for improving maternal, neonatal, and child health in resource-constrained settings. Glob Health Sci Pract. 2015;3:358- Perry H, Morrow M, Borger S, Weiss J, DeCoster M, Davis T, et al. Care Groups I: An innovative community-based strat Wheaton, Il,andArlington,VA: World ReliefandBASICS;1997. Keeley E, Burkhalter RB, Bashir N, editors. The Hearth Nutrition Model: Applications in Haiti, Vietnam, and Bangladesh. Berggren G. Nutritional Education and rehabilitation Program: A Save the Children Project in Vietnam. In: Wollinka O, of CommunityHealth.Sudbury, MA:JonesandBartlett;2010. settings through engagement with the community: an example from Cambodia. In: Gofin J, Gofin R, editors. Essentials Perry H, Sivan O, Bowman G, Casazza L, Edward A, Hansen K, et al. Averting childhood deaths in resource-constrained Baltimore, MD:World Relief,2007. (2003-2007). Evaluation Final (Cambodia): Province Cam Kampong Project, Survival Child W.Life” Relief for “Light 2008;372:1151-62. Medline:18926277 Lancet. trial. controlledcluster-randomised a India: Pradesh, Uttar Shivgarh, in mortality neonatal on management Kumar V, Mohanty S, Kumar A, Misra RP, Santosham M, Awasthi S, et al. Effect of community-based behaviour change Medline:25316706 2015;7:26-31. Health. Int Leone. WeekSierra Health in Child and Maternal integrated an during vaccination measles Sesay FF, Hodges MH, Kamara HI, Turay M, Wolfe A, Samba TT, et al. High coverage of vitamin A supplementation and doi:10.1016/0140-6736(90)91733-Q doi:10.1371/journal.pone.0081304 doi:10.1093/inthealth/ihu073 doi:10.9745/GHSP-D-15-00266 doi:10.9745/GHSP-D-15-00051 doi:10.2105/AJPH.2006.088732 doi:10.1016/S0140-6736(03)13811-1 doi:10.1016/S0140-6736(08)61483-X 413
doi:10.1016/S2214-109X(15)00195-3 Medline:15791272 http://www.mcsprogram.org/wp-content/up- doi:10.1016/S0140-6736(07)61693-6 CBPHC andstrategiesusedbyeffectiveprojects doi:10.1016/S0140-6736(08)60835-1 Medline:15038594 doi:10.1016/S0140-6736(05)71088-6 June 2017 •Vol. 7No. 1•010906 Medline:23434515 doi:10.1038/sj.jp.7211278 Medline:15791283 Medline:17482222
doi:10.1093/ Med------
VIEWPOINTSPaPErS PaPErSVIEWPOINTS June 2017 •Vol. 7No. 1•010906 Perry etal. rEFEr ENCES 32 31 30 36 35 34 33 43 42 41 40 39 38 37 52 51 50 49 48 47 46 45 44 Available: http://apps.who.int/iris/bitstream/10665/70002/1/WHO_FCH_CAH_09.02_eng.pdf WHO. UNICEF. WHO/UNICEF Joint Statement: Home Visits for the Newborn Child: A Strategy to Improve Survival. 2009. meta-analysis of community-based trials. Lancet Infect Dis. 2003;3:547-56. Dis. Infect Lancet trials. community-based of meta-analysis Sazawal S, Black RE. Effect of pneumonia case management on mortality in neonates, infants, and preschool children: a Lancet. 2000;356:550-5.Medline:10950232 Kidane G, Morrow RH. Teaching mothers to provide home treatment of malaria in Tigray, Ethiopia: a randomised trial. line:19289979 2009;28:304-10. J. Dis Infect Pediatr Bangladesh. rural in workers health community by infections newborn of Effectivenessmanagement al. home-based et of SM, Rahman WilliamsI, SE, Mannan Arifeen S, AH, Ahmed Baqui EK, services forchildren.services AmJTrop MedHyg.2012;87:6-10. Medline:23136272 improveto strategy equity-focused an Management: Case treatment Community essential Integrated to on access ment Young M, Wolfheim C, Marsh DR, Hammamy D. World Health Organization/United Nations Children’s Fund Joint State coregroup.org/storage/documents/CCM/CCMbook-internet2.pdf. Accessed:3April2017. nesses in the community. A guide for program managers. Washington, DC: CORE Group; 2010. Available: ill Treatingchildhood essentials: common management case Community BASICS. Children, the Save Group. CORE community healthvolunteersinNepal.JPerinatol.2012;32:959-65. female of experience the sepsis: neonatal possible of treatment community-based for system injection prefilled Uniject the in P,gentamicin Dawson of YV.D, acceptability Pradhan Neupane and KC, Feasibility Gargi J, Sharma PS, Coffey versity Press Limited;1996. Chowdhury AMR, Cash RA. A Simple Solution: Teaching Millions to Treat Diarrhoea at Home. Dhaka, Bangladesh: Uni 3099(03)00737-0 tings. GlobHealthSciPract.2015;3:358-69.Medline:26374798 high-mortality,in set resource-constrainedhealth child and neonatal maternal, improving for strategy delivery based MorrowBorgerH, WeissM, S, Perry T,Davis M, DeCoster J, emergingCareGroupscommunity- an al. innovative I: et Medline:15364188 2004;364:970-9. Lancet. trial. controlledcluster-randomised Nepal: in outcomes birth women’son with groups tion Manandhar DS, Osrin D, Shrestha BP, Mesko N, Morrison J, Tumbahangphe KM, et al. Effect of a participatory interven Lancet. 2013;381:1736-46.Medline:23683640 ing and action to improve maternal and newborn health in low-resource settings: a systematic review and meta-analysis. Prost A, Colbourn T, Seward N, Azad K, Coomarasamy A, Copas A, et al. Women’s groups practising participatory learn doi:10.1093/heapol/czl005 2006;21:217-30. Plan. Policy Health Haiti. in system integrated Schweitzer’s Albert Hopital through mortality under-five Reducing al. et H, Menager F,JR, Philippe Dortonne M, D, Cayemittes Dowell H, Perry J Nutr. 2003;133:4150-7. Medline:14652364 and zinc is associated with lower morbidity due to diarrhea and acute lower respiratory infection in Bangladeshi infants. Baqui AH, Zaman K, Persson LA, El Arifeen S, Yunus M, Begum N, et al. Simultaneous weekly supplementation of iron doi:10.1016/j.socscimed.2006.01.015 South Africa: an exploration using a geographical information system. Soc Sci Med. 2006;63:691-705. Tanser F, Gijsbertsen B, Herbst K. Modelling and understanding primary health care accessibility and utilization in rural health findings.JGlobHealth. 2017;7:010903. neonatal 3. health: child and neonatal maternal, improving in care health primary community-based of effectiveness Sacks E, Freeman P, Sakyi K, Jennings M, Rassekh B, Gupta S, et al. Comprehensive review of the evidence regarding the health findings.JGlobHealth. 2017;7:010902. the effectiveness of community-based primary health care in improving maternal, neonatal and child health: 2. maternal Jennings M, Pradhan S, Schleiff M, Sacks E, Freeman P, Gupta S, et al. Comprehensive review of the evidence regarding CBPHC%20FINAL.pdf. Accessed:3April2017. Available: 2010. survival. child promoting for strategies livery de P,service Freeman community-based H, strengthen to Perry evidence current the on Building B. Rassekh S, Gupta line:24564438 WalkerTamN, Y, ToolSaved Lives the of Friberg Overview IK. 3:S1. Suppl 2013;13 Health. Public BMC (LiST). child health and nutrition services. Food Nutr Bull. 2013;34:412-9. of delivery integrated the for strategy day health child the of Evolution N. Dalmiya T,AC, Diaz Noordam AC, Palmer Health. 2010;15:296-305.Medline:20070638 TropAfrica. approachin Days Health Int Child Med the of assessment multi-country froma programmes:experiences Doherty T, Chopra M, Tomlinson M, Oliphant N, Nsibande D, Mason J. Moving from vertical to integrated child health Countdownto2015FINALREPORT-apr7.pdf/. Accessed:26April2017. Available:2008. Survival. Child UNICEF.and Newborn Maternal, Trackingin Progress BLT.07.045849 2008;86:452-9. Organ. WorldHealth Bull Pakistan. rural in study pilot a from Bhutta ZA, Memon ZA, Soofi S, Salat MS, Cousens S, Martines J. Implementing community-based perinatal care: results 81. Medline:26374799 and child health outcomes achieved in high-mortality, resource-constrained settings. Glob Health Sci Pract. 2015;3:370- Perry H, Morrow M, Davis T, Borger S, Weiss J, DeCoster M, et al. Care Groups II: a summary of the maternal, neonatal doi:10.1097/INF.0b013e31819069e8 doi:10.1186/1471-2458-13-S3-S1 doi:10.1016/S0140-6736(04)17021-9
doi:10.9745/GHSP-D-15-00052 doi:10.1016/S0140-6736(00)02580-0 doi:10.1111/j.1365-3156.2009.02454.x doi:10.1016/S0140-6736(13)60685-6 414
Medline:24605691 doi:10.9745/GHSP-D-15-00051 http://www.mchip.net/sites/default/files/USAID%20 Medline:22422117 Medline:12954560 doi:10.4269/ajtmh.2012.12-0221 www.jogh.org doi:10.1177/156482651303400406 Medline:18568274 doi:10.1038/jp.2012.20 http://www.who.int/pmnch/ • doi:10.7189/jogh.07.010906 . Accessed:26April2017.
Medline:16565151 doi:10.1016/S1473- Medline:16574290
doi:10.2471/ http://www. Med- Med------
www.jogh.org rEFEr ENCES • doi:10.7189/jogh.07.010906 60 59 58 57 56 55 54 53 64 63 62 61 70 69 68 67 66 65 73 72 71 ment inresource-poor settings.Lancet.2001;358:404-9.Medline:11502340 Farmer P, Leandre F, Mukherjee JS, Claude M, Nevil P, Smith-Fawzi MC, et al. Community-based approaches to HIV treat bution ofdifferent World services? JSurg. 2006;30:505-11.Medline:16528459 Gosselin RA, Thind A, Bellardinelli A. Cost/DALY averted in a small hospital in Sierra Leone: what is the relative contri Int JGynaecolObstet.2003;81:83-92.Medline:12676406 McCord C, Chowdhury Q. A cost effective small hospital in Bangladesh: what it can mean for emergency obstetric care. with evidenceoflong–termmortalityimpact.JGlobHealth.2017;7:010309. projects of characteristics shared 7. health: child and neonatal maternal, improving in care health primary nity-based FreemanS, Gupta P.B, Rassekh H, Perry Comprehensive regardingreviewevidence the of effectivenessthe commu of Evaluation: thetoppriorityforglobalhealth.Lancet.2010;375:526. line:20207412 2010;375:1193-202. Lancet. trial. controlled cluster-randomised a Bangladesh: in districts rural three in Effectal. et AR, Rego K, Khan S, Shaha B, Banerjee S, women’sup scaling of Barnett K, Azad groups outcomes birth on 6736(09)62060-2 n et fia a ersetv eauto. act 2010;375:572-82. Lancet. evaluation. retrospective a Africa: west in Bryce J, Gilroy K, Jones G, Hazel E, Black RE, Victora CG. The Accelerated Child Survival and Development programme Development Council;1980. Gwatkin DR, Wilcox JR, Wray JD. Can health and nutrition interventions make a difference? Washington, DC: Overseas findings. JGlobHealth.2017;7:010904. health child 4. health: child and neonatal maternal, improving in care health primary community-based of fectiveness ef the regarding evidence the of review Comprehensive H. Perry S, Gupta P,FreemanB, Rassekh E, Sacks M, Schleiff women’s, children’s and adolescents’ health. Bull World Health Organ. 2016;94:376-82. Marston C, Hinton R, Kean S, Baral S, Ahuja A, Costello A, et al. Community participation for transformative action on Problems Report.Geneva,Switzerland:World inAfrica:AFinalMulti-Country HealthOrganization, 2008. Special Programme for Research and Training in Tropical Diseases. Community-directed Interventions for Major Health Database SystRev. 2013;8:CD010414.Medline:24101553 Cochrane synthesis. evidence qualitative health: child and maternal to access improve to programmes worker health Glenton C, Colvin CJ, Carlsen B, Swartz A, Lewin S, Noyes J, et al. Barriers and facilitators to the implementation of lay key messagesfrom DiseaseControl Priorities3rd Edition.Lancet.2016;388:2811-24. Black RE, Levin C, Walker N, Chou D, Liu L, Temmerman M, et al. Reproductive, maternal, newborn, and child health: doi:10.1093/heapol/czr083 empirical literature and development of a conceptual framework. Health Policy Plan. 2012;27:541-54. Molyneux S, Atela M, Angwenyi V, Goodman C. Community accountability at peripheral health facilities: a review of the line:3388072 Rifkin SB, Muller F, Bichmann W. Primary health care: on measuring participation. Soc Sci Med. 1988;26:931-40. Victora C.Presentation. GlobalForumforHealthResearch; MexicoCity2004. line:25076672 2014;32:356-66. Nutr. Popul Health J defaulters. of tracking and completion immunization polio oral tine DW,Curry HB, Perry Tirmizi SN, Goldstein AL, Lynch MC. Assessing the effectiveness of house-to-house visits on rou 2016. Available: https://www.usaid.gov/sites/default/files/Final-AOTC-file-v2.pdf WHO. The Global Strategy for Women's, Children's and Adolescents' Health (2016-2030): Thrive, Survive, Transform. BLT.15.168492 tries with community health workers. Global Health. 2015;11:43. Mishra SR, Neupane D, Preen D, Kallestrup P, Perry HB. Mitigation of non-communicable diseases in developing coun Contraception. 2005;72:402-7.Medline:16307960 VahidniaN, Prata F, Dries-DaffnerM, Potts programs:distribution community-based Revisiting I. are needed? still they in Bangladesh.Lancet.1997;350:169-72.Medline:9250184 AM, Chowdhury S, Chowdhury Islam MN, Islam A, Vaughan JP. Control of tuberculosis by community health workers doi:10.1016/0277-9536(88)90413-3 doi:10.1016/S0140-6736(10)60142-0 415 doi:10.1016/j.contraception.2005.06.059
doi:10.1016/S0020-7292(03)00072-9 doi:10.1016/S0140-6736(96)11311-8 Medline:26555199 Medline:20071020 doi:10.1016/S0140-6736(01)05550-7 CBPHC andstrategiesusedbyeffectiveprojects doi:10.1007/s00268-005-0609-5 . Accessed:3April2017. June 2017 •Vol. 7No. 1•010906
doi:10.1186/s12992-015-0129-5 Medline:27152056
doi:10.1016/S0140- Medline:22279082 doi:10.2471/ Med- Med- Med------
VIEWPOINTSPaPErS PaPErSVIEWPOINTS June 2017 •Vol. 7No. 1•010907 Paul aFreeman rassekh [email protected] USa Baltimore MD21205 615 NorthWolfeSt. Public Health Johns HopkinsBloombergSchoolof r Henry Perry Correspondence to: 5 4 3 2 1 Henry BPerry with evidenceoflong–termmortalityimpact child health:7.sharedcharacteristicsofprojects health careinimprovingmaternal,neonataland the effectivenessofcommunity–basedprimary Comprehensive reviewoftheevidenceregarding USa of Washington,Seattle, Department ofGlobalHealth,University Washington, USa Independent consultant,Seattle, Medical Epidemiologist,Lusaka,Zambia Columbia, USa The WorldBank,Washington,Districtof Public Health,Baltimore,Maryland,USa Johns HopkinsBloombergSchoolof Department ofInternationalHealth, oom E8537 2
, SundeepGupta
1
, BahieM
4,5 3
, years or less. National demographic and health may surveys show long–term projects that undergo evaluation usually have a relatively short duration of five (MNCH) is an ideal that all MNCH programs seek. However, specially funded Sustainability of effectiveness in improving maternal, neonatal and child health and maternal deathsbytheyear2030. and maternal features will contribute to achieving the goal of ending preventable child these Strengthening health. child and neonatal maternal, improving in effectivenessimprove their might world aroundthe systems health how projects these providefor sharedfeaturesof guidance The Conclusions pital care. hos providefirst–level and capabilities strongreferral have all they and they have develop strong collaborations with the communities they serve, households, all with contact regular maintain who workers health nity vices, they have strong community–based programs that utilize commu ser planning family and health child maternal, comprehensive provide they are: these of notable most the Among characteristics. of number a to 4–year mortality, or under–5 mortality for at least 10 years. They share mortality,infant in reductions demonstrated have 1– all they and years, 30 than more for operating been all have projects four These Results in CommunityHealth(SEARCH)Gadchiroli, India. Research and Action Education, for Society the and India; Jamkhed, in ProjectComprehensiveHealth the Rural Haiti; Deschapelles, (CRHP) in Planning (MCH–FP) Project in Bangladesh; the Hôpital Albert Schweitzer Family and Health Child Maternal Matlab the analysis: this for qualified projectsFour years. 10 least at of period a for impact mortality of dence evi had that intervention single a than more with projects) as to ferred re (hereafter programs and project field studies, research of sessments as these of undertaken was search A series. this in elsewhere described in improving maternal, neonatal and child health has been assembled, as CBPHC of effectiveness the of assessments 700 of database A Methods 2030. vide guidance for ending preventable child and maternal deaths by the year approachesprothe projectscan and byevidence mented usedsuch with ing maternal, neonatal and child health. However, the interventions imple improvin (CBPHC) care health primary community–based integrated of effectiveness long–term the about evidence limited is There Background 416 www.jogh.org • doi:10.7189/jogh.07.010907 global journal of health ------www.jogh.org • doi:10.7189/jogh.07.010907 have evidence of long–term impact on maternal, neonatalorchildhealth. have evidenceoflong–termimpactonmaternal, of which this article is a part, and to describe the features of projects with more than one intervention that The purpose of the current paper is to review the database assembled for the current journal supplement, tality, resource–constrained settings. fectively improve the health of mothers, neonates and children at scale over the longer term in high–mor term effectiveness? The answers to such questions can be helpful in considering how CBPHC can most ef- any common implementation strategies that these programs have in common that might help to explain their long– health care activities have produced evidence of long–term impact on MNCH? on impact long–term of producedevidence carehave health activities is: address can review this However,question important one over aperiodoflessthanfiveyears. child health projects included in our review assessed four or fewer that interventions were implemented neonatal/ of assessments 548 the of Three–fourths years. more or five for assessed were projects the of maternal health assessments in our review were of projects with more than five interventions, only 15.8% 152 the of (66.7%) two–thirds Although interventions. selected of group smaller a of assessments term ness of community–based primary health care (CBPHC) in improving MNCH is based primarily on short– effective regarding the evidence the articles, of series this in seen have we difficult. As improvementsis those for responsible factors programmatic the determining but health, child in improvements national where in this series [ series this in where The database of assessments of the effectiveness of CBPHC in improving MNCH has been described else METHODS plemented onlyoneintervention. from this analysis primarily because no measure of changes in mortality were available or the project im in shown As teria forthisstudythereasons showninFigure. 1 one assessments in the neonatal/child health database were identified. Of these, 14 did not meet the cri the maternal health database were identified, and none of these met the criteria for this analysis. Twenty– in assessments Three more. or years 10 of duration a had that projects) as to referred(hereafter studies morbidity,serious mortality.in in programs/projects/status, or for nutritional database Wethis queried in interventions, survival child moreevidence–based or one of coverage population in changes of terms facility. Each assessment consisted of measurements of changes in maternal, neonatal and child health in tiveness of one or more that interventions have been implemented in the community outside of a health , 17 assessments of projects having a duration of 10 years or more were excluded excluded were more or years 10 of duration a having projects of assessments 17 Table1, 1 ]. In short, it consists of data extracted from 700 documents describing the effec the describing documents 700 from extracted data of consists it short, In ]. 417
CBPHC andprojectswithevidenceoflong–termmortalityimpact What packages of community–based primary primary community–based of packages What June 2017 •Vol. 7No. 1•010907 A related question is: question related A primary healthcareprimary (CBPHC). integrated community-based neonatal/child healththrough or impact onmaternal with long-termevidenceof Figure 1. Selection ofprojects Arethere ------
VIEWPOINTSPaPErS PaPErSVIEWPOINTS June 2017 •Vol. 7No. 1•010907 Perry etal. *Vitamin Ainoneassessment,malariacontrol inoneassessment,andconditionalcashtransfersanotherassessment. sions (2004and2006). project onfourseparateoccasions(1998,2004,2006and2009)theSEARCHproject on twoocca fromHaiti Deschapelles, in Schweitzer Albert Hôpital the of CEO CRHP the visited has He 1999–2003. General/ Director as served He occasions. of number a on site field Matlab the visited and 1995–1999 munications with persons engaged in these projects. By coincidence, Dr Perry worked in Bangladesh from personal experience and field visits of one of the authors (HP) to these projects along with personal com Additional literature on these projects was reviewed, and additional findings were incorporated based on • • • • Table 1. in 1977, and in 1978 the Government of Bangladesh established the International Centre for Diarrheal for Centre International the established Bangladesh of Government the 1978 in and 1977, in socio–economic development. The CRL expanded its work to maternal/child health and family planning and population nutrition, projects research health, regarding field of hundreds of site the is Matlab and on morbidity and mortality. The DSS has become the oldest demographic surveillance system in the world, lance System (DSS) was established at Matlab with the initial goal of assessing the impact of new vaccines ease, including the testing of the effectiveness of new cholera vaccines. In 1966, a Demographic Surveil three hours southeast of Dhaka in a cholera–endemic area to test new approaches for controlling the dis area riverine rural a in site field a established CRL the 1963, In cholera. treating and preventing for ods The Cholera Research Laboratory (CRL) was established in 1960 in Dhaka, Bangladesh to develop meth Project description The IcddrbMCH–FPproject inMatlab(Bangladesh) rent activities. in which the mortality impacts were achieved even though they may not entirely accurately describe cur time the for appropriate are descriptions the Nonetheless, static. remained not have features these time of period long a such over that recognizingprojects, these of features main the below describe we Here • • • • erable evidence of improvements in coverage of key evidence–based These interventions. four projects are: have each been functioning for 30 years or longer and are still functioning. These projects also had consid years 10 for impact mortality under–5 of evidence had that database fromprojects our identified four The r The remaining sevenassessments[2 No measure ofmortalityincluded No baselinemeasure ofmortality Total No comparisonarea No evidenceormortalityimpact implemented Only 1intervention Mortality impactdatacovered lessthan10yearsofprogramming r SEARCH(SocietyforEducation,ActionandResearch inCommunityHealth)Gadchiroli, India. TheJamkhedComprehensive HealthProject inJamkhed,India;and, TheHôpitalAlbertSchweitzerinDeschapelles,Haiti(whichoperatesaCBPHCprogram); TheicddrbMCH–FPproject inMatlab,Bangladesh; SEARCH(SocietyforEducation,ActionandResearch inCommunityHealth)Gadchiroli, India. TheJamkhedComprehensive HealthProject inJamkhed,India;and, TheHôpitalAlbertSchweitzerinDeschapelles,Haiti; eason ESULTS Bangladesh); search field site for icddrb, formerly known as the International Centre for Diarrhoeal Disease Research, re- planning family and health maternal/child (a Bangladesh Matlab, in project MCH–FP icddrb The
For Reasons forexclusionofassessmentsprojects of10ormore years’duration
exclusIon
– 8 ] that qualified for the analysis concerned fourprojects:] thatqualifiedfortheanalysisconcerned 418
Maternal database Neonatal/child health Neonatal/childhealth database Maternal p roJects
1 2 3 oF 10– year www.jogh.org
duratIon
or • doi:10.7189/jogh.07.010907
more
excluded database 14 3* 5 2 1 1 2
From
analysIs ------www.jogh.org • doi:10.7189/jogh.07.010907 tion Area was 97%compared to85% intheComparisonArea. ment health services are provided. Each of these two areas has a population of approximately 112 approximately of population a has areas two these of Each provided. are services health ment govern only whereArea, Comparison a is second project.The (MCH–FP) Planning Family and Health based health and family planning activities operated by icddrb began in 1977. This is the Maternal–Child community– intensified where Area, Intervention an is first The parts. two into divided is site field The activities [9 Disease Research, Bangladesh (now icddrb), which took over responsibility for the Matlab DSS and field proximately 20% nationwide [ nationwide proximately20% ap of rate national a to compared Area Intervention the in 69% was immunizations childhood of ries 71%, 47% in the Comparison Area, and 58% nationwide. In 1987, the coverage rate for the standard se was Area Intervention the in CPR the 2005, In nationwide. 19% and Area Comparison the in 16% was it while compared 46% was Area Intervention the in (CPR) rate prevalence contraceptive the 1984, In outcomes Long–term erations andresearch findingsthrough theearly1990s[10]. book The • • • • • Key componentsforsuccessatMatlabinclude: districthospitalinMatlab[12]. with thegovernment collaboration in provided is section, caesarian including facility,care, Matlab obstetrical emergency but icddrb the at available not are proceduressurgical Major Matlab. in also is area geographic larger a ing operates a 50–bed inpatient facility that serves the Intervention Area. A government district hospital serv also Icddrb infections. tract reproductive and diseases transmitted sexually of treatment and pregnant), (suction curettage of the uterus for women with delayed menstrual periods who do not want to become regulation menstrual IUDs, of insertion including paramedics, the by provided is care health primary comprehensive Basic supplies. of replenishment and education, continuing supervision, for weeks two every sub–center the at meet CHWs works. paramedic full–time a sub–center,where each to assigned 28 about for Area(one Intervention the in arelocated sub–centers Four the healthproject (excludingresearch–related expenses)isaboutUS$5[ nature of the field research activities. The total annual cost per capita for the community–based portion of project managers promptly address any issues raised by the community about the quality of services or the decades. Maintaining good relations with the community is a priority for the Matlab MCH–FP project, and a high level of trust with the population because of the high quality of health care it has provided over four ally by boat in this riverine environment, to the hospital in Matlab operated by the CRL. The project earned of patients with cholera depended on activity promptsince the survival identification and transport, usu CRL initial the of element key a was care hospital available readily and system referral This referrals. for child health care as well as reproductive health care. A hospital operated by the project is readily available refer patients to a nearby sub–center staffed by a full–time paramedic who provides routine and maternal can they and well–supported, and well–trained are project MCH–FP icddrb the in working CHWs The ister injectablecontraceptivesandtrackpregnancies. admin condoms, Finally,arrhea.and controlpills birth distribute planning, promotefamily CHWs the holder,” who is a mother in the neighborhood with additional training in the treatment of childhood di education and treat diarrheal disease. They also leave packs of oral rehydration salts (ORS) with a “depot nutrition provide They guidelines. WHO to according pneumonia childhood treat and care, postnatal and antenatal provide children, and women immunize CHWs the visit, home a of day.time per the At homes 20 visits typically and households 200 approximately for responsible is CHW Each present.) at basis. (The frequency of visits has declined gradually from every two weeks in 1977 to every two months regular a on home each visit Area Intervention the in (CHWs) workers health community paid Eighty persons. Continuouslyavailablesupplies. Awell–developedrecord–keeping system; and Astrong systemofaccompanimentandsupport foralllevelsofworkers; Soundorganizational structure from theoutset; tient referral totheMatlabHospital; Readily available transport throughout the project area, mostly by speedboat, which has facilitated pa Matlab: Women,Matlab: ChildrenHealth and – 11]. ]. In 2005, the childhood immunization coverage rate in the Interven the in rate coverage immunization childhood the 2005, In 12]. 419 provides a full discussion of the history of Matlab, its op its Matlab, of history the of discussion full providesa
CBPHC andprojectswithevidenceoflong–termmortalityimpact June 2017 •Vol. 7No. 1•010907
12]. 000 people), and 20 CHWs are CHWs 20 and people), 000
000 000 ------
VIEWPOINTSPaPErS PaPErSVIEWPOINTS June 2017 •Vol. 7No. 1•010907 Perry etal. • Two atMatlab andreported in1994bearemphasishere: lessonslearned Lessons learned [ less) (25% Area Comparison the in 62.4 and area intervention the in 46.6 was rate tality approximately 120 and 75 respectively (38% less in the Intervention Area). In 2005, the under–five mor imately 200 in the Comparison Area and 150 in the Intervention Area (25% less). In 1995, the rates were has been widely documented and applauded [ [ 2015 by dren chil for Goal Development Millennium the reached that countries high–mortality 68 of out 19 only of throughout– country.services the FP and MCH both – services home–based tending one is Bangladesh ex of job masterful country,a the done for has CBPHC Bangladesh of and model a as served Matlab at parison Areas have narrowed over time. This can partly be explained by the fact that the MCH–FP Project rowed later. The differences in mortality rates for infants and nar children gradually between Areas the Comparison Intervention and and Intervention Com the between difference the how and declined Area Comparison the in mortality child from quickly findings how striking is the example of Matlab One the • vention Area thecorresponding figure was50%[14, In 2004, only 12% of the births in the Comparison Area were taking place in a facility while in the Inter Area. Comparison the in slower was birth giving for facilities of use program.the Progressincreasing in bution of birth control pills and condoms. By the mid–1980s, Bangladesh essentially had a national CHW promoteto workers paid by visits home of distri model the and planning ily family of use the progressThe Controlthe in fam Matlab the Areaof application national the to part in attributed be can rates were essentiallythesameat2.7and2.8,respectively [ vention Area and 6.0 in the Comparison Area; in 1995, they were 3.0 and 3.6 respectively; by 2005, the Inter the in 4.5 were TFRs the 1985, in following: the been has time over (TFR) rate fertility total The carriages andstillbirths[16]. Area, mainly as a result of a lower pregnancy rate and lower case–fatality rates for induced abortion, mis 100 per deaths maternal of number the is, (that rate mortality maternal the 2005, to 1982 from period the strategy nationally, withobviousbenefits. were proven tobeeffective inthe 1970s,there wasanexpliciteffort inthe1980stointroduce thissame based at the 190–bed hospital. In its second decade (in 1967), it established a project of community– of project a established it 1967), (in decade second its In hospital. 190–bed the at based provided HAS existence, its of decade first the For clinic outpatient an at care hospital only services and hours northwestofcapital,Port–au–Prince [20]. of West Africa. The Mellons constructed one of Haiti's first modern hospitals in the Artibonite Valley, three Schweitzer who, for more than a half–century, provided medical care in Gabon, an underserved country Albert missionary medical great the of example the by inspired were Mellon, Grant Gwen and Larimer William couple, American wealthy a after 1956 in operations L’Hôpitalbegan (HAS) Schweitzer Albert Project description Hôpital AlbertSchweitzer(Deschapelles,Haiti) 1978 and 1994 [ between period 15–year a over Area) Control (the area services government the in than lower sistently The infant and 1– to 4–year mortality rates Areafor the Intervention of MCH–FP project area were con riod, lifeexpectancyincreased from 50toaround 65years[10]. 75% in the annual number of childhood deaths over a 25–year period in Matlab, and over a 40-year–pe 54% lower in the Intervention Area than in the Comparison Area [ Between 1988 and 1993, the mortality rate from pneumonia in children younger than 2 years of age was
quality health services intoafamilyplanningprogrammequality healthservices justifytheheavyinputs”[ integrating of benefits [T]he children…. and women of those particularly problems, health other to “Family planning field workers are more likely to gain the confidence of the community if they respond supplies, andrelations withthecommunity”[17]. el of organization in the hands of a competent manager. This applies for staff management, logistics and “The successful operation of such a large and multifaceted project as Matlab requires a professional lev 000 women of reproductive age) was 37% lower in the Intervention Area than in the Comparison Comparison the in than Area Intervention the in lower 37% was age) reproductive of women 000 ], and its national achievements in expanding coverage of community–based services services community–based of coverage expanding in achievements national its and 18],
14, 15]. In 1985, the under–five mortality rate (U5MR) per 1000 live births was approx 420 19]. After the interventions of the Matlab MCH–FP Project 15]. 14, 15]. 13]. There was a reduction by around www.jogh.org • doi:10.7189/jogh.07.010907 17]. 14, ]. Over Over 15]. ------www.jogh.org • doi:10.7189/jogh.07.010907 in which it is located, with 258 with located, is it which in ters opened. The hospital always as served the of Ministry Health’s district hospital for the health district cen health two and posts health seven time, Over facilities. care health primary peripheral any without based primary health care based on community health workers ( years, from 18 the over fluctuated HAS’sprojectcare by health served primary population The assessment. impact the wives, along with supervising and training the [ care of levels and Mobile Clinics, to promote community involvement, and to assist with referral of patients to higher Posts Rally the with assist to women, other to education health peer–to–peer provide to recruited were In the late 1990s, 1500 volunteer community health workers ( and torefer patientswhenneeded. planning) family (including care preventive and curative basic provide to mountains) the in roads few staffed by an auxiliary nurse who, every 1–2 months, visits isolated communities on foot (since there are intermittently.areas isolated are all These reach clinics Mobile hospital. the at care referral and motion, tend Rally Posts where essential services are provided, including immunizations, growth monitoring/pro at to children and mothers mobilize and events, vital register education, health basic provide to home The CBPHC services at HAS have relied on paid Health Agents ( with facility–based primary healthcare,with facility–basedprimary hospitalreferral care andcommunitydevelopmentactivities[ became a comprehensive integrated health and development system with strong CBPHC services together primary education, and promotion of animal husbandry and improved agricultural production. HAS thus providing opportunities for micro–credit and income–generation for women, literacy training, support for reforestation, and gardens vegetable promoting levels, household and village the at sanitation and water improving for projects including activities, development community established also HAS 1960s, the In at HAS include 16 community–based tuberculosis workers (seven numerousin implemented been has and HAS at Finally,originated countries. other services CBPHC the supervising the community–based nutritional rehabilitation project, known as the Hearth Project, which over a three–decade period from 1970 to 1999 [ 1999 to 1970 from period three–decade a over nationally Haiti in than lower much remained area service care health primary HAS the in U5MR The whole [4 data for Haiti as a whole. The HAS project area is similar in socio–economic indicators to rural Haiti as a to these compare to and HAS by served area project care health primary the for U5MR the monitor to ect in 1967 [21] and the intermittent collection of retrospective birth histories since, it has been possible HAS’sof initiation of time the at data events vital of collection the projresultof health a community As the HASproject area, someofwhichrequire eight–hoursbyfoottoreach. [3 15.4%) vs (27.5% areas other in that double nearly was areaproject HAS the in CPR the time, that at Likewise tionally, the U5MR in the HAS service area was less than half of that for Haiti overall (62.3 vs 149.4) [ Addi Haiti. rural in nationwide interventions same the for those approximatelytwice werearea service care health primary HAS the in interventions survival child key of rates coverage population 2000, In outcomes Long–term from throughout Haitihavecomethere fortreatment. is widely regarded as one of the best district hospitals in a rural area of a developing country, and patients hospital The time. of period long a over population the of support and trust the earned has services cal logistical support for its field projects and to provide needed supplies and drugs. The quality of its clini helped HAS to provide high–quality professional leadership and management. It has been able to ensure Steady financial support from external donors has been available to HAS since its inception, and this has now provided aswell. is HIV/AIDS with patients for anti–retroviralprovisionmedication of Community–based home. their in obtain sputum specimens from symptomatic patients and provide directly therapy observed for patients U5MR remained lessthanhalfthatof theU5MRforHaiti[ [ age reproductive of women all of immunization through tetanus quarter of the national level between 1958 and 1973 was due in large part to the elimination of neonatal ]. ]. Great efforts have been made to ensure access to basic services in the most isolated parts of of parts isolated most the in services basic to access ensure to made been effortsGreathave ]. 3 820 in1958to180 ]. In addition, eight Monitors ( Monitors eight addition, In ]. 000 people in its catchment area during most of the period covered by by covered period the of most during area catchment its in people 000 000 in1996andto350 421 Animatrices. The role of
) provide liaison with and training of lay mid lay of training and with liaison provide Monitrices) 4 ]. The rapid decline in under–five mortality to one– to mortality under–five in decline rapid The ]. CBPHC andprojectswithevidenceoflong–termmortalityimpact 4 000 in2016[3 ]. Animatrices), one for every 15 households, Agents de Sante agents de santé) and mobile health teams Accompagnateurs and nine 22, Monitrices at HAS initially involved ]. Between 1970 and 1999, the the 1999, and 1970 Between 23]. June 2017 •Vol. 7No. 1•010907 , 20]. ) who regularly visit every Agents) who 3 3 ]. ]. ]. ------
VIEWPOINTSPaPErS PaPErSVIEWPOINTS June 2017 •Vol. 7No. 1•010907 Perry etal. dia, has been in operation for almost five decades [ In- Maharashtra, of District Ahmednagar in Project(CRHP) Health Rural Comprehensive Jamkhed The Project description India: theJamkhedComprehensive RuralHealthProject for therest ofthecountry. rolemodel a as position its and HAS at experience early the of because part in possible been have Haiti in survival child in achieved gains the to CBPHC of contribution nationwide The Haiti. of population the of two–thirds to services survival child community–based provide NGOs country,these the and in el of CBPHC developed at HAS is now utilized by virtually all other NGOs working in community health the Millennium Development Goals in 2015 for reduction in child and maternal mortality [ the 74 so–called Countdown Countries (with 97% of world’s maternal and child deaths) to have achieved of 34 only of one is Haiti outbreaks, cholera recentmore and 2010 in capital the in earthquake astating throughoutHIV/AIDS of demic country,the dev the of spite in fact, In 1980s. early the in began which particularly in light of the country’s political instability, its deteriorating economic situation, and the epi noting, worth is fromperiod 1970–1999, the during whole a as Haiti for mortality of decline rapid The cilities andthehospitalare keyelementsofsystemeffectiveness. fa care health primary the with activities CBPHC the of integration close The impact. mortality this to contributed likely most – elements these between interactions the as well projects,as development nity based services, primary health care services at health posts and health centers, hospital services, commu community– – development and health of system entire Rather,the reduction. mortality sustained the for responsible interventions of set small a even or intervention single any be to appear not does There Lessons learned live savedwasUS$47;andthecostperDALY savedwasUS$ 90[ past decade. The cost per under–five death averted in current dollars was US$ 3233; the cost per year of the over cutbacks significant undergone have HAS at projects the constraints, resource of Because lars. dol 2016 in 24.77 US$ be would 1999 in existed it projectas entire the for cost annual capita per The pation. from communities partici other to benefits the saw they as involved be to areasought the in villages all Gradually,participation. to themselves committed and assistance requested that communities with only worked CRHP beginning, Fromthe communities. of members disenfranchised most and poorest the of and community empowerment, micro–credit, education, improved agriculture, and prioritizing the needs ill–health through improving access to water and food, nutrition education and kitchen gardens, women’s of determinants the address CHWs, illiterate train to expanded gradually work Their care. curative for prenatal care, complementary infant feeding, ensuring safe delivery, family planning, and a health center Some of the initial activities carried out were: health promotion through health education, immunization, aboutandaddressing theirproblemsticipation bylearning basedontheirownpriorities. par active their through health their improve could communities whereby process a facilitate to came they had a modest amount of additional knowledge and skills. The main purpose of their work soon be ters of health problems could be addressed at the community level, mainly by the villagers themselves, if problems.three–quarmedical realizedover with quickly that them They to came treatingwho patients team physician husband–wife a as 1970 in Jamkhed in working started Arole Mabelle and Rajanikant the scarcity offoodandthelackworkinJamkhedarea. tion was migrating to sugar cane plantations outside of the district to work in jobs temporary because of inhumanely.treated popula often werethe they of One–thirdFurthermore, rights. personal no had en was ingrained, and harmful traditional practices, especially for women, were common. In addition, wom system caste The births. live 1000 per 176 was rate mortality infant the and 1%; werethan vider less all rates of childhood immunizations, family planning, prenatal care, and birth attendance by a trained pro coverage and 40%, was malnutrition childhood of prevalence water.The to access of lack and drought When CRHP began in 1970, the people of the Jamkhed area were living in near–famine conditions from of equity, integration and empowerment have been guiding principles throughout this prolonged period. munity–based health programming in conjunction with first–level hospital referral services. Its principles
422 5 , 26]. It developed a comprehensive approach to com 24]. www.jogh.org • doi:10.7189/jogh.07.010907 25 ]. The mod ------www.jogh.org • doi:10.7189/jogh.07.010907 workers (VHWs), farmers’ clubs and women’s groups ( community, health the village within including volunteers of tainability.groups established project The community,sus futurea ensuresCRHP’semerged with of that processpartnership the into so entering CRHP always insists on major investments of time and energy from community members as a condition 8 in 2011, according to data collected at CRHP by CHWs [ [ 1993 in 19 to 1971 in births live 1000 per fromdeaths declined 176 Jamkhed CRHP at IMR The declined from 1800 to200casesper100 tuberculosishas of incidence the and disappeared, virtually project,has the of start the at common was Leprosy,since. ever which maintained been has undernutrition of prevalence low This measurements. childrenaccordingof 87% werewere5% only and immunized undernourished fully anthropometric to 2011, it reached 99%. The percentage of couples utilizing family planning reached 68% in 2004. In 2004, By 1993, the percentage of pregnant women with antenatal care and a safe delivery reached 82% and, in outcomes Long–term segments ofthecommunityparticipate. techniques are commonly used for assessments and analysis as well as for discussions on what to do. All point fordiscussionsaboutprioritiesthecommunity to address.RuralAppraisal(PRA) Participatory focal a as services and village the in space public a boardin a displayed on is Thiswritten that is information diseases. priority various and issues, environmental and agricultural conditions, economic ly immunized, and the number of children with malnutrition. Also included is information about socio cords of the number of eligible couples who are using family planning, the number of children complete- Each village maintains a record of all births and deaths that take place among its members, as well as re and tovisitvillagesseetheimpactfirsthand. people from over 100 countries have come there to learn from the VHWs, other villagers and CRHP staff 30 than More 1992. in established was ment ence, the Jamkhed International Institute for Training and Research in Community Health and Develop experi CRHP the about learn to world the and throughoutIndia people of interestgreat the of Because five decades,soCRHPnowfocusesonthevillagesthatneedthemmost. lages are now independent, thanks to the sustainable development process that CRHP has nurtured over 500 of population a with villages reach300 to expanded gradually CRHP patients are rarely seenthere now. hospital in Jamkhed was filled with children who had life–threatening infections and malnutrition. Such the project, the of beginning the At there. performed proceduresaresurgical emergency other and tion project area and beyond. A larger 50–bed hospital has recently been completed. Emergency cesarean sec fromthe patients sourcefor referral a as served that hospital 30–bed a operated CRHP years, many For VHWs have been working for more than 30 years. Dropouts are rare, mainly because of old age and death. other.each for the support of providesocial Many and there night the fromspend staff.and They other the other VHWs to discuss problems encountered in their work and to obtain further training from each The VHWs come to the project center in the small town of Jamkhed once a week. There they meet with fromlearning eachother, andadditionaltraining. specialist and a social worker, though they all become multipurpose workers through working together, agricultural an nurse, a of consists team mobile The needed. if often more or month a once village each visits which project’steam, the mobile and community the between link a as serve They activities. ing pay,for work not do VHWs the Although income–generat to access obtain projectthey with assistance ered from anillness(suchastuberculosis) asaresult ofcare provided byCRHP. recov- had who caste (dalit) untouchable the from women illiterate Initially,were VHWs these of many CRHP.of assistance the with undertaken be might marginalizedthat and poor the with especially tions, munity’s assessment of problems and resources, analysis of causes and determinants, and appropriate ac her knowledge with everyone in the community, to organize community groups, and to facilitate the com nity development, communication, organization and personal development. Her primary role is to share commu health, in training receives leprosy).She as (such conditions stigmatized with those and ables) marginalizedpoorestmost the and as especially such village, members her assist to ger The key change agent in the community became the VHW, who is selected by the community. She is ea olescent boys’groups. 000 persons[26]. 423
000 people from throughout India and more than 3000 3000 than more and India throughout from people 000 CBPHC andprojectswithevidenceoflong–termmortalityimpact mahila mandals ), and, more recently, girls’ and ad 26]. In 1971, the IMR at CRHP Jamkhed was June 2017 •Vol. 7No. 1•010907
000 people. Most of these vil these of Most people. 000 (untouch- Dalits 5 ] to to ] ------
VIEWPOINTSPaPErS PaPErSVIEWPOINTS June 2017 •Vol. 7No. 1•010907 Perry etal. CRHP Jamkhed has been half that for rural Maharashtra [ Maharashtra rural for that half been has Jamkhed CRHP at IMR the 1980 since but 110), vs (176 Maharashtra of state the of area rural the for than greater 60% • • • • • The Jamkhedprocess ofsustainedhealthimprovement through CBPHCinvolves: edge, skillsorresources, CRHPhelpsthem. the community learns to work together and solve problems together. If the community needs more knowl process, Jamkhed the In issues. own its address to capacity the and leadership good have must munity dency on outside agencies. In order for the development process to be sustainable by the people, the com it depen- increases and reduces empowerment – self–reliance,Community self–confidence self–esteem, way, whichhavemadethesehealthimprovements possible. qualitative a in communities and persons of transformation the is important more Even important. are that dignity,changes God–given quantitative have the capacity.they only and that not worth is it Thus, affirming villages, their in positions society,leadership of taking mainstream the outside once women, and men self–confident are statistics these Behind decades. four than more over achieved results tive posi show which statistics, health in throughchanges demonstrated is ToCRHP of impact the sure, be Lessons learned gardens foradditional,nutritiousfruitsandvegetables. kitchen have now homes Most use. agricultural and home for groundwater of available the increase to projects development watershed established program.They the for growingfood for land their of some dedicated Clubs Farmers’ the and pulses), or grains, water, salt, firewood, (eg, something contributing everyone with children, of groups for programs feeding organized communities the years early the In and communitygroups. developed. The government now provides these services with the support and cooperation of the VHWs program the as communities the by accepted gradually also were Immunizations monitoring. growth women’sand men for childrenThe the foods. groupsweigh weaning nutritional and feeding, after ing dice). VHWs ensure exclusive breastfeeding for infants during their first 6 months of life, proper burp jaun neonatal for sunlight fever,and for water cool with sponging problems, respiratory for halation home care forcommonproblems (suchashomemadeoralrehydration solutionfordiarrhea, steamin provideearly to how care,prenatal practices, nutritional healthy of importance the know members ily ment together with their understanding of health promotion and disease prevention. For example, fam empower and participation communities’ the of because accomplished were results significant These Maharashtra StateinIndia,were theCRHPislocated[ in district Pune for mortality perinatal and maternal respectivelythe lower than 20.3% wereand 27.8% natal mortality rate of 36.0 per 1000 live births and stillbirths were measured at CRHP [ Jamkhed between 1996 and mortality 1999. ratio A of maternal 70.0 per 100 began, these rates were measured following a careful review of all births and deaths in 25 villages around Although baseline levels of maternal and perinatal mortality were not measured in the 1970s when CRHP lages compared tocontrol villages[28]. onstrated a 30% reduction in the risk of death among children 1–59 months of age in CRHP project vil project villages with those in a surrounding control area was carried out in 2007–8. This in evaluation histories dem birth from findings of comparison a on based CRHP of impact mortality the of evaluation
nity’s knowledge,skills andinterests. commu the of context the in accessible are that resources local and technology appropriate Utilizing motion, prevention, early detection,treatment, andrehabilitation inthe community. prohealth as well as medicines herbal and remedies home including activities low–cost on Focusing engaging intheseactivities. to committed individuals caring and motivated of community a building by volunteerism Promoting analysis, andaction. as active members of the community to solve the problems that concern them most through assessment, ing the whole community, including the poorest and most marginalized members and integrating them Developing a caring and sharing community that promotes reconciliation and peace ( based onwhere theyare andwhattheyhave. communities, and individuals of capabilities the building through skills and knowledge Expanding
424
29]. 27 ]. A large–scale external and independent independent and external large–scale A ]. www.jogh.org 000 live births and a peri- • doi:10.7189/jogh.07.010907 shalom) by engag ]. These rates 29]. These rates ------www.jogh.org • doi:10.7189/jogh.07.010907 sive informationaboutthesebenchmarksintheacclaimedbookbyAroles [ text within which CBPHC operates and how it is actually implemented in the community, there is exten acclaimed book by the Aroles, simply entitled simply Aroles, the by book acclaimed the of publication 1994 the through as well as world the around and India throughout from people of research circles, it is well known in the broader global health community through the visits of thousands women’s empowerment and community participation. In spite of not being well–known in academic and empowerment, community throughworld’s MNCH improvingthe of of examples one leading is CRHP contribute. to all to opportunity an giving and everyone of worth and dignity the embracing by transformed are Lives disease. and poverty from everyone uplifts thereby and change social about brings that community each within from innovation an rather but technology in innovation an not people’sis a This it movement. ing This transformativeprocess isspread toothercommunitiesbythevillagerswhohaveexperienced it, mak • • India, known as Gadchiroli [ Gadchiroli as known India, vided community–based health care services and hospital care in a rural area of the state of Maharashtra, pro has (SEARCH) Health Community in Research and Action, Education, for Society the 1986, Since Project description in Gadchiroli, India SEARCH (SocietyforEducation,ActionandResearch inCommunityHealth) [ Conference the to the prior by years published Organization several monograph Health fluential World tional Conference on Primary Health Care at Alma–Ata. CRHP was one of the projects featured in the in Of historical importance is the fact that the CRHP served as one of the inspirations for the 1978 Interna ber ofdifferent languages. book is one of the best long–term sellers among global health books and has been translated into a num project establishedincollaborationwiththeJohnsHopkinsUniversity1970s[ Bangs obtained important insights for their work from the Narangwal Project, a model community health [ policies health inform to research for and health, in training and education care, health basic Gadchirolifor of communities the with ship philosophy.social Gandhian of context the in work their lished partner collaborative a developed They The founders, Dr Abhay Bang and Dr Rani Bang, were inspired by the life of Mahatma Gandhi and estab partofMaharashtra. pur inthemostwestern other half is composed predominantly of Hindu subsistence farmers. Gadchiroli is 175 km south of Nag trict is largely forested, and half of the inhabitants are indigenous tribal people who live in the forest. The of this area is used as a field site for implementing new interventions while the other half serves as a con project that provides community–based primary health care for a population of 80 Over the past 30 years, Drs. Abhay and Rani Bang and their dedicated staff developed a community health SEARCH. primary health care, community participatory research and training of village people are core activities at [ publications” and training of way by others to available knowledge this health care to meet these health needs, to test these models by way of research studies, and then to make of models community–empowering develop needs, health their identify to communities marginalized name, “Society for Education, Action and Research in Community Health.” The mission is “to work with its in expressed is SEARCH of mission The dependence. from as well as disease from freedom achieve them helping thereby health, own their of charge take to communities and individuals empowering by sion of SEARCH is the realization of vi The policies. health shape researchto conduct (3) and health, in education and training provide (2) population, local the to careprovideorganization: health (1) their for goals three established Bangs The 30
they are still active leaders, still learning andsharingwiththeircommunitiesothers. they are stillactiveleaders,learning service of aredecades who after VHWs, that women motivated supporting so and training Recruiting, conditions outsideofthehealthsectorhavemore impactonhealththancurativecare alone. community, that the recognizing of well–being overall the improving of purpose the for causing are well as building social capital and helping people to recognize the harm that some traditional practices a generation, income and sanitation education, including development, multi–sectoral in Engaging , 31 ]. In contrast to the limited information in most peer–reviewed scientific articles regarding the con 32, ]. The Gadchiroli District is the least developed in the state. The dis The state. the in developed least the is District Gadchiroli The 33]. Aarogya–Swaraj 34 ]. Like the Aroles, who founded the Jamkhed CRHP Project, the the Project, CRHP Jamkhed the founded who Aroles, the Like ]. 425 Jamkhed: A Comprehensive Rural Health Project Health Rural Comprehensive A Jamkhed: (translated as “the people's health in people's hands”) CBPHC andprojectswithevidenceoflong–termmortalityimpact June 2017 •Vol. 7No. 1•010907 ]. Thus, community–based community–based Thus, 32]. 5 ].
000 people. One–half 35, 36]. [ 5 ]. This This ]. s ------
VIEWPOINTSPaPErS PaPErSVIEWPOINTS June 2017 •Vol. 7No. 1•010907 Perry etal. more thanthree decades. have been in operation for more than four decades, while the fourth (SEARCH) has been in operation for projectsthese of four the Threeof resulting frommortality interventions. child community–based or tal among the few examples that exist of projects with evidence of long-term reductions in maternal, neona are India rural in project SEARCH Gadchiroli the and India, rural in CRHP Jamkhed the Haiti, rural in The Matlab MCH–FP project in rural Bangladesh, the HAS integrated project of health and development DISCUSSION spread application. wide and analysis closer merit SEARCH by used CHWs of support and training selection, of methods even if they are illiterate, can provide high–quality technical interventions for mothers and children. The Among other things, their work has demonstrated that properly trained, supervised and supported CHWs, around India. sites various at Research Medical of Council India the by tested being now care neonatal home–based of Maharashtra, and they provided technical assistance for scaled–up versions of the SEARCH model for vided leadership for replication of the home–based neonatal care intervention by other NGOs in the state efficacythe were first interventions [ these 1990s of the in Bangs the by reported care by SEARCH in Gadchiroli have stimulated much additional work by others around the world since The pioneering findings of the community case management of pneumonia and of home–based neonatal Lessons learned reported (1994–2004), the IMR remained essentially unchanged [ in 1988 to 31 in 2003 [ The infant mortality rate in the Intervention Area declined by 74%, from 120 deaths per 1000 live births outcomes Long–term treatment ofchildhoodpneumonia,andprovision ofhome–basedneonatalcare. tendants and community health workers for diagnosis and treatment of common illnesses, diagnosis and health care programs throughout the developing world. The project relies on trained traditional birth at [ pneumonia Bang’s childhood The of management community–case effectivenessof groundbreakingthe research on traditional villagemidwives(dais). the for support and training strong provided also SEARCH 2005, and 1988 Between children. and ers pneumonia treatment and home–based neonatal care along with other basic health care services for childhood providemoth to able is CHW the responsible, is she which for households the with contact close vious visit,andprovides healtheducationandbasicpreventive andcurativehealthcare. Bymaintaining pre the since deaths and births pregnancies,registers basis, monthly a on home every visits CHW This one female community health worker (CHW) for approximatelyand supervises 1000 population. every health provision system that utilizes the government health system for referrals. SEARCH employs, trains SEARCH does not duplicate the government health system. Instead, it has developed a community–based tion andevaluationofitsprojects. Thecommunityhastakenco–ownershipoftheproject. responding to their expressed concerns and priorities, and involving them in the planning, implementa SEARCH established a partnership with communities over the past two decades by listening to members, staff. consists of 30 members, including physicians, paramedics, project supervisors and managers, and research away.minutes 30 about Gadchiroli,is of which city the in hospital government a to staffSEARCH The at – massive hydrocele caused by lymphatic filariasis. Patients requiring higher levels of care are transported disability long–term of cause common a for caresurgical and sections cesarean including hospital, the at which was initiated in response to requests from the community [ addiction, drug and addressalcohol to collaboration community pioneering a research.developed also It reproductivecommunity–based a pioneeredof SEARCH development the carehealth project related and tribalpeoplefromSEARCH alsooperatesa20–bedhospitalandoutpatientfacilitytoserve thearea. period. therestudy implemented the being during not is trolintervention new the that areasense the in the home–basedneonatalcarewasbeingimplementedandevaluated. intervention 38 ] and on the effectiveness of home–based neonatal care [ care neonatal home–based of effectiveness the on and ]
2 , 39]. In the Comparison Area, over the period of time for which data have been 426 2 37]. Basic surgical services are provided , 39]. This was the period during which www.jogh.org 2 ] has had a major impact on on impact major a had has ] • doi:10.7189/jogh.07.010907 38, ]. The Bangs proBangs The 40]. ------www.jogh.org • doi:10.7189/jogh.07.010907 of the first to carefully evaluate the effectiveness of community–based primary health care [ care health primary community–based of effectiveness the evaluate carefully to first the of wal Project, a pioneering field project in north India during the late 1960s and early 1970s that was one Carl Taylor. We noted previously that both CRHP and SEARCH were directly influenced by the Narang Dr and Project Narangwal the to connection strong their is projects four these of feature shared final A on theirfieldexperiences. These projects have been at the forefront of generation of knowledge about effective programming based evaluation, which led to the reporting of outcomes and the inclusion of these four projects in our review. and science of “culture” a has projects four the of each that is characteristic shared interesting Another deaths. ternal ma preventing for foundation the as serve care hospital available readily and systems referral veloped PHC interventions for reproductive and maternal health (including family planning) linked to well–de findings in detail is beyond the scope of this article, but suffice it to say here that presence of strong CB [ MCH–FP Matlab mortality: maternal reductionsin long–term of evidence have also here included projects four the of two that out point to important is it But, health. child and neonatal in improvements long–term on focused have paper this in reported findings The the lateremergence ofnationalCHWprograms inIndiaiswell–known[ on as well as Alma–Ata at defined as care CRHP’s health course, primary emergence of the on influence of And, leaders. health global become later who projects these with field the in experiences personal had have who people younger on influence their through as well researchas their through – world the These four projects have all influenced thinking and practice in CBPHC programming for MNCH around and nearby. community the in sites accessible readily at and homes their in care health essential provide they and areas, service in families all with contact routine maintain They support. supervisory and training ity these workers, all of whom receive some type of financial assistance. These CHWs all receive high–qual for role central a building without compromised be would effectiveness their realized all projects The workers. community–level for strong roles created projects four all that is similarity important final A trust beingdevelopedwiththecommunities. array of high–quality curative services by each of these projects over a long period of time has resulted in provided services health projects.the in by trust of level high a has provisionmunity The broada of com the engagement; community of level high a with sustained and developed been projectshave The a highlevelofrespect. They ensure that essential supplies and drugs are available. They all have a record of treating patients with All projects have strong professional leadership as well as dynamic management and supervisory systems. allincludestrongCBPHC services collaborationsandpartnershipswithcommunities. lished strong CBPHC services that serve as a foundation upon which the other project activities rest. These estab have projects comprehensive integrated importantly,these Most of needed. four when all care of recognize the importance of a functioning referral system to ensure that patients can access higher levels all They serve. they populations the to available is care surgical basic that ensure and services hospital provide all They planning. family and reproductivehealth and maternal on strongfocus a with services system that these projects developed. However, they all also provide comprehensive health primary care health system in which the community is a partner. Improving MNCH is one of many goals of the health Another characteristic these projects have in common is that they all have a strong community–oriented ferral services. –fromtegration oftheirservices allthewaytohospitalre home–basedandcommunity–basedservices – services from preventive to curative to rehabilitative Finally,services. they are similar in the vertical in of types of breadth the in similar also are They adulthood. and adolescent to periods child and natal uum of care for individuals at various points in the life cycle – from pregnancy and childbirth to the neo contin the offeralong they services of broadrange the in projectssimilar arethese of four all onstrates, As projects. these of features the of many of similarity the is striking particularly is What mortality Common characteristicsofprojectsimpacton withevidenceoflong–term versity,directlyinflu was icddrb at work Project.CBPHC Narangwal The the about learned wherethey Aroles and the Bangs were master of public health students of Dr Carl Taylor’s at the Johns Hopkins Uni 427
CBPHC andprojectswithevidenceoflong–termmortalityimpact ]. Exploring these these Exploring [29]. CRHP and 10] June 2017 •Vol. 7No. 1•010907 31, 41]. Tabledem- 2 26, ]. The The 35]. ------
VIEWPOINTSPaPErS PaPErSVIEWPOINTS June 2017 •Vol. 7No. 1•010907 sufficient through theirownincome–generatingactivities.CRHPensures thattheirCHWshaveenoughincometomeetneeds. salary,receivea not do CHWs CRHP the ‡Although self– economically become to them enable creditto to access and training receivespecial do they †The partoftheSEARCHproject area withdocumenteddeclinesininfantmortalityhas40 anddiscussionswithproject*Some ofthisinformationisbasedontheauthors’fieldobservations leadersandisnotcontainedinwrittendocuments. CHW –communityhealthworker Perry etal. Table 2. Do CHWs provide essential child health services in the home? visitation ofallhomes? through families all with contact routine have CHWs Do How strong isthetrainingandsupportofCHWs? Do CHWsreceive financialsupport? Are CHWsanintegralpartoftheproject? Role ofcommunity–basedworkers: project? the in community the of trust of level the is strong How community? How strong is the partnership between the project and the ment: Nature of community partnerships/community involve drugs? and supplies maintaining of record a have project the Does ing patientsandclientswithahighlevelofrespect? Does the project have a record of accomplishment in treat- supervision supervision led by competent and dedicated professionals? Does the project have a strong system of management and Health project managementandsupport: complicated surgical casesatJamkhedandSEARCH. more are as hospital district government the to referred ed into the project. However, all surgical cases at Matlab are In all four projects, a first–level referral hospital is integrat care hospitals? of including levels facilities, higher fixed at to community the from system referral the is strong How vided? pro services inpatient hospital other and/or surgical Are ously illpatientstohigherlevelsofcare. seri of referral and centers, health in ages all of patients among illnesses acute for care community; the in nesses ill childhood serious of indicated) when referral cluding ment of common childhood illnesses and management (in Are general curative services provided? These include treat of familyplanningmethods range wide a of provision and care, postnatal of provision complications, obstetrical of referral and/or management health and nutrition education, provision of antenatal care, include These provided? services planning family and Is a comprehensive array of maternal, reproductive health, ously illchildren toahigherlevelofcare. seri of referral illness, childhood acute of treatment and diagnosis education, nutrition and health include These Is a comprehensive array of child health services provided? provided:Range ofservices Population ofcatchmentarea Year established Basic project characteristics: c haracterIstIc Common characteristicsoffourprojects withlong–termevidenceofimpactonchildmortality* John Wyon, a colleague of Dr Carl Taylor’s in north India and the field director for the Khanna Study [ WarrenDrs. by led was Haiti in HAS at work Gretchen Berggren,and Dr by werethis mentoredwho in CBPHC project.The MCH–FP Matlab the for CHWs of support and training the initiated later Bhutyia, Shusham Dr project, that for fieldwork of director the because well as Project Narangwal the by ence a community–based field research project that served as a predecessor of the Narangwal Project. Dr Carl
------surgical capabilities) h Yes (operatesits referral hospital own first–level own first–level with advanced with advanced ôpItal arysrn arysrn eysrn Very strong Very strong Fairlystrong Fairly strong weItzer eysrn eysrn eysrn Very strong Very strong Very strong Very strong eysrn eysrn eysrn Very strong Very strong Very strong Very strong eysrn eysrn eysrn Very strong Very strong Very strong Very strong 150 1956 a Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes lbert (h 000 428 aItI s ) ch - proJect with nosurgical Yes (operatesits referral hospital own first–level own first–level m capabilities) atlab 100 (b 1965
Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes 000 people. mch–Fp angladesh 000 ) surgical capabilities) crhp–J Yes (operatesits referral hospital own first–level own first–level with advanced with advanced www.jogh.org 300 1970 Yes§ Yes‡ amkhed Yes Yes Yes Yes Yes Yes Yes Yes 000 (I • doi:10.7189/jogh.07.010907 ndIa search–g ) Yes (operatesitsown eg, cesarean section) surgical capabilities, hospital with some hospital withsome first–level referral 80 (I 1986 Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes ndIa 000† adchIrolI ) 42
], ], www.jogh.org • doi:10.7189/jogh.07.010907 tional replication and scale up by the government ofIndia. tional replication andscaleupbythegovernment na guided has intervention care neonatal home–based its and settings, resource–constrained in world dependent confirmatory research. Now, this approach has become the global standard of care around the based management of pneumonia and of home–based neonatal care stimulated interventions further in- program.ASHA moreSEARCH’srecentlyestablished effectivenesscommunity– the of of demonstration the and 1980s the of program Guides Health Village the with both India, in programming CHW tional na for model a as served has CHWs to approach CRHP The Haiti. throughout services munity–based com providing NGOs other all virtually by adopted HAS’sprogramming.been has CHWs to approach CHW of up scale national government of model the as served effective,proven once Project, MCH–FP in every case certain elements of each project have in fact been scaled up in an indirect sense. The Matlab are sustainable because of their long–term operation. None of these projects attempted to go to scale, but they that obvious is projects, it four these of standpoint difficult. From the more magnitude orderof an are which scale, larger at program a operating of challenges logistical and management the with deal to had not have they (ii) and leadership charismatic on dependent are they since scalable not and tainable sus- not are they (i) projectsbecause CBPHC “model” smaller of made been commonly have Criticisms proach toworkingwithcommunitiesandCHWs. ucation of thousands of people from around India and beyond who have come to learn about CRHP’s ap long period of time. CRHP is notable compared to the others in its deep commitment to field–based ed a over beginning the from them led who individuals key two had each projects The leaders. long–term strongof researchsingle set obvious small an international without an center of framework institutional The Matlab MCH–FP project differs importantly from the other three in that it functions within the strong research orientationandmoreorientation. ofaservice Although HAS and CRHP have been the site of important research, these two projects have had less of a care. neonatal home–based and pneumonia of management community–case relatedto health global in research influential most the of some of site the been world’s has SEARCH sites. research field foremost research and reported their results in peer–reviewed The journals. Matlab MCH–FP project is one of the in engaged have they which to degree the of terms in projects four the in differences notable are There well. as note to differences important are there characteristics, shared many these of spite However,in Study andtheyremained closecolleaguessubsequently. Khanna the of implementation and development the during Wyon John to mentor a as Taylor served ing itspopulation’s healthneeds. serv in projectcharacteristics of combination this of strength the to attests projects’characteristics four these of similarity The maintained. been have that characteristics project to led process this time, long a Over population. local their of needs the serve projectsto their adjust to basis regular a on evaluation projectsthese of Each recurrentlyused has reflection/implementation, consultation/planning, of cycle a the peopleitserves. of trust the earned has and engagement, community of level high a sustained and developed has home, the in children for services essential provides also Each them. for support and training strong provides and CHWs uses project Each services. counter–referral and referral facilitate all they and services, pital ductive health services, including family planning. They all provide general curative care, including hos tality. The projects described here provide a comprehensive array of child health and maternal and repro in our database of 700 assessments that have evidence of long–term impact of 10 years of more on mor period of time, usually five years or less. The four projects identified and described here are the only four of single interventions implemented in highly controlled and atypical field settings over a relatively short improving maternal, neonatal and chilld health (MNCH) outcomes comes from assessments of the effect in (CBPHC) care health primary community–based of effectiveness the regarding evidence the of Most important insightsinconsideringthisquestion. some give here cited projects four the of characteristics mortality. common under–five The on impacts high–mortality,projectsin health how for long–lasting dence achieve can resource–constrainedsettings evi less much is there health, child improve can others many and immunizations A, vitamin washing, hand as such interventions specific settings, controlled highly in that evidence strong is there Although CONCLUSIONS 429 CBPHC andprojectswithevidenceoflong–termmortalityimpact June 2017 •Vol. 7No. 1•010907 ------
VIEWPOINTSPaPErS PaPErSVIEWPOINTS June 2017 •Vol. 7No. 1•010907 Perry etal. rEFEr ENCES 3 2 1 6 5Arole M,Arole R.Jamkhed-AComprehensive RuralHealthProject. London:MacmillanPress; 1994. 4 8 7 11 10 9 Hopital Albert Schweitzer’s integrated system in Haiti. Health Policy Plan. 2006;21:217-30. Plan. Policy Health Haiti. in system integrated Schweitzer’s Albert Hopital throughunder-five mortality Reducing al. F,et Philippe H, M, Menager Cayemittes JR, HB, Dortonne Perry D, Dowell roli, India(1993to2003).JPerinatol.2005;25Suppl1:S108-22. AT,Bang Gadchi rural in trial field the of applications and care:neonatal summary Home-based HM. Reddy RA, Bang and databasedescription.JGlobHealth.2017;7:010901. methods rationale, 1. health: child and neonatal improvingmaternal, in care health primary community-based of ness P.Freeman WilhelmJ, effectiveS, regardingthe Gupta evidence B, the Rassekh of reviewHB, comprehensive Perry A lab, Bangladesh.StudFamPlann.1983;14:199-209.Medline:6636221 Chen LC, Rahman M, D’Souza S, Chakraborty J, Sardar AM, Yunus M. Mortality impact of an MCH-FP program in Mat 2007;97:240-6. Medline:17194853 dren under age 5 in rural Haiti: effects of a comprehensive health system in an impoverished setting. Am J Public Health. Perry H, Berggren W, Berggren G, Dowell D, Menager H, Bottex E, et al. Long-term reductions in mortality among chil doi:10.1093/heapol/czl005 lab, Bangladesh?StudFamPlann.2006;37:281-92.Medline:17209285 Hale L, DaVanzo J, Razzaque A, Rahman M. Why are infant and child mortality rates lower in the MCH-FP area of Mat Medline:17547776 doi:10.1186/1475-9276-6-4 2007;6:4. Health. Equity J Int Bangladesh. Matlab, from Evidence mortality? child and fant in neonatal, of improve inequalities socioeconomic intervention health Does DR. Gwatkin Streatfield PK, A, Razzaque demography. Oxford: Oxford University Press; 1997. in studies International countries developing in studies community Prospective editors. G, P,Pison Aaby M, Garenne Aziz K. Mosley H. The History, Methodology, and Main Findings,of the Matlab Project in Bangladesh. In: Das Gupta M, more-on-matlab 2017. . Accessed:26February Available:2016. health. public on impact its and Matlab icddrb. desh; 1994. Fauveau V. Matlab: Women, Children and Health. Dhaka: International Centre for Diarrhoeal Disease Research, Bangla these systemsinawaysothattheyare affordable withlocalresources isoneofourgreat challenges. maintain and build to Learning diabetes. and hypertension as such conditions chronic emerging many even and tuberculosis and malaria HIV/AIDS, of control in improvements long–term also but MNHC improvementsin long–term only not for potential the has here projectsdescribed the of lines the along Building strong and more comprehensive health systems in high–mortality, resource–constrained settings disclosure.pdf (availableuponrequest from thecorresponding author),anddeclare noconflictofinterest. www.icmje.org/coi_ at Form Interest Competing Unified the completed have authors All interests: Competing the finalversion. Authorship declaration: ecution ofthereview. ex the in role no had support financial provided that organizations The Foundation. Gates the and velopment, WorldThe ticle: UNICEF,Organization, Health De International for Agency WorldStates the United the Bank, ar this in described work the conduct to used were that funds providedorganizations following The Funding: tants, DrBahieRassekh,toparticipateasamemberoftheStudyTeam. support to Dr Perry during the initial phase of the review. The World Bank made it possible for one of its consul some of these funds. We thank Future Generations for providing office space, administrative support, and salary administered which staff, Section Health International its particularly and Association Health Public American the to grateful Wealso Foundation. are Gates the and Generations, Future Health)/USAID, Child for Resources and Adolescent Health and Development of the World Health Organization, the CORE Group (Collaboration and that provided small grants to cover the expenses of this review: UNICEF, the Worldorganizations following the Bank, to Wegrateful the aresection). Department SEARCH/Gadchiroli of the on Child comments (for Bang Abhay and section), CRHP Jamkhed the on comments (for Gates AroleConnie Shobha and section), Schweitzer Albert Hôpital the on comments (for Berggren Gretchen section), icddrb Matlab the on comments (for Mosely Henry Acknowledgments:
We are grateful to the following people for comments on an earlier version of this paper: paper: this of version earlier an on comments for people following the to grateful Weare HP wrote the first draft. All authors reviewed and edited subsequent drafts and approved doi:10.2105/AJPH.2006.088732 430 Medline:15791272 http://www.icddrb.org/research/platforms/field-sites/ doi:10.2307/1966412 doi:10.1111/j.1728-4465.2006.00106.x www.jogh.org doi:10.1038/sj.jp.7211278 • doi:10.7189/jogh.07.010907 Medline:16565151 ------
www.jogh.org rEFEr ENCES • doi:10.7189/jogh.07.010907 36 35 29 38 37 34 33 32 31 30 28 27 26 25 24 23 22 21 20 19 18 17 16 15 14 13 12 Narangwal Experiment. Integrated Family Planning and Health Care. Baltimore, MD: Johns Hopkins University Press; 1983. Taylor CE, Sarma RSS, Parker RL, Reinke WA, Faruqee R. Volume 2. Child and Maternal Health Services in Rural India: the World Bank[by]JohnsHopkinsUniversityPress; 1983. the for Published Baltimore: Care. Health and Nutrition Integrated Experiment. Narangwal the India: Rural in vices TaylorAA, Kielmann WA,Reinke C, D. RL. Parker C, DeSweemer CE, Volumeal. et Ser Health Maternal and Child 1. 8-9. discussion 2001;75:297-307, Obstet. Gynaecol J Int home. at are deliveries most where nity commu Indian rural a in care obstetric emergency effective and Efficient R. Arole S, Arole R, Premkumar C, McCord line:1973770 1990;336:201-6. Lancet. India. Gadchiroli, in trial intervention community-based of means by mortality hood Bang AT, Bang RA, Tale O, Sontakke P, Solanki J, Wargantiwar R, et al. Reduction in When CommunitiesOwnTheir Futures.pneumonia Baltimore: JohnsHopkins UniversityPress; 2002. mortality and total child Change: Lasting and Just Taylor-Ideeditors. TaylorIn: C, D, development. to barrier a as Addiction R. Bang A, Bang Suppl 1:S3-10.Medline:15791276 AT,Bang 2005;25 Perinatol. Gadchiroli, J carein India. neonatal home-based of trial field Backgroundthe RA. of Bang sociation ofIndia;1998. VoluntaryDelhi: Health. Community Researchin and As Action Health Education, for Society SEARCH: D. Dasgupta roli.org/ 2017. . Accessed:26February SEARCH. Society for Education, Action and Research in Community Health. 2016. Available: Geneva, Switzerland:World HealthOrganization; 1975. Arole M, Arole R. A comprehensive rural health project in Jamkhed (India). In: Newell KW, editor. Health by the People. Newell KW, editor. HealthbythePeople.Geneva:World HealthOrganization; 1975. doi:10.1016/S0020-7292(01)00526-4 Organ. 2010;88:727-36.Medline:20931057 munity-based primary health care programme on under-5 mortality in villages around Jamkhed, India. Bull World Health V,Mann P.Boone R, com Premkumar Frosta C, A, of Eble impact lasting the of evaluation Retrospectivecomparative Data/Maharashtra_report.pdf 2017. . Accessed:26February Mumbai, Institute India: for International Population Sciences; 2007. Available: 2005-06. India, (NFHS-3), Survey Health Family National Maharashtra: WelfareGoI. Family and Health of Ministry crhpjamkhed.org/impact/impact 2017. . Accessed:27February Jamkhed Comprehensive Rural Health Project. Comprehensive Rural Health Project Impact. 2016. Available: 6736(15)00519-X ress for maternal, newborn, and child survival. Lancet. 2016;387:2049-59. Lancet. survival. child and newborn, maternal, for ress Victora CG, Requejo JH, Barros AJ, Berman P, Bhutta Z, Boerma T, et al. Countdown to 2015: a decade of tracking prog coming 2017. Forth Health. Community Epidemiol J Haiti. rural from findings under-5mortality: reducing in program care health J, Berggren HB, Northrup R, Bryant Perry W, Berggren G. The cost-effectiveness of a long-term comprehensive primary Organ. 1974;8:24-9.Medline:4423926 Berggren WL. Control of neonatal tetanus in rural Haiti through the utilization of medical auxiliaries. Bull Pan Am Health ral Haitianpopulations.AmJTrop MedHyg.1971;20:491-4. Berggren WL, Berggren A retrospective GM. Changing incidence of fatal tetanus of the newborn. study in a defined ru N EnglJMed.1981;304:1324-30.Medline:7219486 Berggren WL, Ewbank DC, Berggren GG. Reduction of mortality in rural Haiti through a primary-health-care program. Schweitzer HA.HôpitalAlbertSchweitzer. 2016.Available: doi:10.1016/S0140-6736(13)62149-2 nerships: innovations in health-service delivery in Bangladesh. Lancet. 2013;382:2012-26. Lancet. Bangladesh. in delivery health-service in innovations nerships: part and approaches Community-based al. et KS, Islam FA,K, Osman Azad L, Reichenbach A, Christou S, Arifeen El Geneva:World andChildSurvival. Newborn Stock ofMaternal, HealthOrganization andUNICEF;2010. Profiles: Country Taking Worldwith UNICEF.Organization.(2000-2010) Health Report Decade 2015 to Countdown Centre forDiarrheoalInternational DiseaseResearch, Bangladesh;1994. Fauveau V, Chakraborty J. Family Planning and Maternal and Child Health Services in Matlab. Fauveau V, editor. Dhaka: lab, Bangladesh.IntPerspectSexReprod Health.2010;36:170-7. Rahman M, DaVanzo J, Razzaque A. The role of pregnancy outcomes in the maternal mortality rates of two areas in Mat graphic Events-2005.Dhaka:ICDDR.B.;2007. Demo and Health of Registration Volume39. Matlab. - System Surveillance Demographic MM. Rahman G, Mostafa 27. RegistrationofVital Events-1995.Dhaka:ICDDR.B.;1996. VolumeMatlab. - System Surveillance Demographic SarderAM. JH, Ginneken van MAK, Shaikh K, Ajmed G, Mostafa in Bangladeshichildren. SocSciMed.2001;52:267-77. TofailM, Emch M, Ali F, mortality infection carerespiratory provisionhealth lower of acute Implications on AH. Baqui desh: UniversityPress Ltd;2000. Twenty-Firstthe for Century.Care Bangla Health Dhaka, Primary in Lessons Bangladesh: in All for Health HB. Perry doi:10.1016/0140-6736(90)91733-Q doi:10.1038/sj.jp.7211267 doi:10.2471/BLT.09.064469 431 doi:10.1056/NEJM198105283042203
Medline:11144783 http://www.hashaiti.org/ 2017. . Accessed:27February Medline:5088398 CBPHC andprojectswithevidenceoflong–termmortalityimpact Medline:21245023 doi:10.1016/S0277-9536(00)00120-9 Medline:26477328 http://www.nfhsindia.org/NFHS-3%20 doi:10.4269/ajtmh.1971.20.491 June 2017 •Vol. 7No. 1•010907 doi:10.1363/3617010 http://www.searchgadchi-
Medline:24268607 doi:10.1016/S0140- Medline:11728493 http://www. Med------
VIEWPOINTSPaPErS PaPErSVIEWPOINTS June 2017 •Vol. 7No. 1•010907 Perry etal. rEFEr ENCES 42 41 40 39 sity Press; 1971. Univer Harvard MA: Cambridge, Punjab. Rural the in Problems Population Study: Khanna The WyonJE. Gordon JB, uploads/2017/01/CHW-CaseStudies-Globes.pdf 2017. . Accessed:26February da, Zambia, and Zimbabwe. 2017. Washington, DC: USAID/MCSP. Available: Programs: Examples from Afghanistan, Bangladesh, Brazil, Ethiopia, India, Indonesia, Iran, Nepal, Niger, Pakistan, Rwan Gergen H,ScottK,JavadiD, Perry Large-ScaleStudies of L, etal.Case K, Crigler J, Shelley Worker CommunityHealth 6736(99)03046-9 sis on neonatal mortality: field trial in rural India. Lancet. 1999;354:1955-61. Lancet. India. rural in trial field mortality: neonatal on sis Bang AT, Bang RA, Baitule SB, Reddy MH, Deshmukh MD. Effect of home-based neonatal care and management of sep doi:10.1038/sj.jp.7211277 of the Gadchiroli field trial: effect of home-based neonatal care. J Perinatol. 2005;25 Suppl 1:S92-107. Bang AT, Reddy HM, Deshmukh MD, Baitule SB, Bang RA. Neonatal and infant mortality in the ten years (1993 to 2003)
432 http://www.mcsprogram.org/wp-content/ Medline:10622298 www.jogh.org • doi:10.7189/jogh.07.010907 doi:10.1016/S0140- Medline:15791283 - - - www.jogh.org 8. summaryandrecommendationsoftheExpertPanel in improvingmaternal,neonatalandchildhealth: effectiveness ofcommunity–basedprimaryhealthcare Comprehensive reviewoftheevidenceregarding material. The onlineversionofthisarticlecontainssupplementary Electronic material: supplementary Claudio FLanata 8 7 6 5 4 3 2 1 r [email protected] USa Baltimore, MD21205 615 NorthWolfeSt. Health Johns HopkinsBloombergSchoolofPublic r Henry Perry Correspondence to: ‡ † * 13 12 11 10 9 Shobha a G Victora Henry BPerry Mushtaque r Phillips r Bang
B
oom E8537 Chairperson, ExpertPanel obert EBlack Kirkwood Member, ExpertPanel Former Chairperson,ExpertPanel(deceased) Institute ofNutritionalr DC, USa for InternationalDevelopment,Washington, Bureau ofGlobalHealth,UnitedStatesa Medicine, London,UnitedKingdom London SchoolofHygieneandTropical a University ofToronto,Ontario,Canada India Community Health,Gadchiroli,Maharashtra, Society forEducation,a Jamkhed, Maharashtra,India Jamkhed Comprehensiver Health, Baltimore,Maryland,USa Johns HopkinsBloombergSchoolofPublic a Capital InstituteofPediatricsandChina Federal UniversityofPelotas,Brazil Vermont, USa Independent Consultant,Southr Health, NewYork,USa Columbia UniversityMailmanSchoolofPublic raC, Dhaka,Bangladesh ga KhanUniversity,Karachi,Pakistan dvisory CenterforChildHealth,Beijing,China 3,*,‡
10,*,‡ • doi:10.7189/jogh.07.010908 , Zulfiqara
12,*,‡ role
, MaryTaylor 7,*,‡ , ZonghanZhu Chowdhury 1 2,*,‡ , NazoKureshy 1,*
, CarlETaylor , a
9,*,‡ esearch, Lima,Peru ction andr Bhutta bhay bhay , James F , JamesF ural Health Project, ural HealthProject, 11,*,‡ oyalton, oyalton, 6,*,‡ 4,5,*,‡
esearch in esearch in , Cesar , Cesar 13,*,‡ , Betty , Betty , a 8,*,‡ gency gency 1,† , , , be needed to guide national and international policiesand programs.be neededtoguide nationalandinternational ing operations research and evaluation will be required, and this evidence will eases. Continuedstrengtheningof CBPHCprograms on rigorousbased ongo - ment of HIV/AIDS, tuberculosis, malaria, hypertension, and other chronic dis planning services as well as for accelerating progress in the detection and treat can also create entry points and synergies for expanding the coverage of family tional Conference on Primary Health Care in 1978. Stronger CBPHC programs Interna the at envisioned as All for Health achieving eventually and world), the around countries UNICEF,many Organization, and World Health the tions), ending preventable child and maternal deaths by 2030 (as called for by for by the Sustainable Development Goals recently adopted by the United Na be essential for achieving universal coverage of health services by 2030 (as called will interventions evidence–based of implementation the with empowerment engagement/ community foster programsStrongerthat CBPHC Conclusions sure thatthosemostinneedare reached. that can be implemented in the community outside of health facilities and as interventions evidence–based key of coverage universal achieve to order in evaluation. CBPHC programs need to reach every community and household and workers), community–level of use full the (including implementation planning, with engagement through – programs CBPHC strengthening and building in them with collaboratively work and partners full as communities value and respect that systems health develop should NGOs and programs, mortality.averting for CBPHC of health portance government Governments, im the reflect and hospitals and facilities care health primary for penditures ex against monitored be CBPHC for expenditures that recommends also el health coverage, and ending preventable child and maternal deaths. The Pan universal achieving in progress accelerating systems, health strengthening for priority a be should CBPHC that recommends Panel Expert The Results in neonatal and/or child health and once for an assessment in maternal health). ments since 39 were analyzed twice (once for an assessment of improvements assess 700 of consisted analysis The literature. gray the from reports 72 and chapters, book 4 books/monographs, 12 articles, journal peer–reviewed 583 including reports, unique 661 of analysis an of consisted review The dations. of the review, the Panel considered the review’s findings and made recommen in New York City with senior UNICEF staff. In 2016, following the completion community–based primary health care (CBPHC). The Expert Panel met in 2008 Methods An Expert Panel convened to guide the review of the effectiveness of been updatedonthebasisofmore recent evidence. Expert Panel for strengthening CBPHC that were formulated in 2008 and have tiveness of CBPHC in improving MNCH and offers recommendations from an by 2015. This article provides a summary of a series of articles about the effec Goals Development Millennium health–related the achieve to failed countries priority few so why reasons the of one is reality unfortunate This preciated. improvement of maternal, neonatal and child health (MNCH) is not widely ap (CBPHC) and engaging with communities as valued partners can make to the Background The contributions that community–based primary health care care health primary community–based that contributions The 433 June 2017 •Vol. 7No. 1•010908 global journal of
health ------
VIEWPOINTSPaPErS PaPErSVIEWPOINTS June 2017 •Vol. 7No. 1•010908 Black etal. second international in a generation [17 , The World Health Organization and UNICEF have called for ending preventable child and “ maternal to deaths access universal with 2030, year the by thrive” can lives all world a for calling Goals, Development Sustainable the ed still toachievetheseloftygoals,muchworkremains tobedone.In2015theUnitedNationshasadopt world’sthe (see of resources use better and fuller a through 2000 year Health ry Care that an acceptable level of health for all the people of the world could be achieved by the time by the World Health Organization and UNICEF affirmed at the International Conference on Prima- that to up convened officials health of gathering largest the 1978, later,in years Forty [13]. assistance” everyone has a right to medical care and that “motherhood and childhood are entitled to special care and In 1948, the United Nations General Assembly affirmed in its Universal Declaration of Human Rights that differently.sibility thattheMDGeramighthaveendedvery therecarestrongbelieve health a morewe is been developed, fully primary pos community–based had resource in global health. Had communities been engaged more fully as partners with health systems, and undervalued most the are Communities CBPHC. strengthening or communities with partnerships ger [ world aroundthe health to commitment political increased and support, donor of harmonization interventions,” high–impact of scale–up matic ing for strengthened health systems to deliver essential services to women, newborns and children,” “dra- The Call for Action arising from this Conference focused on the need for “long–term, predictable financ [ 40% than less is it interventions 21 the of 6 for and 60% than less still is interventions MNCH [ age of years five than younger children of mortality for goal the one–thirdreaching only and mortality maternal for goal the reaching countries The era of the Millennium Development Goals (MDGs) ended in 2015 with only seven of 75 Countdown Why thereview isimportantnow [ series current this in articles earlier the in presented findings the hereWe summarize proving MNCH. crete steps to recognize that communities are a vital resource and key partners with health systems in im con proposes also It (MNCH). health child and neonatal maternal, improving in (CBPHC) care health primary community–based of effectiveness the regarding evidence current the summarizes paper This of CBPHCtoachievingthisgoal ismuted[19, size the importance of community engagement/empowerment, the critical and fundamental contribution that hadevidenceofmortalityimpactfor10yearsorlonger. as the strategies used to achieve health effects, including the common strategies of four projects identified Not only did the authors assess health effects, but they also examined the equity of these projects as well coverage of evidence–based interventions, (2) nutritional status, (3) serious morbidity, and (4) mortality. The health populations. defines geographic review defined effects population the (1) in change broadly: effects of community–based projects, programs and research studies (hereafter referred to as projects) in improving maternal, neonatal and child health (MNCH). The authors identified assessments of the health regarding century a half over for effectivenesscarethe health in (CBPHC) primary community–based of accumulated evidence of reviewcomprehensive in–depth providean series this in articles previous The health postandvisitingfamiliesintheirhomes,asisthecaseEthiopia. village–based a at patients to attending by professionalizedtime which their in divide developed CHWs be to need also approaches Hybrid resource–constrainedsettings. in systems health of effectiveness the improve to order in programs health of mainstream the to brought are services community–based and ditional “facility–centric” approach to health systems and calls for a new paradigm in which communities munities and community–based approaches to improving MNCH is still being overshadowed by the tra activities that may be linked with health facilities but which take place in communities. The role of com CBPHC is defined as a process by which health systems work with communities to improve health through World HealthOrganization (WHO)andtheWorld Bank. the from staff senior with along staff UNICEF senior with 2008 March 27–8 on YorkCity New in ters ment) convened to guide the activities of the Working Group when the Panel met at UNICEF headquar [ previously Workingprepared Groupthe of papers the and Association, Health Public American the of [8 tion Reproductive, Maternal, Newborn and Child Health volume of the Disease Control Priorities, Third Edi 10– ] as well as discussions of an Expert Panel (for membership see membership (for Panel Expert an of discussions as well as 12] , 9 ]. It also is an outgrowth of the Working Group on CBPHC of the International Health Section Section Health WorkingInternational the outgrowthof an CBPHC is Groupon also It ].
18]. However, even though recently released plans for achieving this goal do empha Countdown Countdown to 2015 Conference 8 9 434 ]. The population coverage of 13 of 21 key evidence–based evidence–based key 21 of 13 of coverage population The ]. ]. However, there was no mention or call for building stronbuilding for call However,or ]. mention no was there
20]. met in Cape Town, South Africa, on 17–19 April 2008. “freepoverty,of hunger, wherewant, and disease quality essential health–careservices” [ essential quality ]. As the world seeks seeks world the As [15]. 1) Box www.jogh.org Online Supplementary Docu- Supplementary Online • doi:10.7189/jogh.07.010908 1 – 7 ] and in the the in and ] 8 ]. The The ]. ]. 16]. ------§Integrated communitycasemanagement (thecomponentsincludetreatments fordiarrhea, pneumonia,malaria). ‡Indoor residual spraying. †Insecticide–treated bednet. www.jogh.org pregnant womentoimprove andperinataloutcomes[ maternal for micronutrientsupplementation multiple recommend time this at not does WHO the evidence, the in gaps and risk of evidence some of *Because community oratahealthpost[ Table andchildhealththatcanbeprovided newborn 1.Effective bycommunityhealthworkersinthe formaternal, interventions Promotion ofHIVtesting Food supplementation of malariaduringpregnancy) Intermittent preventive treatment Nutrition education Micronutrient supplementation* care; emergencynewborn planning Preparation forsafebirthand Pregnancy • doi:10.7189/jogh.07.010908 Box 1. of socio–economic status, benefit more from the delivery of one or more CBPHC interventions than does type other some or quintiles income of terms in defined usually population, the of segment advantaged scribed in more detail in Paper 5 of this series [5 de as interventions, CBPHC of effectpro–equitystrong a supports evidence available the effectiveness, intervention overall has as extensively as studied been not have services CBPHC of equity the Although Equity tions research. The number of such interventions will certainly continue to grow with continued experience and opera with appropriate training, supervision and support. All of these interventions are described in the review. Table1 Specific interventions r ESULTS munities toparticipate”(ArticleVII)[14]. tional and other available resources; and to this end develops through appropriate education the ability of com tion in the planning, organization, operation and control of primary health care, making fullest use of local, na participa and self–reliance individual and community promotesmaximum “requirescareand health Primary opment inthespiritofself–reliance andself–determination”(ArticleVI). participation and at a cost that the community and country can afford to maintain at every stage of their devel ods and technology made universally accessible to individuals and families in the community through their full “Primary health care is essential health care based on practical, scientifically sound and socially acceptable meth tation oftheirhealthcare” (ArticleV). “The people have the right and duty to participate individually and collectively in the planning and implemen The DeclarationofAlmaAta contains the evidence–based interventions that can be provided by community–level workers workers community–level by provided be can that interventions evidence–based the contains 21–23 and referral ofcomplications Management oflaboranddelivery (normal) Delivery ] 24 p oInt ].
In
contInuum 435 breastfeeding Promotion of (woman) Postpartum
oF ]. The term pro–equity effect signifies that the most dis-
care for pneumonia Oral antibiotics signs andrefer Assess fordanger care–seeking Promote preterm newborns Provide vitaminA,zinc,andfood Thermal care for Breastfeeding resuscitation Neonatal (newborn) Postpartum
June 2017 •Vol. 7No. 1•010908 CBPHC summaryandrecommendations if dangersignspresent without dangersigns(iCCM§),refer Detect andtreat seriousinfections acute malnutrition Detect andrefer children withsevere Assess fordangersignsandrefer or IRS‡,both Distribute andpromote useofITNs† children’s stoolsandhandwashing Education onsafedisposalof children Co–trimoxazole forHIV–positive Immunizations supplementation feeding complementary Promote breastfeeding and Child ------
VIEWPOINTSPaPErS PaPErSVIEWPOINTS June 2017 •Vol. 7No. 1•010908 Black etal. support fortheimportanceofinvestinginCBPHCimproving MNCH. cilities alone can improve population health in resource–constrained settings [ there [25, services obtain to likely less are quintiles income lower the in those because inequitable is facilities care health primary of utilization that finding observed commonly the to contrast stark in stand findings These surprising. not are findings these so interventions, key of coverage population expand rapidly can and facilities health from furthest those reach can approaches Community–based population. the of segment better–off the MNCH. These are described in greater detail in paper 4 in this series [ improving in effective interventions CBPHC making for systems health with and communities with partnership in working for approaches innovative myriad utilized review the in included projects The Strategies forachievingeffectiveness ulation (only 11% of the projects assessed served more than 25 than more served assessed projects the of 11% (only ulation a relatively short period of time (2–3 years) in highly controlled field settings with a relatively small pop of all, the evidence is largely limited to assessments of a small number of interventions implemented over First recognized. be to need that limitations important have does it extensive, is evidence the Although had strong referral capabilitiesandprovided first–levelhospitalcare. Limitations oftheevidenceidentified tact with all households; they all had strong collaborations with the communities they serve; and they all regularcon programmaintained worker that stronghealth a community had they planning; family ing includ carehealth services, featuresprovided mon all arethey primary striking: comprehensiveof a set com more,their or but years 10 of impact mortality of evidence had four only assessments, 700 the Of a peripheralhealthfacilities.Thesefourstrategiesare notmutuallyexclusive,ofcourse. implementation strategy identified is the provision of community–based servicesby mobile teams based not only improves the health of mothers and children but it empowers women at the same time. A fourth community.their in information this share and children their of process health The the and health own their promote can they which in ways about learn to facilitator a with meet women of groups which in women’s participatory of formation the was assessments the projects in the included groups among fied thirdidenti appropriateA strategy with medications. community the in treatedcases referral,and/or of need in cases identified signs, dangers and warning on providededucation they which in management, case community provided commonly workers these Second, workers. community–level paid and teer often routine visits to all homes as well as visits to targeted groups, were often carried out by both volun Finally, four implementation intervention strategies were commonly encountered. First, home visitations, and thereferral hospitalwere commonfeatures ofeffective projects. referral system. Building strong links among the community–level activities, the peripheral health facility the strengthening and workers, community–level the and facility health peripheral the both for system supply logistics/drug the and workers community–level of system supervisory the strengthening tients, treatreferredand activities pa community–level supervise who facilities health staffperipheral at based of training including types, various of activities strengthening system health in engaged projects Many were involvedinplanningtheproject andin40%theywere involvedintheevaluation. communities projects, the of 39% approximatelywomen’s In present.of promotion was empowerment local resources, and promotion of community engagement/empowerment. In nearly half of the projects, community,the projectand the between partnerships of promotionof use mentation, the promotionof projectin imple involved was community projectsthe the assessed half morethan In workers. these to women, and in three–quarters of the projects included in the review some type of training was provided were workers these cases, most In implementation. project with assisted workers) paid and volunteer both (including types many of workers Community–level education). peer–to–peer and games stories, and devising innovative ways to share key education messages with the community (through skits, songs, leaders (both formal and informal), engagement with existing and/or formation of new women’s groups, tant themes emerged from the review. Many project assessments described engagement with community ations, and contextual considerations have a major influence on project operations. Nevertheless, impor programs thatreach larger populationsoverlongerperiodsof timeislimited. comprehensive more of effectiveness for evidence the Thus, years. more or 5 of period a over mented imple 13% only with and less or year 1 of period a over implemented were projects the of (46%) half
]. This evidence together with the lack of evidence that investments in fa- in investments that evidence of lack the with together evidence This 26]. 436
000 women and children), and almost almost and children), and women 000 6 ]. Clearly no one size fits all situ www.jogh.org 27, • doi:10.7189/jogh.07.010908 28] provide additional ------www.jogh.org • doi:10.7189/jogh.07.010908 0.8 milliondeaths)andinhospitals(whichwouldavert0.9 milliondeaths)( avert would (which centers care health primary in delivery require that interventions the to compared provided at the community level reach all who need them, 2.3 million deaths would be averted each year be childrencan their that and mothers for interventions evidence–based of package complete the if that do, isimportant. to attempts review this which interventions, specific implementing for strategy a as CBPHC on focus a duced in the future, and epidemiological as well contextual conditions will change over time, so keeping packages of interventions are most important. We know that new interventions will continually be intro improving MNCH, is not lost by dwelling on detailed discussions of which specific interventions or for which approach essential and effective an is CBPHC review, that the namely of finding main the theless, None interventions. specific to pertain that conclusions of power the limits This interventions. of ages scriptive and does not undertake a quantitative analysis of effect strength of specific interventions or pack Given the broad scope and heterogeneity of the evidence included, by necessity the review is largely de systematic review [32]. a of quality the judging for AMSTARcriteria quality 11 the of 7 meet to review the allow review entire conclusions of individual articles, and the inclusion of conflict of interest and funding information for the included articles, the presence of a quality assessment of included reviews and incorporation of this into donor support[30, reached only 8% national coverage over an 8–year period as a result of inadequate financial backing and [ approach CBPHC evidence–based effective an where point, mortality.in child case and a is Ghana neonatal maternal, of burden high a with countries in especially CBPHC, up scaling and strengthening en tothepuzzlingquestionofwhy, evidence,more giventheoverwhelming effort hasnotbeengivento MNCH and steps that need to be taken to overcome them. Furthermore, more attention needs to be giv ly, more analyses are needed of the main barriers that hinder the fuller development of CBPHC to improve cation bias needs to be recognized, and the overall findings interpreted accordingly. But more important There is a notable lack of evidence regarding failed attempts to improve MNCH through CBPHC. Publi • must beaddressed byboth programs andcommunities: [ documented been has progress major but simple, Promoting community engagement/empowerment to increase intervention effectiveness is obviously not Promoting communityengagement/empowerment ventions and reducing maternal, neonatal and child mortality [ nity–based interventions can be remarkably effective in expanding the coverage of evidence–based inter Long–standing experience and rapidly growing evidence both show that simplified home– and commu saved byscalingupCBPHC Estimates ofthenumberlivesmothersandtheirchildren thatcouldbe ered atthecommunitylevel,andwhatstepswere takentostrengthen thehealthsystem. weredeliv interventions how wereengaged, communities how implemented, reviewwas the in cluded currently available, with great efforts taken to extract all available information about how each project in erature. The review is composed of 700 assessments. Second, it is one of the most comprehensive reviews lit gray the fromreports and books, evaluations, project unpublished from also but literature reviewed peer– the from only not evidence included reviewers The them. implement to used strategies common most the are as well–known, less is effective be to known interventions of breadth the well–known, is here described interventions the of many effectivenessof the While conditions. treatable or preventable reducing 6 million deaths of mothers and their offspring each year [ progress accelerating in for relevant highly is that topic important this on reviewscurrent in–depth sive The review described in this series has some important strengths. First, it is one of the most comprehen- Strengths ofthe review an of presence The evidence. the of review We make no claim that this is a systematic review of the evidence. We do claim that it is a comprehensive
ing communitiesmainlyastargets andessentiallypassiverecipients of services? financial) to improve MNCH, or will the more common practices continue of health systems consider Willresourcesconsiderable own its bring and partner participating a be community the non– (mostly 31]. 437 design, the inclusion of gray literature, the listing of of listing the literature, gray of inclusion the design, priori a
]. Experience shows that the following questions questions following the that shows Experience 33]. 22, 23]. The best current evidence indicates 8 , 9 ], most of which are from readily June 2017 •Vol. 7No. 1•010908 CBPHC summaryandrecommendations Figure 1)[22]. 29] ------
VIEWPOINTSPaPErS PaPErSVIEWPOINTS June 2017 •Vol. 7No. 1•010908 Black etal. • • • • • • be empoweringforcommunitiesincludethefollowing: can that and interventions of effectiveness the improving to contribute can communities that Activities prove MNCH. munity engagement/empowerment as important for enabling the delivery system to more effectively im com view did it alternatives, either–or as questions these approach not did Panel Expert the Although • platforms [ service Figure 1. all interventions that canbedeliveredprovided allinterventions bya ineachplatformare asfollows.Thecommunityplatform: acute malnutritionwithinfection[22]. operative delivery, fullsupportive care andcare forpreterm of children newborns, withsevere infectionorsevere inhigh–riskwomenorthosewithcomplications,including formanagement oflaboranddelivery services consistingofbothfirst–leveland referral hospitals,includesmore advanced workers. Thehospitalplatform: ofcommunity–based injectable antibiotics,thatcannotbedoneinthecommunity;andtrainingandsupervision pregnancy foruncomplicatedpregnancies; anddelivery provision ofmedicalcare foradultsandchildren, suchas contraceptives (implants,intrauterinedevices);surgical sterilization(vasectomy, tuballigation);care during treatment capabilities.ThePHCprovides includinglong–acting reversible facility–basedcontraceptiveservices, facility withadoctorornursemidwife(orboth),nursesandsupportstaff, aswellbothdiagnosticand health days,immunizations,vitaminA,andotherinterventions. community–based healthworkerwithappropriate trainingandsupportorbyoutreach suchaschild services, Participatinginmonitoring, evaluationandaccountability;and, terventions foracceptabilityand efficiency;terventions in of integration with culture local and realities local to system delivery the adapting in Participating health care utilization; Involving men and mothers–in–law in creative ways that encourage healthy behaviors and appropriate of home–basedcare; Involving women’s groups in participatory learning and action, peer–to–peer education, and provision standing andrespect astheyworktogetherforresults thatare effective andequitable; Clarifying local value systems to help both the delivery system and community develop mutual under resources);cluding themanagementofexternal Involving local leadership in mobilizing the community for planning and management of activities (in als definingthese roles asthe responsibility ofthehealthsystem? and evaluating program activities, in contrast to the much more common practice of health profession implementing as well as priorities setting in participate to opportunity the have community the Will Maternal, perinatal,neonatalandchilddeathsthatcanbeavertedbyhealth–care packagesthroughMaternal, three
22 ]. Thenumbersabovethecolumnswere notintheoriginalfigure.assumedtobe Theservices 438 a a healthcenter(PHC)platform: The primary www.jogh.org • doi:10.7189/jogh.07.010908 ------www.jogh.org • doi:10.7189/jogh.07.010908 tional impact more rapidly. Even though “command and control” approaches can be used for scaling up There is a need to test different approaches for rapid scaling up so that CBPHC programs can achieve na • • • • • but alsoinimproving healthprioritiesinmiddle–andupper–incomecountriesaswell[ providing CBPHC have been shown to be effective in improving not only MNCH in low–income countries grow.to effectivenesscontinues of evidence the as especially workers needed, Community–level now are levels global and international national, local, at engagement/empowerment community and CBPHC for [ context local the to them adapt to need the of result a as greatly vary ties communi and systems delivery intervention health effectivebetween natureof partnerships The central. areas where mortality is the highest and therefore where impact can be greatest. Here also equity issues are facilities. Populations with the most limited access to formal health care are typically in the most unreached receive it (or to take the intervention to where the child is, ie, in the home). CBPHC requires linkages with can they where to taken are intervention an need who children that ensure to important is powerment engagement/em community immunizations), (eg, interventions other For rhea,malaria. and pneumonia diar childhood of management community–based and care neonatal home–based for clearest is idence Effective program planning, implementation and assessment require community involvement, and the ev • • • are thefollowing: Innovative approaches to scaling up CBPHC approaches that improve MNCH are needed. Some examples strengthened CBPHChasmadetotheseachievements. a major contribution to these achievements. More research is needed to fully assess the contribution that Nepal and Rwanda [ ing national examples of improvement in MNCH exist in countries such as Afghanistan, Brazil, Ethiopia, effectivenessthe for evidence The improvingin CBPHC of Yet, limited. still is scale at MNCH encourag- Scaling upcommunity–basedprimaryhealthcare
diarrhea [44]. as was carried out by BRAC through its home–based training of mothers to prevent and treat childhood The gradual expansion of one key intervention to a national level under the direction of a single NGO, [43];and, now governed la Administracion de Salud over health centers and local programs, as has occurred in Peru’s program controlof shared establishing of option the communities local giving framework national a of Creation tries havenowbeentrained[42]; where more than 30 India, Jamkhed, in (CRHP) Program Health Rural Comprehensive the at occurring is as CBPHC, and empowerment learn to center training central a to come countries different in programs and areas ic A “bottom–up” educational approach to scaling up, in which grassroots workers from many geograph dia); Tibet (China),AfghanistanandPeru[ 41]; has done with its SEED–SCALE approach to improve the health of children in Arunachal Pradesh (In el Counties Project [ as nodes to adapt and systematize extension to larger populations, as was done in China with the Mod and an outside agent (such as an NGO or technical support group) first establishes model program sites community,officials, the government which in up, scaling for outset the at partnership three–way A Navrongo Initiativeworkingthrough ofHealthinGhana[ theMinistry the by out carried been has as emerge, to flexibility, champions and local input allowing local strong “Scaling down to scale up” in which a documented successful approach is replicated at other sites with lishing long–termpartnershipsforrobust andsustainablesystems. Collaborating not just in a series of interventions during the initial stages of implementation but estab neonatal care, beginningwithSEARCH’s pioneeringworkinGadchiroli [ home–based for occurred has as India, in underway currently is as system government the into tion program implemented by one NGO, replication by other NGOs, with gradual transfer of the interven effective small a with beginning level national to interventions of package a of expansion gradual The India, Malawi,Nepal,Niger, Rwanda,andmanyothercountries[ with gradual addition of responsibilities, as has happened in Afghanistan, Brazil, Bangladesh, Ethiopia, Establishment of a cadre of government–authorized community–level workers throughout the country 37, 40] (which has now become China’s rural MCH system) and as Future Generations
000 people from around India and more than 3000 people from 100 other coun 38] and these countries have established strong CBPHC programs that have made (CLAS), under which one–third of the government’s 2400 health centers are 439 38]; June 2017 •Vol. 7No. 1•010908 CBPHC summaryandrecommendations 29]; ]. Supportive environments Supportive 34]. 39]; Communidades Locales para 35, 36]. ------
VIEWPOINTSPaPErS PaPErSVIEWPOINTS June 2017 •Vol. 7No. 1•010908 Black etal. mary mary Health Care at Alma–Ata in 1978 and its Declaration of Alma–Ata and worked tirelessly to achieve Pri on Conference International the for leadership provided all They 1973–1988). from WHO of eral 2010), Jim Grant (Executive Director of UNICEF from 1980 to 1995) and Halfdan Mahler (Director Gen Department of International Health at Johns Hopkins and Chair of the Expert Panel prior to his death in Taylorthe Carl of – (founder pioneers health global three the of vision the was CBPHC through – dren chil and mothers namely – family global our of members vulnerable unreachedmost the and Reaching sible forCBPHCtomore effectively neonatalandchildmortality. reduce maternal, pos- it make will that systems health strengthening for and engagement/empowerment community ing Table 2 and 6. 5. 4. 3. 2. 1. The ExpertPanelcallsforthefollowingsteps. Specific recommendations oftheExpertPanel plementation research are greatly needed. are CBPHC of aspects final address adequately not literature. the in im long–term in investments Thus, and Efficiencies available. become interventions evidence-based new as and change, contexts and tions condi as adjustment continuing require will that process long–term a is programs CBPHC effective of prove their effectiveness and to provide guidance for similar programs. The strengthening and scaling up comprehensive package of CBPHC areinterventions needed to enable such programs to continually im- than 3 years at scale is a serious Long–term concern. field studies to assess the ongoing effectiveness ofa more of period a over interventions CBPHC of broadpackage a effectivenessof of evidence limited The funding. international or central on over–dependence without sustainability long–term and effectiveness greater for potential ing of quality and coverage as well as through rigorous implementation research. This would enhance the long–term local sustainability [ and engagement/empowerment community promote that ways in realities local to adapt can that oped cannot be sustained after external funding ends. By contrast, new systematic processes need to be devel- that successes producinginitial period, time limited a only for donors external by supported been have approaches such countries poor most in interventions, community–based of components standardized are tofulfilltheirpotential. for advancing the evidence base for CBPHC program effectiveness will be essential if CBPHC programs support financial Adequate interventions. evidence–based multiple with time of period longer a over scale at effectivenessprogram CBPHC on evidence contextualized the advance continually to agenda Future progress in improving the effectiveness of CBPHC for MNCH will require an expanded research Health forAll. a system will be needed, in fact, for eventually reaching universal comprehensive health coverage and ending preventable child and maternal deaths and achieving universal coverage after of even health term, services. Such long the for needed be will system a such and developed, are they as interventions longer–term process. A strong CBPHC service delivery system will make it possible to incorporate new platform for MNCH is urgent,ery while the inclusion of other elements will need to be a gradual and tious disease outbreaks and registration of vital events). The establishment of the CBHC deliv service diseases such as hypertension, diabetes and mental illness, and for surveillance (identification of infec for ending the HIV/AIDS epidemic, controlling malaria, tuberculosis, and priority non–communicable planning, family for need unmet the reducing for also but development child and improvingMNCH progressaccelerating for in only not established be should platform delivery strongservice A CBPHC order toachievegreater impact. in mortality highest the with populations in CBPHC strengthening to given be should Prioritization tainable programs. sus and equitable, effective, with need in most those reach to order in essential is communities and systems health between partnerships Building potential. full its reach to CBPHC orderfor in fostered be to needs partnership and participation full their and resource, undervalued an are Communities expanding appropriately. is CBPHC for funding that ensure to leaders political and makers policy by given be should tention The amount of resources devoted to CBPHC should be tracked at national and regional levels, and at deaths. universal healthcoverage,andforendingpreventable childandmaternal achieving in progress accelerating for systems, health strengthening for priority a be should CBPHC Table 3 provide additional detailed to the recommendations of the Expert Panel for promot
25]. Different approaches to scaling up should be tested through monitor 440 www.jogh.org • doi:10.7189/jogh.07.010908 ------www.jogh.org Table 3. child health Table 2. effective system healthdelivery Build a stronger, more efficient, and more at greatest risk Promote delivery of interventions to those munity andhousehold com every to system delivery the Extend m tion uation, and use of health–related informa Involve communities in monitoring, eval community andthehealthsystem the between partnerships stronger Build andchildren newborns gaged in improving the health of mothers, these communities to be more actively en in women and communities Empower m aIn aIn
recommendatIons recommendatIons Expert Panelrecommendations forstrengtheningsystemimproved neonatalandchildhealth thedelivery maternal, Expert Panelrecommendations forpromoting communityengagement/empowerment forimproved neonataland maternal, • doi:10.7189/jogh.07.010908 not remain an“unfunded afterthought.” strong CBPHC programs is urgently needed, as is defining the resource needs so that these programs will betray the future of our children and grandchildren” [ indeed shall we justice, distributive of spirit the in equity economic and social for battles global and cal his 2008 address to the 61st World Health Assembly, “unless we all become partisans in the renewed lo us that “morality must march with changing capacity” [ [ However, afterthought” “unfunded an it, calls remains,rightly still colleagues CBPHC and El–Saharty as health ofmothers,neonatesandchildren hasgrown exponentially. [ workers at all levels, including care needs to be brought “as close as possible to where people live and work” and that this requires health that vision, which remains unfilled. They recognized, and the Declaration of Alma–Ata affirms, that health 45] (p. 270) rather than the solid foundation of effective health systems. Jim Grant repeatedly reminded 14]. Over the past three decades, the evidence of what can be achieved through CBPHC to improve the - - - - d d these levelsare appropriate giventheimportanceofCBPHCforavertingdeaths. that ensure and hospitals and centers health primary for those against CBPHC for expenditures Monitor systemcapacityandlocalneed). atthecommunity level(basedondelivery Integrate services Provide delivery. adequatesuppliesforservice strengthening healthcare primary atperipheralhealthfacilities. for and programs community–based of support for level local and community the at investments Foster of community–basedprograms inrelation totheamountbeingspentforfacility–basedcare. Provide adequate, sustainable and flexible global, national and local financing that responds to the needs strategies thatidentifyandreachCreate delivery thoseingreatest equitableservice need ance systemstoremunerate providers andincentiveschemestopromote utilizationofhealthservices). Provide “safety nets” that reduce barriers to accessing and providing services (eg, “CBPHC–friendly” insur vidual practitioners,includingtraditionalhealers). Coordinate the activities of the formal health sector with the informal health sector (drug sellers and indi required foreachtask,giventhedistancetohomesandlevelofremuneration/ incentives). the same time ensuring a suitable workload for an appropriate number of tasks and ensuring enough time at (while intensity service required the for workers community–level of balance appropriate an Develop Train andsupportneighborhood volunteersforpeer–to–peerhealthpromotion. needs andthataddress theepidemiologicalprioritiesofmothersandtheirchildren. health local respondto that tasks health perform to work) will wherethey communities from the by and selected are preferably (who workers community–level of supervision and training appropriate Provide the nearest healthfacility, and(3)are accountabletotheirlocalcommunity. at based stafffrom supervision technical and supportive appropriatereceive (2) involvement, long–term Train and support community–level workers who (1) receive sufficient incentives or salary to support their ofservices. Involve communitymembersinthedelivery sessments ofmortalityimpact. as including programs, CBPHC of evaluation and monitoring the to approaches participatory Develop equity inallstagesofhealthcare). processpromote continuing to a of part as services, of greatestneed in those of identification and deaths Create systems for the community’s generation and use of health data (including registration of births and be accountableforhealthsystemperformance. Create bi–directional linkages between the district health system and communities that can help everyone Create bi–directional communicationflows. Create ahealthsystemculture thatisrespectful ofandcollaborativewithcommunitymembers. planning, implementation,andevaluationoflocalhealthprograms. the on and needs health local on focused organizations community–based of development the Support empowerment, support of micro–credit programs and development of conditional cash transfer programs). government health services, and through building the agency of women (such as the promotion of women’s local of aspects certain controlling or supervising in voice a communities giving through capacity munity com build to processes and communities with partnerships supports that values of foundation a Establish eta eta Ils Ils “physicians, nurses, midwives, auxiliaries and community workers as applicable” 441
47]. Establishing the political will to fund and build 46]. And Halfdan Mahler reminded the world in June 2017 •Vol. 7No. 1•010908 CBPHC summaryandrecommendations - - - - -
VIEWPOINTSPaPErS PaPErSVIEWPOINTS June 2017 •Vol. 7No. 1•010908 Black etal. rEFEr ENCES 6 5 4 3 2 1 J GlobHealth.2017;7:010906. nity-based primary health care in improving maternal, neonatal and child health: 6. strategies used by effective projects. FreemanS, P.Gupta B, Rassekh H, Perry Comprehensiveregarding evidence reviewthe of effectivenessthe commu of fects. JGlobHealth.2017;7:010905. ef- equity 5. health: child and neonatal maternal, improving in care health primary community-based of effectiveness SchleiffFreemanR, Kumapley M, P, ComprehensiveH. review regardingevidence Perry the B, of Rassekh S, Gupta the ings. JGlobHealth.2017;7:010904. tiveness of community-based primary health care in improving maternal, neonatal and child health: 4. child health find Freeman P, Schleiff M, Sacks E, Rassekh B, Gupta S, Perry H. Comprehensive review of the evidence regarding the effec health findings.JGlobHealth.2017;7:010902. neonatal 3. health: child and neonatal maternal, improving in care health primary community-based of effectiveness Sacks E, Freeman P, Sakyi K, Jennings M, Rassekh B, Gupta S, et al. Comprehensive review of the evidence regarding the health findings.JGlobHealth.2017;7:010902. the effectiveness of community-based primary health care in improving maternal, neonatal and child health: 2. maternal Jennings M, Pradhan S, Schleiff M, Sacks E, Freeman P, Gupta S, et al. Comprehensive review of the evidence regarding database description.JGlobHealth.2017;7:010901. and methods rationale, 1. health: child and neonatal maternal, improving in care health primary community-based of WilhelmP.FreemanS, J, Gupta B, Rassekh H, Perry regardingeffectivenessevidence the the of reviewComprehensive local, national and international policiesandprograms.local, nationalandinternational developed through which we can more effectively from learn experience and generate evidence to guide be to need mechanisms Specific programming. CBPHC stronger to contribute all will implementation national and local levels, exchange of information and experiences, training, and evaluations of program important in promoting stronger CBPHC implementation world–wide. Advocacy at global, international, [ vices [ vices ser planning family of coverage synergiesthe and expanding points for programs PHC createentry will [ UNICEF and WHO by convened Care Health Primary on Conference International many other countries) [17], and eventually achieving Health for All as initially envisioned in 1978 at the and WorldUNICEF Organization,the Health by for called (as 2030 by deaths maternal and child able for by the Sustainable Development Goals recently adopted by the United Nations) [ based interventions will be essential for achieving universal coverage of health services by 2030 (as called Stronger CBPHC programs that foster community engagement/empowerment and implement evidence– CONCLUSIONS tually achieveHealthforAll. great challenges for global health in the 21st century and one of the giant steps that can be taken to even the of one is be to needs it that priority the CBPHC making and evidence this on Building deaths. child orities” [34]. The evidence confirms the promise of CBPHC in ending preventable maternal, neonatal and pri mutual around action joint in communities with together come resources and knowledge relevant Carl Taylor, in his final publication, wrote that “[r]eal social change occurs when officials and people with ], and hypertension and other chronic diseases [ diseases chronic other and hypertension and [52], malaria 51] official viewsofUSAID. arenot paper this expressedin views the and (USAID), Development International for Agency States United the officialan as representative of not individual, an as was Panel Expert the of member a as Kureshy'sparticipation Nazo interest. of conflict no declare and author), corresponding the from request upon (available closure.pdf Conflict of interest: and approved thefinaldraft. Authorship declaration: HP and RB wrote the first draft. All of the authors participated in a revision of first draft Funding: None. as ChairoftheExpertPaneluntilhisdeathin2010. care,health project,this of support early his primary community–based in leadership of time leadership his and Acknowledgments: The authors express the gratitude and indebtedness to the late Dr Carl E. Taylor for his life ] and for accelerating progress in the detection and treatment of HIV/AIDS [ HIV/AIDS of treatment and detection the in progress accelerating for and 49]
All authors have completed the Unified Competing Interest Form at www.icmje.org/coi_dis- 442 ]. International cooperation will be be will cooperation International 53]. www.jogh.org • doi:10.7189/jogh.07.010908 48 ]), ending prevent ]. Stronger CB- Stronger 14]. ], tuberculosis 50], ------www.jogh.org rEFEr ENCES • doi:10.7189/jogh.07.010908 12 8 7 29 28 11 9 13 10 14 17 16 15 30 18 22 21 20 19 26 27 23 24 25 ress for maternal, newborn, and child survival. Lancet. 2016;387:2049-59. Lancet. survival. child and newborn, maternal, for ress Victora CG, Requejo JH, Barros AJ, Berman P, Bhutta Z, Boerma T, et al. Countdown to 2015: a decade of tracking prog term impactonmortalityinchildren youngerthanfiveyearsofage.JGlobHealth.2017;7:010907. long- of evidence with programs 7. health: child and neonatal maternal, improving in care health primary nity-based FreemanS, P.Gupta B, Rassekh H, Perry Comprehensiveregarding evidence reviewthe of effectivenessthe commu of who.int/pmnch/Countdownto2015FINALREPORT-apr7.pdf. 2017. Accessed:26February Available:2008. http://www.Report. 2008 UNICEF.The Survival. Child and Newborn Tracking Maternal, in Progress 6736(15)00519-X hlrn hog cmuiybsd approaches. community-based through children hood Illness. Freeman 2017. CBPHC%20FINAL.pdf. Accessed:26February http://www.mchip.net/sites/default/files/USAID%20Available:2010. Survival. Child Promoting for Strategies livery wordpress.com/2009/08/finalcbphcreporttoerp-7july2009.pdf. 2017. Accessed:26February Available:http://www.coregroup.org/storage/documents/finalcbphcreport_july2009.pdf http://aimdb.files. 2009. and Association; Health Public American Panel: Review Expert the to Report and Findings Summary Children? of Health Nyonator Bryce J Perry H UDHR/Documents/UDHR_Translations/eng.pdf. 2017. Accessed:26February Available:Assembly.1948. General Rights. Nations http://www.ohchr.org/EN/Human United of Declaration Universal doi:10.1080/17441690903330305 Perry Available: http://www.who.int/publications/almaata_declaration_en.pdf. 2017. Accessed:27February World Health Organization. UNICEF. Declaration of Alma-Ata: International Conference on Primary Health Care. 1978. Chan mentgoals. Accessed:26April2017. UN. Sustainable Development Goals. 2015. Available: https://sustainabledevelopment.un.org/topics/sustainabledevelop 6-12 September1978;1978;Alma-Ata,USSR.Geneva,Switzerland:World HealthOrganization; 1978. World Health Organization. UNICEF, editors. Declaration of Alma-Ata. International Conference on Primary Health Care, services initiative for scaling up service delivery innovation. delivery service up scaling for initiative services Krumholz doi:10.1093/heapol/czi003 lic Health.2015; evidence-based strategy of community-based primary care: management perspectives from northern Ghana. Glass S0140-6736(12)60908-8 Black andChildHealth.Washington,born, DC:World Bank; 2016. TemmermanR, New- Volume.Laxminarayan WalkerThis Maternal, Reproductive,R, M, of editors. Black sages N, In: randomised controlled trial. cluster-a Bangladesh: district, Sylhet in strategies throughimplemented package careintervention service-delivery two R Black 2017. Thrive, Transform. 2016. Available: http://www.who.int/pmnch/media/events/2015/gs_2016_30.pdf. Accessed: 26 April World Health Organization. The Global Strategy for Women's, Children's and Adolescents' Health (2016-2030): Survive, www.usaid.gov/sites/default/files/Final-AOTC-file-v2.pdf. Accessed:26April2017. USAID. Acting on the Call: Ending Preventable Child and Maternal Deaths: A Focus on Equity. 2016. Available: https:// line:22695930 zation incidenceanalysis. Malik line:20653970 Baqui Lassi key messagesfrom DiseaseControl Priorities3rd Edition.Lancet. 2016;388:2811-24. according to health care need: evidence from the World Health Survey.WorldHealth the from evidence need: care health to according pdf?sequence=3&isAllowed=y https://openknowledge.worldbank.org/bitstream/handle/10986/23833/9781464803482. Available: Bank. WorldWashington,DC: Edition. Third Health, Child and Newborn, Maternal, Reproductive, Priorities: Control ease World Health Organization. Multiple micronutrient supplementation during pregnancy.during Available:http://www.supplementation 2017. micronutrient Multiple Organization.World Health Szwarcwald who.int/elena/titles/micronutrients_pregnancy/en/. 2017. Accessed:27February H
M ZS, Kumar ZS, R RE, Levin C RI, Guttmacher RI, N Ashraf SM, S El-Arifeen AH, , Victora , Walker, N , Freeman P , Lake , Freeman , P FK, Awoonor-Williams , Perry , AR, Stone AR, CL, Souza-Junior CL, Health PolicyPlan.2005; A doi:10.1001/jama.2012.7357 doi:10.1186/1472-6963-10-217 . Towards ending preventable child deaths. CG. Ten methodological lessons from the multi-country evaluation of integrated Management of Child 10:
, Walker N HB, Gupta HB, P , Bhutta , R Laxminarayan , , Gupta S , Gupta , . Equity in the use of public services for mother and newborn child health care in Pakistan: a utili- a Pakistan: in care health child newborn and mother for services public of use the in Equity . 366- AE, Dalaba AE,
AE, Black AE, , Darmstadt , 78. Int JEquityHealth.2016; ZA. Community-based care to improve maternal, newborn, and child health. 2016. In: Dis In: 2016. health. child and careCommunity-based improveZA. newborn, to maternal, S Medline:25436901 , Rassekh B Lancet. 2008; , Chou D
, Rassekh , SK, Rassekh SK, . Accessed: 26 February 2017. . Accessed:26February PR, Damacena PR,
MA, Phillips MA, JK, Phillips RE. Ending preventable child death in a generation. a in death child preventable Ending RE. S Ahmed GL, , Temmerman, M , Liu L 20 B . Building on the Current Evidence to Strengthen Community-Based Service De . How Effective Is Community-based Primary Health Care in Improving the the Improving in Care Health Primary Community-based Is Effective How . B Suppl 1: 371: . Accelerating progress in achieving the millennium development goal for for goal development millennium the achieving in progress Accelerating . , Temmerman M JF, Jones GN. Socioeconomic inequalities in the use of outpatient services in Brazil Brazil in services outpatient of use the in inequalities Socioeconomic GN. JF,Adongo 1936- doi:10.1080/17441692.2014.981831 illiams , W , i94- 443 15: 44. . Reproductive, Maternal, Newborn, and Child Health: Key Mes Key Health: Child and Newborn, Maternal, Reproductive, . TC, Miller
105. 120. lb ulc elh 2012;7: Health. Public Glob Medline:18539225
PB. Factors facilitating and constraining the scaling up of an an of up scaling the constraining and facilitating Factors PB. Lancet. 2012 Medline:16306075 Medline:27459961 EK, Seraji EK, , et al. Reproductive, maternal, newborn, and child health:
Health Policy Plan. 2005; Plan. Policy Health RA. The Ghana community-based health planning and HR, et al. Effect of community-based newborn- community-based of Effect al. et HR, ; 379
BMC Health Serv Res. 2010 Res. Serv Health BMC : 2119 doi:10.1016/S0140-6736(08)60835-1 Medline:26477328 doi:10.1093/heapol/czi056 doi:10.1186/s12939-016-0405-x - 20 June 2017 •Vol. 7No. 1•010908 CBPHC summaryandrecommendations . Medline:22682447
JAMA. 2012 JAMA. 400- 20: 25- 19. 34. Medline:19890758
doi:10.1016/S0140- Medline:15689427 ; 308 ; 10 :
doi:10.1016/ 141 : 217 Glob Pub- -2. . Med Med------
VIEWPOINTSPaPErS PaPErSVIEWPOINTS June 2017 •Vol. 7No. 1•010908 Black etal. rEFEr ENCES 43 42 39 44 41 40 37 45 36 35 38 34 33 47 49 48 46 31 32 51 50 52 53 uploads/2017/01/CHW-CaseStudies-Globes.pdf. 2017. Accessed:26February da, Zambia, and Zimbabwe. 2017. Washington, DC: USAID/MCSP. Available: http://www.mcsprogram.org/wp-content/ Programs: Examples from Afghanistan, Bangladesh, Brazil, Ethiopia, India, Indonesia, Iran, Nepal, Niger, Pakistan, Rwan Paredes P Lasting Change:WhenCommunitiesOwnTheirFutures. Baltimore: JohnsHopkinsUniversity Press; 2002. Arole M munities OwnTheirFutures. Baltimore andLondon:JohnsHopkinsUniversityPress; 2002. Bang ment inresource-poor settings. Chowdhury sity Press; 2002. itors. Just and Lasting Change: When Communities Own Their Futures. Baltimore and London: Johns Hopkins Univer Taylor-IdeD sity Press; 2002.p.284-95. itors. Just and Lasting Change: When Communities Own Their Futures. Baltimore and London: Johns Hopkins Univer roli, India(1993to2003). Taylor C S Arifeen El culture-of-health–lessons-from-u.html. 2017. Accessed:27February Available:http://www.rwjf.org/en/library/research/2016/08/community-health-workers-and-a-2016. Report. Learning Farmer P El-Saharty versity Press Limited;1996. Hannay line:24387091 Perry 6736(10)60449-7 H Perry doi:10.1016/S0140-6736(13)62149-2 Taylor view of their history, recent evolution, and current effectiveness. current and history,evolution, their recent of view ternal, newborn, and child health. esis invtos n elhsrie eiey n Bangladesh. in delivery health-service in innovations nerships: Rosato M doi:10.1186/1471-2288-7-10 to assess the methodological quality of systematic reviews. systematic of quality methodological the assess to H Mahler doi:10.1177/101053959100500209 Prata N 2017. abledevelopmentgoals. Accessed:26February United Nations. Sustainable Development Goals. 2015. Available: https://sustainabledevelopment.un.org/topics/sustain 2017. wha61/hafdan_mahler_speech/en/. Accessed:26February Grant DC: TheWorld Bank;2009. Vujicic M, editors. Improving in Developing Delivery Countries: Health FromService Evidence to Action. Washington, Awoonor-Williams Shea hpm.2304 impoverished region of northern Ghana. Chowdhury Contraception. 2005; Lancet. 2000; Kidane G in Bangladesh. D Neupane Glob Health.2014;2: BJ, Grimshaw AT R Zulliger HB, JP. Reaching the unreached: a miracle in the making. CE. What would Jim Grant say now? say Grant Jim would What CE. , Vahidnia, F , Scott , K J , Arole R , Bang , , Parker , Leandre F , Taylor , Laverack G , Heroux , , Morrow . Address to the 61st World61st the to AddressAssembly. . Health Available:2008. http://www.who.int/mediacentre/events/2008/ S P Kallestrup , A Christou , K Janovsky , , Taylor, C AMR, Cash AM, Chowdhury S 356: RA, Reddy RA, Lancet. 1997; doi:10.1146/annurev-publhealth-032013-182354 , Javadi , D . Jamkhed, India: The Evolution of a World Training Center. In: Taylor-Ide D, Taylor C, editors. Just and R, CE. Peru: Communities and to Learning Government Work Together. In: Taylor-Ide D, Taylor CE, ed 550-5. J , Mukherjee , Potts , M JK, Phillips DongLu D RH. Teaching motherstoprovide hometreatment ofmalariainTigray, Ethiopia:arandomisedtrial. JM, Wells . Community Health Workers and a Culture of Health: Lessons from U.S. and Global Models - A A - Models Global and U.S. from Lessons Health: of Culture a Workersand Health Community . , Rogers , e567.
72: , Grabman . How to go to scale. In: Taylor-IdeIn: scale. to go Com to When How Change: Taylor. Lasting D, and Just editors. CE, L Reichenbach , RA. A Simple Solution: Teaching Millions to Treat Diarrhoea at Home. Dhaka, Bangladesh: Uni- , McLachlan , 402-7. B Siadat , Medline:10950232 , Gergen, J HM. Home-based neonatal care: summary and applications of the field trial in rural Gadchi rural in trial field the of applications and care:neonatal summary Home-based HM. Medline:25304630 J Perinatol.2005; , Dries-Daffner, I 350: MM. Community Health Workers in low-, middle-, and high-income countries: an over an countries: high-income and middle-, Workerslow-, Health in Community MM. GA, Boers M . China's model counties: Going to scale with health care. In: Taylor-Ide D, Taylor CE, ed JF, Bawah JS, Claude M Lancet. 2001; Medline:16307960 , Islam LH, Tripathy P 169-
Lancet. 2008; F Schleimann , , Shelley , K H Perry CS, 72. MN, Islam A
, Osman , Int J Health Plann Manage. 2016 AA. Catalyzing the scale-up of community-based primary healthcare in a rural Medline:9250184 , Andersson N , Nevil P . Revisiting community-based distribution programs:distribution community-based Revisiting . areneeded? still they 25 358: doi:10.1016/S0140-6736(00)02580-0 , Crigler , L , Nair N doi:10.1016/S2214-109X(14)70303-1 372: Suppl 1: Lancet. 2010; Lancet. . Seven Case Studies. In: Peters D, El-Saharty S, Siadat B, Janovsky K, K, Janovsky B, Siadat S, El-Saharty D, Peters In: Studies. Case Seven . . Community health workers for non-communicable diseases. non-communicable for workers health Community . 404-9. FA 444 , Vaughan , Smith-Fawzi doi:10.1016/j.contraception.2005.06.059 962- K Azad , , Mwansambo C
Medline:11502340 , Hamel C S108- , et al. Case Studies of Large-Scaleof Studies Case al. et , WorkerHealth Community 71.
Asia Pac J Public Health. 1991;5:
Medline:18790319 BMC Med Res Methodol. 2007;7: Methodol. Res Med BMC doi:10.1016/S0140-6736(96)11311-8 , Islam , JP. Control of tuberculosis by community health workers 22. 375: Medline:15791272 MC, et al. Community-based approaches to HIV treat , et al. Development of AMSTAR: a measurement tool
Annu Rev Public Health. 2014 Health. Public Rev Annu
1236-7. KS, et al. Community-based approaches and part and approaches Community-based al. et KS, act 2013 Lancet. , et al. Community participation: lessons for ma ; 31: Medline:20382313 doi:10.1016/S0140-6736(01)05550-7 e273
doi:10.1016/S0140-6736(08)61406-3 - 89. ; www.jogh.org 382 Medline:26189569 doi:10.1038/sj.jp.7211278 : 2012 154- • doi:10.7189/jogh.07.010908 - 26 10. 62. . Medline:24268607 doi:10.1016/S0140- Medline:17302989 ; 35 Medline:1818613 : 399
doi:10.1002/ - 421 . Lancet Lancet Med ------
www.jogh.org and solutions with cancer:clinicians’viewsonmainproblems Prioritizing medicationsafetyincareofpeople material. The onlineversionofthisarticlecontainssupplementary Electronic material: supplementary azeem Majeed Car Papachristou 6 5 4 3 2 1 Lorainne TudorCar [email protected] Singapore 308232 11 Mandalayr Lee KongChianSchoolofMedicine Lorainne TudorCar Correspondence to: Department ofExperimentalPsychology, Health ServicesandOutcomesresearch Imperial CollegeHealthcareNHSTrust,St Department of Global Health and Population, Department ofPrimaryCareandPublic Lee KongChianSchoolofMedicine, Oxford, UK Medical SciencesDivision,Universityof Singapore Medicine, NanyangTechnologicalUniversity, Programme, LeeKongChianSchoolof Mary’s Hospital,London,UK Harvard, Boston,Massachusetts,USa Harvard T.H.ChanSchoolofPublicHealth, College London,UK Health, SchoolofPublicImperial Nanyang TechnologicalUniversity,Singapore 2,5 , CharlesVincent • doi:10.7189/jogh.07.011001 oad 2
, CatherineUrch
2 , rifatatun 1,2,3 6 , Nikolaos
3 , Josip 4 , ication errors occurred in 7% of adults and 19% of children [ cancer patients [ priate medication use or patient harm, are a serious and common threat to errors,Medication preventable as defined inappro to lead may that events patient safetypoliciesandevidencebase. implemented currently suggestions, professional care health tween be congruence the assess to opportunity an up open findings Our treatment. cancer of aspects various addressed care cancer in errors medication reducing for priorities Clinician–identified Conclusions strongest agreement amongtheclinicians. the received suggestions ranked highest The threats. safety ication safety. Prescribing stage was identified as the most vulnerable to med medication cancer improving for strategies key as education tient ance procedures (during prescription and monitoring stages) and pa proved communication betweenhealthcare providers, qualityassur im considered clinicians Overall, training. staff better and patients all for work–up pre–chemotherapy unwell, when do to what on ers ranked solutions were provision of guidance to patients and their car top The providers. care health among sharing information adequate cians’ insufficient attention to patients’ psychological distress, and in standing of treatments due to language or education difficulties, clini Results posite listofsuggestionsusingpredetermined criteria. tions. A group of 26 clinicians from the initial cohort ranked the com synthesized into a composite list of 20 distinct problems and 22 solu thematically were which suggestions their submitted providers care cancer Forty care. cancer in errors medication to, solutions for,and causes main prioritize and identify to clinicians care cancer London WestNorth invited Wethen prioritization. for criteria the and text con the scope, the determined group steering project The Methods novel priority–settingapproach. a PRIORITIZE, using WestLondon North in clinicians to according we aimed to identify medication safety priorities in cancer patient care treatment,cancer frontline study,this staffIn gap. this close help can in issues safety main the into insight unique a With strategies. tion ported, little is known about their causal factors and effective preven rebeen - careerrorshas cancer medication frequencyin of the While care. cancer in clinicians of number a of involvement the and ties plex nature of cancer treatment, the common presence of comorbidi Background The top ranked problems focused on patients’ poor under poor patients’ on focused problems ranked top The 445 Cancer care is liable to medication errors due to the com 1 , 2 ]. In an oncology outpatient department in the US, med June 2017 •Vol. 7No. 1•011001 global journal of 3 ]. A system health ------
VIEWPOINTSPaPErS PaPErSVIEWPOINTS June 2017 •Vol. 7No. 1•011001 Tudor Caretal. Figure 1. PRIORITIZE methodologyflow diagram. tiative (CHNRI) approach, to determine the main problems and solutions relating to medication safety in We developed and implemented PRIORITIZE, an adaptation of Child Health and Nutrition Research Ini PRIORITIZE andthestudyscope METHODS [ treatment cancer of quality One way of addressing this lack of evidence is by exploring clinicians’ unique insight into the safety and that couldleadtosignificantimprovements insafety. interventions specific the and errors medication cancer to factors causal main the about known is Little cludes analysis of patient medical records, incident reports and prospective observational studies [ ly focuses on rates and types of medication errors in specific forms of chemotherapy or cancers [ The evidence on cancer medication safety, ie, freedom from preventable harm with medication use, most apy andtheinvolvementofmanydifferent cliniciansinprovision ofcare [ This is coupled with a considerable burden of concurrent illnesses, a common need for supportive ther has to be administered repeatedly, according to protocols and frequently adapted to patients’ conditions. drugs, toxic potentially and potent of regimens complex Chemotherapy,nature.of dynamic consisting appropriate medications [4 atic review reported that approximately 20% of palliative cancer patients were prescribed potentially in cer care cliniciansinNorthWest London. study, we aimed to identify priorities for medication safety in care of people with cancer according to can this In routinelyon. been drawn date formally to and not has which issues safety cancer the of standing
]. Cancer treatment is highly predisposed to errors due to its multifaceted and 12 ]. Cancer care clinicians offer an important source to guide our under our guide to source important an offer clinicians care Cancer ]. 446 problems andsolutions(Box1). to guide prioritisation of collated suggestions, ie, scoring of care delivery [ health of quality improving of aim the with WestLondon North across universities leading and groups missioning zation that unifies NHS health care providers, clinical com [ and international organizations about priorities for research bodies funding policymakers, inform to extensively used cancer care (Figure 1). The CHNRI methodology has been where) [18]. Imperial College Health Partners is an organi - else (presented diagnosis delayed and safety medication to focus on two topics relating to cancer care patient safety: decided Partners), Health groupCollege steering (Imperial approval[16, provement initiative, it did not require ethics or governance im safety and quality innovative an and evaluation vice ser a deemed was study this As custodians. system health for actions c) and hospitals and organisations care health for actions b) clinicians for actions a) implementation: the tation of the top priorities categorized according to level for presen is approach this of output final The solutions. and problems viewpoints: corresponding two using priorities services delivery using clinicians’ as experts and determines ceived feedback (see Appendix S1 in S1 Appendix (see feedback ceived re the on based amended and Department our through recruited trainees and physicians care primary four of ple ication safety in cancer care. It was piloted on a smaller sam to identify the main problems and solutions relating to med clinicians for questionnaire open–ended Wean developed priorities Identifying cancermedicationsafety 13 – 15 ]. PRIORITIZE focuses on priorities in health care health in priorities on focuses PRIORITIZE ]. ]. During the study’sthe During 17]. project the stage, first 19 ]. The steering group also chose the criteria www.jogh.org 5 – 7 ]. • doi:10.7189/jogh.07.011001 Online Supplemen - Online 8 ]. It in 9 – ]. 11]. ------experts are being askedtoevaluatecompetingpatientsafetythreats. www.jogh.org Figure 2. AE AE A A =× =× Formula forcalculatingaverage expertagreement; qisaquestionthat 1 5 1 3 • doi:10.7189/jogh.07.011001 ∑ ∑ q q 5 3 = = 1 1 Ns Ns Box 1. ication delivery and the London Protocol, a framework for aa comprehensive investigation and analysis of of analysis and investigation comprehensive aa for framework Protocol,a London the and delivery ication We classifiedthecollatedsuggestionsformedicationsafetyincancercare usinganadaptedmodelofmed degree ofclinicians’agreement onpriorities.AEA wascalculatedusingtheformulain AEA is the share of scorers selecting the most common score for each research question and indicates the ( (AEA) agreement expert average the using agreement inter–rater evaluated we Instead, scoring. for used criteria different our of number the and statements some to response blank of option rater agreement in this study due to the sample size, the non–standardised categorical nature of data, the criteria and a higher overall score. Kappa statistics was deemed an inappropriate test to determine inter– the of each “Yes”for moreresponses received higher ranked were that Suggestions solutions). for three gestion was then computed as the mean of the scores for each criterion (ie, five criteria for problems and for suggestions were therefore assigned a value between 0 to 100. The overall priority score for every sug answers (“1,” “0” or “0.5”) and dividing the sum by the number of received answers. Intermediate scores calculated the intermediate scores (ie, scores for each criterion for every suggestion) by adding up all the The data from the scoring sheet was collected and analyzed with SPSS (v. 21), IBM, New York, USA. We Computation ofpriorityscores andaverageexpertagreement lected from theinitialcohortofcancercare clinicians. ipants in a form of a £50 voucher. Clinicians who performed scoring of the priorities were arbitrarily se- scoring was time demanding (an average 1 hour to complete), we offered a token payment to the partic in S2 Appendix (see suggestion” this score to confident or unsure whether or not I agree” and blank (no response) for “Unaware – I do not feel sufficiently familiar “Yes – I agree with the statement”, 0 for “No – I do not agree with the statement”, 0.5 for “Unsure – I am for 1 options: four and criteria scoring predetermined the using suggestions the categorize to clinicians text responses. Suggestions which were sufficiently similar were combined. In the second phase, we asked The collected suggestions were examined using content analysis with open coding to categorise the free– Scoring ofcancermedicationsafetypriorities leagues). We targeted oncologyconsultants,generalpractitioners,trainees,nursesandpharmacists. col- to survey the forward to asked were (participants snowballing and lists email via disseminated and version online equivalent an and paper–based a in distributed was questionnaire The Document). tary • Potentialforsavinglives–Thissolutionwouldsavelives. • Cost–effectiveness –Thissolutioniscost–effective. • Feasibility–Theimplementationofthissolutionisfeasible. For solutions: • Responsivenesstosolution–Thisincidentisamenableasolutionwithin5years. • Economicimpact–Theconsequencesofthispatientsafetythreat are costlytothehealthcare system. • • Severity–Thispatientsafetythreat leadstohighratesofmortality, morbidityandincapacity. • Frequency –Thispatientsafetythreat iscommon. For problems:
( ( er groups. Inequity – This patient safety threat affects lower socio–economic groups or ethnic minorities more than oth corers corers Scoring criteriaforprioritizationofcollatedsuggestions wh wh op op
rovide rovide Ns Ns dt dt
() () corers corers he he
most most frequent frequent 447
resp resp
o o nns nns e e ) ) Online Supplementary Document Supplementary Online cer care clinicians and received 40 complet- In the first phase we invited around 780 can r Online Supplementary Document). Online Supplementary clinical incidents [ from learn and reflect to wishing others and researchers managers, safety patient and risk clinicians, by use for incident, safety patient ESULTS People withcancerandmedicationsafety June 2017 •Vol. 7No. 1• 011001 20, 21] (see Appendix S3 in Figure 2. ]. [13]. 2) Figure ). As the the As ). - - - - -
VIEWPOINTSPaPErS PaPErSVIEWPOINTS June 2017 •Vol. 7No. 1•011001 *(Clinicians scored problems using the following criteria: frequency,criteria: following the using severity,problemsscored *(Clinicians ( solution inequity,to responsiveness and impact economic Tudor Caretal. Table 1. Figure 3. clinicians’ verbatimstatementswhich were onlyexceptionallyreworded forclarity. criteria and is ranging from 0 to 100. Higher ranked problems received more “Yes” responses for each of the criteria and a higher score). All tables use lem iscommon)”andblankfor“unaware (eg,Idonotknowifhisproblem is common)”.Total Priority score isthemeanofscores foreachofthefive scoring options were 1 for “yes (eg, this problem is common)”, 0 for “no (eg, this problem is uncommon)”, 0.5 for “unsure (eg, I am unsure if this prob r 10 Too little information on chemotherapy for patients prior to starting treatment meaning meaning treatment starting to prior patients for chemotherapy on Tooinformation little 10 ank 4 3 2 1 5 6 7 9 8 when tocontact that they do not know or recognize signs of complications or serious illness and who and p or delayedtreatments Complications of central access lines inserted for chemotherapy lead to patient morbidity propriate treatments ordelaytreatment whilewaitingfortheinformation care providers eg, palliative care team mean that the oncology team may administer inap health other by or hospitals other in given treatments on information obtain to Inability patient’s condition due to oncological problem and treatment leads to non–compliance and/or worsening of Insufficient attention to recognizing and managing serious psychological distress or illness worsening ofillness may miss treatments or not understand the importance of reporting side effects leading to difficultieseducation or language to treatmentsdue of understanding poor with Patients es (eg,preventable hospitalizations) to delayed treatment of side effects or complications with significant negative consequenc Patients have difficulty accessing acute oncology services outside of routine hours leading Toxicity orsevere allergic reactions from chemotherapy verse eventsforpatientssuchasthromboembolic events ad to restartedleading not proceduresbut for anticoagulants stopped eg, be may Drugs is notaltered andthesideeffects becomeworse Patients do not inform their oncologist of side effects meaning that the chemotherapy dose propriate drugsmaybeadministered together inap that meaning highlighted automatically not are medications between Interactions roposed Top tenmedication–related problems incancercare* Participants’ flowdiagram.
medIcat or poor understanding of chemotherapy (Table S5 in assurance processes; errors during the prescription and monitoring stage and patients’ lack of awareness hanced patientempowermentandeducation(Table S6in en and stage, monitoring and prescribing the during interventions assurance introducingquality vices, lutions overall focused on improving information integration and communication among health care ser Ion – related
problems
In
cancer
care 448 Appendix S4 in S4 Appendix (n = ed questionnaires with the majority by oncology consultants tion among clinicians and with patients; inadequate quality quality inadequate patients; with and clinicians among tion Overall, the proposed problems focused on poor communica ( threats safety medication to vulnerable most the as stage ing tients and better training of staff. Clinicians identified prescrib unwell, an appropriate pre–chemotherapy work up for all pa when do carersto their what and on arepatients to guidance ( ministered in other hospitals or by other health care providers and clinicians’ lack of access to information on treatments ad illness or distress psychological to attention insufficient ties, difficuleducation or language to due treatments of standing cer treatment according to clinicians are patients’ poor under The top ranked problems leading to medication errors in can sheets (Figure 3). ite list of suggestions resulting in 26 fully completed scoring 415 cancer care clinicians were invited to score the compos- cohort, 1 phase the From solutions. 22 and problems tinct medication safety and thematically merged them into 20 dis cancer to relating solutions 53 and problems 101 collated Table 2). Table 1). The top three solutions to medication safety threats 15, 37.5%) and specialty trainees (n trainees specialty and 37.5%) 15, Online Supplementary Document Online Supplementary Document). Online Supplementary - - - - Online Supplementary Document Supplementary Online t otal s 05PeciigPatient Prescribing 50.5 95Administering Task design 59.5 Administering 62.5 Administering/ 75.5 55Administering 55.5 35AmnseigTask design Administering 53.5 p 6Mntrn Individualstaff Monitoring 66 8Mntrn Organisation Monitoring 58 5AmnseigIndividualstaff Administering 55 2Mntrn Patient Monitoring 52 core rIorIty www.jogh.org b reakdown medIcat monitoring Ion
• doi:10.7189/jogh.07.011001 poInt
process
In =
the 15, 37.5%) (see 37.5%) 15,
). Proposed so- c ontrIbutor Patient Box 1 Box – –
). We). Factor ). The The ). ------www.jogh.org Table 2. solution isfeasible)”.Total Priorityscore isthemeanofscores foreachofthetwocriteriaandisrangingfrom 0to100.Higherrankedsolutions). 0 for “no (eg, this solution is unfeasible)”, 0.5 for “unsure (eg, I am unsure if this solution is feasible)” and blank for “unaware (eg, I do not know if this *(Clinicians scored solutions using feasibility and cost–effectiveness solutions ( r 10 11 Advise patients to contact hospital early in day if unwell to ensure to unwell if day in early hospital contact to patients Advise 11 ank 1 5 4 3 2 8 7 6 9 unwell eg,card withcontactnumbers Provide information for patients and their carers on what to do when p clinicians toensure thatappropriate treatments are given community for information written relevant have patients Ensure ortestsaretory notmissed Develop a checklist for clinicians so that important points in the his Improve trainingofstaff All patients should receive an appropriate pre–chemotherapy work up earlier Advise patients to check their temperature regularly to detect sepsis Improve thestaff:patient ratios fully aware ofthepatient’s history Enable staff to access patient records remotely so that on call staff are justments sothatdelaysindrugadministrationdonotoccur ad dose and drugs about pharmacy with Improvecommunication so drugsare notmissed Attach the chemotherapy prescription chart to the routine drug chart appropriate staff available Top tensolutionstomedication-related problems incancercare* roposed • doi:10.7189/jogh.07.011001
solutIon attention to patients’ psychological distress and poor information exchange among health care providers. safety.insufficienttreatments,clinicians’ problemsof ranked understanding top poor wereThe patients’ medication cancer improving for priorities identified Weststudy,London North this from In clinicians DISCUSSION ed as key safety priorities (Table S6 in line Supplementary Document and not knowing whom to inform, and attending their GP rather than oncology services (Table S5 in education difficulties, not informing their oncologist about the side effects, not recognizing complications Document Supplementary line (Tablesafety medication proposedrolecancer the patients’ problemsin of on Several focused in S5 tary Document). tary more on the ranking of solutions compared to the ranking of problems (Table S5 in ranked lower. Proposed solutions received higher AEA scores compared to problems, ie, clinicians agreed those to compared suggestions top the for clinicians among strongerconsensus a was there ie, AEA, est lation errors and the use of personalised medicine approaches. The top ranked suggestions had the high the chemotherapy prescribing system, the need for more frequent blood tests, chemotherapy dose calcu with issues to related overall safety medication cancer for important least as seen were that Suggestions Document). S5 inOnlineSupplementary plications from chemotherapy leading to delays in treatment or inappropriate advice or treatments (Table ceive less information on chemotherapy as well as to visit their GP rather than oncology service for com treatment from other health care providers. This group of patients was also considered more likely to re their on information to access of lack and comorbidities to inattention clinicians’ treatment, of standing Clinicians viewed patients from lower socio–economic group as more commonly affected by poor under body temperature checksandincreased physicalactivity. frequent encouraging as well as care of continuity and education providepatient to specialists nurse cal ensure to records treatment having clini unwell, of number providers,the increasingcommunity appropriate from treatmentthe of administration feeling when do to what on guidance tailored ed
For
medIcat Ion – related
problems
In
cancer
care ). Correspondingly, patient empowerment and education were highlight ). They included poor understanding of treatments due to language or or language to due treatments of understanding poor included They ). Online Supplementary Document 449 - - t Box 1 otal s 93.3 89.2 90.0 Prescribing, transcribing,dispensing, 91.7 92.5 75Peciig oioigTask design 85.8 Prescribing, monitoring 86.7 87.5 85.8 84.2 84.2 p core rIorIty ). The scoring options were 1 for “yes (eg, this solution is feasible)”,
Prescribing, transcribing, dispensing, Prescribing, transcribing,dispensing, administering, monitoring administering, monitoring b reakdown medIcat Administering Administering Transcribing Monitoring Monitoring Monitoring Prescribing Prescribing Ion People withcancerandmedicationsafety
poInt June 2017 •Vol. 7No. 1•011001
process ). Pertinent suggestions includ
In
the
Online Supplemen- environment environment r Task design Task design Task design elated Working Working Patient Patient Patient Patient barrIer Team
deFense On- On-
------
VIEWPOINTSPaPErS PaPErSVIEWPOINTS June 2017 •Vol. 7No. 1•011001 Tudor Caretal. cy), showed reduction in the incidence of errors in cancer patient [ tified solutions in our study (eg, enabling remote access to patient records and closer links with pharma- clinician–iden the of some with aligned reconciliationmedication pharmacist–led on trial intervention, cancer care islacking[27]. in errors medication reduce to interventions effective on evidence the as unclear is solutions collated er tive interventions for medication errorstive interventions in a US oncology outpatient department [ The collated suggestions, while more detailed, correspond in part to the author–nominated list of preven number ofclinicalnursespecialiststoimprove patienteducation. the increasing and activity physical increased undertake to patients encouraging unwell, if day the in ly call if feeling poorly, instructing patients to check their temperature regularly and to contact hospital ear to who and do to what provisionon tailoredincluded information of study our in respondingsolutions structure, delivery mode, potential information overload and a need for message reinforcement [ In educating patients about their cancer treatment, health care professionals should consider the content, this study has also been observed in other settings [ treatmentin threats.cancer noted safety Furthermore,in major fragmentation as studies both in lighted high been have medications home about education patient and sheet ordering standardized nication, tial role patients can have as 'vigilant partners' in prevention of chemotherapy medication errors [ lower socio–economic groups or ethnic minorities. Such lack of guidance is given concerning the essen frompatients morein as common seen was treatmentThis of complications. case in to turn to who and According to the clinicians in our study, cancer patients lack information about the potential side–effects tion andcommunicationwithpatients. medication priorities seem more focused and many addressed the need for improved sharing of informa several suggestions relating to transfer of care between different health care providers. Conversely, cancer included and scope in broader were priorities significantly. safety differ medication orities care Primary formation sharing across different health care providers and quality assurance procedures), particular pri in and communication education, patient (eg, same the were themes overarching the While [22]. care We have also used PRIORITZE to identify primary care clinicians’ medication safety priorities in primary fordable andcouldcontributetoimprovements tomedicationsafetyincancercare. agreement among the clinicians. Many identified suggestions for cancer medication safety are feasible, af ucation as key to ensuring cancer medication safety. The highest ranked suggestions received the strongest ered better communication between health care providers, quality assurance procedures and patient ed- propriate pre–chemotherapy work up for all patients and better staff training. Overall, clinicians consid ap an unwell, when do to carerswhat their on and patients wereto solutions guidance ranked top The shows that that the collective opinion of around 50 experts expressed was sufficient to reach steady find [ individual “average” an of knowledge the than accurate more be will knowledge collective the assumption, most under and cases most in that, reveal assessments validity CHNRI–focused Recent responses. of specificity the guide would which examples providing by enhanced be could this future, by inviting clinicians to identified both problems and solutions, we managed to capture relevant data. In ity or severe allergic reactions from chemotherapy”) rather than causal factors for safety issues. However, “toxic effects (eg, adverse chemotherapy–related to related study our in problemsidentified Some tool. technology information platform–agnostic a of development the through streamlined be could lutions priority–setting methodology, PRIORITIZE could be further refined and validated. The scoring of the so novel a Yet,as transparent. prioritization the and equivalent participants all of contribution larger,the is suggestions discussed of number the PRIORITIZE in approach, Delphi standard a to comparison In tothestudysetting. safety prioritiesandcustomizationofpatientinterventions different outpatient oncology clinics [ clinics oncology differentoutpatient tings. Patient safety incidents are often context–specific as reflected in a study on medication errors across While our findings correspond to the existing literature, it is unclear how applicable they are to other set ity–setting exercises involving health care professionals or employing the CHNRI methodology [ [ questions ended common in physician surveys, especially those focusing on emotionally–laden topics and including is open– rate response low The findings. our of generalizability the influenced have may which study this in part take to refusing clinicians differentthe frompotentially sample, self–selected Wesmall, a recruited Limitations
28, ]. Furthermore, the number of participants corresponds to those in other prior other in those to corresponds participants of number the Furthermore, 29]. 3 ]. The advantage of PRIORITIZE is that allows discovery of local local of discovery allows that is PRIORITIZE of advantage The ]. 450 7 , 23 – 25 ]. A recently published randomized controlled 26]. However, the effectiveness of oth www.jogh.org • doi:10.7189/jogh.07.011001 1 ]. Improved commu ]. It also also It 32]. 30– 9 20, ]. Cor ]. 21]. ]. 32]. ------www.jogh.org rEFEr ENCES • doi:10.7189/jogh.07.011001 2 3 1 National Coordinating Council for Medication Error Reporting and Prevention. About Medication Errors. 2017. AvailErrors.2017. Medication About Prevention.- and ErrorReporting Medication for CoordinatingCouncil National JCO.2008.18.6072 Walsh able: http://www.nccmerp.org/about-medication-errors. Accessed:1May 2017. 2009;27:891-6. Oncol. Clin J setting. outpatient the in cancer with children Walsh KE, Dodd KS, Seetharaman K, Roblin DW, Herrinton LJ, Von Worley A, et al. Medication errors among adults and JCO.2008.18.6072 hlrn ih acr n h otain setting. outpatient the in cancer with children ment oflocallytailored safetystrategies. volvement, evaluate patient safety culture, patient safety comparison and develop enable country–wide polices. Finally, PRIORITIZE could be used as a routine patient safety assessment tool to trigger staff’s in safety into suggestions clinician–identified of inclusion support and strategies safety medication cancer effective determine help to studies experimental robust, for need a also is There needed. is that search re or intervention of type the signpost to priorities clinician–identified collated and safety medication cancer on data organisational policies, safety cancer local of comparison include should steps Future orities anddevelopmentofsolutionstosafetyissues[43]. pri safety patient local of research,uncovering safety patient in staffcare health of inclusion greater for providerscare[ health of largenumber froma concerns safety of voicing incident reporting due to lack of anonymity, time and the risk of victimisation [ Clinicians often report feeling marginalised in patient safety policy development as well as hesitant toward remotely”), reinforcing theimportanceofcertainpriorities. ments given in other hospitals or by other healthcare providers” and “Enable staff to access patient records and stronger patient education. Some suggestions correlated (eg, “Inability to obtain information on treat procedures assurance quality of implementation providers, care cancer among integration information on focused suggestions The priorities. safety medication cancer on suggestions affordable and relevant Using a bottom–up approach with clinicians as change agents, we collated a number of concrete, locally Implications forpracticeandpolicy [ rankings on consensus and ings uncomfortable with exposing safety weaknesses [ weaknesses safety exposing with uncomfortable are professionals care health if error of causes latent detect to unable are analysis cause root as such es analyze, making it hard to spot dangerous trends or problem [ difficult to and incomplete inaccurate, be to found been has system reporting incident the producedby ORITIZE approach. sure.pdf (availableuponrequest from thecorresponding author).Theauthorsdeclare noconflictofinterest. Competing interests: The authors completed the Unified Competing Interest form at www.icmje.org/coi_disclo proved thefinalmanuscript. interpretation of the data and revised the manuscript for important intellectual content. All authors read and ap the in participated CV and JC paper.AM, the CU, of draft LTCinitial data. the wrote the analysed LTCNP and study.the designed LTC,and declaration: conceived Authorship JC collection. and data CV the performed NP NIHR ortheDepartmentofHealth.Professor CharlesVincent issupportedbytheHealthFoundation. Quality. The views expressed in this publication are those of the authors and not necessarily those Service of and the NHS, Safety the Patient for Centre Imperial the and scheme, Centre Research Biomedical NIHR the London, WestNorth for programme (CLAHRC) Care and Research Health Applied in Leadership for Collaborations the lic Health at Imperial College is grateful for support from the National Institute for Health Research (NIHR) under partment of Primary Care and Public Health, Imperial College London. The Department of Primary Care & Pub De Westthe North acrossand communities London) science health and academic the together bringing sation organi partnership (a Partners Health College from Imperial support received the financial study The Funding: Authority guidance. UK’sthe approvalaccordingto research governance or require ethics Researchnot Health did consequently and Ethics approval: This study was deemed to be a service evaluation and quality and safety improvement initiative grateful forthefundingandsupportfrom theNIHRandImperialHealthPartners. Acknowledgments: KE, Dodd KS, Seetharaman K The authors wish to thank the individuals who participated in the study. The authors are study.authors the The in participated who individuals the thank to wish authors The , Roblin ]. These promisingThese PRI 33]. the of part as verified be also could insights DW, Herrinton
Ci Ocl 2009 Oncol. Clin J 451
]. PRIORITIZE enables anonymous and structured and anonymous enables PRIORITIZE 39]. LJ, Von Worley A ; 37, 27 : 38]. Patient safety analytical approach 891 , et al. Medication errors among adults and -6. People withcancerandmedicationsafety Medline:19114695 Medline:19114695 June 2017 •Vol. 7No. 1•011001 40– ]. It corresponds to calls calls to corresponds It 42]. 34– 36]. The information
doi:10.1200/ doi:10.1200/ ------
VIEWPOINTSPaPErS PaPErSVIEWPOINTS June 2017 •Vol. 7No. 1•011001 Tudor Caretal. rEFEr ENCES 23 24 25 22 21 20 19 26 14 4 27 13 16 28 18 17 15 9 8 7 10 6 5 11 12 Hannisdal E safety hazards. Lindsay J ment of medication safety in primary care: a PRIORITIZE study.PRIORITIZE a care: primary in safety medication of ment cate prescriptions among cancer patients. cancer among prescriptions cate cancer patients: evidence to support deprescribing approaches. ment in medical management of elderly patients and potentially inappropriate drug combinations. Tamblyn doi:10.1186/s12875-016-0552-6 84. canada.org/definitions.htm. Accessed:1May2017. https://www.ismp- 2007. Available: terms. of definition Canada: ISMP Canada. Practices Medication Safe for Institute djk130 Riechelmann doi:10.1006/jsre.2002.6449 Ko line:17440151 Norton doi:10.1007/s00520-013-2098-7 H Lipczak views on main problems and solutions. Tudor Car L imperial.ac.uk/resources/C85B6574-7E28-4BE6-BE61-E94C3F6243CE/londonprotocol_e.pdf. Accessed:1May2017. Clinical Safety Research Unit ICL. Systems analysis of clinical incidents: London protocol. 2001. Available: https://www1. Preventing MedicationErrors. QualityChasmSeries.TheNationalAcademiesPress; Washington, DC: 2007. Accessed: 1May2017. are.we Available:Who Partners.2016. Health College Imperial http://imperialcollegehealthpartners.com/who-we-are/ . Vega 2011; I Rudan doi:10.3325/cmj.2008.3.307 search investments: universal challenges and conceptual framework. Vincent line:27231800 I Rudan line:25773812 ity from childhood pneumonia by 2015. by pneumonia childhood from ity reconciliation in oncological patients: a randomized clinical trial. clinical randomized a patients: oncological in reconciliation ResearchAuthority.Health NHS Research.2013. Defining http://www.hra.nhs.uk/documents/2016/06/definAvailable: - pmed.1001099 Wiebe cessed: 9March 2016. Car June_09%2529.doc. Accessed:1May2017. Documents/IRAS_v_2_3_-_Revised_collated_Project_Filter_QSG_%252829_ searchproject.org.uk/help/Help Available:2009. Guidance. ProjectCollated ResearchSystem Filter Integrated Application Guidance IRAS. www.myre ing-research.pdf. Accessed:1May2017. Rudan I doi:10.3325/cmj.2008.49.720 administration: a survey conducted among oncology nurses in Turkey. search investments: guidelines for implementation of CHNRI method. Markert A 2012; cancercare.on.ca/common/pages/UserFile.aspx?fileId=154923. Accessed:1May2017 . 2012. identification. patient ttps://www.and Available: dispensing, ordering,transcribing, chemotherapy during safety Leung M Medline:18989899 Ulas A tients: strategies to efficiently avoid chemotherapy errors in in- and outpatient treatment. Medline:19708929 2010;19:285-92. (Engl). Care Cancer J Eur literature. the of review A prevention. Schwappach j.1365-2354.2009.01127.x C Medline:8612253 LT
, Chaudhry TG-C, TG-C, Sierra-Sánchez 20: 58: , Papachristou N , Silay K
Arifeen S ER, Kaczorowski ER, C , El M Chopra , , Gibson , PG, Baker PG, , Bland R 1052-6. , Dooley M e225-8. RM, McLeod , Thierry V , Amalberti , , Knudsen , , Arianson H , Papachristou N
RP,Tannock DLB, WernliDLB, , Akinci S Jt CommJQualPatientSaf.2013; doi:10.1093/jnci/djk161 doi:10.18553/jmcp.2016.15248 doi:10.7314/APJCP.2015.16.5.1699 S Medline:22611609 Medline:21712371 S Ameratunga JL, . The need for a multidisciplinary approach to cancer care. , Baldassarre F GR. Patient safety in cancer care: a time for action. for time a care: cancer in safety Patient GR. L Kapiriri , doi:10.1002/ijc.23991 , Martin J , Kleber M , Bhutta
A Nissen JL, R PJ, Abrahamowicz M , Urch C . Safer Healthcare. 2016. Available:http://link.springer.com/10.1007/978-3-319-25559-0.Ac- 2016. Healthcare. Safer . , Dede , Braut J
IF,Wang L M , MacKay , JF, Martínez-Bautista , Gallagher J ZA, Black , Fay M . Medication errors in chemotherapy: incidence, types and involvement of patients in patients of involvement and types incidence, chemotherapy: in errors Medication . , Behrens M J Gibson , DS, Akinci , Majeed A GS, Schlichting E , Green E . Safety hazards in cancer care: findings using three different three methods. using findings care: cancer hazardsin Safety . S Arifeen El , J , Kearney A , Saad , doi:10.1136/bmjqs.2010.050856 . Why are response rates in clinician surveys declining? surveys clinician in rates response are Why . J Glob Health. 2016;6: RE, Brooks A , Samra R M Lansang Ann , , Kaizer L , El–Khatib M
, Engelhardt M MB, Sendur J Natl Cancer Inst. 2007; Inst. Cancer Natl J
, Laprise R ED, TabackED, PLoS Med. 2011 Med. PLoS
MJ, MJ, García-Martín , Barras M 39: , Bhutta , , Vinnem , Wazny K , Hertz S 452 511-6. , Chan . MAN, et al. Medication errors in chemotherapy preparation and Do too many cooks spoil the broth? Multiple physician involve- , Aylin P NA, Krzyzanowska NA, M Black ZA, . Chemotherapy safety and severe adverse events in cancer pa Medline:24294679 JE. A risk analysis of cancer care in the Norway: top 16 patient I Carneiro , . Reducing potentially inappropriate medications in palliative , et al. Safe administration of systemic cancer therapy Part 1: KY, et al. Setting research priorities to reduce global mortal , El-Khatib M
020901. Support Care Cancer. 2014 ;8: , et al. Preventing delayed diagnosis of cancer: clinicians’ e1001099 F
BMC Fam Pract. 2016 Pract. Fam BMC , Suárez-Carrascosa J Manag Care Spec Pharm. 2016 Pharm. Spec Care Manag J 99:
Croat Med J. 2008;
Croat Med J. 2008 Asian Pac J Cancer Prev. 2015 , et al. Setting priorities in global child health re health child global in priorities Setting al. et , Medline:28028437 , et al. Setting priorities in global child health re health child global in priorities Setting al. et , 592-
J Surg Res. 2002 , et al. Identification of priorities for improve J Natl Cancer Inst. 2007; Inst. Cancer Natl J . Medline:21980266 MK. Potential drug interactions and dupli- and interactions drug Potential MK. 600. doi:10.1016/S1553-7250(13)39067-9 Medline:17440160 www.jogh.org F ; , Baena-Cañada 49: ; 49
22 Int J Cancer. 2009; ; doi:10.7189/jogh.06.020901 105 ; 17 : : 307- 720 1113 • doi:10.7189/jogh.07.011001 : : 160 53
- CMAJ. CMAJ. 1996 17. 33
-9. -7.
doi:10.1371/journal. . Can Fam Physician Fam Can Medline:27852240 . ; Medline:18581609 Medline:19090596 16 Medline:24362845 Medline:12069502 ; 99: 22 doi:10.1093/jnci/ : 1699
JM. JM. Medication :
579- 734 BMJ Qual Saf. Qual BMJ doi:10.1111/ 124: - ; 154: 705 - 80. 40 . 722-8. . Med- Med 1177 Med ------.
www.jogh.org rEFEr ENCES • doi:10.7189/jogh.07.011001 32 31 30 29 43 35 34 33 42 41 40 39 38 37 36 te: xeine fo Australia. from experiences ities: R Bahl doi:10.2165/00148365-200504020-00007 Wiseman V 2011; VirusSyncytial Respiratory against children.in infections respiratory lower acute (RSV)-associated H Nair review. VanGeest born born infections by 2015. nhs.uk/news-events/news/health-secretary-launches-new-patient-safety-collaboratives.aspx. Accessed:30June2015. Improving Quality NHS. Health Secretary launches new patient safety collaboratives. 2014. Available: http://www.nhsiq. pmed.1000389 Med. 2007;167:302-8.Medline:17296888 Intern Arch algorithm. screeningelectronic an using care primary in errors diagnostic Identifying E. Thomas H, Singh tions. JPatientSaf.2014.Medline:25010194 Bishop AC, Macdonald M. Patient involvement in patient safety: a qualitative study of nursing staff and patient percep of humancollectiveopinion.JGlobHealth.2016;6:010503. properties quantitative VI. method: CHNRI the using researchpriorities health Setting S. Cousens YoshidaI, Rudan S, million deaths from birth asphyxia by 2015. PLoS Med. 2011;8:e1000389. Med. PLoS 2015. by asphyxia birth from deaths million one almost reduceresearch to Setting priorities al. et M, Ellis GL, Darmstadt BergstromZA, R, Bhutta Bahl S, JE, Lawn 2013;10:e1001508. Medline:24019762 Med. morbidity.PLoS and mortality child and maternal reduce to aiming countries: middle-income and low- in Dean S, Rudan I, Althabe F, Webb Girard A, Howson C, Langer A, et al. Setting research priorities for preconception care and middle-income countries. Bull World Health Organ. 2013;91:10-8. Hindin MJ, Christiansen S, Ferguson BJ. Setting research priorities for adolescent sexual and reproductive health in low- 2003;12 Suppl2:ii17-23.Medline:14645891 Nieva VF. Safety culture assessment: a tool for improving patient safety in healthcare organizations. Qual Saf Health Care. (US). Available: http://www.ncbi.nlm.nih.gov/books/NBK133394/. Accessed:7September2016. Safety. Quality of Cultureand Researcha HealthcarePromoting Wilson for R. LH, Agency Lubomski WeaverS, Dy SJ, port-VFinal.pdf. Accessed:1May2017. rial.ac.uk/media/imperial-college/institute-of-global-health-innovation/centre-for-health-policy/Patient-Safety-2030-Re The NIHR Imperial Patient Safety Translational Research Centre. Patient Safety 2030. 2016. Available: http://www.impe Francis R.Freedom tospeakup-Areview ofwhistleblowingintheNHS.NHSImprovement; London:2015. 11 J Martines , , Verma, Eval HealthProf. 2007; Suppl 3: JB, Johnson JB, . Comparing the preferencesthe Comparing . professionalshealth careof health setting for public prior the of members and
VR, Theodoratou VR, S30. N Ali , Medline:21501449 TP,Welch
, Bhan , Pediatr Infect Dis J. 2009; 30: W Carlo MK, VL. Methodologies for improving response rates in surveys of physicians: a systematic systematic a physicians: of improvingfor surveys Methodologies in responseVL. rates E 303- , Zgaga ,
21. pl elh cn elh oiy 2005 Policy. Health Econ Health Appl doi:10.1371/journal.pmed.1001508 Medline:17986667 doi:10.1186/1471-2458-11-S3-S30 L doi:10.1001/archinte.167.3.302 T Huda , , Chan , doi:10.1097/PTS.0000000000000123 doi:10.1136/qhc.12.suppl_2.ii17 28: 453 KY, et al. Research priorities to reduce global mortality from new from mortality global reduce to priorities ResearchKY, al. et , Simões , S43-8. Medline:19106763 Medline:27350874
doi:10.1177/0163278707307899 EAF, et al. An evaluation of the emerging interventions interventions emerging the of EAF,evaluation An al. et Medline:23397346 Medline:21305038 doi:10.1097/INF.0b013e31819588d7 doi:10.7189/jogh.06.010503 People withcancerandmedicationsafety June 2017 •Vol. 7No. 1•011001 ;4: 129-
doi:10.2471/BLT.12.107565 37. Medline:16162032
doi:10.1371/journal.
BMC Public Health Public BMC - - - - - .
VIEWPOINTSPaPErS PaPErSVIEWPOINTS June 2017 •Vol. 7No. 1•011002 health andnutritionplatforms and maternal,newborn,childadolescent implementation ofearlychildhooddevelopment Prioritizing researchforintegrated material. The onlineversionofthisarticlecontainssupplementary Electronic material: supplementary Hamadani Zulfiqar aBhutta S Mclean Pérez–Escamilla Bogard [email protected] M5G 1X8 Canada Toronto, ON 555 Universitya The HospitalforSickChildren Centre forGlobalChildHealth r Professor Zulfiqara Correspondence to: 20 19 18 17 16 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 renee Sharma obert HardingChairinGlobalChildHealth&Policy Friedman SchoolofNutritionScienceandPolicy,TuftsUniversity, Boston,Massachusetts,USa Sinai HealthSystem;LunenfeldTanenbaumresearchInstitute;DepartmentofPsychiatry,University Grand ChallengesCanada,Toronto,Ontario,Canada Institute OfPsychology,HealthandSociety,UniversityofLiverpool,UK The SacklerInstituteforNutritionScienceattheNewYorkacademyofSciences,York,USa World renew,Burlington,Ontario,Canada a Department ofSocialandBehavioralSciences,YaleSchoolPublicHealth,NewHaven,Connecticut,USa Fraser MustardInstituteforHumanDevelopment,UniversityofToronto,Ontario,Canada The InstituteofappliedHealthSciences,Universityaberdeen,UK Centro deInvestigaciónparaelDesarrolloIntegralySostenible,enSaludMaternae School ofPublicHealthandSystems,UniversityWaterloo,Ontario,Canada International CentreforDiarrhealDiseaseresearch,Dhaka,Bangladesh Maternal andChildSurvivalProgram;SavetheChildren,Washington,DC,USa Center ofExcellenceinWomenandChildHealth,agaKhanUniversity,Karachi,Pakistan Department ofPediatrics,StanfordUniversitySchoolMedicine,Stanford,California,USa National academiesofSciences,Engineering,andMedicine,Washington,DC,USa International FoodPolicyresearchInstitute,Washington,DC,USa Patan academyofHealthSciences,Patan,Nepal Centre forGlobalChildHealth,TheHospitalSickChildren,Toronto,Ontario,Canada Toronto, Ontario,Canada Infantil, andSchoolofMedicine,UniversidadPeruanaCayetanoHeredia,Lima,Peru ga KhanUniversity(South–Centralasia,Eastafrica,UK),Nairobi,Kenya
3 , GaryLDarmstadt 15
7 , atifrahman , SusanHorton
venue
1 Bhutta , MichelleFGaffey 12
1 , KerrieProulx
16 , KarleeLSilver 8 , LuisHuicho 4 , JaiKDas 11 , KofiMarfo 1 , Haroldalderman 454 5 , JosephEdeGraft–Johnson
9 , JuliaHussein 17 , DaisyrSingla 13 , VanessaMathews–Hanna 2 , DiegoGBassani 10 , StephenLye 18 www.jogh.org , PatrickWebb • doi:10.7189/jogh.07.011002 6 , JenaD 11 global , rafael journal of 1 , Kimber 14 19,20 , Mireille , health www.jogh.org • doi:10.7189/jogh.07.011002 Integration and implementation of ECD programs into the health and nutrition sectors aims to achieve achieve to aims sectors nutrition and health the into programs ECD of implementation and Integration and infants present readily accessible potential platforms for scaling up delivery of ECD interventions [6 mothers for services nutrition and health Existing years. from 2 days age 1000 to first conception the in periods, there is increasing interest in potential platforms and opportunities to deliver such interventions Although there is a range of options available for promoting ECD during the pre–school and school–age realize theirhumanpotential. fully to communities and children all allow and exclusion and poverty of cycle insidious this disrupt to social justice as well as economic urgency, and must be politically prioritized; it presents an opportunity annual reduction in income–earning potential in adulthood [ [ poverty of demic performance and limit opportunities in adulthood, thereby perpetuating an intergenerational cycle an additional 16.7% experiencing stunting [ in LMICs are not meeting basic milestones in either their cognitive or socio–emotional development, with [ potential developmental their reach to failing of risk at are (LMICs) countries middle–income and low– in children million 250 that indicates poverty extreme language, and socio–emotional competencies [ from conception to age 8 years, that is critical for development of foundational sensory–motor, cognitive, over the life course. ECD encompasses the period of early life, considered by many to include the period ly child development (ECD) does not necessarily translate into long–term health benefits and well-being targets go beyond survival, recognizing that reduction in child mortality without explicit attention to ear ating progress via the Sustainable Development Goals (SDGs) [ recommittedacceler and to globally deaths child and maternal of halving a almost witnessed world the globally,health child and close, maternal a to drew to tention 2015 As survival. on focus specific a with at- unprecedented in ushered 5, and 4 MDGs especially (MDGs), Goals Development Millennium The proaches withinexistinghealthandnutritionservices. tointegratingECDinterventions ap priority on investment and action drive to expected is agenda research generated The all. for suring inclusive and equitable quality education and promotion of life–long opportunities learning en on 4 SDG and all, for well–being and health ensuring on 3 SDG malnutrition, of forms all ing Conclusions Investing in ECD is critical to achieving several of the SDGs, including SDG 2 on end programs. of assessment quality and barriers, financial reduce to strategies cost–effectivenessand velopment, erage of integrated interventions in resource–limited settings, including: workforce and capacity de ranked research priorities varied across the life course and highlighted key aspects of scaling up cov highly– Most outcomes?”. MNCAH&N and ECD poor against protect to populations migrant and source–poor settings?”; and iii) “How can integrated interventions be tailored to vulnerable refugee re in development capacity and requirements resource human affect interventions MNCAH&N and ECD of integration the does “How ii) interventions?”; stimulation and ECD include to panded research question were: i) “How can interventions and packages to reduce neonatal mortality be ex top–ranked The 0.75. of median a with 0.90, to 0.50 from ranged scores agreement expert erage Findings The research priority scores ranged from 61.01 to 93.52, with a median of 82.87. The av each question. for calculated was score agreement expert average the and score, priority research a using ranked were equity.questions answerability,on effectThese ria: deliverability, effectiveness,and impact, diverse group of global health experts to develop and score 57 research questions against five crite a consulted and method, Initiative Research Nutrition and Health WeChild Methods the applied timeframe oftheSustainableDevelopmentGoals(SDGs). 2030 to 2015 the within development child early and nutrition health, for researchpriorities tion implementa integrated of set a identify systematically to aimed we gap, knowledge this of light In manner.integrated an in strategies such for implementation optimize to how on knowledge cient insuffi However, is there years. 2 age to conception from days 1000 first the in especially course, life the across windows sensitive within interventions (ECD) development childhood early of ery Background Existing health and nutrition services present potential platforms for scaling up deliv 4 ]. Risks to development from poverty and stunting are estimated to result in about a 25% 25% a about in result to estimated are stunting and poverty from development to Risks ]. 455 5 ]. Developmental deficits are likely to negatively affect aca 3 ]. A recent estimate based on prevalence of stunting and 4 ]. One third of preschool–aged children living living childrenpreschool–aged thirdof One ]. 6 ]. Investing in ECD is therefore a matter of 1 , 2 ]. However, unlike the MDGs, the SDG June 2017 •Vol. 7No. 1•011002 ------]. ]. - - -
VIEWPOINTSPaPErS PaPErSVIEWPOINTS June 2017 •Vol. 7No. 1•011002 Sharma etal. Figure 1. Pathway from toimproved interventions human development. lined in lined Figure 1isadaptedfrom Vaivada7 etal[ course. life the across windows sensitive within caregivers their and children at–risk reach to platforms [ interventions reachECD equitable of potentially and sustainable, as well as coverage higher ing toresearch priorityscores. researchaccord-setting priorities (v) and scores;agreement expert average and scoressearch priority re- calculating (iv) stakeholders; other from input with values societal wider to according criteria the scoring of research investment options by technical experts using the proposed criteria; (iii) weighing and listing (ii) policy–makers; and investors from input with priority–setting for criteria and context [ management case community ed topics, including but not limited to: birth asphyxia, childhood pneumonia and diarrhea, and integrat- of range wide a to applied been now has approachtransparent and systematic This researchoptions. tors in identifying research gaps and examining the potential risks and benefits of investing in different inves- and policy–makers assist to designed was method CHNRI The [11]. health in searchpriorities re- setting for methodology (CHNRI) Initiative Research Nutrition and Health Child the Weapplied Study design METHODS targets for2030. SDG the of timeline the ie, years, fifteen next the over investments research global informing of aim the with LMICs, in interventions MNCAH&N and ECD of implementation integrated the on orities pri- research top the identify to methods standardized using process consensus expert an undertook these existing programs, underscore the need for appropriate research to accelerate progress [10]. We [ LMICs in programs integrated of paucity The therein. grams come countries (LMICs), there is limited knowledge on how best to integrate and implement ECD pro- (MNCAH&N) interventions and programs in either resource–limited settings or low– and middle–in- nutrition and health adolescent and child newborn, maternal, on years recent in focus the Despite Figure 1 Figure
, this integrated implementation can be accomplished by leveraging existing delivery delivery existing leveraging by accomplished be can implementation integrated this , 12- ]. The CHNRI method involves five stages: (i) defining the defining (i) stages: five involves method CHNRI The 15]. ] andBlacketal[8 456 ]. 9 ] and lack of rigorous evaluations of evaluations rigorous of lack and ] www.jogh.org • doi:10.7189/jogh.07.011002 6 ]. As out- As ]. www.jogh.org Table 1. Effect onequity: Impact: Deliverability: Effectiveness: Answerability: c rIterIon Child HealthandNutritionResearch Initiative(CHNRI)criteria • doi:10.7189/jogh.07.011002 (v) effect on equity [ equity effecton (v) proposeduate research effectiveness; (ii) answerability; (i) questions: and impact; (iv) deliverability; (iii) mon research agenda and develop a consensus on priorities. We applied the five CHNRI criteria to eval researchersapprovecom researchallowed systematicallya ranking alsoto and questions developing of effectiveway.and integrated most the in course life acrossthe well-being and health process ment, This develop- improve might that questions priority and options investment research about researchers and research, health in donors/investors global key inform to was process consensus expert this of aim The Stage 1.Definethecontextandcriteriaforpriority–setting by 20participants. contained 57 research questions (Appendix S1 in scorecardfinal The scoring. for tool marking wereorganizeda questions into beforethe list solidated the exercise. The 27 participants mentioned above were then given an opportunity to review the con- tee compiled the questions, removing overlapping options and questions that fell outside the scope of commit- providedwereproposed.steering who researchquestionsThe participants 92 27 questions, and scores, 12 participants provided only questions, and 5 participants provided only scores. From the researchquestions providedboth experts Fifteen agreed. experts 32 which exercise,of the in ticipate globally.fields these in experts known and series cet par- to wereinvited experts formally 67 total, In and LMICs, with expertise in ECD and/or MNCAH&N. This sample included authors of relevant Lan- We targeted a purposive sample of researchers and program experts from both high–income countries Stage 2.Technical listandscore research criteria experts optionsusingpredetermined evaluate theresearch questions. s 3. 2. 1. Would yousaythatthepresent distribution ofthediseaseburden affects mainlytheunderprivilegedinpopulation? 3. Will theresults ofthisresearch leadtoasignificantandmeasurable reduction indisease burden? 2. Are theresults from thisresearch likelytoshapefuture planningandimplementation? 1. Will theresults ofthisresearch fillanimportantknowledgegap? 3. 2. 1. 3. 2. 1. 3. 2. 1. Would yousaytheresearch questioniswellframedandendpointsare welldefined? ub term (eg,10y)? Would you say that the proposed research has the overall potential to improve equity in disease burden distribution in the long implementation? proposed its research the after of results from the benefit to likely most the be would Wouldunderprivileged the that say you context ofinterest? the within sustainable be researchwould the of endpoints the that say you would coverage), high achieve to climate political and enforcement; governmental intersectoral coordination, partnership with civil society and external donor agencies; favorable monitoring regulation, government of adequacy (eg, requirements partnership and capacity government Takingaccount into would beaffordable withinthecontextofinterest? research the of endpoints the that say you would intervention, the implement to available resources the account Takinginto say thattheendpointsofresearch would bedeliverablewithinthecontextofinterest? frastructure) and users of the intervention (eg, need for change of attitudes or beliefs, supervision, existing demand), would you in transport and communication facilities, health resources, human standardizability,(eg, required infrastructure the safety), design, (eg, itself intervention Taking the of from perspective the delivery difficulty intervention of with level the account into are basedisofhighquality? If the answers to either of the previous two questions are positive, would you say that the evidence upon which these opinions posed research beeffective? Based on the best existing evidence and knowledge, would the intervention which would be developed/improved through pro posed research beefficacious? Based on the best existing evidence and knowledge, would the intervention which would be developed/improved through pro cerns? Do you think that a study needed to answer the proposed research question would obtain ethical approval without major con proposed endpointsoftheresearch? researchthe answer to proposeddesigned the be to can edge study a that say you reachwould to endpoints; and question the Based on: (i) the level of existing research capacity in proposed research and (ii) the size of the gap from current level of knowl – questIons 11]. displays the three specific sub–questions under each criterion used to to used criterion each under sub–questions specific Tablethree the displays 1 457
Online Supplementary Document Prioritizing researchforplatformintegration June 2017 •Vol. 7No. 1•011002 ) that were scored ------
VIEWPOINTSPaPErS PaPErSVIEWPOINTS June 2017 •Vol. 7No. 1•011002 Sharma etal. CHNRI exercises, average expert agreement showed a strong positive association with research priority priority research with association positive strong a showed agreement expert average exercises, CHNRI past to Similar 0.75. of median a with 0.90, to 0.50 from ranged scores agreement expert average The 82.87. of median a with 93.52, to 61.01 from ranged scores researchpriority The criteria. ority–setting present the perceived likelihood that each research question will comply with each of the five chosen pri above 85.00, and Annex I includes the complete list of ranks and scores for all 57 questions. Both tables average of 76.3% (652.4) of fields. The scorecard contained 855 fields in total, and across the 20 scorecards returned, experts completed an the studyparticipantsare summarizedinFigure 2 . 5% were program experts, 85% were researchers, and 10% were involved in both. The characteristics of both research and programing. In contrast, 25% of the 20 respondents who provided scores from LMICs, were19% programAmerica, South wereand 70% Asia, researchers,experts, werein 11% involved and proposedAfrica, that in researchexperts LMICs 27 in approximatelywerethe based questions, 26% Of r tion emailindicatedconsenttoparticipateintheexercise. data and participants provided input within their professional capacity. A positive response to the invita sensitive otherwise or personal any involve requirednot reviewnot did ethics was work formal the A as Ethics statement research priorityscore andaverageexpertagreement scores. ranging from 1 to 15. A Pearson correlation coefficient was calculated to examine the association between where culated asfollows: cal was score agreement expert average The asked. sub–questions 15 the acrossscore) common (most mode the chose who scorers of proportion average the is score agreement expert average The centage. questions. The research priority score is the mean of the scores across the five criteria, expressed as a per The research priority score and average expert agreement score were calculated for each of the 57 research Stage 4.Calculationofresearch agreement priorityscores andaverageexpert ly intheanalysis,aswefelttheywere ofequalimportance. steering committee decided not to assign weights for the present exercise. We scored all five criteria equal [ criteria CHNRI the weigh to polled was stakeholders of range wide a exercise, previous a In stakeholders. among vary may criteria scoring the of importance relative The Stage 3.Solicitinputfrom societalstakeholderstoweighthecriteria the calculationofscores. Twenty completedscoringsheets. expertsreturned a particular question, they were instructed to leave the cell blank. These blank cells were not included in formed but undecided. If the experts did not perceive themselves as sufficiently knowledgeable to answer in were they if 0.5 and no for 0 yes, for 1 score to experts asked we sub–questions, 15 the of each For • • • • • sub–questions: Experts scored each proposed research question against these five predetermined criteria, each with three
ESULTS Effect onequity:likelihoodthattheresearch wouldreduce inequity. planning andimplementation,significantly reduce theburden ofdisease. future shape gaps, knowledge crucial fill researchwould this fromresults the that likelihood Impact: tainable. Deliverability: likelihood that the endpoints of the research would be deliverable, affordable and sus ficacious andeffective. Effectiveness: likelihood that the developed intervention through the proposed research would be ef Answerability: likelihoodthattheresearch questioncouldbeanswered ethically. q is a question that experts are being asked to evaluate competing research investment options, options, investment research competing evaluate to asked being are experts that question a is
Table 2 shows the 23 research questions with a research priority score 458 ]; however, prior to scoring, the the scoring, however,to 16]; prior www.jogh.org • doi:10.7189/jogh.07.011002 ------www.jogh.org • doi:10.7189/jogh.07.011002 who proposed research questions.Theoutergraphsindicatecharacteristicsofthe20expertswhoprovided scores. principle in addressing risks to child development. Moreover, the comprehensive list of highly–ranked highly–ranked of list comprehensiveMoreover, the development. child to risks addressing in principle major a fulfill thus and development, optimize to early intervene to opportunity an provide morbidity and mortality neonatal mortality.address neonatal that Programsreduce to interventions to packages The research question that received the highest research priority score pertained to the integration of ECD and offered greater replicability and transparency thanDelphiorotherconsultativeprocesses [ options, researchinvestment competing of weaknesses and strengths the of some apparent made criteria platforms. The CHNRI method’s systematic ranking of proposed research priorities against predetermined across the continuum of care for MNCAH&N strategies as well as ECD relevant interventions and delivery experience and knowledge with experts health global of group diverse a engaged exercise present The DISCUSSION tions aboutdeterminingtheparametersforqualityassessment ofintegratedprograms (#13,17). tial delivery platform for integrated interventions (#15), and there were two highly ranked research ques grams (#4, 18, 22) were identified as top priorities. Mobile phones and media were proposed as a poten 11), responsive and complementary feeding (#6, 14), and cost–effectiveness and financial incentive pro 8, (#2, workers health community of responsibilities and development capacity to pertaining questions Moreover, 23). (#20, violence to research vulnerable children and mothers and (#10), deficits cognitive and nutritional with children (#5), infants age gestational for small (#3), workers migrant and refugees (#7), and mothers (#19, 20, 23). There was also a particular emphasis on at–risk populations, including: adolescents 23), 20, 17, 12, 10, 6, (#5, children and infants #1), (question neonates interest: of lations The 23 highest–ranked questions varied across the continuum of care, with explicit mention of all popu transfer programs?’ –received aperfectscore fortheeffect onequitycriterion. any, if cash benefits, conditional microcreditthe or programs with are ECD ‘What linking – of question migrant populations to protect against poor ECD and MNCAH&N outcomes?”. The fourth highest–ranked resource–poor settings?”; and iii) “How can integrated interventions be tailored to vulnerable refugee and in development capacity and requirementsresource human affect interventions MNCAH&N and ECD of integration the does “How ii) interventions?”; stimulation and ECD include to expanded be mortality neonatal reduce to packages and interventions can “How i) were: researchquestions top–ranked 3 The there wasstrong agreement amongexpertsaboutwhatwere considered priorityresearch questions. score, as evidenced by a Pearson correlation coefficient of 0.967 ( Figure 2. Background characteristicsofrespondents. Theinnergraphsindicatecharacteristics ofthe27experts 459 P < 0.0001). This finding indicates that Prioritizing researchforplatformintegration June 2017 •Vol. 7No. 1•011002 17]. - - - -
VIEWPOINTSPaPErS PaPErSVIEWPOINTS June 2017 •Vol. 7No. 1•011002 Sharma etal. adolescent health&nutrition and child newborn, maternal, – MNCAH&N development; childhood early – ECD agreement; expert average – AEA score; researchpriority – *RPS Table 2. question r 23 2What is the feasibility and cost–effectiveness of different 22 21 0How can intervention strategies on the prevention of vi 20 9How can maternal health interventions to improve post 19 8Where are the gaps in financing programs that aim to in 18 17 Develop and validate measures of quality and coverage coverage and quality of measures validate and Develop 17 16 Who is the most feasible and acceptable delivery agent agent delivery acceptable and feasible most the is Who 16 5How can mobile phones and/or media be most effective 15 4Does the promotion of high quality, timely complemen 14 13 What are the parameters for assessing the quality of in of quality the assessing for parameters the are What 13 12 What are effective approaches for supporting parents of areeffectiveparentsof What approachessupporting for 12 11 What is the feasibility of integrating ECD interventions interventions ECD integrating of feasibility the is What 11 10 ank 9 8 7 6 5 4 3 2 1 with ECDprograms? on the prevention of violence against mothers and children strategies intervention integrating of impact the is What interventions inresource–limitedinterventions settings? MNCAH&N and ECD integrated up scaling of models most effectively andfeasiblybeintegrated? can interventions ECD and MNCAH&N which in care of continuum the along windows critical the are What integrated withECDprograms? effectively most be children and mothers against olence ECD programs? with integrated effectively most be depression partum tegrate andsupportECDMNCAH&N? infancy andchildhood. early in interventions nutrition and ECD integrated of based settings? community– resource low in interventions integrated of MNCAH&N interventions? ly utilized as a delivery platform for integrated ECD and late intoimproved practice? tary feeding in ECD and MCHN activities actually trans tegrated ECDandMNCAH&Nprograms? that promote childnutrition,healthanddevelopment? young children (under 6 y) to adopt integrated practices and what specific interventions should beprioritized? and whatspecificinterventions workers, health community of responsibilities the into fecting lineargrowth? it possible to improve ECD outcomes with or without af nitive deprivation in the first 1000 d from conception, is For children who have endured either nutritional or cog r income countries? MNCAH&N and ECD interventions in low and middle– integrated of up scale to barriers potential are What tural contexts? cul and socio–economic diverse across strategies tion fective national ECD workforce development and reten ef of design the in required elements key the are What olescent girls? tegrated ECD and MNCAH&N interventions aimed at ad What is the most effective approach for implementing in cognitive andsocio–emotionaldevelopment? children’spromote interventions feeding responsive Do over thelong–term? outcomes developmental their improve significantly to infants age gestational for small for interventions trition How can sensory stimulation best be integrated with nu with microcredit orconditionalcashtransfer programs? programs ECD any, linking if of benefits, the are What poor ECDandMNCAH&Noutcomes? against protect to populations migrant and refugee able vulner to tailored be interventions integrated can How pacity developmentinresource–poor settings? affectterventions human resource requirements and ca in MNCAH&N and ECD of integration the does How interventions? stimulation and ECD include to expanded be mortality neonatal reduce to packages and interventions can How esearch Top 23research questionsaccording totheirachievedresearch priorityscore, withaverageexpertagreement related toeach q uestIon
------a c nswerab rI .208 .708 .58.70.75 85.07 0.85 0.88 0.77 0.84 0.92 .607 .809 .08.20.77 85.32 0.90 0.95 0.78 0.78 0.86 .509 .809 .68.20.80 85.42 0.86 0.94 0.78 0.95 0.75 .207 .608 .28.10.78 85.51 0.82 0.89 0.86 0.79 0.92 .508 .308 .88.80.76 85.78 0.78 0.86 0.83 0.88 0.95 .108 .708 .28.80.78 85.98 0.92 0.86 0.77 0.84 0.91 .208 .808 .18.00.82 86.70 0.81 0.89 0.88 0.83 0.92 .208 .009 .08.20.80 86.72 0.90 0.90 0.90 0.82 0.82 .208 .008 .28.20.80 86.82 0.82 0.89 0.90 0.82 0.92 .408 .208 .98.30.77 87.13 0.89 0.85 0.82 0.86 0.94 .808 .208 .38.20.80 87.32 0.83 0.88 0.92 0.85 0.88 .408 .209 .28.90.77 87.39 0.92 0.93 0.82 0.86 0.84 .708 .208 .38.80.82 87.68 0.93 0.89 0.82 0.87 0.87 .007 .809 .18.30.79 87.73 0.91 0.91 0.88 0.79 0.90 .907 .508 .98.20.82 87.82 0.81 0.89 88.31 0.82 0.88 0.97 89.01 0.81 0.95 89.96 0.93 0.88 0.79 0.84 0.82 0.89 0.94 0.89 90.04 0.85 0.88 0.87 0.88 90.69 0.90 0.87 0.95 0.83 0.81 1.00 0.93 90.81 0.84 0.86 0.95 0.83 0.93 0.97 91.77 0.95 0.90 0.79 0.87 0.95 0.94 93.52 0.88 0.97 0.84 0.94 0.93 0.94 0.83 0.91 0.95 0.94 0.86 0.95 0.94 0.88 0.96 I ter on IlI 1: 1: 460 ty
e c I FFect rI I ter veness on 2: 2:
d c elI rI I ter verab on IlI 3: 3: ty
c rI I I ter mpact on 4: 4: www.jogh.org
e FFect c rI I ter
on • doi:10.7189/jogh.07.011002 on
equ 5: 5: I ty
p aea rps www.jogh.org • doi:10.7189/jogh.07.011002 have an estimated median of 10.2 physicians, nurses and midwives per 10 per midwives and nurses physicians, 10.2 of median estimated an have Collectively, the 75 countries with more than 95% of the current burden of maternal and child mortality as a top priority of scaling up integrated ECD programs [ MNCAH&N This interventions. finding is highly consistent with what others have previously identified and ECD integrated implement to ability the with workers health skilled of shortage global current the reflect strategies retention and development researchcapacity addressingtop–ranked questions The blank whentheydidnotfeelsufficiently knowledgeabletoansweraparticularquestion. tions outside of their area of expertise. To avoid inaccurate scores, experts were instructed to leave the cell da. Lastly, experts might have scored questions about patient populations, interventions or health condi values; thus, eliminating the advantage of more eloquent speakers advocating for their own research agen of set standardized and transparent a against scored anonymously were questions that ensured also ria identifying experts with relevant knowledge to participate in the study. The predetermined CHNRI crite comprehensiveprocess a of employed we responsebias, effortminimize exercise.an to the In in volved in experts the of opinions the to scoreswerelimited Proposedsubsequent researchtheir and questions providedhave may they more challenges. implementation specific their to pertaining questions detailed have been different if there was greater representation of program experts or policy makers; for instance, scored and identified aspriorities. It is also possiblethat the listof highly–ranked research priorities might not included in the list of questions generated by experts. These options, therefore, could not have been were researchthat sound werethereoptions that possibility the is study present the of limitation tential an served appreciable degree of reproducibility with a sample size of only 15 persons. An additional po- ob However,still limitation. they a that be noted scorersmay be 20 should of it sample relatively small our that suggesting experts, 45–55 with achieved was priorities research ranking top of reproducibility of degree high a that found They 90. to fromincreases15 size sample researchas ranking priorities top and colleagues conducted an analysis of the CHNRI methodology [ complex process of research investment priority setting, the approach is not without limitations. Yoshida the in inherent challenges the address to attempt systematic a represents method CHNRI the Although are priorityimplementationchallengesinglobalhealth[ platforms, feature heavily across implementation–focused CHNRIs, indicating strong agreement that they themes, along with questions about harnessing the capacity of information technology and mobile health reduceto programing. health integrated of assessment quality and barriers, financial three These central LMICs, including: human resource and capacity development, cost–effectiveness and incentive schemes in interventions integrated of coverage up scaling of elements key highlighted also priorities research tage of integrating programs is the risk of overloading health services and reducing their effectiveness [ ceptability and uptake of integrated interventions. However, it has been argued that a potential disadvan ac the foster to and norms societal and cultural local to respond to positioned well are workers health Community interventions. essential to access increase to countries several in implemented successfully ters are below the World Health Organization benchmark of 22.8 per 10 and Sub–Saharan Africa, and more recently, in South Asia [ Asia recently,moreSouth and in Africa, Sub–Saharan and America Latin in decades for employed been have households to incentives monetary indirect or direct provide that platforms Support programs. transfer cash conditional or microcredit with programs ECD linking and gaps, financing identifying models, delivery different of cost–effectiveness the lar,assessing particu in implementation; integrated of aspects financial the relatedto questions Threehighly–ranked critical toinformtheintegrationdebate. ties of community health workers, and what specific should interventions be prioritized?” – is especially responsibili the into interventions ECD integrating of feasibility the is “What – eleven question Hence, understanding which programs are working and why.and working programsare into which feed understanding could research questions identified The ity and coverage of essential interventions is necessary for recognizing and reducing inequities, as well as are thus an important component of the monitoring and accountability agenda. Timely data on the qual communities with limited access to quality health services [ A disproportionately high burden of mortality and morbidity is observed among poor, rural, and remote Quality assessment of integrated programs was the central theme of two highly ranked research questions. topromoteilies partakeinplatformsandinterventions health,nutritionandECDinterventions. tions, care seeking and nutrition interventions. They can also offer exceptional opportunities to help fam tional cash transfers, also facilitate the uptake of specific and interventions behaviors such as immuniza condi as such programs, These coverage. intervention up scale and services, health to access improve poverty,alleviate to shown been have and inequities health improve to strategies implemented widely 461
21, 15, 25]. Culturally–informed quality assessments 22], lending further credibility to our results. 18, 24 19]. ]. Such financial incentive programs are programs incentive financial Such ]. 20], examining the concordance among Prioritizing researchforplatformintegration June 2017 •Vol. 7No. 1•011002
000 [1
000 people, and three–quar and people, 000 ]. Task–shifting has been 23 ]. ]. ------
VIEWPOINTSPaPErS PaPErSVIEWPOINTS June 2017 •Vol. 7No. 1•011002 Sharma etal. development ofsocieties. inclusive and sustainable the enabling toward work actively to so, doing in and opportunities; funding transparent and appropriate into recommendations these of translation the support to initiatives; ECD and protection social education, nutrition, health, between siloes of breakdown the for advocate to ers sectors. We call upon the global community of donors, researchers, policy–makers and program manag nutrition and health the with interventions ECD of integration on research implementation streaming munities. The generated research agenda is expected to be a valuable tool that drives discussion on main dows across the life course, thereby reducing pervasive inequities that exist both within and across com win sensitive within interventions development and health of impact the maximize to opportunity an cation and promote life–long learning opportunities for all. The integration of delivery platforms presents sure health and well–being for all, and SDG 4, which aims to ensure inclusive and equitable quality edu- Investing in ECD is critical to achieving a number of SDGs [ SDGs of number a achieving to critical is ECD in Investing the disproportionate burden ofpoorECDinfragilestates. tially a result of limited expertise in this area among the respondents, and it must be acknowledged given wake of the incessant conflict in the Middle East. This gap in the identified research priorities was poten to refugees and displaced populations, despite the latter now numbering in the millions, especially in the structure and exposure to stress, violence, food insecurity, infra and care child health neglect the are in greatest. disruptions The where same – applies settings humanitarian and conflict targeting questions no were there noted we mentioned, weredelivery. populations integrated high–risk via Although ventions inter ECD to from access increased benefit to likely therefore,aremost and exposures adversity to able cognitive deficits, and mothers and children susceptible to violence. These populations are most vulner tion of specific at–risk populations, such as refugees and migrant workers, children with nutritional and what is novel in our research agenda compared with other implementation–focused CHNRIs is the men [ development and growth health, in gains for potential greatest the has days, 1000 first the particularly childhood, targetearly childrensented because on be focus could This [18]. population across the entire continuum of MNCAH&N and, like the present exercise, children were the most repre To our knowledge, there has been one other CHNRI that has explored implementation research priorities agenda, indicatingspecificpriorityareas foraccelerated research. research ECD broader this on expands study present the and implementation, integrated examining of [ services nutrition and health child and newborn, ternal, plementation of interventions, with three priorities pertaining specifically to integration of ECD and ma In a recent CHNRI exercise focused solely on ECD, all top–ranked priorities related to the impact of im ments. children’sprogramsand ECD of environquality learning the of measurement the approachesto curate ac efficientproduce and feasible, to fromworld measurementaround the on expertise together pulling Worldthe CO, is Brookingsinitiative the This at Institute. Education Universal for Center the and Bank projectOutcomes and Quality Learning Early UNICEF,Measuring by the –convened (MELQO) UNES JDH, SH, LH, JH, SL, RP SL, JH, LH, SH, JDH, Research Initiative process, and critically reviewed and revised the manuscript. HA, DG, KB, GLD, JKD, JEdeG-J, study,the designed and conceptualized ZB manuscript. Nutrition the revisedand and Health Child the oversaw reviewed critically protocol,and study the designed MG revisions. subsequent made and manuscript, initial the protocol,study coordinatedthe study, designed the declaration:RS Authorship drafted analysis, the conducted Development. which isfundedbytheUnitedStatesAgencyforInternational Nutrition, for Lab Innovation Future the Feed the by provided time his for support Webbacknowledges rick Pat Children. Sick for Hospital the at Health Child Global for Centre the by funded projectwas This Funding: his contributiontotheconceptualframework. providingfor Buccioni Matthew participation, their for Zlotkin Stanley research Tylerand assistance, Vaivada for and Kristen Yee Peter Waiswa, Ruel, Marie Paul, Vinod Lenters, Lindsey Kak, Lily Daelmans, Bernadette field, Caul Laura Borrazzo, John Black, Maureen Bernis, de Luc acknowledge to Welike Acknowledgments: would sure.pdf (available uponrequest from thecorresponding author),and declare noconflictofinterest. Competing interests: The authors completed the Unified Competing Interest form at www.icmje.org/coi_disclo questions, provided feedbackduringthestudy, andcriticallyreviewed andrevised themanuscript.
E, KP,E, VM KM, H, MSM, AR, KL, DRS, and PW generated and scored priority research priority scored and generated PW and DRS, KL, AR, MSM, H, 462 ]. This finding underscores the importance importance the underscores finding This 26]. 6 ], including SDG 3 [ 3 SDG including ], www.jogh.org • doi:10.7189/jogh.07.011002 2 ], which aims to en to aims which ], ]. However,27]. ------www.jogh.org rEFEr ENCES • doi:10.7189/jogh.07.011002 22 21 20 19 18 17 16 15 13 14 12 11 10 23 9 4 6 8 7 5 3 2 1 nyas.12366 vices in Bangladesh: benefits and challenges. Ann N Y Acad Sci. 2014;1308:192-203. TofailSN, Huda B, Nahar JD, Hamadani F. ser nutrition programsand development health child into early Integrating 6736(13)60937-X science through thelife-course.Lancet.2017;389:77-90.Medline:27717614 Black MM, Walker SP, Fernald LCH, Andersen CT, DiGirolamo AM, Lu C, et al. Early Child Development coming of age: 6736(07)60032-4 development: pathways to scale-up for Early Child Development. Lancet. 2017;389:103-18. Lancet. Development. Child Early for scale-up to pathways development: Richter LM, Daelmans B, Lombardi J, Heymann J, Boo FL, Behrman JR, et al. Investing in the foundation of sustainable weight in low-income and middle-income countries. Lancet. 2013;382:427-51. Black RE, Victora CG, Walker SP, Bhutta ZA, Christian P, de Onis M, et and al. and child Maternal over undernutrition tematic review. Pediatrics.Forthcoming2017. Vaivada T, Gaffey MF, Bhutta ZA. Promoting Early Child Development with interventions in health and nutrition: a sys doi:10.1016/S0140-6736(16)31698-1 2016;13:e1002034. Medline:27270467 and middle-income countries: national, regional, and global prevalence estimates using predictive modeling. PLoS Med. low- in status developmental childhood Early al. et CR, Sudfeld MM, Black G, Danaei M, Ezzati ED, Peet DC, McCoy 2007;369:60-70. Lancet. countries. developing in children for years 5 P,Glewwe S, Cueto YB, first Cheung the S, Grantham-McGregorin potential Developmental al. et B, Strupp L, Richter ber 2015. tainable Development. 2015. Available: United Nations. Resolution adopted by the General Assembly: 70/1. Transforming our world: the 2030 Agenda for Sus nal_report.pdf. Accessed:13June2016. report – countdown to 2015. Available:2015. to countdown – report UNICEF, World Health Organization. A decade of tracking progress for maternal, newborn and child survival: the 2015 improve care for child nutrition and development. Ann N Y Acad Sci. 2014;1308:172-82. to sector health the in building Capacity al. et R, Lingam C, Arnold S, Manji B, Daelmans MA, YousafzaiRasheed AK, an.115.010348 2015;6:852-9. Nutr. Adv considerations. implementation and impact, of dence evi basis, scientific interventions: development SF.child Rao and R, nutrition Pérez-Escamilla Integrating MM, Black of humancollectiveopinion.JGlobHealth.2016;6:010503. propertiesQuantitative VI. method: YoshidaCHNRI researchthe health using Setting priorities S. Cousens I, Rudan S, Med. 2011;8:e1000380.Medline:21245907 I, et al. Setting implementation research priorities to reduce preterm births and stillbirths at the community level. PLoS KY,Chan YoungA, Rudan A, Bang George M, Constraints, and Strategies Based Community on Group Expert GAPPS Open. 2017;5:E82-9.Medline:28401123 Canadian investments in global maternal, newborn, child and adolescent health: a research prioritization exercise. CMAJ Sharma R, Buccioni M, Gaffey MF, Mansoor O, Scott H, Bhutta ZA, et al. Setting an implementation research agenda for Health. 2016;6:010507.Medline:26401271 Yoshida S. Approaches, tools and methods used for setting priorities in health research in the 21st health century. J Glob search investments:addressing valuesofstakeholders.Croat MedJ.2007;48:618-27. re health child global in priorities Setting al. et I, Rudan RE, Black S, Arifeen El M, TomlinsonChopra L, M, Kapiriri exercise. JGlobHealth.2014;4:020413.Medline:25520803 Initiative) Research Nutrition and Health (Child CHNRI a from results (iCCM): management case community grated inte for researchpriorities global Setting T,al. Jacobs et DH, K, Hamer Kallander A, Zipursky WaznyS, Sadruddin K, doi:10.1371/journal.pmed.1001446 2011;8:e1001099. Med. PLoS 2015. by pneumonia childhood from ity Rudan I, El Arifeen S, Bhutta ZA, Black RE, Brooks A, Chan KY, et al. Setting research priorities to reduce global mortal pmed.1000389 morbidity of childhood diarrhoeal disease in the next 15 years. PLoS Med. 2013;10:e1001446. Med. PLoS years. 15 next the in disease diarrhoeal childhood of morbidity Wazny K, Zipursky A, Black R, Curtis V, Duggan C, Guerrant R, et al. Setting research priorities to reduce mortality and pmed.1001099 2011;8:e1000389. Med. PLoS 2015. by asphyxia birth from deaths million one almost reduceresearch to Setting priorities al. et M, Ellis GL, Darmstadt BergstromZA, R, Bhutta Bahl S, JE, Lawn line:19090596 Med- 2008;49:720-33. J. Med Croat method. CHNRI the of implementation for guidelines investments: search re health child global in priorities Setting al. et M, Black ZA, Bhutta S, Arifeen El S, Ameratunga JL, Gibson I, Rudan doi:10.1111/nyas.12284 interventions on child development and nutritional status. Ann N Y Acad Sci. 2014;1308:11-32. Sci. Acad Y N Ann status. nutritional and development child on interventions Grantham-McGregor LCH, SM, Kagawa Fernald RMC, Walker S. Effects of integrated child development and nutrition line:23678926 2013;103:e74-82. Health. Public J Am sustainability? and up scaling about know we do what countries: income Pallas SW, Minhas D, Pérez-Escamilla R, Taylor L, Curry L, Bradley EH. Community health workers in low- and middle- nyas.12322 doi:10.3325/cmj.2008.49.720 doi:10.2105/AJPH.2012.301102 www.un.org/ga/search/view_doc.asp?symbol=A/RES/70/1 doi:10.1371/journal.pmed.1002034 http://countdown2030.org/documents/2015Report/Countdown_to_2015_fi doi:10.9778/cmajo.20160088 doi:10.1371/journal.pmed.1000380 463 Medline:27350874 doi:10.7189/jogh.04.020413 Medline:21980266 Medline:17208643 Medline:21305038 doi:10.1016/S0140-6736(16)31389-7 Medline:23746772 Prioritizing researchforplatformintegration doi:10.7189/jogh.06.010503 June 2017 •Vol. 7No. 1•011002 Medline:26875208 Medline:17948948 Medline:24571219 Medline:24571217 . Accessed: 25 Septem
doi:10.1371/journal. doi:10.1016/S0140- doi:10.1371/journal. Medline:27717610
Medline:23690756 doi:10.1016/S0140- Medline:24673166
doi:10.3945/ doi:10.1111/ doi:10.1111/ Med------
VIEWPOINTSPaPErS PaPErSVIEWPOINTS June 2017 •Vol. 7No. 1•011002 Sharma etal. rEFEr ENCES 26 25 27 24 development in the sustainable development era. Lancet Glob Health. 2016;4:e887-9. Dua T, Tomlinson M, Tablante E, Britto P, Yousfzai A, Daelmans B, et al. Global cef.org/publications/index_83078.htmlresearch. Accessed:13June2016. priorities to accelerate early child UNICEF.Available:2015. Progress2015. PromiseRenewed, Report A Survival: Child to Committing 2458-13-S3-S30 doi:10.1016/S0140-6736(07)61694-8 Maternal and child undernutrition: effective action at national level. Lancet. 2008;371:510-26. Lancet. level. national at action effective undernutrition: child and Maternal Bryce J, Coitinho D, Darnton-Hill I, Pelletier D, Pinstrup-Andersen P; Maternal and Child Undernutrition Study Group. S2214-109X(16)30218-2 ventions: a systematic review and meta-analysis. BMC Public Health. 2013;13:S30. Bassani DG, Arora P, Wazny K, Gaffey MF, Lenters L, Bhutta ZA. Financial incentives and coverage of child health inter
464
Medline:24564520 www.jogh.org Medline:27717631 • doi:10.7189/jogh.07.011002 Medline:18206224
doi:10.1186/1471- https://www.uni- doi:10.1016/ - www.jogh.org exercise byICMr contributing over4000researchideas:aCHNrI experts from256indigenousinstitutions child healthandnutritioninIndiabyengaging Setting researchprioritiesformaternal,newborn, reeta rasaily Meenu Maheshwari Kerri Wazny Shamim aQazi Igor rudan Mohapatra 5 4 3 2 1 Narendra Karora [email protected] India New Delhi–110020 Okhla Industriala F–1/5 (2ndFloor) The INCLENTrustInternational Executive Director NK a Correspondence to: Johns HopkinsBloombergSchoolofPublic World HealthOrganization,Geneva, The IndianCouncilofMedicalresearch, Centre forGlobalHealthresearch,Usher The INCLENTrustInternational,NewDelhi, Health, Baltimore,Maryland,USa Switzerland New Delhi,India Scotland, UK and Informatics,UniversityofEdinburgh, Institute forPopulationHealthSciences India
rora • doi:10.7189/jogh.07.011003
1 2 , HemaSGopalan 2 , VasanthaThavaraj
3 rea (Phase1)
, ManojKDas
4 , robertEBlack 1 , archisman 1 , rajivBahl
1 , 1 , 4 , 3 5 , , andINCLEN will takeadvantage ofthisprioritizedlist research options. tions are only valuable if they are put to use, and we hope that donors op research of However,prioritization levels. regional and national at priorities research of identification for analyses enabling odology meth CHNRI the on expanded It ways. multiple in method CHNRI Conclusions andWesternSouthern partsofIndia. in territories Union and States (iii) Westand (including Bengal); India Northern in territories Union and States (ii) States; North–Eastern and tional and three regional levels: (i) the Empowered Action Group (EAG) Reference Group. Given India’s diversity, priorities were identified at na Larger the from members 79 by assigned were criteria the to weights Relative research). on investment thinking, out–of–box and vation (answerability, criteria scoredfive against these equity,relevance, inno 122; newborn health: 56; child health: 101; nutrition: 94); 893 experts health: (maternal options research 373 into consolidated were which experts 498 from generated were ideas research 4003 – India across institutions 256 from participants engaged exercise The areas. matic the the respective refinement of research for and ideas criteria scoring specific Research Sub–Committees technically supported finalizing the tion). A National Steering Group oversaw the exercise and four theme– nutri and health child newborn, (maternal, CHNRIs theme–specific four into divided was that CHNRI umbrella large a on reports paper ers, to generate and score research options against a set of criteria. This researchusually stakeholders, of group a of intelligence collective the Methods generated andalsoexpandedonthemethodology. ideas and participants of terms methodology,in CHNRI both the of use to–date largest the was exercise The 2016–2025. of timeline the setting method for maternal, neonatal, child health and nutrition with Initiative’spriority researchResearch (CHNRI) Nutrition and Health Child the employed Trust(INCLEN) INCLEN International the and ResearchMedical of Council Indian the (ICMR) Goals, Development children and infants. In order to look forward toward the Sustainable trition. India has a high burden of morbidity and mortality in women, nu and health population for goals set achieve to fail countries why of donor’s priorities with local research need may be one of the reasons of the countries where the research takes place. This lack of alignment needs the by than rather priorities donor by driven often is (LMICs) countries income middle– and researchlow– BackgroundHealth in CHNRI is a crowdsourcing–based exercise that involves using 465 The exercise leveraged the inherent flexibility of the of flexibility inherent the leveraged exercise The June 2017 •Vol. 7No. 1•011003 global journal of health ------
VIEWPOINTSPaPErS PaPErSVIEWPOINTS June 2017 •Vol. 7No. 1•011003 Arora etal. creasingly popular and to date, over 50 CHNRI research priority setting exercises have been reported [ in- become has method CHNRI The method. CHNRI the called priorities, research setting for method systematic yet flexible a developed processes, existing in weaknesses by informed (CHNRI), Initiative [ 1999–2004 Research, Health for Forum al Challenges inGlobalHealth2003;and,(v)theCombinedApproach Matrix(CAM)toolbytheGlob Grand The (iv) 1996–2000; (ENHR), Setting Priority and Research Health National Essential the (iii) ResearchHealth (COHRED); on Development Council and The (ii) 1996; Options, FutureIntervention to Relating Research Health on Committee Hoc Ad the (i) were: these, among Prominent priorities. tify iden to methods objective and structured developing at made were attempts several report, 10–90 the tion, particularly in low– and middle– income countries (LMICs) [ alloca fund of patterns and quantum the and researchneeds health local between mismatch prevailing the on emphasized (1990) Development for Research Health on Commission the of report 10–90 The cies inmakingtheirinvestmentdecisionsmore co–aligned,efficient andimpactful[ systematic, transparent, objective and inclusive process could help policy makers and research funding agen attainment of policy goals [ context, it can perpetuate disharmony, inequity and inefficiency in health servicesand contribute to lack of “Today’s health research is tomorrow’s health service” [ of prioritization to the various constituencies of stakeholders (end–users of health research funding) [ scription” and “discovery”). The CHNRI method systematically delegates, ie, “crowdsources,” [8 instruments of research) with need for generation of new knowledge through long–term investment (“de ‘development’ and ‘delivery’ (the needs translational contextual immediate balances It process. making This method recognizes research priority setting as a multi–dimensional and multi–stakeholder decision– [ funder–driven being to vulnerable and subjectivity difficult, to is liable This tive. and far exceeds the available resources, relative prioritization among competing research options is impera inclusion ofalarge numberofstakeholdersfor representativeness. tional and sub–national (regional) levels with a 10–year reference time period (2016–2025) and through na at identified be would researchpriorities that decided was it Hence, economics. and milieu cultural socio– performance, program governance, in heterogeneity state–level and regional– with along versity di population large has India continuum. life–course the along prioritization for population target the be to identified were mothers) lactating and women pregnant including years, (15–49 women age tive reproduc- and methodology.years), CHNRI (0–18 childrenthe using Newborns, MNCHN for exercise researchsetting nationwide priority this undertake to together came 2010) since Secretariat CHNRI the apex institution for medical research in India) and the INCLEN Trust International (INCLEN; which was the (ICMR; Research Medical for Council Indian the 2030, Goals Development Sustainable upcoming the and goals, Mission Health National MDG5), (MDG4, 5 and 4 Goals Development Millennium able constrained by individual and organizational preferences. In 2011, in response to the seemingly unachiev are and metropolis the in mostly located are who experts of group small a by guided been traditionally together contribute to the largest burden of disease in India. Public health research decisions in India (MNCHN) have nutrition and health child neonatal, Maternal, statistics. health global the determine hugely India is the second most populous country in the world with many pressing health problems that, in fact, ment policies[11]. invest research influence can stakeholders local from input wherein level national the at effective be to an organization, such as to solve a problem or complete a task [ or and individual an of benefit the for tasks collective or wisdom collective of Crowdsourcinguse the is kim, Assam, Meghalaya, Tripura,Meghalaya, Assam, kim, Government (The Manipur,Pradesh); Mizoram, Arunachal Nagaland, tisgarh, Odisha, Jharkhand, Bihar, Uttar Pradesh and Uttarakhand) and (NE) North–Eastern States (Sik Chat Pradesh, Madhya (Rajasthan, States (EAG) Group Action Empowered (i) were: regions three The priorities. sub–national enable to order in regions three into grouped were territories union and States communicate andengageparticipantsfrom across thecountry. gynecology, to able hence, and multilingual was and nutrition) health public and medicine, community CLEN, New Delhi. The team had experts in the four core MNCHN disciplines (pediatrics, obstetrics and 2016. The exercise was coordinated by the RPS project management team at the Executive Office of IN The ICMR–INCLEN National Research Priority Setting (RPS) exercise was completed between 2012 and METHODS
2 , 3 ]. Given that the scope of research in health and nutrition is ever–expanding 466 3 , 6 ]. In 2006–07, the Child Health and Nutrition Research Nutrition and Health Child the 2006–07, In ]. 1 ]. If the research agenda is not aligned to local needs and 10]. The CHNRI method has been shown 5 ]. Between 1990 and 2005, following www.jogh.org • doi:10.7189/jogh.07.011003 4 ]. Prioritization using a a using Prioritization ]. 2 ]. ] the task 7 9 ]. ]. ]. ]. ------www.jogh.org Table 1. Total Editors Biomedical Journal State Program Managers(ICDS,NRHM, Directorate ofHealthServices) Technical Experts(MNCHN) Multilateral/ BilateralDonorAgencies/Foundation–Funders Policy–Decision MakersandProgram ofIndia Managers(MNCHN),Government e xpertIse Profile oftheNationalSteering Group • doi:10.7189/jogh.07.011003 ary 2016), attheconclusiontoreview,ary refine andfinalizethe results. Four keystructures were created toaccomplishthetask,outlinedasfollows. ala, Tamil AndhraPradeshandTelangana, Nadu,Karnataka, Maharashtra,Gujarat,Goa,Puducherry). Westand TerritoriesUnion and States (iii) and Bengal); Westernand Southern (Ker in country the of part Delhi, Chandigarh, Haryana, Pradesh, Kashmir,Himachal & Punjab, (Jammu territories Union and states action. EAG and NE states share similarities in MNCHN contexts and program performance); (ii) Northern focused for states EAG as indicators development and health poor with states eight identified has India of Box 1. at the initiation of the exercise to ratify the context also members of the NSG ( were RSCs four all of chairs The agencies. multilateral and donors international and national of tatives partment of Science & Technology). Its membership also included invited subject experts and represen Food and Nutrition Board), of Science and Technology and Ministry (Department of Biotechnology, De and DHR–ICMR), Ministry of Women and Child Development (Integrated Child Development Services, Services Health of General Directorate divisions, Nutrition and Health Maternal Health, Child Mission, rector of INCLEN. It included key officials from the Ministry of Health & Family Welfare (National Health the Secretary, Department of Health Research (DHR) & Director General (DG–ICMR) and Executive Di and, (iv) dissemination of the final national and regional research priorities. The NSG was co–chaired by sub–committees (RSCs); (iii) critical review, interpretation and endorsement of the results of the exercise; cluded (i) setting the rationale and contour of the MNCHN research themes; in (ii) responsibilities establishment Its exercise. of the researchfor oversight and making policy for body highest the was NSG The 1. TheNationalSteeringGroup (NSG) health andnutrition ment inacademic&research institutions. as future programs, national and international institutions & organizations funding research, research environ Translation and implementation context: society, donoragenciesandindustries. civil leadership, their and communities makers, policy functionaries, Researchers,health professionals,public Stakeholder constituencies (operating in civil, public and private sectors, health and non–health sectors): da forachievingforthcomingSustainableDevelopmentGoals2030). opment Goals 1, 4 and 5, and National Health Mission targets and the challenge of preparing the national agen- For the next ten years ie, 2016–2025 (with due consideration to unachieved Millennium Devel Millennium unachieved to consideration due (with 2016–2025 ie, years ten next the For Timeframe: able evidenceandexpertopinion. avail the per as 2012–2013 during India in Nutrition and Health Child Newborn, burdenbidity Maternal, in Conditions that together contributed to 75% of the mortality and mor and mortality the of 75% research:to for contributed together that Conditions areasconcern Major of States, StatesandUnionTerritories andWestern India,andthoseinSouthern inNorhtern India. Geography: Priorities at National and three Regional levels: Empowered Action Group States and North–Eastern (0–28days),under–fivechildren (0–59months)andchildrenborns uptotheageof18years. Target population: Women of reproductive age (15–49 years) including pregnant and lactating women, new vestment inresearch usingasystematic,transparent, inclusive, objectiveandquantitativemethod. rewarding in efficient and for nutrition and health child and newborn, maternal, in setting Priority Purpose: Context oftheINCLENICMRnationalresearch children prioritysettingexercise newborn, inmaternal, Table 1). Two NSG meetings were organized – the first (on 18th April 2013), Public and private health systems of India and their existing as well 467
( Box 1) ICMR–INCLEN CHNRIforresearchprioritysettinginMNCHN and protocol, and the second (on 4th Febru 18 a June 2017 •Vol. 7No. 1•011003 pr 78 29 15 22 3 9 21 4F 2013 eb 78 11 21 19 24 3 2016 ------
------
VIEWPOINTSPaPErS PaPErSVIEWPOINTS Figure 1. CLEN NationalResearch PrioritySettingExercise. June 2017 •Vol. 7No. 1•011003 Arora etal. Sequence ofactivitiesundertaken intheICMR–IN- through snow–ballingandinvitationsinpersonalcapacity. These were from departments of health and of women and child development. Experts were also identified exercise. the in participate to invited also were managers program and policy–makers state–level and tral Nutrition Society of India, and the Indian Dietetic Association) were also contacted for participation. Cen the Association, Health Public Indian the PreventiveMedicine, of Social Association and Indian the India, of Societies Gynecological and Obstetrics of Federation the Forum, Neonatology National the Pediatrics, of Academy Indian (the MNCHN in associations professional major of office–bearers zonal and National experience. of years 5–10 with level junior/middle are who those and faculty) ‘senior’ (ie, experience ing secureto proportionsimilar members/researchersfaculty of moreresearchwith of years 10 than teach or was effort The heads. institutional respective their through demography,contacted were agriculture) and thology,microbiology, midwifery, and statistics sciences, social sciences, home and nutrition health public gynecology,& obstetrics (eg, neonatology,pediatrics, biochemistry,medicine, community physiology, pa Faculty/researchers from departments that were directly or indirectly engaged in work pertaining to country. the MNCHN across departments and institutions from identified experts with established was network A 3. TheNationwideNetworkforcrowd sourcing definitions. RespectiveRSCspresented thestudyfindingstosecondmeetingofNSGfor review. tive refinement and consolidation of the research options (ROs) and in finalizing the scoring criteria and their ticipated in the crowdsourcing processes along with the nationwide network. They also helped in the itera- a literature search for active research contribution to respective MNCHN domains ( woman and child development), and donor agency representatives. Technical experts were identified through and (health, programspecialists scientists, social specialists), health public and scientists basic experts, ject An RSC was constituted for each of the four themes. The RSCs’ membership included technical experts (sub 2. ThethematicResearch Sub–Committees (RSCs)
from research funding organizations (n organizations funding research from from central and state governments (n ment), senior researchers (n = Develop Resource Human Development, Child Womanand Welfare,Family and Health eg, Ministries, key from reaucrats erate criteria–weighedpriority ranksfortheROs. uted relative weights to the scoring criteria which helped to gen for crowdsourcing (Table 3). departments and donor agencies participated in the two rounds health state NGOs, organizations, research institutions, ICMR colleges, medical including institutions 256 Overall, pated. respond/ logged in but did not score, and 893 (58.1%) partici not did 628 declined, 15 which of round) first the during ed activity), 1536 experts were contacted (including those contact perts contributed research ideas. For the second round (scoring perts (Table 2), 668 did not respond. Overall, 498 (42.3%) ex Of these, 12 declined to participate. Of the remaining 1166 ex in the RSCs) were identified, of whom 1178 could be contacted. first round of crowd sourcing, 1423 experts (including the 112 the manner,for this In theme. each in representation plinary and publication history to achieve equitable regional and disci expertise their to according themes four the of one into cated allo be to consented network nationwide the in members The decision makers (n makers decision policy of composed was which LRG a employed we exercise, incorporate broader societal perspectives and values within the that could, most have been conducted at a national level [ those Of group). the composing trouble to due (mostly LRG a Beyond 75% of CHNRI exercises published have not employed 4. TheLargerReference Group (LRG) Figure 1showstheschematicflowof activitieswithtimelines. Processes 468 = 24; Central and State politicians and bu and politicians State and Central 24; 17), MNCHN program managers www.jogh.org = • doi:10.7189/jogh.07.011003 Table 2). The RSCs par 24) and representatives = 19). The LRG attrib LRG The 19). 7 ]. To ------*The expertsinthesecategorieswere requested toidentifytheirtheme/component ofexpertise. www.jogh.org Table 2. RSCs) (beyond network Nation–wide (RSC) Committee Sub– Research g roup Profile of research sub–committeesandnation–widenetwork(1 • doi:10.7189/jogh.07.011003 Grand total Sub-total Technical Expertsfrom donoragencies* State program managers* health andalliedsciences)* Scientists from research institutes (public ofIndia)* Policy Makers(Government Pediatricians andneonatologists Obstetricians andgynecologists Miscellaneous* Nursing & midwifery experts Nursing &midwifery Experts from ICMRInstitutes* Dietitians &nutritionists Community medicineexperts Sub-total Basic scientists* Agriculturists Technical Expertsfrom donoragencies* State program managers* health andalliedsciences)* Scientists from research institutes (public Policy makers (Government ofIndia)* Policy makers(Government Basic scientists* Pediatricians andneonatologists Dietitians andnutritionists Obstetricians andgynecologists experts Nursing &midwifery Experts from ICMRinstitutes e xpertIse (improving ie,design,deliverability, theexisting intervention, affordability andsustainability). and systems research, including program evaluation and implementation research); and (iv) development policy (health delivery (iii) innovations); and discoveries inventions, technology interventions/package, descriptions); (ii) discovery (identification of novel pathways, discovery of novel clinical and public health pathophysiological biomarkers, factors, epidemiology,risk disease, and of etiology (burden description research: of (i) domains four the into segregated be research(RIs) all ideas that suggested also NSG The approved listofAOCsunderthefourthemesis presented in final The interventions.” health public innovative & “novel and packages,” health public composite ing exist of improvement and “impact determinants,” “social themes: the of each in AOCs additional three ed the nutrition theme into and maternal childhood nutrition components. The NSG advised to include stillbirths (most stillbirths occur in–utero and are thus are a maternal health concern). Similarly, it divid morbidity,viz., components three encompassed and health mortality maternal that suggested NSG The collectively contributedtoatleast75%ofthemortalityand morbidityburden intherespective theme. themes and presented to the first meeting of the NSG for review. The AOCs accounted for conditions that at: able literaturewereof (avail databases searched.compendium Embase the and on CINAHL Based PubMed, policy documents and reports, program reviews and grey literature for the period of 1990 to 2012/2013. India’s of Government non–indexed), and (indexed literature published Wesearched data. Indian on Extensive review of literature on burden of MNCHN related conditions was done in 2012–13 with focus Review ofliterature andidentificationofareas ofconcern ), a draft list of “areas of concern” (AOCs) was prepared for each of the RPS RPS the of each for prepared was (AOCs) concern” of “areas of list draft www.inclentrust.orga ), Mortality 5 5 5 0 1 4 5 1166 157 145 212 200 150 150 152 3 4 4 7 8 3 4 1054 142 133 186 175 141 141 136 16 87 82 74 68 44 23 86 5304 35 61 38 39 42 45 44 4 4 5 3 8 7 4 4 7 3 3 3 9 9 8 3 1 1 1 1 2 2 2 1 7 7 1 2 3 1 3 2 1 1 4 4 7 Maternal health Newborn Newborn health Maternal obdt Still- Morbidity 469 9 9 1 st round ofcrowd–sourcing) births ICMR–INCLEN CHNRIforresearchprioritysettinginMNCHN 1 2 2 2 2 1 1 t heme Table 4. ( health wIth 1 122 111 52 21 112 15 12 26 25 012 10 313 13 5 5 1 2 3 3 4 4 1 2 1 2 1 2 1 1
components June 2017 •Vol. 7No. 1•011003 health Child Child ) aenlChild- Maternal 17 33 66 33 33 4 6 1 8 5 3 9 8 urto Total Nutrition hood 7280 47 2 2 3 1 242 13 22 38 58 11 23 10 11 26 15 17 9 1 5 1 1 - - -
VIEWPOINTSPaPErS PaPErSVIEWPOINTS June 2017 •Vol. 7No. 1•011003 Arora etal. *States are ordered according totheregion/territory. Table 3. ternal health (n health ternal by INCLEN for submission of RIs by the network. The software had seven separate electronic forms: ma Solicitation of research ideas (RIs) from the Nationwide Network: F Crowdsourcing closely examined each RI and rephrased, split, and combined the RIs (as required) keeping the core idea Refinement of the research ideas: The RPS project management team at INCLEN along with the RSCs 3497 RIswere obtained across theMNCHNthemesfrom 498 experts(42.3%participation). of total A internet. the accessing in difficulty had who participants with shared and prepared was form the of version offline An sessions. multiple over completion for allowed forms electronic The domain. livery and development). The expert was not limited in the number of RIs s/he could submit under each researchof discovery,domains (description, four the in RIs contribute to AOCs two any select to ed de tutorial. The list of AOCs was then displayed on his/her computer screen and the participant was instruct power–point self–orientation a through taken then was S/he territory,ID). state/union email alternative asked to enter personal details (name, area(s) of work, employment status (working/retired), institution, was s/he logging–in, him/her.After to pre–assigned as forms electronic seven the of one only to in log with an and individualized passwordlog–in username for the dedicated software. The participant could provided was participant crowdsourcing. Each of round first the of purpose the with them provide the the research priority setting exercise; (ii) to provide them the context and scope of the exercise; and, (iii) of method the about them sensitize to (i) phone: the over contacted then and email initial an sent were (n nutrition = Kerala Karnataka Andhra Pradesh Tamil Nadu Puducherry Rajasthan Maharashtra Gujarat Goa Grand total Madhya Pradesh Jharkhand Chattisgarh Bihar Uttarakhand Uttar Pradesh Punjab Jammu &Kashmir Himachal Pradesh Haryana Delhi Chandigarh West Bengal Tripura Odisha Sikkim Nagaland Meghalya Manipur Assam s tate irst / u nIon
Profile ofparticipatinginstitutionsintheNationwideNetwork* round t errItory 2: maternal & child nutrition) themes]. The experts in the nationwide network and RSCs RSCs and network nationwide the in experts The themes]. nutrition) child & maternal 2:
= oF
3: mortality, morbidity & stillbirth); newborn health (n health newborn stillbirth); mortality,& 3: morbidity
crowdsourcing colleges m 6 15 169 edIcal 15 15 14 19 10 10 14 11 9 3 1 8 1 1 1 4 4 5 2 1 2 1 8 2 2 1 7 1 1 1 3
InstItutIons Icmr 1 1 3 1 1 470 o research ther
publIc 38
1 6 3 1 1 1 2 3 1 1 8 2 3 1 3 1 InstItutes
health
n tal on
organIzat – governmen 12 1 2 1 1 5 2 Ions - An online software was designed www.jogh.org ( health s = tate 1); child health (n health child 1);
and departments 19 1 2 2 2 1 2 2 1 1 1 1 1 1 1
• doi:10.7189/jogh.07.011003 nutrItIon
) agencIes d onor 3 2 7 1
= 1); and and 1); total 256 17 18 25 12 26 13 14 14 25 15 14 11 2 2 1 1 4 3 3 8 3 2 3 1 1 2 1 8 - - - www.jogh.org Table 4. Child HealthTheme: HealthTheme:StillbirthComponent Maternal MorbidityComponent: HealthTheme:Maternal Maternal MortalityComponent: HealthTheme:Maternal Maternal 14 13 12 11 10 15 14 13 12 11 10 13 12 11 10 9 8 7 6 5 4 3 2 1 9 8 7 6 5 4 3 2 1 8 7 6 5 4 3 2 1 9 8 7 6 5 4 3 2 1 (RBSK) –ChildHealthScreening Services] andEarlyIntervention “Novel & Innovative” composite public health packages with potential impact on Under–5 Mortality Rate [eg, Rashtriya Bal Swasthya Karyakram ChildandAdolescent Health(RMNCH+A)] Newborn, “Existing” composite public health packages with potential impact on Under 5 Mortality Rate [eg, IMNCI, F-IMNCI, Reproductive, Maternal, Social determinantsofunder5mortalityrate[eg,immunizationrefusal, inappropriate feedingpractices,poorhealthseekingbehavior.] Others (Pleasespecify____) CP, neuro–motor impairment,audio–visualimpairment] Neuro–developmental disorders (NDD) [eg, early developmental delays, autism, speech & language disorders, intellectual disability, epilepsy, Other infections¶siticdiseases Meningitis/encephalitis Acute bacterialsepsis Unintentional injuries Malaria Congenital anomalies Measles andvaccinepreventable diseases Diarrheal diseases Pneumonia (andARI) “Novel &Innovative”compositepublichealthpackageswithpotentialtoinfluencestillbirths[eg,Innovativesolutionspromote accesstocare] (JSSK)] “Existing” compositepublichealthpackageswithpotentialtoinfluencestillbirths[eg,JananiShishuSurakshaKaryakram Social determinantsofstillbirths[eg,prevailing harmfultraditionalbirthpractices,lackofwomens’empowerment,poverty, illiteracy] Non-health factors[eg,Indoorairpollution,tobaccosmoke] placentalandfetalconditions] Unexplained [Byknownmaternal, Complications ofplacenta,cord andmembranes Intra partumcause:Obstetriccomplications Intra partumcause:Acutehypoxicinsult Fetal cause:Congenitalmalformations Fetal cause:Pre–term birth Fetal cause:Intrauterinegrowth restriction BMI,gestationaldiabetes] malnutrition[eg,lowmaternal cause:Maternal Maternal disease, lupus] Maternal cause: Underlying chronic maternal illness [eg, chronic hypertension, epilepsy, liver disease, diabetes mellitus, renal disease, thyroid infectionsinpregnancy [eg,TORCHgroup cause:Maternal Maternal ofinfections] cause:HypertensivedisordersMaternal ofpregnancy cess tocare] “Novel & Innovative” composite public health packages with potential impact on maternal morbidity [eg, innovative solutions to promote ac childandadolescenthealth(RMNCH+A)] newborn, ductive, maternal, RTIReproof syndromicSTI; management [eg, & morbidity maternal on impact potential with packages health public composite “Existing” Economic [eg,impoverishmentandpoverty] caste, religion, teenagepregnancy, culturalpractices] status; community of loss violence, disharmony,domestic marital dissolution, stigmatization, household isolation, divorce, social [eg, Social Strong fearofpregnancy andchildbirth Post partumdepression andpsychosis Post partummorbiditiesandlongtermdisabilities[eg,obstetricfistula,utero–vaginal prolapse, incontinence, dyspaerunia,infertility] urinary morbidities(SAMMs) andNearmissevents Severe acutematernal “Novel &Innovative”compositepublichealthpackageswithpotentialimpactonMMR[eg,Innovativesolutionstopromote accesstocare] (JSSK)] “Existing” compositepublichealthpackageswithpotentialimpactonMMR[eg,JananiShishuSurakshaKaryakram Economic [eg,impoverishmentandpoverty] violence, lossofcommunitystatus;caste,religion, teenagepregnancy, culturalpractices] disharmony, domestic marital dissolution, stigmatization, household divorce,isolation, social [eg, mortality maternal of determinants Social Others (Pleasespecify______) Malaria inpregnancy Medical disorders inpregnancy [eg,chronic hypertension,epilepsy, liverdisease,diabetes mellitus,renal disease,thyroid disease,lupus] Anemia andothernutritionalproblems Unsafe abortion Obstructed labor Sepsis Hypertensive disorders ofpregnancy Hemorrhage Areas of concern finalizedbytheNationalSteeringGroupAreas ofconcern • doi:10.7189/jogh.07.011003 471
ICMR–INCLEN CHNRIforresearchprioritysettinginMNCHN June 2017 •Vol. 7No. 1•011003 - -
VIEWPOINTSPaPErS PaPErSVIEWPOINTS June 2017 •Vol. 7No. 1•011003 Arora etal. Table 4. Nutrition Theme: Maternal NutritionComponent Nutrition Theme:Maternal Nutrition Theme:ChildhoodComponent Newborn HealthTheme: Newborn 12 11 10 16 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 9 8 7 6 5 4 3 2 1 8 7 6 5 4 3 2 1 ntial impact on maternal nutrition “Novel &Innovative”compositepublichealthpackageswithpotentialimpactonmaternal nutrition “Existing” compositepublichealthpackageswithpotentialimpactonmaternal Socio–cultural andeconomicdeterminants:others(Pleasespecify______) Socio–cultural andeconomicdeterminants:women’s statusinthecommunity, familystructures andnorms Socio–cultural andeconomicdeterminants:globalization&marketforces influencingfoodhabits Socio–cultural andeconomicdeterminants:competinguseofresources otherthannutrition/food forgoodsandservices Socio–cultural andeconomicdeterminants:care andfeedingpractices Socio–cultural andeconomicdeterminants:timeconstraintwithmothersenteringintotheworkforce &obesityandothernon-communicablediseases overweight Maternal Vitamin Ddeficiencyamongwomen Iodine deficiencydisorders amongwomen Anemia amongwomenofreproductive agegroup of kitchengardens/organic farming,deworming,convergent-innovation coalitiontoaddress issuesofanemia,under-nutrition, obesity] program, promotionfortification Food [eg, nutrition child on impact potential with packages health public composite Innovative” & “Novel “Existing” compositepublichealthpackageswithpotentialimpactonChildnutrition[eg,ICDS,Mid-dayMealProgram] Socio–cultural andeconomicdeterminants:others(Pleasespecify____) Socio–cultural andeconomicdeterminants:potablewater, hygieneandsanitation Socio–cultural andeconomicdeterminants:useofpesticides&fertilizers Socio–cultural andeconomicdeterminants:statusofgirlchildwomeninthecommunity Socio–cultural andeconomicdeterminants:globalization&marketforces influencingfoodhabits Socio–cultural andeconomicdeterminants:competinguseofresources otherthannutrition/food forgoodsandservices Socio–cultural andeconomicdeterminants:care andfeedingpractices Socio–cultural andeconomicdeterminants:timeconstraintwithmothersenteringintotheworkforce Fetal andchildnutritionoriginofadultchronic non–communicablediseases[eg,cardiovascular diseases,metabolicsyndrome, obesityetc.] Nutrition deficiencyassociatedcongenitalmalformations andobesity Childhood overweight Micro-nutrient deficiencies(iron/folic acid/zinc/iodine/Vitamin A) Low birthweight Protein energy malnutrition(PEM) to promote accesstocare] “Novel & Innovative” composite public health packages with potential to influence neonatal morbidity and mortality [eg, Innovative solutions care, childandadolescenthealth(RMNCH Reproductive, newborn, maternal, newborn based Home IMNCI, [eg, mortality and morbidity neonatal influence to potential with packages health public composite “Existing” care practices,poverty,Social determinantsofNMR[eg,newborn poorhealthseekingbehaviour] Others (Pleasespecify____) Congenital malformations Birth asphyxia&trauma Neonatal sepsisincludingpneumonia Preterm birth Continued bers, international CHNRI experts, and experts from the World Health Organization who had been close exercises.past Twoin used criteria retrievescoring to mem RSC with wereheld roundsconsultation of scoring: for criteria of Finalization s 101, nutrition:94)(Table 6). The ROs were finally categorized into four themes (maternal health: 122, newborn health: 56, child themes. health: and components domains, across issues cross–cutting to pertained these of several and AOCs portfolio of inter–related RIs that addressed a central research concept. Thus, the ROs addressed multiple a represented redundancy. RO and Each duplication avoid to refinement iterative through crystallized (ROs): researchoptions of Development from theoriginal3497RIs.(Table 5). RIs 4003 of compendium a to processled This (PICO). outcome control,and intervention, population, the described that way a in refined were RIs the possible, as far As sittings). several over teams by ing erence at any time. The process was intuitive, consultative and iterative (completed through brainstorm ref ready for file separate a as maintained was list RI original The RI. discardingany without and intact econd
round
oF
crowdsourcing Previously published CHNRI exercises were reviewed extensively extensively reviewed were exercises CHNRI published Previously 472 The 4003 RIs were consolidated onto 373 ROs. These were These ROs. 373 onto consolidated were RIs 4003 The
+ A)] www.jogh.org • doi:10.7189/jogh.07.011003 - - - - www.jogh.org • doi:10.7189/jogh.07.011003 *Research ideasreceived from theNationalSteeringGroup asandwhenthrough hand–writtensubmissions. Table 5. Box 2. Table 6. high agreement [ der each criterion, we chose to forego sub–questions as we were advised that sub–questions usually had search option against the particular criterion. While other CHNRI exercises employed sub–questions un and not, did responses:following the of one ing the scoring criteria to be used. The scorers were expected to evaluate the ROs against the criteria by choos plines across India that were to score the research options were the key considerations while deciding on presentthe of exercise,scope and largethe and natureROs scorersthe of of number from disci various ( ROs and themes domains, across applicable consistently be to believed were teria equity,research)on cri investment These wereand finalized. thinking, out–of–the–box and innovation (answerability,criteria wordedrelevance, succinctly Five exercises. CHNRI previous with associated ly Nutrition Child health Total (N = >1 domain Newborn health Newborn Description Single domain Discovery Development Delivery Maternal health Maternal t d heme omaIn 5. 4. 3. 2. 1. nutrition willoutweightheconsideration ofinvestmentsonresearch? Investment on research. Is it likely that the potential impact and benefits of the new knowledge on health/ graphic disadvantages,thereby reducing inequities? tematically associated with social, cultural and economic hierarchies, ethnicity, gender, environment and geo Equity.researchthe that likely it Is product addresswill differences the arethat nutrition sys and health in search havethepotentialfor transformativechangeinthehealthsystem/care? challenges. refractory and complex, resolve to thinking out–of–box and Innovation edge? knowl in gap critical and condition burden high a address would research the that likely it Is Relevance. global collaboration? strengtheningby throughor capacities and existing capacities the national and local existing the regional or Answerability.
oF Scoring criteriaandtheirdefinitions
Distribution oftheresearch optionsinthedomainsofresearch Research ideasobtainedthrough thefirst round ofcrowd–sourcing andsubsequent refinement research 373) ‘Not my expertise’ my ‘Not 12] and also because our exercise had a large number of ROs to be scored and we were Can the research be done through ethical, transparent, well–designed, “do–able” studies with Total Subtotal Lateral submissions* Childhood nutrition nutrition Maternal – – Subtotal Lateral submissions* Morbidity Stillbirths Mortality c omponent Maternal health Maternal 122 (100.0) if the scorer felt that s/he was not sufficiently informed to adjudge the the sufficientlyre- adjudge not to was informed s/he that scorerfelt the if 93 (76.2) 42 (34.4) 29 (23.8) 44 (36.1) 57 (46.7) 8 (6.6) ‘Yes’researchquery,the criterion if the met favorably option t 473 otal
number
concern F 16 12 12 15 13 requency 8 8
oF ebr elhCidhat Nutrition Childhealth health Newborn
areas 56 (100.0) 48 (85.7) 15 (26.8) 21 (37.5) 24 (42.9) (%) 8 (14.3) ICMR–INCLEN CHNRIforresearchprioritysettinginMNCHN 4 (7.1)
oF
oF
research n umber
optIons
( oF receIved 3497 1042 1218 596 641 590 450 353 418 436
11 research 2
June 2017 •Vol. 7No. 1• 011003 In )
themes
0 100 94(100.0) 101 (100.0) Ideas 6(58 12(12.8) 16 (15.8) 5(42 82(87.2) 85 (84.2) 9(86 35(37.2) 39 (38.6) 7(66 27(28.7) 42(44.7) 37 (36.6) 37 (36.6) 4 (4.0)
n umber ). The context context The 2). Box ( Does the new re new the Does aFter
oF
4003 1421 1308 reFInement
648 626 831 590 243 542 523 research – – 2 (2.1) if it it if ‘No’
Ideas ) - - - -
- - - -
VIEWPOINTSPaPErS PaPErSVIEWPOINTS June 2017 •Vol. 7No. 1•011003 EAG –Empowered ActionGroup, UT–UnionTerritories Arora etal. Table 7. Total States andUTs Western and Southern (including West Bengal) UTs and States Northern States Eastern EAG StatesandNorth r egIon Distribution ofexpertswhoparticipatedinthe2 constituency (Table 8). Of 84 members approached for the LRG, 79 participated (94.0% participation). LR6 each in criterion respective the by received amount average the of mean arithmetic the calculating portant and the minimum to the least important. The relative weight for each criterion was computed by (INR) 100 across the five criteria, giving the maximum amount to the criteria they felt to be the most im ed to assign relative weights to the scoring criteria by distributing a hypothetical amount of Indian Rupees The LRG members were given an in–depth explanation of the CHNRI exercise. They were then request Assignment ofrelative criteriaweightsbytheLRG ticipated (58.1%participationrate)(Table 7). vided with the same for recording the responses. In the second round of crowdsourcing, 893 scorers par progress. Scorers who requested hard copies of the questionnaires instead of the online process were pro mediate troubleshooting and timely reminders, and used continuous real–time data monitoring to check im- for telephone and throughemail network nationwide the with vigorouslyengaged remained CLEN IN at team projectmanagement RPS The constraints. overcometime and fatigue scorer effectsof avoid his previous responses once s/he had moved forward. Completion over multiple sessions was allowed to edit and review could However,scorer disabled). the was logic skip (ie, RO next the to on moving fore It was mandatory for the scorer to evaluate the RO on the screen against both of the assigned criteria be was restarted6thexpert). withevery Questionnaire 1; the next in line got Survey Questionnaire 2 and so on; the questionnaire allocation cycle Survey got 1 number serial with expert (the allocation throughconsecutive theme the within tionnaires region. Subsequently, the experts within each region were equally distributed across the five survey ques sourcing (‘participated’, ‘could not participate’, or ‘newly invited’ experts); and, second, according to their of crowd first round the in status participation their according to first, levels: two at stratified was work Relevance and Investment on Research; and (v) Equity and Investment on Research. The nationwide net (iv) Innovation; and Relevance (iii) Equity; and Answerability (ii) Innovation; and Answerability (i) ing: scor preparedfor been had questionnaires) (survey combinations criteria such Five scoring. for criteria five the of two of combination a allocated randomly was scorer each attrition, and burden scorer high to led have could and high was scored be to ROs of number the As theme. assigned the for ROs the all score to requested was scorer The scorer’sscreen. the phone on computer/smart time, a at one quence, se random a in Thereafter,appearedprocess. ROs and criteria scoring the and exercise, the of method by the scorer. Once the scorer logged in, s/he underwent a comprehensive orientation of the context and selected responses and addresses) IP and (email details access archive readily could software The team. management project RPS INCLEN the from email invitation an through accessed be could that counts interface online user–friendly Network: Nationwide the by researchoptions the of Scoring the validityofourplannedregional analyses. aiming to maximize retention of participants by minimizing scorer fatigue, especially in order to preserve as follows[13]: were excluded from the calculations. Relative ranking and Research Priority Scores (RPS) were calculated The Data managementandanalysis Box Thinking(0.199),Equity(0.193),Answerability(0.192),andInvestmentonResearch (0.161). Out–of– and Innovation by followed (0.254), Relevance to relativeweight maximum ascribed LRG The “Yes”and aeFml oa aeFml oa aeFemale Male Total Female Male Total Female Male 0 4 5 3 518165 3 0 0 1 1 7 893 376 517 212 107 105 233 57 176 198 65 133 250 147 103 95 43 76 92 63 08 7 5 336 159 269 177 104 82 288 165 113 50 60 175 32 29 70 96 31 28 27 68 42 69 16 69 64 52 14 27 72 55 37 15 62 94 57 23 55 69 39 39 44 87 25 48 39 m aternal responses were scored as “1” and “0” respectively.“0” and The “1” as scored were responses “No”
health ( n www.surveymonkey.com) ewborn nd
round ofcrowd–sourcing (theScoringExercise) health 474
c hIld that allowed for having individualized scorer ac scorer individualized having for allowed that
health Total The scoring exercise was done using a a using done exercisewas scoring The Male Female n www.jogh.org utrItIon responses expertise” my “Not oa aeFml Total Female Male Total • doi:10.7189/jogh.07.011003 o verall ------www.jogh.org • doi:10.7189/jogh.07.011003 LRG –Larger Reference ChildHealth andNutrition Group, Newborn, MNCHN–Maternal, Average Expert Agreement (AEA) [ cation ofthescorer asentered byhim/heratthetimeofscoringdeterminedregional ranking. The ROs were arranged in descending order of their RPS to get national and regional rankings. Work lo 2. 1. Average scores received againsteachofthefivecriteriawere calculatedforeachRO. Table 8. stable estimate of priority ranks at national and sub–national (regional) levels [ and saturation to led have should size”) (“sample scorers of number large The ROs. the score to theme 893 experts were involved in the scoring process. About 75 (60–96) experts were involved per region per CLEN’sfromexperts 498 studies, multi–centric for network across researchcontributed India and ideas IN regionally.on building interventions and tailor technology to of country Withuse effectivethe able which, in a country as large and diverse as India, is imperative to truly explore research priorities and en health and nutrition community. Moreover, the exercise is the first to conduct subnational–level analysis by,driven India’sand of, representation a truly is exerciseway, CHNRI this this In exercises. previous from unique it makes exercise this to scorers and contributors exclusive the as nationals Indian Having to prioritizationofROsthatwasimportantboth. field of MNCHN. This helped in including a variety of viewpoints in the scoring process and possibly, led the researchin of (techno–managerial) users and doers of range diverse include to base stakeholder the expanded we contrast, In scorers. for pre–requisite selection a as research/policy to contribution active methodological robustness. CHNRI exercises hitherto had approachtaken a conservative in considering further built and method CHNRI systematic the of flexibility inherent the leveraged exercise The els. (regional)sub–national lev and national at nutrition and health child and neonatal maternal, for orities pri identified setting, LMIC a in stakeholders national of range diverse of inclusion systematic through [ LMICs the in so more priorities, setting for process the impeded frequently capacity, and societal values and ethics [ stakeholders is essential to identify priorities that reflect research needs, available technical and financial of spectrum wide a of engagement that highlighted Setting Priority Workingon GroupCOHRED The DISCUSSION way forward fortheexercise hasbeenacceptedforpublicationinthe to the Journal of Global Health. The overall discussions by the National Steering Group on the results and submission for prepared manuscripts in separately in–depth reported are exercises all from results The RPS foreachofthese. accordingtheir lists to ranked generate and ROs, 373 of fromcompendium methods the biotechnology requiring areas (iii) and, systems; health and health on impact greater for themes MNCHN four across group further suggested to identify ROs relevant to three more themes: (i) adolescence; (ii) issues cutting The second meeting of the NSG reviewed the ranked list of national and regional research priorities. The scored thatRO. who scorers of number total the by divided RO particular a for score common most the scored who ers Overall (N = Funding agencies(N = (N governments = MNCHN program managersfrom centralandstate Eminent researchers (N = Policy decisionmakers,politicians(N = lrg each RO. Research Priority Scores (RPS) were calculated by adding together each criterion’s weighted scores for rion. crite each by received score mean the to applied were LRG) the by assigned (as weights criteria The categorIes Relative weightsassignedtothescoringcriteriabyLarger Reference Group 79) 24) 20) 17) 18) 14] was also calculated for each RO. The AEA is a proportion of scor 15]. Stakeholder engagement, and data and capacity constraints 475
a nswerabIlIty .9 .5 .9 .9 0.161 0.199 0.193 0.254 0.192 .7 .8 .9 .0 0.145 0.162 0.200 0.177 0.162 0.198 0.195 0.197 0.288 0.201 0.203 0.169 0.173 0.254 0.209 0.245 0.186 0.229 0.212 0.197 ICMR–INCLEN CHNRIforresearchprioritysettinginMNCHN r elevance Indian Journal ofMedicalResearchIndian Journal . e quIty June 2017 •Vol. 7No. 1• 011003 8 ]. The current exercise, current The 16]. ]. The improved response I nnovatIon I nvestment research
on
------
VIEWPOINTSPaPErS PaPErSVIEWPOINTS June 2017 •Vol. 7No. 1•011003 Arora etal. rates between first and second rounds of crowdsourcing should have reduced bias [ sources of avoidable waste once research priorities are set [ the addressing of responsibility primary the regulators, research and funders research the with affixes [ priorities exercisesand CHNRI making of decision impact funder the on a buy–in. It is also anticipated that there is an imminent challenge to develop tools to detect and evaluate search priorities. It has been opined that prioritization processes will have an impact only if funders have robust,participatory,a inclusive, as method of re identification for technique objective transparentand gaged. It expanded on the CHNRI methodology and thus, contributes to further evolution of the CHNRI en constituencies stakeholder of spectrum and participants of number the and scored, and processed The exercise was the largest to–date use of the CHNRI methodology in terms of research ideas collected, for theseadditionallists. unknown remain shall ranking relative their determine to scores RO of validity the themes, across high was AEA overall the Although expertise. professional their in differences with experts of set dissimilar by groups,scoredfromdifferent thematic up picked were ROs the because list priority indicative an be and biotechnology related ROs from the 373 ROs spread across different themes. These lists will, at best, of services, the NSG suggested developing ranked priority lists for adolescent health, cross cutting themes delivery and implementation the in burden systems disease health the the of of view significance In and priorities shouldbesuitedtothecontext. to observe that “Relevance” was accorded the highest weight by all the LRG sub–groups highlighting that interesting was It [18]. priorities research about perspective collective their in differences have to likely are constituencies different since strength a as viewed be to is LRG the LRG; the among constituencies distinct four were There scoring. setting priority the of validity the affecting without compliance ipant criteria scoring adopted in the current exercise and appears to be a pragmatic approach for better partic partial to due bias minimal indicates also This scorers. the among consistency indicating 0.869–0.923) nutrition: 0.899–0.923; health: child 0.871–0.902; health: newborn 0.887–0.929; health: (maternal tional level across the themes, the AEA for both individual and aggregate of the five criteria was fairly high - na at ROs 10 top the For [14]. response (modal) frequent most the gave that scorers of proportion the represents option research evaluated each for AEA The criteria. five all of instead them to randomly ed To minimize scorer fatigue, we asked the participants to score against predefined pairs of criteria allocat respective thematicpapersprepared forsubmissiontoJoGH. demia for the themes considered in this exercise. The scorer profiles have been discussed in details in the program in participation gender reflectionscorers skewed management, a of is researchtion aca and priorities nationally. researchsetting in populations, their among diversity with those especially LMICs, other for guidance a should further help in matching the exercise’s findings to other LMIC contexts. This exercise can serve as course correction of currently funded research portfolio as needed. Sub–national (regional) prioritization dia and donor agencies funding research in India and in similar LMIC contexts, and also inform any mid– tive effort helps in rational distribution of health and nutrition research budget by the Government of In
476
20]. We hope that ICMR–INCLEN collabora www.jogh.org ]. A recent article in Lancet Lancet recentin A article 19]. • doi:10.7189/jogh.07.011003 17]. Gender distribu ------www.jogh.org rEFEr ENCES • doi:10.7189/jogh.07.011003 14 13 12 11 10 9 8 7 6 5 4 3 2 1 doi:10.3325/cmj.2008.3.307 2008;49:307-17. J. Med Croat framework. conceptual and challenges universal investments: Rudan I, Chopra M, Kapiriri L, Gibson J, Lansang MA, Carneiro I, et al. Setting priorities in global child health research of humancollectiveopinion.JGlobHealth.2016;6:010503. propertiesQuantitative VI. method: CHNRI researchYoshidathe health using Setting priorities S. Cousens I, Rudan S, method: VII.Areview ofthefirst50applications oftheCHNRImethod.JGlobHealth.2017;7:011004. CHNRI the using priorities research health Setting al. et H, Nair Wazny K, D, KY, Sridhar Chan Yoshida S, I, Rudan Medline:19535830 2009;129:368-75. Res. Med J approachIndian combined matrix. the researchapplying GhaffarSetting by A. priorities http://announcementsfiles.cohred.org/gfhr_pub/assoc/s14791e/s14791e.pdf 2017. . Accessed:15January Available:1999. Switzerland. Geneva Research. Health for Forum Global 1999. Research Health on Report 10/90 The 2016;6:010507. Medline:26401271 Yoshida S. Approaches, tools and methods used for setting priorities in health research in the 21st century. J Glob Health. loads/publications/files/0535210001249198837-604_file_PRIORITY_SETTING_pdf Available:2006. Level. Global the at Mortality Child and 4 Goal Development nium Fontaine O, Martines J, Bahl R, Briend A, Muhe L, El Arifeen S, et al. Setting health research priorities to address Millen vestments: assessmentofprinciplesandpractice.Croat MedJ.2007;48:595-604. Rudan I, Gibson J, Kapiriri L, Lansang MA, Hyder AA, Lawn J, et al. Setting priorities in global child health research in files/0535210001249198837-604_file_PRIORITY_SETTING_.pdf. Accessed:20August2016. Available: 2006. (CHNRI) Initiative Research Nutrition and Health Child Bangladesh: 2006. Investment. Research Health Child in Setting Priority Systematic in Approach New A INF.0b013e31819588d7 on netos y 05 Pdar net i J 20;81 Suppl):S43-8. 2009;28(1 J. Dis Infect Pediatr 2015. by infections born new from W,mortality Carlo global reduceMK, to Bhan priorities ResearchN, KY, al. Ali Chan et J, Martines R, Bahl line:19090596 2008;49:720-33. J. Med Croat Method. CHNRI the of implementation for guidelines investments: search re health child global in priorities Setting al. et M, Black ZA, Bhutta S, Arifeen El S, Ameratunga JL, Gibson I, Rudan land, UK).12May2015. Wazny, Kerri (University of Edinburgh, Scotland, UK). In conversation Available:with: Rudan, Igor (University of Edinburgh, Scot Africa. files/0535210001249198837-604_file_PRIORITY_SETTING_.pdf. Accessed:20August2016. South for investments search Tomlinson M, Chopra M, Sanders D, Bradshaw D, Hendricks M, Greenfield D, et al. Setting priorities in child health re Wazny K.‘Crowdsourcing’ tenyearsin:areview. JGlobHealth.2017;7:forthcoming. rial organizations. has journal been addressed, this article was reviewed according to best practice guidelines of edito international is the editor–in–chief of the coi_disclosure.pdf (available upon request from the corresponding author) and declare no conflict of interest. IR http://www.icmje.org/at form disclosure uniform ICMJE the completed have authors All interests: Competing and editingthemanuscriptforintellectualcontentinapproving thefinalversionforsubmission. providingexercisethe to guidance technical differentat projectthe of points reviewing critically in and lifecycle, critical technical and strategic inputs for the design and execution of the exercise. All authors were responsible for chiving and retrieval, supporting AM in data analysis and in preparing the draft manuscript. KW ar and IR data provided network, the of management for responsible were MM and HSG manuscript. the drafting for and data the of analysis and management network, the with communication participants, of identification project, the of execution literature, of review for responsible was AM team. projectmanagement RPS INCLEN the of part were with the network and request for participation in the exercise were made on his behalf. AM, HSG, MM and MKD of data, revising the manuscript critically and for the final approval of the version submitted. All communications interpretation and acquisition exercise, the of design and conception the for responsible was He meetings. NSG Authorship contributions: NKA was the PI and provided overall leadership to the entire exercise by chairing the Funding: Thisexercise wasco–fundedbyICMR,theCHNRISecretariat andINCLEN, NewDelhi. CLEN. IN of Committee Ethics Institutional Independent the by approved and reviewed was protocol study The Ethics: views andopinionsdonotnecessarilyexpress thepoliciesofWorld HealthOrganization. Disclaimer: phases oftheexercise. Dr Mr Kriti Amit Agarwal, K Sagar, Mr Gaurav Banyal and Mr Chandan Singh for their assistance during various pation in this exercise and kind support. We are also thankful to Dr Geeta Chhibber, Dr Manish Singh Chundawat, Acknowledgments: The team is grateful to the experts from different parts of India for their enthusiastic partici doi:10.3325/cmj.2008.49.720 Rajiv Bahl and Shamim Qazi are staff members of the World Health Organization. The expressed The Organization.World Health the of members staff are Qazi Shamim and Bahl Rajiv Journal Journal of Global Health. To ensure that any possible conflict of interest relevant to the doi:10.7189/jogh.06.010302 477
http://www.chnri.org/secured/uploads/publications/ Medline:27350874 http://www.chnri.org/secured/uploads/publications/ ICMR–INCLEN CHNRIforresearchprioritysettinginMNCHN doi:10.7189/jogh.06.010503 Medline:17948946 Medline:19106763 June 2017 •Vol. 7No. 1• 011003 . Accessed:20August2016. http://www.chnri.org/secured/up- Medline:18581609
doi:10.1097/ Med------
VIEWPOINTSPaPErS PaPErSVIEWPOINTS June 2017 •Vol. 7No. 1•011003 Arora etal. rEFEr ENCES 16 15 20 19 18 17 when research prioritiesare set.Lancet.2014;383:156-65.Medline:24411644 Chalmers I, Bracken MB, Djulbegovic B, Garattini S, Grant J, Gülmezoglu AM, et al. How to increase value and reduce waste ytmtc eiw f ulse rprs Po Oe 2014;9:e108787. One. PLoS reports. published of review systematic A countries? income middle and low in set priorities research health are How JM. Kaldor KJ, Henderson S, McGregor Policy Plan.2000;15:130-6.Medline:10837035 Health countries. developing from research:lessons health for setting Priority WorkingSetting. The Priority on Group method: I.Involvingfunders.JGlobHealth.2016;6:010301. CHNRI the using priorities research health Setting al. et R, Bahl D, Sridhar S, KY,Cousens Chan YoshidaS, I, Rudan Medline:19535827 Rudan I. The complex challenge of setting priorities in health research investments. Indian J Med Res. 2009;129:351-3. searchers. JGlobHealth.2016;6:010302.Medline:27350870 YoshidaWaznyS, Cousens S, KY.Chan K, researchhealth Setting re Involving II. method: CHNRI the using priorities pone.0108787
doi:10.1093/heapol/15.2.130 478
Medline:26401269 doi:10.7189/jogh.06.010302 Medline:25275315 doi:10.1016/S0140-6736(13)62229-1 www.jogh.org • doi:10.7189/jogh.07.011003
doi:10.1371/journal. - www.jogh.org of theCHNr method: VII.a Setting healthresearchprioritiesusingtheCHNrI material. The onlineversionofthisarticlecontainssupplementary Electronic material: supplementary a Mickey Chopra E Lawn Nair 7 6 5 4 3 2 1 Igor r [email protected] Scotland, UK Edinburgh EH89aG Teviot Place The UniversityofEdinburgh and Informatics The UsherInstituteforPopulationHealthSciences Centre forGlobalHealthr Professor Igorr Correspondence to: 14 13 12 11 10 9 8 Chan
Department ofInfectiousDiseaseEpidemiology,London International CentreforDiarrhoealDiseaseresearch, Child HealthresearchFoundation,DhakaShishuHospital, TheWorldBank,Washington,DC,USa Centre ofExcellenceinWomenandChildHealth,theaga Centre forGlobalChildHealth,theHospitalSick Centre forMaternal,adolescent,reproductiveandChild Department ofPsychology,StellenboschUniversity, NrF CentreofExcellenceinHumanDevelopment,DVC Centre forMedicalInformatics,TheUsherInstitute Nossal InstituteforGlobalHealth,UniversityofMelbourne, Department forMaternal,Newborn,Childandadolescent Centre forGlobalHealthresearch,TheUsherInstitute School ofPublicHealth,Baltimore,Maryland,USa Institute forInternationalPrograms,JohnsHopkinsBloomberg School ofHygieneandTropicalMedicine,London,UK Bangladesh, Dhaka,Bangladesh Dhaka, Bangladesh Khan UniversityKarachi,Pakistan Children, Toronto,Canada Keppel Street,London,UnitedKingdom Health, LondonSchoolofHygieneandTropicalMedicine, Stellenbosch, Southafrica South africa research Office,UniversityofWitwatersrand,Johannesburg, of Edinburgh,Scotland,UK Population HealthSciencesandInformatics,TheUniversity Victoria, australia Health, WorldHealthOrganization,Geneva,Switzerland of Edinburgh,Scotland,UK Population HealthSciencesandInformatics,TheUniversity rifeen 1 , a 1,3 udan , DeviSridhar • doi:10.7189/jogh.07.011004 11,12 7 ziz Sheikh , Zulfiqara
, r 1
, SachiyoYoshida udan obert EBlack 10
, HarryCampbell 4 , MarkTomlinson Bhutta esearch 1 , KerriWazny
I method
8,9 13 , SimonCousens , r 2 review of the first 50 applications reviewofthefirst50applications , Kit Yee , KitYee ajiv Bahl 1 , Shams El , ShamsEl 1 , Harish , Harish 5,6
, Joy , Joy 2 , 14 479 Results and so–calledgrey literature. order.chronological We Scholar, Google searchedPubMed in method CHNRI the of application of examples 50 first Methods periences. ex those from emerged that messages important most the method, published between 2007 and 2016, and summarize CHNRI the of application of examples 50 first the review we paper this In decade. past the over used widely and lar search Initiative”) as an approach that clearly became popu (acronym derived from the “Child Health and Nutrition Re method CHNRI the identified have priorities research set Backgroundto recentused Several methods reviews the of those inlow–and middle–incomecountries. globally,groups particularly many by uptake its for sons pensive to conduct, which we believe is one of the main rea Finally,needs. inex relativelyand and contexts simple is it different many to adjustable and flexible very is It process. the in invest groups various that ensuring by results the of “ownership” fostering inclusive, is It “crowd–sourcing”. on text and priority–setting criteria. It is democratic, as it relies transparent and replicable, because it clearly defines the con also is It researchquestions. many handling for framework acceptable an offering nature, in systematic is method The conceptual advances that have addressed common concerns. ting health research priorities can be attributed to several key set in method CHNRI the of popularity The Conclusions “relevance” andothers). “feasibility”, “acceptability”, “sustainability”, cost”, “low (eg, teria used, but also by introducing some entirely new criteria cise. This was done not only by changing the number of cri exer particular each suit to method CHNRI the modifying of all conducted exercises departed from recommendations, two–thirds morethan 2016 by but criteria, priority–setting recommended five the to adhered articles CHNRI–based tries, and national–level applications are on the rise. The first coun middle–income and low– to relevant only were that cation. The majority of the exercises were focused on issues adolescent health, dementia, national health policy and edu as such topics the widely,into more expanding used was method CHNRI the 2012 recorded.Since also were noses) tion outside this field (eg, mental health, disabilities and zoo global child health issues, although the first cases of applica od was mainly used for setting research priorities to address
Initially, between 2007 and 2011, the CHNRI meth We conducted a literature review to identify the the identify to review literature a Weconducted June 2017 •Vol. 7No. 1• 011004 global journal of health ------
VIEWPOINTSPaPErS PaPErSVIEWPOINTS June 2017 •Vol. 7No. 1•011004 Rudan etal. ity, (iv)thepotentialforasubstantialreduction ofdiseaseburden and(v)theimpactonequity [ erates new knowledge. The five suggested criteria were (i) answerability, (ii) effectiveness, (iii) deliverabil teria followed a simple conceptual framework that demonstrated how the process of health research gen criteria that could discriminate between many competing research options. CHNRI’s “standard” set of cri researchtypical Finally,a article). correspondto (which tions introducedtransparentmethod a the specific very and program),research 5–year a to correspond (which researchoptions (usually upto200)isconsolidated byremoving overlappingideasandintegrating related ideas,anum tribute hundreds of research ideas [ searchers (but also policy–makers and program managers, depending on focus of the exercise) who con processCHNRI typical The reachesthat largeteam a management to small out a involves reof number and breadth of suggested research questions by categorizing them in broad includes delivery, operations and implementation research). Moreover, it addressed the difference in depth ment (throughresearch),epidemiological (ii) tified four fundamental instruments of health research – “ The method also introduced a systematic approach to listing many competing research questions. It iden ment (eg,burden reduction, patents,orvariousformsofpublicrecognition) [ frominvest returns expected the (v) and risk–seeking); or aversive risk (eg, investment of style the (iv) timeframe within which the impact of supported research was expected (eg, short, medium or long term); is focused; (ii) the affected population that would benefit from the investments in health research; (iii) the research priority setting. The components of the context were: (i) the health issue on which the research ities in health research investments [ investments research health in ities addressto gether relatedchallenges key of number a problemmulti–dimensional the to prior setting of Worldfromthe funding to with worked supported experts experts, The 15 Bank. of panel disciplinary The CHNRI method was developed between 2005 and 2007 through 12 consecutive meetings of a trans– reducing bothdiseaseburden andtheinequitiesamongworld'schildren [ in research health of potential the maximize to order in research implementation and research lational trition. Their method also sought to achieve an acceptable balance between fundamental research, trans sist decision–making and priority setting in health research investments to improve child health and nu for Health Research in Geneva, Switzerland [ The Child Health and Nutrition Research Initiative (CHNRI) started as an initiative of the Global Forum The CHNRImethodforsettinghealthresearch priorities improve thelegitimacyofpriority–settingexercises atalllevels[ could that tool desirable a be would priorities research health setting for process replicable and parent [3 century 21st the in prioritization research health in used tools priority–setting compared and described review effectivenessthe differentbut of forms, various difficult A in recent approaches evaluate. cised to very is locally. Therefore, the process for setting health research priorities is a genuine need and it is being exer oritize between competing research questions arises at different levels – globally, regionally, nationally and Given that a spectrum of possible ideas for health research is extremely broad and diverse, a need to pri our planet'spopulation[2 for outcomes health improve and disease and health human on knowledge new generate to is aim tive tinuously to conduct, facilitate, support and promote health research and utilize its results. Their collec yers and many other stakeholders [1 conference organizers, guidelines developers, but also science–focused journalists and media, patent law editors, journal and publishers professionalbodies, governments, of hands the in is results the of lation reviewerspeer committees, proposalsethics grant of researchand trans and dissemination The articles. [ researchersof breed new a – scientists” “citizen by even and researchtools, new develop that industries sector.private the in also supporting the scientists, for opportunities education life–long by assisted is It [ industry and philanthropy–oriented foundations, all of which invest in health research with different aims biotech and pharmaceutical organizations, international and regional national, agencies, funding large includes It actors. diverse many of network complex extremely an is system research health global The 2 1 ]. Scrutiny over the health research process is in the hands of many individual research policy makers, ]. The research itself thrives in well–managed and meritocratic universities and research institutes, but (through translational research) and (iv) ]. There seems to be a general consensus among researchers that a flexible, systematic, trans systematic, flexible, a researchersthat among consensus general a be to seems There ].
]. 9 ]. All of these individuals, groups and organizations act together con- , 10]. Once a list of a manageable number of research ideas/questions 5 – 7 (through basic, ie, fundamental research),fundamental (throughie, discovery (iii) basic, ]. The method aimed to carefully define the define carefully to aimed method The ]. 480 4 delivery ]. One of its main aims was to develop a tool that could as (through health policy and systems research, which the four D’s” – research to achieve (i) 3 , 4 ]. www.jogh.org research avenues 6 5 – ]. • doi:10.7189/jogh.07.011004 8 ]. researchideas/ques- context , more focused for health health for description 6 develop- – 8 set of of set ]. ------
www.jogh.org date related tothecontextofexercise priority–setting exercises basedontheCHNRImethodpublishedto searchers and/ortechnicalexpertsinvolved intheexercise. *Population groups other than funders of research and their representatives, re Table 1. No Yes stakeholders:* Involvement ofexternal All agegroups People withHIV/mentalhealthillnessesdisability Population aged60andabove Adolescents andyoungadults Children olderthan5years Children aged1month–5years Stillbirths orneonates(<1month) Population thatwouldbenefitfrom research: More than10years 10 years Less than10years Time frameuntiltheexpectedimpactofresearch: Crisis setting Sub–national National Low– andmiddle–incomecountries Global Context oftheCHNRIexercise: All–cause morbidityandmortality Health andeducationsystemrelated research Dementia Mental health All–cause disability Major infectiousdiseases(eg,tuberculosis, zoonoses) Sexual health Child morbidityandsuboptimaldevelopment Child mortality(all–causeorindividualcauses) h ealth
Issue The maincharacteristicsofthedesign50research • doi:10.7189/jogh.07.011004
addressed
through eral keycriteriaforprioritization[8 sev using community this of community,subset research a the by to judged questions research mitted several transparent priority–setting criteria [7 against scores their by ideas/questions research 200 to up ranks that list a is process CHNRI the of put others, so that the overall score also includes the value system of a wider community [ community wider a of system value the includes also score overall the that so others, olds and weights for each of the priority–setting criteria, giving some criteria greater importance over the tive optimism” on a scale 0–100. In the final step, external stakeholders are invited to set different thresh proposed research questions against each priority–setting criterion [ ber ofresearchers (from 20toupseveralhundreds, dependingonthecontext)are invitedtoscore all study is available in Table S1 in each of details with list full (the 2016 and 2007 between published exercises, priority–setting CHNRI 50 first The Initiative”. Research Nutrition and Health “Child or “CHNRI” term search the using nels) defined as papers produced by organizations outside of the traditional publishing and distribution chan (usually literature” “grey so–called and Scholar,PubMed WeGoogle published. searched and reviewed been have that publications 50 first the on focus to decided we implementation, method's in milestone not all of them have reached their final stage of peer–reviewed publication. Therefore,a to acknowledge but planned, or conducted being exercises CHNRI further with application, of examples 50 than more presently are There method. the of uptake the of evolution order,chronologicalthe in study method to We conducted a review of the literature to identify the first 50 examples of the application of the CHNRI The examplesofimplementation
research Online Supplementary Document n umber 774 26 37 13 34 17 74 20 37 10 50 32 25 16 52 26 8 2 9 4 8 1 8 4 6 7 3 2 2 1 1 7 6 4 2 3 2 2 1 6 8 8 1 4 3 4 2 ]. tIon p ropor 18 16 14 14 16 (%) 481 ]. This serves to reveal strengths and weaknesses of all sub- -
- 10 perts submitting research ideas. They submitted about all exercises was above 60%, with more than 3000 ex acrossresponserate initial proposedThe the criteria. according questions those to of scoring the and tions participation in the generation of research ideas/ques ers, policy makers and program officers, seeking their causes) (52%) ( addressed child mortality (either all–cause or specific is not surprising that the majority of the exercises have mortality,child of reduction it the on focus initial its and method CHNRI the of development the of tory his the Given method. CHNRI the of advances tual concep key the of one is prioritization for used teria cri the and prioritization of context the over Clarity Document). in OnlineSupplementary “calls for action” within the Lancet series (see Table S1 three were a part of policy recommendation papers or and exercises stand–alone as published were three the in published exercises six the Health Public BMC including journals in lished age of 48 per exercise). Most of the papers were pub aver (an scorers participating 2403 by exercise) per Eventually, 86 of average werescored(an ideas 4282 ideas. duplicate of rate high relatively a indicating cy rate in submitted questions was slightly above 50%, of its initial focus – such as mental health (16%), all– method started to find its application in areas outside the Then, (6%). zoonoses and tuberculosis as such diseases, infectious major several by followed (8%), health sexual and perinatal maternal, of questions address the key global health issues, it was applied to to application method's the of progression logical a hood morbidity and improved development (4%). In child to related questions to extended then was od
000 ideas (more than 3 per expert). The redundan Table 1 ), reached out to nearly 5000 research 7 , (14%) and (14%) 10]. Their input measures “collec- June 2017 •Vol. 7No. 1•011004 ). The use of the CHNRI meth The first50applicationsofCHNRI PLoS Medicine PLoS (12%). Among Among (12%). Lancet journal, journal, Lancet The 2 ]. The final out final The ]. (20%), ------
VIEWPOINTSPaPErS PaPErSVIEWPOINTS June 2017 •Vol. 7No. 1•011004 Rudan etal. *Less than a third (n third a than *Less ified thesettoadjustitneed ofaparticularexercise. of the CHNRI criteria; more than two–thirds (n Table 2. to daterelated tothecriteriausedforprioritization priority–setting exercises basedontheCHNRImethodpublished scale–up/need/quality/operationalizability to impact/obstacles Sensitivity/immediacy/long–term Usefulness (eg, for guiding policies and programmes) Local ownership Potential fortranslation Clarity Fills akeygap/potentialforbreakthrough Fundability Attractiveness andoriginality Ethical Applicability Relevance Feasibility Acceptability Sustainability Low cost Effectiveness Deliverability Impact ondisease/disabilityburden Answerability Equity Priority–setting criteriamostfrequently used: Seven ormore Six Five Four Three Number ofpriority–settingcriteriaused:* The maincharacteristicsofthedesign50research exercises), acceptability (22%), low cost (22%), sustainability (22%) and relevance (12%). This shows shows This (12%). relevance and (22%) sustainability (22%), cost low (22%), acceptability exercises), all of 22% (in feasibility were criteria added frequently most number.The their reduce even or ercises, conducting the CHNRI processes felt a need to replace them and/or introduce further criteria in their ex groups the that clear is it effectiveness), for 70% to equity for 86% (fromfrequently, expected most as of criteria applied – up to 13 in one exercise. Interestingly, although the five “standard” criteria were used originally suggested, 12% reduced their number to only four or three, while 32% expanded the number as criteria, 5 used exercises all of 56% Although themselves. criteria the in changes the and used, teria cri of number the in changes two–thirds.included in Modification weremodified exercises,they while the of one–third in only used were criteria 5 proposed originally The exercises. 50 the across analyzed method through its implementations is particularly apparent when the criteria used for prioritization are timeframes, while 6% had longer time frames ( shorter exerciseshad the of 20% conducted; exerciseswere the which to contexts the suit to timeframe recommended the from deviated exercises the of minority sizable A method. CHNRI the of mentation imple for guidelines the in suggested originally years, 10 of frame time “standard” a used (74%) cises research,exer of the largeimpact of the expected majority the until frame time adopted the of terms In cluding newborns), 16% on adolescent and young adults, and 28% on adults or all age groups ( (in children on focused being exercises of 56% in reflected further is research.This health of oritization tial focus on child health, and to national and sub–national levels, where there is also a lot of need for pri This shows that application of the CHNRI method is beginning to expand to health issues beyond the ini there were also 14% of exercises conducted at the national level, and 2% at a sub–national level ( on low– and middle–income countries (50%). Further 32% of CHNRI exercises were global in scope, but exercisesfocused wereMost mortality,(2%). cause dementia and (8%) adults in disability and morbidity
=
16) of all exercises used the original, “standard” set set “standard” original, the used exercises all of 16)
=
34) of the exercises mod n umber 122 22 11 22 11 22 11 70 11 72 35 78 36 84 39 86 42 43 22 11 56 28 2 1 4 2 4 2 4 2 4 2 4 2 4 2 6 3 6 3 8 4 6 5 8 4 4 2 p roportIon (%) 12 10 482
- Table 1). The evolution of the originally proposed CHNRI greater urgency to reduce child mortality among the un the among mortality child reduceurgencygreater to impact was expected in most exercises (eg, 10 years) and priority is the relatively short time frame within which the research a as identified frequently was research delivery strongly encouraged(Table 2). be should exercise specific each of needs the to process the of Adjustments criteria. priority–setting different of use the allowing in process CHNRI the of flexibility the [ countries middle–income and low– in particularly cise, exer researchthe implementation dominated frequently and/or operations with along systems, and policy health delivery,on including research that surprising not is it implemented, being not are but exist do burden the trol factors, and that effective to reduceinterventions or con risk and population the in burden well–defined sonably rea a have issues health contemporary most that Given available. un was priority,researchknowledge ing such wherever effective interventions) was usually identified as the lead- and its “architecture” (in terms of contributing factors and issue health the of burden the on knowledge the erating leading research priority as a rule. This showed that gen the as identified was research (epidemiological) tive tive in controlling and mitigating the issue, then descrip uted to the issue, or the interventions that could be effec burden in the population, or the risk factors that contrib its of terms in understood well not was exercise zation First, if the health issue that was the focus of the prioriti- policy.research health for relevant messages broad very several generated exercises CHNRI 50 the whole, a As exercises The mainmessagesfrom theconducted ]. An additional important factor that explains why why explains that factor important additional An 11]. www.jogh.org • doi:10.7189/jogh.07.011004 Table 1). Table 1). ------www.jogh.org • doi:10.7189/jogh.07.011004 research prioritisation. is explicit or implicit and how this is decided – as the time frame of research questions clearly influences this whether 2) and agencies; acrossdiffer these how and adopt agencies grant horizon(s) time what 1) health issue (eg, the effect of exercise on dementia and Alzheimer disease [ the control or reduce to available were interventions effective any hardly where and years 10 than ger required discovery (fundamental) research were prioritised in the exercises where the time frame was lon temperatures).Research that questions external high at stable vaccines (eg, settings middle–income and low– in scale–up their enable to as so modification straight–forward and defined clearly some required research questions were scored highly wherever there were pre–existing and effective interventions which Translationalpriorities. of list the of top the to close it made researchquestions fundamental) ie, (basic, ation [ understandable to users, replicable, amenable to agreement statistics, post–exercise validation and evalu measure collective optimism of a group of experts toward each component of each research question are and 0–100% between range scoresthat results.Intuitive the obtain to computation statistical or matical mathe complicated any (Yes/No),avoiding input qualitative simple, a on based is method CHNRI the be able to easily organize and conduct it within any other setting. Although quantitative in its outcomes, been trained in the application of the method. It is enough to study any previously conducted exercise to ly setting criteria, as demonstrated through these first 50 applications. Sixth, the CHNRI process is extreme very easy to modify it by adjusting the components of the context and adding additional useful priority– extremelyprocess is CHNRI the Fifth, others [2 and other stakeholders, through being able to assign more importance (weight) to some criteria over the “ where examples many were there Still, have shiftedtoward developmentresearch research anddiscovery [ eases), and the specified time frame longer (eg, 20–30 years), it is very likely that research priorities would der–privileged populations of the world. Had the health issue been less devastating (eg, mild chronic dis ful addressing. First, there is a risk that the spectrum of research ideas submitted and evaluated in the the in evaluated and submitted ideas research of spectrum the that risk a is there First, addressing. ful There are several concerns that were expressed in relation to the CHNRI process and they will need care toaddress inthefutureThe mainpointsofconcern ing the context and criteria [ holders, all of whom can have a substantial influence on the final list of priorities: donors, through defin invest in the process. This means that an appropriate role is given to donors, researchers and other stake Fourth, the CHNRI process is decisions andpredictions betterthananyexpertsinthegreat majorityofcases[ of types certain make to likely be will individuals independently–deciding of collection diverse a that is contributing only a minor fraction to the overall scores. The central idea of the crowd–sourcing principle input individual community,researcheach the with from experts other and researchers of sample the cise can have a decisive (or undue) influence on the final ranks. The scores reflect the way,this exer In the proposedquestions. in the participant of single scoring no and research questions Third, theCHNRIprocess is in theformofanumericaldatasetuponwhichprioritiescanbeset. stored and documented easily be can input processall the and of replicableapproach.stages a All vides also is it Second, equal provided opportunity toquestionsfrom which different categoriesofhealthresearch. questions, research of spectrum endless an handling for framework able previousthe following exercises.ed is method CHNRI the First, uted to several key advances that it proposed. These advances addressed common that concerns persist attrib be can priorities research health setting in method CHNRI the of popularity the Wethat believe The keyadvantagesoftheCHNRImethod ting isstructured, objective, replicable andtransparent. priority–set processfor the as audience, global the to disseminate to easy relativelyare method CHNRI simple, which we believe is one of the main reasons for its uptake by many groups globally that haven't 14, ]. 15]. Seventh, the CHNRI method is reasonably because it clearly defines the context and priority–setting criteria and pro and criteria priority–setting and context the defines transparent,clearly it because 9 democratic. Itrelies ona“crowd–sourcing” approach tobothsubmissionof ]; researchers, through providing research questions and scoring them [ inclusive , fostering “ownership” of the results by ensuring the various groups and adjustable to many different contexts and needs. It is is It needs. differentand many contexts to adjustable and flexible development 483 ” (translational) research questions and “ and questions research (translational) ” inexpensive to conduct. Finally, the results of the because it offeredit because systematic, accept an 11]. June 2017 •Vol. 7No. 1•011004 12]). This begs the questions: The first50applicationsofCHNRI 13]. of collective opinion of discovery ]; 10]; ” ------
VIEWPOINTSPaPErS PaPErSVIEWPOINTS June 2017 •Vol. 7No. 1•011004 Rudan etal. there were stillsomeindividualswhomanagedtoout–performthegroup's prediction [ but cases, of majority vast the in predictions individual out–perform indeed predictions collective that tative properties of human collective knowledge and opinion was designed and conducted to demonstrate the results, and these thresholds should be respected [ exercise to achieve “stable” scores and ranks, above which further addition of experts is unlikely to change per required scorers expert of number minimum the established exercises CHNRI conducted the from exercisesdiffered fromsignificantly [ declined who those [ step this at introduced be could bias response significant a that means which 30–70%, between ranges researchers,invited the programof and responsetypically rate policy–makers the leaders Second, tions. researchques promising particularly some missing is it that and comprehensive not is process CHNRI making. decision– their in exercises conducted the of results the used or priorities, research set to themselves it should be conducted to learn whether they are aware of the CHNRI method and if they have been using institutions funding key at makers policy research with interviews importantly,of More series lished. a comparing the intensity of research on identified priorities before and after each of the exercises was pub ysis of bibliometric indicators, showing the impact of the CHNRI papers on the research community and some impact on health research funders and research communities. This could be achieved through anal least at have process CHNRI the on based publications the that demonstrated be Ultimately,should it may influencethe responses andcouldintroduce biasatthisstep. broadlyareframed, how they or arephrased, questions way groupprocessThe small ofmanagers. very is an important step. It requires knowledge of the subject matter and is therefore usually performed by a several hundreds of research ideas/questions to a number that is feasible for scoring, such as 200 or less, Reducing researchquestions. combining and compiling of process the from comes bias of risk Another of implemented health interventions inreal–time,of implementedhealthinterventions andmany others. implementation over large geographic areas, estimating disease burden, effects of risk factors and impact program monitoring shortage, supplies medical of areas identifying epidemics, of development the ing mation in real time and solve a diverse set of problems ranging from coordinating funding support, alert mobile phones and crowd–sourcing could potentially serve to generate a massive amount of useful infor priorities among further ideas for crowd–sourcing–based solutions in global health. The world–wide web, set to conducted be could exercise CHNRI crowd–sourcing.The and crowds” of wisdom “the on based Finally, the CHNRI method shows how the area of global health may be particularly receptive to solutions based onwidelyavailablespreadsheet software. form, which would further simplify the exercise and the computation of scores and agreement statistics, plat application CHNRI automated fully and phone–based mobile and web–based free a of velopment in CHNRI implementation may facilitate its wider adoption. Another welcome progress would be the de- at a global, national and sub–national level. The development of a massive open online course (MOOC) implementation its help should Moreover,cost issues. low health and lation implementation of ease the opportunities to implement the CHNRI method to address research priorities relevant to all other popu However, its advantages have helped its expansion beyond its initial boundaries. There are clearly many investments in international child health research at a regional level (low– and middle–income countries). The CHNRI method for setting health research priorities was developed to support decision–making for Opportunities forfurtherdevelopmentandimplementation tional andlocalfundingagencies. na regional, international, by implementation its facilitate to tools support of number a developing be and to address health problems outside of child health and nutrition. To encourage its wider use, we will parent and replicable. We believe that it has the potential to be scaled up, especially at the national level, trans- acceptable, widely is method the that shown has domains and contexts of range acrossa method method was developed specifically to address this need. A decade of experience with applying the CHNRI CHNRI The transparent. fully and systematic rarely are setting priority and decision–making of cesses pro the but agencies, funding of variety a by made being continuously are decisions investment Major CONCLUSIONS 9 ]. It should be explored whether those who responded to the invitation to participate in the CHNRI CHNRI the in participate to invitation the to responded who those whether explored be should It ].
484 14, 9 15]. Fourth, a series of experiments on quanti ]. Third, statistical simulations using data sets Third,sets ]. data using simulations statistical www.jogh.org • doi:10.7189/jogh.07.011004 14, 15]. ------www.jogh.org rEFEr ENCES • doi:10.7189/jogh.07.011004 2 1 15 12 11 10 9 8 5 4 3 7 6 14 13 stakeholders. JGlobHealth.2016;6:010303.Medline:27303649 involving III. method: CHNRI the using priorities research health KY.Setting Chan S, Cousens WaznyK, YoshidaS, 2016;6:010505. Medline:26401270 Health. Glob J system. research health global the of needs basic five and function Structure, D. Sridhar I, Rudan method: I.involvingfunders.JGlobHealth.2016;6:010301. CHNRI the using priorities research health Setting al. et R, Bahl D, Sridhar S, KY,Cousens Chan YoshidaS, I, Rudan 2016;6:010501. Medline:27418959 Health. Glob J advances. IV.conceptual method: key CHNRI the using priorities research health Setting I. Rudan Infect Dis.2007;7:56-61.Medline:17182344 Lancet right. priorities our setting diarrhoea: and pneumonia Childhood H. Campbell RE, Black S, Arifeen El I, Rudan Medline:26955468 2016;6:010101. Health. Glob J (CHNRI). Initiative Research Nutrition and Health Child the of legacy The RE. Black 2016;6:010507. Medline:26401271 Yoshida S. Approaches, tools and methods used for setting priorities in health research in the 21st century. J Glob Health. Croat MedJ.2008;49:720-33. Medline:19090596 method. CHNRI of implementation for guidelines investments: research health child global in priorities Setting tive. ResearchNutrition and Health Child al; Initia et M, Black ZA, Bhutta S, Arifeen El S, Ameratunga JL, Gibson I, Rudan doi:10.3325/cmj.2008.3.307 search investments: universal challenges and conceptual framework. Croat Med J. 2008;49:307-17. Medline:18581609 re health child global in priorities Setting al. et I, Carneiro M, Lansang Ann J, Gibson L, Kapiriri M, Chopra I, Rudan of humancollectiveopinion.JGlob Health.2016;6:010503. propertiesQuantitative VI. method: YoshidaCHNRI researchthe health using Setting priorities S. Cousens I, Rudan S, jogh.06.010502 of dementiaby2025.LancetNeurol. 2016;15:1285-94.Medline:27751558 Researchal. et MC, Carrillo KM, Langa I, reduceRudan to C, priorities burdenDuggan global E, the Albanese H, Shah line:22325672 2012;126:237-40. Health. Public world. developing the mobilizing priorities: research health Global I. Rudan searchers. JGlobHealth.2016;6:010302. Medline:27350870 YoshidaWaznyS, Cousens S, KY.Chan K, researchhealth Setting re Involving II. method: CHNRI the using priorities titative properties of human collective knowledge. J Glob Health. 2016;6:010502. Health. Glob J knowledge. collective human of properties titative Rudan I, Yoshida S, Wazny K, Chan KY, Cousens S. Setting health research priorities using the CHNRI method: V. quan Surowiecki J.Thewisdomofcrowds. NewYork: RandomHouse;2004. ties ofhighlymotivatedchampions. activi throughthe historically achieve to managed we than development, and health improveglobal to to be seen whether, as a collective and assisted with modern technology, we could indeed achieve far more that can be particularly highlighted as important across most of the conducted exercises? It also remains system, health of integration or models implementation around eg, questions, priority of set integrated an is there whether explore to interesting be should it addressed, research health of areas different and published, being exercises CHNRI Finally,the horizons? of time number and increasingtargets an with as priorities remain valid beyond 2015, or do some of the CHNRI exercises need to be repeated with new identified researchideas/questions the whether explore Weto 2030. need until will period the to evant which mainly focused on the context defined by the Millennium Development Goals, would remain rel application, method's exploreresultsCHNRI to the the useful whether of be will it years, coming the In organizations. editorial international of reviewedguidelines was accordingpractice addressed,article been this best has to nal editors–in–chief of the sure.pdf (available on request from the corresponding author), and declare no conflict of interest. IR and HC are Competing interests: The authors completed the Unified Competing Interest form at www.icmje.org/coi_disclo reviewed thedraftandprovided importantintellectualcontenttothefinal versionofthepaper. KW,DS, KYC, SC, analyses. the performed and designed SY MT,REB AS, and HN, SEA HC, MC, RB, ZAB, JEL, Authorship declaration: IR and SY conducted the review of the CHNRI exercises and drafted the paper. SC and Funding: ThisworkwassupportedbytheCHNRIFoundationaward totheUniversityofEdinburgh. their personalpositions,andnottheofficial viewsoftheWHO. Disclaimer: Ethical approval: Notrequired. Acknowledgments: None. doi:10.1016/j.puhe.2011.12.001 The views expressed by the authors affiliated to the World Health Organization (WHO) represent (WHO) Organization Health the World to affiliated authors the by expressed views The Journal Journal of Global Health. To ensure that any possible conflict of interest relevant to the jour doi:10.7189/jogh.06.010501 doi:10.7189/jogh.06.010302 doi:10.1016/S1473-3099(06)70687-9 doi:10.3325/cmj.2008.49.720 485 Medline:27350874 Medline:26401269 doi:10.7189/jogh.06.010302 doi:10.7189/jogh.06.010303 doi:10.1016/S1474-4422(16)30235-6 doi:10.7189/jogh.06.010503 June 2017 •Vol. 7No. 1•011004 Medline:27350873 The first50applicationsofCHNRI
doi:10.7189/ Med ------
VIEWPOINTSPaPErS PaPErSVIEWPOINTS June 2017 •Vol. 7No. 1•011101 Huijun Han robert WScherpbier Wei Wang Li Chen Helena vanVelthoven Zhang Qiong Wu China: amixedmethodsevaluationstudy Bao) amongyoungchildreninruralQinghai, complementary foodsupplement(YingYang Monitoring andevaluatingtheadherencetoa material. The onlineversionofthisarticlecontainssupplementary Electronic material: supplementary [email protected] China Beijing 100600 12 SanlitunLu UNICEF China ronment andSanitationSection Health andNutrition,Water,Envi- Suying ChangPhD [email protected] China Beijing 100020 Chaoyang District No. 2Yabaor Capital InstituteofPediatrics Childhood Development Department ofIntegratedEarly Yanfeng ZhangMSc Correspondence to: 5 4 3 2 1
Beijing, China Peking UnionMedicalCollege, a Basic MedicalSciences,Chinese and Biostatistics,Instituteof Department ofEpidemiology United Kingdom Imperial CollegeLondon, of PrimaryCareandPublicHealth, Global eHealthUnit,Department UNICEF China,Beijing,China Environment and Sanitation Section, Health andNutrition,Water, Institute ofPediatrics,Beijing,China Childhood Development,Capital Department ofIntegratedEarly China Health EducationCenter,Qinghai, Framing andPastorala Department ofHealthEducationin cademy of Medical Sciences, cademy ofMedicalSciences,
1 , SuyingChang 1 , XiaozhenDu 1 1
oad
, Michelle , Yanfeng 4 , MinXing
reas, Qinghai reas, Qinghai
2 2 3 5 1 , ,
, ,
January 2013; 2) N 2) 2013; January (N = survey (baseline months 6–23 aged children of caregivers with surveys tional proving children's health status in rural areas in China. We conducted five cross–sec trial aimed to evaluate the effectiveness of community–based YYB distribution on im The trial. evaluation interventional controlled a during place took that groups cus Methods This mixed methods evaluation study combined data from surveys and fo ity ofYYB;and3)evaluatecommunity–basedstrategiestoimprove theprogram. children’sassess 2) YYB; acceptabil of and distribution to adherencemonitor 1) to: We China. aimed in programs YYB implementing of challenges and experience the young children in rural areas in China. However, there is a lack of knowledge about of nutrition improve to programs YYB) YangBao, (Ying supplement food mentary Background mentation inothersettings Chinaandelsewhere. ensuring continuity of YYB supply. Future programs need to monitor program imple the taste of YYB; strengthening health education of village doctors and caregivers; and herence trends. This led to strategic changes to the intervention over time: improving ad and implementation program understanding for important critically was China Westrural in programmonitoring that showed study evaluation This Conclusions CI 1.072–1.393)duringthepasttwoweekshadsignificantly loweradherence. diarrheawith (OR children that showed analysis Multivariate YYB. taking after children their in ment improve health positive perceive not did caregivers of fortunately,60% than more Un YYB. taking disliked children that reason main the was which YYB, of taste the Self–reported acceptability increased from 43.2% to 71.8%, partly due to improving 2014; (January survey follow–up third the in 53.6% to 2013) (August have led to a decrease of high adherence from 64.1% in the second follow–up survey may which months, six for YYB receive not did caregivers stock–out, YYB planned training sessions with village doctors could increase adherence. However, due to un 2013) to 81.4% in the last follow–up survey (August 2014; (August survey follow–up last the in 81.4% to 2013) High adherence to YYB increased from 49.4% in the first follow–up survey (January dren ever took YYB. Caregivers mainly knew about YYB through their village doctors. Results deeper insightintothequantitativeresults. gain to used were which findings, qualitative explore to used was analysis Content focus groups with73 caregivers and health workers involved in the YYB distribution. 10 conducted we Also, adherence. high with associated factors of intervals fidence A logistic regression model was developed to obtain odds ratios (OR) with 95% con defined as children who consumed at least four YYB sachets during the previous week. was survey each in adherence High surveys. the in acceptability and consumption YYB characteristics, household on caregivers from collected vey.was Information stage cluster sampling technique to select mothers with eligible children for each sur N = 2186 in August 2014) in one rural county in Qinghai Province. We used a two– 1804) in August 2012 and four follow–up cross–sectional surveys: 1) N Around 90% of caregivers had ever received YYB and more than 80% of chil Large investments are currently made in community–based comple community–based in made currently are investments Large = 1.216, 95% CI 1.025–1.442), cough or fever (OR fever or cough 1.025–1.442), CI 95% 1.216, = 2187 in August 2013; 3) N 3) 2013; August in 2187 486 = www.jogh.org 504 in January 2014; and 4) 4) and 2014; January in 504 • doi:10.7189/jogh.07.011101 P < 0.0001). Repeated Repeated 0.0001). global journal of = P 1.222, 95% 95% 1.222, < 0.0001). 0.0001). = 494 in health ------www.jogh.org • doi:10.7189/jogh.07.011101 intervention thatmightbehelpfulforimprovementintervention of larger scaleprograms inChinaandelsewhere. the in changes strategic to led monitoring how illustrate will This improveadherence. to strategies based community– evaluate 3) and YYB; of acceptability and children'sadherenceto assess 2) YYB; of tribution dis monitor 1) to: aimed paper current this Qinghai, in trial controlledinterventional the on Based tion. approach. High adherence to MNPs is critical for achieving the maximum health benefits of the interven distribution community–based the with program of challenges and experience program adherence, the Although such a large–scale national nutritional program was carried out in China, no study documented tices andreduce anemiaprevalence [22]. prac feeding improve can counseling dietary with combined supplements food complementary based community– that reported and China, in areas rural in status health children's improving on tribution evaluation trial in Qinghai from 2012 to 2014 to evaluate the effectiveness of community–based YYB dis wereprogramwhich national 2012, the with consistent [ viding free YYB to all children aged 6–23 months in 15 out of 34 poor counties in Qinghai Province since In addition, as a part of Qinghai–Tibet Plateau, the provincial of government Qinghai has also been pro months inruralareas [20]. 6–23 aged children million one than more reach to estimated was which China, in provinces 21 in ties gram was scaled up between 2013 and 2014 [ pro This [17]. areas rural poor in months 6–23 aged children for YYB free provides which areas, rural eration initiated a national community–based nutritional program to improve children's nutrition in poor Women'sAll–China Fed and Commission Planning Family and Health National Chinese the 2011, In Children’s Fund (UNICEF), the United States Centers for Disease Control (US CDC) and China CDC [ earthquake–affected counties in Sichuan, Gansu and Shaanxi provinces, supported by the United Nations 30 around to provided was YYB free 2011, and 2010 Between YYB was recommended for scale–up in disaster and poor rural areas to improve Chinese children’s health. vitamin B vitamin itoring of YYB distribution and evaluation of children’s adherence to and acceptability of YYB only took took only YYB children’sof of acceptability evaluation and and to distribution adherence YYB of itoring Caregivers and their children aged between 6–23 months were main participants of our evaluation. Mon [ Qinghai in trial interventional controlled the in embedded was evaluation methods mixed current This Study designanddatasources METHODS on how to use MNPs for home fortification of foods for children aged 6–23 months [ desh [8 grams are currently being scaled up at a national level in several developing countries, including Bangla mented in many countries [ vitamin and mineral deficiencies and improving nutritional status of young children has been well docu- micronutrient deficiencies among children younger than two years [ were developed and have been proposed as an important intervention for addressing undernutrition and During the last decade, multi–nutrient powders (MNPs), which are home nutrition fortification products, dren’s nutritionandhealthinruralChina. and did not change between 2005 and 2009 [ respectively,2010, in 20.5% and 28.2% was months children13–24 Chinese and of months 6–12 aged 2010, respectively, which is more than two times as in the areas national rural average. poor Furthermore, in anemia 20.3% prevalence and 8.0% was under–five children Chinese stunted and underweight of malnutrition of children is still a prominent problem, particularly in poor rural areas [ Although China has made great achievements in improving children's health during the past two decades, ] in 2009 and made YYB commercially available on the market [ market the on available commercially YYB made and 2009 in [16] (GB/T22570–2008) Chinese government approved and issued the National Standard for Complementary Food Supplements and improve children’s developmental quotient (DQ) [ small–scale efficacy study carried out in Gansu from 2001 to 2004 showed YYB can reduce anemia [ carbohydrate (3.0 g), vitamin A (250 A vitamin g), (3.0 carbohydrate multiple micronutrients [ as well as flour soy fat proteinfull and throughof acids inclusion fatty essential contains which veloped Yingde Yangcalled was children (YYB) young Bao and infant for producedMNP domestic a China, In 22]. We combineddatafrom andfocus groups. surveys ], Mongolia[9 12 (0.5 g), folic acid (75 mg), elemental iron (7.5 mg), zinc (5 mg), and calcium (200 mg). A A mg). (200 calcium and mg), (5 zinc mg), iron(7.5 elemental mg), (75 acid folic g), ], Kenya[5 12, 3 13]. Each sachet of YYB contained the following: protein (3.0 g), fat (1.0 g), - 6 ]. In 2011, the World Health Organization (WHO) developed a guideline ], Nepal[10]andNigeria11]. g),vitamin D g),vitamin 487 1 18– ]. Therefore, more efforts are still required to improve chil
20] and until 2014 the program had covered 341 coun 3 (5 15]. With this evidence on the efficacy of YYB, the ]. We21]. controlleda out carried interventional g), vitamin B vitamin g),
000 children aged 6–23 months in eight eight in months 6–23 aged children 000 2 ]. The efficacy of MNPs in reducing 1 (0.5 mg), vitamin B vitamin mg), (0.5 June 2017 •Vol. 7No. 1•011101 7 ]. Also, MNPs pro 1 ]. The prevalence ]. Moreover,12]. 2 (0.5 mg), mg), (0.5 ], 14], 13] ------
VIEWPOINTSPaPErS PaPErSVIEWPOINTS June 2017 •Vol. 7No. 1•011101 Wu etal. Table 1. Six focusgroups Children aged6–23monthsandtheircaregivers (n = Intervention started Baseline survey d nln uvyChildren aged6–23months and theircaregivers (n = ended Intervention Children aged6–23monthsand theircaregivers (n = Endline survey Children aged6–23monthsandtheircaregivers (n = Mini 2survey Midterm survey Children aged6–23monthsandtheircaregivers (n = Mothers(n = Four focusgroups Mini 1survey ata
source Sources foodsupplementYing ofmonitoringdatafor consumedcomplementary Yang Bao(YYB)intervention program. Inaddition,wereport qualitativedataonlyondifficulties withYYBdistribution. ence of children to YYB, caregivers’ experience with YYB, YYB awareness and lessons learnt from the YYB adher distribution, YYB for data qualitative by followed data We quantitative [23]. report first findings quantitative the validate to data quantitative with findings qualitative the compared we Also mentation. implementation. We integrated quantitative and qualitative data to show findings on intervention imple program of understanding our increase to caregivers and workers health local with discussions group with YYB, YYB awareness and lessons learnt from the YYB program. Qualitative data were from ten focus children’s experience distribution, caregivers’ YYB YYB, of to adherencecoverage assess to aimed which The intervention county lies in the northeast of Qinghai province, with the area of 3423.9 km the adultfemaleliteracyrate,andproportion ofpipedwatercoverage. residents, rural for income capita per annual including: counties, two the between conditions economic socio– and cooperate, to government local the of willingness the considered we selection, For Qinghai. Wefromcounties. these controlthe county,selected County,Guinan in counties fromremaining19 the Province before we designed the trial and therefore we selected one county,intervention Huzhu County, Qinghai in counties 15 in program the implement to decided already had government provincial The Study setting ( county tervention countyinthetrial,andthereforeplace intheintervention thedatainthiscurrent paperare from thein infant and young child feeding, and morbidity status. Trained fieldworkers from the School of Public Public of School the from Trainedfieldworkers status. morbidity and feeding, child young and infant characteristics, socio–demographic included which data, follow–up and characteristics baseline collect We used the adapted Maternal, Newborn and Child Health household survey (MNCH HHS) tool [ Data collection one survey. We conducted the in surveys the same villages; this meant that children could be included in more than size of504inbothsurveys. mini 1 and mini we 2 only surveys, collected the data on hemoglobin levels, and thus we used a sample However,collected. survey,were endline level the and hemoglobin in and height weight, on data the as survey,baseline the survey in midterm 1973 of size sample Wea [22]. used paper study interventional controlled the of effectiveness the in reported procedurewere sampling two–stage and size sample The Sample sizeandsampling aged between6–23monthswere participantsofourevaluation. children their and caregivers Main (Table1 ). county respectively,intervention the survey), in (endline 2013 August and survey), 2 (mini 2014 January survey), (midterm 2013 August survey), 1 (mini 2013 January in surveys cross–sectional follow–up four and 2012 August in Wesurvey baseline a conducted Quantitative approach intervention intervention county. The annual per capita income of rural residents is ¥ 5691 (US$ 872.43) in 2011 [ population of 370 of population p MCH workers in township hospitals (n hospitals township in workers MCH (n = artIcIpants 6), mothers(n = 12) andgrandparents (n =
4), and grandparents (n =
540, with 93.1% of rural population. There are 19 townships and 294 villages in the the in villages 294 and townships 19 are There population. rural of 93.1% with 540, Table1 ). Quantitative data were from five representative cross–sectional surveys, surveys, cross–sectional representative five from were data Quantitative ). = 1 fcide gd62 otsQaiaieArl21 7months April2013 Qualitative 11) ofchildren aged6–23months 11); village doctors (n doctors village 11); 9) of children aged 6–23 months 488 1804) 2186) 504) 2187) 494) = 20); fathers fathers 20); uniaieAugust2012 Quantitative uniaieAgs 0423 months 16months August2014 11 months Quantitative 2014 January August2013 Quantitative 4months Quantitative 2013 January Quantitative ulttv oebr21 2months November2012 Qualitative research t ype
oF www.jogh.org
September 2012 August 2014 • doi:10.7189/jogh.07.011101 d ate n aFter 2 . It has total umber
InterventIon
oF 25] to
months 24]. - - - www.jogh.org • doi:10.7189/jogh.07.011101 two–tailed P considered (CI). We intervals confidence 95% and (OR) Ratios Odds Weregression.present stepwise using covariates relevant the for adjusted were Models presented. are model multivariate final the in ed gle factor analysis. Multivariate analysis was used to assessed, and only those that were significant includ data from the four follow–up surveys to explore the factors. All relevant factors were first selected by sin WeYYB. to adherence high Wewith the associated combined factors identify regressionto logistic used eral questions. survey,each in participants “Don’t/Didn’tthe answered all who weresev those tors in including know” line Supplementary Document). line Supplementary formation that we obtained from a pilot text messaging survey in October 2012 (see Appendix S1 in For the four follow–up surveys, we developed questions on YYB distribution and consumption using in- views withcaregivers invillageclinics. inter conducted interviewers then and registration, for clinics village to come first to caregivers asked Health, Qinghai University collected data for the five surveys using smartphones. During each survey, we the permission of each participant. Tape recordings were transcribed verbatim in Chinese by a medical medical a by Chinese in verbatim Tapetranscribed participant. wererecordings each of permission the sions were conducted in Mandarin, typically lasting around 30 minutes, and were digitally recorded with were done at a place convenient for them. Caregivers were invited to village clinics to participate. Discus in S1 pendix (Ap guides group focus the developed team study The discussions. group focus conducted Pediatrics researcherone and Center of fromEducation fromInstitute Health facilitator Capital Qinghai local One Data collection child aged6–23months(7focusgroups). fromdifferentgroups). focus (2 township a Caregiversin werevillages a fromhad and villages same the .MCH workers came from different township hospitals in the county (1 focus group), village doctors were We used convenience sampling. The participants in the focus group were independent from the surveys. Sampling YYB implementation:sixinNovember2012andfourApril2013( We conducted 10 focus group discussions in the intervention county to obtain a better understanding of Qualitative approach Pearson used compare differences for the mean sachets. For binary or categorical variables, we present percentages. We survey.WeANOVA [ used each in week previous the during surveyed children by consumed YYB of sachets mean and children, We present the mean and standard deviation (SD) to describe the age of mothers and main caregivers of sis. We carried out statistical analysis with SAS 9.2 for Windows (SAS Institute Inc., North Carolina, USA). each survey. After the data cleaning, we converted the database into databasefile (dbf) for the final analy for Washington,sheet WA,Seattle, USA) (Microsoft, Excel Microsoft a into data each pool and formed Data of each interview was automatically stored as “.txt” form in each smartphone, and we manually trans Statistical analysis ter awhile;5=Disliked,reasons fordislike….;8=Don’t know.” child like taking YYB? 1=Like very much; 2=Liked; 3=Neutral; 4=Disliked at the beginning, but liked af Children’s acceptability was measured though one question in the questionnaire “How do you think your Definition ofYYBacceptability with thedefinitioninotherstudies[6 proportion of children who consumed at least four YYB sachets during the previous week, which consist the as defined was adherence High week?” previous the during consumed child your did YYB of chets adherencevariable measuredoutcome was The through questionnairesa the many in “How question a Definition ofhighadherence –values of<0.05forasignificantdifference. Online Supplementary Document Supplementary Online 2 –test and Fisher exact test to compare binary and categorical variables. The denomina The variables. categorical and binary compare to test exact Fisher and –test ] analysis to detect statistically significant differences in age, and T–test to to T–test and age, in differences significant statistically detect to analysis 26] ]. 489 ). Discussion with MCH workers and village doctors doctors village and workers MCH with Discussion ). Adherence toacomplementaryfoodsupplement Table 1). June 2017 •Vol. 7No. 1•011101 On------
VIEWPOINTSPaPErS PaPErSVIEWPOINTS June 2017 •Vol. 7No. 1•011101 ‡Two were interviewees missingforthis calculation. †Two were interviewees missingforthis calculation. wasmissingforthiscalculation. *One interviewee Wu etal. Table 2. Primary school Primary Illiterate Education,% (n) Do notknow Senior highschoolorabove Junior highschool Age (year),mean(SD) Other Grandparent Father Mother Relationship withchildren, %(n) Main Father workingoutsidehometown Mother workingoutsidehometown Age (year),mean(SD) Mothers w–ekpeaec fcuho ee 90(8)3. 11 53(7)4. 21 39.8(870) 43.9(221) 35.3(772) 38.7 (191) 49.0(884) Two–week prevalence ofdiarrhea Two–week prevalence ofcoughorfever Currently breastfeeding Girl Boy Sex, %(n) 18–23 months 12–17 months 6–11 months Age, %(n) Children s urveys caregivers Characteristics of surveyed caregiversCharacteristics ofsurveyed andtheirchildren 30% of children were currently breastfed. Two–week prevalence of ( cough/fever, surveys cross–sectional and the diarrhea in were participate aroundto agreed invited were who caregivers All Population inquantitativesurveys r viewees read theinformationsheetandprovided writteninformedconsent. The evaluation study was approved by the Ethical Committee of Capital Institute of Pediatrics. All inter Ethical considerations reviewed thetranslatedthemes.We listallthekeythemesthatweidentified. terpretation Finally,findings. the of DXZ and themes English the into translated quotes WQ related in and themes the on reached was consensus until themes the refined further They discrepancies. and key themes independently. Then the researchers compared the themes and discussed areas of agreement MAXQDA 11 (VERBI Software GmbH, Berlin, Germany) to organize data along the previously identified use and transcripts the read first DXZ) and (WQ study the in Twoinvolved researchersdata. Chinese We conducted content analysis [ Analysis ed thetranscripts. again to correct any errors. Finally, the study team member who participated in the focus groups validat- student from Qinghai University, and then checked by another medical student by listening to the tapes ESULTS
b aselIne 94(37 631.)4. 1.) 83(32 38.8(14.1) 38.3(13.2) 40.0(13.6)* 36.3(12.3) 39.4 (13.7) 20(9)2. 17 14(6) 73(7 18.9(414)† 17.3 (87) 21.4(467)‡ 23.7(117) 22.0 (396) 50(1)3. 10 67(00*3. 13 47.8(1045)† 51.8(1131)† 34.3(173) 46.7(1020)* 58.5(295) 57.5(1257) 52.4(1146)* 34.4(170) 41.1(207) 64.8(320) 45.0 (812) 63.9(1397) 53.2 (960) 47.6(235) 39.2 (707) 41(3)1. 5)2. 59 23(2 13.6(515) 12.3 (62) 26.1(569) 11.9(59) 24.1 (435) 67(0)1. 8)1. 34 45(3 15.9(348) 14.5 (73) 25.3(553) 45.2(988) 14.8(324) 27.8(140) 54.8(1198) 41.5(209) 35.5(776) 27.1(593) 58.5(295) 17.6(87) 29.1(635) 45.0(984) 35.5(775) 33.7(170) 55.0(1203) 36.2(179) 37.7(190) 16.7 (302) 45.7(226) 25.6(144) 30.9(676) 54.3(268) 29.5(645) 26.8 (484) 39.6(866) 46.8 (844) 28.9(143) 53.2 (960) 41.5(205) 29.6(146) 39.4 (710) 26.8 (484) 33.8 (610) 1375 34(6)4. 89‡4. 22 40.3(880)† 40.1 (202) 40.2(879)‡ 33.4(165) 41.3(745) 2058 50(7)3. 69‡3. 11 34.1(746)† 37.9 (191) 31.5(689)‡ 35.0(173) 32.0(578) 69(.)2. 46 91(11 92(04 28.6(9.6) 29.2(10.4) 29.1(11.1) 27.4(4.6) 26.9 (4.9) 0.6 (11) 4.1 (74) 1.2 (21) 0.6 (11) ( 84 m (n = 1804) 27] by examining the major themes and patterns that emerged from the InI 490 1( 9)m 1(n = 494) . 3)57(2) . 2)5.9(128)† 4.6(23) 5.7(125)‡ 7.3 (36) . 3 . 2) . 1 0.8(17)† 0.2(1) 1.1(25)‡ 0.6 (3) . 0 . 1) . 2)0.2(5)† 0.2(4)† 4.0(20) 3.2(16) 0.5(12)* 0.4(8)* 0.0 (0) 0.8 (4) Idterm ( 17 m (n = 2187) www.jogh.org InI 2( 0)e 2(n = 504) • doi:10.7189/jogh.07.011101 ). Around Table2). ndlIne (n = 2186) - - Table 3. ¶We didnotaskcaregivers thisquestionintheMini1survey. ||Mini 1vsEndline. §Mini 1vsMini2. ‡Mini 1vsMidterm. †Data missingfor8children. *Data missingfor1child. www.jogh.org Table 4. hours Proportion of children who took YYB within the last 24 ing YYB¶ consum- still currentlywere who childrenProportion of rprino hlrnwoee osmdYB8.%(0)9.%(06 36 44 98.1%(2144) 93.6% (464) 95.9%(2096) 82.0%(405) Proportion ofchildren whoeverconsumedYYB Proportion of children whose caregivers ever received YYB children duringtheprevious surveyed week Mean (standard deviation) sachets of YYB consumed by sumed 4sachetsofYYBormore) (con adherence high had who children of Proportion Didn't know Others There isnoYYBinthevillageclinic Didn't wantYYB Not athomewhendistribution Children were justsixmonths Didn't knowthedistributionofYYB r easons Complementary food supplementYingComplementary Yang Bao(YYB)distributionandconsumptionbychildren Reasons for “not received complementary complementary foodsupplementYingReasons for“notreceivedcomplementary complementary Yang Bao(YYB)” • doi:10.7189/jogh.07.011101 fathers, 16mothers,and20grandparents. 6 doctors, village 20 workers, MCH township 11 groups: focus the in participated people 73 of total A Population infocusgroups them were illiterate. of around40% and years 40 was caregivers main of age mean The grandparents.were around30–40% and mothers were caregivers main the of respectively. half 15%, than and more 40% surveys, five all In six months”infourfollow–upsurveys. reasons for “not received” were “caregivers didn’t know the distribution of YYB”, and “children were just YYB. Although most caregivers of children had ever received YYB, still a small part of caregivers did not receive program.public healthservice basic the of supervision regular their undertaking were they while distribution YYB monitor to villages catchment their to went workers MCH round.month one for YYB Every distribute to days seven to one doctors village took it villages, different in children of number different to Due received. had they YYB givers had to use empty YYB bags and boxes to exchange a new box of YYB to ensure children consumed Furthermore, YYBwasgiventochildren whoreceived vaccinationsintheclinics.Insomevillages, care uted to caregivers through home visits or by asking caregivers to visit clinics through mobile phone calls. line survey). Once a month, village doctors received YYB from their township hospitals, and then distrib cus groups said that YYB was mainly distributed by village doctors from September 2012 (after the base aged 6–23 months in villages had ever received YYB ( children of 90%) (around caregivers most that surveys cross–sectional follow–up four the Wein found YYB distribution Table 4 shows the distribution of reasons why caregivers did not receive YYB and we found in main - 35 16 .%(33 82 29 78.8%(1722) 48.2% (239) 3.3%(1383) 23.5% (116) 77 43 71 22)9.%(7)99.0%(2164) 95.6% (474) 97.1%(2123) 87.7% (433) 94(4)6. 10)5. 26 81.4(1780) 53.6(266) 64.1(1402) 49.4 (244) (n = 494) (4 . 30 . 29 . 31 . 22 .02<.01<0.0001 <0.0001 0.0002 5.8(2.2) 3.9(3.1) 4.7(2.9) 4.0 (3.0) m months InI – 1 ) m 12 28 InI 5 4 0 2 1 1 23 10)7.%(6)92.9%(2032) 73.1% (363) 82.3% (1800) (n = 2186*) 491 (11 m Idterm months
) Table 3).Both health workers and caregivers in fo- (n = 496)† (16 m m Idterm 14 15 13 6 2 0 7 months InI 2 ) Adherence toacomplementaryfoodsupplement (n = 2186) (23 e ndlIne months June 2017 •Vol. 7No. 1•011101
m ) 25 InI 2 3 1 0 6 5 2 <0.0001 <0.0001 <0.0001 <0.0001 1 2 p3 p2§ p1‡ – – – 000 <0.0001 <0.0001 .23<0.0001 0.0243 .76<0.0001 0.0746 .84<0.0001 0.1824 e ndlIne 5 1 3 1 3 5 4 || - - -
VIEWPOINTSPaPErS PaPErSVIEWPOINTS June 2017 •Vol. 7No. 1•011101 Wu etal. stopped consumingYYBcurrently, with89,386,133,154ineachfollow–upsurvey, respectively. denominator ofthisfigure were thenumbersofcaregivers whosechildren hadnever received orconsumedYYB sachets children consumed during the previous week increased from 4.0 sachets to 5.8 sachets ( from 23.5% in the mini 1 survey to 78.8% in the endline survey ( survey endline the in 78.8% to survey 1 mini the in 23.5% from (Tableimplementation proportionThe 3). childrenof increased hours 24 last the within YYB took who program YYB the of progress the with increased children by consumed YYB of coverage the general, In Adherence ofchildren toYYB Figure 1. survey,midterm the in (179/313) 57.2% survey,2 Mini the in (43/50) 86.0% the in (48/71) 67.6% and survey,1 Mini the in (27/53) 50.9% YYG of taste the disliked children that were caregivers by reported 70% caregivers reported that their children liked taking YYB. The most common reason for disliking YYB than more survey endline the of time the at time; over increased YYB taking liked who children of tion Figure 3 shows that most children’s perceptions on YYB were either “neutral” or “like” YYB. The propor YYB acceptability together.” (agrandparent, focusgroup inApril2014) “ dren becauseofforgetting increased throughout thefourfollow–upsurveys. undertaken in August. In addition, the proportion of caregivers who skipped sachets of YYB to their chil were which surveys endline and midterm January,the in to undertaken compared were which surveys 2 mini and 1 mini the in higher were sickness to due YYB of sachets skipped temporarily who children sachets of YYB to their children was because children got cold or diarrhea (Figure 2). The proportions of skipped temporarily they which in situation common most the that reported Moreover,also caregivers ence. (OR = herence ( herence outside hometown (OR = (OR children the of age higher analysis, multivariate In disliked takingYYB.”(Figure 1). currently was “Not received”; however, in the last three follow–ups, the main reason changed to “children YYB consuming survey, stopped mini or first consumed the never In had children why reason main the ( ed indicatorsbetweenthemidtermandmini2surveys tively, which indicated a similar trend. However, there were a significant decrease in all adherence–relat respec 81.4%, and 53.6% 64.1%, 49.4%, was survey cross–sectional follow–up each for week) vious proportionThe childrenof adherenceYYB prehigh the had during who sachets YYB four least at (took Not feed (YYB) when my child got a cold, (I am) afraid that YYB could not be given with medication for a cold cold a for medication with given be not could YYB that afraid am) (I cold, a got child my when (YYB) feed Not 1.222, 95% CI 1.072–1.393) during the past two weeks were associated with having low adher low having with associated were weeks two past the during 1.072–1.393) CI 95% 1.222, Reasons children foodsupplementYing hadneverconsumedcomplementary Yang Bao(YYB).The Table 5 Table
). However, had diarrhea (OR diarrhea had However, ). 0.795, 95% CI 0.692–0.913) were significantly associated with having high ad- 492 = 1.216, 95% CI 1.025–1.442), had cough or fever or cough had 1.025–1.442), CI 95% 1.216, = P < 0.976, 95% CI 0.962–0.990), father working working father 0.962–0.990), CI 95% 0.976, 0.0001) (Table 3). P < 0.0001); meanwhile, the average average the meanwhile, 0.0001); www.jogh.org • doi:10.7189/jogh.07.011101 P < 0.0001). 0.0001). ------their children. Thedenominatorofthisfigure isthenumbersofcaregivers whosechildren are stillconsumingYYB. www.jogh.org Figure 2. Situations where caregiversfoodsupplementYing temporarilyskipsachetsofconsumedcomplementary Yang Bao(YYB)to • doi:10.7189/jogh.07.011101 ). However, only less than 20% of caregivers had per had caregivers of 20% (Tablethan However,children less 6). their only to YYB giving started they In all follow–up surveys, more than 60% of caregivers did not perceive any change in their children after cus group inNovember 2013) (YYB) in the follow–up surveys. Caregivers’(YYB) inthefollow–upsurveys. experiencewithYYB. soybean milk powder. Children even refused eating meals, when YYB added to their meals.” “ ple addingsomesugar. and smell. They suggested that it should be changed into a sweeter taste that children like, by for exam taste soybean the to just YYB taking like not did children their that said caregivers Some improved. be In all focus groups, MCH health workers, village doctors and caregivers said that the taste of YYB should it”, and“diarrhea”. endline survey). Other reasons were: “did not know why children disliked taking YYB”, “nausea and vom Figure 3. The critical problem is that the taste of YYB is not good, and children are not willing to consume. (YYB) tasted like Caregivers’ perception foodsupplement Ying ofchildacceptanceconsumedcomplementary Yang Bao 493 Adherence toacomplementaryfoodsupplement June 2017 •Vol. 7No. 1•011101 (a village doctor, fo- - - -
VIEWPOINTSPaPErS PaPErSVIEWPOINTS June 2017 •Vol. 7No. 1•011101 Wu etal. Table 6. §Mini 1vsEndline. ‡Mini 1vsMini2. †Mini 1vsMidterm. *The denominatorofthistablewere thenumbersofcaregivers whosechildren were stillconsumedYYBcurrently. oiiehih an . 2)72(3)39(4 . 16 .64008 0.1195 0.0585 0.0002 0.9654 0.0089 5.2(106) 0.7703 <0.0001 12.1(246) 0.8334 3.9(14) 0.5025 12.1(44) 74.7(1517) 7.2(130) 0.5(2) 18.4 (331) 64.2(233) 7.2(29) Increased cognitive ability Prevented diseases 61.7(1110) 19.0(77) Increased appetite Positive heightgains 63.5(257) Positive weightgains Perceived changesinchild’s healthafterfeedingYYB: No changesobserved v arIables ( 8 m (n = 78) Mothers’ experience with complementary foodsupplementYingMothers’ experiencewithcomplementary Yang Bao(YYB)* Table 5. in November2013). children”to vitamins and iron,zinc providecalcium, can YYB that me told doctor Qualitative data showed that village doctors played very important roles in YYB health education. increased throughout thefourfollow–upsurveys. YYB box, and the proportion of caregivers reporting that they ever received YYB information from the box OR –oddsratio,CIconfidenceinterval lage doctor were around 80% in each survey (Figure 4). The other major source of YYB information was from information vil YYB caregivers who received and information source YYB first of werethe doctors survey.endline Village the in 64.1% to survey mini–1 the in 43.5% from increased YYB on formation in the received ever had they that reported who caregivers proportionof the surveys, follow–up the In YYB awareness ber 2013) but he is able to walk by holding something, and grasp things himself.” (a grandfather, focus group in Novem- “ any changesintheirchildren. improved and less colds occurred after eating YYB; however, some caregivers said it was too short to see had children their of immunity growth, appetite, the that mentioned groups focus in caregivers Several ceived positiveheightgains,increased appetiteandimproved children’s cognitiveability. per caregivers of 10% than less children; their in prevented diseases and gains weight positive ceived Endline –Mini1 1.126(0.966,1.312) Mini 2–1 Midterm –Mini1 0.1281 Surveys: Cough andfever Diarrhea 2.3153 Child wascurrently breastfed Main incomesource offamilywaswork Father workingoutsidehometown 0.1188 Mother workingoutsidehometown Main caregiver attendmiddleschoolorabove Age ofmaincaregiver (years) Mother –others Mother –father Mother –grandparents Main caregiver: Age ofchild(months) F My child is heavier than before, and he has never got cold, even if I took him out every day. Now he is nine months, actors Factors associatedwithhighadherencefoodsupplementYing tocomplementary Yang Bao(YYB) InI 31(3 67(0)83(0 20(4)005 .360.5492 0.0316 0.0758 12.0(244) 8.3(30) 16.7(300) 13.1 (53) 1(n = 405) . 3)82(4)88(2 . 16 .35067 0.1879 0.6973 0.3385 7.7(156) 8.8(32) 8.2(147) 9.6 (39) % ( n
) m Idterm 3.6 (64) % ( (n = 1800) n ) m 494 InI 007 .10057 0.926(0.706,1.214) 0.251(0.201,0.313) 0.641(0.517, 0.794) <0.0001 0.5770 <0.0001 148.7340 0.3110 16.6102 0.890(0.769, 1.030) –1.3838 0.795(0.692,0.913) –0.0771 0.890(0.728, 1.089) 0.1181 –0.4450 0.0012 0.994(0.984, 1.003) 0.2585 2.4428 10.5547 0.2038 1.2767 –0.1163 –0.2293 1.6150 –0.1161 0.976(0.962,0.990) –0.0063 0.0012 10.5029 –0.0244 .05902 .071.222(1.072,1.393) 1.216(1.025,1.442) 1.103(0.930,1.309) 0.0027 0.0245 0.2603 9.0021 5.0589 1.2669 0.2005 1.006(0.769,1.030) 0.1958 1.599(0.664,3.851) 0.0982 1.136(0.856,1.507) 0.9904 0.2951 0.3785 0.0001 1.0964 0.7757 0.0063 0.4695 0.1271 2(n = 363) 1.9 (7) % (
n ) e ndlIne w ald . 6)001 .610.0021 0.0651 0.0011 3.3 (66) (n = 2030) % ( n ) www.jogh.org p – value p (a grandfather,group (a focus 1† • doi:10.7189/jogh.07.011101 p 2‡ or (95%cI) “Village p 3§ - - - www.jogh.org Figure 4. Source ofcaregivers'foodsupplementYing informationonconsumedcomplementary Yang Bao(YYB). • doi:10.7189/jogh.07.011101 the leaflets in four follow–up surveys (Figurethe leafletsinfourfollow–up surveys 4). from YYB of information got they that reported caregivers of 10% than less However,only description. through face–to–face counseling and we modified the leaflet by adding more pictures to replace the text information YYB deliver to doctors village encouraged we Therefore, illiterate. were caregivers main However, usage. and benefit YYB more that found we 2012) (August of 40% than survey baseline the at of description detailed a contained which caregivers”, to letter “a called leaflet a developed Weinitially fromLessons learnt YYBprogram implementation “ their children YYBorsomechildren dislikedtakingYYB. feed to reluctant and uncooperative were villages the in caregivers several that indicated Villagedoctors November 2013) “Although we work very hard on YYB distribution, caregivers are ungrateful.” “We are verybusy, andstillhavetodistributeYYB.” village doctorsalsomadecomplaints. towardhowever,attitude distribution; positive YYB a had townships their in doctors village some the of half than more that agreed generally county intervention the in hospitals township from workers MCH Difficulties withYYBdistribution home, Iread from theintroduction book[ontheYYBbox].”(amother, focusgroup inNovember2013) “ village doctorsdemonstratedintheirclinicstocaregivers howtogiveYYBchildren. When distributing YYB to caregivers, village doctors encouraged them to give YYB to their children. Some be givenallowance. during the YYB distribution. Both MCH workers and village doctors asked whether village doctors could transportation and calls for fees the all pay to had doctors Village clinics. village to come not did givers care if door to door YYB bring to had they Sometimes YYB. distribute and caregivers of inform to days working four to two use to had they month, Every load. work their increased distribution YYB the that said doctors village Several workers. health for distribution YYB on allowance no was there addition, In courage hertofeed,butshestilltoldmechilddislikedtakingYYB.” There is a grandmother in my village who did not feed YYB to her grandchild. I called her for threefor her en called to I times grandchild. her to YYB feed not did who village Theremy in grandmother a is back came I After YYB. of benefit the me told doctor village the YYB, bring to clinic village the to came I When 495 Adherence toacomplementaryfoodsupplement (a village doctor, focus group in June 2017 •Vol. 7No. 1•011101 - -
VIEWPOINTSPaPErS PaPErSVIEWPOINTS June 2017 •Vol. 7No. 1•011101 Wu etal. out ( out took YYB within the last 24 hours and high adherence significantly decreased as a result of the YYB stock– In the Mini–2 survey (16 months after intervention start, January 2014), the percentage of children who was stillunavailable,becausetheapproval procedure hadnotbeencompletedyet. county for two months from November to December 2013. However, in January and February 2014 YYB intervention the to YYB provide to decided UNICEF implemented, continuously was program the sure the county MCH hospital in the intervention county to monitor the process of implementation. To make We 2013. with October communicated frequentlyto September from stopped be to had provision YYB result, a As pricing. reasonable and quality good the ensure to implementation intervention of months program,a governmental purchasing YYBisregulated underacomplexandstrictprocess 12 afterevery is YYB As process. in procurementwas YYB procedureapprovalfor lengthy new a because department, health provincial the fromprovision further without YYB of stock–out a was there 2013, September In were stillbeused. givers who forgot to give YYB and children who disliked YYB. Meanwhile, banners, calendars and posters fore, we continued encouraging village doctors to explain again the benefit of YYB, with a focus on care time of the endline survey in August 2014 (23 months later survey) ( survey) later months (23 2014 August in survey endline the of time liked YYB more which was reflected by a decrease in the proportion of children who disliked YYB at the sessions said that the taste of the new YYB was much better than before and that children in their village lated knowledge, health education and complementary feeding skills. Most village doctors in the training re YYB their strengthen to county intervention the in doctors village the all retrain to sessions training three–day a out carried we 2014, June In onward. 2014 March from re–supplied was YYB and 格森)” After approval,governmental the YYB supplier changed from “Tian Tian Ai ( Also werequested themanufacturer toimprove thetasteofYYB. tinue usingmultiplechannelstopromote caregivers’ awareness oftheprogram. ers who forgot to give YYB continuously increased ( vey high adherence to YYB increased significantly (Table 3 ). However, because the proportion of caregiv- of high adherence got to 81.4% at the endline survey. Caregivers reported children’s acceptability to YYB of YYB [ receipt the guarantee could doctors village to then clinics, township to hospitals MCH to manufacturer from system distribution the proved also China in areas earthquake–affected in study previous A nel. chan multi–tiered distribution the programefficiently by the was caregivers county in deliveredYYB to high; the majority of caregivers ever received YYB and most children ever took YYB, which indicated that munity–based complementary food supplement program in China. The coverage of YYB distribution was out in one Chinese rural county, it provides an important insight into successfully implementing a com YYB program implementation experiences and challenges in China. Although our study was only carried on published data no was However,there [18]. 2014 since China in counties rural children’sin health implement the community–based complementary food supplement program (YYB program) to improve to yearly million) 75.24 (US$ RMB million ¥500 than moreCurrently, invests government Chinese the DISCUSSION ( 2 ure diarrhea)or were(Fig- children(cold they their sick when to YYB of sachets skipped temporarily givers care of 61.3% that found we 2013), January start, intervention after months (4 survey mini first the In side effects, suchasdiarrhea afterchildren tookYYB. liked the taste of YYB; children refused to take YYB; stopped giving YYB when their children got sick; and workers, village doctors, and caregivers of children. The following problems were identified: health children MCH dis with groups focus out carried we implementation, program of months two first the After nificantly increased to 22.0% ( identified (August 2013).However, caregivers who forgot to give YYB and children who disliked YYB sig In the midterm survey (11 months after intervention start, August 2013), no new program problems were banners, calendarsandposterstopromote theprogram. leaflets, used Moreover,we problems. those with dealing with them help to counties project YYB ghai 12.6% of caregivers sometimes forgot to give YYB ( Table 3 ). We coordinated with provincial health department to speed up the approval procedure.approval the up speed to department health provincialWe Tablewith 3). coordinated ), and 17.0% of mothers reported that their children disliked consuming YYB ( YYB consuming disliked children their that reported mothers of 17.0% and ), 13]. Generally, children’s adherence to YYB increased over time in our study, and the proportion ). Therefore, additional training was conducted in March 2013 to train the village doctors in Qin
P < 0.001, Figure 2)and 23.1% (P 496 Figure 2) and 12.3% of caregivers did not receive YYB Figure 2), we advised the local MCH hospital to con = 0.008, ). Also at the endline sur endline the at Also Figure3). www.jogh.org Figure 3), respectively . There 天添爰)” to “Fu Ge Sen ( • doi:10.7189/jogh.07.011101 Figure 3 Figure ). Also, Also, ). 福 ------www.jogh.org • doi:10.7189/jogh.07.011101 giving YYB due to their forgetfulness in our study. The possible reason was that more than half caregivers stopped temporarily who caregivers of number the in increase continual a was there that was challenge Another well. as future the in maintained be to has community the to MNP of flow uninterrupted fore, al process, which likely will have caused a significant decrease in adherence in the Mini 2 survey. There YYB was stocked out for twice due approvto a period of over six months of the complex governmental - of micronutrient powders was one of the key success factors to MNP program [ chain supply the maintaining that suggested also program MNP Bangladesh form Experience [36]. tion depends on caregivers to be motivated to offer sachets MNP for children properly and without interrup Real–life program implementation challenges can be hard to predict. It is known that effectiveness of MNP opment process tominimizecasesofrejection andtoincrease adherence[ tointervention devel the during characteristics sensory supplements’ the to given be must attention carefulTherefore, acceptable [ more be could sesame and peanut added which YYB improved that showed study previous A children. micronutrient powder. The soy flavor of YYB may explain that the taste of YYB was unacceptable by some [ proved. A study in Lao People’s Democratic Republic also report sprinkles had unpleasant smell and im taste be to needed YYB of taste the that implied which YYB, eat to refused even or disliked children that was adherence and acceptability poor for reasons main the of one that suggested study our addition, In days, andnearlyhalfofthosewhoskippedasachetYYBhadfeverinthepast15days[ 60 past the children’sin any illness of because MMNP of sachet a skipped they that reported mothers the of 19% around that Bangladesh in found also al. et Mirak diarrhea. or cold got children their when nificant lower adherence, which consist with most caregivers’ report that they would skip sachets of YYB We found in our study that children who got cough/fever or diarrhea during the past two weeks had sig lenges, adherence canincrease overtime. the endline survey (August 2014), which indicates that when active program monitoring to address chal that the high adherence increased from 2013) to more 49.4% (January in the mini1 survey than 80% in tion of the intervention decreased people’s motivations and adherence [ dura- longer a that found Studies [34]. sachets 60 all took children of 39% only but adherence, 75% of [ ducted in a controlled setting where field workers deliver and monitor the intervention on a regular basis implementation [28, implementation Program monitoring is critical for understanding program implementation and enabling more strategic given informationonYYBandthemainsources were villagedoctorsandYYBboxes. were YYB received ever who caregivers of 60% than More YYB. taking after children their provementin moretaking and Unfortunately,doses. more caregiversof 60% than perceivenot did im health positive the YYB manufacturer to improve the taste of YYB, which appeared to result in children liking YYB better increased over time as well, and the main reason for dislike was the taste of YYB. Therefore, we requested 100% [30– 100% [ 90% around to 32% from ranged MNP of provision daily to week) per more or sachets four of consumption as (defined adherence high [ reachedareadherence and acceptance of levels high Interventions like MNP that aim to reduce anemia prevalence in rural communities will only work when the training. after increased YYB to adherence high the that showed surveys endline and midterm the in Data YYB. ing sessions with village doctors (March 2013 and June 2014), which helped increase the adherence to low,was YYB to adherencetrain- overall additional that conducted showed we data monitoring When [ caregivers and doctors village educating and awareness raising for sessions training quality took Wedoctors. under village repeatedly on mainly relied education health as well as YYB of distribution well–trained village doctors played an important roles in successful program implementation. Therefore, that survey,indicated each which in doctors village from YYB on information received they reported also used leaflets, calendar, banners, posters and blackboards to promote the program, most caregivers tion on the leaflet into the pictures, which was easy for illiterate caregivers to understand. Although we survey,informa baseline text the at changed illiteracy we were caregivers main of 40% than more program implementation and allowed us to make real–time modifications. For example, when we found on information comprehensive and dynamic us provided which years, two for program the monitor 4 average an had months 2 childrenfor micronutrientto providingsprinkles powders study daily A 31]. ]. Different to MNP in other counties, Chinese YYB was a full fat soybean powder mixed with multiple ]. The highest adherence to MNP in those studies was observed in a study that was con was that study a in observed was studies those in MNP to adherence highest The 33]. 12].Furthermore, high mineral concentration in MNP sachetscouldalso influence the taste. ]. In our study, we carried out both quantitative and quantitative interviews to to interviews quantitative and quantitative both study,out our carried In we 29]. 6 ]. Studies of MNP in Bangladesh even found an adherence of 70– of adherence an found even Bangladesh in MNP of Studies ]. 497
6 , ]. Previous studies in other countries showed showed countries other in studies Previous 30]. Adherence toacomplementaryfoodsupplement 31, June 2017 •Vol. 7No. 1•011101 35]. Data in our study showed 37 ]. However, in our study 30]. 36]. ]. 29]. ------
VIEWPOINTSPaPErS PaPErSVIEWPOINTS June 2017 •Vol. 7No. 1•011101 Wu etal. time andcostthisisanobstacletosustainabilityoftheprogram. [ gram minder, which has been proved could improve the compliance of caregivers to a home fortification pro re message text as such YYB, of benefits the on knowledge caregivers’ reinforce in planned be should duration of the intervention decreased people’s motivations and adherence [ adherence people’sand decreased motivations intervention the of duration status of children is another key success factor to MNP program [ contributed to high acceptability among caregivers [ creased over time as well. It is documented that perceived benefits to children’s health was one of factors expressed they did not perceive improvements in their children’s after taking YYB, and the proportion in other settingsinChinaandelsewhere. in implementation program monitor to need supply.also programssupplement Future food of tinuity the food supplement, strengthening health education of village doctors and caregivers, and ensuring con of taste the improving include: supplements food complementary community–based the in adherence improve to Efforts YYB. to adherence children improve thus YYB, of awareness caregivers’ the improve tribute the supplements and educate caregivers; quality training conducted among village doctors could dis they as programs YYB community–based Chinese the of success the to critical were doctors Village implementation. strategic more direct and YYB, of acceptability and to adherenceaffecting issues reveal Monitoring YYB distribution and consumption promoted the YYB program implementation, which could CONCLUSIONS ting inChina. tings. When similar evaluations in different settings are conducted, this data can be compared to our set set other to study this from findings the generalizing when needed is caution and county Chinese one card to keep track of children under the program [ immunization an to similar card compliance a introduce could monitoring Future period. intervention representcompletely whole not the could realduring data the consumption week one the basis, weekly [6 sis ous week, which may have recall bias. Previous studies defined “high adherence to MNP” on weekly ba sachets during the previous week)” in this paper was based the caregivers’ recalled data during the previ YYB four least at consumed who (children YYB to adherence “high indicator main the First, limitations. several has also study evaluation Our implementation. program in time over trends the showed which cross-sectionalsurveys follow–up from four data collected we that is study evaluation this of strength A Strengths andlimitations interest. coi_disclosure.pdf (available upon request from the corresponding author), and declare no conflict of Competing interests: The authors completed the Unified Competing Interest form at www.icmje.org/ the finalmanuscript. QW,by ed CSY,RS, by revised and reviewed,WW, YFZ, MV, approved and read authors All LC. and WW, and MV participated in the explanation and discussion of the results. The manuscript was draft- QW,analysis. data performed QW data. QW,the HJH. coded and and WW,collected MX and XZD declaration: Authorship wasfundedbytheUnitedNationsChildren'sFunding: Thesurvey Fund(UNICEF). givers whoparticipatedinoursurvey. care- and mothers the Weall interviewers. to as indebted arehardwork their for University Qinghai from students all thank to want we and arrangements, logistic and coordination for Hospital Health Acknowledgments: ] that is “consumption of four sachets or more per week”; however, we could not get the data on on data the get not could however,we week”; per more or sachets four of “consumption is that ] ]. Moreover, currently no governmental subsidies are in place to compensate village doctors’ doctors’ village compensate to place in are subsidies governmental no Moreover,currently 38].
The authors wish to thank all colleagues from the Huzhu Maternal and Child Child and Maternal Huzhu the from colleagues all thank to wish authors The The study was initiated, conceptualized, and supervised by RS, CSY, YFZ RS, by supervised and conceptualized, initiated, was study The 498 30]. Furthermore, this evaluation study took place in 36] and a visible and convincing change in nutrition 37]. Studies also indicated that a longer www.jogh.org 31, • doi:10.7189/jogh.07.011101 ]. Additional efforts Additional 35]. ------www.jogh.org rEFEr ENCES • doi:10.7189/jogh.07.011101 25 24 23 22 21 20 19 18 17 16 4 15 10 9 3 14 6 5 2 1 8 7 11 13 12 shtml. Accessed:18August2012. The story of “Ying Yang Bao”. [website in Chinese]. Available: http://www.ce.cn/cysc/sp/info/201510/22/t20151022_6779753. S Kounnavong hood anaemia. Sprinkles-Evaluation-Report-2005.pdf. Accessed:10September2016. Available:http://www.wvi.org/sites/default/files/Mongolia-2005. anemia. and rickets address to program nutrition ed Lao People’s Democratic Republic: a randomised trial. home fortification with multiple micronutrient powder on haemoglobin concentration of young children in a rural area, Mongolia World Vision. Effectiveness of home-based fortification of complementary foods with Sprinkles in an integrat egy forcombatingchildhoodiron deficiencyanemia.BRAC;Dhaka: 2006. controlled trial. der reduces anemia, iron deficiency, and vitamin A deficiency in young children in Western Kenya: a cluster-randomized Zlotkin S0140-6736(13)60996-4 De-Regil Suchdev 10-129 ternal ternal and child nutrition: what can be done and at what cost? Lancet. 2013;382:452-77. Bhutta ZA, Das JK, Rizvi A, Gaffey MF, Walker N, Horton S, et al. Evidence-based interventions for improvement of ma ofHealth;2012. Beijing: Ministry years. 0-6 aged children of status nutritional on report National China. of Republic People China. Health of Ministry Karim F and children 6–23monthsofage.Geneva:World HealthOrganization; 2011. World Health Organization. Use of multiple micronutrient powders for home fortification of foods consumed by infants Medline:21901727 powders for health and nutrition in children under two years of age. of years two under children in nutrition and health for powders a controlled interventional study. and dietary counseling on anemia and stunting in children aged 6-23 months in poor areas in Qinghai Province, China: Zhang S Qinghai StatisticsYear Book.2010.Beijing;ChinaStatistics Press: 2011. Thousand Oaks,CA:Sage;2007 . Chinese children. Creswell Zhang Y 2012. News Agency. Beijing. 2011. Available: http://www.gov.cn/jrzg/2011-11/17/content_1996437.htm. Accessed: 18 August Gu L 9241a3a3553e19dec77421.shtml. Accessed:8October2015. Available:Chinese]. in http://www.nhfpc.gov.cn/fys/s3585/201411/254523446fprotocol.[website 2014: in areas rural People’sthe of Commission Planning Family and Health National childrenImprovein for China. nutrition of Republic 774177bd30e0bcc516c67e.shtml. Accessed:8October2015. Available:Chinese]. in protocol.[website 2013: in areashttp://www.nhfpc.gov.cn/fys/s3585/201311/25bcc2db06 rural People’sthe of Commission Planning Family and Health National childrenImprovein for China. nutrition of Republic Liu S ofHealththePeople’ssupplements. Beijing:Ministry RepublicofChina; 2008. food complementary fortified for standard Chinese China. People’sof the Republic of Administration Standardization Chen doi:10.1016/S0895-3988(09)60045-3 mia of infants and young children in poor rural of Gansu. of rural poor in children young and infants of mia Rah infant feedingpracticesinrural China. Wang current evidence. Korenromp Suppl 3: Huo J supplement (Ying Yang Bao) and impacts on anaemia and feeding practices in Shanxi, China. Maternal, Neonatal and Child Health Weeks in Nigeria: process evaluation of feasibility and use. and feasibility of evaluation process Nigeria: Weeksin Health Child and Neonatal Maternal, 2015; China. of regions rural poor in months 23 to 6 aged children young and infants among anemia of Sun J 2016; JH, dePee S . Qinghai will input 2.5 million RMB every year to implement Integrate Early Child Development program. Xinhua . Eliminateanemiaamonginfantsandyoungchildren inChina.ChinaYouth Daily;Beijing: 2011. , Dai Y , Sun J 36: 19: CM, Wang YY,Chen , Choudhury N , C Schauer SH, , 96-
JW, Plano Clark PS, Ruth 405- 1882- Wu LM, Suchdev LM, Wu
, Zhang S EL, Adeosun 111. , Fang Z Q Q 14. T Sunahara , 92. , Wang W PLoS Med.2005;2: , van Velthoven , Kraemer K Am JClinNutr. 2012; CM, WangCM, Medline:21929638 Medline:26612420 YY, Chang J Nutr. 2012; Biomed Environ Sci.2010; LJ, Woodruff Medline:26370070 , Chang S , Huang J , Pasha A A Christofides , PS, Vist PS,
O , van Velthoven VL. The nature of mixed methods research. Designing and conducting mixed methods research. , Adegoke , Mascie-Taylor, , Steiger G FZ, Jia FZ, M SY. Effect of in-home fortification of complementary feeding on intellectual development of , Zhao L , Yang Z 142: GE, WalleserGE, S MH, Chen L BA, Mbakaya C , Hasan K
BMJ Open. 2016;6: e1. 191S- F , Wang, doi:10.1111/j.1740-8709.2011.00353.x doi:10.1177/0379572115616001 Medline:15696200 , Akerele , Bloem , Fu P 95: W Sharieff , , Huo J doi:10.1017/S1368980015002499 J Med Internet Res.2012; J MedInternet MH, Chang , Hyder 1223- 6S. , et al. Ef M Hashizume CG, , Car J fect of home-based complementary food fortification on prevalence KA. Effects of nutrient fortified complementary food supplements on ane on supplements food Effects complementary KA. fortified nutrient of Medline:22131547 23: MW, Spiegel P , et al. Implementation of a programme food to market a complementary A , Pena-Rosas , , Mandava U 30. , Anger 83- SZ, Zlotkin , Tondeur, , Rudan I Medline:22492366 499 91. e011234
SY, Han Nutr J. 2011 Medline:20514982 C , Ohajinwa doi:10.1371/journal.pmed.0020001
Biomed Environ Sci. 2009; Sci. Environ Biomed , et al. Smartphone versus pen-and-paper data collection of , et al. Program experience with micronutrient powders and JP. micronutrient multiple with foods of fortification Home , Flores-Ayala R MC, Hyder MC, HJ, et al. Effectiveness of complementary food supplement . SH. Reaching sprinkles to millions: social marketing strat Medline:27799239 J Okumura , 14: doi:10.3945/jn.111.140004 ; 10 e119. C :
129 Cochrane Database Syst Rev.Syst Database Cochrane 2011;9: , et al. Micronutrient powder distribution through doi:10.3945/ajcn.111.030072 SM. Micronutrient sprinkles to control child control to sprinkles Micronutrient SM. . Medline:22111770 Medline:22989894 K Moji , , et al. Selling Sprinkles micronutrient pow- doi:10.1016/S0895-3988(10)60036-0 Adherence toacomplementaryfoodsupplement
doi:10.1136/bmjopen-2016-011234 , et al. Effect of daily versus weekly weekly versus daily of Effect al. et , June 2017 •Vol. 7No. 1•011101 Medline:23746776 22: 194-
Matern Matern Child Nutr. 2011;7
doi:10.1186/1475-2891- 200. doi:10.2196/jmir.2183
Medline:19725461 Public Health Nutr Health Public
Food Nutr Bull Nutr Food
doi:10.1016/ CD008959. - - - - - . .
VIEWPOINTSPaPErS PaPErSVIEWPOINTS June 2017 •Vol. 7No. 1•011101 Wu etal. rEFEr ENCES 38 37 26 29 28 27 36 35 31 33 34 30 32 fortification program against child anemia in rural western China: a cluster-randomized controlled trial. controlled cluster-randomized a China: western rural in anemia child against program fortification Health. 2016; Zhou H 2014; Afsana K 016-2978-0 aged 6 to 23 months: a systematic review.systematic a months: 23 to 6 aged SAS Institute Inc. The ANOVA Procedure. Available: https://support.sas.com/documentation/cdl/en/statug/63033/HTML/ Suchdev World Health Organization. Managing programmes to improve child health. Geneva: World Health Organization; 2009. Washington AccountabilityOffice: DC;USGovernment 1989. Accountability US Office.Government Content analysis: a methodology for structuring and analyzing written material. default/viewer.htm#anova_toc.htm. Accessed:24April2017. de Barrosde Soc SciMed.1994; R Galloway Bull. 2010; reduces anemia among children 6 to 36 months of age in the Kyrgyz Republic: a cluster-randomizedtrial. a Republic: Kyrgyz the in age of months 36 to 6 children among anemia reduces bidity: a randomized, double-blind, placebo-controlled, noninferiority safety trial. safety noninferiority placebo-controlled, double-blind, randomized, a bidity: children with moderate-to-severe malnutrition increases hemoglobin concentrations but not the risk of infectious mor Ip H s12889-015-1752-z dren in rural Bangladesh: a cross-sectional study.cross-sectional a Bangladesh: rural in dren Hyder Medline:21715512 E Lundeen 2007; micronutrient Sprinkles on hemoglobin and iron status among young children in rural Bangladesh. rural in children young among status iron and hemoglobin on Sprinkles micronutrient Angdembe doi:10.1177/15648265100312S209 Kenya. western rural in powders micronutrient Sprinkles tration of micronutrient Sprinkles: a new public health approach to anaemia control. Lemaire M Medline:17895911 , Hyder , 23: 28: F Haseen SM, , Sun S , Haque , 377- 156- L Ruth PS, SF,Cardoso , Islam T Schueth , 31: MR, Choudhury N J McGuire , SM, Haseen SM, F 106: 84. 64. 446- , Luo R Medline:25164447 Medline:24683674 MR, Sobhan MR, S QS, Shen H 1256 60. 39: doi:10.3945/ajcn.110.009316 , ObureA , doi:10.1038/sj.ejcn.1602917 M Rahman ,
, Sylvia S MA. Adherence to and acceptability of home fortification with vitamins and minerals in children in Adherence minerals MA. and vitamins with fortification home of acceptability and to , Toktobaev, N Medline:20973465 381- . Determinants of compliance with iron supplementation: supplies, side effects, or psychology? or effects, side supplies, supplementation: iron with compliance of Determinants . - 62. , Rahman , M 90. Medline:27077354 , Khan , , Haque Medline:7939855 , Were , Shahin , Yue , Tondeur, A V S Zlotkin , MA, Parveen M , Shi Y doi:10.1177/156482650702800204 C Ochieng , , Zlotkin , MR, Ahmed T SA. BRAC’sSA. Bangladesh. in MNP scaling-up in experience
BMC Public Health. 2016; Health. Public BMC doi:10.1177/156482651003100307 MC, Zlotkin MC, , et al. Impact of text message reminders on caregivers’ adherence to a home , Hyder , SH. ImprovedSH. adminis adherencecureflexible anaemia with and rates doi:10.2105/AJPH.2016.303140
doi:10.1016/0277-9536(94)90135-X L Ogange , BMC Public Health. 2015 Health. Public BMC 500 , Abedin F . Adherence to multiple micronutrient powder among young chil SM. Daily use of Sprinkles micronutrient powder for 2 months months 2 for powder micronutrient Sprinkles of use Daily SM. SH. Effect of daily versus once-weekly home fortification with with fortification home once-weekly versus daily of Effect SH.
, et al. Monitoring the marketing, distribution, and use of of use and distribution, marketing, the Monitoring al. et , Food Nutr Bull. 2010 Bull. Nutr Food , et al. Iron-containing micronutrient powder provided to 16: 299. ; Medline:27056182 15 : 440 www.jogh.org
. Am J Clin Nutr.2011; Clin J Am ;
Medline:25925874 Eur J Clin Nutr. 2009; 31 : S168 • doi:10.7189/jogh.07.011101 - 78 . doi:10.1186/s12889- Medline:20715601 Asia Pac J Clin Nutr Clin J Pac Asia
Food Nutr Bull Nutr Food
doi:10.1186/ Am J Public Public J Am
94: 63: Food Nutr Nutr Food 585- 165- 93. 72. - - - . .
www.jogh.org investigation andhealthcarechallenges Diabetes inShenzhen,China:epidemiological Jianfeng Zou Lizhen Yang Haitao Li * 2 1 Xinfeng Yan [email protected] China Shenzhen Nanhai a Shenzhen University School ofMedicine Haitao LiMS,PhD Correspondence to: Jointfirstauthorship School ofMedicine,ShenzhenUniversity, Longhua DistrictCenterforChronicDisease Shenzhen, China Prevention andControl,Shenzhen,China
• doi:10.7189/jogh.07.011102 ve 3688
2 , XiaotingDeng 1,* 1
, ShaojuanZhao 1
, YiLuo , HuiXia
1 , SijingCao 1,* , 1 , 1 , 1 that China is home to the largest diabetic population in the world. Statistics middle–income countries disproportionately affected [ worldwide [2 the past several decades. The current estimate of diabetes prevalence is 9% during prevalence diabetes of increase rapid a show (WHO) Organization [ diseases non–cardiovascular and lar Diabetes is associated with increased mortality from a range of cardiovascu 0.7% in 1980 [ sharply,increased prevalence diabetes decades, several from past the Over In China, diabetes has also emerged as an important public health problem. developing countries[5 for especially worldwide, concern health public major a represents betes vices andbettermanagementofdiabetes. betes. Primary care may be crucial an improved access to medical ser dia of detection early as well as prevention, secondary and primary diagnosed. Our findings highlight the need of public health efforts for care system. Approximately half of the subjects with diabetes are un blood glucose imposes a public health threat and burden to the health fasting impaired of prevalence high average, Chinese the of that half Conclusions corded foronly11.1%ofthepatients. macological treatments. Primary care management of diabetes was re non–phar undergoingwere patients diabetic of one–third than Less pharmacologically.treated being not were patients diabetic of 54.3% and (51.9%) poor was diabetes of awareness The 1.15–3.22). (CI) with diabetes only (odds ratio (OR) ratio (odds only diabetes with Results of 1676participantscompletedthesurvey. Face–to–face interview surveys were conducted to collect data. A total participants. the select to methods sampling random cluster tistage Methods A cross–sectional study was conducted. We employed mul zhen, China. Shen in management care primary and non–pharmacological ical, the prevalence and awareness of diabetes, as well as its pharmacolog investigate to aimed study This policy–making. and diabetes of trol tes in a specific population will potentially benefit prevention and con Background Understanding epidemiological characteristics of diabe with age ( age with increased glucose blood fasting impaired and diabetes both of rates prevalence The 6.0%. was glucose blood fasting impaired of alence We found that the prevalence of diabetes was 4.8%. The prev P ]. In 2014, diabetes caused 1.5 million deaths, with low– and 501 < 6 Although diabetes prevalence in Shenzhen is about a a about is Shenzhen in prevalence diabetes Although ], to 2.7% in 2002 [ 0.001), whereas hypertension was strongly associated associated strongly was hypertension whereas 0.001),
]. 7 ], to 11.6% in 2010 [ 1 = ]. Statistics from the World Health the from Statistics ]. 1.93, 95% confidence interval interval confidence 95% 1.93, June 2017 •Vol. 7No. 1•011102 3 , 4 global journal of ]. Therefore, dia 8 ]. This implies health ------
VIEWPOINTSPaPErS PaPErSVIEWPOINTS June 2017 •Vol. 7No. 1•011102 Yan etal. deff = method was adopted for participant selection within each household. The overall response rate was 89%. Kish A members. household current of list the obtain to contacted was household Each district. by fied neighborhoodstrati each in distributed wereevenly Households district. each of size population the to random sampling method. The total number of households selected from each district was proportional systematic a employing sampled were households and listed were neighborhood each in dwellings All then randomly drawn from each randomly selected district. A total of 20 clusters were randomly selected. approach. sampling random simple a were neighborhoods using Ten selected randomly first were tricts dis Twoten design. the sampling of random cluster multistage a using participants the sampled study dence interval. The final sample size was 2000, taking into consideration a 10% non–response rate. This macological and primary caremacological andprimary managementinShenzhen,China. aimed to investigate the prevalence and awareness of diabetes, as well as its pharmacological, non–phar study current The policy–making. and diabetes of control and prevention the benefit potentially will veloped area in China. Understanding epidemiological characteristics of diabetes in a specific population recorded. On an appointed date after the interview, blood sample was obtained from participants. Twelve– be would measurements three all of mean the and performed was measurement additional an Hg, mm 5 than larger was measurements two the between difference the If measurements. two of means the as pressureblood werepressurediastolic recorded(DBP) blood and Systolic were(SBP) performed. ments management, blood pressure was measured using calibrated mercury sphygmomanometer. Two measure hypertension for guidelines national the by recommended protocol the stadiometer.to Accordingtable nearest 0.1 kg using a digital scale, and height to the nearest 0.1 cm in the standing position with a por the to measured was weight Body obtained. were interview,measurements the anthropometric During ment?” cological approaches are taken for management of diabetes?”, and “Are you under primary care manage you have diabetes diagnosed by a health professional?”, “What kind of pharmacological or non–pharma were“Do participants month?”, The past income. the household in monthly smoke you “Do asked, also registration, status, marital occupation, level, education age, their about asked were participants The and confidentiality of the survey, and informed consent was obtained before the survey was commenced. The survey was conducted by specially trained interviewers. The participants were assured of anonymity interview survey was adopted for the collection of socio–demographic factors and clinical measurements. Face–to–face parameters. biochemical of assessment for sample blood subsequent a and measurements clinical characteristics, socio–demographic on questionnaire a included which [13], surveillance factor Data were collected using the World Health Organization (WHO) STEPS approach to chronic disease risk Data collectionprocedure regularthe of members Forces,Chinese were formula the Using excluded. fore the survey was performed and aged 18–70 years. Those living in institutions like nursing homes, and April and May 2015. The study included subjects living in Shenzhen ≥6months in the past one year be- between conducted survey population household community–based a was study cross–sectional This Study population Control EthicsCommittee. This study was approved by the Shenzhen Longhua District Center for Chronic Diseases Prevention and Ethics METHODS area of 1996.8 km total a within living migrants being population its of 70% about with city migrant a is Shenzhen China. southern of region Delta River Pearl the in situates status, administrative sub–provincial holding Zone economy,on pendent [ etc. regions living and culture de are factors risk its well as diabetes, of management awareness, prevalence, that shown have Studies which are crucialtodecrease itsrelated complicationsanditsfinancialburden [ [ in 2013 showed that approximately one–fourth of worldwide diabetes–related deaths occurred in China 3 ]. However, there have been no obvious improvements in diabetes awareness, treatment and control, and treatment awareness, diabetes in improvements obvious no been However,have ]. there 1.5 and p =
0.05, we calculated the sample size of 1752 for a 95% confidence level and 2.5% confi 2 . Shenzhen is an important economic powerhouse, and represents one of the most de 502 10– ]. Shenzhen, China’sShenzhen, 12]. Economic Special first www.jogh.org n = deff 9 • doi:10.7189/jogh.07.011102 ]. ´ u 2
´ p(1 – p)/d 2 , where , ------www.jogh.org • doi:10.7189/jogh.07.011102 and/or systolicbloodpressure (SBP)≥140mmHgand/ordiastolicbloodpressure (DBP)≥90mm Hg. treatment, antihypertensive under currently and condition physician–diagnosed self–reported as defined daily.or occasionally smoke who individuals included and self–reportedwas was Hypertension smoking were performedbyusingtheSPSS19.0software (SPSSInc.,Chicago,IL,USA). were presented as prevalence rates. A rates. prevalence as presented were diabetes of control and Awareness,management model. the in inclusion variable for 0.10 of threshold a with elimination, backward using conducted were fittings Model diabetes. and glucose blood fasting (CI) to estimate the strength of associations between socio–demographic and lifestyle factors and impaired interval confidence 95% and (ORs) ratios odds of calculation the for constructed were models gression Twore characteristics. logistic lifestyle multivariate and socio–demographic different with participants individual’s body mass index (BMI) of 24.0–27.9 kg/m an as defined were obesity and Overweight hypertension. and status smoking self–reported index, mass hold income level (RMB10 groups according to the monthly household poverty line (RMB 5000, US$ 725) and mean monthly house Controlled diabeteswasdefinedasFBG<7.0mmol/L. monitoring bloodglucoseregularly. and/or exercise, in engaging and/or diet, changing as defined was management Non–pharmacological larly orinsulininjectionfordiabetes. regu diabetes for use participant’s medication a of as report defined was management Pharmacological professionals, and/ortheuseofinsulinormedicationfordiabetes. Awareness of diabetes referred to participants’ self–report of any previous diagnosed condition by health blood glucosewasdefinedas5.6mmol/L≤FBG<7.0mmol/L. currently. diabetes fasting Impairedfor treatment drug reported participants’ and/or condition, nosed Diabetes was defined as fasting blood glucose (FBG) ≥7.0 mmol/L, and/or self–reported physician–diag Key definitions ing calibratedbloodglucosemetersandreagent trips. hour fasting blood glucose levels were assessed according to WHO standardized fingertip prick tests, us oe Ma SP a 198 m H, hl ma DP a 7.3 m g Te rvlne of prevalence The Hg. mm 77.63 kg/m 23.50 was BMI Mean was 17.6%. was hypertension DBP mean while Hg, mm 119.81 was SBP Mean come. in household monthly their know not did or question the answer to rejected participants of 33.1% as mary school or below. More than one–third of the participants were in the middle–income group, where (88.1%). Around one–third of the participants had middle-school education, and just over 10% had pri than half of participants were women. The majority of participants were married or living with a partner More migrants. and years, 44 to 18 between aged were participants the Approximatelyof three–fourths Characteristics ofparticipants r characteristics. lifestyle and socio–demographic to ing means (SD). Prevalence estimates of impaired fasting blood glucose and diabetes were computed accord or percentages as presented were participants of factors lifestyle and characteristics Socio–demographic Statistical analysis [ registration Hukou official their changing without location new a to moved who individuals as defined were Migrants registration. the to cording ac locals, the and migrants the into classified were participants The income. household monthly pation, occu level, education registration, status, gender,marital age, included information Socio–demographic Descriptive variables mmol/L (Table 1). weight or obese. Approximately one–fifth of the participants were current smokers. Mean FBG was 4.81 ≤18.4 kg/m ≤18.4 ESULTS 2 and 18.5–23.9 kg/m 18.5–23.9 and
000, US$ 1450) in Shenzhen in 2011 [ 2 indicated underweight and normal weight, respectively [ respectively weight, normal and underweight indicated P value <0.05 was considered statistically significant. All analyses analyses All significant. statistically considered was <0.05 value 503 14 ]. Monthly household income was categorized into three into categorized was income household Monthly ]. 2 2 . Around two–fifths of the participants were over were participants Around the two–fifths. of and ≥28.0 kg/m χ 2 –tests were performed for comparison between between comparison for wereperformed –tests 15]. Lifestyle factors included the body 2 , respectively, whereas the BMI of June 2017 •Vol. 7No. 1•011102 Diabetes inShenzhen,China ]. Current Current 16]. ------
VIEWPOINTSPaPErS PaPErSVIEWPOINTS June 2017 •Vol. 7No. 1•011102 Yan etal. Table 2. glucose anddiabetes Table 1. stolic bloodpressure, BMI–bodymassindex SD – standard deviation, SBP – systolic blood pressure, DBP – dia 60–70 50– 40– 30– 18– a ≥ 45–59 18–44 Age (years),mean(SD): c Never inunion Marital status: Migrants Locals Registration: Female Male Gender: Fasting bloodglucose No Yes Current smoking: Overweight/obese weight Underweight/normal BMI, mean(SD) DBP, mean(SD) SBP, mean(SD) No Yes Hypertension: Do notknow Rejected High Middle Low Household income: 12.8 Not working Other workers Clerical 214 Professional, technical and managerial Sales andservices Manual workers Occupation: 3–year collegeandabove 2.5 High schoolandequivalent 88.1 Middle school schoolandbelow Primary 42 1476 Education: Widowed, divorced andseparated Married orlivingwithpartner ge haracterIstIcs 60
group Characteristics ofparticipants Age–specific prevalence ofimpaired fastingblood I mpa I red
F ast 17/103 (16.5) 30/232 (12.9) I 27/394 (6.9) 20/617 (3.2) ng 7/329 (2.1)
(mmol/L) blood
glucose , n o
./ n (%) 1970.3 39.26(11.13) 1179 1675 2672.6 1216 664.81(1.55) 78.2 1676 1310 23.50(3.71) 77.63(10.79) 1665 82.4 1676 1676 1381 0 6.1 23.4 103 393 4 8.9 149 26.4 52.8 442 47.2 885 791 6 21.8 40.6 366 59.4 680 996 17.6 22.8 295 10.3 20.6 382 36.2 172 10.1 346 606 27.2 170 15.0 10.5 456 251 19.1 176 15.1 320 18.3 253 31.5 36.0 306 13.8 528 603 231 n o u . d 11/103 (10.7) 29/232 (12.5) Iabetes 30/394 (7.6) 10/617 (1.6) 119.81 (15.65) 1/329 (0.3) nweIghted , n o ./ n (%) % - cation level of 3–year college and above (2.3%, edu an with those among lowest was and level, education with vorced or separated (7.1%, The prevalence of diabetes was the highest in those widowed, di blood glucose was not significant (OR 95% CI 1.15, 3.22), whereas the association for impaired fasting adjusting for socio–demographic and lifestyle factors (OR = after even diabetes for significant was hypertension with ciation asso The made. were adjustments similar after non–significant were glucose blood fasting impaired and diabetes of prevalence with factors socio–demographic other of associations ever,the after adjusting for socio–demographic and lifestyle factors. How significant statistically still were glucose blood fasting impaired and diabetes of prevalence and age between relationships The P 4.9%, vs (7.6% counterparts their with compared when alent participants who were either overweight or obese was more prev P 5.0%, vs (10.8% counterparts their of that than higher times ing blood glucose among participants with hypertension was two ent education levels (P for impaired fasting blood glucose across participants with differ 2 and was highest among those aged ≥60 (16.5%, prevalence of impaired fasting blood glucose increased with age, The prevalence of impaired fasting blood glucose was 6.0%. The the participantswithindifferent BMIgroups (Table). 3 ( counterparts their than hypertension with participants in (7.7%, working not those among highest being differentoccupation, with pants was a significant difference in diabetes prevalence across partici- with age up to 50–59 age group (12.5%, group age 50–59 to up age with The overall prevalence of diabetes was 4.8%. The prevalence rose and diabetes Prevalence ofimpaired fasting bloodglucose Significant relationships between BMI and diabetes (OR diabetes and BMI between relationships Significant 95% CI 0.91, 2.43) and impaired fasting blood glucose (OR 4 founding effects of socio–demographic and lifestyle factors (Table con for adjusting after diminished were 1.93) 0.81, CI 95% their condition. Among all participants with diabetes, 45.7% were Among 81 participants with diabetes, 42 (51.9%) were aware of Diabetes awareness, managementandcontrol ment (Table 5). one–tenth of the participants were careunder primary manage respectively.over 45.2%, Only and 38.1% 64.3%, condition: their aware were who participants in common more two-fold almost were approaches regularly.management Non-medical 19.8% engaged in exercise and 23.5% monitored blood glucose while participants, of 33.3% by adopted were changes Dietary diabetes. diagnosed previously with participants among 88.1% was percentage this while insulin, or medications with treated P = < ). Like the prevalence of diabetes, a similar trend was observed ). < 504 0.021) (Table 3). in glucose blood fasting impaired of prevalence The 0.001). 0.001). There was a noteworthy 1.8–fold difference between
P = 0.021). Diabetes was 3 times more frequent more times 3 was Diabetes 0.021). = 0.011). The prevalence of impaired fast- P = 0.040). The prevalence decreased www.jogh.org = 1.48, 95% CI 0.91, 2.40). • doi:10.7189/jogh.07.011102 P < 0.001) ( 0.001) P P < = 0.001) ( 0.030). There Table2 = = Table 1.49, 1.49, 1.93, 1.93, 1.25, 1.25, ). ). ------www.jogh.org • doi:10.7189/jogh.07.011102 * BMI –bodymassindex Table 3. zhen, China.Ahighresponse ratewasachieved.Rigorous randomsampling approach wasadoptedand This is a representative study with 1676 participants to investigate the epidemiology of diabetes in Shen of theparticipants. ing non–pharmacological treatments. Primary care management of diabetes was reported by only one–tenth weretients treatedbeing not pharmacologically. one–thirdthan Less werepatients diabetic of undertak pa diabetic of half than more and poor was diabetes of awareness The diabetes. with only associated ly of both diabetes and impaired fasting blood glucose increased with age, whereas hypertension was strong prevalencerates The 6.0%. was glucose blood impairedfasting prevalenceof The 4.8%. was diabetes of Our study on a representative sample of 1676 participants in Shenzhen, China, found that the prevalence DISCUSSION 18–44 Age group: Gender: ≥ 45–59 Male Registration: Female Locals Marital status: Migrants Other workers Manual workers Occupation: schoolandbelow Primary Education: Married orlivingwithpartner Never inunion Household income: 9(4.2) Not working Clerical Professional, technicalandmanagerial Sales andservices 3–year collegeandabove High schoolandequivalent Middle school Widowed, divorced andseparated Do notknow Low BMI: No Hypertension: Rejected High Middle Underweight/normal weight Underweight/normal Yes Current smoking: Overweight/obese Yes No All participants c χ haracterIstIcs 60 2 –test wasusedforcomparisons. Prevalence ofimpaired fastingbloodglucoseanddiabetesbysocio–demographiclifestylecharacteristics 17 (16.5) 41 (10.5) 32 (10.8) 101 (6.0) No. (%) 43 (3.6) 45 (5.1) 56 (7.1) 73 (6.0) 26 (5.9) 36 (7.9) 11 (4.4) 22 (6.9) 14 (5.5) 29 (5.5) 41 (6.8) 21 (9.1) 90 (6.1) 69 (5.0) 27 (7.1) 14 (8.1) 17 (4.9) 32 (5.3) 11 (6.5) 52 (7.6) 49 (4.9) 74 (5.6) 27 (7.4) 7 (4.0) 8 (2.6) 4 (9.5) 5 (3.4) I mpa 505 Ired
FastIng
blood
glucose <0.001 <0.001 0.087 0.927 0.245 0.209 0.011 0.478 0.021 0.219 P* – June 2017 •Vol. 7No. 1•011102 11 (10.7) 42 (10.7) 32 (10.8) No. (%) 28 (2.4) 81 (4.8) 42 (4.7) 39 (4.9) 58 (4.8) 22 (5.0) 35 (7.7) 10 (4.0) 13 (4.1) 10 (4.0) 24 (4.5) 30 (5.0) 18 (7.8) 76 (5.1) 49 (3.5) 24 (6.3) 11 (3.2) 30 (5.0) 10 (5.9) 45 (6.6) 36 (3.6) 64 (4.9) 17 (4.6) 3 (1.7) 7 (2.3) 3 (7.1) 1 (0.7) 9 (4.2) 6 (3.5) Diabetes inShenzhen,China d Iabetes <0.001 <0.001 0.860 0.862 0.040 0.021 0.030 0.297 0.005 0.849 P* – - - - -
VIEWPOINTSPaPErS PaPErSVIEWPOINTS June 2017 •Vol. 7No. 1•011102 Yan etal. and smokingstatus. † *Model adjustedforage,gender, maritalstatus,registration, education,occupationandmonthlyhouseholdincome. BMI–bodymassindex OR –oddsratio,CIconfidenceinterval, DMs – diabetes mellitus patients, PC – primary careDMs –diabetesmellituspatients,PC –primary Table 5. Table 4. Yes Hypertension: No Overweight/obese weight Underweight/normal BMI: Yes Current smoking: No Do notknow Rejected High Middle Low Household income: Not working Other workers Clerical Professional, technicalandmanagerial Sales andservices Manual workers Occupation: 3–year collegeandabove High schoolandequivalent Middle school Primary schoolandbelow Primary Education: ioe,dvre rsprtd12 04–.0 .6(.430)39 05–97)3.62(0.48–27.25) 3.98(0.53–29.77) 1.16(0.44–3.06) 1.22(0.46–3.20) Widowed, divorced orseparated Married orlivingwithpartner Never inunion Marital status: Migrants Locals Registration: Female Male Gender: ≥ 45 – 18 – Age group: Control Do notknow Under PCmanagement Blood glucosemonitoring Exercise Diet Non–pharmacological: Insulin injection Medications Pharmacological Management: Awareness v c Model adjusted for age, gender, marital status, registration, education, occupation, monthly household income, hypertension, BMI arIables haracterIstIcs 60 59 44 Awareness, managementandcontrol ofdiabetes Multivariate analysisonfactorsassociatedwithprevalence ofimpaired fastingbloodglucoseanddiabetes
a mong .2(.21.3 .7(.489)41 23–.0 3.54(2.01–6.25) 4.15(2.39–7.20) 4.17(1.94–8.96) 4.72 (2.22–10.03) .1(.119)09 05–.3 .5(.627)1.20(0.54–2.66) 1.25(0.56–2.76) 0.98(0.50–1.93) 1.01 (0.51–1.97) .3(.930)13 06–.3 .8(.827)0.93(0.30–2.88) 0.88(0.28–2.70) 1.38(0.61–3.13) 1.33 (0.59–3.00) .5(.624)10 04–.4 .5(.637)1.51(0.58–3.93) 1.45(0.56–3.75) 1.07(0.47–2.44) 1.25 (0.49–3.19) 1.05 (0.46–2.40) 1.22(0.48–3.09) 1.06(0.45–2.54) 1.05 (0.44,2.49) .9(.716)08 04–.7 .6(.930)1.44(0.68–3.04) 1.46(0.69–3.08) 0.88(0.47–1.67) 0.89 (0.47–1.68) .0(.718)07 02–.4 .5(.123)0.88(0.32–2.44) 0.85(0.31–2.36) 0.71(0.27–1.84) 0.70 (0.27–1.81) .7(.830)10 03–.7 .0(.319)0.49(0.12–1.95) 0.50(0.13–1.99) 1.05(0.37–2.97) 1.07 (0.38–3.02) .4(.225)10 04–.2 .8(.224)0.91(0.33–2.49) 0.88(0.32–2.40) 1.06(0.43–2.62) 1.04 (0.42–2.54) .9(.031)13 06–.1 .5(.724)0.97(0.38–2.51) 0.95(0.37–2.46) 1.39(0.61–3.21) 1.39 (0.60–3.18) .2(.747)19 08–.7 .6(.434)1.29(0.52–3.25) 1.36(0.54–3.41) 1.96(0.85–4.57) 2.02 (0.87–4.70) .5(.553)21 09–.0 .0(.228)1.02(0.40–2.64) 1.10(0.42–2.84) 2.19(0.92–5.20) 2.25 (0.95–5.36) .1(.055)20 07–.1 .3(.434)1.13(0.40–3.22) 1.23(0.44–3.48) 2.01(0.76–5.31) 2.11 (0.80–5.55) .7(.987)19 04–.1 .1(.73.6 2.97(0.25–35.91) 3.21(0.27–38.46) 1.97(0.46–8.41) 2.07 (0.49–8.74) .8(.816)09(.216)09(.116)0.92(0.52,1.65) 0.91(0.51,1.62) 0.92(0.52–1.65) 0.98 (0.58–1.65) .2(.225)15 09–.5 .2(.822)1.30(0.72–2.37) 1.32(0.78–2.24) 1.56(0.93–2.65) 1.62 (1.02–2.57) .6(.845)25 15–.7 .8(.683)2.78(1.14–6.78) 3.48(1.46–8.33) 2.51(1.51–4.17) 2.76 (1.68–4.53) R(5 I*O 9%C) R(5 I*OR(95%CI)† OR(95%CI)* OR(95%CI)† OR (95%CI)* 29.6 (24/81) 13.6 (11/81) 23.5 (19/81) 19.8 (16/81) 33.3 (27/81) 39.5 (32/81) 45.7 (37/81) 51.9 (42/81)
11.1 (9/81) all 6.2 (5/81) (n
pat I o 506 mpa 1 – – – – – – 1 1 1 1 1 1 ., %) Ients Ired
wIth
FastIng dm
blood 1.48 (0.91–2.40) 1.25 (0.81–1.93) 0.99 (0.57–1.73)
glucose a mong 1 1 1 1 1 1 1 1 1 1
aware 57.1 (24/42) 26.2 (11/42) 45.2 (19/42) 38.1 (16/42) 64.3 (27/42) 76.2 (32/42) 88.1 (37/42) 21.4 (9/42) 11.9 (5/42) (n
o pat ., %) Ients www.jogh.org wIth dm 1 – – – – – – 1 1 1 1 1 1 • doi:10.7189/jogh.07.011102 a mong drug d Iabetes
pat 67.6 (25/37)
treatment 1.93 (1.15–3.22) 1.49 (0.91–2.43) 0.89 (0.46–1.73) Ients
wIth (n 1 1 1 1 1 1 1 1 1 1 dm o ., %) under
www.jogh.org • doi:10.7189/jogh.07.011102 by the Shenzhen Center for Chronic Diseases Prevention and Control [ estimated 5.2% the with agreement in is which 4.8%, was Shenzhen in prevalence diabetes overall The least, the cross–sectional nature of the current study does not allow establishing any causal relationships. not but Last findings. the of extrapolation for need is caution Therefore, distribution. gender and age to prevalence ofdiabetesduetotheunavailabilityShenzhenoverallpopulationinformationwithrespect and impaired fasting blood glucose. Fourth, we did not provide age and gender standardized estimate of of diabetes was based on fasting blood glucose, which may underestimate the prevalence rates of diabetes self-reported. We were not able to construct a criterion standard for rigid validation. Third, the diagnosis on awareness, pharmacological and non–pharmacological treatments, and primary care management were data Second, high. was rate response the although non–respondents, of characteristics the knowing out gram. However, the study had some limitations. First, the selection bias might pro have assurance quality been vigorous introduced a with using supervised and interviewers trained specially by collected were Data literature. international the with compatibility facilitate to diabetes of control and treatment ness, aware prevalence, the of definition international used commonly most Wemeasurement.the followed implemented. Standard protocols and instruments were employed for blood pressure and blood glucose relationship between diabetes prevalence and age [ positive the recognized widely have studies as phenomenon, the explain to help may average, on 39.26 9.0% [19] and 15.91% [20], respectively. Younger age of the participants in the current study, which was hai, which have economic context similar to that of Shenzhen, also yielded much higher prevalence rates, that the national average prevalence of diabetes was 11.6% [ that prevalence of diabetes was 9.7% between 2007 and 2008 [ showed Study Disorders Metabolic and Diabetes National China The level. national the at that than er corresponds to previous reports [ [ al. et Bragg by study national group, the with conflict in is decreasingwhich a then trend observed, was diabetes [21]. However, the prevalence rate of diabetes in the current study had the peak in 50 to 59 age for factor risk important an is age that shown have studies of number A studies. reported the with line in is which increasedage, diabetes with and glucose impairedprevalenceblood of the fasting rates Both tem’s capacityandcapability. tance of both primary and secondary prevention of diabetes, which challenges Shenzhen health care sys oping diabetes, which indicates a substantially greater disease burden. Our findings highlight the impor devel for risk increased at are impairment glucose with Subjects system. care health the to burden and ly 50% in 2010), its relatively higher prevalence rate than that of diabetes imposes a public health threat observed lower prevalence of impaired fasting blood glucose than that at the national level (approximate [ time and budget limited to due diabetes of diagnosis for tests tolerance glucose oral alence in the current study [ in the current study may have caused misclassifications and subsequent underestimation of diabetes prev campaign and has been implemented across China, including Shenzhen. Documentation of blood glu- blood of Documentation Shenzhen. including China, across implemented been has and campaign national the of part a is resident community every recordsfor health of establishment burden.The ease dis and social related and complications subsequent the decrease a and management in improvements [ condition wereawaretheir who patients diabetic ings comply with a previous study by Liu et al., which showed that drug treatment rate was 93.5% among the pharmacological treatment rate was high (88.1%) among participants aware of their disease. Our find diabetes, with patients of population general the in rates treatment non–pharmacological and cological (72% in 2014) [26], which suggests a room for improvement. Although our study showed low pharma- 2010 (30.1%) [8 in average national the higher times two almost is which 51.9%, was study our in rate awareness betes We also showed that the management of diabetes was not optimism, especially the control rate. The dia study, andwarrantsfurtherinvestigations. risk factors for diabetes, including poor education and low–income level, are not observed in the current socioeconomic Potential results. our influenced have might which diabetes, with diagnosed being after lifestyles their changed have may study.participants our Some in observed not were associations these [ diabetes with associated closely be to known well are smoking and Overweight/obesity of diabetes. several non–communicable diseases. Generally speaking, reducing blood pressure could reduce the risk for factors risk modifiable shared, include should diabetes addressing efforts health public and abetes, 1 ]. We alsorecorded higherprevalence ofco–morbidhypertensionamongdiabeticparticipants,which ]. However, the awareness rate was lower than that in developed countries like the USA 22 ]. However, it is impractical for large–scale epidemiological studies to adopt 24]. Our finding implies that hypertension is a modifiable factor for di 507
21]. Un–implementation of oral glucose tolerance tests ]. This implies that early screening may lead to the the to lead screeningmay early that implies This 21]. 8 ]. Studies conducted in Beijing and Shang 18]. The newest statics in 2010 indicated 17]. However, our estimate is low June 2017 •Vol. 7No. 1• 011102 Diabetes inShenzhen,China ]. Although we we Although 23]. ]. However,25]. ------
VIEWPOINTSPaPErS PaPErSVIEWPOINTS June 2017 •Vol. 7No. 1•011102 Yan etal. rEFEr ENCES 1 5 4 3 2 6 7 9 10 8 rural andurbanareas ofChina.JAMA.2017;317:280-9.Medline:28114552 F,Bragg in mortality cause-specific and MV,diabetes Holmes between Association Y,al. Chen Y,et H, Guo Du A, Iona Zimmet PZ, Magliano DJ, Herman WH, Shaw JE. Diabetes: a 21st century challenge. Lancet Diabetes Endocrinol. Diabetes Lancet challenge. century 21st a Diabetes: JE. Shaw WH, Herman DJ, Magliano PZ, Zimmet 2013. cause-specific mortality for 240 causes of death, 1990-2013: a systematic analysis for the and Global all-cause Burdenspecific ofage-sex Diseasenational Study and regional, Global, Collaborators. Death of Causes and Mortality 2013 GBD WHO. Globalhealthestimates:deathsbycause,age,sexandcountry, 2000-2012.Geneva:WHO; 2014. World HealthOrganization. Globalstatusreport onnoncommunicablediseases2014.Geneva:WHO; 2014. ticle inChinese].ZhonghuaNeiKeZaZhi.1981; China. of cities and provinces fourteen in 300,000 of survey mellitus Diabetes Group.Research Diabetes National 2014;2: Liu S tion].Brisbane: QueenslandUniversity ofTechnology; 2008. H Shen 59. China NationalNutritionandHealthSurvey, 2002. Xu Y Williams betes Care. 2012 ; tus and incidence of abnormal glucose metabolism: the Australian Diabetes, Obesity and Lifestyle (AusDiab) study. Medline:24002281 relevant foranimproved andbettermanagementofdiabetes. accesstomedicalservices be may care Primary strategies. intervention formulating when years 59 and 50 between aged residents the to paid be should attention More diabetes. of detection early for as well as prevention, secondary and primary efforts for health public of need the highlight findings Our undiagnosed. are diabetes with arehypertension Approximatelypopulation. Shenzhen in diabetes of factors risk the subjects the of half In conclusion, diabetes prevalence in Shenzhen (4.8%) is about half that of the Chinese average. Age and CONCLUSIONS management ofdiabetesisaneffective approach inreducing bloodglucoseneedsfurtherinvestigations. ship between primary care management and control rate of diabetes. Whether primary care standardized relation the test to designed not was study Our care. primary in diabetes of management standardized for guidelines launched also has government Chinese The facilities. care primary by provided services aging chronic diseases [ and national studies although have international shown the relevance care of primary approach in man We found that primary care management of diabetes was just over one–tenth of all treatment modalities, early detectionofpre–diabetes ordiabetes. on light shed may hypertension, developing of risk high at are who ≥35–year aged individuals for tests pressure blood as such China, in performed been has that settings care primary in screening pertension cose information for everyone may be an alternative for early detection of diabetes and pre–diabetes. Hy disclosure.pdf (availableuponrequest from thecorresponding author),anddeclare nocompetinginterests. www.icmje.org/coi_ at form Interest Competing Unified the completed have authors All interests: Competing All authorsread andapproved thefinalmanuscript. lection and analysis. XTD, YL and SJC helped to draft the manuscript and revised the draft for intellectual content. drafted the manuscript and were responsible for data interpretation. LZY, SJZ and JFZ participated in the data col Authorship contributions: data collection,interpretation ofthedata,andwritingorsubmittingmanuscript. Research Program (Grant number: JCYJ20160427183317387). The funding body had no role in research design, Basic Shenzhen and Government Shenzhen fromthe grant operating an by supported was study This Funding: this study. for data the contributed who staff and participants the all thank to like would authors The Acknowledgments: , Wang L , Wang W Lancet. 2015; . Effectiveness of a peer-led self-management program for older people with type 2 diabetes in China [disserta China in diabetes 2 type with people older peer-ledprogramfor a self-management Effectivenessof . 56- ED, Magliano 64. , He J , Zhang J Medline:24622669 35: , Bi Y 385:
1455- , He Y , Li M DJ, Zimmet doi:10.1001/jama.2013.168118 117- 61. , , Wang T 71. 27, Yao Medline:22619081 HTL and XFY conceived of the study, and took part in its design. HTL, XFY and HX Medline:25530442 28]. Chronic diseases management is designed to be one of the six integrated C doi:10.1016/S2213-8587(13)70112-8 PZ, Kavanagh , Zeng Z , et al. Prevalence and control of diabetes in Chinese adults. , et al. Prevalence of diabetes and impaired fasting glucose in Chinese adults, 20: AM, Stevenson 508 doi:10.2337/dc11-1410 Prev Chronic Dis.2011;8: 678- doi:10.1016/S0140-6736(14)61682-2 83. Medline:7341098 CE, Oldenburg doi:10.1001/jama.2016.19720 A13. BF, et al. Area-level socioeconomic sta Medline:21159225 www.jogh.org • doi:10.7189/jogh.07.011102 JAMA. 2013; 310: 948- Dia-
[ar ------www.jogh.org rEFEr ENCES • doi:10.7189/jogh.07.011102 28 25 26 27 14 11 13 12 23 24 15 16 17 18 20 19 21 22 mary mary care between Shanghai and Shenzhen: a cohort study of 3196 patients. Hammouche S doi:10.1097/MD.0000000000000455 World Health Organization. Obesity and overweight. 2011. Available: http://www.who.int/mediacentre/factsheets/fs311/ line:27711179 gov/diabetes/pubs/statsreport14/national-diabetes-report-web.pdf. Accessed:8June2016. Centers for Disease Control and Prevention. 2014 National Diabetes Statistics Report. 2014. Available: https://www.cdc. en/. Accessed:5June2016. vation? vation? An observational study. Li H Tanzania.from study population-based a Region: Kilimanjaro the in betes 2016; Li H opportunities. and challenges, methods, Surveillance: Risk-Factor disease noncommunicable to tematic review andmeta-analysis. Yach Riley L Agardh E diabetes. Motala Stanifer Risk. Shenzhen Statistics Bureau, Office NBS Survey in Shenzhen. Shenzhen Statistical Yearbook 2015. Beijing: China Statis 2014; China. Shenzhen, in centres health community by delivered care health primary using patients Yang tics Press; 2015. Zhou doi:10.1097/MD.0000000000002103 Shenzhen. District, Longhua in participants 6934 of study sectional Yang nese]. ZhonghuaYu FangYi XueZaZhi. Y Qin in Chinese].Beijing:People'sMedicalPublishingHouse; 2012. [book 2011. People’sin city Municipality.Beijing Beijing of of report status Government health population and Health author reply 6. ment ment and healthcare challenges. population based study in Shanghai, China. Shanghai, in study based population risk factors in Chinese rural population: the RuralDiab study. Liu X doi:10.3390/ijerph13050512 C Silva-Matos srep31426 , , Chung D W SH, Dou , Li Y Wei 2003; , WangR , 106: 15: HB, Peng J , Guthold R , Stuckler , , Li H AA, Omar AA, JW, Cleland 76. , Allebeck P X Nat Med.2006; , Li L 74-8. 10: , Wong Medline:24779564 , Fu X RY A Gomes , KF, Song 77- , Holland R Medline:20578276 X Ma , , Zhang L doi:10.1371/journal.pone.0164428 , Wei Medline:26696288 , Liu D 83. , Cowan M , Lu J MA, Pirie MA, MC, Yang N , Brownell , CR, Makuka , Hallqvist J X Medline:12668904 XL, Lin Y Zhao , , Mou J , Xue WJ. Prevalence of diabetes among men and women in China. A Azevedo , , Ren Y 12: , Steel N 62-6. Z FJ. Diabetes in Africa. Epidemiology of type 1 and type 2 diabetes in Africa. in diabetes 2 type and 1 type of Epidemiology Africa. in Diabetes FJ. HC, Zhang D , Savin S
, Wong
KD. Epidemiologic and economic consequences of the global epidemics of obesity and and obesity of epidemics global the of consequences economic and Epidemiologic KD. J Lu , , BMC BMC Health Serv Res. , Wong
, Zhou H Wu Diabetes Diabetes Metab. 2011 , Moradi T doi:10.1186/1471-2296-15-76 GJ, Egger Medline:16397571 Int JEpidemiol.2011; . Does quality of care for hypertension in primary care vary with postcode area depri C doi:10.1056/NEJMoa0908292 , doi:10.2105/AJPH.2015.302962 A Damasceno , . Inequalities in cardiovascular health between local and migrant residents: a cross- Wu , Bhatti L SY, Wong SY, Lao X , et al. Prevalence, awareness, treatment, control of type 2 diabetes mellitus and 2011; C doi:10.1177/174182670301000202 , Sidorchuk A . Prevalence of diabetes in Shenzhen between 1997 and 2009. JR, Maro V , et al. Prevalence, awareness, treatment and control of diabetes mellitus-a mellitus-a diabetes of control and treatment awareness, Prevalence, al. et ,
, Armstrong T 45: Int J Environ Res Public Health. 2016 Health. Public Res Environ J Int , et al. A comparison of the quality of hypertension management in pri MC, et al. Comparison of perceived quality amongst migrant and local 815-9. 509 A Prista , 2011 ; 37 , Maro H doi:10.1038/nm0106-62 . Type 2 diabetes incidence and socio-economic position: a sys : 40: Medline:22177305 237 ; 11 804- , et al. The World Health Organization STEPwise approach , Lunet , -
: Sci Rep. 44 297 , et al. Prevalence, risk factors, and complications of dia . 18. Medline:21236716 . Medline:22047508 Medline:21335614 N 2016;6:
. Diabetes in Mozambique: prevalence, manage prevalence, Mozambique: in Diabetes . Medicine. 2015; Medicine.
Medicine. Medicine. 2015 31426.
PLoS One. 2016; One. PLoS June 2017 •Vol. 7No. 1•011102
doi:10.1016/j.diabet.2010.10.006 Medline:27510966 N Engl J Med. 2010;
doi:10.1186/1472-6963-11-297 ; 94: doi:10.1093/ije/dyr029 13: ; 94 e2103. E512 : e455 Diabetes inShenzhen,China .
11: Medline:27213415 Am J Public Health Public J Am Medline:26656335 . Medline:25654383 e0164428. BMC Fam Pract. Fam BMC [article in Chi-
doi:10.1038/ 362: J Cardiovasc J 2425- Med- 6; 6; ------.
VIEWPOINTSPaPErS PaPErSVIEWPOINTS June 2017 •Vol. 7No. 1•011103 Taiwan: asystematicreview 2 diabetesinmainlandChina,HongKongand Socioeconomic statusandprevalenceoftype material. The onlineversionofthisarticlecontainssupplementary Electronic material: supplementary Edinburgh, Scotland,UK and Informatics,UniversityofEdinburgh, Usher InstituteofPopulationHealthSciences Caroline aJackson Sarah HWild,DanijelaGasevic, Hongjiang Wu,XiangruiMeng, [email protected] UK Edinburgh, EH89aG Teviot Place University ofEdinburgh and Informatics Usher InstituteofPopulationHealthSciences Hongjiang Wu Correspondence to:
adults aged 18 years or older in 2010 [ 2010 in older or years 18 aged adults in 11.6% to 1980 in older or years 30 aged adults in 0.9% about from na representative surveys indicate an increase in prevalence of diabetes in Chi [ decades recent over countries) income high in than faster much (and markedly increased has China in diabetes of prevalence The and type2diabetesisunclear. type 2 diabetes, while the association between income and occupation low education is probably associated with an increased prevalence of individual SES and prevalence of type 2 diabetes in China found that between association the of review systematic first This Conclusions occupational groups evenafteradjustingforage. jobs were reported to have a higher risk of type 2 diabetes than other collar white in working people and people Retired diabetes. 2 type and occupation between association significant a identified studies of number small a Only studies. between inconsistent was diabetes 2 type and income between association The level. education lowest the to compared highest the for 2.54) – (0.91 1.52 to reported) not (CI 0.39 from ranging (CI) interval confidence 95% and (OR) ratios odds with diabetes, 2 type and education between association verse and 12 studies, respectively. Most, but not all, studies reported an in 19 27, by reported was diabetes 2 type and occupation and income education, review.between systematic association the The in cluded in were and criteria inclusion the met studies Thirty–three Results tive synthesis. meta–analyses, therefore we summarized study results using a narra- modified Newcastle–Ottawa Scale. Heterogeneity of studies precluded and occupation. We appraised the quality of included studies using a income education, by measured groups SES different for population ies reporting prevalence or odds ratio for type 2 diabetes in a Chinese base and Global Health electronic databases for English language stud Methods and Taiwan. Kong Hong China, mainland in Chinese in diabetes 2 type of alence view to describe the cross–sectional association between SES and prev re systematic a Weconducted countries. middle–income and low– from studies in found been has pattern opposite the whereas tries, coun high–income in (SES) status socioeconomic low with people 2030. Type 2 diabetes has been reported to be more prevalent among least at until continue will situation this that suggesting data isting ex of extrapolation with 2015, in world the in diabetes with people Background We conducted a systematic literature search in Medline, Em 510 China is estimated to have had the largest number of of number largest the had have to estimated is China
2 , 3 www.jogh.org ]. China is thought to have had had have to thought is China ]. • doi:10.7189/jogh.07.011103 global journal of 1 ]. Nationally Nationally ]. health ------www.jogh.org • doi:10.7189/jogh.07.011103 http://www.crd.york.ac.uk/PROSPERO/display_record.asp?ID=CRD42016047913 Document Supplementary line This systematic review was conducted using the PRISMA guideline (see checklist in Appendix S1 in Literature search METHODS between SES indicators and type 2 diabetes; a greater number of participants; or the most recent publi recent most the or participants; of number greater a diabetes; 2 type and indicators SES between association the on information more either: with publication the including thereby given, order the in data from the same study were reported in multiple publications, we applied the following three criteria selected for specific characteristics such as hypertension or obesity; and non–English language articles. If written in English. We excluded: case–control and hospital–based studies; studies limited to populations were and occupation; or income education, as exposure SES individual defined groups; SES different in populations for diabetes 2 type of ratio odds or diabetes 2 type of prevalence on data reported older; studies which: included Chinese populations in mainland China, Hong Kong or Taiwan aged 18 years or We included cross–sectional population–based studies and baseline surveys of population–based cohort Study selectionanddataextraction input, are notChinesespeakers.Nolimitswere appliedforlanguageorpublicationtime. include Chinese databases because other members of the research team, who provided additional review dix S2 in (1980–May 2016) and Global Health (1973–May 2016) using a comprehensive search strategy (Appen 2 diabetes in mainland China, Hong Kong and Taiwan. We searched Medline (1946–May 2016), Embase atic literature search of published studies describing the association between SES and prevalence of type is more prevalent among lower than higher socioeconomic groups [ in China [28– China in diabetes 2 type of prevalence and SES between associations inconsistent reported have studies Previous exists within and between regions, but it is not clear how this is associated with diabetes prevalence [ perienced extremely rapid economic development over the past 30 years and major economic inequality cable diseases occur initially in high SES groups, before appearing in low SES groups [ noncommuni transition, epidemiological the during that, indicates fromcountries Evidence developed velopment [22– tern has been found in studies from low– and middle–income countries undergoing rapid economic de [ society structureof the in occupies individual an position the describes that concept complex a is (SES) status Socioeconomic gesting thatthissituationwillcontinueuntilatleast2030[ the largest number of people with diabetes in the world in 2015, with extrapolation of existing data sug land China. Hong Kong and Taiwan is useful for helping estimate future diabetes prevalence in urban areas of main differ to that of mainland China [ in urbanised environments and developing related lifestyle habits than in China. Health care systems also stage of economic development and epidemiological transition, with a larger proportion of people living advanced more a at are former However,the China. mainland in counterparts their to similar netically Taiwan.and Kong Hong TaiwanChina, and mainland Kong in Hong ulations in people Chinese arege SES (measured by education, income and occupation) and prevalence of type 2 diabetes in Chinese pop between association Wethe China. describe to cross–sectionalstudies of review systematic a conducted in diabetes 2 type of prevalence and SES of review systematic published no is Tothere knowledge, our prevention andsecondary ofdiabetesintheChinesepopulation. to primary der to attempt to address socioeconomic health disparities in diabetes as well as for planning approaches SES and diabetes is not the same in all populations [ betes across populations, including age, overweight/obesity and physical inactivity, the association between [7 populations ent differentwith diseases, non–communicable differand in nicable strengthsdirectionsassociation andof of a population’s health [ cators such as income, education and occupation. SES has been recognized as an important determinant Online Supplementary Document). Online Although Supplementary the reviewersprimary are Chinese, we did not ]. Understanding the association between SES and diabetes in China is necessary in or in necessary is China in diabetes and SES between association the Understanding 31]. 25]. – ]. Unlike many risk factors that have consistently shown an association with dia with association an shown consistently have that factors risk many Unlike 12]. 5 6 ]. It consists of many dimensions and is often measured by using several indi several using measuredby often is and dimensions many of consists It ]. ]. SES is closely linked to a wide range of health problems, including commu 32]. Understanding the association between SES and type 2 diabetes in ). The protocol was registered on PROSPERO and can be accessed at at accessed be can and PROSPERO on registered was protocol The ). Status andprevalenceoftype2diabetesinmainlandChina,HongKongTaiwan 511
4 , 13– 15]. In high–income countries, type 2 diabetes 4 ]. 10, 16– June 2017 •Vol. 7No. 1•011103 21], whereas the opposite pat . We carried out a system a out Wecarried . 26]. China has ex ]. 27]. On------
VIEWPOINTSPaPErS PaPErSVIEWPOINTS June 2017 •Vol. 7No. 1•011103 Wu etal. review (Figure 1). systematic the in included were and criteria inclusion the met review.studies text Thirty–three full ter wereaf studies excluded further screening,131 abstract and and title wereafter studies excluded 1771 The literature search initially identified 3003 studies, with 1935 remaining after de–duplication. Of these, Selection ofstudies r between studiesprecluded meta–analyses. heterogeneity since findings, study the of synthesis narrative a by accompanied were tables and figures lence; only odds ratios; and both prevalence and odds ratios, and ranked from high to low quality. These preva only presented: studies whether to according grouped materials, supplementary in summarized For each SES indicator, the full results from each study, including stratification by urban/rural status were ings foroccupationinfigures, giventhemarkedheterogeneity indefinitionofoccupation. presentedwe gender), resultthe largestthe with find the summarize to possible not was It size. sample and age by stratified (eg, subgroups several in only results reporting studies For factors. additional for sex only, but unfortunately few studies reported these minimally adjusted estimates, with most adjusting and Ideally,age confounding. for for adjusted adjustment ratios plete odds summarized have would we tios of type 2 diabetes for the highest compared with lowest SES level from the model with the most com ra odds and level SES highest and figures lowest the prevalence in the diabetes showing 2 type mary of 2 diabetes for each of education, income and occupation. For education and income, we presented sum We reported type 2 diabetes prevalence and odds ratios for associations between SES indicators and type Synthesis ofstudyfindings representing higherquality. score higher a with fromscored0–8, was study Each studies. in reportedprobability level and intervals of two scores were given for the assessment of the diagnosis of diabetes and one score for the confidence of two scores were given for the control of confounding factors. Within the outcome category, a maximum tion of the sample; and ascertainment of SES exposures. Within the comparability category, a maximum study can be awarded one score for each of the following items: representativeness of the sample; descrip comparability,selection, category,including selection dimensions Withinthe three outcome. into and a [ quality study of assessment quantitative a allows which Newcastle–Ottawa Scale (NOS) for cohort studies (Appendix S3 in modified a using studies included of quality the appraised independently XM) Twoand (HW authors Quality assessment not bemade. could decision a if arbiter as acting thirdauthor a with XM) and (HW authors two the between cussion el with the most complete adjustment for confounding was chosen. Disagreements were resolved by dis reporting several models to estimate the association between SES and diabetes, the result from the mod werewerestudies they prevalencereportedratio For if for not odds authors. calculated and by intervals sures (prevalence and odds ratio); and adjustments for potential confounders. Where possible, confidence graphics; participant selection; study location; SES measures; diabetes diagnosis method; outcome mea- demo diabetes; 2 type with people of number size; sample publication; of year year; study author; on: Weinformation articles. extracted included for characteristics key extracted independently and articles Two authors (HW and XM) screened the titles, abstracts and (for potentially relevant studies) full text of lished in 2011, which reported no studies of incidence of diabetes and SES were identifiedpub inreview China [ systematic a on based 2010 after published studies longitudinal for search literature pilot in different SES groups in mainland China were identified in our pilot literature search. Wea conducted diabetes of incidence of studies longitudinal no as studies longitudinal include Wenot date. did cation were conducted in mainland China (three in urban areas, five in rural areas and 16 in both urban and and urban both in 16 and areas rural in five areas, urban in (three China mainland in conducted were in presented is studies included the of characteristics the of overview An Study characteristics ESULTS
512
]. This scale contains six items, categorized categorized items, six contains scale This 33]. Online Supplementary Document) www.jogh.org . Twenty–fourTable1. studies • doi:10.7189/jogh.07.011103 15]. ------www.jogh.org Figure 1. Flowchart ofselectionstudiesinsystematicreview. • doi:10.7189/jogh.07.011103 vide confidence intervals orpvaluesforstatisticaltests[ vide confidenceintervals diabetes solely based on self–reported diagnosis [ ). The studies from Hong Kong reported an inverse association [ association inverse an reported Kong Hong from studies The Document ). tary other two reported no evidence of an association [ association an of evidence reportedno two other est compared to the lowest education level ( ciation, with odds ratios (95% CI) ranged from 0.39 (CI not reported) to 1.52 (0.91, 2.54) for the high inverse association between education level and type 2 diabetes or a possible trend toward such an asso S4 and S6 in ( level education higher to compare lower a with those in higher was diabetes Generally, 2 confounders. type potential of various prevalence for controlled that ratios odds presented which of 14 ratios, odds reported studies Fifteen prevalence. standardized reported five which among estimates, prevalence reported 16 diabetes, 2 type and education on reporting studies 27 the Among Association betweenSESandtype2diabetes or asimpleclassificationofemployedandunemployed. differed greatly between studies, with the definition based on: job titles; skills (manual or non–manual); four personal income measures. Occupation was reported in 12 studies, but the measures of occupation of school years completed. Income was reported in 19 studies, including 15 family income measures and ported in 27 studies, and was classified either as highest educational level (in 21 studies) or the number studies) or three (8 studies) SES indicators. Education was the most commonly used indicator, being re- A single measure of SES was reported in 15 studies, with the remaining studies reporting data for two (10 Measures ofSES association between education and type 2 diabetes, two reported an inverse association [ Taiwanfrom an studies reporting four Among diabetes. 2 type and education between [55] association ratio for the association between SES and diabetes [ diabetes and SES between association the for ratio tors for type 2 diabetes [ Online Supplementary Document). Most, but not all, studies reported either a significant 4 ]. Eight studies only reported crude prevalence of diabetes or unadjusted odds Status andprevalenceoftype2diabetesinmainlandChina,HongKongTaiwan 513 Figure 3 and Appendices S5 and S6 in
31, 58, 43, 35 tion since both sex and age are important risk fac risk important are age and sex both since tion not report sex or age distribution, which is a limita any sampling techniques [ to cooperate with the research team, without using willing were who participants included study ond [ area study the of population older from an association for elders to represent the total sample the selected One population. study lected ified NOS assessment. Two studies had a highly se 4 to 7 with a mean score of 6.0 based on the mod The quality scores of included studies ranged from Quality ofincludedstudies and sixprovided bothprevalence andoddsratios. groups, SES different for diabetes of ratio odds ed diabetes in different SES groups, 15 studies provid Twelve diabetes. of prevalenceof provided studies diagnosis for glucose blood postprandial and cose glu blood random (OGTT), test tolerance glucose ods to diagnose diabetes. Some studies used an oral blood glucose were the most commonly used meth fasting and diabetes Self–reported sex. by betes ported the association between SES and type 2 dia re five only but women, and men both included [ 988 from ranged size Sample 2009. since published studies 24 with marked increase in studies on this topic over time, a with 2012, to 1986 from ranged year Study rural areas), three in Hong Kong and six in Taiwan. ]. Among all studies, four studies reportedsex– studies four studies, all Among 62]. , 47, 42 35, , 55, 45 41, , 56]. In addition, four studies did not pro 48 42, , 49 57]. , 55 , ] to 512 to 63] June 2017 •Vol. 7No. 1•011103 57 , 61 ] and five studies defined defined studies five and ] and Appendices Appendices and 2 Figure 52]. Fourteen studies did ] and no significant significant no and 57]
]. All studies studies All [41]. 891 Online Supplemen- 49, ], and a sec a and 50], 61], and the ------
VIEWPOINTSPaPErS PaPErSVIEWPOINTS June 2017 •Vol. 7No. 1•011103 Wu etal. Table 1. Characteristics of studies identified in systematic review with evaluation of association between prevalence of type 2 diabetes and SES in mainland China, Hong Kong and Taiwan study study year partIcIpants selectIon sample, n gender age (mean ± sd, years) study loca- ses IndIca- dIabetes dIagnosIs outcome qualIty (dIabetes) (% men) tIon* tor method measure score (prevalence/ odds ratIo) Wu et al. 2016 [ 34] 2007–2011 Using a random multistage stratified sampling method, 2–3 cit - 23 010 46.9 ≥18 (43.0, Urban Education Self–report or Prevalence 7 ies within each of 6 provinces from south and north China were (983†) 30.4–56.3)‡ and rural FBG selected, from which several communities and villages were randomly selected Liu et al. 2016 [ 35] 2001, 2010 Using a two–stage cluster random sampling method, 9 residen - 2001: 2001: ≥60 (67.9 ± 5.8 in Urban Education Self–report or Prevalence 5 tial communities were randomly selected in Wanshoulu district 2277 41.4; 2001; 71.2 ± 6.6 in FBG in Beijing, from which all households were chosen and one per - (487§), 2010: 2010) son aged ≥60 was randomly selected in each household 2010: 40.3 2102 (521§) Zhou et al. 2015 2010 Using a multistage probability sampling design, 3 communities 98 058 45.7 ≥18 (NR) Urban Education, Self–report or Odds ratio 6 [36] or villages were selected in each of 4 subdistricts with probabil - (12 237§) and rural occupation FBG or OGTT or ity proportional to size from each National Disease Surveillance HbA1c Point; within each community 50 households were randomly selected, and one person randomly selected in each household Yu et al. 2015 [ 37] 2012 Using a multistage stratified random cluster sampling method, 16 834 45.9 18–79 (42.7 ± 14.5) Urban Education, Self–report or Prevalence 7 32 counties were selected from 9 cities, from which 3 or 4 (1380||) and rural income, FBG towns were randomly selected; within each town, 3 neighbor - occupation hood committers were randomly selected, from each of which
514 one village was randomly selected, before randomly selecting people aged 18 to 79 years
Xue et al. 2015 [ 38] 2006, 2009 Using a stratified random cluster sampling method, people who 6894 39.4 35–74 (51.2 ± 10.6) Urban Education, HbA1c Odds ratio 7 lived in Qingdao city for at least 5 years in 3 urban areas and 3 (360‡) and rural income rural areas were selected Xu et al. 2015 [ 39] 2010–2011 Using a multistage stratified random sampling method, 3 cen - 1659 49.5 ≥18 (44.0 ± 15.2) Urban Education, Self–report or Prevalence 7 tral temples and 3 counties in Chengdu region were selected (106||) and rural income FBG or OGTT from each altitude level;4 townships were selected from each county, and within each townships 3 villages were selected, from which all people aged ≥18 were selected Bu et al. 2015 [ 40] 2007–2008 Using a multistage stratified random sampling method, cities 39 071 39.2 ≥30 (NR) Urban Education Self–report or Odds ratio 6 within 14 provinces in China were selected, from which 152 (3254||) and rural FBG or OGTT city districts and 112 rural villages were randomly selected; www.jogh.org people aged ≥20 years who had lived at their current residence for ≥5 years were selected Bragg et al. 2014 2004–2008 People aged 30–79 were selected from five urban and five rural 512 891 41.0 30–79 (NR) Urban Education, Self–report or Prevalence 5
• doi:10.7189/jogh.07.011103 41] areas in China; these were permanent residents identified [ (30 773||) and rural income FBG or random through official residential and invited by letter after extensive blood glucose publicity campaigns Zhang et al. 2013 2005 Using a multistage stratified cluster random sampling method, 7315 NR 20–79 (NR) Urban Education, Self–report or Odds ratio 6 [28] 3 communities were randomly selected from two urban and (688§) income, FBG or OGTT one suburban district(s) in Tianjin; 3 neighborhoods were ran - occupation domly selected from each community and all people who had lived in the selected neighborhoods for >5 years and were aged ≥15 years were selected Xia et al. 2013 [ 42] 2010–2011 Using a stratified random sampling method, 3 communities 12 000 51.0 >18 (49.1 ± 0.26¶) Urban Education, FBG Prevalence 5 within each of 4 districts of Haikou were randomly selected, (636§) occupation from which 1000 people were selected www.jogh.org Table 1. Continued
study study year partIcIpants selectIon sample, n gender age (mean ± sd, years) study loca- ses IndIca- dIabetes dIagnosIs outcome qualIty
• doi:10.7189/jogh.07.011103 (dIabetes) (% men) tIon* tor method measure score (prevalence/ odds ratIo) Wu et al. 2013 [ 43] 2010 Using a probability sampling design and a multistage cluster 13 157 48.1 ≥50 (62.6 ± 0.3) Urban Income Self–report Both 6 sampling method, 1 county from rural National Disease Sur - (868‡) and rural veillance Points (DSPs) and one district from urban DSPs were selected from 8 provinces, resulting in 64 principle sample units Wang et al. 2013 2011 Using a multistage stratified random sampling method, all 4801 44.8 25–86 (51.1) Rural Education, Self–report or Odds ratio 7 [44] townships within two counties in Yunnan province were se - (341||) income FBG or OGTT lected and within each township 2 villages were randomly se - lected Cai et al. 2013 [ 45] 2011 Using a multistage stratified random sampling method, 1 coun - 9396 46.0 ≥18 (51.7 ± 19.6) Rural Education Self–report or Prevalence 6 ty with high wealth and 1 county with low wealth were ran - (614§) FBG domly selected in Yunnan province; people aged ≥18 were ran - domly selected from 20 villages within each county Yan et al. 2012 [ 46] 2009 Using a multistage random cluster sampling method, people 8458 47.1 ≥18 (NR) Urban Income FBG or HbA1c Odds ratio 6 aged ≥7 were randomly selected from 228 communities in 9 (NR) and rural provinces Chen and Chen. NR Using a multistage random cluster sampling method, 2–12 13 741 57.0 18–64 (NR) Taiwan Occupa- Self–report Both 6 2012 [ 47] townships were randomly selected from each 23 counties in (NR) tion Taiwan, within which 12–123 neighborhoods were randomly Status andprevalenceoftype2diabetesinmainlandChina,HongKongTaiwan selected; within each neighborhood, 4 households were ran - 515 domly selected
Shi et al. 2011 [ 48] 2002 Using a multistage random sampling method, households were 2849 45.9 ≥20 (47.0) Urban Education FBG Prevalence 6 randomly selected from 6 counties and 2 cities; all people in (79||) and rural the households were selected Lin et al. 2011 [ 49] 2004 Using a multistage random sampling method with a sampling 2332 48.6 ≥40 (56.9) Taiwan Education, Self–report or Prevalence 6 rate proportional to size within each stage, 39 Li units were ran - (284§) income FBG domly selected from each 8 city districts; people were random - ly selected from each sample Li Kavikondala et al. 2005–2008 People were randomly selected from ‘The Guangzhou Health 19 818 26.7 50–96 (60.4) Urban Education, Self–report or Odds ratio 6 2011 [ 50] and Happiness Association’ who are permanent residents in (2193‡) occupation FBG Guangzhou Fu et al. 2011 [ 30] 2006–2007 All adult residents aged 18–64 were selected with exclusion of 4506 41.4 18–64 (46.1 ± 10.0) Rural Education, Self–report or Both 7
June 2017 •Vol. 7No. 1•011103 those who were temporary workers or university students not (99‡) income, FBG living in the county from four rural communities in Deqing, occupation Zhejiang province Cai et al. 2011 [ 29] 2008–2010 Using a multistage stratified random sampling method, 3 coun - 10 007 46.2 ≥18 (NR) Rural Education, Self–report or Odds ratio 6 ties with low, high and high level of wealth were randomly se - (657§) income FBG lected from Yunnan province; all townships in counties were selected and 3 villages in each township were selected by prob - ability proportional to size, from which people aged ≥18 years were randomly selected Wei et al. 2010 [ 51] 2005 Using a multistage random cluster sampling, communities were 1058 50.1 >20 (NR) Rural Education, Self–report or Both 7 randomly selected from 5 areas in a region in Heilongjiang (75§) income FBG or OGTT Zhou et al. 2009 2007–2008 People aged ≥20 years in 10 communities in Beijing were se - 2801 27.2 35–79 (54.7) Urban Education, Self–report or Both 7 [52] lected where their committees would like to cooperate with the (580§) and rural income, FBG or OGTT research team occupation
VIEWPOINTSPaPErS PaPErSVIEWPOINTS June 2017 •Vol. 7No. 1•011103 Wu etal. Table 1. Continued
study study year partIcIpants selectIon sample, n gender age (mean ± sd, years) study loca- ses IndIca- dIabetes dIagnosIs outcome qualIty (dIabetes) (% men) tIon* tor method measure score (prevalence/ odds ratIo) Ning et al. 2009[ 53] 2001–2002, Using a stratified random cluster sampling method, people aged 11 624 39.6 35–74 (54.4 in Urban Education, Self–report or Odds ratio 7 2006 35–74 years living in Qingdao city for at least 5 years were ran- (1383‡) 2001–2002; 51.5 in and rural income, FBG or OGTT domly selected from 3 urban districts and 4 rural counties 2006) occupation Hu et al. 2009 [ 54] 2000–2001 Using a multistage stratified random sampling method, 1 rural 15 236 48.4 35–74 (50.1 ± 0.12¶) Urban Education, Self–report or Prevalence 7 and 1 urban county within each of four provinces from North (986§) and rural income, FBG and 4 provinces from South China were randomly selected; 1 occupation township/street was randomly selected from each county, from
which people aged 35–74 years were randomly selected Xu et al. 2006 [ 31] 2000–2001 Using a multistage random sampling method, 3 urban districts 29 340 49.8 ≥35 (NR) Urban Education, Self–report Both 6 and 2 rural counties were randomly selected in Nanjing, from (556§) and rural income, each of which 3 streets/towns were selected; 3 villages were occupation randomly selected in each street/town, from which people aged ≥35 y who had been a local resident for at least 5 years in each village were selected Chou and Chi. 1996 People aged ≥60 years in 6000 households were randomly se - 2003 47.0 ≥60 (NR) Hong Education Self–report Prevalence 4 55] lected from a continuous sample survey, which use a full list of 2005 [ (246§) Kong addresses of quarters in Hong Kong as the sampling frame Yu and Wong. 2004 NR Households in Tai Po Hong Kong were randomly selected by 2670 NR ≥20 (NR) Hong Income Self–report Odds ratio 5 [56] telephone survey using a residential telephone directory (NR) Kong 516 Woo et al. 2003 1995–1996 People aged 25–74 from 3 major of Hong Kong were random - 988 (59§) 49.4 25–74 (45.6 ± 11.7) Hong Education FBG or OGTT Prevalence 5
[57] ly selected by telephone survey Kong Chen et al. 2001 1996–1997 Using a multistage proportional stratified random cluster sam - 1293 41.8 50–79 (63.8) Taiwan Education Self–report or Odds ratio 7 [58] pling method, people aged 50–79 years were randomly select - (182‡) FBG ed from 3 townships Chen et al. 1999 1995–1996 Using a proportional stratified random sampling method, peo- 1601 48.7 40–79 (57.4) Taiwan Occupa- Self–report or Odds ratio 7 [59] ple aged 40–79 years were randomly selected from 6 areas (295§) tion FBG Pan et al. 1997 [ 60] 1994 People aged ≥25 years were selected from cities and rural areas 213 515 52.9 25–64 (NR) Urban Income Self–report or Odds ratio 6 in 19 provinces (4864§) and rural FBG or OGTT Chou et al. 1994 1991 All people aged >30 years in each village from Kin–Hu Town 3236 47.5 >30 (NR) Taiwan Education FBG or OGTT Odds ratio 5 [61] were selected (193§) Tai et al. 1992 [ 62] 1986 8 subdistricts of Ta–An District in Taipei City and 5 villages of 11 478 50.3 ≥40 (NR) Taiwan Education, Self–report or Odds ratio 6 www.jogh.org 11 counties of Taiwan Province were randomly selected (715§) income FBG or postprandial blood glucose or
• doi:10.7189/jogh.07.011103 OGTT SD – standard deviation, Self–report – self–reported history of type 2 diabetes or using medication for type 2 diabetes, FBG – fasting blood glucose, OGTT – oral glucose tolerance test, NR – not reported *Urban and/or rural are in mainland China. †Median and interquartile range. ‡Number is estimated based on the crude prevalence of diabetes. §Number reported in the study. ||Number is estimated based on the adjusted prevalence of diabetes. ¶Standard error. www.jogh.org Figure 2. Prevalence oftype2diabetesinthelowestandhighestlevelseducationincomeincludedstudies. • doi:10.7189/jogh.07.011103 association [49] and no significant association [ inverse association between income and type 2 diabetes [56]. The studies from Taiwan reported an inverse in S9 and S8 Appendices tios, the evidence for an association between income level and type 2 diabetes was inconsistent ( in S9 and S7 pendices [ association between income and type 2 diabetes only in men in rural areas. Zhou et al. [ [ al. et Ning which among analyses, reportedsex–specific studies four ies, ference, small samplesize[39]. butincludedavery 2 diabetes by income level across studies, with considerable inconsistency between studies ( type of prevalence of pattern clear no was There confounders. various for adjusted ratios odds presented reportedfour standardizeda only which of one prevalence.reportedbut all studies ratios, Fourteen odds Of the 19 studies reporting on income and type 2 diabetes, 10 reported prevalence estimates, among which did notfindanygenderdifferences inthe association betweeneducationandtype2diabetes. type 2 diabetes among men, with the opposite observed in women [ of prevalence increased with associated was education higher that reporting two with analyses, specific 46] also found a positive association in men but not in women. A fourth study did not find a gender dif ). Similarly,ra Document). odds reporting Supplementary studies Online among ). The study from Hong Kong reported an an reported Kong Hong from study The Document). Supplementary Online Status andprevalenceoftype2diabetesinmainlandChina,HongKongTaiwan 517
62] between income and type 2 diabetes. Among all stud 34, 52]. The other two studies [ June 2017 •Vol. 7No. 1•011103 ] found a significant positive positive significant a found 53] 52] and Yan et al. Figure 2, - Ap Figure 3, 39, 53] - - -
VIEWPOINTSPaPErS PaPErSVIEWPOINTS June 2017 •Vol. 7No. 1•011103 Wu etal. income inincludedstudies. Document). occupation and prevalence of type 2 diabetes (Appendices S10, S11 and S12 in sons andassemblers.However, moststudiesdidnotreport astatisticallysignificantassociationbetween officials, salesper as such occupation of kinds other to compared diabetes 2 type of risk lowest the had lar occupations, even after controlling for confounding factors. Chen and Chen [ the prevalence of type 2 diabetes was much higher in people with white collar occupations than blue col [ al. et Xu age. for adjusting after people employed to compared retired in diabetes 2 type of using a figure as for education and income. Zhou et al. [ ings from studies reporting on occupation and type 2 diabetes and we were unable to present the results easy to rank the occupation classification from high to low SES. This affects the comparability of the find controlled for various potential confounders. As the measures of occupation were heterogeneous, it is not which of all ratios, odds reported studies Nine factors. various standardizedfor estimates reporting two Of the 12 studies reporting on occupation and type 2 diabetes, eight reported prevalence estimates, with Figure 3. hest vs lowest levels of education and Study specificoddsratiosfortype2diabetescomparingthehighestvslowestlevelsofeducationand
518
36] and Zhang et al. [ www.jogh.org 28] found an increased risk Online Supplementary Online Supplementary • doi:10.7189/jogh.07.011103 47] found professionals ] found found 31] - - - www.jogh.org • doi:10.7189/jogh.07.011103 and women, who may not be the main in earners the family [ adults young for especially income, personal than reliable more be to believed is income ever,family total studies in this systematic review used personal income as individual’s measure of SES [ <10 to 360) (US$ <2500¥ from ranging category lowest the with studies between different very is level income of classification the Furthermore, measure among older people, especially retired people, where income is low but actual wealth can be high. [ association null a toward non–differentialbias of risk the increases and reliability the decreases obviously which it, report to under– or over–estimated in studies as people may consider income sensitive information and be reluctant [ SES life whole of indicator good a necessarily not completed in young adulthood, income is unstable and sensitive to change in life circumstances and so it is usually is which education, review.unlike our First, in diabetes and income between association sistent pared tootheroccupations.Thesefindingswere notobviouslyinfluencedbystudyyearorqualityscore. prevalence, a few did report higher prevalence among people who were retired or in white–collar jobs com diabetes and occupation between association no found studies most While studies. between inconsistent was diabetes 2 type and income between association The diabetes. 2 type of prevalence decreased a with associated probably is education higher that Taiwansuggests and Kong Hong China, mainland in lations popu- Chinese in diabetes 2 type of prevalence and SES between association the of review systematic This DISCUSSION only one study considered family size [ size consideredfamily study one only may have higher outgoing costs than a smaller family [ come to all family members, family size should be accounted for, since for the same income, a larger family ple from low– and middle–income countries [ tries with low income were more likely to have type 2 diabetes [ review.our in fromcoun- people high–income found has which previousstudies with inconsistent is This differed diabetes 2 studies type between prevalenceof and income between association directionof The froment economicreturn schoolyearscompletedcompared toeducationallevelachieved[ diabetes measured education as school years completed [ to note that all three studies reporting a positive association between education and prevalence of type 2 important is It studies. between heterogeneity the observed of some explain might education of initions diabetes and education may have been distorted by confounding factors. Furthermore, differences in def justing for several variables. This means that the crude positive association between prevalence of type 2 education level, but the logistic regression model revealed a non–significant inverse association after ad [ al. et Xu cation) [64] and so choosing 7 years as the cut–point may have different effects in different birth cohorts. from the 1960s a large proportion of Chinese started to receive middle school education (9 years of edu However,years. 7 cut–offof survey.a 2001 using the education in dichotomised association study This survey,2010 a in older groupsor education years higher 60 in aged diabetes people in no found having [ al. et Liu methods same the using and area same the in differentconducted this being to result.Despite contribute partly may which China, of parts other with Tibetan population. Tibet is an undeveloped region at an earlier economic development stage compared small relatively a in diabetes 2 type of prevalence and education between association positive a ported [ al. et Xu example, For association. opposite the found some but creaseddiabetes, prevalenceof de with associated are education of levels higher review,that systematic suggested this studies In most Explanation forfindingsinthissystematic review velopment during the past several decades [ association between SES and diabetes. According to the latest China nationally representative diabetes diabetes representative nationally China latest the to According diabetes. and SES between association geneity between studies. Different diagnostic criteria may have a different effect on the magnitude of the The methods used to diagnose type 2 diabetes varied across studies, which was another source of hetero for diabeteschallenging. factor risk a as classification its makes that activity physical of levels differing also but income and tion cupation across studies was complex and heterogeneous. Occupation in China is associated with educa systematic review, though a few studies reported statistically significant findings. The classification of oc this in diabetes 2 type of prevalence and occupation between association consistent a find not Wedid lated behaviorsmaylagbehindchangesineconomicconditions andmayalsodiffer indifferent settings. 31 ] found a significantly higher crude prevalence of type 2 diabetes in people with a higher higher a with people in diabetes 2 type of prevalence crude higher significantly a found ] ]. In addition, income is only one part of an individual’san of part one good only is income very addition, a In not 70]. is and assets ]. Furthermore, China has undergone a very rapid economic de economic rapid undergonevery has Furthermore,a China 31]. Status andprevalenceoftype2diabetesinmainlandChina,HongKongTaiwan 000¥ (US$ 1440) for a family’s whole year income [ income family’syear a whole for 1440) (US$ 000¥ 72]. However, changes in an individual’s lifestyle and health–re 519 22,
68]. There are several potential explanations for the incon ]. Second, self–reported income is more likely to be be to likely more is income self–reported Second, 69]. 71]. Among 15 studies reporting total family income, 31, ] found a much higher prevalence of type 2 2 type of prevalence higher much a found 35] 69]. However, when applying total family in 35, 66, 39]. However, people may receive differ 67], but an opposite association in peo June 2017 •Vol. 7No. 1•011103 28, 53, 54, 41, 65]. 60]. How- ]. Four Four 52]. ] re 39] ------
VIEWPOINTSPaPErS PaPErSVIEWPOINTS June 2017 •Vol. 7No. 1•011103 Wu etal. and occupation. tes. However, further work is needed to determine whether similar associations are observed with income diabe 2 type of prevalenceincreased an with probablyassociated is education low that found China in diabetes 2 type of prevalence and SES individual between association the of review systematic first This CONCLUSIONS ic inequalitiesinhealththerefore haveimportant implicationsforglobalhealth. The epidemiological transition in China and the challenges of identifying and addressing socio–econom the largest number of people with diabetes in the world and is undergoing rapid economic development. with country the is China countries, between varies diabetes and SES between association the Although how associationsbetweenSESanddiabeteschangeover time inChina. tion and in different regions of China. Additionally, repeated cross–sectional studies are needed to explore differentdifferpopula in associations the sub–groups whether of identify to and diabetes and cupation of those in Chinese language publications, are needed to explore the association between income and oc review including studies, More inconsistent. were diabetes and occupation income, between sociations However,populations. Chinese in diabetes 2 as type prevalenceof and education relationshipbetween inverse an of evidence some found review This understood. fully is diabetes and SES between sociation as the when made be only can diabetes in disparities health reducingsocioeconomic for polices Health Implications forhealthpolicyandfuture research were observed. graphical locationsmeantthatnoobviouspatterns Taiwan. However, the inconsistent findings and limited number of studies within each of these study geo and Kong Hong China, mainland rural urban, in location study by vary diabetes 2 type and SES tween be association the whether examine review,to this made In were effortsKong. Hong in living those to ly income of US$ 1500 provides vastly different standards of living for a family in west China compared ferent implications for people living in developed and undeveloped areas. For example, an average month dif have may income of level same the Also, work. farming or agricultural in engage to level education [76– areas undeveloped and veloped SES indicators may have different values and implications in different urban and rural settings and in de car andhomeownership)orarea–based SESmeasures [75]. not include the association between other indicators of individual SES (such as wealth, house condition, [ Scotland in men than women in stronger be to appears diabetes 2 type of prevalence in gradient SES the example, For [73]. studies previous by reported been has outcomes health in gradient SES sex–specific A populations. different in consistent is this whether clear not is it differedbut diabetes sex 2 by type prevalenceof and SES between association directionof and strength the that found review systematic this in studies few A diabetes. 2 type and indicators SES between association the of over–adjustment to led have may factors these of Inclusion diabetes. 2 type various factors in addition to age and sex, many of which may lie on the causal pathway between SES and tion between SES and diabetes. Additionally, all but three studies reporting odds ratios were adjusted for alence estimates presented in studies were unadjusted for age, which is a key confounder of the associa of relationships between other variables and health outcomes. It is also important to note that most prev ies. SES was generally considered as a descriptive variable of the study sample or a potential confounder ciation between SES and diabetes was rarely the main research aim or hypothesis of most identified stud asso- the that is limitation important Another bias. potential exclude to needed also is Chinese in lished Our review was limited to papers published in the English language. A systematic review of studies pub Limitations ofthestudy groups. Bragg et al. [ SES by differ may misclassification this and group non–diabetic the to assigned erroneously been have that defined diabetes based on self–report, a large proportion of those with diabetes in these studies may survey, around 70% of Chinese adults with diabetes were undiagnosed [ diagnosed andundiagnoseddiabetesare required. both and SES between association the on reporting studies more association, this examine clearly more likely to be aware of their health conditions. However, another study [ SES typically have more access to health resources such as routine health checks, thus they may be more cation and low income groups, while the opposite was found for self–reported diabetes. People with high
41] found that undiagnosed diabetes was more common among people in low edu ]. For example, people in rural areas may not need a very high high very a need not may areas rural in people example, For 79]. 520 ]. Furthermore, the scope of this review did did review this of scope the Furthermore, 74]. 28] did not find this difference. To www.jogh.org 3 ]. Thus, among the five studies • doi:10.7189/jogh.07.011103 ------www.jogh.org rEFEr ENCES • doi:10.7189/jogh.07.011103 1 8 7 6 5 4 3 2 9 12 11 10 18 17 16 15 14 13 21 20 19 22 ies with4.4millionparticipants.Lancet.2016;387:1513-30. NCD Risk Factor Collaboration. Worldwide trends in diabetes since 1980: a pooled analysis of 751 population-based stud McLaren L. Socioeconomic status and obesity.Rev.and Epidemiol status 2007;29:29-48. Socioeconomic L. McLaren mxm001 line:18520717 Med- 2008;23:335-9. Cardiol. Opin Curr status. socioeconomic and Y.Hypertension Sharabi M, Huerta I, Grotto National Bureau ofEconomicResearch; 2008. MA: Cambridge, mechanisms. and dimensions Voglhealth: A, and T.Lleras-Muney status DM, Socioeconomic Cutler Lynch J,KaplanG.Socioeconomicposition.2000:Socialepidemiology. NewYork: Oxford UniversityPress. resources/2015-atlas.html. Accessed:11May2017. Available:2015. edition. seventh atlas diabetes IDF Federation. Diabetes International 59. Medline:24002281 Xu Y, Wang L, He J, Bi Y, Li M, Wang T, et al. Prevalence and control of inces andmunicipalitiesinChina.[articleChinese].ZhonghuaNeiKeZaZhi.1981;20:678-83. diabetes in Chinese adults. JAMA. 2013;310:948- National Diabetes Co-operative Study Group. A mass survey of diabetes mellitus in a population of 300 000 in 14 prov implications forcare. NatRevCardiol. 2009;6:712-22.Medline:19770848 and risks disease: cardiovascularW, and Luo status M, Socioeconomic DesMeules WielgoszA. AM, AS, Clark Duncan Marmot M. Social determinants of health inequalities. Lancet. 2005;365:1099-104. Lancet. inequalities. health of determinants Social M. Marmot Clin Oncol.1991;9:1500-9.Medline:2072149 DF,Cella KW,Lee PM, Silberfarb JC, Holland AB, Kornblith EJ, Orav J survival. cancer and status Socioeconomic al. et doi:10.2337/diacare.27.3.716 2004;27:716-21. Care. Diabetes women. and men Swedish in diabetes 2 type of risk excess differencesin V,socioeconomic T, Grill of Andersson S, Explanations A, Efendic al. Ahlbom et EE, AgardhJ, Hallqvist and Type 2 diabetes mellitus. Diabet Med. 2000;17:478-80. Evans JM,NewtonRW, RutaDA,MacDonaldTM,MorrisAD.Socio-economic status, obesityandprevalence ofType 1 doi:10.1007/s00125-008-1146-1 century.2008;51:1971-9. 21st Diabetologia. the of beginning the at Europe across mellitus betes Espelt A, Borrell C, Roskam AJ, Rodríguez-Sanz M, Stirbu I, Dalmau-Bueno A, et al. Socioeconomic inequalities in dia 2000;54:173-7. Medline:10746110 Health. Community Epidemiol J areas. deprived in mellitus diabetes 2 type of prevalence increased showing study V,Connolly P,Sherriff N, Unwin W. Kelly based R, population Bilous a status: socioeconomic and prevalence Diabetes tematic review andmeta-analysis.IntJEpidemiol.2011;40:804-18. Agardh E, Allebeck P, Hallqvist J, Moradi T, Sidorchuk A. Type 2 diabetes incidence and socio-economic position: a sys- atic review. DiabetesCare. 2007;30:744-52.Medline:17327354 CY,Jeon system - a diabetes RP,2 Lokken type of risk and intensity moderate of activity Physical RM. Dam van FB, Hu 2007;356:213-5. Medline:17229948 Med. J Engl N challenge. growing world—a developing the in diabetes and Obesity M. P,Nahas Hossain El B, Kawar S0140-6736(05)74234-3 ed diabetes.IntJEpidemiol.2003;32:381-5. Medline:12777423 Tang M, Chen Y, Krewski D. Gender-related differences in the association between socioeconomic status and self-report Public Health.2001;91:76-83.Medline:11189829 Survey. J Examination Am Nutrition and Health National Third the from evidence men: and women white Hispanic non- and American African in diabetes 2 type VaccarinoSV.and V,JM, Kasl status Robbins H, Socioeconomic Zhang 2000;17:478-80. Medline:10975218 Med. Diabet deprivation. to relation its and practice general in P.prevalence Meadows mellitus diabetes Variationof line:9096979 ence in prevalence of diabetes between rural and urban populations in Bangladesh. Diabetes Care. 1997;20:551-5. abu Sayeed M, Ali L, Hussain MZ, Rumi MA, Banu A, Azad Khan AK. Effect of socioeconomic risk factors on the differ ing author)anddeclare nootherconflictsofinterest. uniform disclosure form at http://www.icmje.org/coi_disclosure.pdf (available upon request from the correspond Ed/Astra Zeneca and on the Scottish Diabetes Register from Novo Nordisk. All authors have completed the ICMJE Conflict of interest ical writing.Allauthorsapproved thefinaldraftofmanuscript. wrote the first draft of the paper. All authors contributed to the interpretation of the findings and the paper’s crit Author’scontributions: Funding: HWandXMare fundedbyscholarshipsfrom theChinaScholarshipCouncil. tion oftheinstitutionorfunder. posi official an not and own our are article submitted this in expressed views We the Disclaimer: that confirm doi:10.2337/diacare.20.4.551 doi:10.1097/HCO.0b013e3283021c70 doi:10.1001/jama.2013.168118 : Sarah H Wild reports honoraria for lectures on epidemiology of diabetes from Global Med HW and XM independently conducted the literature search and data extraction. HW HW extraction. data and search literature the conducted independently XM and HW doi:10.1136/jech.54.3.173 doi:10.1056/NEJMp068177 Status andprevalenceoftype2diabetesinmainlandChina,HongKongTaiwan doi:10.1200/JCO.1991.9.8.1500 521 doi:10.2105/AJPH.91.1.76
Medline:10975218 Medline:27061677 doi:10.2337/dc06-1842 doi:10.1093/ije/dyg075 Medline:21335614 doi:10.1038/nrcardio.2009.163
Medline:17478442 doi:10.1046/j.1464-5491.2000.00309.x doi:10.1016/S0140-6736(16)00618-8 June 2017 •Vol. 7No. 1•011103 Medline:15781105 doi:10.1093/ije/dyr029 http://www.diabetesatlas.org/ Medline:14988291
Medline:7341098 Medline:18779946 doi:10.1093/epirev/ doi:10.1016/ Med------
VIEWPOINTSPaPErS PaPErSVIEWPOINTS June 2017 •Vol. 7No. 1•011103 Wu etal. rEFEr ENCES 29 28 27 26 30 25 24 23 34 33 32 31 43 42 35 41 40 39 38 37 36 45 44 47 46 come in transforming rural China. BMC Public Health. 2011;11:862. Fu C, Chen Y, Wang F, Wang X, Song J, Jiang Q. High prevalence of hyperglycaemia and the impact of high household in ua rsdns n einl anad hn. ibt e. 2006;23:1239-46. Med. Diabet China. mainland regional doi:10.1111/j.1464-5491.2006.01965.x in residents rural and urban in diabetes 2 type diagnosed and income average Family N. WareOwen F,E, R, Xu Leslie M, Zhang XM, Yin multilevel spatialanalysisof98,058adults.DiabetesCare. 2015;38:72-81. Zhou M, Astell-Burt T, Bi Y, Feng X, Jiang Y, Li Y, et al. Geographical variation in diabetes prevalence and detection in china: rural southwest China. Trop Med Int Health. 2011;16:1070-6. Le C, Jun D, Zhankun S, Yichun L, Jie T. Socioeconomic differences in diabetes prevalence, awareness, and treatment in 62. Medline:23397898 2 diabetes: findings based on a large population-based cross-sectional study in Tianjin, China. Diabet Med. 2013;30:e157- Zhang H, Xu W, Dahl AK, Xu Z, Wang HX, Qi X. Relation of socio-economic status to impaired fasting glucose and Type doi:10.1073/pnas.1403158111 2014;111:6928-33. A. S U Sci Acad today’sNatl Procin China. inequality Income X. Y,Zhou Xie 2017. lications/PUB_9789290614746/en/. Accessed:1February A multicountry analysis of noncommunicable disease surveillance data. 2010. Available: World Health Organization. Noncommunicable disease risk factors and socioeconomic inequalities – what are the links? doi:10.1007/s001250100627 glucose tolerance in India: National Urban Diabetes Survey. Diabetologia. 2001;44:1094-101. Survey.Diabetologia. Diabetes Urban National India: in tolerance glucose impaired and diabetes of V,prevalence Vijay High A, V,al. Kapur Mohan et C, AK, Snehalatha Das A, Ramachandran munity inSriLanka.JRSocPromot Health.2004;124:92-4. com rural a Tennakoonin S, mellitus Rambodagalla diabetes TemporalU, of S. Illangasekera prevalence the in trends 1993;16:68-75. Medline:8422835 Care. Diabetes factors. nutritional and health, sociodemographic, to relationship their and Malaysia in Malays and nes Ali O, Tan TT, Sakinah O, Khalid BA, Wu LL, Ng ML. Prevalence of NIDDM and impaired glucose tolerance in aborigi a recent cross-sectional survey. Medicine.2016;95:e2712.Medline:26962771 Wu J, Cheng X, Qiu L, Xu T, Zhu G, Han J, et al. Prevalence and clustering of major cardiovascular risk factors in China: oxford.asp. Accessed:23June2016. Available:2000. meta-analyses. in studies nonrandomised of ity Wells G, Shea B, O’Connell D, Peterson J, Welch V, Tugwell P. The Newcastle-Ottawa Scale (NOS) for assessing the qual pedia ofPublicHealth.2ndedition.Oxford: AcademicPress; 2017. Encyclo International In Taiwan,of. and Systems Kong, Health Hong China, Mainland Q. Meng J-FR, Lu GM, Leung Study on Global AGEing and Adult Health (SAGE) Wave 1. PLoS One. 2013;8:e74176. Medline:24069278 Wu F, Guo Y, Kowal P, Jiang Y, Yu M, Li X, et al. Prevalence of major chronic conditions among older Chinese City,adults: the HainanIsland,China.IranJPublicHealth.2013;42:222-30. Haikou in adults the in diabetes 2 type of factors YangT.risk X, and Zhang YangPrevalence J, Q, WangCai Z, Z, Xia 2016;16:278. Health. Public BMC decade. past for trend changing its and elderly Liu M, Wang J, He Y, Jiang B, Wu L, Wang Y, et al. Awareness, treatment and control of type 2 diabetes among Chinese line:24344928 2014;31:540-51. Med. Diabet Biobank. Kadoorie China the Chinese: adult 000 500 in disease cardiovascular of Bragg F, Li L, Smith M, Guo Y, Chen Y, Millwood I, et al. Associations of blood glucose and prevalent diabetes with risk Disorders Study. PLoSOne.2015;10:e0139039.Medline:26406982 in the middle-aged population of China: a subgroup analysis of the 2007-2008 China National Diabetes and Metabolic Bu S, Ruan D, Yang Z, Xing X, Zhao W, Wang N, et al. Sex-specific prevalence of diabetes and cardiovascular risk factors nal.pone.0129966 Tibetan adults in China: a population-based study. PLoS One. 2015;10:e0129966. Xu S, Jiayong Z, Li B, Zhu H, Chang H, Shi W, et al. Prevalence and clustering of cardiovascular disease risk Medline:26777908 factors among betes diagnosed using HbA1c criteria among Chinese Adults in Qingdao, China. Biomed Environ Sci. 2015;28:884-93. Xue B, Tan JB, Ning F, Sun JP, Zhang KY, Liu L, et al. Association between serum uric acid and prevalence of type 2 dia Jilin (China).IntJEnviron ResPublicHealth.2015;13:ijerph13010070. Yu J, Ma Y, Yang S, Pang K, Yu Y, Tao Y, et al. Risk factors for cardiovascular disease and their clustering among adults in s12889-016-2874-7 southwest China.IntJCardiol. 2013;165:273-7.Medline:21908062 rural in mellitus diabetes 2 type of burden economic The al. et WenlongC, Z, Keying H, Jianhui D, Jun L, Lin C, Le Medline:23874667 fects on diabetes mellitus and glucose intolerance in rural southwest China: a multilevel analysis. PLoS One. 2013;8:e68553. WangKW, WuWeiL, JQ, ef- Cai demographic ZK, W.individual Shu and contextual of magnitude the of Assessment journal.pone.0074176 53. Medline:22322107 Chen FL, Chen PY. Health disparities among occupations in Taiwan: a population study. J Occup Health. 2012;54:147- China Health and Nutrition Survey. Obes Rev. The 2012;13:810-21. China: in risk cardiometabolic of burdenP, Gordon-Larsenexpanding S, The Du al. B, et Zhang S, Li J, Li YanS, doi:10.1111/dme.12392 doi:10.1371/journal.pone.0068553
doi:10.1111/dme.12156 doi:10.1539/joh.11-0052-FS doi:10.2337/diacare.16.1.68 522 Medline:21668591 Medline:22738663 Medline:15067982 Medline:23641399 http://www.ohri.ca/programs/clinical_epidemiology/ Medline:22078718 doi:10.1371/journal.pone.0139039 doi:10.1016/j.ijcard.2011.08.039 Medline:25352654 doi:10.1097/MD.0000000000002712
doi:10.1111/j.1365-3156.2011.02805.x doi:10.1111/j.1467-789X.2012.01016.x Medline:26047133 doi:10.1177/146642400412400214 www.jogh.org Medline:26987372 doi:10.1186/1471-2458-11-862 http://www.wpro.who.int/pub- • doi:10.7189/jogh.07.011103 doi:10.2337/dc14-1100 Medline:17054602 Medline:11596662 Medline:24778237
doi:10.1371/jour
doi:10.1186/ doi:10.1371/ Med------
www.jogh.org rEFEr ENCES • doi:10.7189/jogh.07.011103 71 70 69 68 67 66 65 64 63 62 61 58 57 56 55 60 59 49 48 75 74 73 72 54 53 52 51 50 tion andriskfactors.DiabetesResClinPract.2001;51:59-66. Chen KT, Chen CJ, Gregg EW, Engelgau MM, Narayan KM. Prevalence of type 2 diabetes mellitus in Taiwan: Ethnic varia guidelines. AnnuRevPublicHealth.1997;18:341-78.Medline:9143723 and methodologies, concepts, research: health public US in class social Measuring NE. Moss DR, Williams N, Krieger 2000;54:207-14. Medline:10746115 Health. Community Epidemiol J research. inequalities health for implications non-reporting: Income TurrellG. miol CommunityHealth.2006;60:7-12.Medline:16361448 Epide J Lynch1). DA, (part JW,Lawlor position M, socioeconomic Shaw of B, Indicators Galobardes G. Smith Davey factors associated with urbanization in India. Diabetes Care. 2008;31:893-8. Ramachandran A, Mary S, Yamuna A, Murugesan N, Snehalatha C. High prevalence of diabetes and cardiovascular risk healthpol.2010.07.018 results of the Canadian Community Health Survey.Health Policy.Community Health Canadian the 2011;99:116-23. of results T, Bryant M, income: Dinca-Panaitescu prevalenceand S, Diabetes Dinca-Panaitescu D. Raphael B, Pilkington I, Daiski Medline:19527580 Wallach JB, Rey MJ. A socioeconomic analysis of obesity and diabetes in New York City. Prev Chronic Dis. 2009;6:A108. doi:10.1086/227355 1981;86:1093-111. J Sociol. Am research. stratification in variable neglected a certification: by Selection MA. Faia Saywell WG.EducationinChinasinceMao.CanJHighEduc.1980;10:1-27. Kong Chineseaged70yearsandover. Neuroepidemiology. 1994;13:50-8. Hong elderly among factors associated and impairment cognitive of YuenPrevalence J, YK. Lau S, Lau SC, WooHo J, doi:10.1093/ije/21.3.511 1992;21:511-7. Epidemiol. J study.Int follow-up 4-year a adults: Chinese among pertension Tai TY, Chuang LM, Wu HP, Chen CJ. Association of body build with non-insulin-dependent diabetes mellitus and hy 35. Medline:7736904 Chou P, Liao MJ, Tsai ST. Associated risk factors of diabetes in Kin-Hu, Kinmen. Diabetes Res Clin Pract. 1994;26:229- Nutr. 2003;57:523-30.Medline:12700613 Woo J, Ho SC, Sham A, Sea MM, Lam KS, Lam TH, et al. Diet and glucose tolerance in a Chinese population. Eur J Clin Biomed Environ Sci.2004;17:234-45.Medline:15386950 Yu TS, Wong TW. Socioeconomic distribution of health and health care utilization in a new town in Hong Kong, China. 2005;51:334-9. Medline:16110236 Gerontology.adults. Chinese Kong Hong older in mellitus diabetes to related disability Functional I. Chi KL, Chou bres.2009.02.021 9. Medline:9353605 Pan XR, Yang WY, Li GW, Liu J. Prevalence of diabetes and its risk factors in China, 1994. Diabetes Care. 1997;20:1664- 69. Medline:10414941 abetes mellitus in Penghu Islets, Taiwan: evidence of a rapidly emerging epidemic? Diabetes Res Clin Pract. 1999;44:59- Chen KT, Chen CJ, Gregg EW, Williamson DF, Narayan KM. High prevalence of impaired fasting glucose and type 2 di type 2 diabetes in Chinese. Eur J Clin Invest. 2011;41:659-66. Lin WY, Xaiver Pi-Sunyer F, Chen CC, Davidson LE, Liu CS, Li TC, et al. Coffee consumption is inversely associated with China. Nutrition.2011;27:194-8.Medline:20471806 Jiangsu, adults, in diabetes of risk the and consumption Egg G. Holmboe-Ottesen M, Zhou C, Zhang YuanB, Z, Shi line:3066632 Liberatos P, Link BG, Kelsey JL. The measurement of social class in epidemiology. Epidemiol Rev. 1988;10:87-121. Diabetes Care. 2011;34:1127-32.Medline:21421800 mortality among people With Type 2 Diabetes A study from the Scottish Diabetes Research Network Epidemiology Group. on status socioeconomic Effectof al. et AD, Morris JA, McKnight RS, Lindsay HM, Colhoun WalkerSJ, Livingstone JJ, Action. 2015;8:27259.doi:10.3402/gha.v8.27259 Phillips SP, Hamberg K. Women’s relative immunity to the socio-economic health gradient: artifact or real? Glob Health Accessed: 29August2016. Available:Equity,2016. & China. Poverty World Bank. The population: the InterASIA Study. Diabetes Res Clin Pract. 2009;84:288-95. Pract. Clin Res Study. Diabetes InterASIA the population: Hu D, Sun L, Fu P, Xie J, Lu J, Zhou J, et al. Prevalence and risk factors for type 2 diabetes mellitus in the Chinese adult line:19719705 Med- 2009;26:855-63. Med. Diabet China. Qingdao, in population Chinese adult the in diabetes of prevalence F,Ning WangHR, Nan WG, Gao YH, Dong ZC, increasePang the dramatic in the with associated factors Risk al. et SJ, munities inBeijing.DiabetesResClinPract.2009;86:233-8. Zhou X, Ji L, Luo Y, Han X, Zhang X, Sun X, et al. Risk factors associated with the presence of diabetes in Chinese com na-based rural community adults. Public Health. 2010;124:593-601. WeiW, SY,Liu FF,Zeng Yao SP, HT,Zhang Wan Typeal. et G, impairedand North-Chi diabetes in 2 tolerance glucose doi:10.1002/ajhb.21206 a developing population: the Guangzhou Biobank Cohort Study.Cohort Biobank Guangzhou the population: developing a 2011;23:747-54. Biol. J Hum Am Kavikondala S, Jiang CQ, Zhang WS, Cheng KK, Lam TH, Leung GM, et al. Intergenerational influences on diabetes in doi:10.1093/oxfordjournals.epirev.a036030 doi:10.1111/j.1464-5491.2009.02791.x doi:10.2337/diacare.20.11.1664 doi:10.1016/0168-8227(94)90065-5 doi:10.1016/S0168-8227(99)00025-X doi:10.1159/000086371 doi:10.1136/jech.54.3.207 doi:10.1038/sj.ejcn.1601586 Status andprevalenceoftype2diabetesinmainlandChina,HongKongTaiwan 523 doi:10.2337/dc10-1862 doi:10.1016/j.nut.2010.01.012 Medline:21226707 Medline:11137183 Medline:19836096 doi:10.1136/jech.2004.023531 http://povertydata.worldbank.org/poverty/country/CHN Medline:20846702 doi:10.1146/annurev.publhealth.18.1.341 Medline:8190206 Medline:18310309 Medline:19442859
doi:10.1111/j.1365-2362.2010.02455.x doi:10.1016/S0168-8227(00)00200-X doi:10.1016/j.diabres.2009.09.014 June 2017 •Vol. 7No. 1•011103 Medline:20724018
doi:10.1016/j.puhe.2010.05.014 doi:10.1159/000110358
doi:10.2337/dc07-1207 Medline:21987430
Medline:1634313 doi:10.1016/j.dia doi:10.1016/j. Med------.
VIEWPOINTSPaPErS PaPErSVIEWPOINTS June 2017 •Vol. 7No. 1•011103 Wu etal. rEFEr ENCES 79 78 77 76 2000;32:735-51. doi:10.1068/a32130 Martin D, Brigham P, Roderick P, S, Diamond Barnett I. The (mis) representation of rural deprivation. Environ Plann A. Sci Med.2003;57:2055-63.Medline:14512237 Gilthorpe MS, Wilson RC. Rural/urban differences in the association between deprivation and healthcare utilisation. Soc doi:10.1136/jech.56.10.754 2002;56:754-61. Health. Community Epidemiol J comparison. urban/rural an level: area small the P,at Roderick need S, care health Barnett proxiesof Wrigleythree I, between InterrelationsDiamond H. D, Martin 85. Medline:11027953 Haynes R, Gale S. Deprivation and poor health in rural areas: inequalities hidden by averages. Health Place. 2000;6:275-
doi:10.1016/S1353-8292(00)00009-5 doi:10.1016/S0277-9536(03)00071-6 524 www.jogh.org • doi:10.7189/jogh.07.011103 Medline:12239201
(continued from the outside cover)
Emmanuelle Daviaud, Donnela Besada, Natalie Leon, Sarah rohde, David Sanders, Nicholas Oliphant, Tanya Doherty PE rS Costs of implementing integrated community case management (iCCM) in six Pa african countries: implications for sustainability 100 rynaz rabiee, Emilie agardh, Matthew M Coates, Peter allebeck, anna–Karin Danielsson alcohol–attributed disease burden and alcohol policies in the BrICS–countries during the years 1990–2013 111 allison I Daniel, robert H Bandsma, Lyubov Lytvyn, Wieger P Voskuijl, Isabel Potani, Meta van den Heuvel Psychosocial stimulation interventions for children with severe acute malnutrition: a systematic review 119 Evelyn reinke, Supriyatiningsih, Jörg Haier Maternal mortality as a Millennium Development Goal of the United Nations: a systematic assessment and analysis of available data in threshold countries using Indonesia as example 131 Marco Floridia, Valentina Frisina, Marina ravizza, anna Maria Marconi, Carmela Pinnetti, Irene Cetin, Matilde Sansone, atim Molinari, Francesca Cervi, alessandra Meloni, Kety Luzi, Giulia Masuelli, Enrica Tamburrini; The Italian Group on Surveillance on antiretroviral Treatment in Pregnancy Evolving treatment implementation among HIV–infected pregnant women and their partners: results from a national surveillance study in Italy, 2001–2015 140 Pallab K Maulik, Sudha Kallakuri, Siddhardha Devarapalli, Vamsi Krishna Vadlamani, Vivekanand Jha, anushka Patel Increasing use of mental health services in remote areas using mobile technology: a pre–post evaluation of the SMarT Mental Health project in rural India 145 Shanshan Zhang, Beatrice Incardona, Shamim a Qazi, Karin Stenberg, Harry Campbell, Harish Nair; Severe aLrI Working Group Cost–effectiveness analysis of revised WHO guidelines for management of childhood pneumonia in 74 Countdown countries 158 Franziska Meinck, Deborah Fry, Choice Ginindza, Kerri Wazny, aldo Elizalde, Thees F Spreckelsen, M Catherine Maternowska, Michael P Dunne Emotional abuse of girls in Swaziland: prevalence, perpetrators, risk and protective factors and health outcomes 172 Trevor Duke, Ilomo Hwaihwanje, Magdalynn Kaupa, Jonah Karubi, Doreen Panauwe, Martin Sa’avu, Francis Pulsan, Peter Prasad, Freddy Maru, Henry Tenambo, ambrose Kwaramb, Eleanor Neal, Hamish Graham, rasa Izadnegahdar Solar powered oxygen systems in remote health centers in Papua New Guinea: a large scale implementation effectiveness trial 184 Mejbah Uddin Bhuiyan, Stephen P Luby, Nadia Ishrat alamgir, Nusrat Homaira, Katharine Sturm–ramirez, Emily S. Gurley, Jaynal abedin, rashid Uz Zaman, aSM alamgir, Mahmudur rahman, Ismael r. Ortega–Sanchez, Eduardo azziz–Baumgartner Costs of hospitalization with respiratory syncytial virus illness among children aged <5 years and the financial impact on households in Bangladesh, 2010 193 Oghenebrume Wariri, Lucia D’ambruoso, rhian Twine, Sizzy Ngobeni, Maria van der Merwe, Barry Spies, Kathleen Kahn, Stephen Tollman, ryan G Wagner, Peter Byass Initiating a participatory action research process in the agincourt health and socio–demographic surveillance site 202 Xiaolin Xu, Gita D Mishra, Mark Jones Mapping the global research landscape and knowledge gaps on multimorbidity: a bibliometric study 219 PaPErS Triin Habicht,KaijaLukka,Elinr of Mother–to–ChildTransmissionProgram how besttomonitorearlyinfantinfectionswithinthePreventi on r Livia Puljak,IvonaBuc´an Kajo Buc´an, a in Estonianhealthsystem:analysisofnationalpaneldata Shifting chronicdiseasemanagementfromhospitalstoprimarycare r Toward eliminationofmother–to–childtransmissionHIVinSoutha Gayle GSherman,a rESEarCH THEME3:MONITOr Eastern r Sociodemographic, behavioral,andenvironmentalfactorsofchildmortalityin a rESEarCH THEME2:VErBaL families’ financialburden:evidencefromJamaica User–fee–removal improvesequityofchildren’shealthcareutilizationandreduces Zhihui Li,MingqiangGüntherFink,PaulBourne,TillBärnighausen,r and resilienthealthsystems Fukushima aftertheGreatEastJapanEarthquake:lessonsfordevelopingresponsive Shingo Fukuma,Shahiraa rESEarCH THEME1:HEa intervention study Integrating palliativecareintonationalhealthsystemsina Mackuline a Liz Grant,JuliaDowning,EmmanuelLuyirika,MaireadMurphy,LizNamukwaya,FatiaKiyange, Gender biasincareseekingpractices57low–andmiddle–incomecountries Janaína CaluCosta,FernandoCWehrmeister,a base andchallengesindatasynthesis r Paul Eze,EvelynBalsells,MoeHKyaw,HarishNair adolescents: amulti–country,cross–sectionalstudy Prevalence andassociatedriskfactorsofviolenceagainstconflict–affectedfemale JLindsay Stark,Khudejhaa a 16–yearretrospectivecohortstudy Epidemiology ofoculartraumainchildrenrequiringhospitaladmission: and childhealth:1.rationale, methods anddatabasedescription community–based primaryhealth careinimprovingmaternal,neonatal Comprehensive reviewofthe evidenceregardingtheeffectivenessof Henry BPerry,BahieMr IN rESEarCH THEME5:EVIDENCE health: gapsandopportunities Improving coveragemeasurementforreproductive,maternal, neonatalandchild Coverage MeasurementforMNCH* Melinda KMunos,CynthiaStanton,JenniferBryce;theCoreGr oup forImproving rESEarCH THEME4:IMPr Henry DKalter Kathryn LFalb onelle Niit,MargaretOkobi,a ifat a of the evidence isk factorsforClostridiumdifficileinfections–anoverviewoftheevidence lain KKoffi,r IMPr s, tun, IpekGurol–Urganci,ThomasHone,LisaPell,JonathanStokes, OVING MNCH OVING MNCH egion ofCameroon:resultsfromasocialautopsystudy tieno, EmillyKemigisha–Ssali,JennyHunt,KalySnell,Scotta nita Matas,JosipaMarinLovric´, DarkoBatistic´, IvnaPleštinaBorjan, omain SWounang,FélicitéeNguefack,SeidouMoluh,Paul–r hmad Haeri Mazanderani, Peter Barron, Sanjana Bhardwaj, hmad HaeriMazanderani,PeterBarron,SanjanaBhardwaj, assekh, SundeepGupta,JessWilhelm, Paula hmed, r sghar, GaryYu,CarolineBora,a drian Puren,DebraJJackson,a aaper, JarnoHabicht LTH POLICYaNDSYSTEMS OVING COVEraGEMEa ei Goto,ThomasSInui,r /SOCIaL ING E FOr COMMUNITY-Ba luísio JDBarros,CesarGVictora MTCT INSOUTH aUTOPSY sham a ifat a
frica: a multi–country frica: amulti–country meena EbrahimGoga tun, ShunichiFukuhara SUr ssazenew Baysa, ssazenew Baysa, rESEarCH SED PHC SED PHC aFr Freeman EMENT ifat a
Murray,MhoiraLeng
ICa
tun
oger Libite, oger Libite, frica: frica:
309 219 298 286 278 265 257 248 237 230 328 317 Mary Carol Jennings, Subarna Pradhan, Meike Schleiff, Emma Sacks, Paul a Freeman, Sundeep Gupta, Bahie M rassekh, Henry B Perry
PE rS a comprehensive review of the evidence regarding the effectiveness
Pa of community–based primary health care in improving maternal, neonatal and child health: 2. maternal health findings 342 Emma Sacks, Paul a Freeman, Kwame Sakyi, Mary Carol Jennings, Bahie M rassekh, Sundeep Gupta, Henry B Perry Comprehensive review of the evidence regarding the effectiveness of community–based primary health care in improving maternal, neonatal and child health: 3. neonatal health findings 355 Paul a Freeman, Meike Schleiff, Emma Sacks, Bahie M rassekh, Sundeep Gupta, Henry B Perry Comprehensive review of the evidence regarding the effectiveness of community–based primary health care in improving maternal, neonatal and child health: 4. child health findings 367 Meike Schleiff, richard Kumapley, Paul a Freeman, Sundeep Gupta, Bahie M rassekh, Henry B Perry Comprehensive review of the evidence regarding the effectiveness of community–based primary health care in improving maternal, neonatal a nd child health: 5. equity effects for neonates and children 388 Henry B Perry, Emma Sacks, Meike Schleiff, richard Kumapley, Sundeep Gupta, Bahie M rassekh, Paul a Freeman Comprehensive review of the evidence regarding the effectiveness of community–based primary health care in improving maternal, neonatal and child health: 6. strategies used by effective projects 402 Henry B Perry, Bahie M rassekh, Sundeep Gupta, Paul a Freeman Comprehensive review of the evidence regarding the effectiveness of community–based primary health care in improving maternal, neonatal and child health: 7. shared characteristics of projects with evidence of long–term mortality impact 416 robert E Black, Carl E Taylor, Shobha arole, abhay Bang, Zulfiqar a Bhutta, a Mushtaque r Chowdhury, Betty r Kirkwood, Nazo Kureshy, Claudio F Lanata, James F Phillips, Mary Taylor, Cesar G Victora, Zonghan Zhu, Henry B Perry Comprehensive review of the evidence regarding the effectiveness of community–based primary health care in improving maternal, neonatal and child health: 8. summary and recommendations of the Expert Panel 433
rESEarCH THEME 6: GLOBaL HEaLTH rESEarCH PrIOrITIES Lorainne Tudor Car, Nikolaos Papachristou, Catherine Urch, azeem Majeed, rifat atun, Josip Car, Charles Vincent Prioritizing medication safety in care of people with cancer: clinicians’ views on main problems and solutions 445 renee Sharma, Michelle F Gaffey, Harold alderman, Diego G Bassani, Kimber Bogard, Gary L Darmstadt, Jai K Das, Joseph E de Graft–Johnson, Jena D Hamadani, Susan Horton, Luis Huicho, Julia Hussein, Stephen Lye, rafael Pérez–Escamilla, Kerrie Proulx, Kofi Marfo, Vanessa Mathews–Hanna, Mireille S Mclean, atif rahman, Karlee L Silver, Daisy r Singla, Patrick Webb, Zulfiqar a Bhutta Prioritizing research for integrated implementation of early childhood development and maternal, newborn, child and adolescent health and nutrition platforms 454 Narendra K arora, archisman Mohapatra, Hema S Gopalan, Kerri Wazny, Vasantha Thavaraj, reeta rasaily, Manoj K Das, Meenu Maheshwari, rajiv Bahl, Shamim a Qazi, robert E Black, Igor rudan Setting research priorities for maternal, newborn, child health and nutrition in India by engaging experts from 256 indigenous institutions contributing over 4000 research ideas: a CHNrI exercise by ICMr and INCLEN 465 PaPErS 50 applicationsoftheCHNr Setting healthresearchprioritiesusingtheCHNr Igor r Hong KongandTaiwan:asystematicreview Socioeconomic statusandprevalenceoftype2diabetesinmainlChina, Hongjiang Wu,XiangruiMeng,SarahHWild,DanijelaGasevic,Carolinea Diabetes inShenzhen,China:epidemiologicalinvestigationandhealthcarechallenges Jianfeng Zou,YiLuo,SijingCao Xinfeng Yan,HuiXia,HaitaoLi,XiaotingDeng,LizhenYang,ShaojuanZhao, evaluation study (Ying YangBao)amongyoungchildreninruralQinghai,China:amixedmethods Monitoring andevaluatingtheadherencetoacomplementaryfoodsupplement Huijun Han,MinXing,LiChen,XiaozhenDu,r Qiong Wu,YanfengZhang,SuyingChang,WeiWang,MichelleHelenavanVelthoven, rESEarCH THEME7:HEa Harry Campbell,ShamsEla a ziz Sheikh,MarkTomlinson,JoyELawn,Zulfiqara Harish Nair, udan, SachiyoYoshida,KitYeeChan,DeviSridhar,KerriWazny,HarishNair, rifeen, r I method LTH TraNSITIONS obert EBlack,SimonCousens obert WScherpbier Bhutta,r I method:VII.a IN CHINa ajiv Bahl, Mickey Chopra, ajiv Bahl,MickeyChopra, review of the first reviewofthefirst Jackson
479 510 501 486 journal of global health www.jogh.org
Journal of Global Health (JoGH) is a peer-review general medical journal focusing on issues relevant with the Global Health Academy of the University of Edinburgh, while also maintaining collaboration with for global health. The mission of the journal is to serve the community of researchers, funding agencies, academic partners throughout the world. international organizations, policy-makers and other stakeholders by providing an independent critical The views expressed by the authors do not reflect the views of JoGH or its publisher. assessment of the key issues that dominated the global health community. Journal design: Snježana Engelman Džafić for LASERplus, Zagreb, Croatia JoGH publishes original articles, viewpoints, research protocols and other publication items, making Realisation: LASERplus, Zagreb, Croatia, www.laser-plus.hr them available in full text online as soon as they are ready for publication. The articles are collected in Creative Commons Licence: CC BY 4.0. two annual print issues, at the end of June and December. All submissions and editorial inquiries should Print ISSN: 2047-2978 be directed to editor[at]jogh.org. Online ISSN: 2047-2986 Publisher: Edinburgh University Global Health Society (EUGHS). The aim of the EUGHS is to inspire and JoGH editors and editorial board members are independent researchers based at academic institutions, educate individuals, groups, organizations and communities about global health issues. The society is thus being well placed to provide professional evaluation of the key topics and ongoing activities and affiliated to the Centre for Population Health Sciences of the University of Edinburgh, and it works closely programs.
Neonatal Health; Deputy Director of the Intensive Care Unit and of Epidemiology; Croatian Centre for Global Health; Department EDITORS-IN-CHIEF Clinical Director of the General Intensive Care Unit; The University of Public Health; The University of Split School of Medicine, Split, of Melbourne and Royal Children’s Hospital, Melbourne, Australia Croatia • Assistant Professor Tomislav Lauc – Assistant Professor Ana Marušić – Chair, Department of Research in • Professor Wei Wang – Professor of Postgraduate Medicine; Professor of Orthodontics and Dental Anthropology; Editor-in-Chief, Biomedicine and Health, The University of Split School of Medicine, School of Medical Sciences, Edith Cowan University, Perth, Australia South European Journal of Orthodontics and Dentofacial Research; Split, Croatia Department of Anthropology, School of Social Sciences and Professor Igor Rudan – Chair in International Health and Molecular EDITORIAL COUNCIL Humanities; University of Zagreb, Zagreb, Croatia • Professor Joy Medicine; Co-Director, WHO Collaborating Centre for Population E. Lawn – Director, Global Evidence and Policy, Saving Newborn Health Research and Training; Co-Director, Centre for Global Health Professor James Smith – Vice-Principal International; Professor Lives programme, Save the Children; Director, MARCH Centre; The Research, College of Medicine and Veterinary Medicine, The of African and Development Studies; School of Social and Political London School of Hygiene and Tropical Medicine, London, UK • University of Edinburgh, Edinburgh, Scotland, UK Science; The University of Edinburgh, Edinburgh, Scotland, UK • Clinical Assistant Professor James A. Litch – Director, Perinatal Professor Harry Campbell – Professor of Genetic Epidemiology Jake Broadhurst – Head, Global Academies Directorate; The Interventions Program; Global Alliance to Prevent Prematurity and and Public Health; Co-Director, WHO Collaborating Centre for University of Edinburgh, Edinburgh, Scotland, UK • Professor Stillbirth (GAPPS), Seattle Children's; Department of Global Health Population Health Research and Training; Co-Director, Centre Mickey Chopra – Honorary Professor of Global Health; Centre and Department of Epidemiology; University of Washington, Seattle, for Global Health Research, College of Medicine and Veterinary for Global Health Research, College of Medicine and Veterinary WA, USA • Dr Li Liu – Assistant Scientist; Division of Global Medicine, The University of Edinburgh, Edinburgh, Scotland, UK Medicine, The University of Edinburgh, Edinburgh, Scotland, UK; Disease Epidemiology and Control; The Johns Hopkins Center for Lead health Specialist, The World Bank, Washington, DC, USA • Global Health; Johns Hopkins University, Baltimore, MD, USA • Dr Elizabeth Grant – Director, Global Health Academy; Senior REGIONAL EDITORS Professor Azeem Majeed – Head, Department of Primary Care Lecturer in Global Health and Development; Programme Director, & Public Health; Imperial College London, UK • Professor Julie North America: Professor Robert E. Black – Director for the Distance Learning Masters Programmes; Centre for Population Meeks Gardner – Head, Caribbean Child Development Centre; Institute of International Programs; Johns Hopkins Bloomberg School Health Sciences; College of Medicine and Veterinary Medicine; Consortium for Social Research and Development; The University The University of Edinburgh, Edinburgh, Scotland, UK • Professor of Public Health; Johns Hopkins University, Baltimore, MD, USA • of the West Indies, Mona, Jamaica • Professor Katherine L. Devi Sridhar – Chair in Global Public Health; Centre for Global Professor Rifat Atun - Professor of Global Health Systems; Director O'Brien – Executive Director, The International Vaccine Access Health Research, College of Medicine and Veterinary Medicine, The Center (IVAC); Global Disease Epidemiology and Control, Center for of Global Health Systems Cluster; Harvard T. H. Chan School of Public University of Edinburgh, Edinburgh, Scotland, UK and Co-Director, Health, Harvard University, Cambridge, MA, USA Global Health; Johns Hopkins University Bloomberg School of Public Centre for AIDS Interdisciplinary Research (CAIRO), Blavatnik School Health, Johns Hopkins University, Baltimore, MD, USA • Associate South America: Professor Claudio F. Lanata – Nutritional of Government, Oxford University, Oxford, UK • Professor Sarah Professor Ozren Polašek – Associate Professor of Biostatistics; Research Institute; Lima, Peru • Professor Cesar G. Victora – Cunningham-Burley – Dean of Molecular, Genetic and Population Director, Croatian Centre for Global Health; Department of Public Professor of Epidemiology; Federal University of Pelotas; Pelotas, Health Sciences, Professor of Medical and Family Sociology, The Health; The University of Split School of Medicine, Split, Croatia Brazil University of Edinburgh, Edinburgh, Scotland, UK • Professor Aziz • Professor Carine Ronsmans – Professor of Epidemiology; Sheikh – Co-Director of the Centre of Medical Informatics and Europe: Professor Sue C. Welburn – Vice-Principal, Global Head, Department of Infectious Diseases Epidemiology, The London Professor of Primary Care Research and Development, The University School of Hygiene and Tropical Medicine, London, UK • Professor Journal of Global Health is published biannually by the Edinburgh University Global Health Society (EUGHS). Access; Professor, Medical and Veterinary Molecular Epidemiology; of Edinburgh, Edinburgh, Scotland, UK The University of Edinburgh, Edinburgh, Scotland, UK • Dr Evropi Craig E. Rubens – Affiliate Professor of Global Health and Affiliate The aim of the EUGHS is to inspire and educate individuals, groups, organizations and communities about global Theodoratou – Centre for Global Health Research, College of Professor of Pediatrics; Executive Director, Global Alliance to Prevent EDITORIAL BOARD health issues. Its key objectives are to provide a platform for University of Edinburgh students to share ideas and/ Medicine and Veterinary Medicine, The University of Edinburgh, Prematurity and Stillbirth (GAPPS); Seattle Children’s and University of Washington School of Medicine, Seattle, WA, USA Professor Samir or experiences in Global Health with each other; organize meetings and other events to raise awareness of Global Edinburgh, Scotland, UK Dr Davies Adeloye – Covenant University Ota Ogun State, K. Saha – Head, Department of Microbiology; Executive Director, Nigeria; and Centre for Global Health Research, The University of Health issues within the University and more widely; and to give students opportunities to present their work. The Africa: Professor Shabir A. Madhi – Executive Director, National The Child Health Research Foundation; Bangladesh Institute of Child Edinburgh, Edinburgh, Scotland, UK • Professor Abdullah H. Institute for Communicable Diseases; DST/NRF South African Health, Dhaka Shishu Hospital, Dhaka, Bangladesh • Professor society is affiliated to the Centre for Population Health Sciences of the University of Edinburgh, and it works close- Baqui – Director, International Center for Maternal and Newborn Research Chair in Vaccine Preventable Diseases; University of the Joshua A. Salomon – Professor of Global Health; Department of Witwatersrand, Johannesburg, South Africa • Professor Mark R. Health; The Johns Hopkins Center for Global Health; Johns Hopkins ly with the Global Health Academy of the University of Edinburgh. University, Baltimore, MD, USA • Dr Luciana Brondi – Centre Global Health and Population; Harvard School of Public Health and Tomlinson – Department of Psychology; Stellenbosch University, Harvard Global Health Institute; Harvard University, Boston, MA, USA Matieland, South Africa for Global Health Research, College of Medicine and Veterinary Medicine, The University of Edinburgh, Edinburgh, Scotland, UK • • Dr Giorgio Tamburlini – Research Director, Institute of Child Asia: Professor Zulfiqar A. Bhutta – The Noordin Noormahomed Dr W. Abdullah Brooks – Director, Kamalapur Field Site; Head, Health IRCCS Burlo Garofolo; Scientific Director, European School for Sheriff Endowed Professor and Founding Chair; Department of Infectious Diseases Unit; The Johns Hopkins Center for Global Health; Maternal, Newborn, Child and Adolescent Health; Director, WHO Paediatrics and Child Health; The Aga Khan University, Karachi, Johns Hopkins University, Baltimore, MD, USA Professor Graham Collaborating Centre on Maternal and Child Health; The Institute Pakistan • Professor Shams El Arifeen – Professor of Brown – Foundation Director; Nossal Institute for Global Health; Burlo Garofolo, Trieste, Italy • Assistant Professor Alexander C. Subscription (including postage): GBP 30 per any single issue, GBP 50 for annual subscription, Epidemiology; Director, Centre for Child and Adolescent Health; The University of Melbourne, Melbourne, Australia • Ruth M. Tsai – Assistant Professor of Psychiatry; Department of Psychiatry; International Centre for Diarrhoeal Disease Research, Bangladesh Campbell – Honorary Associate/Researcher; Harvard Global Equity Massachusetts General Hospital and Harvard Medical School; Boston, GBP 25 for students, GBP 20 for EUGHS members. Please contact JoGH office for details. (ICDDR, B); James P. Grant School of Public Health, BRAC University, Initiative, Cambridge, MA, USA • Associate Professor Josip Car USA • Professor Peter Tugwell – Professor of Medicine and Dhaka, Bangladesh – Director of Health Services and Outcomes Research Programme; Epidemiology & Community Medicine; co-Director, WHO Collaborating Contact for subscription information: Lee Kong Chian School of Medicine; Imperial College &Nanyang Centre for Knowledge Translation and Health Technology Assessment China: Professor Yan Guo – Professor of Public Health; Vice Technological University, Singapore and Director, Global eHealth in Health Equity; Director, Centre for Global Health; Institute of Journal of Global Health President, Peking University Health Science Center; Vice Chairman Unit, School of Public Health, The Imperial College London, London, Population Health and Department of Medicine; Faculty of Medicine, Rachael Atherton of the Chinese Rural Health Association; Vice Director of the China UK • Professor Simon Cousens – Professor of Epidemiology The University of Ottawa, Ottawa, Canada • Associate Professor Academy of Health Policy (PKU); School of Public Health, Peking and Medical Statistics; Director of Research Degrees; Department Anita K. Zaidi – Director of Research, Department of Paediatrics Edinburgh University Global Health Society University,Beijing, China • Dr Kit Yee Chan – Visiting Scholar, of Infectious Disease Epidemiology; London School of Hygiene and and Microbiology; The Aga Khan University, Karachi, Pakistan and School of Public Health, Peking University Health Science Center; Tropical Medicine, London, UK • Professor Karen Edmond – Director, Enteric and Diarrheal Diseases, Bill and Melinda Gates Centre for Global Health Research Peking University, Beijing, China and Reader, Center for Population Winthrop Professor; School of Paediatrics and Child Health; The Foundation, Seattle, WA, USA • Associate Professor Lina Zgaga College of Medicine and Veterinary Medicine Health Sciences; College of Medicine and Veterinary Medicine; The University of Western Australia, Perth, Australia • Dr Christa L. – Associate Professor of Epidemiology; Department of Public Health University of Edinburgh, Scotland, UK Fischer Walker – Associate Scientist; Division of Global Disease and Primary Care; Trinity College Centre for Health Sciences, Dublin, The University of Edinburgh India: Professor K. Srinath Reddy – President, Public Health Epidemiology and Control; The Johns Hopkins Center for Global Ireland • Professor Stanley H. Zlotkin – Professor, Departments Teviot Place Foundation of India; President, World Heart Federation; Chair, Core Health; Johns Hopkins University, Baltimore, MD, USA • Professor of Paediatrics, Nutritional Sciences and Dalla Lana School of Public Edinburgh EH8 9AG, Scotland, UK Advisory Group on Health and Human Rights, National Human Rights Wendy J. Graham – Chair in Obstetrics and Gynaecology; The HealthVice President, Medical and Academic Affairs, Department Commission, India; President, National Board of Examinations, Institute of Applied Sciences; The University of Aberdeen, Aberdeen, of Pediatrics and Surgery, Division of Clinical Medicine; The Hospital Phone +44 (0) 131 651 5142 Fax +44 (0) 131 650 6909 India; Bernard Lown Visiting Professor of Cardiovascular Health, UK • Dr David B. Hipgrave – Honorary Senior Researcher; Nossal for Sick Children, The University of Toronto, Toronto, Canada • Dr Henry Kalter – Institute of International Programs; Johns Hopkins [email protected] Harvard School of Public Health, Boston, MA, USA • Dr Harish Institute of Global Health; University of Melbourne, Melbourne, Bloomberg School of Public Health; Johns Hopkins University, Nair – Public Health Foundation of India; Indian Institute of Public Australia and UNICEF, New York, NY, USA • Dr Hope L. Johnson – Center for Global Health; Johns Hopkins University Bloomberg Baltimore, MD, USA Health, New Delhi, India; Centre for Global Health Research, College School of Public Health, Johns Hopkins University, Baltimore, MD, of Medicine and Veterinary Medicine, The University of Edinburgh, USA and Head, Programme Outcomes and Impact, GAVI Alliance, MANAGING EDITOR/NEWS SECTION EDITOR Edinburgh, Scotland, UK Geneva, Switzerland • Professor Keith P. Klugman – William H. Australia and Western Pacific: Professor Trevor Duke Foege Chair of Global Health; Hubert Department of Global Health, Rachael Atherton – Centre for Global Health Research; College Journal design: Snježana Engelman Džafic´ for LASERplus, Zagreb, Croatia – Director, Centre for International Child Health; Director, WHO Rollins School of Public Health; The Emory Universtiy, Atlanta, GA, of Medicine and Veterinary Medicine, The University of Edinburgh, Collaborating Centre for Research and Training in Child and USA • Assistant Professor Ivana Kolčić – Assistant Professor Edinburgh, Scotland, UK Realisation: LASERplus, Zagreb, Croatia, www.laser-plus.hr PAPERS VIEWPOINTS NEWS EDITORIAL Resources Agencies Regions sustainable developmentgoalsforhealth Community–based primaryhealthcare:acorestrategyforachieving Zulfiqar ABhutta Journal ofGlobalHealth:TheMissionStatement Sagar Dugani,ZulfiqarA.Bhutta,NiranjanKissoon development inThailand Surveillance ofantimicrobialconsumption:methodologicalreviewforsystems on behalfoftheThaiSACWorkingGroup Nithima Sumpradit,RungpetchSakulbumrungsil,SasiJaroenpoj,VaravootSermsinsiri; Viroj Tangcharoensathien,AngkanaSommanustweechai,BoonratChanthong, Men’s health:timeforanewapproachtopolicyandpractice? Peter Baker,TimShand Zika: informationinthenickoftime Jessica LWalker,JamesHConway,ESvenson ultrasound restriction Stopping femalefeticideinIndia:thefailureandunintendedconsequenceof Sheida Tabaie Lessons fromBrazil:onthedifficultiesofbuildingauniversalhealthcaresystem Valbona Muzaka performance? Does SDG3haveanadequatetheoryofchangeforimprovinghealthsystems Gabriel Seidman four parameterstoachieveindividualandcollectiveaccountability A newparadigmonhealthcareaccountabilitytoimprovethequalityofsystem: Michelangelo Casali Umberto Genovese,SaraDelSordo,GabriellaPravettoni,IgorMAkulin,RiccardoZoja, communicable diseaseinterventionsinKenya Research forActionablePolicies:implementationscienceprioritiestoscaleupnon– von Rège,DavidWata,PamWilliams,GeraldYonga;ParticipantsfromtheNCDSymposiuminKenya Elijah Ogola,CarolOlale,DeborahOlwal–Modi,RoseRao,SarasRosin,OnyangoSangoro,Daniel Walter Mwanda,DanielMwai,JuliusMwangi,EstherMunyoro,ZacharyMuriuki,JamesNjoroge, James Kayima,AlfredKaragu,DorcasKiptui,AnneKorir,NkathaMeme,BredaMunoz, Robai Gakunga,GladwellGathecha,RainerHilscher,MuhammadJamiHusain,LydiaKaduka, Amuyunzu–Nyamongo, GiselaAbbam,NaftaliBusakhala,AbigailChakava,JonathanDick, Sujha Subramanian,JosephKibachio,SonjaHoover,PatrickEdwards,EvansAmukoye,Mary appropriate casemanagement ofchildhoodfeverinMozambique Contribution ofcommunity health workerstoimprovingaccesstimelyand Jeanne Koepsell,IbadulHaque Khan,AgbessiAmouzou Tanya Guenther,SalimSadruddin,KarenFinnegan,EricaWetzler , FatimaIbo,PauloRapaz, year estimates Economics in“GlobalHealth2035”:asensitivityanalysis ofthevaluealife Angela YChang,LisaARobinson,JamesKHammitt,Stephen C Resch Reducing maternalmortalityinsub–SaharanAfrica:therole ofethicalconsumerism Dileep Wijeratne,AndrewDavidWeeks Empowering peopleforsustainabledevelopment:theOttawa Charterandbeyond (continued ontheinside)
21 12 66 61 58 55 50 43 39 30 88 80 76 72 4 2 1
journal of global health June, 2017 Vol. 7 No. 1 PART TWO ISSN 2047-2978 Vol. 7No.1/2 June 2017 global journal of health