BMJ Open BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from

Can training non-physician clinicians/associate clinicians (NPCs/ACs) in emergency obstetric, neonatal care and clinical leadership make a difference to practice and help towards reductions in maternal and neonatal mortality, in rural ? The ETATMBA Project. For peer review only

Journal: BMJ Open

Manuscript ID: bmjopen-2015-008999

Article Type: Research

Date Submitted by the Author: 05-Jun-2015

Complete List of Authors: Ellard, David; Warwick Medical School, Clinical trials Unit Shemdoe, Aloisia; Ifakara Health Institute, Mazuguni, Festo; Ifakara Health Institute, Mbaruku, Godfrey; Ifakara Health Institute, Davies, David; The University of Warwick, Educational Development & Research Team, Warwick Medical School Kihaile, Paul; Ifakara Health Institute, Pemba, Senga; Tanzanian Training centre for International Health, Bergström, Staffan; Division of Global Health (IHCAR), Karolinska Institute, Stockholm, Department of Public Health Sciences Nyamtema, Angelo; Tanzanian Training centre for International Health,

Mohamed, Hamed-Mahfoudh; Ifakara Health Institute, http://bmjopen.bmj.com/ O'Hare, Paul; University of Warwick, Warwick Medical School Group, The ETATMBA Study; University of Warwick, Warwick Medical School

Primary Subject Global health Heading:

Secondary Subject Heading: Obstetrics and gynaecology, Medical education and training

human resources, maternal mortality, Tanzania, Non-physician clinicians, on October 1, 2021 by guest. Protected copyright. Keywords: Associate clinicians, medical education and training

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 1 of 23 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 1 2 Can training non-physician clinicians/associate clinicians (NPCs/ACs) in 3 emergency obstetric, neonatal care and clinical leadership make a difference to 4 5 practice and help towards reductions in maternal and neonatal mortality, in rural 6 Tanzania? The ETATMBA Project. 7 8 David R Ellard* PhD 9 Warwick Clinical Trials Unit, Division of Health Sciences, Warwick Medical School, The University 10 of Warwick, Coventry, CV4 7AL, UK 11 [email protected] 12 13 Aloisia Shemdoe,MSc 14 Ifakara Health Institute, Dar es Salaam, Tanzania 15 [email protected] peer review only 16 17 Festo Mazuguni, BSc 18 Ifakara Health Institute, Dar es Salaam, Tanzania 19 [email protected] 20 21 Godfrey Mbaruku, PhD 22 Ifakara Health Institute, Dar es Salaam, Tanzania 23 [email protected] 24 25 David Davies, PhD 26 Educational Development & Research Team, Warwick Medical School, The University of Warwick, 27 Coventry, CV4 7AL, UK 28 [email protected] 29 30 Paul Kihaile, MD,PhD 31 Ifakara Health Institute, Dar es Salaam, Tanzania 32 [email protected] 33

34

Senga Pemba, PhD http://bmjopen.bmj.com/ 35 Tanzanian Training centre for International Health 36 Ifakara, Tanzania 37 [email protected] 38

39 40 Staffan Bergström, MD,PhD 41 Department of Public Health Sciences, Karolinska Institutet, Sweden 42 [email protected]

43 on October 1, 2021 by guest. Protected copyright. 44 Angelo Nyamtema, MD,PhD 45 Tanzania Training centre for International Health 46 Ifakara, Tanzania 47 [email protected] 48 49 Hamed-Mahfoudh Mohamed, MD 50 Ifakara Health Institute, Dar es Salaam, Tanzania 51 [email protected] 52 53 Joseph Paul O'Hare, MD 54 Division of Metabolic & Vascular Health, Warwick Medical School, The University of Warwick, 55 Coventry, CV4 7AL, UK 56 [email protected] 57 58 On behalf of The ETATMBA Study Group 59 *corresponding author 60 1 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 2 of 23 BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 1 2 Abstract 3 4 Objectives 5 6 The enhancing human resources and the use of appropriate technologies for maternal and perinatal 7 8 survival in sub-Saharan Africa (ETATMBA) project is training Non-Physician Clinicians (NPCs)/ 9 10 Associate Clinicians (ACs) as advanced clinical leaders in emergency obstetric care in Tanzania and 11 12 13 Malawi. The main aim of this study in Tanzania was to evaluate the effect of the ETATMBA training 14 15 on health outcomesFor including peer maternal and neonatalreview morbidity and mortalityonly in the facilities where 16 17 trainees were based. 18 19 Design 20 21 A mixed methods study was carried out in health facilities in rural Tanzania where trainees were 22 23 based. The study included a before and after examination of key mortality indicators, and a survey of 24 25 infrastructure and the availability of equipment, supplies, and drugs. 26 27 Participants 28 29 30 During late 2010 and 2011 approximately 18 pairs (36 trainees) of assistant medical officers (AMOs) 31 32 and nurse midwives/nurses (anaesthesia) were recruited from districts across Tanzania and invited to 33 34 join the ETATMBA training programme. 35 http://bmjopen.bmj.com/ 36 37 Results 38 39 36 trainees completed the training, thereafter returning to 17 facilities. Maternal deaths show a non- 40 41 significant downward trend over the two years (282 to 232 cases per 100,000 live births). There were 42

43 no significant differences for any of the key maternal, neonatal and birth complication variables on October 1, 2021 by guest. Protected copyright. 44 45 across the lifetime of the project. The survey of facilities revealed that most have important shortages 46 47 in some areas, while some had serious shortages. 48 49 Conclusion 50 51 This study demonstrates that enhancing the knowledge and skills of NPCs/ACs in maternal and 52 53 neonatal care and in clinical leadership is possible in Tanzania, with the new skills and knowledge 54 55 being applied into practice. There were significant challenges not least the need for more investment 56 57 in infrastructure. 58 59 60 2 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 3 of 23 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 1 2 Key words: Non-physician clinicians, Associate clinicians, maternal mortality, training, medical 3 4 education, human resources, Tanzania. 5 6 Strengths and limitations of the study 7 8 9 • This is one of the first study taking an in-depth look at the impact on health outcomes, in 10 11 districts across rural Tanzania, of a programme of knowledge, skills and clinical leadership 12 13 training for non-physician clinicians/Associate clinicians (NPCs, ACs); 14 15 • LookingFor to see if thepeer up-skilling of thisreview important cadre of healthonly workers can impact on 16 17 district maternal and neonatal mortality ratios and key obstetric and birth complications; 18 19 • The study highlighted a number of challenges in delivering the training and in working in 20 21 22 rural Tanzania; 23

24 • This cadre is an important component in helping relieve the chronic shortages of trained 25 26 medical professionals in sub-Saharan Africa and helping countries move towards realisation 27 28 of millennium development goals. Further evaluations of the up-skilling of this cadre are 29 30 needed. 31 32

33 34 35 http://bmjopen.bmj.com/ 36 37 38 39 40 41 42

43 on October 1, 2021 by guest. Protected copyright. 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 3 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 4 of 23 BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 1 2 Background 3 4 In 2013 it was estimated that there was a global shortage of 7.2 million health-care workers, and that 5 6 by 2035 this is expected to rise to 12.9 million. [1] A recent review of global surgery, obstetric, and 7 8 anaesthesia workforce literature highlights the crisis. Countries like Tanzania only have a physician 9 10 density of 1 per 100,000 people. [2] It is estimated that currently there is a shortage of one million 11 12 healthcare workers in sub-Saharan Africa.[3] This shortage is partly because not enough people are 13 14 appropriately trained but is aggravated by meagre salaries, poor working conditions, low morale, 15 For peer review only 16 inadequate remuneration and few opportunities for continuous professional development. [4] Even 17 18 with a proliferation of new medical and nursing schools in recent years, the rise is not proportional to 19 20 the existing large populations. [5] For those working in rural areas there is professional isolation, 21 22 inadequate communication with peers and consultants in the cities, and a lack of appropriate 23 24 equipment and technologies. [3] 25 26 27 28 In Tanzania, the lack of basic items in many health facilities has been hindering timely and 29 30 appropriate quality obstetric and neonatal care, particularly in rural and remote health facilities. A 31 32 number of studies conducted in the country have also indicated that poor quality of care has been 33 34

experienced at health facilities due to the lack of an enabling environment (drugs, equipment, and http://bmjopen.bmj.com/ 35 36 supplies), poor skills of providers or hostile attitudes of providers, and a lack of trained staff. [6-9] 37 38 Many African countries have a cadre of mid-level health workers hitherto called Non Physician 39 40 Clinicians (NPCs), now more usually referred to as Associate Clinicians (ACs). In Tanzania this cadre 41 42 is often referred to as assistant medical officers (AMOs) or clinical officers (COs). These workers are 43 on October 1, 2021 by guest. Protected copyright. 44 trained by both government and non-government organisations and are often the most experienced 45 46 47 health workers in hospitals and health centres across the country, particularly away from urban 48 49 centres.[10] In Tanzania, all of these AMOs/COs are trained in emergency obstetric care (EmOC) and 50 51 are in the frontline providing care for mothers and babies.[11] In rural areas where medical doctors 52 53 (MDs) are few in number, the use of AMOs/COs and nurse midwives (NMWs) has been identified as 54 55 a viable solution, as these groups can be trained during short course programmes to provide effective 56 57 comprehensive emergency obstetric care (CEmOC) services in remote health centres. Some of the 58 59 benefits of using AMOs/COs in CEmOC services include; reducing training and employment costs 60 4 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 5 of 23 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 1 2 and enhancing retention within local health systems. Studies have shown that unlike MDs, 3 4 AMOs/COs remain in rural areas and continue working there. [12] 5 6 7 8 Major surveys consistently show that extra training and support can achieve task shifting and improve 9 10 maternal and foetal mortality and morbidity in the areas where these schemes have been piloted. [11, 11 12 13, 14] Training skilled attendants to prevent, detect, and manage major obstetric complications 13 14 including undertaking emergency caesarean surgery in complicated deliveries is arguably the single 15 For peer review only 16 most important factor in preventing maternal deaths and protecting the human rights of women. [11, 17 18 13-15] To be effective AMOs/COs need appropriate knowledge, skills, equipment, drugs and 19 20 technology essential for managing obstetric complications in rural or deprived communities. 21 22 23 24 The overall aim of enhancing human resources and use of appropriate technologies for maternal and 25 26 perinatal survival in Sub-Saharan Africa (ETATMBA) project was to develop, implement and 27 28 evaluate a programme of locally based clinical service improvement including clinical guidelines and 29 30 pathways, workforce development through structured education and leadership training.[16, 17] This 31 32 was linked to specialist on-site support and mentoring. The clinical service improvement involved 33 34

implementing best existing practice, and providing the context for understanding the additional health http://bmjopen.bmj.com/ 35 36 gain from the use of appropriate available technologies designed to reduce morbidity-specific 37 38 maternal case-fatality rates and fresh stillbirth rates (intra-partum foetal mortality) across different 39 40 African communities (Malawi and Tanzania). [18] 41 42

43 on October 1, 2021 by guest. Protected copyright. 44 The main objective of this study in Tanzania, was to evaluate the effect of the ETATMBA training on 45 46 47 health outcomes including maternal and neonatal morbidity and mortality in the facilities where the 48 49 trainees were based. 50 51 52 53 Methods 54 55 Design 56 57 The study used mixed methods with two approaches. Firstly, a within-facility before and after 58 59 comparison of the maternal, neonatal and perinatal mortality rates and obstetric complications for the 60 5 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 6 of 23 BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 1 2 year before the CEmOC training compared to rates for the year after the training. Secondly, a survey 3 4 of all facilities where one or more ETATMBA trainees were based to include leadership capacity, 5 6 infrastructure, and the availability of equipment, supplies, and drugs. 7 8 9 10 Outcomes 11 12 The primary outcomes for the study were early neonatal mortality (only including deaths that occur 13 14 before discharge) and maternal mortality (case specific) in the district. 15 For peer review only 16 Secondary outcomes include: 17 18 • Recorded data: obstetric complications (e.g. stillbirths, post-partum haemorrhage, caesarean 19 20 section, eclampsia, sepsis); 21 22 • Availability of resources (e.g. infrastructure, equipment, supplies and drugs) and leadership 23 24 capacity. 25 26 27 28 Research team 29 30 The primary data collection team consisted of two local research assistants based at the Ifakara Health 31 32 Institute (IHI), Dar es Salaam, Tanzania. Both of the research assistants are experienced researchers. 33 34

The principal investigator at the IHI gave local support, with management/oversight provided by DE http://bmjopen.bmj.com/ 35 36 at Warwick. 37 38

39 40 41 Participants 42

43 During late 2010 and 2011 approximately 18 pairs (36 trainees) of assistant medical officers (AMOs) on October 1, 2021 by guest. Protected copyright. 44 45 and nurse midwives/nurses (anaesthesia) were recruited from districts across Tanzania and invited to 46 47 join the ETATMBA training programme.[18] 48 49 50 51 Procedure 52 53 The research assistants identified the facilities in which trainees were working and extracted the study 54 55 variables from the IHI database to create an ETATMBA database. Baseline data was data for the 56 57 facility for the year 2011 (collected at the outset of the ETATMBA study by research associates from 58 59 the IHI). The follow-up data were the same variables for the year 2013. The follow-up data was 60 6 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 7 of 23 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 1 2 gathered during the ‘grand tour’. A survey instrument was used to carry out the facility survey. This 3 4 covered equipment, supplies and infrastructure, and recorded whether there was a sufficient 5 6 supply/number of the listed items for the facility’s daily caseload of deliveries, and whether the items 7 8 had been available and functional, available but NOT functional, or not available. 9 10 11 12 Essential drugs: Recording the availability and supply of drugs for each room (emergency room, 13 14 labour / delivery room, maternity ward, operating theatre and pharmacy). Checks were done to 15 For peer review only 16 confirm whether the listed drug was available and if the supply was sufficient to last for less than one 17 18 week, up to one week, up to two weeks, up to three weeks, or up to four or more weeks. 19 20 21 22 Data analysis 23 24 Descriptive and summary statistics were produced for the two years, change scores were produced, 25 26 and appropriate paired statistics carried out. Significance is set at 5%. Data are presented in tables and 27 28 graphs as appropriate. 29 30 31 32 Ethical approval 33 34

The study was reviewed and approved by the Biomedical Research Ethics Committee (BREC) at the http://bmjopen.bmj.com/ 35 36 University of Warwick, UK (REGO-2013-572) and The National Institute for Medical Research, 37 38 Institutional review board, Dar es Salaam, Tanzania (no.35). 39 40

41 42 Results 43 on October 1, 2021 by guest. Protected copyright. 44 Thirty-six received the ETATMBA training including 18 assistant medical officers (AMOs) and 18 45 46 47 nurse midwives (NMW). During the project period one AMO and one NMW left the programme to 48 49 pursue other interests and one NMW died. Thus attrition at the end of the programme was around 8%. 50 51 52 53 The trainees were based in health centres and district hospitals across Tanzania, with some in urban 54 55 areas but most in remote or very remote areas. The plan was to recruit trainees from health facilities 56 57 that were due to be upgraded with a theatre and maternity ward including equipment and resources so 58 59 that trainees could implement their new skills and knowledge. However, the reality was that of the 33 60 7 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 8 of 23 BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 1 2 trainees who completed the programme only 19 returned to the place from where they were selected 3 4 and seven of these returned to facilities that had not been upgraded or where upgrading was still in 5 6 process. Fourteen trainees did not return to the facility from which they were recruited because the 7 8 facilities had not been upgraded. Of these 10/14 were returned to district hospitals in the area they had 9 10 originally come from. Often these decisions were made by local District Medical Officers responding 11 12 to need and not to the strategic planning of the central Ministry. Table 1 gives an overview of where 13 14 the trainees were based and the type of facility they worked in. This also notes the availability of an 15 For peer review only 16 operating theatre in the facility as this is one of the key areas to be upgraded. Upgrading of facilities 17 18 was not part of the ETATMBA project but rather was on-going work with the Government and other 19 20 funding agencies. 21 22 23 24 25 26 27 28 Table 2summarises the key obstetric complication figures across Tanzania for 2011 and 2013 from 29 30 the included health facilities. It is important to note here that we were unable to collect our primary 31 32 outcome of neonatal mortality. Whilst stillbirths are recorded as a matter of course neonatal death is 33 34

not recorded at the facilities visited. http://bmjopen.bmj.com/ 35 36 37 38 No significant differences were found for any of the key obstetric complication variables across the 39 40 lifetime of the project. The number of deliveries seemed to decrease slightly overall (-604) but the 41 42 number of deliveries in health centres did rise (from 7326 to 7961). There is only a slight increase in 43 on October 1, 2021 by guest. Protected copyright. 44 overall fresh stillbirths (+16, an increase of 1 case per 1000 births) whilst macerated ones appear to 45 46 47 worsen in health centres (from 8.3 to 13.9 cases per 1000 live births). Maternal death shows a 48 49 downward improving trend over the two years (down from 282 to 232 cases per 100,000 live births), 50 51 but this is not statistically significant. There was a reduction in the number of caesarean sections 52 53 overall down from 80.2 to 77.2 (cases per 1000 live births) with the largest reduction indicated in 54 55 health centres where rates are down from 10.6 to 6.2 (cases per 1000 live births). The three birth 56 57 complication variables collected all show a slight increase overall but each shows a differing trend in 58 59 where the complications were reported. The rates of post-partum haemorrhage change little over time. 60 8 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 9 of 23 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 1 2 Obstructed labour rates increased in district hospitals (6.4 to 9.5 cases per 1000 live births), while in 3 4 health centres there was a decrease (6.7 to 2.9 cases per 1000 live births). Sepsis follows a similar 5 6 trend with an increase in hospitals (1.7 to 3.1 cases per 1000 live births) and a decrease in health 7 8 centres (1.6 to 0.5 cases per 1000 live births). This could imply earlier recognition and transfer of at- 9 10 risk mothers from health centres to hospitals as a result of training. 11 12 13 14 15 For peer review only 16 17 18 The following sections are the results related to infrastructure, drugs and supplies at the health 19 20 facilities where we have ETATMBA in Tanzania. These results originate from the survey undertaken 21 22 in early 2014 by IHI researchers. As noted in table 1 above there were 17 facilities across the country 23 24 that housed ETATMBA trainees during this survey. One of these facilities (due to its distance and 25 26 remoteness) was not visited. All results are based on 16 facilities, nine health-centres and seven 27 28 district hospitals. 29 30 31 32 Facilities: overall capacity and infrastructure 33 34 Running water and functioning toilets are a very significant problem with only one of nine health 35 http://bmjopen.bmj.com/ 36 centres and four of seven district hospitals found to have availability of running water and only just 37 38 over half of facilities a functioning toilet. Most facilities had sufficient access to lighting to perform 39 40 tasks at night but clearly some still struggle. Delivery beds were found to be available in 5/9 health 41 42 centres and 6/7 district hospitals. Ambulance availability was poor at health centres with only one 43 on October 1, 2021 by guest. Protected copyright. 44 having availability, whereas six of the seven district hospitals had an ambulance available. Referrals 45 46 from within the maternity area are problematic as only four health centres had a working (land line) 47 48 phone in this area and none of the district hospitals had. The availability of health related 49 50 registers/records is variable varying from 100% for items like the delivery register and 51 52 monthly/annual reports to 6% or less for the gynaecology register, patient records and discharge 53 54 registers. 55 56 57 58 59 60 9 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 10 of 23 BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 1 2 3 4 Drugs and equipment for normal delivery and infection prevention 5 6 Generally, supplies and equipment availability were good but there are a number of exceptions. Only 7 8 about 50% of facilities had needles and syringes available and similarly availability of suction and 9 10 vacuum extraction equipment was low. The availability of drugs for normal delivery purposes was 11 12 very variable with some drugs readily available (e.g. Lignocaine) whilst others had very low 13 14 availability (e.g. injectable antibiotic and Diazepam) (Table 4). 15 For peer review only 16 17 18 Infection prevention services in labour delivery/operating theatres 19 20 Overall only 75% or less of the facilities surveyed had the basics for infection prevention. None 21 22 seemed to have regular availability of soap for hand washing although antiseptics and bleach were 23 24 available and may be alternatives (Table 4). 25 26 27 28 29 30 31 32 Comprehensive services for provision of anaesthesia 33 34

Most of the district hospitals surveyed had availability of equipment and supplies for anaesthesia http://bmjopen.bmj.com/ 35 36 although Halothane is only available in 3/7 and less than 40% overall. Health centres seemed to lack 37 38 access to oxygen with only 2/9 having supplies when surveyed (Table 5). 39 40

41 42 Items for management of birth complications and caesarean section 43 on October 1, 2021 by guest. Protected copyright. 44 Overall, unsurprisingly, district hospitals had better availability of equipment, drugs and supplies for 45 46 47 managing birth complications and for performing caesarean sections (Table 5). 48 49 50 51 52 53 54 55 56 57 Discussion 58 59 60 10 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 11 of 23 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 1 2 The main objective of this study was to evaluate the effect of the ETATMBA training on health 3 4 outcomes including maternal and neonatal morbidity and mortality in the facilities where the trainees 5 6 were based. Changes in maternal and perinatal mortality was explored by comparing data from before 7 8 the training was implemented to data from a point not less than a year after the trainees completed 9 10 their training. We believe we have been successful in achieving these objectives. This being said, the 11 12 study has highlighted a number of challenges. 13 14 15 For peer review only 16 Neonatal mortality was one of our two primary endpoints in this study. However, it was found that 17 18 neonatal mortality is not recorded on Ministry of Health monthly summary sheets in facilities and thus 19 20 was not available for us. Whilst we had the number of stillbirths recorded, early neonatal deaths were 21 22 not. Our study has acted as a ‘wake-up call’ to the Ministry of Health and Social Welfare (MoHSW) 23 24 in Tanzania, who have now updated the current HIMS (Health Management Information System) to 25 26 ensure neonatal data are collected. 27 28 29 30 Interestingly, the number of actual births had decreased overall between 2011 and 2013. The 31 32 reduction was seen mostly at the district hospitals with numbers increasing at health centres. There 33 34

was a slight increase in fresh stillbirths but again most of this is at the district hospitals rather than at http://bmjopen.bmj.com/ 35 36 the health centres. This may suggest that health centres are referring more women with this problem 37 38 but the number of macerated stillbirths increased in both district hospitals and health centres, with the 39 40 latter being the biggest rise, so this may not be the case. 41 42

43 on October 1, 2021 by guest. Protected copyright. 44 Maternal deaths decreased. It is not a statistically significant reduction but there is a downward trend. 45 46 47 This could be simply a reflection of the reduction in maternal mortality reported in recent years across 48 49 Tanzania. A control group would have been necessary to be able to ascertain whether this was due to 50 51 the ETATMBA training. However, the concept of “maternal death audits”, which is being emphasized 52 53 by the MoHSW, is not systematically being undertaken. Whereas, in all ETATMBA districts, the 54 55 ETATMBA training has resulted in these audits being conducted in a more regular fashion with a lot 56 57 of confidence and appreciation. 58 59 60 11 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 12 of 23 BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 1 2 Looking at the three key birth complication data (post-partum haemorrhage, obstructed labour and 3 4 sepsis) all are seen to rise from 2011 to 2013 in both district hospitals and health centres with one 5 6 exception. Sepsis rates in health centres decreased, though great caution must be observed, since 7 8 registration of morbidities is often incomplete. Despite the increase in the numbers of mothers with 9 10 obstructed labour and postpartum haemorrhage maternal mortality ratio sat these facilities are falling. 11 12 Together with the qualitative data we believe this supports the view that the training and facility 13 14 upgrades have improved healthcare for these mothers. Once again we need to be cautious with only 15 For peer review only 16 before and after data as there is no control to detect temporal trends that might be occurring across 17 18 Tanzania. 19 20 21 22 The positive nature of the clinical results is surprising given the results of the survey of facilities in 23 24 early 2014 and how many facilities were not given the upgrade that was promised. The plan was to 25 26 provide upgraded facilities in remote areas with staff who had received ETATMBA training. Many 27 28 trainees had just returned to district hospitals or back to facilities that were not upgraded. While this 29 30 may seem not to have been the original aim, in reality, the district hospitals, which were themselves 31 32 greatly understaffed and in fact were the referral centres for outlying health centres, were glad to 33 34

receive these more “advanced leaders”. They covered the staff shortage and were found to work, http://bmjopen.bmj.com/ 35 36 mobilise, and align resources better and with more confidence. They were also instrumental in the 37 38 training of other staff. 39 40

41 42

43 on October 1, 2021 by guest. Protected copyright. 44 The cadres of AMOs and NMWs seem to be hard working and dedicated groups of individuals whose 45 46 47 efforts are still poorly rewarded with little or no housing provision in the poorer areas where health 48 49 needs are greatest. This and the poor basic fundamental infrastructure across the facilities with the 50 51 lack of running water and electricity supply constitute huge problems. Transport and communication 52 53 links are also poor with the lack of fuel and functioning vehicles, and few landline telephones making 54 55 it very difficult to move difficult cases to centres with specialist care. Often the most basic of 56 57 equipment or drugs were not available and infection prevention services were extremely poor. Basic 58 59 items like soap for hand washing were mostly absent. However, sepsis rates although rising slightly 60 12 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 13 of 23 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 1 2 overall were not significantly different to baseline (2011) levels, suggesting that despite enormous 3 4 challenges and lack of even basic supplies and equipment these clinicians manage to contain sepsis in 5 6 their facilities. 7 8 9 10 This study has a number of limitations not least that one of the primary outcomes was not available to 11 12 us. In addition, health centres were not being upgraded. The sample is small and with generally only 13 14 two trainees in each facility yet a large throughput of cases/births. We are not comparing our facilities 15 For peer review only 16 to control districts so it is difficult to attribute changes just to ETATMBA training. 17 18 19 20 Our findings support the early findings from this study with suggestions that the training had impact, 21 22 at the local level, on maternal mortality. [19] Delivery of ETATMBA training was challenging in 23 24 Tanzania, not least because of the huge distances involved and the effect this had on follow-up and 25 26 close support. What appears to have happened is that ETATMBA trainees became the stimulus to 27 28 upgrade more health centres at the district level in tandem with the MOH objective of upgrading at 29 30 least 50% of all health centres in a particular province to provide comprehensive EmOC. [20] We 31 32 were also able to influence the participants in understanding the importance of continuing medical 33 34

education by using current approaches such as electronic literature searches and exposure to e- http://bmjopen.bmj.com/ 35 36 learning modules, hence the demands for computers from them. The latter is a very powerful outcome 37 38 since due to the existing shortage of faculty and continuous medical education in these contexts this 39 40 has given us confidence that the graduates have gained tremendously in this exposure. This has been 41 42 explained by other researchers. [21] Participants also gained an understanding of the importance of 43 on October 1, 2021 by guest. Protected copyright. 44 locally generated data for planning instead of the past use of sending all the information to the 45 46 47 MoHSW. Lastly participants have shown a move to starting their own professional association. 48 49 Indeed, there is now a very active network called ANAC (African Network of Associate Clinicians). 50 51 This has enabled formation of a Community of Practice and has promoted the shift from the term 52 53 NPC to the more dignifying and respectful AC. 54 55 56 57 58 59 60 13 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 14 of 23 BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 1 2 Comparing our results with those from the Malawi of this project we see a strong indication that our 3 4 training can make a difference. [22] There are similarities and differences between this study and that 5 6 carried out in Malawi but in both countries it seems that overall the outcomes have been very positive. 7 8 9 10 The ETATMBA training was successfully implemented and appears to have been taken up and 11 12 applied in practice. We do, however, interpret these results with an air of caution, presenting exactly 13 14 what we found. There are trends in the data, which suggest an improving picture. However it seems 15 For peer review only 16 that the full impact of the training at a community level does not as yet show in the results. We 17 18 believe that the dedication shown by the trainees coupled with their new skills and knowledge will 19 20 have a positive impact over the coming years as more health centres are upgraded. 21 22 23 24 Competing Interests: None declared 25 26 27 28 Author contributions 29 30 DE, JPOH, GM, SB, and SP were involved in conception and design of the study. DE drafted the 31 32 manuscript supported by all authors. JPOH, GM, SB, SP and DD were responsible for the design of 33 34

the training. GM, SB, SP, PK, AN, HMM and DD were responsible for the management and delivery http://bmjopen.bmj.com/ 35 36 of the training. AS and FM, carried out the fieldwork and collated results supervised by DE. 37 38 39 40 Funding and Acknowledgements 41 42 Enhancing Human Resources and Use of Appropriate Technologies for Maternal and Perinatal 43 on October 1, 2021 by guest. Protected copyright. 44 Survival in Sub-Saharan Africa (ETATMBA) is a collaborative project funded by the European 45 46 47 Union, Seventh Framework Programme THEME [HEALTH.2010.3.4-2] [Project no. 266290]. This 48 49 study was embedded within this programme of work. All authors are part of the ETATMBA team. 50 51 The ETATMBA team would like to thank all of the AMOs and the district medical and nursing 52 53 officers for their hard work and support. This project benefited from facilities funded through 54 55 Birmingham Science City Translational Medicine Clinical Research and Infrastructure Trials 56 57 Platform, with support from Advantage West Midlands 58 59 60 14 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 15 of 23 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 1 2 The ETATMBA Study Group 3 Malawi University of Malawi College Sweden Karolinska Institutet, Sweden 4 of Medicine Staffan Bergström 5 Francis Kamwendo 6 Chisale Mhango 7 WanangwaChimwaza United GE Healthcare 8 ChikayikoChiwandira Kingdom Alan Davies 9 Queen Dube 10 The University of Warwick, UK 11 Ministry of Health, Malawi Paul O'Hare 12 Fannie Kachale Siobhan Quenby 13 ChimwemweMvula Douglas Simkiss 14 David Davies 15 Tanzania ForIfakara Health peer Institute, reviewDavid only Ellard 16 Tanzania Frances Griffiths 17 Godfrey Mbaruku Ngianga-bakwin, Kandala 18 Paul Kihaile Anne-Marie Brennan 19 Hamed Mohamed Edward Peile 20 Aloisia Shemdoe Anne-Marie Slowther 21 FestoMazuguni SaliyaChipwete 22 Tanzanian Training Centre Paul Beeby 23 for International Health Gregory Eloundou 24 Senga Pemba Harry Gee 25 Sidney Ndeki Vinod Patel 26 Angelo Nyamtema 27

28 29 30 31 32 33 34 35 http://bmjopen.bmj.com/ 36 37 38 39 40 41 42

43 on October 1, 2021 by guest. Protected copyright. 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 15 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 16 of 23 BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 1 2 REFERENCES 3 1. WHO, GHWA: A universal truth: no health without a workforce. In. Geneva: World 4 Health Organisation; 2014. 5 2. Hoyler M, Finlayson SG, McClain C, Meara J, Hagander L: Shortage of Doctors, Shortage 6 of Data: A Review of the Global Surgery, Obstetrics, and Anesthesia Workforce 7 Literature. World J Surg 2014, 38(2):269-280. 8 3. WHO: Human Resources for Health Observer. In. Geneva: World Health Organisation; 9 2012. 10 4. Anyangwe SC, Mtonga C: Inequities in the global health workforce: the greatest 11 impediment to health in sub-Saharan Africa. International journal of environmental 12 research and public health 4 13 2007, (2):93-100. 14 5. Kinfu Y, Dal Poz MR, Mercer H, Evans DB: The health worker shortage in Africa: are 15 enoughFor physicians peer and nurses being review trained? Bulletin of theonly World Health Organization 16 2009, 87(3):225-230. 17 6. Magoma M, Requejo J, Campbell OM, Cousens S, Filippi V: High ANC coverage and low 18 skilled attendance in a rural Tanzanian district: a case for implementing a birth plan 19 intervention. BMC pregnancy and childbirth 2010, 10:13. 20 7. MoHSW: Tanzania service availability and readiness assessment (SARA) 1012. In. Dar es 21 Salaam: Ifakara Health Institute; 2013. 22 8. Mselle LT, Moland KM, Mvungi A, Evjen-Olsen B, Kohi TW: Why give birth in health 23 facility? Users' and providers' accounts of poor quality of birth care in Tanzania. BMC 24 Health Serv Res 2013, 13:174. 25 9. Sarker M, Schmid G, Larsson E, Kirenga S, De Allegri M, Neuhann F, Mbunda T, Lekule I, 26 Muller O: Quality of antenatal care in rural southern Tanzania: a reality check. BMC 27 Res Notes 2010, 3:209. 28 10. Mullan F, Frehywot S: Non-physician clinicians in 47 sub-Saharan African countries. 29 Lancet 2007, 370(9605):2158-2163. 30 11. McCord C, Mbaruku G, Pereira C, Nzabuhakwa C, Bergstrom S: The quality of emergency 31 obstetrical surgery by assistant medical officers in Tanzanian district hospitals. Health 32 Aff (Millwood) 2009, 28(5):w876-885. 33 12. Nyamtema AS, Pemba SK, Mbaruku G, Rutasha FD, van Roosmalen J: Tanzanian lessons in 34 using non-physician clinicians to scale up comprehensive emergency obstetric care in 35 remote and rural areas. Human resources for health 2011, 9:28. http://bmjopen.bmj.com/ 36 13. Chilopora G, Pereira C, Kamwendo F, Chimbiri A, Malunga E, Bergstrom S: Postoperative 37 outcome of caesarean sections and other major emergency obstetric surgery by clinical 38 officers and medical officers in Malawi. Human resources for health 2007, 5:17. 39 14. Pereira C, Cumbi A, Malalane R, Vaz F, McCord C, Bacci A, Bergstrom S: Meeting the 40 need for emergency obstetric care in : work performance and histories of 41 medical doctors and assistant medical officers trained for surgery. BJOG : an 42 international journal of obstetrics and gynaecology 2007, 114(12):1530-1533. 43 15. Bergstrom S: Who will do the caesareans when there is no doctor? Finding creative on October 1, 2021 by guest. Protected copyright. 44 solutions to the human resource crisis. BJOG : an international journal of obstetrics and 45 gynaecology 2005, 112(9):1168-1169. 46 16. Davis D, Evans M, Jadad A, Perrier L, Rath D, Ryan D, Sibbald G, Straus S, Rappolt S, 47 Wowk M et al: The case for knowledge translation: shortening the journey from 48 evidence to effect. BMJ 2003, 327(7405):33-35. 49 17. McGlynn EA, Asch SM, Adams J, Keesey J, Hicks J, DeCristofaro A, Kerr EA: The quality 50 of health care delivered to adults in the United States. The New England journal of 51 medicine 2003, 348(26):2635-2645. 52 18. ETATMBA: Enhancing human resources and the use of appropriate technologies for 53 maternal and perinatal survival in sub-Saharan Africa 54 [http://www2.warwick.ac.uk/fac/med/about/global/etatmba/about/] 55 19. Kihaile P, Mbaruku G, Pemba S: Improved Maternal and Perinatal Mortalities by 56 Trained Medical Assistant Staffs in Rural Tanzania. J Health Med Informat 2013, 57 S11:007. 58 59 60 16 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 17 of 23 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 1 2 20. Ministry of Health and Social Welfare: Mpango wa Maendeleo wa Afya ya Msingi 3 (MMAM) 2007-2017 (Primary Health Services Development Programme,PHSDP). In. 4 United Republic of Tanzania: Ministry of Health and Social Welfare; 2007. 5 21. Frehywot S, Vovides Y, Talib Z, Mikhail N, Ross H, Wohltjen H, Bedada S, Korhumel K, 6 Koumare AK, Scott J: E-learning in medical education in resource constrained low- and 7 middle-income countries. Human resources for health 2013, 11:4. 8 22. Ellard DR, Chimwaza W, Davies D, O'Hare JP, Kamwendo F, Quenby S, Griffiths F: Can 9 training in advanced clinical skills in obstetrics, neonatal care and leadership, of non- 10 physician clinicians in Malawi impact on clinical services improvements (the 11 ETATMBA project): a process evaluation. BMJ Open 2014, 4(8). 12 13 14 15 For peer review only 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 http://bmjopen.bmj.com/ 36 37 38 39 40 41 42

43 on October 1, 2021 by guest. Protected copyright. 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 17 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 18 of 23 BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 1 2 Table 1. Health facilities where the Tanzanian ETATMBA trainees were based in 2013 3 District Name of facility Operating CEmOC or No. Trainees 4 5 Theatre BEmOC 6 1 Bukombe Bukombe District Hospital Yes CEmOC 1 AMO 7 8 2 Bukombe Uyovu Health Centre No BEmOC 1 AMO, 1CO 9 10 3 Geita Nzela Health Centre Yes CEmOC 1 NMW, 1 Nurse 11 4 Geita Katoro Health Centre No BEmOC 1 NMW 12 13 5 Inyonga Mamba Health Centre Yes CEmOC 1 NMW 14 15 6 KaramboFor Matai Healthpeer Centre reviewNo only BEmOC 1 AMO, 1NMW 16 7 Liwale Liwale District Hospital No CEmOC 2 AMOs 17 18 8 Meatu Mwandoya Health Centre No BEmOC 1 AMO, 1 NMW 19 20 9 Mpanda Mpanda District Hospital Yes BEmOC 1 AMO, 1 Nurse 21 22 10 Nachingwea Nachingwea District Yes CEmOC 2 AMOs 23 Hospital 24 11 Nkasi Kirando Health Centre Yes CEmOC 2 AMOs 25 26 12 Nyanghwale Nyanghwale Health Centre No BEmOC 1 AMO, 1 NMW 27 a 28 13 Nyanghwale Kharumwa District Hospital Yes CEmOC 1 AMO, 1 NMW 29 14 Ruangwa Ruangwa District Hospital Yes CEmOC 1 AMO, 1 NMW 30 31 15 Sumbawanga Laela Health Centre No BEmOC 1 AMO, 1 NMW 32 33 16 Chato Chato District Hospital Yes CEmOC 1 AMO, 1 NMW

34 b 35 17 Lindi Nyangao Mission Hospital unknown CEmOC 2 NMWs http://bmjopen.bmj.com/ 36 a Upgraded to a district hospital between 2011 &2013. b This hospital not visited so not included in in 37 analysis. AMO – Assistant medical officer, NMW – Nurse midwife, Nurse – nurse/anaesthetics. 38 39 CEmOC – Comprehensive Emergency Obstetric Care, BEmOC - Basic Emergency Obstetric Care 40 41 42

43 on October 1, 2021 by guest. Protected copyright. 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 18 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 19 of 23 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 1 2 Table 2. Comparison of key maternal, neonatal and birth complication figures from baseline (2011) to 3 follow-up (2013) 4 5 2011 2013

6 a b * 7 DH HC Total DH HC Total Difference b-a 8 (n=7) (n=9) (n=7) (n=9) 9 10 Total Deliveries 17893 7326 25219 16654 7961 24615 -604 11 Fresh stillbirths (FSB) 287 65 352 300 68 368 16·0 12 13 FSB per 1000 births 16·0 8·9 14·0 18·0 8·5 15·0 1·0 14 15 Macerated StillbirthsFor (MSB) peer 312 review 61 373 305 only 111 416 43·0 16 MSB per 1000 births 17·4 8·3 14·8 18·3 13·9 16·9 2·1 17 18 Maternal deaths (n) 68 3 71 55 2 57 -14·0 19 20 MD per 100,000 births 380 41 282 330 25 232 -50 21 Caesarean deliveries 1944 78 2022 1851 49 1900 -122 22 23 CS per 1000 births 108·6 10·6 80·2 111·1 6·2 77·2 -3·0 24 25 Post-partum Haemorrhage 200 77 277 225 86 311 34·0 26 PPH per 1000 births 11·2 10·5 11·0 13·5 10·8 12·6 1·7 27 28 Obstructed labour 114 49 163 159 23 182 19·0 29 30 Obst/Lab per 1000 births 6·4 6·7 6·5 9·5 2·9 7·4 0·9 31 Sepsis 31 12 43 51 4 55 12·0 32 33 Sepsis per 1000 births 1·7 1·6 1·7 3·1 0·5 2·2 0·5

34 * 35 DH – District hospitals, HC – Health centres· Note: there are NO significant differences here so p- http://bmjopen.bmj.com/ 36 values not shown. 37

38 39 40 41 42

43 on October 1, 2021 by guest. Protected copyright. 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 19 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 20 of 23 BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 1 2 3 Table 3. Survey findings from health facilities in Tanzania related to infrastructure 4 5 Facilities with the items 6 HC (%) DH (%) 7 Overall n=9 n=7 8 Health facility Infrastructure Availability of Power & Availability of water 9 Sufficient light source to perform tasks at night 12 (75%) 6 (67%) 6 (86%) 10 Means of ventilation 5 (31%) 1 (11%) 4 (57%) 11 12 Running water 5 (31%) 1 (11%) 4 (57%) 13 Functioning toilet 9 (56%) 6 (67%) 3 (43%) 14 Functional fan/air conditioning 5 (31%) 1 (11%) 4 (57%) 15 ForCurtains/means peer of providing review patient privacy only14 (88%) 9 (100%) 5 (71%) 16 17 Waiting area for visitors and family 6 (38%) 4 (43%) 2 (33%) 18 Facility with electricity 14 (89%) 8 (86%) 6 (86%) 19 Motor Vehicle Ambulance Available and functional 6 (38%) 1 (22%) 5 (71%) 20 Available and functional landline telephone in the maternity area 4 (25%) 4 (43%) 0 (0%) 21 Delivery bed / table 11 (69%) 5 (56%) 6 (86%) 22

23 24 Availability of health related registers 25 General admission register 5 (56%) 6 (86%) 11 (69%) 26 Delivery register 9 (100%) 7 (100%) 16 (100%) 27 Maternity ward register 4 (44%) 5 (71%) 9 (56%) 28 Female ward register 4 (44%) 5 (71%) 9 (56%) 29 30 Operating theatre register 4 (44%) 6 (86%) 10 (63%) 31 Gynaecology register 0 0 0 (0%) 32 Post-abortion register 4 (44%) 5 (71%) 9 (56%) 33 Individual patient records 0 1 (14%) 1 (6%) 34 35 Discharge register 0 1 (14%) 1 (6%) http://bmjopen.bmj.com/ 36 Death register 6 (67%) 5 (71%) 11 (69%) 37 Mortuary register 2 (22%) 5 (71%) 7 (44%) 38 Monthly / annual facility summary reports 9 (100%) 7 (100%) 16 (100%) 39 40

41 42 43 on October 1, 2021 by guest. Protected copyright. 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 20 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 21 of 23 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 1 2 Table 4.Survey findings from health facilities in Tanzania related to the availability of equipment, 3 supplies and drugs 4 5 facilities with the equipment 6 Overall (%) HC (%) DH (%) 7 N=16 N=9 N=7 8 Drugs and equipment: Availability of Items for normal delivery 9 Equipment and Supplies 10 Blood pressure cuff/machine 13 (81%) 7 (78%) 6 (86%) 11 12 Stethoscope 15 (94%) 8 (89%) 7 (100%) 13 Fetal stethoscope 16 (100%) 9 (100%) 7 (100%) 14 Clinical thermometer 13 (81%) 6 (67%) 7 (100%) 15 For peer reviewSterile gloves 16 (100%) only 9 (100%) 7 (100%) 16 Non-sterile protective clothing/apron 15 (94%) 8 (89%) 7 (100%) 17 18 Scissors or razor blade for cutting cord 15 (94%) 9 (100%) 6 (86%) 19 Cord ties 10 (63%) 5 (56%) 5 (71%) 20 Needles and Syringes 8 (50%) 4 (44%) 4 (57%) 21 IV fluid set (giving set) 15 (94%) 9 (100%) 6 (86%) 22 Suture needles and suture materials 10 (63%) 5 (56%) 5 (71%) 23 24 suction apparatus 8 (50%) 3 (33%) 5 (71%) 25 Manual vacuum extractor 5 (31%) 2 (33%) 2 (29%) 26 Obstetric forceps 11 (69%) 8 (89%) 3 (43%) 27 Drugs 28 29 Pitocin (Oxytocin) 13 (81%) 6 (67%) 7 (100%) 30 Ergometrine (injectable) 4 (25%) 3 (33%) 1 (14%) 31 Normal saline 14 (88%) 8 (89%) 6 (86%) 32 Ringers lactate 7 (44%) 2 (22%) 5 (71%) 33 Dextrose / glucose 9 (56%) 3 (33%) 6 (86%) 34 35 Lignocaine 2% or 1% 15 (94%) 8 (89%) 7 (100%) http://bmjopen.bmj.com/ 36 injectable antibiotic 5 (31%) 3 (33%) 2 (29%) 37 Magnesium sulphate 14 (88%) 8 (89%) 6 (86%) 38 Diazepam 6 (38%) 3 (33%) 3 (43%) 39 Skin disinfectant 12 (75%) 7 (78%) 5 (71%) 40 41 42 Availability of Infection prevention services in labour delivery/operating theatres

43 Decontamination container with prepared solution 11 (69%) 5 (56%) 6 (86%) on October 1, 2021 by guest. Protected copyright. 44 Covered contaminated trash bin 11 (69%) 6 (67%) 5 (71%) 45 Sharps container 12 (75%) 6 (67%) 6 (86%) 46 47 Soap 0 0 0 48 Antiseptics 10 (63%) 5 (56%) 5 (71%) 49 Chlorine/ Bleach 6 (38%) 2 (22%) 4 (57%) 50 Sterile gloves 12 (75%) 6 (67%) 6 (86%) 51 Other items 52 53 Regular trash bin 12 (75%) 6 (67%) 6 (86%) 54 Non sterile gloves 12 (75%) 6 (67%) 6 (86%) 55 Non-sterile protective clothing 12 (75%) 6 (67%) 6 (86%) 56 57 58 59 60 21 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 22 of 23 BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 1 2 Table 5.Survey findings from health facilities in Tanzania related to the availability of items 3 for management of anaesthesia, birth complications and caesarean section 4 5 Facilities with the items 6 Equipment Overall HC DH 7 N=16 N=9 N=7 8 Items for provision of anaesthesia 9 Suction machine 6 (38%) 4 (44%) 2 (29%) 10 Filled oxygen cylinder with cylinder carrier 8 (50%) 2 (22%) 6 (86%) 11 and key to open valve 12 Intubating forceps (Magill) 6 (38%) 4 (44%) 2 (29%) 13 14 Adult laryngoscope 11 (69%) 6 (67%) 5 (71%) 15 ForAdult ventilator peer bag and maskreview 11 (69%) 6only (67%) 5 (71%) 16 IV fluid set (giving set) 10 (63%) 5 (56%) 5 (71%) 17 Spinal needles (18-gauge to 25-gauge) 3 (19%) 1 (11%) 2 (29%) 18 Endotracheal tubes with cuffs (8 – 10mm) 9 (56%) 4 (44%) 5 (71%) 19 20 Halothane 6 (38%) 3 (33%) 3 (43%) 21 Ketamine 11 (6%) 5 (56%) 6 (86%) 22 Aesthetic face masks 9 (56%) 5 (56%) 4 (57%) 23 24 Items for management of pre-eclampsia/ 25 26 eclampsia 27 Magnesium Sulphate 7 (44%) 4 (44%) 3 (43%) 28 Diazepam-injectable 10 (63%) 4 (44%) 6 (86%) 29 Niphedipine 1 (6%) 0 (0%) 1 (14%) 30 Blood pressure cuff/machine 13 (81%) 7 (78%) 6 (86%) 31 Stethoscope 15 (94%) 8 (89%) 7 (100%) 32 33 Adult ventilator bag and mask 13 (81%) 7 (78%) 6 (86%) 34 Needles and Syringes 4 (25%) 1 (11%) 3 (43%) 35 Urinary catheters (Foleys) 8 (50%) 3 (33%) 5 (71%) http://bmjopen.bmj.com/ 36 Uristix 4 (25%) 1 (11%) 3 (43%) 37

38 39 Items for management of haemorrhage (parenteral uterotonics) 40 Needles and Syringes 8 (50%) 4 (44%) 4 (57%) 41 IV fluid set (giving set) 9 (56%) 3 (33%) 6 (86%) 42 43 on October 1, 2021 by guest. Protected copyright. 44 Items for Caesarean Section (not including anaesthesia) 45 Operating table 46 Light- adjustable, shadow less 11 (69%) 6 (56%) 5 (86%) 47 Antiseptics 10 (63%) 5 (56%) 5 (71%) 48 Sterile gloves 12 (75%) 6 (67%) 6 (86%) 49 50 Cord ties 10 (63%) 5 (56%) 5 (71%) 51 Needles and Syringes 6 (38%) 4 (44%) 2 (29%) 52 Benzyl Penicillin 4 (25%) 3 (33%) 1 (14%) 53 Metronidazole (IV) 2 (13%) 1 (11%) 1 (14%) 54 Gentamycin (IV) 1 (6%) 1 (11%) 0 (0%) 55 56 Caesarean section pack 57 needle holder 13 (81%) 7 (78%) 6 (86%) 58 scalpel handle with blade 10 (63%) 5 (56%) 5 (71%) 59 60 22 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 23 of 23 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 1 2 retractor 12 (75%) 6 (67%) 6 (86%) 3 surgical scissors 12 (75%) 6 (67%) 6 (86%) 4 suction apparatus / 8 6 (38%) 4 (44%) 2 (29%) 5 oxygen 8 (50%) 2 (22%) 6 (86%) 6 7 sutures 11 (69%) 5 (56%) 6 (86%) 8 ketamine 11 (69%) 5 (56%) 6 (86%) 9 lidocaine / 5 12 (75%) 6 (67%) 6 (86%) 10 11 12 13 14 15 For peer review only 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 http://bmjopen.bmj.com/ 36 37 38 39 40 41 42

43 on October 1, 2021 by guest. Protected copyright. 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 23 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from

Can training non-physician clinicians/associate clinicians (NPCs/ACs) in emergency obstetric, neonatal care and clinical leadership make a difference to practice and help towards reductions in maternal and neonatal mortality in rural Tanzania? The ETATMBA Project. For peer review only

Journal: BMJ Open

Manuscript ID bmjopen-2015-008999.R1

Article Type: Research

Date Submitted by the Author: 28-Aug-2015

Complete List of Authors: Ellard, David; Warwick Medical School, Clinical trials Unit Shemdoe, Aloisia; Ifakara Health Institute, Mazuguni, Festo; Ifakara Health Institute, Mbaruku, Godfrey; Ifakara Health Institute, Davies, David; The University of Warwick, Educational Development & Research Team, Warwick Medical School Kihaile, Paul; Ifakara Health Institute, Pemba, Senga; Tanzanian Training centre for International Health, Bergström, Staffan; Division of Global Health (IHCAR), Karolinska Institute, Stockholm, Department of Public Health Sciences Nyamtema, Angelo; Tanzanian Training centre for International Health,

Mohamed, Hamed-Mahfoudh; Ifakara Health Institute, http://bmjopen.bmj.com/ O'Hare, Paul; University of Warwick, Warwick Medical School Group, The ETATMBA Study; University of Warwick, Warwick Medical School

Primary Subject Global health Heading:

Secondary Subject Heading: Obstetrics and gynaecology, Medical education and training

human resources, maternal mortality, Tanzania, Non-physician clinicians, on October 1, 2021 by guest. Protected copyright. Keywords: Associate clinicians, medical education and training

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 1 of 66 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 1 2 1 Can training nonphysician clinicians/associate clinicians (NPCs/ACs) in 3 2 emergency obstetric, neonatal care and clinical leadership make a difference to 4 5 3 practice and help towards reductions in maternal and neonatal mortality in rural 6 4 Tanzania? The ETATMBA Project. 7 8 5 David R Ellard* PhD 9 6 Warwick Clinical Trials Unit, Division of Health Sciences, Warwick Medical School, The University 10 7 of Warwick, Coventry, CV4 7AL, UK 11 8 [email protected] 12 9 13 10 Aloisia Shemdoe, MSc 14 11 Ifakara Health Institute, Dar es Salaam, Tanzania 15 12 [email protected] peer review only 16 13 17 14 Festo Mazuguni, BSc 18 15 Ifakara Health Institute, Dar es Salaam, Tanzania 19 16 [email protected] 20 17 21 18 Godfrey Mbaruku, MD PhD 22 19 Ifakara Health Institute, Dar es Salaam, Tanzania 23 20 [email protected] 24 21 25 22 David Davies, PhD 26 23 Educational Development & Research Team, Warwick Medical School, The University of Warwick, 27 24 Coventry, CV4 7AL, UK 28 25 [email protected] 29 26 30 27 Paul Kihaile, MD, PhD 31 28 Ifakara Health Institute, Dar es Salaam, Tanzania 32 29 [email protected] 33 30 34

31 Senga Pemba, PhD http://bmjopen.bmj.com/ 35 32 Tanzanian Training centre for International Health 36 33 Ifakara, Tanzania 37 34 [email protected] 38 35 39 40 36 Staffan Bergström, MD, PhD 41 37 Department of Public Health Sciences, Karolinska Institutet, Sweden 42 38 [email protected]

43 39 on October 1, 2021 by guest. Protected copyright. 44 40 Angelo Nyamtema, MD, PhD 45 41 Tanzania Training centre for International Health 46 42 Ifakara, Tanzania 47 43 [email protected] 48 44 49 45 HamedMahfoudh Mohamed, MD 50 46 Ifakara Health Institute, Dar es Salaam, Tanzania 51 47 [email protected] 52 48 53 49 Joseph Paul O'Hare, MD 54 50 Division of Metabolic & Vascular Health, Warwick Medical School, The University of Warwick, 55 51 Coventry, CV4 7AL, UK 56 52 J.P.O[email protected] 57 53 58 54 On behalf of The ETATMBA Study Group 59 55 *corresponding author 60 1 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 2 of 66 BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 1 2 1 Abstract 3 4 2 Objectives 5 6 3 During 2011, 18 pairs of assistant medical officers and nurse midwives/nurses were recruited from 7 8 4 districts across rural Tanzania and invited to join the enhancing human resources and the use of 9 10 5 appropriate technologies for maternal and perinatal survival in subSaharan Africa (ETATMBA) 11 12 13 6 training programme. The ETATMBA project is training Associate Clinicians (ACs) as advanced 14 15 7 clinical leadersFor in emergency peer obstetric care. reviewThe trainees returned to healthonly facilities across the 16 17 8 country with the hope of being able to apply their new skills and knowledge. The aim of this study in 18 19 9 was to explore the impact of the ETATMBA training on health outcomes including maternal and 20 21 10 neonatal morbidity and mortality in their facilities. Secondly, to explore the challenges faced in 22 23 11 working in these health facilities. 24 25 12 Design 26 27 13 The study is a pre/post examination of maternal and neonatal health indicators and a survey of health 28 29 14 facilities in rural Tanzania. The facilities surveyed were those in which ETATMBA trainees were 30 31 15 placed posttraining. The maternal and neonatal indicators were collected for 2011 and 2013 the 32 33 16 survey of the facilities was early 2014. 34 35 http://bmjopen.bmj.com/ 36 17 Results 37 38 18 Sixteen of seventeen facilities were surveyed. Maternal deaths show a nonsignificant downward 39 40 19 trend over the two years (282 to 232 cases per 100,000 live births). There were no significant 41 42 20 differences in maternal, neonatal and birth complication variables across the timepoints. The survey

43 on October 1, 2021 by guest. Protected copyright. 44 21 of facilities revealed shortages in key areas and some of these shortages are a serious concern. 45 46 22 Conclusion 47 48 23 This study represents a snapshot of rural health facilities providing maternal and neonatal care in 49 50 24 Tanzania. Enhancing knowledge, practical skills, and clinical leadership of ACs may have a positive 51 52 25 impact on health outcomes. However, any impact may be confounded by the significant challenges in 53 54 26 delivering a service in terms of infrastructure and supplies. Thus training may be beneficial it requires 55 56 27 an environment that supports it. 57 58 59 28 60 2 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 3 of 66 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 1 2 1 Key words: Nonphysician clinicians, Associate clinicians, maternal mortality, training, medical 3 4 2 education, human resources, infrastructure, Tanzania. 5 6 3 7 8 9 4 Strengths and limitations of the study 10 11 12 5 • This is one of the first studies taking an indepth look at the impact on health outcomes in 13 14 6 districts across rural Tanzania, of a programme of knowledge, skills and clinical leadership 15 For peer review only 16 7 training for associate clinicians; 17 18 8 • This cadre is an important component in helping relieve the chronic shortages of trained 19 20 9 medical professionals in subSaharan Africa and helping countries move towards realisation 21 22 10 of millennium development goals; 23 24 11 • One of the primary outcomes (neonatal mortality) was found to be not recorded or poorly 25 26 12 recorded at health facilities at the time preventing us from reporting on this important 27 28 13 outcome; 29 30 14 • This was a before and after design and there was no control group on which to draw 31 32 15 comparisons; 33 34

16 A number of facilities where trainees were returned to posttraining were not upgraded, as http://bmjopen.bmj.com/ 35 • 36 37 17 planned, thus preventing them from putting into practice their new skills and knowledge. 38 39 40 41 42

43 on October 1, 2021 by guest. Protected copyright. 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 3 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 4 of 66 BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 1 2 1 Background 3 4 2 In 2013 it was estimated that there was a global shortage of 7.2 million healthcare workers, and that 5 6 3 by 2035 this is expected to rise to 12.9 million. [1] A recent review of global surgery, obstetric, and 7 8 4 anaesthesia workforce literature highlights the crisis. Countries like Tanzania only have a physician 9 10 5 density of 1 per 100,000 people. [2] It is estimated that currently there is a shortage of one million 11 12 6 healthcare workers in subSaharan Africa.[3] This shortage is partly because not enough people are 13 14 7 appropriately trained but is aggravated by meagre salaries, poor working conditions, low morale, 15 For peer review only 16 8 inadequate remuneration, and few opportunities for continuous professional development. [4] Even 17 18 9 with a proliferation of new medical and nursing schools in recent years, the rise is not proportional to 19 20 10 the existing large populations. [5] For those working in rural areas there is professional isolation, 21 22 11 inadequate communication with peers and consultants in the cities, and a lack of appropriate 23 24 12 equipment and technologies. [3] 25 26 13 27 28 14 In Tanzania, the lack of basic items in many health facilities has hindered timely and appropriate 29 30 15 quality obstetric and neonatal care, particularly in rural and remote health facilities. A number of 31 32 16 studies conducted in the country have also indicated that poor quality of care has been experienced at 33 34

17 health facilities due to the lack of an enabling environment (drugs, equipment, and supplies) [6], poor http://bmjopen.bmj.com/ 35 36 18 skills of providers or hostile attitudes of providers, and a lack of trained staff. [710] As part of the 37 38 19 solution many African countries have created a cadre of midlevel health workers called Non 39 40 20 Physician Clinicians (NPCs), now more usually referred to as Associate Clinicians (ACs). In Tanzania 41 42 21 this cadre is often referred to as Clinical Officers (COs) or Assistant Medical Officers (AMOs) (COs 43 on October 1, 2021 by guest. Protected copyright. 44 22 who have received some additional training). These workers are trained by both government and non 45 46 47 23 government institutions and are often the most experienced health workers in hospitals and health 48 49 24 centres across the country, particularly away from urban centres.[11] Moreover, all of these 50 51 25 AMOs/COs are trained in Emergency Obstetric Care (EmOC) and are in the frontline providing care 52 53 26 for mothers and babies.[12] In rural areas where medical doctors (MDs) are few in number, the use of 54 55 27 AMOs/COs and nurse midwives (NMWs) has been identified as a viable solution, as these groups can 56 57 28 be trained through short course programmes to provide effective Comprehensive Emergency Obstetric 58 59 29 Care (CEmOC) services in remote health centres. Key benefits of using AMOs/COs in CEmOC 60 4 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 5 of 66 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 1 2 1 services include; reducing training and employment costs, promoting task sharing/shifting and 3 4 2 enhancing retention within local health systems. Studies have shown that unlike MDs, AMOs/COs 5 6 3 remain in rural areas and continue working there. [13] 7 8 4 9 10 5 Major surveys consistently show that extra training and support can enhance task sharing/shifting and 11 12 6 reduce maternal and neonatal mortality and morbidity in the areas where these schemes have been 13 14 7 piloted. [12 14 15] Training skilled attendants to prevent, detect, and manage major obstetric 15 For peer review only 16 8 complications including undertaking emergency caesarean surgery in complicated deliveries is 17 18 9 arguably the single most important factor in preventing maternal deaths and protecting the human 19 20 10 rights of women. [12 1416] To be effective AMOs/COs need appropriate knowledge, skills, 21 22 11 equipment, drugs and technology essential for managing obstetric complications in rural or deprived 23 24 12 communities. 25 26 13 27 28 14 The aim of the Enhancing human resources and use of appropriate technologies for maternal and 29 30 15 perinatal survival in subSaharan Africa (ETATMBA) project was to develop, implement, and 31 32 16 evaluate a programme of locally based clinical service improvement including clinical guidelines and 33 34

17 pathways, workforce development through structured education, and leadership training.[17 18] This http://bmjopen.bmj.com/ 35 36 18 was linked to specialist onsite support and mentoring. 37 38 19 39 40 20 The ETATMBA project in Tanzania 41 42 21 The ETATMBA Project combined two main interventions: First, the training of ACs and nurses in 43 on October 1, 2021 by guest. Protected copyright. 44 22 CEmOC and anaesthesia. Secondly, post training mentoring and supervision of participants at their 45 46 47 23 working places. Within this project, the clinical service improvement involved implementing best 48 49 24 existing practice, linked to training in clinical leadership, and providing the context for understanding 50 51 25 the additional health gain from the use of appropriate available technologies designed to reduce 52 53 26 morbidityspecific maternal casefatality rates and fresh stillbirth rates (intrapartum foetal mortality) 54 55 27 across different African communities (Malawi and Tanzania). [19] 56 57 28 58 59 60 5 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 6 of 66 BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 1 2 1 During 2011, 18 pairs (36 trainees) of AMOs and nurse midwives/nurses (anaesthesia) were recruited 3 4 2 from rural districts across Tanzania and invited to join the ETATMBA training programme. The 5 6 3 ETATMBA training provided specific skills consisting of indepth theoretical reviews and 7 8 4 demonstrations on the prevention and management of the four major killers of mothers, notably 9 10 5 haemorrhage, sepsis, hypertension and unsafe abortion and the three most common causes of neonatal 11 12 6 death; asphyxia, treatment of maternal and neonatal sepsis, and hypothermia. Key components of the 13 14 7 training included leadership, advanced surgical skills for the management of emergency obstetric 15 For peer review only 16 8 complications, and the prevention of preterm labour/birth. The aim of the training was to empower 17 18 9 the trainee ACs and enhance their leadership skills, allowing them to think and to maximise the use of 19 20 10 their limited material resources to provide the best care possible for their patients and facilities. 21 22 11 Included in the training there were key components of clinical service improvement and values based 23 24 12 practice, which can be expected to yield returns outside the study period. [19] Whilst not the topic of 25 26 13 this paper it is important to note here that trainees were assessed pre and post training with regard to 27 28 14 their skills and knowledge and all were examined at the end of the training and passed. They were 29 30 15 also provided with ongoing mentorship and support from the team. 31 32 16 The trainees were based in health centres and district hospitals across rural Tanzania. The plan was to 33 34

17 recruit trainees from health facilities that were due to be upgraded with a theatre and maternity ward http://bmjopen.bmj.com/ 35 36 18 including equipment and resources so that trainees could implement their new skills and knowledge. 37 38 19 However, the reality was that of the 33 trainees who completed the programme only 19 returned to the 39 40 20 place from where they were selected and seven of these returned to facilities that had not been 41 42 21 upgraded or where upgrading was still in process. Fourteen trainees did not return to the facility from 43 on October 1, 2021 by guest. Protected copyright. 44 22 which they were recruited because the facilities had not been upgraded. Of these 10/14 were returned 45 46 47 23 to district hospitals in the area they had originally come from. Often these decisions were made by 48 49 24 local District Medical Officers responding to need and not to the strategic planning of the central 50 51 25 Ministry. Table 1 gives an overview of where the trainees were based and the type of facility they 52 53 26 worked in. This also notes the availability of an operating theatre in the facility as this is one of the 54 55 27 key areas to be upgraded. Upgrading of facilities was not part of the ETATMBA project but rather 56 57 28 was ongoing work with the government and other funding agencies. 58 59 29 60 6 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 7 of 66 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 1 2 1 The aim of this study is to explore the impact of the ETATMBA training on health outcomes 3 4 2 including maternal and neonatal morbidity and mortality in the facilities where trainees were based. 5 6 3 Secondly, to explore the challenges faced in working in these health facilities. In addition, a 7 8 4 qualitative study was undertaken at the same time with the trainees and other stakeholders but this 9 10 5 will be reported elsewhere. 11 12 6 13 14 7 Methods 15 For peer review only 16 8 Design 17 18 9 The study is a pre and post examination of maternal and neonatal health indicators and a survey of a 19 20 10 sample of health facilities in rural Tanzania. The survey includes: infrastructure, availability of 21 22 11 equipment, supplies and drugs. The facilities surveyed were those in which ETATMBA trainees were 23 24 12 placed posttraining. The health indicators were collected for the whole of 2011 (pre) and the whole of 25 26 13 2013 (post): the survey of the facilities was early 2014. 27 28 14 29 30 15 Outcome measure 31 32 16 Maternal and neonatal health outcomes were collected from each health facility where a trainee was 33 34

17 based (posttraining) for the whole of 2011 (pre) and 2013 (post). This included early neonatal http://bmjopen.bmj.com/ 35 36 18 mortality (only including deaths that occur before discharge) and maternal mortality (case specific) 37 38 19 and other obstetric indicators including: numbers of birth events, stillbirths, postpartum haemorrhage, 39 40 20 caesarean sections, obstructed labor and sepsis. It is important to note here that neonatal mortality 41 42 21 rates were not reported in the baseline data, we believed because they had been overlooked. We 43 on October 1, 2021 by guest. Protected copyright. 44 22 planned to rectify this by retrospectively collecting the data. However, after we visited the sites it 45 46 47 23 became clear that neonatal mortality rates had not been recorded, or at least were not available. These 48 49 24 data were therefore unavailable for either baseline or follow up. 50 51 25 52 53 26 The outcomes selected all relate to ETATMBA knowledge and skills training, and rely upon data 54 55 27 believed to be available in monthly / annual summary reports stored at each facility covering the 56 57 28 project time period. A 10% sample of variables were crosschecked with the actual registers for 58 59 29 accuracy at each facility. 60 7 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 8 of 66 BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 1 2 1 3 4 2 A predesigned instrument was used to capture the survey data (see online supplementary appendix 1). 5 6 3 This captured the availability of resources including; equipment, supplies and infrastructure, and 7 8 4 recorded whether there was a sufficient supply/number of the listed items for the facility’s daily 9 10 5 caseload of deliveries, and whether the items had been available and functional, available but NOT 11 12 6 functional, or not available. (e.g. infrastructure, equipment, supplies and drugs). Essential drugs: the 13 14 7 availability and supply of drugs for each room (emergency room, labour / delivery room, maternity 15 For peer review only 16 8 ward, operating theatre and pharmacy) were recorded. Checks were done to confirm whether the 17 18 9 listed drug was available and if the supply was sufficient to last for less than one week, up to one 19 20 10 week, up to two weeks, up to three weeks, or up to four or more weeks. 21 22 11 23 24 12 Research team 25 26 13 The primary data collection team consisted of two local research assistants based at the Ifakara Health 27 28 14 Institute (IHI), Dar es Salaam, Tanzania. Both of the research assistants are experienced researchers. 29 30 15 The principal investigator at the IHI gave local support, with management/oversight provided by DE 31 32 16 at Warwick. 33 34

17 http://bmjopen.bmj.com/ 35 36 18 Procedure 37 38 19 The research assistants identified the facilities in which trainees were working and extracted the 2011 39 40 20 study variables from data collected by colleagues at IHI for ETATMBA in 2012 (baseline data). The 41 42 21 followup data were the same variables for the year 2013. The followup data and the facility survey 43 on October 1, 2021 by guest. Protected copyright. 44 22 data were gathered during site visits to the facilities in early 2014. 45 46 47 23 48 49 24 Data analysis 50 51 25 Descriptive and summary statistics were produced for the two years, change scores were produced, 52 53 26 and ttests were used to look for differences. Significance was set at 5%. Data are presented in tables 54 55 27 and graphs as appropriate. Survey data are presented as descriptive statistics. Data will be grouped by 56 57 28 facility type (i.e. district hospitals and health centres) as they are different. In simple terms we would 58 59 60 8 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 9 of 66 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 1 2 1 expect the hospital to be larger than a health centre, have more staff and better availability of essential 3 4 2 infrastructure, supplies, equipment and drugs. 5 6 3 7 8 4 Ethical approval 9 10 5 The study was reviewed and approved by the Biomedical Research Ethics Committee (BREC) at the 11 12 6 University of Warwick, UK (REGO2013572) and The National Institute for Medical Research, 13 14 7 Institutional review board, Dar es Salaam, Tanzania (no.35). 15 For peer review only 16 8 17 18 9 Results 19 20 10 Post training ETATMBA trainees returned to 17 rural health facilities in Tanzania. Sixteen of these 21 22 11 health facilities were included in this study. Table 1 gives an overview of the facilities and the 23 24 12 ETATMBA trainees who were based their after the training. Thirtysix ACs received the ETATMBA 25 26 13 training, half of whom were assistant medical officers (AMOs) and half nurse midwives NMW/ 27 28 14 anaesthesia. During the project period one AMO and one NMW left the programme to pursue other 29 30 15 interests and one NMW died. Thus attrition at the end of the training programme was around 8%. 31 32 16 Fourteen trainees did not return to the facility from which they were recruited because the facilities 33 34

17 had not received an expected facility upgrade. http://bmjopen.bmj.com/ 35 36 18 37 38 19 39 40 20 41 42 21 Figure 1. Provides a United Nations definition of basic and comprehensive emergency obstetric and 43 on October 1, 2021 by guest. Protected copyright. 44 22 newborn care (BEmOC & CEmOC). 45 46 47 23 48 49 24 50 51 25 52 53 26 Table 2 summarises the key obstetric indicator figures from the 16 health facilities for 2011 and 2013. 54 55 27 56 57 28 No significant differences were found for any of the key obstetric variables across the lifetime of the 58 59 29 project. The number of deliveries seemed to decrease slightly overall (604) but the number of 60 9 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 10 of 66 BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 1 2 1 deliveries in health centres did rise (from 7326 to 7961). There is only a slight increase in overall 3 4 2 fresh stillbirths (+16, an increase of 1 case per 1000 births) whilst macerated ones appear to worsen in 5 6 3 health centres (from 8.3 to 13.9 cases per 1000 live births). Maternal death ratios show a downward 7 8 4 improving trend over the two years (down from 282 to 232 cases per 100,000 live births), but this is 9 10 5 not statistically significant. There was a reduction in the caesarean section rate overall down from 11 12 6 80.2 to 77.2 (cases per 1000 live births) with the largest reduction indicated in health centres where 13 14 7 rates are down from 10.6 to 6.2 (cases per 1000 live births). The birth complication variables and 15 For peer review only 16 8 caesarean section rates collected all show a slight increase overall but each shows a differing trend in 17 18 9 where they were reported. The rates of postpartum haemorrhage change little over time. Obstructed 19 20 10 labour rates increased in district hospitals (6.4 to 9.5 cases per 1000 live births), while in health 21 22 11 centres there was a decrease (6.7 to 2.9 cases per 1000 live births). Sepsis follows a similar trend with 23 24 12 an increase in hospitals (1.7 to 3.1 cases per 1000 live births) and a decrease in health centres (1.6 to 25 26 13 0.5 cases per 1000 live births). 27 28 14 29 30 15 31 32 16 Facility survey results 33 34

17 These results originate from the survey undertaken in early 2014 by IHI researchers. As noted in table http://bmjopen.bmj.com/ 35 36 18 1 above there were 17 facilities across the country that housed ETATMBA trainees during this 37 38 19 survey. One of these facilities (due to its distance and remoteness) was not visited. All results are 39 40 20 based on 16 facilities, nine healthcentres and seven district hospitals. 41 42 21 43 on October 1, 2021 by guest. Protected copyright. 44 22 Facilities: overall capacity and infrastructure 45 46 23 Running water and functioning toilets are a very significant problem with only one of nine health 47 48 24 centres (11%) and four of seven (57%) district hospitals found to have the availability of running 49 50 25 water and only just over half of facilities a functioning toilet (56%). Most facilities had sufficient 51 52 26 access to lighting to perform tasks at night (75%) but clearly some still struggle. Delivery beds were 53 54 27 found to be available in 56% of the health centres and 86% of the district hospitals. Ambulance 55 56 57 28 availability was poor at health centres with only one (11%) having availability, whereas five of the 58 59 29 seven (71%) district hospitals had an ambulance available. Referrals from within the maternity area 60 10 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 11 of 66 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 1 2 1 are problematic as only four health centres had a working (land line) phone in this area and none of 3 4 2 the district hospitals had. The availability of health related registers/records is variable varying from 5 6 3 100% for items like the delivery register and monthly/annual reports to 6% or less for the 7 8 4 gynaecology register, patient records and discharge registers. 9 10 5 11 12 6 13 14 7 15 For peer review only 16 8 Drugs and equipment for normal delivery and infection prevention 17 18 9 Generally, supplies and equipment availability were acceptable but there are a number of exceptions. 19 20 10 Only about 50% of facilities had needles and syringes available and similarly availability of suction 21 22 11 and vacuum extraction equipment was low. The availability of drugs for normal delivery purposes 23 24 12 was very variable with some drugs readily available (e.g. Lignocaine) whilst others had very low 25 26 13 availability (e.g. injectable antibiotic and Diazepam) (Table 4). 27 28 14 29 30 15 Infection prevention services in labour delivery/operating theatres 31 32 16 Overall only 75% or less of the facilities surveyed had the basics for infection prevention. None 33 34

17 seemed to have regular availability of soap for hand washing although antiseptics and bleach were http://bmjopen.bmj.com/ 35 36 18 available and may be alternatives (Table 4). 37 38 19 39 40 20 41 42 21 43 on October 1, 2021 by guest. Protected copyright. 44 22 Comprehensive services for provision of anaesthesia 45 46 47 23 Most of the district hospitals surveyed had availability of equipment and supplies for anaesthesia 48 49 24 although Halothane is only available in 3/7 and less than 40% overall. Health centres seemed to lack 50 51 25 access to oxygen with only 2/9 having supplies when surveyed (Table 5). 52 53 26 54 55 27 Items for management of birth complications and caesarean section 56 57 28 Overall, unsurprisingly, district hospitals had better availability of equipment, drugs and supplies for 58 59 29 managing birth complications and for performing caesarean sections (Table 5). 60 11 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 12 of 66 BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 1 2 1 3 4 2 5 6 3 7 8 4 Discussion 9 10 5 The main objectives of this study were to explore the impact of the ETATMBA training on health 11 12 6 outcomes including maternal and neonatal morbidity and mortality in the facilities where trainees 13 14 7 were based. Secondly, to explore the challenges faced in working in these health facilities. We were 15 For peer review only 16 8 successful in collecting data for the pre and post comparisons and also the survey data. 17 18 9 19 20 10 Interestingly, the number of actual births had decreased overall, in the 16 facilities measured, between 21 22 11 2011 and 2013. The reduction was seen mostly at the district hospitals with numbers increasing at 23 24 12 health centres. There was a slight increase in fresh stillbirths but again most of this is at the district 25 26 13 hospitals rather than at the health centres. This may suggest that health centres are referring more 27 28 14 women with this problem but the number of macerated stillbirths increased in both district hospitals 29 30 15 and health centres, with the latter being the biggest rise. This trend should be interpreted with caution, 31 32 16 since distinction of type of stillbirth is known to be variable in quality and indeed it may just suggest 33 34

17 that women are presenting late at the health facilities. http://bmjopen.bmj.com/ 35 36 18 37 38 19 Maternal deaths decreased which is encouraging as this was a goal of the ETATMBA training. 39 40 20 However, it is not a statistically significant reduction rather a downward trend. This could be simply 41 42 21 be a reflection of the reduction in maternal mortality reported in recent years across Tanzania. 43 on October 1, 2021 by guest. Protected copyright. 44 22 45 46 47 23 Neonatal mortality was one of our key health indicators in this study. However, it was found that 48 49 24 neonatal mortality was not recorded on Ministry of Health monthly summary sheets in facilities and 50 51 25 thus was not available for us. Reducing neonatal mortality is one of the WHO millennium 52 53 26 development goals. [20] Whilst the number of stillbirths was routinely recorded, early neonatal deaths 54 55 27 were not. This was a very disappointing outcome, a key component of the ETATMBA training was 56 57 28 related towards interventions to prevent neonatal deaths (i.e. deaths at or around the time of birth and 58 59 29 before discharge from hospital). [19] Indeed in Malawi, we have very positive indications that the 60 12 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 13 of 66 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 1 2 1 ETATMBA training has helped to save neonate lives. [21] Our study has acted as a ‘wakeup call’ to 3 4 2 the Ministry of Health and Social Welfare (MoHSW) in Tanzania, who have now updated the current 5 6 3 HIMS (Health Management Information System) to ensure neonatal data are collected. 7 8 4 9 10 5 Looking at the birth complication data (postpartum haemorrhage, obstructed labour and sepsis) all 11 12 6 are seen to rise from 2011 to 2013 in both district hospitals and health centres with one exception. 13 14 7 Sepsis rates in health centres decreased, though great caution must be observed, since registration of 15 For peer review only 16 8 morbidities is often incomplete and the facility survey showed centres lacking basic hygiene resources 17 18 9 such as soap. It is of some concern that the number of complications is increasing but this could be a 19 20 10 reflection of more women getting to a health facility where there are health staff who can deal with 21 22 11 the problems. Despite the increase in the numbers of mothers with obstructed labour and postpartum 23 24 12 haemorrhage it is encouraging that maternal mortality ratios at these facilities appear to be falling. 25 26 13 The observed incidence increase in these two registered morbidities by all probability implies an 27 28 14 enhanced recognition and registration of them, rather than a higher incidence in the facility population 29 30 15 under study. We do need to be cautious in our interpretation of these data with only before and after 31 32 16 data as there is no control to detect temporal trends that might be occurring across Tanzania. 33 34

17 http://bmjopen.bmj.com/ 35 36 18 Whilst our focus in this study was on the facilities where ETATMBA trainees returned to after their 37 38 19 training it is important to draw attention to events that were outside of the control of the ETATMBA 39 40 20 team, events that may have influenced the outcomes. Prior to recruitment the ETATMBA trainees 41 42 21 were based in health centres and district hospitals across rural Tanzania. The original Ministry of 43 on October 1, 2021 by guest. Protected copyright. 44 22 Health (MOHSW) plan was to recruit trainees from health facilities that were due to be upgraded with 45 46 47 23 a theatre and maternity ward including equipment and resources so that trainees could implement their 48 49 24 new skills. However, the reality was that of the 33 trainees who completed the programme only 19 50 51 25 returned to the place from where they were selected and seven of these returned to facilities that had 52 53 26 not been upgraded or where upgrading was still in process. Fourteen trainees did not return to the 54 55 27 facility from which they were recruited because the facilities had not been upgraded. Of these 10/14 56 57 28 were returned to district hospitals in the area they had originally come from. Often these decisions 58 59 29 were made by local District Medical Officers pragmatically responding to need and not to the 60 13 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 14 of 66 BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 1 2 1 strategic planning of the MoHSW. Upgrading of facilities was not part of the ETATMBA project but 3 4 2 rather was ongoing work with the Tanzanian Government and other funding agencies. It is clear that 5 6 3 in a number of cases trainees would have struggled to put their new found skills into practice as 7 8 4 facilities were not conducive. For those who returned to a district hospital it could have been a double 9 10 5 edged sword. On one hand a district hospital could give many more opportunities to put their new 11 12 6 found skills into practice but on the other the current senior staff may have been reluctant to allow 13 14 7 them to practice. 15 For peer review only 16 8 17 18 9 The survey reveals some alarming trends in the availability of resources in these facilities. A facility 19 20 10 designated as a CEmOC where there was no functioning operating theatre and a District hospital 21 22 11 designated as a BEmOC rather than a CEmOC when we surveyed. The latter clearly did not meet all 23 24 12 the requirements for a CEmOC at the time of the survey. There are considerable shortages in basic 25 26 13 infrastructure like running water, electricity and toilet facilities. Lack of telephones is interesting as 27 28 14 the survey specifically asked for a landline phone to be available. However, more and more now in 29 30 15 Africa mobile phones are used and are more reliable in terms of service provision. Future surveys 31 32 16 should take this into account. 33 34

17 http://bmjopen.bmj.com/ 35 36 18 Record keeping in the facilities is also very varied. We were lucky to find that the monthly/annual 37 38 19 summary reports (containing the data we required) were available in all facilities and in some we were 39 40 20 able to crosscheck the data in these with register records but some registers were missing and we have 41 42 21 already noted the issues surrounding neonatal mortality rates. 43 on October 1, 2021 by guest. Protected copyright. 44 22 45 46 47 23 The survey also reveals shortages in equipment, supplies and drugs that could impact on patient care. 48 49 24 The district hospitals are supplied than the health centres and this may be due to remoteness of the 50 51 25 health centres but there are disturbing shortages of the basics for infection/hygiene control and the 52 53 26 provision of oxygen. Infection prevention services were extremely poor. Basic items like soap for 54 55 27 hand washing were mostly absent. However, sepsis rates although rising slightly overall were not 56 57 28 significantly different to baseline (2011) levels, suggesting that despite enormous challenges and lack 58 59 29 of even basic supplies and equipment these clinicians manage to contain sepsis in their facilities. Our 60 14 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 15 of 66 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 1 2 1 survey findings are all the more alarming as they seem to mirror a more comprehensive survey done 3 4 2 back in 2005/2006 suggesting that things have not changed a great deal. [6] Despite all of this it does 5 6 3 seem that in the face of all of these challenges things are not getting any worse. 7 8 4 9 10 5 This study has a number of limitations not least that one of the primary outcomes was not available to 11 12 6 us. The sample is small and with generally only two trainees in each facility with large throughputs of 13 14 7 cases/births may not be totally representative of all facilities. We also are not comparing our facilities 15 For peer review only 16 8 to control districts so it is difficult to attribute changes just to ETATMBA training. Another limiting 17 18 9 factor is that ETATMBA had no control over where trainees returned to post training and a significant 19 20 10 number returned to facilities where they could not practice their new found skills. This however did 21 22 11 mean that our sample was more random (not chosen by us). Finally, this project needs to be seen in 23 24 12 the context of the vast distances between facilities and how the terrain and weather impacts on the 25 26 13 health service provision in rural Tanzania. Indeed, in 2009 EvjenOlsen et al., suggest the need for an 27 28 14 integrated and comprehensive hospital/community based approach to obstetric healthcare in rural 29 30 15 Tanzania but our experience here has not shown this being put into practice. [22] 31 32 16 33 34

17 Earlier findings from this project suggested that the training had impact, at the local level, on maternal http://bmjopen.bmj.com/ 35 36 18 mortality. [23] Sadly in this larger current study we cannot be certain of this conclusion. It is 37 38 19 acknowledged that maternal mortality is still a significant problem, particularly in rural Tanzania. [24] 39 40 20 Nelissen et al., suggest that there is a great need for the up scaling and use of evidencebased 41 42 21 interventions that could help to save lives. [24] We can only hope that the, evidenced based, 43 on October 1, 2021 by guest. Protected copyright. 44 22 ETATMBA training and its trainees will be a stimulus to improve care. But for a full impact the 45 46 47 23 implementation of the training needs to be linked to the provision of well supplied health care 48 49 24 facilities in the remote areas. We note that in one province the ETATMBA training has influenced the 50 51 25 upgrade of more health centres at the district level in tandem with the MoHSW objective of upgrading 52 53 26 at least 50% of all health centres in a particular province to provide CEmOC. [25] 54 55 27 56 57 28 A number of papers still highlight that women are reluctant to attend rural health facilities as they 58 59 29 believed the standard of care they will receive will be poor and many still give birth at home without 60 15 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 16 of 66 BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 1 2 1 skilled birth attendance. [9 26] We can only hope that the upskilling of health providers in these rural 3 4 2 areas cascade within the communities to encourage women to seek skilled help during birth. 5 6 3 7 8 4 Whilst not a direct result of our work during the lifetime of this project there has been a shift in 9 10 5 acknowledging the importance of this cadre of health workers. The negative label nonphysician 11 12 6 clinician has been replaced with the more dignifying and respectful associate clinician. Associate 13 14 7 clinicians are now coming together across Africa starting their own professional association. Indeed, 15 For peer review only 16 8 there is now a very active network called ANAC (African Network of Associate Clinicians) enabling 17 18 9 the formation of a community of practice. 19 20 10 21 22 11 Comparing our results with those from Malawi in this project we see an indication that ETATMBA 23 24 12 training can make a difference. [21] There are similarities and differences between this study and that 25 26 13 carried out in Malawi but in both countries it seems that overall the outcomes have been very positive. 27 28 14 29 30 15 We know that The ETATMBA training was successfully implemented (we were able to train the ACs 31 32 16 and we know we have improved their leadership, knowledge and clinical skills) but we are still 33 34

17 unclear about the impact in Tanzania. We interpret our results here with caution, presenting just http://bmjopen.bmj.com/ 35 36 18 exactly what we found. There are trends in the data, which suggest an improving picture. However it 37 38 19 seems that the full impact of the training at a community level does not as yet show in the results. We 39 40 20 believe that the dedication shown by the trainees coupled with their new skills and knowledge will 41 42 21 have a positive impact over the coming years as more health centres are upgraded. 43 on October 1, 2021 by guest. Protected copyright. 44 22 45 46 47 23 Competing Interests: None declared 48 49 24 50 51 25 Author contributions 52 53 26 DE, JPOH, GM, SB, and SP were involved in conception and design of the study. DE drafted the 54 55 27 manuscript supported by all authors. JPOH, GM, SB, SP and DD were responsible for the design of 56 57 28 the training. GM, SB, SP, PK, AN, HMM and DD were responsible for the management and delivery 58 59 29 of the training. AS and FM, carried out the fieldwork and collated results supervised by DE. 60 16 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 17 of 66 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 1 2 1 3 4 2 Funding and Acknowledgements 5 6 3 Enhancing Human Resources and Use of Appropriate Technologies for Maternal and Perinatal 7 8 4 Survival in subSaharan Africa (ETATMBA) is a collaborative project funded by the European 9 10 5 Commission, Seventh Framework Programme THEME [HEALTH.2010.3.42] [Project no. 266290]. 11 12 6 This study was embedded within this programme of work. All authors are part of the ETATMBA 13 14 7 team. The ETATMBA team would like to thank all of the AMOs and the district medical and nursing 15 For peer review only 16 8 officers for their hard work and support. This project benefited from facilities funded through 17 18 9 Birmingham Science City Translational Medicine Clinical Research and Infrastructure Trials 19 20 10 Platform, with support from Advantage West Midlands 21 22 11 23 The ETATMBA Study Group 24 Malawi University of Malawi College Sweden Karolinska Institutet, Sweden 25 of Medicine Staffan Bergström 26 Francis Kamwendo 27 Chisale Mhango 28 WanangwaChimwaza United GE Healthcare 29 ChikayikoChiwandira Kingdom Alan Davies 30 Queen Dube 31 The University of Warwick, UK 32 Ministry of Health, Malawi Paul O'Hare 33 Fannie Kachale Siobhan Quenby 34

ChimwemweMvula Douglas Simkiss http://bmjopen.bmj.com/ 35 David Davies 36 Tanzania Ifakara Health Institute, David Ellard 37 Tanzania Frances Griffiths 38 Godfrey Mbaruku Ngiangabakwin, Kandala 39 Paul Kihaile AnneMarie Brennan 40 Hamed Mohamed Edward Peile 41 Aloisia Shemdoe AnneMarie Slowther 42 FestoMazuguni SaliyaChipwete 43 on October 1, 2021 by guest. Protected copyright. Paul Beeby 44 Tanzanian Training Centre 45 for International Health Gregory Eloundou 46 Senga Pemba Harry Gee 47 Sidney Ndeki Vinod Patel 48 Angelo Nyamtema 49 12 50 13 51 14 52 53 54 55 56 57 58 59 60 17 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 18 of 66 BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 1 2 1 REFERENCES 3 4 2 1. WHO, GHWA. A universal truth: no health without a workforce. 2014. 5 6 3 http://www.who.int/workforcealliance/knowledge/resources/GHWA 7 8 4 a_universal_truth_report.pdf?ua=1. 9 10 5 2. Hoyler M, Finlayson SG, McClain C, et al. Shortage of Doctors, Shortage of Data: A Review of the 11 12 6 Global Surgery, Obstetrics, and Anesthesia Workforce Literature. World J Surg 13 14 7 2014;38(2):26980 doi: 10.1007/s002680132324y. 15 For peer review only 16 8 3. WHO. Human Resources for Health Observer. 2012; (11). 17 18 9 http://www.who.int/hrh/resources/observer/en/. 19 20 10 4. Anyangwe SC, Mtonga C. Inequities in the global health workforce: the greatest impediment to 21 22 23 11 health in subSaharan Africa. International journal of environmental research and public 24 25 12 health 2007;4(2):93100 26 27 13 5. Kinfu Y, Dal Poz MR, Mercer H, et al. The health worker shortage in Africa: are enough 28 29 14 physicians and nurses being trained? Bulletin of the World Health Organization 30 31 15 2009;87(3):22530 32 33 16 6. National Bureau of Statistics [Tanzania]. Tanzania Service Provision Assessment Survey 2006: 34 35 17 Key Findings on Family Planning, Maternal and Child Health, and . 2006. http://bmjopen.bmj.com/ 36 37 18 http://dhsprogram.com/pubs/pdf/SR130/SR130.pdf. 38 39 19 7. Magoma M, Requejo J, Campbell OM, et al. High ANC coverage and low skilled attendance in a 40 41 20 rural Tanzanian district: a case for implementing a birth plan intervention. BMC pregnancy 42

43 21 and childbirth 2010;10:13 doi: 10.1186/147123931013. on October 1, 2021 by guest. Protected copyright. 44 45 22 8. MoHSW. Tanzania service availability and readiness assessment (SARA) 2012. 2013. 46 47 23 http://ihi.eprints.org/2448/1/SARA_2012_Report.pdf. 48 49 24 9. Mselle LT, Moland KM, Mvungi A, et al. Why give birth in health facility? Users' and providers' 50 51 25 accounts of poor quality of birth care in Tanzania. BMC Health Serv Res 2013;13:174 doi: 52 53 26 10.1186/1472696313174. 54 55 27 10. Sarker M, Schmid G, Larsson E, et al. Quality of antenatal care in rural southern Tanzania: a 56 57 28 reality check. BMC Res Notes 2010;3:209 doi: 10.1186/175605003209. 58 59 60 18 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 19 of 66 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 1 2 1 11. Mullan F, Frehywot S. Nonphysician clinicians in 47 subSaharan African countries. Lancet 3 4 2 2007;370(9605):215863 doi: 10.1016/S01406736(07)607855. 5 6 3 12. McCord C, Mbaruku G, Pereira C, et al. The quality of emergency obstetrical surgery by assistant 7 8 4 medical officers in Tanzanian district hospitals. Health Aff (Millwood) 2009;28(5):w87685 9 10 5 doi: 10.1377/hlthaff.28.5.w876. 11 12 6 13. Nyamtema AS, Pemba SK, Mbaruku G, et al. Tanzanian lessons in using nonphysician clinicians 13 14 7 to scale up comprehensive emergency obstetric care in remote and rural areas. Human 15 For peer review only 16 8 resources for health 2011;9:28 doi: 10.1186/14784491928. 17 18 9 14. Chilopora G, Pereira C, Kamwendo F, et al. Postoperative outcome of caesarean sections and 19 20 10 other major emergency obstetric surgery by clinical officers and medical officers in Malawi. 21 22 11 Human resources for health 2007;5:17 doi: 10.1186/14784491517. 23 24 12 15. Pereira C, Cumbi A, Malalane R, et al. Meeting the need for emergency obstetric care in 25 26 13 Mozambique: work performance and histories of medical doctors and assistant medical 27 28 14 officers trained for surgery. BJOG : an international journal of obstetrics and gynaecology 29 30 15 2007;114(12):15303 doi: 10.1111/j.14710528.2007.01489.x. 31 32 16 16. Bergstrom S. Who will do the caesareans when there is no doctor? Finding creative solutions to 33 34

17 the human resource crisis. BJOG : an international journal of obstetrics and gynaecology http://bmjopen.bmj.com/ 35 36 18 2005;112(9):11689 doi: 10.1111/j.14710528.2005.00719.x. 37 38 19 17. Davis D, Evans M, Jadad A, et al. The case for knowledge translation: shortening the journey 39 40 20 from evidence to effect. BMJ 2003;327(7405):335 doi: 10.1136/bmj.327.7405.33. 41 42 21 18. McGlynn EA, Asch SM, Adams J, et al. The quality of health care delivered to adults in the 43 on October 1, 2021 by guest. Protected copyright. 44 22 United States. The New England journal of medicine 2003;348(26):263545 doi: 45 46 47 23 10.1056/NEJMsa022615. 48 49 24 19. Brennan AM. ETATMBA: Enhancing human resources and the use of appropriate technologies 50 51 25 for maternal and perinatal survival in subSaharan Africa. Secondary ETATMBA: Enhancing 52 53 26 human resources and the use of appropriate technologies for maternal and perinatal survival 54 55 27 in subSaharan Africa [Web Pages] 20 Nov. 2013. 56 57 28 http://www2.warwick.ac.uk/fac/med/about/global/etatmba/about/. 58 59 60 19 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 20 of 66 BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 1 2 1 20. WHO. The World Health Report: Working together for health. Geneva: World Health 3 4 2 Organisation, 2006. 5 6 3 21. Ellard DR, Chimwaza W, Davies D, et al. Can training in advanced clinical skills in obstetrics, 7 8 4 neonatal care and leadership, of nonphysician clinicians in Malawi impact on clinical 9 10 5 services improvements (the ETATMBA project): a process evaluation. BMJ Open 2014;4(8) 11 12 6 doi: 10.1136/bmjopen2014005751. 13 14 7 22. EvjenOlsen B, Olsen OE, Kvale G. Achieving progress in maternal and neonatal health through 15 For peer review only 16 8 integrated and comprehensive healthcare services experiences from a programme in 17 18 9 northern Tanzania. International journal for equity in health 2009;8:27 doi: 10.1186/1475 19 20 10 9276827. 21 22 11 23. Kihaile P, Mbaruku G, Pemba S. Improved Maternal and Perinatal Mortalities by Trained Medical 23 24 12 Assistant Staffs in Rural Tanzania. J Health Med Informat 2013;S11:007 doi: 10.4172/2157 25 26 13 7420.S11007. 27 28 14 24. Nelissen EJ, Mduma E, Ersdal HL, et al. Maternal near miss and mortality in a rural referral 29 30 15 hospital in northern Tanzania: a crosssectional study. BMC pregnancy and childbirth 31 32 16 2013;13:141 doi: 10.1186/1471239313141. 33 34

17 25. Ministry of Health and Social Welfare. Mpango wa Maendeleo wa Afya ya Msingi (MMAM) http://bmjopen.bmj.com/ 35 36 18 20072017 (Primary Health Services Development Programme,PHSDP). United Republic of 37 38 19 Tanzania: Ministry of Health and Social Welfare, 2007. 39 40 20 26. Ng'anjo Phiri S, Kiserud T, Kvale G, et al. Factors associated with health facility childbirth in 41 42 21 districts of Kenya, Tanzania and Zambia: a population based survey. BMC pregnancy and 43 on October 1, 2021 by guest. Protected copyright. 44 22 childbirth 2014;14:219 doi: 10.1186/1471239314219. 45 46 47 23 48 49 50 51 52 53 54 55 56 57 58 59 60 20 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 21 of 66 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 1 2 Table 1. Health facilities where the Tanzanian ETATMBA trainees were based in 2013 3 District Name of facility Operating CEmOC or No. Trainees 4 5 Theatre BEmOC 6 1 Bukombe Bukombe District Hospital Yes CEmOC 1 AMO 7 8 2 Bukombe Uyovu Health Centre No BEmOC 1 AMO, 1CO 9 10 3 Geita Nzela Health Centre Yes CEmOC 1 NMW, 1 Nurse 11 4 Geita Katoro Health Centre No BEmOC 1 NMW 12 13 5 Inyonga Mamba Health Centre Yes CEmOC 1 NMW 14 15 6 KaramboFor Matai Healthpeer Centre reviewNo only BEmOC 1 AMO, 1NMW 16 7 Liwale Liwale District Hospital No CEmOC 2 AMOs 17 18 8 Meatu Mwandoya Health Centre No BEmOC 1 AMO, 1 NMW 19 20 9 Mpanda Mpanda District Hospital Yes BEmOC 1 AMO, 1 Nurse 21 10 Nachingwea Nachingwea District Yes CEmOC 2 AMOs 22 23 Hospital 24 11 Nkasi Kirando Health Centre Yes CEmOC 2 AMOs 25 26 12 Nyanghwale Nyanghwale Health Centre No BEmOC 1 AMO, 1 NMW 27 a 28 13 Nyanghwale Kharumwa District Hospital Yes CEmOC 1 AMO, 1 NMW 29 14 Ruangwa Ruangwa District Hospital Yes CEmOC 1 AMO, 1 NMW 30 31 15 Sumbawanga Laela Health Centre No BEmOC 1 AMO, 1 NMW 32 33 16 Chato Chato District Hospital Yes CEmOC 1 AMO, 1 NMW

34 b 35 17 Lindi Nyangao Mission Hospital unknown CEmOC 2 NMWs http://bmjopen.bmj.com/ 36 a Upgraded to a district hospital between 2011 &2013. b This hospital not visited so not included in 37 analysis. AMO – Assistant medical officer, NMW – Nurse midwife, Nurse – nurse/anaesthetics. 38 39 CEmOC – Comprehensive Emergency Obstetric Care, BEmOC Basic Emergency Obstetric Care 40 1 41 42 2

43 on October 1, 2021 by guest. Protected copyright. 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 21 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 22 of 66 BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 1 2 1 Figure 1. Defining Basic and Comprehensive Emergency Obstetric and newborn care 3 4 2 (BEmOC & CEmOC) 5 6 7 3 Basic emergency obstetric and newborn care (BEmOC) is critical to reducing maternal and 8 9 4 neonatal death. This care, which can be provided with skilled staff in health centres, large or 10 11 5 small, includes the capabilities for: 12 13 14 6 • Administering antibiotics, uterotonic drugs (oxytocin) and anticonvulsants 15 For peer review only 16 7 (magnesium sulphate); 17 18 19 8 • Manual removal of the placenta; 20 21 9 • Removal of retained products following miscarriage or abortion; 22 23 10 • Assisted vaginal delivery, preferably with vacuum extractor; 24 25 26 11 • Basic neonatal resuscitation care. 27 28 12 Comprehensive emergency obstetric and newborn care (CEmOC), typically delivered in 29 30 31 13 hospitals, includes all the basic functions above, plus capabilities for: 32 33 34 14 • Performing Caesarean sections; 35 http://bmjopen.bmj.com/ 36 15 • Safe blood transfusion; 37 38 16 • Provision of care to sick and lowbirth weight newborns, including resuscitation. 39 40 41 17 Adapted from United Nations Population fund material. For more information see: 42 43 18 http://www.unfpa.org/resources/settingstandardsemergencyobstetricandnewborn on October 1, 2021 by guest. Protected copyright. 44 45 46 19 care#sthash.5rcjLhLA.dpuf 47 48 20 49 50 51 52 53 54 55 56 57 58 59 60 22 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 23 of 66 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 1 2 Table 2. Comparison of key maternal, neonatal and birth complication figures from baseline (2011) to follow 3 up (2013) 4 5 2011 2013

6 a b * 7 DH HC Total DH HC Total Difference ba 8 (n=7) (n=9) (n=7) (n=9) 9 10 Total Deliveries 17893 7326 25219 16654 7961 24615 604 11 Fresh stillbirths (FSB) (n) 287 65 352 300 68 368 16·0 12 13 FSB rate (per 1000 births) 16·0 8·9 14·0 18·0 8·5 15·0 1·0 14 15 Macerated StillbirthsFor (MSB) peer (n) 312review 61 373 only 305 111 416 43·0 16 MSB rate (per 1000 births) 17·4 8·3 14·8 18·3 13·9 16·9 2·1 17 18 Maternal deaths (n) 68 3 71 55 2 57 14·0 19 20 MD Ratio (per 100,000 births) 380 41 282 330 25 232 50 21 Caesarean deliveries (CS) (n) 1944 78 2022 1851 49 1900 122 22 23 CS rate (per 1000 births) 108·6 10·6 80·2 111·1 6·2 77·2 3·0 24 25 Postpartum Haemorrhage (PPH) (n) 200 77 277 225 86 311 34·0 26 PPH rate (per 1000 births) 11·2 10·5 11·0 13·5 10·8 12·6 1·7 27 28 Obstructed labour (Obst/Lab) (n) 114 49 163 159 23 182 19·0 29 30 Obst/Lab rate per 1000 births 6·4 6·7 6·5 9·5 2·9 7·4 0·9 31 Sepsis (n) 31 12 43 51 4 55 12·0 32 33 Sepsis rate per 1000 births 1·7 1·6 1·7 3·1 0·5 2·2 0·5

34 * 35 DH – District hospitals, HC – Health centres· Note: there are NO significant differences here so p values not http://bmjopen.bmj.com/ 36 shown. 37 1 38 39 2 40 41 3 42

43 on October 1, 2021 by guest. Protected copyright. 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 23 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 24 of 66 BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 1 2 Table 3. Survey findings from health facilities in Tanzania related to infrastructure 3 4 5 Facilities with the items 6 HC (%) DH (%) 7 Overall n=9 n=7 8 Health facility Infrastructure Availability of Power & Availability of water 9 Sufficient light source to perform tasks at night 12 (75%) 6 (67%) 6 (86%) 10 Means of ventilation 5 (31%) 1 (11%) 4 (57%) 11 12 Running water 5 (31%) 1 (11%) 4 (57%) 13 Functioning toilet 9 (56%) 6 (67%) 3 (43%) 14 Functional fan/air conditioning 5 (31%) 1 (11%) 4 (57%) 15 ForCurtains/means peer of providing review patient privacy only14 (88%) 9 (100%) 5 (71%) 16 Waiting area for visitors and family 6 (38%) 4 (43%) 2 (33%) 17 18 Facility with electricity 14 (89%) 8 (86%) 6 (86%) 19 Motor Vehicle Ambulance Available and functional 6 (38%) 1 (11%) 5 (71%) 20 Available and functional landline telephone in the maternity area 4 (25%) 4 (43%) 0 (0%) 21 Delivery bed / table 11 (69%) 5 (56%) 6 (86%) 22

23 24 Availability of health related registers 25 General admission register 11 (69%) 5 (56%) 6 (86%) 26 Delivery register 16 (100%) 9 (100%) 7 (100%) 27 Maternity ward register 9 (56%) 4 (44%) 5 (71%) 28 Female ward register 9 (56%) 4 (44%) 5 (71%) 29 30 Operating theatre register 10 (63%) 4 (44%) 6 (86%) 31 Gynaecology register 0 (0%) 0 0 32 Postabortion register 9 (56%) 4 (44%) 5 (71%) 33 Individual patient records 1 (6%) 0 1 (14%) 34 35 Discharge register 1 (6%) 0 1 (14%) http://bmjopen.bmj.com/ 36 Death register 11 (69%) 6 (67%) 5 (71%) 37 Mortuary register 7 (44%) 2 (22%) 5 (71%) 38 Monthly / annual facility summary reports 16 (100%) 9 (100%) 7 (100%) 39 DH – District hospitals, HC – Health centres 40 41 1 42 2 43 on October 1, 2021 by guest. Protected copyright. 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 24 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 25 of 66 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 1 2 Table 4.Survey findings from health facilities in Tanzania related to the availability of equipment, 3 supplies and drugs 4 5 facilities with the equipment 6 Overall (%) HC (%) DH (%) 7 N=16 N=9 N=7 8 Drugs and equipment: Availability of Items for normal delivery 9 Equipment and Supplies 10 Blood pressure cuff/machine 13 (81%) 7 (78%) 6 (86%) 11 12 Stethoscope 15 (94%) 8 (89%) 7 (100%) 13 Fetal stethoscope 16 (100%) 9 (100%) 7 (100%) 14 Clinical thermometer 13 (81%) 6 (67%) 7 (100%) 15 For peer reviewSterile gloves 16 (100%) only 9 (100%) 7 (100%) 16 Nonsterile protective clothing/apron 15 (94%) 8 (89%) 7 (100%) 17 18 Scissors or razor blade for cutting cord 15 (94%) 9 (100%) 6 (86%) 19 Cord ties 10 (63%) 5 (56%) 5 (71%) 20 Needles and Syringes 8 (50%) 4 (44%) 4 (57%) 21 IV fluid set (giving set) 15 (94%) 9 (100%) 6 (86%) 22 Suture needles and suture materials 10 (63%) 5 (56%) 5 (71%) 23 24 suction apparatus 8 (50%) 3 (33%) 5 (71%) 25 Manual vacuum extractor 5 (31%) 2 (33%) 2 (29%) 26 Obstetric forceps 11 (69%) 8 (89%) 3 (43%) 27 Drugs 28 29 Pitocin (Oxytocin) 13 (81%) 6 (67%) 7 (100%) 30 Ergometrine (injectable) 4 (25%) 3 (33%) 1 (14%) 31 Normal saline 14 (88%) 8 (89%) 6 (86%) 32 Ringers lactate 7 (44%) 2 (22%) 5 (71%) 33 Dextrose / glucose 9 (56%) 3 (33%) 6 (86%) 34 35 Lignocaine 2% or 1% 15 (94%) 8 (89%) 7 (100%) http://bmjopen.bmj.com/ 36 injectable antibiotic 5 (31%) 3 (33%) 2 (29%) 37 Magnesium sulphate 14 (88%) 8 (89%) 6 (86%) 38 Diazepam 6 (38%) 3 (33%) 3 (43%) 39 Skin disinfectant 12 (75%) 7 (78%) 5 (71%) 40 41 42 Availability of Infection prevention services in labour delivery/operating theatres

43 Decontamination container with prepared solution 11 (69%) 5 (56%) 6 (86%) on October 1, 2021 by guest. Protected copyright. 44 Covered contaminated trash bin 11 (69%) 6 (67%) 5 (71%) 45 Sharps container 12 (75%) 6 (67%) 6 (86%) 46 47 Soap 0 0 0 48 Antiseptics 10 (63%) 5 (56%) 5 (71%) 49 Chlorine/ Bleach 6 (38%) 2 (22%) 4 (57%) 50 Sterile gloves 12 (75%) 6 (67%) 6 (86%) 51 Other items 52 53 Regular trash bin 12 (75%) 6 (67%) 6 (86%) 54 Non sterile gloves 12 (75%) 6 (67%) 6 (86%) 55 Nonsterile protective clothing 12 (75%) 6 (67%) 6 (86%) 56 DH – District hospitals, HC – Health centres 57 1 58 59 60 25 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 26 of 66 BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 1 2 1 3 Table 5.Survey findings from health facilities in Tanzania related to the availability of items 4 for management of anaesthesia, birth complications and caesarean section 5 6 Facilities with the items 7 Equipment Overall HC DH 8 N=16 N=9 N=7 9 10 Items for provision of anaesthesia 11 Suction machine 6 (38%) 4 (44%) 2 (29%) 12 Filled oxygen cylinder with cylinder carrier 8 (50%) 2 (22%) 6 (86%) 13 and key to open valve 14 Intubating forceps (Magill) 6 (38%) 4 (44%) 2 (29%) 15 For peerAdult laryngoscope review 11 (69%) 6only (67%) 5 (71%) 16 17 Adult ventilator bag and mask 11 (69%) 6 (67%) 5 (71%) 18 IV fluid set (giving set) 10 (63%) 5 (56%) 5 (71%) 19 Spinal needles (18gauge to 25gauge) 3 (19%) 1 (11%) 2 (29%) 20 Endotracheal tubes with cuffs (8 – 10mm) 9 (56%) 4 (44%) 5 (71%) 21 Halothane 6 (38%) 3 (33%) 3 (43%) 22 23 Ketamine 11 (6%) 5 (56%) 6 (86%) 24 Anaesthetic face masks 9 (56%) 5 (56%) 4 (57%) 25 26 Items for management of pre-eclampsia/

27 eclampsia 28 Magnesium Sulphate 7 (44%) 4 (44%) 3 (43%) 29 Diazepaminjectable 10 (63%) 4 (44%) 6 (86%) 30 31 Niphedipine 1 (6%) 0 (0%) 1 (14%) 32 Blood pressure cuff/machine 13 (81%) 7 (78%) 6 (86%) 33 Stethoscope 15 (94%) 8 (89%) 7 (100%) 34

Adult ventilator bag and mask 13 (81%) 7 (78%) 6 (86%) http://bmjopen.bmj.com/ 35 36 Needles and Syringes 4 (25%) 1 (11%) 3 (43%) 37 Urinary catheters (Foleys) 8 (50%) 3 (33%) 5 (71%) 38 Uristix 4 (25%) 1 (11%) 3 (43%) 39 40 Items for management of haemorrhage (parenteral uterotonics) 41 42 Needles and Syringes 8 (50%) 4 (44%) 4 (57%)

43 IV fluid set (giving set) 9 (56%) 3 (33%) 6 (86%) on October 1, 2021 by guest. Protected copyright. 44 45 Items for Caesarean Section (not including anaesthesia) 46 Operating table 47 48 Light adjustable, shadow less 11 (69%) 6 (56%) 5 (86%) 49 Antiseptics 10 (63%) 5 (56%) 5 (71%) 50 Sterile gloves 12 (75%) 6 (67%) 6 (86%) 51 Cord ties 10 (63%) 5 (56%) 5 (71%) 52 Needles and Syringes 6 (38%) 4 (44%) 2 (29%) 53 54 Benzyl Penicillin 4 (25%) 3 (33%) 1 (14%) 55 Metronidazole (IV) 2 (13%) 1 (11%) 1 (14%) 56 Gentamycin (IV) 1 (6%) 1 (11%) 0 (0%) 57 Caesarean section pack 58 needle holder 13 (81%) 7 (78%) 6 (86%) 59 60 26 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 27 of 66 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 1 2 scalpel handle with blade 10 (63%) 5 (56%) 5 (71%) 3 retractor 12 (75%) 6 (67%) 6 (86%) 4 surgical scissors 12 (75%) 6 (67%) 6 (86%) 5 a suction apparatus / 8 6 (38%) 4 (44%) 2 (29%) 6 7 oxygen 8 (50%) 2 (22%) 6 (86%) 8 sutures 11 (69%) 5 (56%) 6 (86%) 9 ketamine 11 (69%) 5 (56%) 6 (86%) 10 a lidocaine / 5 12 (75%) 6 (67%) 6 (86%) 11 DH – District hospitals, HC – Health centres 12 a The numbers against these items (8 & 5) are the units for this item to be classed as 13 available (e.g. there has to be 8 suction apparatus for it to be classed as available) 14 1 15 For peer review only 16 2 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 http://bmjopen.bmj.com/ 36 37 38 39 40 41 42

43 on October 1, 2021 by guest. Protected copyright. 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 27 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from

BMJ Open Page 28 of 66

1 2 3 4 5 ETATMBA Facility Assessment (A) 6 7 Module 1: Moduli kuu (KAMILI) 8 9 Name of the Interviewer ______10 Date of the interview: ______11 For peer review only 12 GIS coordinates [filled by supervisor later]______13 14 Instructions: 15 SEHEMU YA 1: TAARIFA ZA MSINGI ZA KITUO 16 NA MASWALI NA MCHUJO MPANGILIO WA MAKUNDI RUKA 17

1101 http://bmjopen.bmj.com/ 18 District: ______19 1102 20 Name of the Facility: ______1103 21 ID of the Facility: ______22 23 24 25 Data Collection Sources and Quality

26 Please record the following information about the registers used to collect obstetric complication and maternal death information. on October 1, 2021 by guest. Protected copyright. 27  List as per MOH guidelines. 28 29 30 Check if used to collect (check all Answer the following only for the registers that collect delivery and 31 that apply): obstetric complication data and/or maternal death data: Is register 32 Registers and Data Is the register 33 available at this Obstetric Is the register Is the register Is the register Sources Maternal death regularly 34 facility? complication easily completely up to date? data reviewed by 35 CSa data accessible? filled out? CSc staff? 36 CSb CSd CSe CSf 37 CSg 1 General admission 38 □ 0.No □ 1.Yes □ 0.No □ 1.Yes □ 0.No □ 1.Yes □ 0.No □ 1.Yes □ 0.No □ 1.Yes □ 0.No □ 1.Yes □ 0.No □ 1.Yes 39 register 40 2 Delivery register □ 0.No □ 1.Yes □ 0.No □ 1.Yes □ 0.No □ 1.Yes □ 0.No □ 1.Yes □ 0.No □ 1.Yes □ 0.No □ 1.Yes □ 0.No □ 1.Yes 41 42 43 44 45 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from

Page 29 of 66 BMJ Open

1 2 3 3 Maternity ward 4 □ 0.No □ 1.Yes □ 0.No □ 1.Yes □ 0.No □ 1.Yes □ 0.No □ 1.Yes □ 0.No □ 1.Yes □ 0.No □ 1.Yes □ 0.No □ 1.Yes register 5 6 4 Female ward register □ 0.No □ 1.Yes □ 0.No □ 1.Yes □ 0.No □ 1.Yes □ 0.No □ 1.Yes □ 0.No □ 1.Yes □ 0.No □ 1.Yes □ 0.No □ 1.Yes 7 5 Operating theatre 8 □ 0.No □ 1.Yes □ 0.No □ 1.Yes □ 0.No □ 1.Yes □ 0.No □ 1.Yes □ 0.No □ 1.Yes □ 0.No □ 1.Yes □ 0.No □ 1.Yes 9 register 10 6 Gynecology register □ 0.No □ 1.Yes □ 0.No □ 1.Yes □ 0.No □ 1.Yes □ 0.No □ 1.Yes □ 0.No □ 1.Yes □ 0.No □ 1.Yes □ 0.No □ 1.Yes 11 For peer review only 12 7 Post-abortion register □ 0.No □ 1.Yes □ 0.No □ 1.Yes □ 0.No □ 1.Yes □ 0.No □ 1.Yes □ 0.No □ 1.Yes □ 0.No □ 1.Yes □ 0.No □ 1.Yes 13 8 Individual patient 14 □ 0.No □ 1.Yes □ 0.No □ 1.Yes □ 0.No □ 1.Yes □ 0.No □ 1.Yes □ 0.No □ 1.Yes □ 0.No □ 1.Yes □ 0.No □ 1.Yes 15 records 16 9 Discharge register □ 0.No □ 1.Yes □ 0.No □ 1.Yes □ 0.No □ 1.Yes □ 0.No □ 1.Yes □ 0.No □ 1.Yes □ 0.No □ 1.Yes □ 0.No □ 1.Yes 17 18 10 Death register □ 0.No □ 1.Yes □ 0.No □ 1.Yes □ 0.No □ 1.Yes □ 0.No □ 1.Yes □ 0.No □ 1.Yeshttp://bmjopen.bmj.com/ □ 0.No □ 1.Yes □ 0.No □ 1.Yes 19 20 11 Mortuary register □ 0.No □ 1.Yes □ 0.No □ 1.Yes □ 0.No □ 1.Yes □ 0.No □ 1.Yes □ 0.No □ 1.Yes □ 0.No □ 1.Yes □ 0.No □ 1.Yes 21 22 12 Monthly / annual 23 facility summary □ 0.No □ 1.Yes □ 0.No □ 1.Yes □ 0.No □ 1.Yes □ 0.No □ 1.Yes □ 0.No □ 1.Yes □ 0.No □ 1.Yes □ 0.No □ 1.Yes 24 reports/forms 25 13 Otherooo (specify): 26 □ 0.No □ 1.Yes □ 0.No □ 1.Yes □ 0.No □ 1.Yes □ 0.No □ 1.Yes □ 0.No □ 1.Yes on October 1, 2021 by guest. Protected copyright. □ 0.No □ 1.Yes □ 0.No □1.Yes 27 28 29 NB: UN Process Indicator Data: 30 31 32 33 34 35 36 37 38 39 40 41 42 Utafiti wa Vituo vya Tiba wa ETATMBA Page 2 of 5 43 Modulli ya 1 ya 4: Moduli Kuu Jina la Kituo cha Tiba______44 45 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from

BMJ Open Page 30 of 66

1 2 3 4 5 6 CHUKUA TAKWIMU ZA KUANZIA January 2012- December 2012 7 8 Year Total 9 Jan Feb Mar Apr May Jun Jul Aug Sept Oct Nov Dec 10

11 No. obstetric admissions For peer review only 12 Total No of all deliveries (SVD+CS) 13 14 Total No deliveries (SVD) 15 Deliveries (Breech) 16 17 Twins http://bmjopen.bmj.com/ 18 BBA 19 20 No. cesarean deliveries 21 Other mal-presentations

22 (Transverse, Compound etc) 23 24 Direct obstetric complications: 25 Hemorrhage (ante & post-partum) 26 on October 1, 2021 by guest. Protected copyright. 27 Obstructed / prolonged labor 28 Ruptured uterus 29 30 Post-partum sepsis 31 Severe pre-eclampsia / eclampsia 32 Complications of abortion (with hemorrhage 33 and/or sepsis) 34 35 Ectopic pregnancy 36 Total direct obstetric complications 37 Other obstetric complications (from all other 38 39 causes) – Specify: 40 Other abortion complications 41 42 Utafiti wa Vituo vya Tiba wa ETATMBA Page 3 of 5 43 Modulli ya 1 ya 4: Moduli Kuu Jina la Kituo cha Tiba______44 45 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from

Page 31 of 66 BMJ Open

1 2 3 4 5 6 Year: Jan Feb Mar Apr May Jun Jul Aug Sept Oct Nov Dec Total 7 8 Maternal deaths – direct

9 obstetric causes: 10 11 Hemorrhage (ante & post-partum)For peer review only 12 Obstructed / prolonged labor 13 14 Ruptured uterus 15 16 Post-partum sepsis 17 18 Severe pre-eclampsia / eclampsia http://bmjopen.bmj.com/ 19 Complications of abortion (with 20 21 hemorrhage and/or sepsis) 22 23 Ectopic pregnancy 24 Total maternal deaths from 25

26 direct obstetric causes on October 1, 2021 by guest. Protected copyright. 27 Other maternal deaths (direct 28 causes) – Specify: 29 30 31 32 33 34 35 36 37 38 39 40 41 42 Utafiti wa Vituo vya Tiba wa ETATMBA Page 4 of 5 43 Modulli ya 1 ya 4: Moduli Kuu Jina la Kituo cha Tiba______44 45 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from

BMJ Open Page 32 of 66

1 2 3 ** List indirect obstetric complications and maternal deaths relevant for local country context (examples: HIV, severe anemia, malaria, etc. 4

5 Year: Jan Feb Mar Apr May Jun Jul Aug Sept Oct Total 6 Nov Dec 7 Indirect obstetric complications: f26a-f26x 8 9 26 malaria 10 11 27 hiv/aids For peer review only 12 28 severe anaemia 13 14 28_1 hepatitis 15 16 28_2 other indirect complications 17 18 Indirect maternal deaths: http://bmjopen.bmj.com/ 19 20 29 malaria 21 30 hiv/aids 22 23 31 severe anaemia 24 Neonatal outcomes Stillbirths & neonatal

25 deaths: 26 on October 1, 2021 by guest. Protected copyright. 27 32. Total live birth 28 33 Fresh stillbirths  2.5 Kg 29 30 34 Fresh stillbirths  2.5 Kg 31 32 35 Macerated stillbirths 33 36 Early neonatal deaths (< 24 hrs) 34 37 Early neonatal deaths (> 24 hrs) 35 36 Referrals: 37 38 Referrals out due to direct obstetric causes 38 39 39 Referrals in due to direct obstetric causes 40 41 42 Utafiti wa Vituo vya Tiba wa ETATMBA Page 5 of 5 43 Modulli ya 1 ya 4: Moduli Kuu Jina la Kituo cha Tiba______44 45 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 33 of 66 BMJ Open

1 2 3 ETATMBA Facility Assessment (B) 4 BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 5 EmOC Signal Functions and Other Essential Services 6 7 8 9 Instructions: Answer the following questions regarding the EmOC Signal Functions by interviewing health 10 workers in the maternity ward and other departments, reviewing facility registers, and through observation. 11 Record whether the function has been performed in the past 3 months; if not, why it has not been performed; 12 and whether it was performed in the last 12 months. Remember that “parenteral” means by injection, either 13 intramuscular or intravenous. 14 15 16 17 For peer review only 18 Signal Function 1: Parenteral Antibiotics 19

20 No. Item Responses Skip to 21 22 1 Have antibiotics been Yes ...... 1 If “Yes,” skip 23 administered parenterally in the No...... 0 to Item 4 24 Last 3 months? 25 2 If parenteral antibiotics were NOT Spontaneously Not mentioned 26 administered in the last 3 months, mentioned 27 Why? 28 29 (circle 1 for all spontaneous answers; otherwise circle 0) 30 1 0 31 a. availability of human resources

32 b. training issues 1 0 33 c. supplies/equipment/drugs 1 0 34 35 d. management issues 1 0 36 e. policy issues 1 0 37 f. no indication 1 0 38 g. other (specify) 1 0 39 http://bmjopen.bmj.com/ ______40 41 3 If parenteral antibiotics were NOT Yes ...... 1

42 administered in the last 3 months, No ...... 0 43 were they administered in the last 44 12 months? 45 46

47 on October 1, 2021 by guest. Protected copyright. 48 · We have found that the following categories are useful and cover most of the likely answers. 49 50 51 52 53 54 55 56 57 58 59 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml

BMJ Open Page 34 of 66

1 2 3 Signal Function 2: Administer Uterotonic Drugs 4 BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 5 6 No. Item Responses Skip to 7 8 4 Have oxytocics been administered Yes ...... 1 If “No,” skip 9 parenterally in the last 3 months? to Item 6 No ...... 0 10 11 12 5 If parenteral oxytocics were Oxytocin...... 1 All administered in the last 3 months, responses 13 Ergometrine ...... 2 14 which type of oxytocic was used? to this item Both ...... 3 skip to 15 (circle one) Item 9 16 Other (specify) ...... 4 17 For peer review only 18 ______19 20 6 If parenteral oxtyocics were NOT Spontaneously Not mentioned 21 administered in the last 3 months, Mentioned why? 22 23 (circle 1 for all spontaneous 24 answer; otherwise circle 0) 25 1 0 26 a. availability of human resources

27 b. training issues 1 0 28 c. supplies/equipment/drugs 1 0 29 30 d. management issues 1 0 31 32 e. policy issues 1 0

33 f. no indication 1 0 34 g. other (specify) 1 0 35 36 ______37 38 7 If parenteral oxytocics were NOT Yes ...... 1 If “No,” skip

http://bmjopen.bmj.com/ 39 administered in the last 3 months, to Item 9 No ...... 0 40 were they administered in the last 41 12 months? 42 43 8 If parenteral oxytocics were Oxytocin...... 1 44 administered in last 12 months, Ergometrine ...... 2 45 which type of oxytocic was used? 46 Both ...... 3 (circle one) 47 on October 1, 2021 by guest. Protected copyright. 48 Other (specify) ...... 4 49 ______50

51 9 Is misoprostol used in this facility Yes ...... 1 52 for obstetric indications? 53 No ...... 0 54 55 56

57 58 Page 2 of 14 ETATMBA 59 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml

Page 35 of 66 BMJ Open

1 2 Signal Function 3: Parenteral Anticonvulsants 3 4 BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 5 6 No. Item Responses Skip to 7 10 Have anticonvulsants been Yes ...... 1 If “No,” skip 8 administered parenterally in the to Item 12 9 No ...... 0 last 3 months? 10 11 11 If parenteral anticonvulsants were Magnesium sulfate ...... 1 All 12 administered in the last 3 months, responses 13 Diazepam ...... 2 14 which type of anticonvulsant was to this item Both ...... 3 15 used? skip to Item 15 16 (circle one) Other (specify) ...... 4 17 For peer review only ______18 19 20 12 If parenteral anticonvulsants were Spontaneously Not mentioned NOT administered in the last 3 Mentioned 21 22 months, why? 23 (circle 1 for all spontaneous 24 answers; otherwise circle 0) 25 1 0 a. availability of human resources 26 27 b. training issues 1 0 28 29 c. supplies/equipment/drugs 1 0 30 d. management issues 1 0 31 e. policy issues 1 0 32 33 f. no indication 1 0 34 g. other (specify) 35 1 0 36 ______37

38 13 If parenteral anticonvulsants were Yes ...... 1 If “No,” skip 39 NOT administered in the last 3 to Item 15 http://bmjopen.bmj.com/ No ...... 0 40 months, were they administered in 41 the last 12 months? 42 43 14 If parenteral anticonvulsants were Magnesium sulfate ...... 1 44 administered in last 12 months, Diazepam ...... 2 45 which type of medication was 46 used? Both ...... 3

47 on October 1, 2021 by guest. Protected copyright. 48 (circle one) Other (specify) ...... 4 49 ______50 51 52 53 54 55

56 57 58 59 60

Page 3 of 14 ETATMBA For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml

BMJ Open Page 36 of 66

1 2 Signal Function 4: Manual Removal of Placenta 3 4 BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 5 No. Item Responses Skip to 6 7 15 Has manual removal of placenta Yes ...... 1 If “Yes,” skip 8 been performed in the last 3 to Item 18 No ...... 0 9 months? 10 11 16 If manual removal of placenta Spontaneously Not mentioned 12 was NOT performed in the last 3 Mentioned 13 months, why? 14 (circle 1 for all spontaneous 15 answers; otherwise circle 0) 16 1 0 17 Fora. availability peer of human review only 18 resources 19 b. training issues 1 0 20 21 c. supplies/equipment/drugs 1 0 22 d. management issues 1 0 23 24 e. policy issues 1 0 25 f. no indication 1 0 26 27 g. other (specify) 1 0 28 ______29 30 31 17 If manual removal of placenta was Yes ...... 1 NOT performed in the last 3 32 No...... 0 33 months, has it been performed in the last 12 months? 34 35 36 Signal Function 5: Removal of Retained Products 37

38 39 No. Item Responses Skip to http://bmjopen.bmj.com/ 40 18 Has removal of retained products Yes ...... 1 If “No,” skip 41 been performed in the last 3 to Item 20 42 No ...... 0 months? 43 44 19 If removal of retained products Yes No All answers 45 46 was performed in last 3 months, to this item which method was used? skip to 47 on October 1, 2021 by guest. Protected copyright. Item 23 48 (read options) 49 1 0 a. Vacuum aspiration 50 51 b. Dilatation and curettage (D&C) 1 0 52 c. Dilatation and evacuation 1 0 53 (D&E) 54 55 d. Misoprostol 1 0 56

57 58 59 60

Page 4 of 14 ETATMBA For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml

Page 37 of 66 BMJ Open

1 2

3 No. Item Responses Skip to

4 BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from

5 20 If removal of retained products Spontaneously Not mentioned 6 was NOT performed in the last 3 Mentioned 7 months, why? 8 (circle 1 for all spontaneous 9 answers; otherwise circle 0) 10 0 1 11 a. availability of human 12 resources 13 b. training issues 1 0 14 15 c. supplies/equipment/drugs 1 0 16 d. management issues 1 0 17 For peer review only 18 e. policy issues 1 0 19 f. no indication 1 0 20 21 g. other (specify) 1 0 22 ______23 24 21 If removal of retained products Yes ...... 1 If “No,” skip 25 was NOT performed in the last 3 to Item 23 26 No ...... 0 months, has it been performed in 27 the last 12 months? 28 29 22 If removal of retained products Yes No 30 31 was performed in last 12 months, 32 which method was used? 33 (read options) 34 1 0 a. Vacuum aspiration 35 36 b. Dilatation and curettage 1 0 37 (D&C) 38 c. Dilatation and evacuation 1 0 39 http://bmjopen.bmj.com/ (D&E) 40 41 d. Misoprostol 1 0 42 43 Signal Function 6: Assisted Vaginal Delivery 44

45 46 No. Item Responses Skip to

47 on October 1, 2021 by guest. Protected copyright. 23 Has assisted vaginal delivery Yes ...... 1 If “No,” skip 48 (vacuum or forceps) been to Item 25 49 No ...... 0 performed in the last 3 months? 50

51 52 24 If assisted vaginal delivery was Vacuum extractor ...... 1 All performed in last 3 months, what responses 53 Forceps...... 2 54 instrument was used? to this item Both ...... 3 skip to 55 (circle one) 56 Item 28 57

58 59 60

Page 5 of 14 ETATMBA For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml

BMJ Open Page 38 of 66

1 2

3 No. Item Responses Skip to

4 BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from

5 25 If assisted vaginal delivery Spontaneously Not mentioned 6 (vacuum or forceps) was NOT Mentioned 7 performed in the last 3 months, 8 why?

9 (circle 1 for all spontaneous 10 answers; otherwise circle 0) 0 11 1 12 a. availability of human resources 13 b. training issues 1 0 14 15 c. supplies/equipment/drugs 1 0 16 d. management issues 1 0 17 For peer review only 18 e. policy issues 1 0 19 f. no indication 1 0 20 21 g. other (specify) 1 0 22 ______23 24 26 If assisted vaginal delivery Yes ...... 1 If “No,” skip 25 (vacuum or forceps) was NOT to Item 28 26 No ...... 0 performed in the last 3 months, 27 has it been performed in the last 28 12 months? 29 30 27 If assisted vaginal delivery was Vacuum extractor ...... 1 31 performed in last 12 months, what 32 Forceps...... 2 instrument was used? 33 Both ...... 3 34 (circle one) 35 36 Signal Function 7: Newborn Resuscitation 37 38

http://bmjopen.bmj.com/ 39 No. Item Responses Skip to 40 28 Has newborn resuscitation with Yes ...... 1 If “Yes,” 41 bag and mask been performed in skip to 42 No...... 0 43 the last 3 months? Item 31 44 29 If newborn resuscitation with bag Spontaneously Not mentioned 45 46 and mask was NOT performed in Mentioned the last 3 months, why? 47 on October 1, 2021 by guest. Protected copyright. 48 (circle 1 for all spontaneous 49 answers; otherwise circle 0) 50 1 0 a. availability of human resources 51 52 b. training issues 1 0 53 c. supplies/equipment/drugs 1 0 54 55 d. management issues 1 0 56 e. policy issues 1 0 57 58 f. no indication 1 0 59 g. other (specify 1 0 60 ______

Page 6 of 14 ETATMBA For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml

Page 39 of 66 BMJ Open

1 2

3 No. Item Responses Skip to

4 BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from

5 30 If newborn resuscitation with bag Yes ...... 1 6 and mask was not performed in No...... 0 7 the last 3 months, has it been 8 performed in the last 12 months? 9 10 Signal Function 8: Obstetric Surgery (Cesarean Delivery) 11

12 13 No. Item Responses Skip to 14 31 Has a cesarean been performed in Yes ...... 1 If “Yes,” 15 the last 3 months? skip to 16 No ...... 0 Item 34 17 For peer review only 18 32 If a cesarean was NOT performed Spontaneously Not mentioned 19 in the last 3 months, why? Mentioned 20 21 (circle 1 for all spontaneous 22 answers; otherwise circle 0) 23 1 0 a. availability of human resources 24 25 b. training issues 1 0 26 c. supplies/equipment/drugs 1 0 27 28 d. management issues 1 0 29 e. policy issues 1 0 30 31 f. no indication 1 0 32 g. other (specify) 1 0 33 34 ______35

36 33 If a cesarean was NOT performed Yes ...... 1 If “No,” skip 37 in the last 3 months, has it been to Item 35 No ...... 0 38 performed in the last 12 months? 39 http://bmjopen.bmj.com/ 40 34 What type of anesthesia is Yes No 41 currently used when performing a 42 cesarean delivery? (read options 43 out loud) 44 1 0 a. General 45 46 b. Spinal/epidural 1 0

47 on October 1, 2021 by guest. Protected copyright. c. Ketamine 1 0 48 49 d. Other (specify) 1 0 50 ______51 52 53 54 Signal Function 9: Blood Transfusion 55 56 No. Item Responses Skip to 57 35 Has blood transfusion been Yes ...... 1 If “No,” skip 58 performed in the last 3 months? to Item 37 59 No ...... 0 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 7 of 14 ETATMBA

BMJ Open Page 40 of 66

1 2

3 No. Item Responses Skip to

4 BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from

5 36 If blood transfusion was performed Blood comes from central blood bank ... 1 All 6 in the last 3 months, describe the responses Blood comes from a facility blood bank . 2 7 primary supply of blood. to this item Blood is collected from family or skip to 8 (circle one) 9 friends as needed Item 40 10 (i.e., direct transfusion) ...... 3 11 Other (specify) ...... 4 12 13 ______14 15 37 If blood transfusion was NOT Spontaneously Not mentioned 16 performed in the last 3 months, Mentioned 17 why? For peer review only 18 (circle 1 for all spontaneous 19 answers; otherwise circle 0) 20 1 0 21 a. availability of human resources 22 b. training issues 1 0 23 24 c. supplies/equipment/drugs 1 0 25 d. management issues 1 0 26 27 e. policy issues 1 0 28 f. no indication 1 0 29 30 g. other (specify 1 0 31 ______32 33 38 If blood transfusion was NOT Yes ...... 1 If “No,” skip 34 performed in the last 3 months, has to Item 36 35 No ...... 0 it been performed in the last 12 36 37 months? 38 39 If blood transfusion was performed Blood comes from central blood bank .. 1 39 http://bmjopen.bmj.com/ in the last 12 months, describe the 40 Blood comes from a facility blood bank . 2 41 primary supply of blood: Blood is collected from family 42 (circle one) 43 or friends as needed 44 (i.e., direct transfusion) ...... 3 45 Other (specify) ...... 4 46 ______47 on October 1, 2021 by guest. Protected copyright. 48 Other Maternal and Newborn Health-Related Services 49 50 51 No. Item Responses Skip to 52 53 40 Does staff routinely practice active Yes ...... 1 54 management of the third stage of No ...... 0 55 labor? 56 57 41 Has a partograph been used to Yes ...... 1 If “Yes,” 58 manage labor in the last 3 months? skip to No ...... 0 59 Item 43 60

Page 8 of 14 ETATMBA For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml

Page 41 of 66 BMJ Open

1 2

3 No. Item Responses Skip to

4 BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from

5 42 If a partograph has NOT been used Spontaneously Not mentioned 6 to manage labor in the last 3 Mentioned 7 months, why? 8 (circle 1 for all spontaneous 9 answer; otherwise circle 0) 10 0 1 11 a. availability of human resources 12 b. training issues 1 0 13 14 c. supplies/equipment/drugs 1 0

15 d. management issues 1 0 16

17 For e.peer policy issues review1 only0 18 f. no indication 1 0 19 20 g. other (specify) 1 0 21 ______22

23 43 Has a breech delivery been Yes ...... 1 If “Yes,” 24 performed in the last 3 months? skip to No ...... 0 25 Item 45 26 27 44 If a breech delivery was NOT Spontaneously Not mentioned 28 performed in the last 3 months, Mentioned 29 why? 30 (circle 1 for all spontaneous 31 answers; otherwise circle 0) 32 1 0 33 a. availability of human resources 34 b. training issues 1 0 35 36 c. supplies/equipment/drugs 1 0 37 d. management issues 1 0 38

39 e. policy issues 1 0 http://bmjopen.bmj.com/ 40 f. no indication 1 0 41 42 g. other (specify) 1 0 43 ______44 45 45 For mothers with unknown HIV Yes ...... 1 If “Yes,” 46 status, has rapid testing been skip to No ...... 0 47 performed in the maternity/labor Item 47 on October 1, 2021 by guest. Protected copyright. 48 ward in the last 3 months? 49 50 51 52

53 54 55 56 57 58

59 60

Page 9 of 14 ETATMBA For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml

BMJ Open Page 42 of 66

1 2

3 No. Item Responses Skip to

4 BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from

5 46 If rapid HIV testing was NOT Spontaneously Not mentioned 6 provided in the maternity/labor ward Mentioned 7 in the last 3 months, why? 8 (circle 1 for all spontaneous 9 answers; otherwise circle 0) 10 0 1 11 a. availability of human resources 12 b. training issues 1 0 13 14 c. supplies/equipment/drugs 1 0

15 d. management issues 1 0 16

17 For e.peer policy issues review1 only0 18 f. no indication 1 0 19 20 g. other (specify) 1 0 21 ______22

23 47 Have ARVs been given to Yes ...... 1 If “Yes,” 24 seropositive mothers in skip to No ...... 0 25 maternity/labor ward in the last Item 49 26 3 months? 27 28 48 If ARVs were NOT given to Spontaneously Not mentioned

29 seropositive mothers in the Mentioned 30 maternity/labor ward in the last 3 31 months, why? 32 (circle 1 for all spontaneous 33 answers; otherwise circle 0) 34 0 1 35 a. availability of human resources 36 b. training issues 1 0 37 38 c. supplies/equipment/drugs 1 0

39 http://bmjopen.bmj.com/ d. management issues 1 0 40

41 e. policy issues 1 0 42 f. no indication 1 0 43 44 g. other (specify) 1 0 45 ______46

47 49 Have ARVs been given to Yes ...... 1 If “Yes,” on October 1, 2021 by guest. Protected copyright. 48 newborns in maternity/labor ward in skip to No ...... 0 49 the last 3 months? (PMTCT) Item 51 50 51 52

53 54 55 56 57 58

59 60

Page 10 of 14 ETATMBA For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml

Page 43 of 66 BMJ Open

1 2

3 No. Item Responses Skip to

4 BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from

5 50 If ARVs were NOT given to Spontaneously Not mentioned 6 newborns in the maternity/labor Mentioned 7 ward in the last 3 months, why? 8 (circle 1 for all spontaneous 9 answers; otherwise circle 0) 10 0 1 11 a. availability of human resources 12 b. training issues 1 0 13 14 c. supplies/equipment/drugs 1 0

15 d. management issues 1 0 16

17 For e.peer policy issues review1 only0 18 f. no indication 1 0 19 20 g. other (specify) 1 0

21 ______22 23 51 Has special or intensive care been Yes ...... 1 If “Yes,” 24 provided to a preterm or low birth skip to 25 No ...... 0 weight baby in the last 3 months? Item 53 26 27 52 If special or intensive care has NOT Spontaneously Not mentioned 28 been provided to a premature or 29 Mentioned low birth weight baby in the last 3 30 31 months, why? 32 (circle 1 for all spontaneous

33 answers; otherwise circle 0) 0 34 1 a. availability of human resources 35 36 b. training issues 1 0 37 c. supplies/equipment/drugs 1 0 38

39 d. management issues 1 0 http://bmjopen.bmj.com/ 40 e. policy issues 1 0 41 42 f. no indication 1 0 43 g. no pediatric or intensive care unit 1 0 44 for infants 45 46 h. other (specify) 1 0

47 on October 1, 2021 by guest. Protected copyright. ______48 49 53 Has a craniotomy been performed Yes ...... 1 If “Yes,” 50 in the last 3 months? skip to 51 No ...... 0 52 Item 55

53 54 55 56 57 58

59 60

Page 11 of 14 ETATMBA For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml

BMJ Open Page 44 of 66

1 2

3 No. Item Responses Skip to

4 BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from

5 54 If a craniotomy was NOT performed Spontaneously Not mentioned 6 in the last 3 months, why? Mentioned 7 (circle 1 for all spontaneous 8 answers; otherwise circle 0) 9 1 0 10 a. availability of human resources 11 b. training issues 1 0 12 13 c. supplies/equipment/drugs 1 0 14 d. management issues 1 0 15 16 e. policy issues 1 0 17 For peer review only f. no indication 1 0 18 19 g. other (specify 1 0 20 ______21 22 55 Has an episiotomy been performed Yes ...... 1 If “Yes,” 23 in the last 3 months? skip to 24 No ...... 0 Item 57 25 26

27 56 If an episiotomy was NOT Spontaneously Not mentioned 28 performed in the last 3 months, Mentioned why? 29 30 (circle 1 for all spontaneous 31 answer;, otherwise circle 0) 32 1 0 a. availability of human resources 33 34 b. training issues 1 0 35 c. supplies/equipment/drugs 1 0 36 37 d. management issues 1 0 38 e. policy issues 1 0 39 http://bmjopen.bmj.com/ 40 f. no indication 1 0 41 g. other (specify 1 0 42 43 ______44 45 57 Is there a health worker at this Yes ...... 1 If “No,” skip

46 facility who is trained to repair to Item 59 No ...... 0 47 obstetric fistula? on October 1, 2021 by guest. Protected copyright. 48 49 58 If there is a health worker trained to Yes ...... 1 50 repair obstetric fistula, has at least No ...... 0 51 one fistula been repaired in this 52 facility in the last 3 months? 53 54 59 Have temporary family planning Yes ...... 1 If “Yes,” 55 methods been provided in the last 3 skip to No ...... 0 56 months? Item 61 57 58

59 60

Page 12 of 14 ETATMBA For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml

Page 45 of 66 BMJ Open

1 2

3 No. Item Responses Skip to

4 BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from

5 60 If temporary family planning Spontaneously Not mentioned 6 methods were NOT provided in the Mentioned 7 last 3 months, why? 8 (circle 1 for all spontaneous 9 answers; otherwise circle 0) 10 0 1 11 a. availability of human resources 12 b. training issues 1 0 13 14 c. no methods available 1 0

15 d. management issues 1 0 16

17 For e.peer policy issues review1 only0 18 f. no indication 1 0 19 20 g. other (specify 1 0

21 ______22 23 61 Has a surgical method of Yes ...... 1 If “Yes,” 24 permanent contraception been skip to 25 No ...... 0 performed in the last 3 months? Item 63 26 27 62 If a surgical method of permanent Spontaneously Not mentioned 28 29 contraception was NOT performed Mentioned in the last 3 months, why? 30 31 (circle 1 for all spontaneous 32 answers; otherwise circle 0) 33 1 0 a. availability of human resources 34

35 b. training issues 1 0 36 c. supplies/equipment/drugs 1 0 37 38 d. management issues 1 0

39 http://bmjopen.bmj.com/ e. policy issues 1 0 40

41 f. no indication 1 0 42 g. other (specify) 1 0 43 44 ______45 46 63 Is there a health worker that can do Yes ...... 1 47 tubal ligation (surgical method of on October 1, 2021 by guest. Protected copyright. No ...... 0 48 permanent contraception)? 49 50 64 Is there a health worker that can do Yes ...... 1 51 vasectomy? 52 No ...... 0 53 65 Does the facility provide Yes ...... 1 54 postabortion contraception to 55 No ...... 0 56 women?

57 58

59 60

Page 13 of 14 ETATMBA For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml

BMJ Open Page 46 of 66

1 2 3 4 BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from Comments: 5 6 7 8 9

10 11 12 13 14 15

16 17 For peer review only 18 19 20 21

22 23 24 25 26 27

28 29 30 31 32 33

34 35 36 37 38

39 http://bmjopen.bmj.com/

40 41 42 43 44 45

46 47 on October 1, 2021 by guest. Protected copyright. 48 49 50 51

52 53 54 55 56 57

58 59 60

Page 14 of 14 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml ETATMBA

Page 47 of 66 BMJ Open

1 2 3

4 BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 5 6 7 8 9 10 11 12 13 14 15 16 17 For peer review only 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38

39 http://bmjopen.bmj.com/ 40 41 42 43 44 45 46

47 on October 1, 2021 by guest. Protected copyright. 48 49 50 51 52 53 54 55 56 57 58 59 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 48 of 66 Facility ID: BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 1 2 3 (A.) EQUIPMENT, SUPPLIES & ESSENTIAL DRUGS 4 5 Name of district: ______Name of Village: 6 ______7 8 Name of health facility: ______9 10 Date of Interview: ___ / ___ / ___ Start time ______:______11 12 Instructions: 13 14 Equipment and supplies: 15  Record whetherFor there is a sufficientpeer supply/number review of the following only items for the facility’s daily caseload of 16 deliveries, and whether the items are available & functional, available but NOT functional, or not available. 17  Equipment, supplies and infrastructure are organized by room. 18 19 Essential drugs: 20  Record the availability and supply of drugs for each room (emergency room, labor / delivery room, maternity 21 ward, operating theatre and pharmacy). Check whether the listed drug is available and if the supply is 22 sufficient to last for less than one week, up to one week, up to two weeks, up to three weeks, or up to four or 23 more weeks. 24  Drug lists are organized by room. 25  List according to MOH guidelines. 26 27 A. Emergency room 28 29 30 Code Instructions Available 31 A1 Does this facility have an emergency room? Kama hakuna, □ 0.NO □ 1. YES 32 nenda kwenye chumba kingine (Nenda B1) 33 A2 Are obstetric complications managed in the emergency room? □ 0.NO □ 1. YES 34 http://bmjopen.bmj.com/ 35 Infrastructure 36 37 ID Instructions Available 38 A3 Electricity □ 0.NO □ 1. YES 39 A4 Sufficient light source to perform tasks during the day □ 0.NO □ 1. YES 40 A5 Sufficient light source to perform tasks at night □ 0.NO □ 1. YES 41 42 A6 Means of ventilation (Fan,AC Windows) □ 0.NO □ 1. YES

43 A7 Running water □ 0.NO □ 1. YES on October 1, 2021 by guest. Protected copyright. 44 A8 Functioning toilet □ 0.NO □ 1. YES 45 A9 Curtains/means of providing patient privacy □ 0.NO □ 1. YES 46 A10 Waiting area for visitors □ 0.NO □ 1. YES 47 48 Equipments 49 50 ID A. Availability B. Supply 51 Are there enough 52 At least 1 Available but Not for the daily 53 Essential Items available & NONE Availabl caseload of 54 Functional functional e deliveries? 55 Filled oxygen cylinder with cylinder 56 A11 2 1 0 □0.NO □1.YES 57 carrier and key to open valve 58 A12 BP cuff 2 1 0 □0.NO □1.YES 59 60 1

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 49 of 66 BMJ Open Facility ID: BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 1 2 3 4 ID A. Availability B. Supply 5 Are there enough 6 At least 1 Available but Not for the daily 7 Essential Items available & NONE Availabl caseload of 8 Functional functional e deliveries? 9 10 A13 Stethoscope 2 1 0 □0.NO □1.YES 11 A14 Fetal stethoscope 2 1 0 □0.NO □1.YES 12 13 A15 Cannular 2 1 0 □0.NO □1.YES 14 Kidney basins/Beseni la kunawia A16 2 1 0 □ NO □1.YES 15 mikono For peer review only 16 A17 Clinical thermometer 2 1 0 □0.NO □1.YES 17 A18 Needles and Syringes (5-10-20cc) 2 1 0 □0.NO □1.YES 18 19 A19 Suture needles/suture materials 2 1 0 □0.NO □1.YES 20 A20 IV Drip Stand(s) 2 1 0 □0.NO □1.YES 21 □0.NO □1.YES 22 A21 Urinary catheters 2 1 0 23 A22 Adult ventilator bag and mask 2 1 0 □0.NO □1.YES 24 25 A23 Mouth gag 2 1 0 □0.NO □1.YES 26 Patient transport (wheelchair, trolley, A24 2 1 0 □0.NO □1.YES 27 hammock) 28 A25 Examination table with privacy 2 1 0 □0.NO □1.YES 29 A26 Uris tix /Albustix (dip stick for protein 2 1 0 30 □0.NO □1.YES in urine 31 Infection prevention 32 33 A 27 Soap 2 1 0 □0.NO □1.YES 34 35 A28 Antiseptics (Kama Detol nk) 2 1 0 □0.NO □1.YES http://bmjopen.bmj.com/ 36 Sterile gloves (Pea ya glovu A29 2 1 0 □0.NO □1.YES 37 zilizofungwa) 38 Non-sterile gloves (Glovu □0.NO □1.YES A30 2 1 0 39 hazijafungwa) 40 A31 Non-sterile protective clothing 2 1 0 □0.NO □1.YES 41 42 A32 Decontamination container 2 1 0 □0.NO □1.YES 43 A33 Bleach or bleaching powder (Jik) 2 1 0 □0.NO □1.YES on October 1, 2021 by guest. Protected copyright. 44 □0.NO □1.YES 45 A34 Prepared disinfection solution 2 1 0 46 A35 Regular trash bin 2 1 0 □0.NO □1.YES 47 48 A36 Puncture proof sharps container 2 1 0 □0.NO □1.YES 49 50 Essential drugs 51 ID Drugs Available 52 A37 Antibiotics 53 A37a Amoxicillin □0.NO □1.YES 54 A37b Ampicillin □0.NO □1.YES 55 A37c Benzyl penicillin (x-pen) □0.NO □1.YES 56 A37d Cloxacillin □0.NO □1.YES 57 A37e Erythromicin □0.NO □1.YES 58 A37f Gentamicin □0.NO □1.YES 59 60 2

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 50 of 66 Facility ID: BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 1 2 3 ID Drugs Available 4 A37g Metranidazole (Flagyl) □0.NO □1.YES 5 A37h Nitrofurantoin □0.NO □1.YES 6 A37i Penicillin G □0.NO □1.YES 7 A37j Procaine penicillin G (PPF) □0.NO □1.YES 8 A37k Trimethoprim/Sulfamethoxazole (Septrine) □0.NO □1.YES 9 A38 Anticonvulsants 10 A38a Magnesium sulfate □0.NO □1.YES 11 □ □ 12 A38b Phenytoin □0.NO □1.YES 13 A39 Antihypertensives 14 A39a Hydralazine □0.NO □1.YES 15 A39b NifedipineFor peer review □only0.NO □1.YES 16 A40 Dawa zinazotumika wakati wa dharura 17 A40a Adrenaline □0.NO □1.YES 18 A40b Aminophylline □0.NO □1.YES 19 A40c Atropine sulfate □0.NO □1.YES 20 A40d Calcium gluconate □0.NO □1.YES 21 A40e Digoxin □0.NO □1.YES 22 A40f Ephedrine □0.NO □1.YES 23 A40g Frusemide(Lasix) □0.NO □1.YES 24 A40h Naloxone □0.NO □1.YES 25 A40i Nitroglycerine □0.NO □1.YES 26 A40j Prednisolone □0.NO □1.YES 27 A40k Promethazine (Phenergan) □0.NO □1.YES 28 A41 Analgesics 29 A41a Paracetamol/Asprin □0.NO □1.YES 30 A41b Pethidine/ramadol □0.NO □1.YES 31 A42 Sedatives 32 A42a Diazepam (Valium) □0.NO □1.YES 33 A42b Phenobarbitone □0.NO □1.YES 34 A43 IV Fluids http://bmjopen.bmj.com/ 35 A43a Dextrose 5% □0.NO □1.YES 36 A43b Normal saline □0.NO □1.YES 37 A43c Ringer’s lactate □0.NO □1.YES 38 A44 PMTCT/HIV care 39 A44a ART □0.NO □1.YES 40 A44b Niverapine □0.NO □1.YES 41 A44c Rapid testing kit □0.NO □1.YES 42

43 on October 1, 2021 by guest. Protected copyright.

44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 3

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 51 of 66 BMJ Open Facility ID: BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 1 2 3 B. Labor / Delivery room 4 Kama chumba cha kusubiri wakati wa uchungu ni tofauti na kile cha kujifungulia, tafadhali chukulia kama ni chumba 5 kimoja 6 7 ID Available 8 B1 Does this facility have a labor / delivery room? Kama jibu ni NO, □ 0.NO □ 1. YES 9 nenda kwenye chumba kingine (Nenda C1) 10 11 Infrastructure 12 13 ID Available 14 B2 Electricity □ 0.NO □ 1. YES 15 B3 Sufficient lightFor source to pepeerrform tasks during review the day □ 0.onlyNO □ 1. YES 16 B4 Sufficient light source to perform tasks at night □ 0.NO □ 1. YES 17 B5 Means of ventilation (Fan,AC or Window) □ 0.NO □ 1. YES 18 19 B6 Running water □ 0.NO □ 1. YES 20 B7 Functioning toilet □ 0.NO □ 1. YES 21 B8 Panga boi (Fan) □ 0.NO □ 1. YES 22 B9 Curtains/means of providing patient privacy (Screen) □ 0.NO □ 1. YES 23 B10 Waiting area for visitors (iliyo na viti na kivuli) □ 0.NO □ 1. YES 24 25 Vifaa 26 ID A. Available B. Supply 27 At least 1 Available but NOT Are there enough for 28 Essential Items available & NONE Availabl the daily caseload of 29 Functional functional e deliveries? 30 B11 Delivery bed with stirrups 2 1 0 □0.NO □1.YES 31 B12 Delivery bed (no stirrups) 2 1 0 □0.NO □1.YES 32 B13 BP cuff 2 1 0 □0.NO □1.YES 33 B14 Stethoscope 2 1 0 □0.NO □1.YES 34 □ □ 35 B15 Baby weighing scale 2 1 0 □0.NO □1.YES http://bmjopen.bmj.com/ 36 B16 Fetal scope 2 1 0 □0.NO □1.YES 37 B17 Kidney dish 2 1 0 □0.NO □1.YES 38 B18 Clinical thermometer 2 1 0 □0.NO □1.YES 39 B19 Scissors 2 1 0 □0.NO □1.YES 40 B20 Needles and Syringes (10-20cc) 2 1 0 □0.NO □1.YES 41 B21 Suture needles/suture materials 2 1 0 □0.NO □1.YES 42 B22 IV Drip Stand(s) 2 1 0 □0.NO □1.YES

43 B23 Urinary catheters 2 1 0 □0.NO □1.YES on October 1, 2021 by guest. Protected copyright. 44 Uristix/Albustix (dip stick for protein □0.NO □1.YES B24 2 1 0 45 in urine) 46 Filled oxygen cylinder with cylinder □0.NO □1.YES B25 2 1 0 47 carrier and key to open valve 48 B26 Mouth gag 2 1 0 □0.NO □1.YES 49 Patient transport (wheelchair, □0.NO □1.YES B27 2 1 0 50 gurney, hammock) 51 Incubator/warm Room □0.NO □1.YES 52 B28 2 1 0 53 54 Infection prevention (Kuzuia maambukizi) 55 B29 Soap 2 1 0 □0.NO □1.YES 56 B30 Antiseptics 2 1 0 □0.NO □1.YES 57 B31 Sterile gloves 2 1 0 □0.NO □1.YES 58 B32 Non-sterile protective clothing 2 1 0 □0.NO □1.YES 59 60 4

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 52 of 66 Facility ID: BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 1 2 3 ID A. Available B. Supply 4 At least 1 Available but NOT Are there enough for 5 Essential Items available & NONE Availabl the daily caseload of 6 Functional functional e deliveries? 7 (Gauns/Apron) 8 □0.NO □1.YES 9 B33 Decontamination container 2 1 0 □ □ Bleach or bleaching powder □0.NO □1.YES 10 B34 2 1 0 11 (Jick/Chlorinated lime) 12 B35 Prepared disinfection solution 2 1 0 □0.NO □1.YES 13 B36 Covered contaminated trash bin 2 1 0 □0.NO □1.YES 14 B37 Puncture proof sharps container 2 1 0 □0.NO □1.YES Mayo stand (or equivalent to establish □0.NO □1.YES 15 B38 For peer2 review1 only0 16 sterile field) 17 B39 Sterilizer/autoclave 2 1 0 □0.NO □1.YES 18 B40 Placenta pit 2 1 0 □0.NO □1.YES 19 B41 Daily caseload of deliveries? 20 B41a Artery forceps 2 1 0 □0.NO □1.YES 21 B41b Cord-cutting/blunt-ended scissors 2 1 0 □0.NO □1.YES 22 □0.NO □1.YES 23 B41c Cord ties 2 1 0 □ □ 24 B41d Gloves 2 1 0 □0.NO □1.YES 25 B41e Plastic sheets/ Macking tosh 2 1 0 □0.NO □1.YES 26 B41f Gauze swabs 2 1 0 □0.NO □1.YES 27 B41g Cloth (Green towels) 2 1 0 □0.NO □1.YES 28 Perneal/Vaginal/Cervical repair B42 29 pack 30 B42a Sponge forceps 2 1 0 □0.NO □1.YES 31 B42b Artery forceps large/small 2 1 0 □0.NO □1.YES 32 B42c Needle holder 2 1 0 □0.NO □1.YES 33 B42d Stitch scissors 2 1 0 □0.NO □1.YES 34 B42e Dissecting forceps, toothed 2 1 0 □0.NO □1.YES http://bmjopen.bmj.com/ 35 B42f Vaginal speculum, (Sims) 2 1 0 □0.NO □1.YES 36 B42g Vaginal speculum (Bivalve) 2 1 0 □0.NO □1.YES 37 B43 Vacum extractior/ 38 □0.NO □1.YES 39 B43a Vacuum extractor 2 1 0 □ □ 40 B43b forceps delivery 2 1 0 □0.NO □1.YES Uterine evacuation 41 B44 42

43 B44a Vaginal speculum (Sims) 2 1 0 □0.NO □1.YES on October 1, 2021 by guest. Protected copyright. Sponge (ring) forceps or uterine □0.NO □1.YES 44 B44b 2 1 0 45 packing forceps 46 B44c Single tooth tenaculum forceps 2 1 0 □0.NO □1.YES 47 B44c Long dressing forceps 2 1 0 □0.NO □1.YES 48 B44d Uterine dilators, sizes 13-27 (French) 2 1 0 □0.NO □1.YES 49 B44e Sharp uterine curettes, size 0 or 00 2 1 0 □0.NO □1.YES 50 B44f Blunt uterine curettes, size 0 or 00 2 1 0 □0.NO □1.YES 51 B44g Metal uterine sound 2 1 0 □0.NO □1.YES 52 B45 Manual Vacuum evacuation 53 □0.NO □1.YES Basic uterine evacuation instruments 54 B45a 2 1 0 (B43a-e & h)PLUS: 55 56 Vacuum syringes (single / double □0.NO □1.YES B45b 2 1 0 57 valve) 58 B45c Silicone lubricant 2 1 0 □0.NO □1.YES 59 60 5

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 53 of 66 BMJ Open Facility ID: BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 1 2 3 ID A. Available B. Supply 4 At least 1 Available but NOT Are there enough for 5 Essential Items available & NONE Availabl the daily caseload of 6 Functional functional e deliveries? 7 B45d Adapters 2 1 0 □0.NO □1.YES 8 □0.NO □1.YES 9 B45e Flexible cannulae, 4 - 12 mm 2 1 0 □ □ Kipo japo Kipo lakini Vinatolewa kiasi cha 10 NOT kimoja hakuna kutosha kwa siku? 11 B46 Neonatal resuscitation park Availabl kinafanya kinachofanya 12 e 13 kazi kazi 14 B46a Mucus extractor 2 1 0 □0.NO □1.YES 15 B46b Infant face maskFor peer2 review1 only0 □0.NO □1.YES 16 B46c Ventilatory bag 2 1 0 □0.NO □1.YES 17 B46d Suction catheter Ch 12 2 1 0 □0.NO □1.YES 18 B46e Suction catheter Ch 10 2 1 0 □0.NO □1.YES 19 Infant laryngoscope with spare bulb & □0.NO □1.YES B46f 2 1 0 20 batteries 21 B46g Endotracheal tubes 3.5 2 1 0 □0.NO □1.YES 22 B46h Endotracheal tubes 3.0 2 1 0 □0.NO □1.YES 23 Suction apparatus: Foot- or □0.NO □1.YES B46i 2 1 0 24 electrically-operated 25 B46j Oxygen cylinders 2 1 0 □0.NO □1.YES 26 27 Essential Drugs 28 If yes, enough supply to last for up to Code Drugs Available 29 (Check NUMBER OF WEEKS): 30 B47 Antibiotics 31 B47a Amoxicillin □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 32 B47b Ampicillin □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 33 B47c Benzyl penicillin (X-Pen) □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 34

B47d Cloxacillin □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ http://bmjopen.bmj.com/ 35 B47e Erythromicin □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 36 B47f Gentamicin □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 37 B47g Metronidazole (Flagyl) □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 38 B47h Nitrofurantoin □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 39 B47i Penicillin G □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 40 □ □ □ □ □ □ □ B47j 41 Procaine penicillin G (PPF) □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ Trimethoprim/Sulfamethozazol □0.NO □1.YES 42 B47k □ <1 □ 1 □ 2 □ 3 □ 4+ 43 e (Septrine) on October 1, 2021 by guest. Protected copyright. 44 B47l Magnesium sulfate □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 45 B47m Phenytoin □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 46 B48 Antihypertensives 47 B48a Hydralazine □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 48 B48b Nifedipine □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ Oxytocics, Prostaglandins & 49 B49 50 other 51 B49a Ergometrine □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 52 B49b Methylergometrine □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 53 B49c Misoprostol □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 54 B49d Oxytocin □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 55 B49e Prostaglandin E2 □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ Dawa zinazotumika wakati 56 B50 57 wa dharura 58 B50a Adrenaline □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 59 60 6

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 54 of 66 Facility ID: BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 1 2 3 If yes, enough supply to last for up to Code Drugs Available 4 (Check NUMBER OF WEEKS): 5 B50b Aminophylline □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 6 B50c Atropine sulfate □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 7 B50d Calcium gluconate □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 8 B50e Digoxin □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 9 B50f Ephedrine □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 10 B50g Frusemide (Lasix) □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 11 B50h Naloxone □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 12 □ □ □ □ □ □ □ 13 B50i Nitroglycerine □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 14 B50j Prednisolone □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 15 B50k Promethazine (Phenergan)For peer□0.NO □review1.YES □

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 55 of 66 BMJ Open Facility ID: BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 1 2 3 If yes, enough supply to last for up to Code Drugs Available 4 (Check NUMBER OF WEEKS): 5 prophylactic treatment 6 7 8 9 10 11 C. Operating theatre 12 13 Code Available 14 C1 Does this facility have an operating theatre? □0.NO □1.YES 15 Kama jibu niFor NO ruka nenda peer chumba kingine review (NENDA D1) only 16 Miundo mbinu 17 18 Code Available 19 C2 Electricity □0.NO □1.YES 20 C3 Sufficient light source to perform tasks during the day □0.NO □1.YES 21 C4 Sufficient light source to perform tasks at night □0.NO □1.YES 22 □0.NO □1.YES 23 C5 Running water 24 C6 Means of ventilation (Fan) □0.NO □1.YES 25 26 Vifaa 27 Code A. Availability B. Supply 28 At least 1 Available but NOT Are there enough for 29 Essential Items available & NONE Availa the daily caseload of 30 Functional functional ble deliveries? 31 Infection prevention 32 C7 Soap 2 1 0 □0.NO □1.YES 33 C8 Antiseptics 2 1 0 □0.NO □1.YES 34 C9 Sterile gloves 2 1 0 □0.NO □1.YES http://bmjopen.bmj.com/ 35 C10 Non-sterile gloves 2 1 0 □0.NO □1.YES 36 C11 Non-sterile protective clothing 2 1 0 □0.NO □1.YES 37 C12 Decontamination container 2 1 0 □0.NO □1.YES 38 C13 Bleach or bleaching powder 2 1 0 □0.NO □1.YES 39 C14 Prepared disinfection solution 2 1 0 □0.NO □1.YES 40 C15 Regular trash bin 2 1 0 □0.NO □1.YES 41 C16 Covered contaminated waste trash bin 2 1 0 □0.NO □1.YES 42 C17 Puncture proof sharps container 2 1 0 □0.NO □1.YES 43 on October 1, 2021 by guest. Protected copyright. Mayo stand (or equivalent to establish □0.NO □1.YES 44 C18 2 1 0 sterile field) 45 C19 Sterilizer/autoclave 2 1 0 □0.NO □1.YES 46 Obstetric laparotomy / cesarean 47 C20 section pack 48 Stainless steel instrument tray with □0.NO □1.YES 49 C20a 2 1 0 50 cover 51 C20b Towel clips 2 1 0 □0.NO □1.YES 52 C20c Sponge forceps 2 1 0 □0.NO □1.YES 53 C20d Straight artery forceps 2 1 0 □0.NO □1.YES 54 C20e Uterine haemostasis forceps 2 1 0 □0.NO □1.YES 55 C20f Needle holder 2 1 0 □0.NO □1.YES 56 C20g Surgical knife handle 2 1 0 □0.NO □1.YES 57 C20h Surgical knife blades 2 1 0 □0.NO □1.YES 58 C20i Triangular point suture needles/7.3 2 1 0 □0.NO □1.YES 59 60 8

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 56 of 66 Facility ID: BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 1 2 3 Code A. Availability B. Supply 4 At least 1 Available but NOT Are there enough for 5 Essential Items available & NONE Availa the daily caseload of 6 Functional functional ble deliveries? 7 cm/size 6 8 □0.NO □1.YES 9 C20j Round-bodied needles/No 12/size 6 2 1 0 □ □ Abdominal retractors/double-ended □0.NO □1.YES 10 C20k 2 1 0 11 (Richardson) Curved operating scissors/blunt □0.NO □1.YES 12 C20l 2 1 0 13 pointed (Mayo)17cm Straight operating scissors/blunt □0.NO □1.YES 14 C20m 2 1 0 15 pointed (Mayo)17cForm peer review only 16 C20n Scissors, straight, 23 cm 2 1 0 □0.NO □1.YES 17 C20o Suction nozzle 2 1 0 □0.NO □1.YES Suction tube, 22.5 cm, 23 French □0.NO □1.YES 18 C20p 2 1 0 19 gauge 20 C20r Intestinal clamps 2 1 0 □0.NO □1.YES 21 C20s Dressing (non-toothed tissue) forceps 2 1 0 □0.NO □1.YES 22 C20t Sutures (different sizes and types) 2 1 0 □0.NO □1.YES 23 C21 Anesthesia equipment 24 C21a Anesthetic face masks 2 1 0 □0.NO □1.YES 25 C21b Oropharyngeal airways 2 1 0 □0.NO □1.YES 26 Laryngoscopes (with spare bulbs and □0.NO □1.YES C21c 2 1 0 27 batteries) 28 C21d Intubating forceps (Magill) 2 1 0 □0.NO □1.YES 29 Endotracheal tube connectors: 15 mm □0.NO □1.YES 30 plastic (can be connected directly to C21e 2 1 0 31 the breathing valve; three for each 32 tube size) 33 C21f Spinal needles (18-gauge to 25-gauge) 2 1 0 □0.NO □1.YES 34

C21g Suction apparatus: Foot-operated 2 1 0 □0.NO □1.YES http://bmjopen.bmj.com/ 35 Suction apparatus: Electrically □0.NO □1.YES 36 C21h 2 1 0 operated 37 Anesthesia apparatus (EMO/draw-over □0.NO □1.YES 38 C21i 2 1 0 system) 39 Oxygen cylinders c manometer and 40 C21j flowmeter tubes and connectors/o2 2 1 0 □0.NO □1.YES 41 □ □ concentrator 42

Perineal / vaginal / cervical repair on October 1, 2021 by guest. Protected copyright. 43 C22 pack 44 45 C22a Sponge forceps 2 1 0 □0.NO □1.YES 46 C22b Artery forceps large/small 2 1 0 □0.NO □1.YES 47 C22c Needle holder 2 1 0 □0.NO □1.YES 48 C22d Stitch scissors 2 1 0 □0.NO □1.YES 49 C22e Dissecting forceps, toothed 2 1 0 □0.NO □1.YES 50 C22f Vaginal speculum, large (Sims) 2 1 0 □0.NO □1.YES 51 C23 Uterine evacuation 52 C23a Vaginal speculum (Sims) 2 1 0 □0.NO □1.YES 53 Sponge (ring) forceps or uterine C23b 2 1 0 □0.NO □1.YES 54 cpacking forceps 55 C23c Single tooth tenaculum forceps 2 1 0 □0.NO □1.YES 56 C23d Long dressing forceps 2 1 0 □0.NO □1.YES 57 C23e Uterine dilators, sizes 13-27 (French) 2 1 0 □0.NO □1.YES 58 C23f Sharp/blunt uterine curettes, 2 1 0 □0.NO □1.YES 59 60 9

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 57 of 66 BMJ Open Facility ID: BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 1 2 3 Code A. Availability B. Supply 4 At least 1 Available but NOT Are there enough for 5 Essential Items available & NONE Availa the daily caseload of 6 Functional functional ble deliveries? 7 C23g Malleable metal uterine sound 2 1 0 □0.NO □1.YES 8 9 C23 Manual vacuum evacuation(MVA) Basic uterine evacuation instruments 10 C23a 2 1 0 □0.NO □1.YES 11 PLUS: Vacuum syringes (single / double 12 C23b 2 1 0 □0.NO □1.YES 13 valve) 14 C23c Silicone lubricant 2 1 0 □0.NO □1.YES 15 C23d Adapters For peer2 review1 only0 □0.NO □1.YES 16 C23e Flexible cannulae, 4 - 12 mm 2 1 0 □0.NO □1.YES 17 C24 Neonatal resuscitation pack 18 C24a Mucus extractor 2 1 0 □0.NO □1.YES 19 C24b Infant face mask 2 1 0 □0.NO □1.YES 20 C24c Ventilatory bag 2 1 0 □0.NO □1.YES 21 C24d Suction catheter 2 1 0 □0.NO □1.YES 22 Infant laryngoscope with spare bulb& C24e 2 1 0 □0.NO □1.YES 23 batteries 24 C24f Endotracheal tubes 2 1 0 □0.NO □1.YES 25 C24g Suction apparatus: warmer 2 1 0 □0.NO □1.YES 26 Craniotomy equipment for C25 27 destructive operation 28 C25a Craniotomy set/kit 2 1 0 □0.NO □1.YES 29

30 Essential Drugs 31 If yes, enough supply to last for up to (Check 32 Code Drugs Available NUMBER OF WEEKS): 33 C26 Antibiotics 34 35 C26a Ampicillin □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ http://bmjopen.bmj.com/ 36 C26b Benzyl penicillin □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 37 C26c Gentamicin □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 38 C26d Metronidazole □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 39 C26e Penicillin G □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 40 C27 Anticonvulsants 41 C27a Magnesium sulfate □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 42 C27b Phenytoin □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+

43 C28 Antihypertensives on October 1, 2021 by guest. Protected copyright. 44 C28a Hydralazine □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 45 C28b Nifedipine □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 46 Oxytocics, Prostaglandins C29 47 & other 48 C29a Ergometrine □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 49 C29b Misoprostol □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 50 C29c Oxytocin □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 51 Drugs used in C30 52 Emergencies 53 C30a Adrenaline □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 54 C30b Aminophylline □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 55 C30c Atropine sulfate □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 56 C30d Calcium gluconate □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 57 C30e Digoxin □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 58 C30f Ephedrine □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 59 60 10

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 58 of 66 Facility ID: BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 1 2 3 C30g Frusemide □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 4 C30h Naloxone □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 5 C30i Nitroglycerine □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 6 C30j Prednisone □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 7 C30k Promethazine □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 8 C31 Anesthetics 9 C31a Halothane □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 10 C31b Ketamine □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 11 □ □ □ □ □ □ □ 12 C31c Lignocaine 2% or 1% □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 13 C32 Analgesics 14 C32a Morphine □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 15 C32b Paracetamol For peer□0.NO □1. YESreview □ <1 □only 1 □ 2 □ 3 □ 4+ 16 C32c Pethidine □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 17 C33 Sedatives 18 C33a Diazepam □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 19 C33b Phenobarbitone □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+

20 C34 Tocolytics 21 22 C34a Nifedipine □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 23 C34b Salbutamol □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 24 C34c Steroids □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 25 C34d Betamethasone □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 26 C34e Dexamethasone □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 27 C34f Hydrocortisone □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 28 C35 IV Fluids 29 C35a Dextrose □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 30 C35b Normal saline □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 31 C35c Ringer’s lactate □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 32 C36 Antimalarial 33 C36a Artemether □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 34 C36b Artesunate □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ http://bmjopen.bmj.com/ 35 C36c Coartem (ALU) □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 36 C36d Quinine dihydrochloride □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 37 C36e Quinine sulfate □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 38 C37 Other 39 C37a Anti-tetanus serum □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 40 C37b Magnesium trisilicate □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 41 C37c Tetanus toxoid □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 42 C37d Anti-retrovirals – Mother □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 43 on October 1, 2021 by guest. Protected copyright. C37e Anti-retrovirals - Newborn □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 44 C37f HIV rapid testing kit □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 45 Post-HIV exposure □0.NO □1.YES 46 C37g □ <1 □ 1 □ 2 □ 3 □ 4+ prophylactic treatment 47

48

49

50

51

52

53 54 55 56 57 58 59 60 11

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 59 of 66 BMJ Open Facility ID: BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 1 2 3 D. Obstetric /Maternity ward 4 5 ID Available 6 D1 Does this facility have an obstetric / maternity ward? □0.NO □1.YES 7 Kama jibu ni NO, ruka nenda chumba kingine(NENDA E1) 8 9 10 Infrastructure 11 12 ID Miundo mbinu kwa ujumla Available 13 D2 Electricity □0.NO □1.YES 14 D3 Sufficient light source to perform tasks during the day □0.NO □1.YES 15 D4 Means of ventilationFor peer review □only0.NO □1.YES 16 D5 Running water □0.NO □1.YES 17 18 D6 Fan / air conditioning (if applicable) □0.NO □1.YES 19 D7 Curtains/means of providing patient privacy □0.NO □1.YES 20 D8 Waiting area for visitors (iliyo na viti na kivuli) □0.NO □1.YES 21 22 23 Vifaa 24 ID A. Availability B. Supply 25 At least 1 Available Are there enough for NOT 26 Essential Items available & but NONE the daily caseload of Available 27 Functional functional deliveries? 28 D9 Beds 2 1 0 □0.NO □1.YES 29 D10 Linens 2 1 0 □0.NO □1.YES 30 D11 Blankets for cold weather 2 1 0 □0.NO □1.YES 31 D12 BP cuff 2 1 0 □0.NO □1.YES 32 D13 Stethoscope 2 1 0 □0.NO □1.YES 33 D14 Baby weighing scale 2 1 0 □0.NO □1.YES 34

D15 Fetal stethoscope 2 1 0 □0.NO □1.YES http://bmjopen.bmj.com/ 35 □ □ 36 D16 Sponge bowls 2 1 0 □0.NO □1.YES 37 D17 Clinical oral thermometer 2 1 0 □0.NO □1.YES 38 D18 Clinical oral thermometer 2 1 0 □0.NO □1.YES 39 D19 Scissors 2 1 0 □0.NO □1.YES 40 D20 Low reading thermometer 2 1 0 □0.NO □1.YES Surgeon’s handbrush w/ white nylon □0.NO □1.YES 41 D21 2 1 0 42 bristles

43 D22 Needles and Syringes (10-20cc) 2 1 0 □0.NO □1.YES on October 1, 2021 by guest. Protected copyright. 44 D23 Suture needles/suture materials 2 1 0 □0.NO □1.YES 45 D24 IV Stand(s) 2 1 0 □0.NO □1.YES 46 Filled oxygen cylinder with cylinder □0.NO □1.YES 47 D25 carrier and key to open valve/ 02 2 1 0 48 concentrator 49 D26 Adult ventilator bag and mask 2 1 0 □0.NO □1.YES 50 D27 Mouth gag 2 1 0 □0.NO □1.YES 51 Patient transport (wheelchair, trolley, □0.NO □1.YES D28 2 1 0 52 hammock) 53 At least 1 Available Are there enough for NOT 54 Infection and Prevention available & but NONE the daily caseload of Available 55 Functional functional deliveries? 56 D29 Soap 2 1 0 □0.NO □1.YES 57 D30 Antiseptics 2 1 0 □0.NO □1.YES 58 D31 Sterile gloves 2 1 0 □0.NO □1.YES 59 60 12

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 60 of 66 Facility ID: BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 1 2 3 ID A. Availability B. Supply 4 D32 Non-sterile gloves 2 1 0 □0.NO □1.YES 5 □0.NO □1.YES 6 D33 Decontamination container 2 1 0 □ □ 7 D34 Bleach or bleaching powder 2 1 0 □0.NO □1.YES 8 D35 Prepared disinfection solution 2 1 0 □0.NO □1.YES 9 D36 Covered contaminated waste trash bin 2 1 0 □0.NO □1.YES 10 D37 Puncture proof sharps container 2 1 0 □0.NO □1.YES 11 12 ID Instructions Jibu 13 D38 Is food provided by the hospital to patients? □0.NO □1.YES 14 D39 Are there empty beds for the next patients? □0.NO □1.YES 15 D40 If yes, are the Forempty beds cleanpeer and ready reviewto receive new patients? only □0.NO □1.YES 16 17 Essential Drugs 18 ID Drugs Available 19 D41 Antibiotics 20 D41a Amoxicillin □0.NO □1.YES 21 D42 Ampicillin 22 D42a Benzathine penicillin □0.NO □1.YES 23 D42b Benzyl penicillin □0.NO □1.YES 24 D43c Cloxacillin □0.NO □1.YES 25 D43d Erythromicin □0.NO □1.YES 26 D43e Gentamicin □0.NO □1.YES 27 D43f Metronidazole □0.NO □1.YES 28 D43g Nitrofurantoin □0.NO □1.YES 29 D43h Penicillin G □0.NO □1.YES 30 D43i Procaine penicillin G □0.NO □1.YES 31 D43j Trimethoprim/Sulfamethozazole □0.NO □1.YES 32 D44 Anticonvulsants □0.NO □1.YES 33 D44a Magnesium sulfate □0.NO □1.YES 34

D44b Phenytoin □0.NO □1.YES http://bmjopen.bmj.com/ 35 □ □ Antihypertensives 36 D45 □0.NO □1.YES 37 D45a Hydralazine □0.NO □1.YES 38 D45b Nifedipine □0.NO □1.YES 39 D46 Drugs used in Emergencies □0.NO □1.YES 40 D46a Adrenaline □0.NO □1.YES 41 D46b Aminophylline □0.NO □1.YES 42 D46c Atropine sulfate □0.NO □1.YES

43 D46d Calcium gluconate □0.NO □1.YES on October 1, 2021 by guest. Protected copyright. 44 D46e Digoxin □0.NO □1.YES 45 D46f Ephedrine □0.NO □1.YES 46 D46g Frusemide □0.NO □1.YES 47 D46h Naloxone □0.NO □1.YES 48 D46i Nitroglycerine □0.NO □1.YES 49 D46j Prednisone □0.NO □1.YES 50 D46k Prednisolone □0.NO □1.YES 51 D46l Promethazine □0.NO □1.YES 52 D47 Analgesics 53 D47a Morphine □0.NO □1.YES 54 D47b Paracetamol □0.NO □1.YES 55 D47c Pethidine □0.NO □1.YES 56 D48 Sedatives 57 D48a Diazepam □0.NO □1.YES 58 D48b Phenobarbitone □0.NO □1.YES 59 60 13

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 61 of 66 BMJ Open Facility ID: BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 1 2 3 ID Drugs Available 4 D49 IV Fluids 5 D49a Dextrose □0.NO □1.YES 6 D49b Glucose □0.NO □1.YES 7 D49c Normal saline □0.NO □1.YES 8 D49d Ringer’s lactate □0.NO □1.YES 9

10

11

12 E. Laboratory 13 14 ID Available 15 E1 Does this facilityFor have a laboratory?peer review only□0.NO □1.YES 16 17 Kama jibu ni NO, ruka nenda chumba kinachofuata (Nenda F1) 18 19 Infrastructure 20 21 ID Infrastructure kwa ujumla Available 22 E2 Electricity □0.NO □1.YES 23 E3 Chanzo cha mwanga kinachotosheleza kufanya kazi mchana □0.NO □1.YES 24 E4 Njia za kuingizia hewa (ex. ceiling fan?) □0.NO □1.YES 25 26 E5 Maji ya bomba □0.NO □1.YES 27 28 Vifaa 29 ID A. Availability B. Supply 30 Provision of donor blood for At least 1 Available Are there enough for NOT 31 transfusion available & but NONE the daily caseload of AVAILABLE 32 Functional functional deliveries? 33 E6 Jokofu 2 1 0 □0.NO □1.YES 34 E7 Test tubes - various sizes 2 1 0 □0.NO □1.YES 35 E8 Slides (microscope) 2 1 0 □0.NO □1.YES http://bmjopen.bmj.com/ 36 E9 Compound microscope 2 1 0 □0.NO □1.YES 37 E10 Microscope illuminator 2 1 0 □0.NO □1.YES 38 E11 Blood lancets 2 1 0 □0.NO □1.YES 39 E12 Cotton wool 2 1 0 □0.NO □1.YES 40 E13 Rack 2 1 0 □0.NO □1.YES 41 E14 8.5 g/l Sodium Chloride solution 2 1 0 □0.NO □1.YES 42 E15 20% Bovine albumin 2 1 0 □0.NO □1.YES 43 E16 Centrifuge 2 1 0 □0.NO □1.YES on October 1, 2021 by guest. Protected copyright. 44 E17 Blood typing and cross-marching □0.NO □1.YES 2 1 0 45 reagents 46 E18 Blood collection bags 2 1 0 □0.NO □1.YES 47 E19 Artery forceps 2 1 0 □0.NO □1.YES 48 E20 Pilot bottles (containing 1 ml ACD □0.NO □1.YES 49 2 1 0 solution) 50 E21 Hepatitis Test 2 1 0 □0.NO □1.YES 51 E22 HIV Test 2 1 0 □0.NO □1.YES 52 E23 Syphilis Test 2 1 0 □0.NO □1.YES 53 Laboratory supplies 54 55 E24 Microscope 2 1 0 □0.NO □1.YES 56 E25 Immersion oil 2 1 0 □0.NO □1.YES 57 E26 Glass rods 2 1 0 □0.NO □1.YES 58 E27 Sink or staining tank 2 1 0 □0.NO □1.YES 59 60 14

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 62 of 66 Facility ID: BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 1 2 3 ID A. Availability B. Supply 4 Provision of donor blood for At least 1 Available Are there enough for 5 NOT transfusion available & but NONE the daily caseload of 6 AVAILABLE Functional functional deliveries? 7 E28 Measuring cylinder (10-50 ml) 2 1 0 □0.NO □1.YES 8 E29 Wash bottle containing buffered water 2 1 0 □0.NO □1.YES 9 E30 Interval timer clock 2 1 0 □0.NO □1.YES 10 □ □ 11 E31 Rack for drying slides 2 1 0 □0.NO □1.YES 12 E32 Leishman stain 2 1 0 □0.NO □1.YES 13 E33 Methanol 2 1 0 □0.NO □1.YES 14 E34 Refrigerator 2 1 0 □0.NO □1.YES 15 E35 Field stains A andFor B peer2 review1 only0 □0.NO □1.YES 16 E36 Glass containers 2 1 0 □0.NO □1.YES 17 E37 Counting chamber (Neubauer) 2 1 0 □0.NO □1.YES 18 E38 Pipette (various sizes) 2 1 0 □0.NO □1.YES 19 E39 Tork diluting solution 2 1 0 □0.NO □1.YES 20 E40 Tally counter, differential if possible 2 1 0 □0.NO □1.YES 21 E41 Haemoglobinometer 2 1 0 □0.NO □1.YES 22 E42 Hydrochloric acid solution 2 1 0 □0.NO □1.YES E43 Microhaematocrit centrifuge (manual □0.NO □1.YES 23 2 1 0 24 or electric) 25 E44 Scale for reading results 2 1 0 □0.NO □1.YES 26 E45 Heparinized capillary tubes (75 mm x □0.NO □1.YES 2 1 0 27 1.5 mm) 28 E46 Spirit lamp 2 1 0 □0.NO □1.YES 29 E47 Ethanol 2 1 0 □0.NO □1.YES 30 Laboratory Supplies 2 1 0 □0.NO □1.YES 31 E48 Indicator papers and tablets 2 1 0 □0.NO □1.YES 32 E49 Benedict solution 2 1 0 □0.NO □1.YES 33 E50 Test-tube holder 2 1 0 □0.NO □1.YES 34 E51 Beakers (various sizes) 2 1 0 □0.NO □1.YES http://bmjopen.bmj.com/ 35 E52 Spirit lamp 2 1 0 □0.NO □1.YES 36 E53 Sodium nitroprusside 2 1 0 □0.NO □1.YES 37 E54 Glacial acetic acid 2 1 0 □0.NO □1.YES 38 E55 Ammonia 2 1 0 □0.NO □1.YES 39 Sulfosalicyclic acid (300 g/I aqueous □0.NO □1.YES E56 2 1 0 40 solution) 41 E57 Lugol’s iodine solution 2 1 0 □0.NO □1.YES 42 E58 Ehrlich reagent 2 1 0 □0.NO □1.YES 43 E59 Uristix (dip stick for protein in urine) 2 1 0 □0.NO □1.YES on October 1, 2021 by guest. Protected copyright. 44 Infection Prevention 2 1 0 □0.NO □1.YES 45 E60 Soap 2 1 0 □0.NO □1.YES 46 E61 Antiseptics 2 1 0 □0.NO □1.YES 47 E62 Sterile gloves 2 1 0 □0.NO □1.YES 48 E63 Non-sterile gloves 2 1 0 □0.NO □1.YES 49 E64 Decontamination container 2 1 0 □0.NO □1.YES 50 □ □ E65 Bleach or bleaching powder 2 1 0 □0.NO □1.YES 51 □ □ 52 E66 Prepared disinfection solution 2 1 0 □0.NO □1.YES 53 E67 Regular trash bin 2 1 0 □0.NO □1.YES 54 E68 Covered contaminated waste trash bin 2 1 0 □0.NO □1.YES 55 E69 Puncture proof sharps container 2 1 0 □0.NO □1.YES 56 57 58 59 60 15

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 63 of 66 BMJ Open Facility ID: BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 1 2 3 F: Pharmacy 4 5 Code Available 6 F1 Does this facility have a pharmacy? □0.NO □1.YES 7 8 F2 Is drug inventory register up to date? □0.NO □1.YES 9 F3 Are records on supply requests from wards up to date? □0.NO □1.YES 10 F4 Is ‘First-in-First-out’ system for supply management used? □0.NO □1.YES 11 F5 Is there a regularly used mechanism to ensure that expired drugs are not □0.NO □1.YES 12 distributed? 13 F6 Are drugs protected from moisture, heat or infestation (e.g., placed on shelves or □0.NO □1.YES 14 slats, ventilated)? 15 F7 Does it have a Forbuffer stock? peer review only □0.NO □1.YES 16 17 F8 Is there a minimum stock level for ordering new drugs? □0.NO □1.YES 18 F9 Do you receive what you order (Accuracy)? □0.NO □1.YES 19 F10 Time taken from ordering till receiving the supply □0.---Wiki □1.----- 20 Miezi 21 22 Essential Drugs If yes, enough supply to last for up to 23 ID Drugs Available 24 (Check NUMBER OF WEEKS): 25 F11 Antibiotics 26 F11a Amoxicillin □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 27 F11b Ampicillin □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 28 F11c Benzathine penicillin □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 29 F11d Benzyl penicillin □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 30 F11e Ceftriaxone □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 31 F11f Cloxacillin □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 32 F11g Erythromicin □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 33 F11ih Gentamicin □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 34 F11i Kanamycin □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ http://bmjopen.bmj.com/ 35 F11j Metronidazole □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 36 F11k Nitrofurantoin □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 37 F12 Penicillin G □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 38 F12a Procaine penicillin G □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 39 F12b Trimethoprim / Sulfamethozazole □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 40 F13 Anticonvulsants 41 F13a Magnesium sulfate □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 42 F13b Phenytoin □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 43 on October 1, 2021 by guest. Protected copyright. F14 Antihypertensives 44 F14a Hydralazine □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 45 F14b Aldomet/Metheldopa □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 46 F14c Nifedipine □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 47 □ □ □ □ □ □ □ 48 F15 Oxytocics, Prostaglandins & other 49 F15a Ergometrine □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 50 F15b Methylergometrine □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 51 F15c Misoprostol □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 52 F15d Oxytocin □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 53 F15e Prostaglandin E2 □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 54 F16 Drugs used in Emergencies □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 55 F16a Adrenaline □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 56 F16b Aminophylline □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 57 F16c Atropine sulfate □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 58 F16d Calcium gluconate □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 59 60 16

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 64 of 66 Facility ID: BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 1 2 3 If yes, enough supply to last for up to ID Drugs Available 4 (Check NUMBER OF WEEKS): 5 F16e Digoxin □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 6 F16f Ephedrine □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 7 F16g Frusemide □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 8 F16h Naloxone □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 9 F16i Nitroglycerine □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 10 F16j Prednisone □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 11 F16k Prednisolone □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 12 □ □ □ □ □ □ □ 13 F16l Promethazine □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 14 F17 Anaethetics 15 F17a Halothane For peer□0. NOreview □1.YES □only <1 □ 1 □ 2 □ 3 □ 4+ 16 F17b Ketamine □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 17 F17c Lignocaine 2% or 1% □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 18 F18 Analgesics 19 F18a Morphine □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 20 F18b Paracetamol □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 21 F18c Pethidine □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 22 F18d Sedatives □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 23 F18e Diazepam □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 24 F18f Phenobarbitone □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 25 F19 Tocolytics 26 F19a Nifedipine □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 27 F19b Salbutamol □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 28 F20 Steroids 29 F20a Betamethasone □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 30 F20b Dexamethasone □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 31 F20c Hydrocortisone □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 32 F21 IV Fluids 33 F21a Dextrose □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 34 F21b Glucose □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ http://bmjopen.bmj.com/ 35 F21c Normal saline □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 36 F21d Ringer’s lactate □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 37 F22 Antimalarial 38 F22a ALU □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 39 F22b Artesunate □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 40 F22c Quinine dihydrochloride □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 41 F22d Quinine sulfate □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 42 F22e Sulfadoxine/Pyrimethamine □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 43 on October 1, 2021 by guest. Protected copyright. F23 Other □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 44 F23a Anti-tetanus serum □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 45 F23b Ferrous sulfate □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 46 F23c Folic acid □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 47 F23d Magnesium trisilicate □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 48 □ □ □ □ □ □ □ 49 F23e Sodium citrate □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 50 F23f Tetanus antitoxin □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 51 F23g Tetanus toxoid □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 52 F23h Anti-retrovirals – Mothers □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 53 F23i Anti-retrovirals – Newborn □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 54 F23j HIV rapid testing kits □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ Post-HIV exposure prophylactic □0.NO □1.YES 55 F23k □ <1 □ 1 □ 2 □ 3 □ 4+ 56 treatment 57 58 59 60 17

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 65 of 66 BMJ Open Facility ID: BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 1 2 3 G. Autoclave room 4 5 Code Available 6 G1 Does this facility have an autoclave room? □ 0.No □ 1. Yes 7 Kama jibu ni NO, ruka nenda chumba (Nenda MWISHO) 8 9 Infrastructure 10 11 Code Available 12 G2 Electricity □0.NO □1.YES 13 G3 Sufficient light source to perform tasks during the day □0.NO □1.YES 14 G4 Sufficient light source to perform tasks at night □0.NO □1.YES 15 For peer review□0.NO only □1.YES 16 G5 Means of ventilation 17 Running water 18 19 Equipments 20 ID A. Availability B. Usambazwaji 21 At least 1 Available but Are there enough for NOT 22 Infection prevention available & NONE the daily caseload of Available 23 Functional functional deliveries? 24 Autoclave (with temperature and □0.NO □1.YES G6 2 1 0 25 pressure gauges) 26 G7 Soap 2 1 0 □0.NO □1.YES 27 G8 Antiseptics 2 1 0 □0.NO □1.YES 28 G9 Sterile gloves 2 1 0 □0.NO □1.YES 29 G10 Non-sterile gloves 2 1 0 □0.NO □1.YES 30 G11 Non-sterile protective clothing 2 1 0 □0.NO □1.YES 31 G12 Decontamination container 2 1 0 □0.NO □1.YES 32 G13 Bleach or bleaching powder 2 1 0 □0.NO □1.YES 33 G14 Prepared disinfection solution 2 1 0 □0.NO □1.YES 34

G15 Regular trash bin 2 1 0 □0.NO □1.YES http://bmjopen.bmj.com/ 35 Covered contaminated waste 2 1 0 □0.NO □1.YES 36 G16 trash bin 37 G17 Puncture proof sharps container 2 1 0 □0.NO □1.YES 38 Mayo stand (or equivalent to 2 1 0 □0.NO □1.YES 39 G18 40 establish sterile field) 41 REFERRAL 42

on October 1, 2021 by guest. Protected copyright. 43 44 CODE ITEM Availability 45 R Not available Available and functional Available but not 46 functional 47 R1 Land Telephones 0 1 2 48 R2 Mobile phones 0 1 2 49 R3 Radio communication set with repeater 0 1 2 50 station 51 R4 Motor vehicle ambulance 0 1 2 52 R5 Ox carte 0 1 2 53 R6 Motor vehicle 0 1 2 54 R7 Motorcycle 0 1 2 55 R8 Bicycle 0 1 2 56 R9 Boat 0 1 2 57 R10 Who provides fuel for the 58 ambulance(s)? 59 60 18

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 66 of 66 Facility ID: BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 1 2 3 4 5 Finish Time: Hour__/___ Minutes___/____ 6 7 8 9 End 10 11 12 13 14 15 For peer review only 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 http://bmjopen.bmj.com/ 36 37 38 39 40 41 42

43 on October 1, 2021 by guest. Protected copyright. 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 19

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from

Can training non-physician clinicians/associate clinicians (NPCs/ACs) in emergency obstetric, neonatal care and clinical leadership make a difference to practice and help towards reductions in maternal and neonatal mortality in rural Tanzania? The ETATMBA Project. For peer review only

Journal: BMJ Open

Manuscript ID bmjopen-2015-008999.R2

Article Type: Research

Date Submitted by the Author: 15-Oct-2015

Complete List of Authors: Ellard, David; Warwick Medical School, Clinical trials Unit Shemdoe, Aloisia; Ifakara Health Institute, Mazuguni, Festo; Ifakara Health Institute, Mbaruku, Godfrey; Ifakara Health Institute, Davies, David; The University of Warwick, Educational Development & Research Team, Warwick Medical School Kihaile, Paul; Ifakara Health Institute, Pemba, Senga; Tanzanian Training centre for International Health, Bergström, Staffan; Karolinska Institute, Division of Global Health (IHCAR), Department of Public Health Sciences Nyamtema, Angelo; Tanzanian Training centre for International Health,

Mohamed, Hamed-Mahfoudh; Ifakara Health Institute, http://bmjopen.bmj.com/ O'Hare, Paul; University of Warwick, Warwick Medical School Group, The ETATMBA Study; University of Warwick, Warwick Medical School

Primary Subject Global health Heading:

Secondary Subject Heading: Obstetrics and gynaecology, Medical education and training

human resources, maternal mortality, Tanzania, Non-physician clinicians, on October 1, 2021 by guest. Protected copyright. Keywords: Associate clinicians, medical education and training

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 1 of 72 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 1 2 1 Can training nonphysician clinicians/associate clinicians (NPCs/ACs) in 3 2 emergency obstetric, neonatal care and clinical leadership make a difference to 4 5 3 practice and help towards reductions in maternal and neonatal mortality in rural 6 4 Tanzania? The ETATMBA Project. 7 8 5 David R Ellard* PhD 9 6 Warwick Clinical Trials Unit, Division of Health Sciences, Warwick Medical School, The University 10 7 of Warwick, Coventry, CV4 7AL, UK 11 8 [email protected] 12 9 13 10 Aloisia Shemdoe, MSc 14 11 Ifakara Health Institute, Dar es Salaam, Tanzania 15 12 [email protected] peer review only 16 13 17 14 Festo Mazuguni, BSc 18 15 Ifakara Health Institute, Dar es Salaam, Tanzania 19 16 [email protected] 20 17 21 18 Godfrey Mbaruku, MD PhD 22 19 Ifakara Health Institute, Dar es Salaam, Tanzania 23 20 [email protected] 24 21 25 22 David Davies, PhD 26 23 Educational Development & Research Team, Warwick Medical School, The University of Warwick, 27 24 Coventry, CV4 7AL, UK 28 25 [email protected] 29 26 30 27 Paul Kihaile, MD, PhD 31 28 Ifakara Health Institute, Dar es Salaam, Tanzania 32 29 [email protected] 33 30 34

31 Senga Pemba, PhD http://bmjopen.bmj.com/ 35 32 Tanzanian Training centre for International Health 36 33 Ifakara, Tanzania 37 34 [email protected] 38 35 39 40 36 Staffan Bergström, MD, PhD 41 37 Department of Public Health Sciences, Karolinska Institutet, Sweden 42 38 [email protected]

43 39 on October 1, 2021 by guest. Protected copyright. 44 40 Angelo Nyamtema, MD, PhD 45 41 Tanzania Training centre for International Health 46 42 Ifakara, Tanzania 47 43 [email protected] 48 44 49 45 HamedMahfoudh Mohamed, MD 50 46 Ifakara Health Institute, Dar es Salaam, Tanzania 51 47 [email protected] 52 48 53 49 Joseph Paul O'Hare, MD 54 50 Division of Metabolic & Vascular Health, Warwick Medical School, The University of Warwick, 55 51 Coventry, CV4 7AL, UK 56 52 J.P.O[email protected] 57 53 58 54 On behalf of The ETATMBA Study Group 59 55 *corresponding author 60 1 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 2 of 72 BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 1 2 1 Abstract 3 4 2 Objectives 5 6 3 During late 2010, 36 trainees including 19 assistant medical officers (AMOs) one senior clinical 7 8 4 officer (CO) and 16 nurse midwives/nurses were recruited from districts across rural Tanzania and 9 10 5 invited to join the use of appropriate technologies for maternal and perinatal survival in subSaharan 11 12 13 6 Africa (ETATMBA) training programme. The ETATMBA project is training Associate Clinicians 14 15 7 (ACs) as advancedFor clinical peerleaders in emergency review obstetric care. The traineesonly returned to health 16 17 8 facilities across the country with the hope of being able to apply their new skills and knowledge. The 18 19 9 aim of this study in was to explore the impact of the ETATMBA training on health outcomes 20 21 10 including maternal and neonatal morbidity and mortality in their facilities. Secondly, to explore the 22 23 11 challenges faced in working in these health facilities. 24 25 12 Design 26 27 13 The study is a pre/post examination of maternal and neonatal health indicators and a survey of health 28 29 14 facilities in rural Tanzania. The facilities surveyed were those in which ETATMBA trainees were 30 31 15 placed posttraining. The maternal and neonatal indicators were collected for 2011 and 2013 the 32 33 16 survey of the facilities was early 2014. 34 35 http://bmjopen.bmj.com/ 36 17 Results 37 38 18 Sixteen of seventeen facilities were surveyed. Maternal deaths show a nonsignificant downward 39 40 19 trend over the two years (282 to 232 cases/per 100,000 live births). There were no significant 41 42 20 differences in maternal, neonatal and birth complication variables across the timepoints. The survey

43 on October 1, 2021 by guest. Protected copyright. 44 21 of facilities revealed shortages in key areas and some are a serious concern. 45 46 22 Conclusion 47 48 23 This study represents a snapshot of rural health facilities providing maternal and neonatal care in 49 50 24 Tanzania. Enhancing knowledge, practical skills, and clinical leadership of ACs may have a positive 51 52 25 impact on health outcomes. However, any impact may be confounded by the significant challenges in 53 54 26 delivering a service in terms of resources. Thus training may be beneficial but it requires an 55 56 27 infrastructure that supports it. 57 58 59 28 60 2 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 3 of 72 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 1 2 1 Key words: Nonphysician clinicians, Associate clinicians, maternal mortality, training, medical 3 4 2 education, human resources, infrastructure, Tanzania. 5 6 3 7 8 9 4 Strengths and limitations of the study 10 11 12 5 • This is one of the first studies taking an indepth look at the impact on health outcomes in 13 14 6 districts across rural Tanzania, of a programme of knowledge, skills and clinical leadership 15 For peer review only 16 7 training for associate clinicians; 17 18 8 • This cadre is an important component in helping relieve the chronic shortages of trained 19 20 9 medical professionals in subSaharan Africa and helping countries move towards realisation 21 22 10 of millennium development goals; 23 24 11 • One of the primary outcomes (neonatal mortality) was found to be not recorded or poorly 25 26 12 recorded at health facilities at the time preventing us from reporting on this important 27 28 13 outcome; 29 30 14 • This was a before and after design and there was no control group on which to draw 31 32 15 comparisons; 33 34

16 A number of facilities where trainees were returned to posttraining were not upgraded, as http://bmjopen.bmj.com/ 35 • 36 37 17 planned, thus preventing them from putting into practice their new skills and knowledge. 38 39 40 41 42

43 on October 1, 2021 by guest. Protected copyright. 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 3 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 4 of 72 BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 1 2 1 Background 3 4 2 In 2013 it was estimated that there was a global shortage of 7.2 million healthcare workers, and that 5 6 3 by 2035 this is expected to rise to 12.9 million. [1] A recent review of global surgery, obstetric, and 7 8 4 anaesthesia workforce literature highlights the crisis. Countries like Tanzania only have a physician 9 10 5 density of 1 per 100,000 people. [2] It is estimated that currently there is a shortage of one million 11 12 6 healthcare workers in subSaharan Africa. [3] This shortage is partly because not enough people are 13 14 7 appropriately trained but is aggravated by meagre salaries, poor working conditions, low morale, 15 For peer review only 16 8 inadequate remuneration, and few opportunities for continuous professional development. [4] Even 17 18 9 with a proliferation of new medical and nursing schools in recent years, the rise is not proportional to 19 20 10 the existing large populations. [5] For those working in rural areas there is professional isolation, 21 22 11 inadequate communication with peers and consultants in the cities, and a lack of appropriate 23 24 12 equipment and technologies. [3] 25 26 13 27 28 14 In Tanzania, the lack of basic items in many health facilities has hindered timely and appropriate 29 30 15 quality obstetric and neonatal care, particularly in rural and remote health facilities. A number of 31 32 16 studies conducted in the country have also indicated that poor quality of care has been experienced at 33 34

17 health facilities due to the lack of an enabling environment (drugs, equipment, and supplies) [6], poor http://bmjopen.bmj.com/ 35 36 18 skills of providers or hostile attitudes of providers, and a lack of trained staff. [710] As part of the 37 38 19 solution many African countries have created a cadre of midlevel health workers called Non 39 40 20 Physician Clinicians (NPCs), now more usually referred to as Associate Clinicians (ACs). In Tanzania 41 42 21 this cadre is often referred to as Clinical Officers (COs) or Assistant Medical Officers (AMOs) (COs 43 on October 1, 2021 by guest. Protected copyright. 44 22 who have received some additional training). These workers are trained by both government and non 45 46 47 23 government institutions and are often the most experienced health workers in hospitals and health 48 49 24 centres across the country, particularly away from urban centres.[11] Moreover, all of these 50 51 25 AMOs/COs are trained in Emergency Obstetric Care (EmOC) and are in the frontline providing care 52 53 26 for mothers and babies.[12] In rural areas where medical doctors (MDs) are few in number, the use of 54 55 27 AMOs/COs and nurse midwives (NMWs) has been identified as a viable solution, as these groups can 56 57 28 be trained through short course programmes to provide effective Comprehensive Emergency Obstetric 58 59 29 Care (CEmOC) services in remote health centres. Key benefits of using AMOs/COs in CEmOC 60 4 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 5 of 72 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 1 2 1 services include; reducing training and employment costs, promoting task sharing/shifting and 3 4 2 enhancing retention within local health systems. Studies have shown that unlike MDs, AMOs/COs 5 6 3 remain in rural areas and continue working there. [13] 7 8 4 9 10 5 Major surveys consistently show that extra training and support can enhance task sharing/shifting and 11 12 6 reduce maternal and neonatal mortality and morbidity in the areas where these schemes have been 13 14 7 piloted. [12 14 15] Training skilled attendants to prevent, detect, and manage major obstetric 15 For peer review only 16 8 complications including undertaking emergency caesarean surgery in complicated deliveries is 17 18 9 arguably the single most important factor in preventing maternal deaths and protecting the human 19 20 10 rights of women. [12 1416] To be effective AMOs/COs need appropriate knowledge, skills, 21 22 11 equipment, drugs and technology essential for managing obstetric complications in rural or deprived 23 24 12 communities. 25 26 13 27 28 14 The aim of the Enhancing human resources and use of appropriate technologies for maternal and 29 30 15 perinatal survival in subSaharan Africa (ETATMBA) project was to develop, implement, and 31 32 16 evaluate a programme of locally based clinical service improvement including clinical guidelines and 33 34

17 pathways, workforce development through structured education, and leadership training.[17 18] This http://bmjopen.bmj.com/ 35 36 18 was linked to specialist onsite support and mentoring. 37 38 19 39 40 20 The ETATMBA project in Tanzania 41 42 21 The ETATMBA Project combined two main interventions: First, the training of ACs and nurses in 43 on October 1, 2021 by guest. Protected copyright. 44 22 CEmOC and anaesthesia. Secondly, post training mentoring and supervision of participants at their 45 46 47 23 working places. Within this project, the clinical service improvement involved implementing best 48 49 24 existing practice, linked to training in clinical leadership, and providing the context for understanding 50 51 25 the additional health gain from the use of appropriate available technologies designed to reduce 52 53 26 morbidityspecific maternal casefatality rates and fresh stillbirth rates (intrapartum foetal mortality) 54 55 27 across different African communities (Malawi and Tanzania). [19] See also web supplement for 56 57 28 addition information. 58 59 29 60 5 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 6 of 72 BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 1 2 1 3 4 2 The aim of this study is to explore the impact of the ETATMBA training on health outcomes 5 6 3 including maternal and neonatal morbidity and mortality in the facilities where trainees were based. 7 8 4 Secondly, surveying these health facilities and looking at their ability to support the trainees in terms 9 10 5 of infrastructure, supplies and drugs. Looking for the facilitators and barriers to good clinical practice 11 12 6 and the daytoday challenges faced by the health workers in these facilities. In addition, a qualitative 13 14 7 study was undertaken time with the trainees and other stakeholders but this will be reported 15 For peer review only 16 8 elsewhere. 17 18 9 19 20 10 Methods 21 22 11 Design 23 24 12 The study is a pre and post examination of maternal and neonatal health indicators and a survey of a 25 26 13 sample of health facilities in rural Tanzania. The survey includes: infrastructure, availability of 27 28 14 equipment, supplies and drugs. The facilities surveyed were those in which ETATMBA trainees were 29 30 15 placed posttraining. The health indicators were collected for the whole of 2011 (pre) and the whole of 31 32 16 2013 (post): the survey of the facilities was early 2014. 33 34

17 Participants http://bmjopen.bmj.com/ 35 36 18 During late 2010 and 2011 36 trainees (assistant medical officers (AMOs) and nurse midwives/nurses 37 38 19 (anaesthesia)) were recruited from districts across Tanzania and invited to undertake the ETATMBA 39 40 20 training programme (see web appendix for more information). 41 42 21 Outcome measure 43 on October 1, 2021 by guest. Protected copyright. 44 22 Maternal and neonatal health outcomes were collected from each health facility where a trainee was 45 46 47 23 based (posttraining) for the whole of 2011 (pre) and 2013 (post). This included early neonatal 48 49 24 mortality (only including deaths that occur before discharge) and maternal mortality (case specific) 50 51 25 and other obstetric indicators including: numbers of birth events, stillbirths, postpartum haemorrhage, 52 53 26 caesarean sections, obstructed labor and sepsis. It is important to note here that neonatal mortality 54 55 27 rates were not reported in the baseline data, we believed because they had been overlooked. We 56 57 28 planned to rectify this by retrospectively collecting the data. However, after we visited the sites it 58 59 60 6 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 7 of 72 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 1 2 1 became clear that neonatal mortality rates had not been recorded, or at least were not available. These 3 4 2 data were therefore unavailable for either baseline or follow up. 5 6 3 7 8 4 The outcomes selected all relate to ETATMBA knowledge and skills training, and rely upon data 9 10 5 believed to be available in monthly / annual summary reports stored at each facility covering the 11 12 6 project time period. A 10% sample of variables were crosschecked with the actual registers for 13 14 7 accuracy at each facility. 15 For peer review only 16 8 17 18 9 A predesigned instrument was used to capture the survey data (see online supplementary appendix 1). 19 20 10 This captured the availability of resources including; equipment, supplies and infrastructure, and 21 22 11 recorded whether there was a sufficient supply/number of the listed items for the facility’s daily 23 24 12 caseload of deliveries, and whether the items had been available and functional, available but NOT 25 26 13 functional, or not available. (e.g. infrastructure, equipment, supplies and drugs). Essential drugs: the 27 28 14 availability and supply of drugs for each room (emergency room, labour / delivery room, maternity 29 30 15 ward, operating theatre and pharmacy) were recorded. Checks were done to confirm whether the 31 32 16 listed drug was available and if the supply was sufficient to last for less than one week, up to one 33 34

17 week, up to two weeks, up to three weeks, or up to four or more weeks. http://bmjopen.bmj.com/ 35 36 18 Research team 37 38 19 The primary data collection team consisted of two local research assistants based at the Ifakara Health 39 40 20 Institute (IHI), Dar es Salaam, Tanzania. Both of the research assistants are experienced researchers. 41 42 21 The principal investigator at the IHI gave local support, with management/oversight provided by DE 43 on October 1, 2021 by guest. Protected copyright. 44 22 at Warwick. 45 46 47 23 Procedure 48 49 24 The research assistants identified the facilities in which trainees were working and extracted the 2011 50 51 25 study variables from data collected by colleagues at IHI for ETATMBA in 2012 (baseline data). The 52 53 26 followup data were the same variables for the year 2013. The followup data and the facility survey 54 55 27 data were gathered during site visits to the facilities in early 2014. 56 57 28 58 59 29 Data analysis 60 7 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 8 of 72 BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 1 2 1 Descriptive and summary statistics were produced for the two years, change scores were produced, 3 4 2 and ttests were used to look for differences. Significance was set at 5%. Data are presented in tables 5 6 3 and graphs as appropriate. Survey data are presented as descriptive statistics. Data is grouped by 7 8 4 facility type (i.e. district hospitals and health centres) as they are different. In simple terms we 9 10 5 expected the hospitals to be larger than health centres, have more staff and better availability of 11 12 6 essential infrastructure, supplies, equipment and drugs. 13 14 7 15 For peer review only 16 8 Ethical approval 17 18 9 The study was reviewed and approved by the Biomedical Research Ethics Committee (BREC) at the 19 20 10 University of Warwick, UK (REGO2013572) and The National Institute for Medical Research, 21 22 11 Institutional review board, Dar es Salaam, Tanzania (no.35). 23 24 12 25 26 13 Results 27 28 14 Post training ETATMBA trainees returned to 17 rural health facilities in Tanzania. Sixteen of these 29 30 15 health facilities were included in this study. Table 1 gives an overview of the facilities and the 31 32 16 ETATMBA trainees who were based their after the training. Thirtysix received the ETATMBA 33 34

17 training including19 assistant medical officers (AMOs), one CO, and 14 nurse midwives (NMW) and http://bmjopen.bmj.com/ 35 36 18 two nurses (anaesthesia). During the project period one AMO and one NMW left the programme to 37 38 19 pursue other interests and one NMW died. Thus attrition at the end of the training programme was 39 40 20 around 8%. Fourteen trainees did not return to the facility from which they were recruited because the 41 42 21 facilities had not received an expected facility upgrade. 43 on October 1, 2021 by guest. Protected copyright. 44 22 45 46 47 23 48 49 24 50 51 25 Table 2. Provides a United Nations definition of basic and comprehensive emergency obstetric and 52 53 26 newborn care (BEmOC & CEmOC). 54 55 27 56 57 28 58 59 29 60 8 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 9 of 72 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 1 2 1 Table 3 summarises the key obstetric indicator figures from the 16 health facilities for 2011 and 2013. 3 4 2 5 6 3 No significant differences were found for any of the key obstetric variables across the lifetime of the 7 8 4 project. The number of deliveries/births decreases slightly overall (604) but the number of 9 10 5 deliveries/birth in health centres rises (from 7326 to 7961). There is only a slight increase in overall 11 12 6 fresh stillbirths (+16, an increase of 1 case per 1000 births) whilst there is an increase in macerated 13 14 7 stillbirths in health centres (from 8.3 to 13.9 cases per 1000 live births). Maternal death ratios show a 15 For peer review only 16 8 downward, improving, trend over the two years (down from 282 to 232 cases per 100,000 live births), 17 18 9 but this is not statistically significant. There is a reduction in the caesarean section rate overall down 19 20 10 from 80.2 to 77.2 (cases per 1000 live births) with a large reduction in health centres where rates are 21 22 11 down from 10.6 to 6.2 (cases per 1000 live births), in the hospitals there is an increase in rate from 23 24 12 108.2 to 111.1 (cases per 1000 live births). The birth complication variables collected all show a slight 25 26 13 increase overall but each shows a differing trend in where they were reported. The rates of post 27 28 14 partum haemorrhage change little over time. Obstructed labour rates increased in district hospitals 29 30 15 (6.4 to 9.5 cases per 1000 live births), while in health centres there was a decrease (6.7 to 2.9 cases 31 32 16 per 1000 live births). Sepsis follows a similar trend with an increase in hospitals (1.7 to 3.1 cases per 33 34

17 1000 live births) and a decrease in health centres (1.6 to 0.5 cases per 1000 live births). http://bmjopen.bmj.com/ 35 36 18 37 38 19 39 40 20 41 42 21 Facility survey results 43 on October 1, 2021 by guest. Protected copyright. 44 22 These results originate from the survey undertaken in early 2014 by IHI researchers. As noted in table 45 46 47 23 1 above there were 17 facilities across the country that housed ETATMBA trainees during this 48 49 24 survey. One of these facilities (due to its distance and remoteness) was not visited. All results are 50 51 25 based on 16 facilities, nine healthcentres and seven district hospitals. 52 53 26 Facilities: overall capacity and infrastructure 54 55 27 Availability of running water and functioning toilets are a very significant problem with only one of 56 57 28 nine health centres (11%) and four of seven (57%) district hospitals found to have the availability of 58 59 29 running water and only just over half of facilities a functioning toilet (56%). Most facilities had 60 9 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 10 of 72 BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 1 2 1 sufficient access to lighting to perform tasks at night (75%) but clearly some still struggle. Delivery 3 4 2 beds were found to be available in 56% of the health centres and 86% of the district hospitals. 5 6 3 Ambulance availability was poor at health centres with only one (11%) having availability, whereas 7 8 4 five of the seven (71%) district hospitals had an ambulance available. Referrals from within the 9 10 5 maternity area are problematic as only four health centres had a working (land line) phone in this area 11 12 6 and none of the district hospitals had. The availability of health related registers/records is variable 13 14 7 varying from 100% for items like the delivery register and monthly/annual reports to 6% or less for 15 For peer review only 16 8 the gynaecology register, patient records and discharge registers. (Table 4) 17 18 9 19 20 10 21 22 11 23 24 12 Drugs and equipment for normal delivery and infection prevention 25 26 13 Generally, supplies and equipment availability were at an acceptable level but there are a number of 27 28 14 exceptions. Only about 50% of facilities had needles and syringes available and similarly availability 29 30 15 of suction and vacuum extraction equipment was low. The availability of drugs for normal delivery 31 32 16 purposes was very variable with some drugs readily available (e.g. Lignocaine) whilst others had very 33 34

17 low stocks (e.g. injectable antibiotic and Diazepam) (Table 5). http://bmjopen.bmj.com/ 35 36 18 Infection prevention services in labour delivery/operating theatres 37 38 19 Overall only 75% or less of the facilities surveyed had the basics for infection prevention. None 39 40 20 seemed to have regular availability of soap for hand washing although antiseptics and bleach were 41 42 21 available and may be used as alternatives (Table 5). 43 on October 1, 2021 by guest. Protected copyright. 44 22 45 46 47 23 48 49 24 50 51 25 Comprehensive services for provision of anaesthesia 52 53 26 Most of the district hospitals surveyed had good availability of equipment and supplies for anaesthesia 54 55 27 although Halothane is only available in 3/7 and less than 40% overall. Health centres seemed to lack 56 57 28 access to oxygen with only 2/9 having supplies when surveyed (Table 6). 58 59 29 Items for management of birth complications and caesarean section 60 10 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 11 of 72 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 1 2 1 Overall, unsurprisingly, district hospitals had better availability of equipment, drugs and supplies for 3 4 2 managing birth complications and for performing caesarean sections (Table 6). 5 6 3 7 8 4 9 10 5 11 12 6 Discussion 13 14 7 The main objectives of this study were to explore the impact of the ETATMBA training on health 15 For peer review only 16 8 outcomes including maternal and neonatal morbidity and mortality in the facilities where trainees 17 18 9 were based. Secondly, surveying these health facilities and looking at their ability to support the 19 20 10 trainees in terms of infrastructure, supplies and drugs. Looking for the facilitators and barriers to good 21 22 11 clinical practice and the daytoday challenges faced by the health workers in these facilities. We were 23 24 12 successful in collecting data for the pre and post comparisons and also the survey data. 25 26 13 27 28 14 Interestingly, our study shows the number of actual births decreased overall, in the 16 facilities 29 30 15 measured, between 2011 and 2013. The reduction was seen mostly at the district hospitals with 31 32 16 numbers increasing at health centres. There was a slight increase in fresh stillbirths but again most of 33 34

17 this is at the district hospitals rather than at the health centres. This may suggest that health centres are http://bmjopen.bmj.com/ 35 36 18 referring more women with this problem but the number of macerated stillbirths increased in both 37 38 19 district hospitals and health centres, with the latter being the biggest rise. This trend should be 39 40 20 interpreted with caution, since distinction of type of stillbirth is known to be variable in quality and 41 42 21 indeed it may just suggest that women are presenting late at the health facilities. 43 on October 1, 2021 by guest. Protected copyright. 44 22 45 46 47 23 Maternal deaths decreased which is encouraging as this was a goal of the ETATMBA training. 48 49 24 However, it is not a statistically significant reduction rather a downward trend. This could be simply 50 51 25 be a reflection of the reduction in maternal mortality reported in recent years across Tanzania. 52 53 26 54 55 27 Neonatal mortality was one of our key health indicators in this study. However, it was found that 56 57 28 neonatal mortality was not recorded on Ministry of Health monthly summary sheets in facilities and 58 59 29 thus was not available for us. Reducing neonatal mortality is one of the WHO millennium 60 11 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 12 of 72 BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 1 2 1 development goals. [20] Whilst the number of stillbirths was routinely recorded, early neonatal deaths 3 4 2 were not. This was a very disappointing outcome, a key component of the ETATMBA training was 5 6 3 aimed at interventions to prevent neonatal deaths (i.e. deaths at or around the time of birth and before 7 8 4 discharge from hospital). [19] Indeed in Malawi, we have very positive indications that the 9 10 5 ETATMBA training has helped to save neonate lives. [21] Our study has acted as a ‘wakeup call’ to 11 12 6 the Ministry of Health and Social Welfare (MoHSW) in Tanzania, who have now updated the current 13 14 7 HIMS (Health Management Information System) to ensure neonatal data are collected. 15 For peer review only 16 8 17 18 9 Looking at the birth complication data (postpartum haemorrhage, obstructed labour and sepsis) all 19 20 10 are seen to rise from 2011 to 2013 in both district hospitals and health centres with one exception. 21 22 11 Sepsis rates in health centres decreased, though great caution must be observed, since registration of 23 24 12 morbidities is often incomplete and the facility survey showed centres lacking basic hygiene resources 25 26 13 such as soap. It is of some concern that the number of complications is increasing but this could be a 27 28 14 reflection of more women getting to a health facility where there are health staff who can deal with 29 30 15 the problems. Despite the increase in the numbers of mothers with obstructed labour and postpartum 31 32 16 haemorrhage it is encouraging that maternal mortality ratios at these facilities appear to be falling. 33 34

17 The observed incidence increase in these two registered morbidities by all probability implies an http://bmjopen.bmj.com/ 35 36 18 enhanced recognition and registration of them, rather than a higher incidence in the facility population 37 38 19 under study. We do need to be cautious in our interpretation of these data with only before and after 39 40 20 data as there is no control to detect temporal trends that may be occurring across Tanzania. 41 42 21 43 on October 1, 2021 by guest. Protected copyright. 44 22 Whilst our focus in this study was on the facilities where ETATMBA trainees returned to after their 45 46 47 23 training it is important to draw attention to events that were outside of the control of the ETATMBA 48 49 24 team, events that may have influenced the outcomes. Prior to recruitment the ETATMBA trainees 50 51 25 were based in health centres and district hospitals across rural Tanzania. The original Ministry of 52 53 26 Health (MOHSW) plan was to recruit trainees from health facilities that were due to be upgraded with 54 55 27 a theatre and maternity ward including equipment and resources so that trainees could implement their 56 57 28 new skills. However, the reality was that of the 33 trainees who completed the programme only 19 58 59 29 returned to the place from where they were selected and seven of these returned to facilities that had 60 12 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 13 of 72 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 1 2 1 not been upgraded or where upgrading was still in process. Fourteen trainees did not return to the 3 4 2 facility from which they were recruited because the facilities had not been upgraded. Of these 10/14 5 6 3 were returned to district hospitals in the area they had originally come from. Often these decisions 7 8 4 were made by local District Medical Officers pragmatically responding to need and not to the 9 10 5 strategic planning of the MoHSW. Upgrading of facilities was not part of the ETATMBA project but 11 12 6 rather was ongoing work with the Tanzanian Government and other funding agencies. It is clear that 13 14 7 in a number of cases trainees would have struggled to put their new found skills into practice as 15 For peer review only 16 8 facilities they returned to were not conducive to good clinical practice. For those who returned to a 17 18 9 district hospital it could have been a double edged sword. On one hand a district hospital could give 19 20 10 many more opportunities to put their new found skills into practice but on the other the current senior 21 22 11 staff may have been reluctant to allow them to practice. 23 24 12 25 26 13 The survey reveals some alarming trends in the availability of resources in these facilities. A facility 27 28 14 designated as a CEmOC where there was no functioning operating theatre and a District hospital 29 30 15 designated as a BEmOC rather than a CEmOC when we surveyed. The latter clearly did not meet all 31 32 16 the requirements for a CEmOC at the time of the survey. There are considerable shortages in basic 33 34

17 infrastructure like running water, electricity and toilet facilities. Lack of telephones is interesting as http://bmjopen.bmj.com/ 35 36 18 the survey specifically asked for a landline phone to be available. However, more and more now in 37 38 19 Africa mobile phones are used and are more reliable in terms of service provision. Future surveys 39 40 20 should take this into account. 41 42 21 43 on October 1, 2021 by guest. Protected copyright. 44 22 Record keeping in the facilities is also very variable. Monthly/annual summary reports (containing the 45 46 47 23 data we required) were available in all facilities and in most we were able to crosscheck the data with 48 49 24 register records but some registers were missing and we have already noted the issues surrounding 50 51 25 neonatal mortality rates. 52 53 26 54 55 27 The survey reveals shortages in equipment, supplies and drugs that could impact on patient care. The 56 57 28 district hospitals are better supplied than the health centres. This may be due to remoteness of the 58 59 29 health centres but there are disturbing shortages of the basics for infection/hygiene control and the 60 13 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 14 of 72 BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 1 2 1 provision of oxygen. Infection prevention services were extremely poor. Basic items like soap for 3 4 2 hand washing were mostly absent. However, sepsis rates although rising slightly overall were not 5 6 3 significantly different to baseline (2011) levels, suggesting that despite enormous challenges and lack 7 8 4 of even basic supplies and equipment these clinicians manage to contain sepsis in their facilities. Our 9 10 5 survey findings are all the more alarming as they seem to mirror a more comprehensive survey done 11 12 6 back in 2005/2006 suggesting that things have not changed a great deal. [6] Despite all of this it does 13 14 7 seem that in the face of all of these challenges things are not getting any worse but they could be 15 For peer review only 16 8 better. 17 18 9 19 20 10 This study has a number of limitations not least that one of the primary outcomes was not available to 21 22 11 us. The sample is small and may not be representative of all facilities across Tanzania and with 23 24 12 generally only two trainees in each facility with large throughputs of cases/births. We are not 25 26 13 comparing our facilities to control districts so it is difficult to attribute changes just to ETATMBA 27 28 14 training. Another limiting factor is that ETATMBA had no control over where trainees returned to 29 30 15 post training and a significant number returned to facilities where they could not practice their new 31 32 16 found skills. This however did mean that our sample was more random (not chosen by us). Finally, 33 34

17 this project needs to be seen in the context of the vast distances between facilities and how the terrain http://bmjopen.bmj.com/ 35 36 18 and weather impacts on the health service provision in rural Tanzania. Indeed, in 2009 EvjenOlsen et 37 38 19 al., suggest the need for an integrated and comprehensive hospital/community based approach to 39 40 20 obstetric healthcare in rural Tanzania but our experience here has not shown this being put into 41 42 21 practice. [22] 43 on October 1, 2021 by guest. Protected copyright. 44 22 45 46 47 23 Earlier findings from this project suggested that the training had impact, at the local level, on maternal 48 49 24 mortality. [23] Sadly in this larger current study we cannot be certain of this conclusion. It is 50 51 25 acknowledged that maternal mortality is still a significant problem, particularly in rural Tanzania. [24] 52 53 26 Nelissen et al., suggest that there is a great need for the upscaling and use of evidencebased 54 55 27 interventions that could help to save lives. [24] We can only hope that the, evidenced based, 56 57 28 ETATMBA training and its trainees will be a stimulus to improve care. But for a full impact the 58 59 29 implementation of the training needs to be linked to the provision of well supplied health care 60 14 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 15 of 72 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 1 2 1 facilities in the remote areas. We note that in one province the ETATMBA training has influenced the 3 4 2 upgrade of more health centres at the district level in tandem with the MoHSW objective of upgrading 5 6 3 at least 50% of all health centres in a particular province to provide CEmOC. [25] 7 8 4 9 10 5 A number of papers still highlight that women are reluctant to attend rural health facilities as they 11 12 6 believed the standard of care they will receive will be poor and many still give birth at home without 13 14 7 skilled birth attendance. [9 26] We can only hope that the upskilling of health providers in these rural 15 For peer review only 16 8 areas cascade within the communities to encourage women to seek skilled help during birth. 17 18 9 19 20 10 Whilst not a direct result of our work during the lifetime of this project there has been a shift in 21 22 11 acknowledging the importance of this cadre of health workers. The negative label nonphysician 23 24 12 clinician has been replaced with the more dignifying and respectful associate clinician. Associate 25 26 13 clinicians are now coming together across Africa starting their own professional association. Indeed, 27 28 14 there is now a very active network called ANAC (African Network of Associate Clinicians) enabling 29 30 15 the formation of a community of practice. 31 32 16 33 34

17 Comparing our results with those from Malawi in this project we see an indication that ETATMBA http://bmjopen.bmj.com/ 35 36 18 training can make a difference. [21] There are similarities and differences between this study and that 37 38 19 carried out in Malawi but in both countries it seems that overall the outcomes have been very positive. 39 40 20 41 42 21 We know that The ETATMBA training was successfully implemented (we were able to train the ACs 43 on October 1, 2021 by guest. Protected copyright. 44 22 and we know we have improved their leadership, knowledge and clinical skills) but we are still 45 46 47 23 unclear about the impact in Tanzania. We interpret our results here with caution, presenting just 48 49 24 exactly what we found. There are trends in the data, which suggest an improving picture. However it 50 51 25 seems that the full impact of the training at a community level does not as yet show in the results. We 52 53 26 believe that the dedication shown by the trainees coupled with their new skills and knowledge will 54 55 27 have a positive impact over the coming years as more health centres are upgraded and adequately 56 57 28 resourced. 58 59 29 60 15 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 16 of 72 BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 1 2 1 Competing Interests: No, there are no competing interests. 3 4 2 Author contributions: No additional data available. 5 6 3 Author contributions 7 8 4 DE, JPOH, GM, SB, and SP were involved in conception and design of the study. DE drafted the 9 10 5 manuscript supported by all authors. JPOH, GM, SB, SP and DD were responsible for the design of 11 12 6 the training. GM, SB, SP, PK, AN, HMM and DD were responsible for the management and delivery 13 14 7 of the training. AS and FM, carried out the fieldwork and collated results supervised by DE. 15 For peer review only 16 8 17 18 9 Funding and Acknowledgements 19 20 10 Enhancing Human Resources and Use of Appropriate Technologies for Maternal and Perinatal 21 22 11 Survival in subSaharan Africa (ETATMBA) is a collaborative project funded by the European 23 24 12 Commission, Seventh Framework Programme THEME [HEALTH.2010.3.42] [Project no. 266290]. 25 26 13 This study was embedded within this programme of work. All authors are part of the ETATMBA 27 28 14 team. The ETATMBA team would like to thank all of the AMOs and the district medical and nursing 29 30 15 officers for their hard work and support. This project benefited from facilities funded through 31 32 16 Birmingham Science City Translational Medicine Clinical Research and Infrastructure Trials 33 34

17 Platform, with support from Advantage West Midlands http://bmjopen.bmj.com/ 35 36 18 37 The ETATMBA Study Group 38 39 40 41 42

43 on October 1, 2021 by guest. Protected copyright. 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 16 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 17 of 72 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 1 2 Malawi University of Malawi College Sweden Karolinska Institutet, Sweden 3 of Medicine Staffan Bergström 4 Francis Kamwendo 5 Chisale Mhango 6 WanangwaChimwaza United GE Healthcare 7 ChikayikoChiwandira Kingdom Alan Davies 8 Queen Dube 9 The University of Warwick, UK 10 Ministry of Health, Malawi Paul O'Hare 11 Fannie Kachale Siobhan Quenby 12 ChimwemweMvula Douglas Simkiss 13 David Davies 14 Tanzania Ifakara Health Institute, David Ellard 15 ForTanzania peer reviewFrances only Griffiths 16 Godfrey Mbaruku Ngiangabakwin, Kandala 17 Paul Kihaile AnneMarie Brennan 18 Hamed Mohamed Edward Peile 19 Aloisia Shemdoe AnneMarie Slowther 20 FestoMazuguni SaliyaChipwete 21 Tanzanian Training Centre Paul Beeby 22 for International Health Gregory Eloundou 23 Senga Pemba Harry Gee 24 Sidney Ndeki Vinod Patel 25 Angelo Nyamtema 26 1 27 28 2 29 3 30 31 32 33 34 35 http://bmjopen.bmj.com/ 36 37 38 39 40 41 42

43 on October 1, 2021 by guest. Protected copyright. 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 17 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 18 of 72 BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 1 2 1 REFERENCES 3 4 2 1. WHO, GHWA. A universal truth: no health without a workforce. 2014. 5 6 3 http://www.who.int/workforcealliance/knowledge/resources/GHWA 7 8 4 a_universal_truth_report.pdf?ua=1. 9 10 5 2. Hoyler M, Finlayson SG, McClain C, et al. Shortage of Doctors, Shortage of Data: A Review of the 11 12 6 Global Surgery, Obstetrics, and Anesthesia Workforce Literature. World J Surg 13 14 7 2014;38(2):26980 doi: 10.1007/s002680132324y. 15 For peer review only 16 8 3. WHO. Human Resources for Health Observer. 2012; (11). 17 18 9 http://www.who.int/hrh/resources/observer/en/. 19 20 10 4. Anyangwe SC, Mtonga C. Inequities in the global health workforce: the greatest impediment to 21 22 23 11 health in subSaharan Africa. International journal of environmental research and public 24 25 12 health 2007;4(2):93100 26 27 13 5. Kinfu Y, Dal Poz MR, Mercer H, et al. The health worker shortage in Africa: are enough 28 29 14 physicians and nurses being trained? Bulletin of the World Health Organization 30 31 15 2009;87(3):22530 32 33 16 6. National Bureau of Statistics [Tanzania]. Tanzania Service Provision Assessment Survey 2006: 34 35 17 Key Findings on Family Planning, Maternal and Child Health, and Malaria. 2006. http://bmjopen.bmj.com/ 36 37 18 http://dhsprogram.com/pubs/pdf/SR130/SR130.pdf. 38 39 19 7. Magoma M, Requejo J, Campbell OM, et al. High ANC coverage and low skilled attendance in a 40 41 20 rural Tanzanian district: a case for implementing a birth plan intervention. BMC pregnancy 42

43 21 and childbirth 2010;10:13 doi: 10.1186/147123931013. on October 1, 2021 by guest. Protected copyright. 44 45 22 8. MoHSW. Tanzania service availability and readiness assessment (SARA) 2012. 2013. 46 47 23 http://ihi.eprints.org/2448/1/SARA_2012_Report.pdf. 48 49 24 9. Mselle LT, Moland KM, Mvungi A, et al. Why give birth in health facility? Users' and providers' 50 51 25 accounts of poor quality of birth care in Tanzania. BMC Health Serv Res 2013;13:174 doi: 52 53 26 10.1186/1472696313174. 54 55 27 10. Sarker M, Schmid G, Larsson E, et al. Quality of antenatal care in rural southern Tanzania: a 56 57 28 reality check. BMC Res Notes 2010;3:209 doi: 10.1186/175605003209. 58 59 60 18 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 19 of 72 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 1 2 1 11. Mullan F, Frehywot S. Nonphysician clinicians in 47 subSaharan African countries. Lancet 3 4 2 2007;370(9605):215863 doi: 10.1016/S01406736(07)607855. 5 6 3 12. McCord C, Mbaruku G, Pereira C, et al. The quality of emergency obstetrical surgery by assistant 7 8 4 medical officers in Tanzanian district hospitals. Health Aff (Millwood) 2009;28(5):w87685 9 10 5 doi: 10.1377/hlthaff.28.5.w876. 11 12 6 13. Nyamtema AS, Pemba SK, Mbaruku G, et al. Tanzanian lessons in using nonphysician clinicians 13 14 7 to scale up comprehensive emergency obstetric care in remote and rural areas. Human 15 For peer review only 16 8 resources for health 2011;9:28 doi: 10.1186/14784491928. 17 18 9 14. Chilopora G, Pereira C, Kamwendo F, et al. Postoperative outcome of caesarean sections and 19 20 10 other major emergency obstetric surgery by clinical officers and medical officers in Malawi. 21 22 11 Human resources for health 2007;5:17 doi: 10.1186/14784491517. 23 24 12 15. Pereira C, Cumbi A, Malalane R, et al. Meeting the need for emergency obstetric care in 25 26 13 Mozambique: work performance and histories of medical doctors and assistant medical 27 28 14 officers trained for surgery. BJOG : an international journal of obstetrics and gynaecology 29 30 15 2007;114(12):15303 doi: 10.1111/j.14710528.2007.01489.x. 31 32 16 16. Bergstrom S. Who will do the caesareans when there is no doctor? Finding creative solutions to 33 34

17 the human resource crisis. BJOG : an international journal of obstetrics and gynaecology http://bmjopen.bmj.com/ 35 36 18 2005;112(9):11689 doi: 10.1111/j.14710528.2005.00719.x. 37 38 19 17. Davis D, Evans M, Jadad A, et al. The case for knowledge translation: shortening the journey 39 40 20 from evidence to effect. BMJ 2003;327(7405):335 doi: 10.1136/bmj.327.7405.33. 41 42 21 18. McGlynn EA, Asch SM, Adams J, et al. The quality of health care delivered to adults in the 43 on October 1, 2021 by guest. Protected copyright. 44 22 United States. The New England journal of medicine 2003;348(26):263545 doi: 45 46 47 23 10.1056/NEJMsa022615. 48 49 24 19. Brennan AM. ETATMBA: Enhancing human resources and the use of appropriate technologies 50 51 25 for maternal and perinatal survival in subSaharan Africa. Secondary ETATMBA: Enhancing 52 53 26 human resources and the use of appropriate technologies for maternal and perinatal survival 54 55 27 in subSaharan Africa [Web Pages] 20 Nov. 2013. 56 57 28 http://www2.warwick.ac.uk/fac/med/about/global/etatmba/about/. 58 59 60 19 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 20 of 72 BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 1 2 1 20. WHO. The World Health Report: Working together for health. Geneva: World Health 3 4 2 Organisation, 2006. 5 6 3 21. Ellard DR, Chimwaza W, Davies D, et al. Can training in advanced clinical skills in obstetrics, 7 8 4 neonatal care and leadership, of nonphysician clinicians in Malawi impact on clinical 9 10 5 services improvements (the ETATMBA project): a process evaluation. BMJ Open 2014;4(8) 11 12 6 doi: 10.1136/bmjopen2014005751. 13 14 7 22. EvjenOlsen B, Olsen OE, Kvale G. Achieving progress in maternal and neonatal health through 15 For peer review only 16 8 integrated and comprehensive healthcare services experiences from a programme in 17 18 9 northern Tanzania. International journal for equity in health 2009;8:27 doi: 10.1186/1475 19 20 10 9276827. 21 22 11 23. Kihaile P, Mbaruku G, Pemba S. Improved Maternal and Perinatal Mortalities by Trained Medical 23 24 12 Assistant Staffs in Rural Tanzania. J Health Med Informat 2013;S11:007 doi: 10.4172/2157 25 26 13 7420.S11007. 27 28 14 24. Nelissen EJ, Mduma E, Ersdal HL, et al. Maternal near miss and mortality in a rural referral 29 30 15 hospital in northern Tanzania: a crosssectional study. BMC pregnancy and childbirth 31 32 16 2013;13:141 doi: 10.1186/1471239313141. 33 34

17 25. Ministry of Health and Social Welfare. Mpango wa Maendeleo wa Afya ya Msingi (MMAM) http://bmjopen.bmj.com/ 35 36 18 20072017 (Primary Health Services Development Programme,PHSDP). United Republic of 37 38 19 Tanzania: Ministry of Health and Social Welfare, 2007. 39 40 20 26. Ng'anjo Phiri S, Kiserud T, Kvale G, et al. Factors associated with health facility childbirth in 41 42 21 districts of Kenya, Tanzania and Zambia: a population based survey. BMC pregnancy and 43 on October 1, 2021 by guest. Protected copyright. 44 22 childbirth 2014;14:219 doi: 10.1186/1471239314219. 45 46 47 23 48 49 50 51 52 53 54 55 56 57 58 59 60 20 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 21 of 72 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 1 2 Table 1. Health facilities where the Tanzanian ETATMBA trainees were based in 2013 3 District Name of facility Operating CEmOC or No. Trainees 4 5 Theatre BEmOC 6 1 Bukombe Bukombe District Hospital Yes CEmOC 1 AMO 7 8 2 Bukombe Uyovu Health Centre No BEmOC 1 AMO, 1CO 9 10 3 Geita Nzela Health Centre Yes CEmOC 1 NMW, 1 Nurse 11 4 Geita Katoro Health Centre No BEmOC 1 AMO, 1 NMW 12 13 5 Inyonga Mamba Health Centre Yes CEmOC 1 NMW 14 15 6 KaramboFor Matai Healthpeer Centre reviewNo only BEmOC 1 AMO, 1NMW 16 7 Liwale Liwale District Hospital No CEmOC 2 AMOs 17 18 8 Meatu Mwandoya Health Centre No BEmOC 1 AMO, 1 NMW 19 20 9 Mpanda Mpanda District Hospital Yes BEmOC 1 AMO, 1 Nurse 21 10 Nachingwea Nachingwea District Yes CEmOC 2 AMOs 22 23 Hospital 24 11 Nkasi Kirando Health Centre Yes CEmOC 2 AMOs 25 26 12 Nyanghwale Nyanghwale Health Centre No BEmOC 1 AMO, 1 NMW 27 a 28 13 Nyanghwale Kharumwa District Hospital Yes CEmOC 1 AMO, 1 NMW 29 14 Ruangwa Ruangwa District Hospital Yes CEmOC 1 AMO, 1 NMW 30 31 15 Sumbawanga Laela Health Centre No BEmOC 1 AMO, 1 NMW 32 33 16 Chato Chato District Hospital Yes CEmOC 1 AMO, 1 NMW

34 b 35 17 Lindi Nyangao Mission Hospital unknown CEmOC 2 NMWs http://bmjopen.bmj.com/ 36 a Upgraded to a district hospital between 2011 &2013. b This hospital not visited so not included in 37 analysis. AMO – Assistant medical officer, NMW – Nurse midwife, Nurse – nurse/anaesthetics. 38 39 CEmOC – Comprehensive Emergency Obstetric Care, BEmOC Basic Emergency Obstetric Care 40 1 41 42 2

43 on October 1, 2021 by guest. Protected copyright. 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 21 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 22 of 72 BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 1 2 1 Table 2. Defining Basic and Comprehensive Emergency Obstetric and newborn care 3 4 2 (BEmOC & CEmOC) 5 6 7 3 Basic emergency obstetric and newborn care (BEmOC) is critical to reducing maternal and 8 9 4 neonatal death. This care, which can be provided with skilled staff in health centres, large or 10 11 5 small, includes the capabilities for: 12 13 14 6 • Administering antibiotics, uterotonic drugs (oxytocin) and anticonvulsants 15 For peer review only 16 7 (magnesium sulphate); 17 18 19 8 • Manual removal of the placenta; 20 21 9 • Removal of retained products following miscarriage or abortion; 22 23 10 • Assisted vaginal delivery, preferably with vacuum extractor; 24 25 26 11 • Basic neonatal resuscitation care. 27 28 12 Comprehensive emergency obstetric and newborn care (CEmOC), typically delivered in 29 30 31 13 hospitals, includes all the basic functions above, plus capabilities for: 32 33 34 14 • Performing Caesarean sections; 35 http://bmjopen.bmj.com/ 36 15 • Safe blood transfusion; 37 38 16 • Provision of care to sick and lowbirth weight newborns, including resuscitation. 39 40 41 17 Adapted from United Nations Population fund material. For more information see: 42 43 18 http://www.unfpa.org/resources/settingstandardsemergencyobstetricandnewborn on October 1, 2021 by guest. Protected copyright. 44 45 46 19 care#sthash.5rcjLhLA.dpuf 47 48 20 49 50 51 52 53 54 55 56 57 58 59 60 22 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 23 of 72 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 1 2 Table 3. Comparison of key maternal, neonatal and birth complication figures from baseline (2011) to follow 3 up (2013) 4 5 2011 2013

6 a b * 7 DH HC Total DH HC Total Difference ba 8 (n=7) (n=9) (n=7) (n=9) 9 10 Total Deliveries 17893 7326 25219 16654 7961 24615 604 11 Fresh stillbirths (FSB) (n) 287 65 352 300 68 368 16·0 12 13 FSB rate (per 1000 births) 16·0 8·9 14·0 18·0 8·5 15·0 1·0 14 15 Macerated StillbirthsFor (MSB) peer (n) 312review 61 373 only 305 111 416 43·0 16 MSB rate (per 1000 births) 17·4 8·3 14·8 18·3 13·9 16·9 2·1 17 18 Maternal deaths (n) 68 3 71 55 2 57 14·0 19 20 MD Ratio (per 100,000 births) 380 41 282 330 25 232 50 21 Caesarean deliveries (CS) (n) 1944 78 2022 1851 49 1900 122 22 23 CS rate (per 1000 births) 108·6 10·6 80·2 111·1 6·2 77·2 3·0 24 25 Postpartum Haemorrhage (PPH) (n) 200 77 277 225 86 311 34·0 26 PPH rate (per 1000 births) 11·2 10·5 11·0 13·5 10·8 12·6 1·7 27 28 Obstructed labour (Obst/Lab) (n) 114 49 163 159 23 182 19·0 29 30 Obst/Lab rate per 1000 births 6·4 6·7 6·5 9·5 2·9 7·4 0·9 31 Sepsis (n) 31 12 43 51 4 55 12·0 32 33 Sepsis rate per 1000 births 1·7 1·6 1·7 3·1 0·5 2·2 0·5

34 * 35 DH – District hospitals, HC – Health centres· Note: there are NO significant differences here so p values not http://bmjopen.bmj.com/ 36 shown. 37 1 38 39 2 40 41 3 42

43 on October 1, 2021 by guest. Protected copyright. 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 23 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 24 of 72 BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 1 2 3 Table 4. Survey findings from health facilities in Tanzania related to infrastructure 4 5 Facilities with the items 6 HC (%) DH (%) 7 Overall n=9 n=7 8 Health facility Infrastructure Availability of Power & Availability of water 9 Sufficient light source to perform tasks at night 12 (75%) 6 (67%) 6 (86%) 10 Means of ventilation 5 (31%) 1 (11%) 4 (57%) 11 12 Running water 5 (31%) 1 (11%) 4 (57%) 13 Functioning toilet 9 (56%) 6 (67%) 3 (43%) 14 Functional fan/air conditioning 5 (31%) 1 (11%) 4 (57%) 15 ForCurtains/means peer of providing review patient privacy only14 (88%) 9 (100%) 5 (71%) 16 Waiting area for visitors and family 6 (38%) 4 (43%) 2 (33%) 17 18 Facility with electricity 14 (89%) 8 (86%) 6 (86%) 19 Motor Vehicle Ambulance Available and functional 6 (38%) 1 (11%) 5 (71%) 20 Available and functional landline telephone in the maternity area 4 (25%) 4 (43%) 0 (0%) 21 Delivery bed / table 11 (69%) 5 (56%) 6 (86%) 22

23 24 Availability of health related registers 25 General admission register 11 (69%) 5 (56%) 6 (86%) 26 Delivery register 16 (100%) 9 (100%) 7 (100%) 27 Maternity ward register 9 (56%) 4 (44%) 5 (71%) 28 Female ward register 9 (56%) 4 (44%) 5 (71%) 29 30 Operating theatre register 10 (63%) 4 (44%) 6 (86%) 31 Gynaecology register 0 (0%) 0 0 32 Postabortion register 9 (56%) 4 (44%) 5 (71%) 33 Individual patient records 1 (6%) 0 1 (14%) 34 35 Discharge register 1 (6%) 0 1 (14%) http://bmjopen.bmj.com/ 36 Death register 11 (69%) 6 (67%) 5 (71%) 37 Mortuary register 7 (44%) 2 (22%) 5 (71%) 38 Monthly / annual facility summary reports 16 (100%) 9 (100%) 7 (100%) 39 DH – District hospitals, HC – Health centres 40 41 1 42 2 43 on October 1, 2021 by guest. Protected copyright. 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 24 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 25 of 72 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 1 2 Table 5.Survey findings from health facilities in Tanzania related to the availability of equipment, 3 supplies and drugs 4 5 facilities with the equipment 6 Overall (%) HC (%) DH (%) 7 N=16 N=9 N=7 8 Drugs and equipment: Availability of Items for normal delivery 9 Equipment and Supplies 10 Blood pressure cuff/machine 13 (81%) 7 (78%) 6 (86%) 11 12 Stethoscope 15 (94%) 8 (89%) 7 (100%) 13 Fetal stethoscope 16 (100%) 9 (100%) 7 (100%) 14 Clinical thermometer 13 (81%) 6 (67%) 7 (100%) 15 For peer reviewSterile gloves 16 (100%) only 9 (100%) 7 (100%) 16 Nonsterile protective clothing/apron 15 (94%) 8 (89%) 7 (100%) 17 18 Scissors or razor blade for cutting cord 15 (94%) 9 (100%) 6 (86%) 19 Cord ties 10 (63%) 5 (56%) 5 (71%) 20 Needles and Syringes 8 (50%) 4 (44%) 4 (57%) 21 IV fluid set (giving set) 15 (94%) 9 (100%) 6 (86%) 22 Suture needles and suture materials 10 (63%) 5 (56%) 5 (71%) 23 24 suction apparatus 8 (50%) 3 (33%) 5 (71%) 25 Manual vacuum extractor 5 (31%) 2 (33%) 2 (29%) 26 Obstetric forceps 11 (69%) 8 (89%) 3 (43%) 27 Drugs 28 29 Pitocin (Oxytocin) 13 (81%) 6 (67%) 7 (100%) 30 Ergometrine (injectable) 4 (25%) 3 (33%) 1 (14%) 31 Normal saline 14 (88%) 8 (89%) 6 (86%) 32 Ringers lactate 7 (44%) 2 (22%) 5 (71%) 33 Dextrose / glucose 9 (56%) 3 (33%) 6 (86%) 34 35 Lignocaine 2% or 1% 15 (94%) 8 (89%) 7 (100%) http://bmjopen.bmj.com/ 36 injectable antibiotic 5 (31%) 3 (33%) 2 (29%) 37 Magnesium sulphate 14 (88%) 8 (89%) 6 (86%) 38 Diazepam 6 (38%) 3 (33%) 3 (43%) 39 Skin disinfectant 12 (75%) 7 (78%) 5 (71%) 40 41 42 Availability of Infection prevention services in labour delivery/operating theatres

43 Decontamination container with prepared solution 11 (69%) 5 (56%) 6 (86%) on October 1, 2021 by guest. Protected copyright. 44 Covered contaminated trash bin 11 (69%) 6 (67%) 5 (71%) 45 Sharps container 12 (75%) 6 (67%) 6 (86%) 46 47 Soap 0 0 0 48 Antiseptics 10 (63%) 5 (56%) 5 (71%) 49 Chlorine/ Bleach 6 (38%) 2 (22%) 4 (57%) 50 Sterile gloves 12 (75%) 6 (67%) 6 (86%) 51 Other items 52 53 Regular trash bin 12 (75%) 6 (67%) 6 (86%) 54 Non sterile gloves 12 (75%) 6 (67%) 6 (86%) 55 Nonsterile protective clothing 12 (75%) 6 (67%) 6 (86%) 56 DH – District hospitals, HC – Health centres 57 1 58 59 60 25 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 26 of 72 BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 1 2 1 3 Table 6.Survey findings from health facilities in Tanzania related to the availability of items 4 for management of anaesthesia, birth complications and caesarean section 5 6 Facilities with the items 7 Equipment Overall HC DH 8 N=16 N=9 N=7 9 10 Items for provision of anaesthesia 11 Suction machine 6 (38%) 4 (44%) 2 (29%) 12 Filled oxygen cylinder with cylinder carrier 8 (50%) 2 (22%) 6 (86%) 13 and key to open valve 14 Intubating forceps (Magill) 6 (38%) 4 (44%) 2 (29%) 15 For peerAdult laryngoscope review 11 (69%) 6only (67%) 5 (71%) 16 17 Adult ventilator bag and mask 11 (69%) 6 (67%) 5 (71%) 18 IV fluid set (giving set) 10 (63%) 5 (56%) 5 (71%) 19 Spinal needles (18gauge to 25gauge) 3 (19%) 1 (11%) 2 (29%) 20 Endotracheal tubes with cuffs (8 – 10mm) 9 (56%) 4 (44%) 5 (71%) 21 Halothane 6 (38%) 3 (33%) 3 (43%) 22 23 Ketamine 11 (6%) 5 (56%) 6 (86%) 24 Anaesthetic face masks 9 (56%) 5 (56%) 4 (57%) 25 26 Items for management of pre-eclampsia/

27 eclampsia 28 Magnesium Sulphate 7 (44%) 4 (44%) 3 (43%) 29 Diazepaminjectable 10 (63%) 4 (44%) 6 (86%) 30 31 Niphedipine 1 (6%) 0 (0%) 1 (14%) 32 Blood pressure cuff/machine 13 (81%) 7 (78%) 6 (86%) 33 Stethoscope 15 (94%) 8 (89%) 7 (100%) 34

Adult ventilator bag and mask 13 (81%) 7 (78%) 6 (86%) http://bmjopen.bmj.com/ 35 36 Needles and Syringes 4 (25%) 1 (11%) 3 (43%) 37 Urinary catheters (Foleys) 8 (50%) 3 (33%) 5 (71%) 38 Uristix 4 (25%) 1 (11%) 3 (43%) 39 40 Items for management of haemorrhage (parenteral uterotonics) 41 42 Needles and Syringes 8 (50%) 4 (44%) 4 (57%)

43 IV fluid set (giving set) 9 (56%) 3 (33%) 6 (86%) on October 1, 2021 by guest. Protected copyright. 44 45 Items for Caesarean Section (not including anaesthesia) 46 Operating table 47 48 Light adjustable, shadow less 11 (69%) 6 (56%) 5 (86%) 49 Antiseptics 10 (63%) 5 (56%) 5 (71%) 50 Sterile gloves 12 (75%) 6 (67%) 6 (86%) 51 Cord ties 10 (63%) 5 (56%) 5 (71%) 52 Needles and Syringes 6 (38%) 4 (44%) 2 (29%) 53 54 Benzyl Penicillin 4 (25%) 3 (33%) 1 (14%) 55 Metronidazole (IV) 2 (13%) 1 (11%) 1 (14%) 56 Gentamycin (IV) 1 (6%) 1 (11%) 0 (0%) 57 Caesarean section pack 58 needle holder 13 (81%) 7 (78%) 6 (86%) 59 60 26 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 27 of 72 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 1 2 scalpel handle with blade 10 (63%) 5 (56%) 5 (71%) 3 retractor 12 (75%) 6 (67%) 6 (86%) 4 surgical scissors 12 (75%) 6 (67%) 6 (86%) 5 a suction apparatus / 8 6 (38%) 4 (44%) 2 (29%) 6 7 oxygen 8 (50%) 2 (22%) 6 (86%) 8 sutures 11 (69%) 5 (56%) 6 (86%) 9 ketamine 11 (69%) 5 (56%) 6 (86%) 10 a lidocaine / 5 12 (75%) 6 (67%) 6 (86%) 11 DH – District hospitals, HC – Health centres 12 a The numbers against these items (8 & 5) are the units for this item to be classed as 13 available (e.g. there has to be 8 suction apparatus for it to be classed as available) 14 1 15 For peer review only 16 2 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 http://bmjopen.bmj.com/ 36 37 38 39 40 41 42

43 on October 1, 2021 by guest. Protected copyright. 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 27 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 28 of 72

1 2 3

4 Supplementary appendix to go with: Can training non- BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 5 6 physician clinicians/associate clinicians (NPCs/ACs) in 7 8 emergency obstetric, neonatal care and clinical leadership 9 10 make a difference to practice and help towards reductions 11 12 in maternal and neonatal mortality in rural Tanzania? The 13 14 ETATMBA Project. 15 16 17 For peer review only 18 19 20 21 22 23

24 25 26 27 28 29 Table of Contents 30 31 32 33 34 35 ETATMBA Training overview (Tanzania)*(Supplementary appendix 1) ...... 2 36 37 38 ETATMBA, Tanzania. Survey Instruments (Supplementary Appendix 2) ...... 6

39 http://bmjopen.bmj.com/ 40 41 42 43 44 45 46

47 on October 1, 2021 by guest. Protected copyright. 48 49 50 51 52 53 54 55 56 57 58 59 60

1

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 29 of 72 BMJ Open

1 2 3

4 ETATMBA Training overview (Tanzania)*(Supplementary appendix 1) BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 5 6 7 TRAINING OF NPCS IN MATERNAL HEALTH AND LEADERSHIP IN TANZANIA 8 9 BACKGROUND 10 11 With a physician to population ratio of close to 1:25,000 and further challenged by 80-90% of 12 medical doctors practicing in urban areas the majority of the Tanzania population has no access to 13 14 care by a physician 15 16 However, there is potential of using the existing mid-level cadres in the system to provide essential 17 services if their skillsFor can be upgradedpeer through review targeted short courses only or long courses 18 19 Assistant Medical officers (AMOs) and Nurse Midwives (NMW) for instance can effectively be 20 upgraded to provide Comprehensive Emergency Obstetric Care CEmOC in rural areas where Medical 21 doctors are not willing or ready to work. 22 23 THE ETATMBA PROJECT: TRAINING OF ADVANCED LEADERS IN MATERNAL HEALTH 24 25 To implement the ETATMBA Project, The Ifakara Health Institute (IHI) in collaboration with Tanzania 26 Training centre for International Health (TTCIH) set out to train Non Physician Clinicians (NPCs) from 27 28 disadvantaged Health Centres (HC) and District Hospitals in CEmOC, anaesthesia and leadership and 29 thereafter assess its impact on Maternal and Perinatal Mortality in their respective areas. 30 31 It is important to note here that the ETATMBA training package, whilst delivered by National and 32 International experts and based on evidenced based practice, was designed specifically for work in 33 low resource settings. 34 35 SELECTION AND RECRUITMENT OF TRAINEES 36 37 Selection and recruitment was carried out by the Tanzanian ETATMBA Obstetricians in collaboration 38 with the Ministry of Health and local District Medical Officers (who were made aware of the project 39 http://bmjopen.bmj.com/ 40 and its requirements) 41  To be able to show an impact, a pair (AMO, NMW) were recruited from one health centre 42 43  The NPCs were selected from facilities that were remote and hardly accessible during rainy 44 season. 45  The Facilities given priority were Heath Centres (HCs) that were already upgraded with 46 theatres or were in the process of being upgraded.

47 on October 1, 2021 by guest. Protected copyright. 48  However, in certain disadvantaged districts there were no such HCs and therefore the 49 District hospital was selected. 50 51 DURATION OF TRAINING 52 53  Duration of Training was 16 weeks undertaken at the Tanzanian Training Centre and St. 54 Francis Referral Hospital in Ifakara. 55

56  AMOs: 10 weeks on CEmOC, 2 weeks on Leadership and 4 weeks of internship. 57  Nurses: 10 weeks on Anaesthesia, 2 weeks on leadership and 4weeks on internship 58  The internship was done in their Regional Hospitals 59 60 DAILY TIME TABLE FOR NPC TRAINING

1. 7.45 am - 9.00 am: Review of Emergency cases managed by AMO within 24 hrs

2

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 30 of 72

1 2 3 2. 9.00 am-1.00 pm Major ward round 4 BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 5 3. 3.00-5.00 pm: Lecture on EMOC topics including Neonatal Resuscitation. 6 4. 5.00 pm-7:30 am: Night Duty for one AMO daily including weekends and public holidays 7 8 WHAT TRAINING STRATEGY ARE WE USING? 9 10  We are training the AMOs and NMWs using a competence based education approach 11  Competence-based education is an approach that is related to on-the-job performance and 12 has a close relationship between the knowledge or skill required and on-the-job application 13 14 TRAINING CURRICULUM 15 16  Two competence based training curricula are used (one for AMOs in EmOC and another for 17 For peer review only Nurses in Anaesthesia) 18 19  The actual training duration is three months (12 weeks) full time with an addition of one 20 month (4weeks) for internship 21  The implementation of the curriculum requires the use of training facilities at a training 22 23 centre as well as opportunities to practice at the hospital 24  Training materials addressing the various areas of CEMOC are prepared in advance and 25 availed to trainees 26  A schedule detailing the day to day topics is usually prepared to standardise the 27 28 teaching/learning process 29  A logbook to guide the training of skills is given to each trainee 30 31 All NPC (AMO and Nurses) underwent 10 day- Leadership course during which management and 32 leadership skills were taught by experts in this area. 33 34 ADVANCED OBSTETRICS CARE COURSE 35 36 The training tools for the Advanced Obstetrics Care Course includes the following: 37 38  Curriculum in Maternal and Leadership for Non Physician Clinicians 39  Course Programme http://bmjopen.bmj.com/ 40  Modules and power Point presentations on: 41 42 - MNH and Health Systems 43 - ANC and Diseases in Pregnancy 44 - Haemorrhage 45 - Partograph 46 - Prolonged Obstructed Labour 47 on October 1, 2021 by guest. Protected copyright. 48 - Abortion 49 - Pre-eclampsia 50 - Immediate New-born Care 51 52 - Criterion based Audit in Obstetrics 53 - Clinical Leadership (see below) 54  E-Learning package in maternal health 55  56 Advanced Life Saving Skills Booklet by MOHSW 57  Beyond the Numbers: A Manual by WHO 58  Monitoring Emergency Obstetric Care by WHO 59 60 LEADERSHIP: OBJECTIVES and LEARNING OUTCOMES

 Provide leadership in key areas of maternal health services

3

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 31 of 72 BMJ Open

1 2 3  Conduct organizational capacity assessment 4 BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 5  Manage change in a health facility 6  Improve quality of services 7 8 TOPICS IN LEADERSHIP AND MANAGEMENT 9 10 11 12 13 14 15 16 17 For peer review only 18 19 20 21 22 23 24 TRAINING METHOD 25 26 27 28 29 30 31 32 33 34 35 36 37 WHO ARE THE TRAINERS? 38 The trainers comprised the following: 39 http://bmjopen.bmj.com/ 40  Obstetricians local and international 41 42  Anaesthetists/Anaesthetic Officers 43  General Medical Officers working in OBGY Department 44  Nurse midwives working in the in maternity ward 45 46  Leadership and Management experts

47 on October 1, 2021 by guest. Protected copyright. 48 TEACHING METHODS 49 A variety of teaching/learning methods were used including: 50 51  Lectures and discussions 52 53  Involving the trainees in the actual doing through practicing in the theatre as well as in the 54 maternity and surgical wards at the teaching hospital 55  Encouraging trainees to conduct their own group discussions and presentations and availing 56 57 trainees the opportunity for self-study 58  Use of Clinical Skills Lab for demonstrations and simulations 59  A team approach to learning is emphasised where AMOs and Nurse Midwives learn together 60 TRAINING CURRICULUM: ASSESSMENT

4

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 32 of 72

1 2 3  The assessment process is designed to promote the highest possible standards of 4 BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 5 demonstrable achievement within and beyond the scope and content of the curriculum. 6  To achieve this, assessment of the trainees involves the following: 7 o Course work and individual assignments 8 o Class participation and group assignments 9 10 o Self-assessment 11 o Monthly tests 12 o End of module assessment using written exams and OSCE 13 14 INTERNSHIP 15 16 The last 4 weeks of the course were on Internship in their Regional Hospitals 17 For peer review only 18  No Lectures were given at the hospitals. 19  AMOs managed the Maternity wards as well as labour wards under our supervision and did 20 21 surgical procedures (C/S, vacuum etc.) on patients with indications. 22  The nurses administered anaesthesia and did neonatal resuscitation in the theatre 23 24 MODE OF INTERNSHIP 25 26  Clinical meeting- attended daily by NPCs, all senior hospital staff and supervisor 27  Lively discussion of all cases admitted during the 24 hrs. 28  Administrative problems discussed and sorted out 29 30  Acts as a forum for continuing education and leadership role in action 31  Daily ward rounds in the maternity ward 32  Discussion on management of cases 33  34 Focus on correct use and interpretation of the partograms 35  Demonstration of practical procedures 36  Applying skills and mentorship both in labour ward and theatre 37 38

39 http://bmjopen.bmj.com/ 40 *Adapted from material on the ETATMBA website 41 http://www2.warwick.ac.uk/fac/med/about/global/etatmba/ 42 43 44 45 46

47 on October 1, 2021 by guest. Protected copyright. 48 49 50 51 52 53 54 55 56 57 58 59 60

5

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 33 of 72 BMJ Open

1 2 3

4 ETATMBA, Tanzania. Survey Instruments (Supplementary Appendix 2) BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 5 6 7 ETATMBA Facility Assessment (A) 8 9 10 Module 1: Moduli kuu (KAMILI) 11 12 Name of the Interviewer ______13 14 Date of the interview: ______15 16 GIS coordinates [filled by supervisor later]______17 Instructions: For peer review only 18 19 SEHEMU YA 1: TAARIFA ZA MSINGI ZA KITUO 20 21 NA MASWALI NA MCHUJO MPANGILIO WA MAKUNDI RUKA 22 1101 23 District: ______24 25 1102 26 Name of the Facility: ______27 1103 28 ID of the Facility: ______29 30 31 32 33 34 Data Collection Sources and Quality 35 36 Please record the following information about the registers used to collect obstetric complication 37 and maternal death information. 38 39  List as per MOH guidelines. http://bmjopen.bmj.com/ 40 41 42 43 44 45 46

47 on October 1, 2021 by guest. Protected copyright. 48 49 50 51 52 53 54 55 56 57 58 59 60

6

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 34 of 72

1 2 3

4 Answer the following only for the registers BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 5 Check if used to collect that collect delivery and obstetric 6 (check all that apply): complication data and/or maternal death 7 Is data: 8 register 9 Registers and 10 available Is the Is the Data Sources Is the Is the 11 at this Obstetric Maternal register register 12 register register facility? complication death up to regularly 13 easily completely data data date? reviewed 14 CSa accessible? filled out? 15 by staff? CSb CSc 16 CSd CSe 17 For peer review only CSf CSg 18 19 1 General 20 □ 0.No □ 0.No □ □ 0.No □ □ 0.No □ □ 0.No □ □ 0.No □ 0.No □ admission 21 □ 1.Yes 1.Yes 1.Yes 1.Yes 1.Yes □ 1.Yes 1.Yes 22 register 23 24 2 Delivery □ 0.No □ 0.No □ □ 0.No □ □ 0.No □ □ 0.No □ □ 0.No □ 0.No □ 25 register □ 1.Yes 1.Yes 1.Yes 1.Yes 1.Yes □ 1.Yes 1.Yes 26 27 3 Maternity □ 0.No □ 0.No □ □ 0.No □ □ 0.No □ □ 0.No □ □ 0.No □ 0.No □ 28 ward register □ 1.Yes 1.Yes 1.Yes 1.Yes 1.Yes □ 1.Yes 1.Yes 29 30 4 Female ward □ 0.No □ 0.No □ □ 0.No □ □ 0.No □ □ 0.No □ □ 0.No □ 0.No □ 31 register □ 1.Yes 1.Yes 1.Yes 1.Yes 1.Yes □ 1.Yes 1.Yes 32 33 5 Operating 34 □ 0.No □ 0.No □ □ 0.No □ □ 0.No □ □ 0.No □ □ 0.No □ 0.No □ theatre 35 □ 1.Yes 1.Yes 1.Yes 1.Yes 1.Yes □ 1.Yes 1.Yes 36 register 37 38 6 Gynecology □ 0.No □ 0.No □ □ 0.No □ □ 0.No □ □ 0.No □ □ 0.No □ 0.No □

39 register □ 1.Yes 1.Yes 1.Yes 1.Yes 1.Yes □ 1.Yes 1.Yes http://bmjopen.bmj.com/ 40 41 7 Post-abortion □ 0.No □ 0.No □ □ 0.No □ □ 0.No □ □ 0.No □ □ 0.No □ 0.No □ 42 register □ 1.Yes 1.Yes 1.Yes 1.Yes 1.Yes □ 1.Yes 1.Yes 43 44 8 Individual 45 □ 0.No □ 0.No □ □ 0.No □ □ 0.No □ □ 0.No □ □ 0.No □ 0.No □ patient 46 □ 1.Yes 1.Yes 1.Yes 1.Yes 1.Yes □ 1.Yes 1.Yes records 47 on October 1, 2021 by guest. Protected copyright. 48 49 9 Discharge □ 0.No □ 0.No □ □ 0.No □ □ 0.No □ □ 0.No □ □ 0.No □ 0.No □ 50 register □ 1.Yes 1.Yes 1.Yes 1.Yes 1.Yes □ 1.Yes 1.Yes 51 52 10 Death □ 0.No □ 0.No □ □ 0.No □ □ 0.No □ □ 0.No □ □ 0.No □ 0.No □ 53 register □ 1.Yes 1.Yes 1.Yes 1.Yes 1.Yes □ 1.Yes 1.Yes 54 55 11 Mortuary □ 0.No □ 0.No □ □ 0.No □ □ 0.No □ □ 0.No □ □ 0.No □ 0.No □ 56 register □ 1.Yes 1.Yes 1.Yes 1.Yes 1.Yes □ 1.Yes 1.Yes 57 58 12 Monthly / 59 □ 0.No □ 0.No □ □ 0.No □ □ 0.No □ □ 0.No □ □ 0.No □ 0.No □ 60 annual facility □ 1.Yes 1.Yes 1.Yes 1.Yes 1.Yes □ 1.Yes 1.Yes summary

7

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 35 of 72 BMJ Open

1 2 3 reports/forms 4 BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 5 13 Otherooo 6 (specify): 7 □ 0.No □ 0.No □ □ 0.No □ □ 0.No □ □ 0.No □ □ 0.No □ 0.No 8 9 □ 1.Yes 1.Yes 1.Yes 1.Yes 1.Yes □ 1.Yes □1.Yes 10 11 12

13 14 NB: UN Process Indicator Data: 15 16 17 For peer review only CHUKUA TAKWIMU ZA KUANZIA January 2011- December 2011 18 19

20 21 Total 22 Year 23 F A J No Dec 24 Ja M M Ju A Se Oc e p u v 25 n ar ay l ug pt t 26 b r n 27 28 No. obstetric admissions 29 Total No of all deliveries 30 31 (SVD+CS) 32 33 Total No deliveries (SVD) 34 35 Deliveries (Breech) 36 37 Twins 38

39 BBA http://bmjopen.bmj.com/ 40 41 No. cesarean deliveries 42 43 Other mal-presentations 44 45 (Transverse, Compound etc) 46 Direct obstetric complications: 47 on October 1, 2021 by guest. Protected copyright. 48 Hemorrhage (ante & post- 49 50 partum) 51 52 Obstructed / prolonged

53 labor 54 55 Ruptured uterus 56 57 Post-partum sepsis 58 59 Severe pre-eclampsia /

60 eclampsia

8

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 36 of 72

1 2 3 Complications of abortion 4 BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 5 (with hemorrhage and/or 6 sepsis) 7 8 Ectopic pregnancy 9 10 Total direct obstetric

11 complications 12 13 Other obstetric 14 complications (from all 15 16 other causes) – Specify: 17 For peer review only Other abortion 18 19 complications 20 21 22 23 24 Year: Jan Feb Mar Apr May Jun Jul Aug Sept Oct Nov 25 Dec Total 26 27 Maternal deaths 28 – direct obstetric 29 causes: 30 31 Hemorrhage (ante

32 & post-partum) 33 34 Obstructed / 35 36 prolonged labor 37 38 Ruptured uterus

39 http://bmjopen.bmj.com/ Post-partum 40 41 sepsis 42 43 Severe pre- 44 eclampsia / 45 eclampsia 46

47 Complications of on October 1, 2021 by guest. Protected copyright. 48 abortion (with 49 50 hemorrhage 51 and/or sepsis) 52 53 Ectopic pregnancy 54 55 Total maternal 56 deaths from

57 direct obstetric 58 59 causes 60 Other maternal deaths (direct

9

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 37 of 72 BMJ Open

1 2 3 causes) – Specify: 4 BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 5 6 7 8 9 10 11 12 13 14 15 16 17 For peer review only 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38

39 http://bmjopen.bmj.com/ 40 41 42 43 44 45 46

47 on October 1, 2021 by guest. Protected copyright. 48 49 50 51 52 53 54 55 56 57 58 59 60

10

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 38 of 72

1 2 3 ** List indirect obstetric complications and maternal deaths relevant for local country context 4 BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 5 (examples: HIV, severe anemia, malaria, etc. 6 7 8 Year: Jan Feb Mar Apr May Jun Jul Aug Sept Oct Total 9 Nov Dec 10 11 Indirect obstetric 12 complications: f26a- 13 f26x 14 15 26 malaria 16 17 27 hiv/aids For peer review only 18 19 28 severe anaemia 20 21 28_1 hepatitis 22 23 28_2 other indirect

24 complications 25 26 Indirect maternal

27 deaths: 28 29 29 malaria 30 31 30 hiv/aids 32 33 31 severe anaemia 34 35 Neonatal outcomes 36 Stillbirths & 37 38 neonatal deaths:

39 http://bmjopen.bmj.com/ 40 32. Total live birth 41 33 Fresh stillbirths  42 43 2.5 Kg 44 45 34 Fresh stillbirths 

46 2.5 Kg

47 on October 1, 2021 by guest. Protected copyright. 48 35 Macerated

49 stillbirths 50 51 36 Early neonatal 52 53 deaths (< 24 hrs) 54 37 Early neonatal 55 56 deaths (> 24 hrs) 57 58 Referrals: 59 60 38 Referrals out due to direct obstetric

11

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 39 of 72 BMJ Open

1 2 3 causes 4 BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 5 39 Referrals in due 6 7 to direct obstetric 8 causes 9 10 11 12 13 14 15 16 17 For peer review only 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38

39 http://bmjopen.bmj.com/ 40 41 42 43 44 45 46

47 on October 1, 2021 by guest. Protected copyright. 48 49 50 51 52 53 54 55 56 57 58 59 60

12

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 40 of 72

1 2 3 (A.) EQUIPMENT, SUPPLIES & ESSENTIAL DRUGS

4 BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 5 6 7 Name of district: ______Name of Village: 8 ______9 10 11 Name of health facility: ______12 13 14 15 Date of Interview: ___ / ___ / ___ Start time ______:______16 17 For peer review only 18 Instructions: 19 20 21 22 Equipment and supplies: 23 24  Record whether there is a sufficient supply/number of the following items for the facility’s daily 25 caseload of deliveries, and whether the items are available & functional, available but NOT 26 functional, or not available. 27  Equipment, supplies and infrastructure are organized by room. 28 29 Essential drugs: 30 31  Record the availability and supply of drugs for each room (emergency room, labor / delivery 32 room, maternity ward, operating theatre and pharmacy). Check whether the listed drug is 33 available and if the supply is sufficient to last for less than one week, up to one week, up to 34 two weeks, up to three weeks, or up to four or more weeks. 35  Drug lists are organized by room. 36  List according to MOH guidelines. 37 38

39 A. Emergency room http://bmjopen.bmj.com/ 40 41 42 43 Code Instructions Available 44 45 A1 Does this facility have an emergency room? Kama □ 0.NO □ 1. YES 46 hakuna, nenda kwenye chumba kingine (Nenda B1)

47 on October 1, 2021 by guest. Protected copyright. 48 A2 Are obstetric complications managed in the emergency □ 0.NO □ 1. YES 49 room? 50 51 52 53 Infrastructure 54 55 ID Instructions Available 56 57 A3 Electricity □ 0.NO □ 1. YES 58 59 A4 Sufficient light source to perform tasks during the □ 0.NO □ 1. YES 60 day

13

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 41 of 72 BMJ Open

1 2 3 A5 Sufficient light source to perform tasks at night □ 0.NO □ 1. YES 4 BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 5 A6 Means of ventilation (Fan,AC Windows) □ 0.NO □ 1. YES 6 7 A7 Running water □ 0.NO □ 1. YES 8 9 A8 Functioning toilet □ 0.NO □ 1. YES 10 11 A9 Curtains/means of providing patient privacy □ 0.NO □ 1. YES 12 A10 Waiting area for visitors □ 0.NO □ 1. YES 13 14 15 16 Equipments 17 For peer review only 18 ID A. Availability B. Supply 19 20 Are there 21 At least 1 Available Not enough for the 22 Essential Items available & but NONE Availa daily caseload of 23 Functional functional ble deliveries? 24 25 26 Filled oxygen cylinder with 27 A11 cylinder carrier and key to 2 1 0 □0.NO □1.YES 28 29 open valve 30 A12 BP cuff 2 1 0 □0.NO □1.YES 31 32 A13 Stethoscope 2 1 0 □0.NO □1.YES 33 34 A14 Fetal stethoscope 2 1 0 □0.NO □1.YES 35 36 A15 Cannular 2 1 0 □0.NO □1.YES 37 Kidney basins/Beseni la 38 A16 2 1 0 □ NO □1.YES kunawia mikono 39 http://bmjopen.bmj.com/ 40 A17 Clinical thermometer 2 1 0 □0.NO □1.YES 41 42 Needles and Syringes (5-10- A18 2 1 0 □0.NO □1.YES 43 20cc) 44 45 Suture needles/suture A19 2 1 0 □0.NO □1.YES 46 materials

47 on October 1, 2021 by guest. Protected copyright. 48 A20 IV Drip Stand(s) 2 1 0 □0.NO □1.YES 49 50 A21 Urinary catheters 2 1 0 □0.NO □1.YES 51 A22 Adult ventilator bag and mask 2 1 0 □0.NO □1.YES 52 53 A23 Mouth gag 2 1 0 □0.NO □1.YES 54 55 Patient transport (wheelchair, A24 2 1 0 □0.NO □1.YES 56 trolley, hammock) 57 58 A25 Examination table with privacy 2 1 0 □0.NO □1.YES 59 A26 Uris tix /Albustix (dip stick for 2 1 0 60 □0.NO □1.YES protein in urine

14

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 42 of 72

1 2 3 ID A. Availability B. Supply 4 BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 5 Are there 6 enough for the 7 At least 1 Available Not Essential Items available & but NONE Availa daily caseload of 8 deliveries? 9 Functional functional ble 10 11 12 Infection prevention 13 A □0.NO □1.YES 14 Soap 2 1 0 15 27 16 17 A28 AntisepticsFor (Kama Detolpeer nk) 2 review1 only0 □0.NO □1.YES 18 Sterile gloves (Pea ya glovu 19 A29 2 1 0 □0.NO □1.YES zilizofungwa) 20 21 Non-sterile gloves (Glovu □0.NO □1.YES A30 2 1 0 22 hazijafungwa) 23 24 A31 Non-sterile protective clothing 2 1 0 □0.NO □1.YES 25 26 A32 Decontamination container 2 1 0 □0.NO □1.YES 27 Bleach or bleaching powder 28 A33 2 1 0 □0.NO □1.YES 29 (Jik) 30 31 A34 Prepared disinfection solution 2 1 0 □0.NO □1.YES 32 A35 Regular trash bin 2 1 0 □0.NO □1.YES 33 34 Puncture proof sharps A36 2 1 0 □0.NO □1.YES 35 container 36 37 38

39 Essential drugs http://bmjopen.bmj.com/ 40 ID Drugs Available 41 42 A37 Antibiotics 43 44 A37a Amoxicillin □0.NO □1.YES 45 46 A37b Ampicillin □0.NO □1.YES

47 on October 1, 2021 by guest. Protected copyright. 48 A37c Benzyl penicillin (x-pen) □0.NO □1.YES 49 A37d Cloxacillin □0.NO □1.YES 50 51 A37e Erythromicin □0.NO □1.YES 52 53 A37f Gentamicin □0.NO □1.YES 54 55 A37g Metranidazole (Flagyl) □0.NO □1.YES 56 57 A37h Nitrofurantoin □0.NO □1.YES 58 A37i Penicillin G □0.NO □1.YES 59 60 A37j Procaine penicillin G (PPF) □0.NO □1.YES

15

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 43 of 72 BMJ Open

1 2 3 ID Drugs Available

4 BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 5 A37k Trimethoprim/Sulfamethoxazole (Septrine) □0.NO □1.YES 6 7 A38 Anticonvulsants 8 9 A38a Magnesium sulfate □0.NO □1.YES 10 A38b Phenytoin □0.NO □1.YES 11 12 A39 Antihypertensives 13 14 A39a Hydralazine □0.NO □1.YES 15 16 A39b Nifedipine □0.NO □1.YES 17 For peer review only 18 A40 Dawa zinazotumika wakati wa dharura 19 A40a Adrenaline □0.NO □1.YES 20 21 A40b Aminophylline □0.NO □1.YES 22 23 A40c Atropine sulfate □0.NO □1.YES 24 25 A40d Calcium gluconate □0.NO □1.YES 26 27 A40e Digoxin □0.NO □1.YES 28 A40f Ephedrine □0.NO □1.YES 29 30 A40g Frusemide(Lasix) □0.NO □1.YES 31 32 A40h Naloxone □0.NO □1.YES 33 34 A40i Nitroglycerine □0.NO □1.YES 35 36 A40j Prednisolone □0.NO □1.YES 37 A40k Promethazine (Phenergan) □0.NO □1.YES 38 39 A41 Analgesics http://bmjopen.bmj.com/ 40 41 A41a Paracetamol/Asprin □0.NO □1.YES 42 43 A41b Pethidine/ramadol □0.NO □1.YES 44 45 A42 Sedatives 46 A42a Diazepam (Valium) □0.NO □1.YES

47 on October 1, 2021 by guest. Protected copyright. 48 A42b Phenobarbitone □0.NO □1.YES 49 50 A43 IV Fluids 51 52 A43a Dextrose 5% □0.NO □1.YES 53 A43b Normal saline □0.NO □1.YES 54 □ □ 55 A43c Ringer’s lactate □0.NO □1.YES 56 57 A44 PMTCT/HIV care 58 59 A44a ART □0.NO □1.YES 60 A44b Niverapine □0.NO □1.YES

16

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 44 of 72

1 2 3 ID Drugs Available

4 BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 5 A44c Rapid testing kit □0.NO □1.YES 6 7 8 9 B. Labor / Delivery room 10 Kama chumba cha kusubiri wakati wa uchungu ni tofauti na kile cha kujifungulia, tafadhali chukulia 11 kama ni chumba kimoja 12 13 14 ID Available 15 16 B1 Does this facility have a labor / delivery room? Kama jibu □ 0.NO □ 1. YES 17 ni NO, nendaFor kwenye chumba peer kingine (Nenda review C1) only 18

19 20 Infrastructure 21 22 23 ID Available 24 25 B2 Electricity □ 0.NO □ 1. YES

26 Sufficient light source to perform tasks during the □ 0.NO □ 1. YES 27 B3 day 28 29 B4 Sufficient light source to perform tasks at night □ 0.NO □ 1. YES Vifaa 30 31 B5 Means of ventilation (Fan,AC or Window) □ 0.NO □ 1. YES 32 33 B6 Running water □ 0.NO □ 1. YES 34 35 B7 Functioning toilet □ 0.NO □ 1. YES 36 B8 Panga boi (Fan) □ 0.NO □ 1. YES 37 38 Curtains/means of providing patient privacy □ 0.NO □ 1. YES

B9 http://bmjopen.bmj.com/ 39 (Screen) 40 41 B10 Waiting area for visitors (iliyo na viti na kivuli) □ 0.NO □ 1. YES 42 43 ID A. Available B. Supply 44 Are there enough 45 At least 1 Available NOT 46 for the daily Essential Items available & but NONE Availa

47 caseload of on October 1, 2021 by guest. Protected copyright. Functional functional ble 48 deliveries? 49 50 B11 Delivery bed with stirrups 2 1 0 □0.NO □1.YES 51 52 B12 Delivery bed (no stirrups) 2 1 0 □0.NO □1.YES 53 B13 BP cuff 2 1 0 □0.NO □1.YES 54 55 B14 Stethoscope 2 1 0 □0.NO □1.YES 56 57 B15 Baby weighing scale 2 1 0 □0.NO □1.YES 58 59 B16 Fetal scope 2 1 0 □0.NO □1.YES 60 B17 Kidney dish 2 1 0 □0.NO □1.YES

17

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 45 of 72 BMJ Open

1 2 3 ID A. Available B. Supply 4 BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 5 Are there enough At least 1 Available NOT 6 for the daily Essential Items available & but NONE Availa 7 caseload of Functional functional ble 8 deliveries? 9 10 B18 Clinical thermometer 2 1 0 □0.NO □1.YES 11 12 B19 Scissors 2 1 0 □0.NO □1.YES 13 Needles and Syringes (10- □0.NO □1.YES 14 B20 2 1 0 15 20cc) 16 Suture needles/suture □0.NO □1.YES 17 B21 For peer2 review1 only0 18 materials 19 B22 IV Drip Stand(s) 2 1 0 □0.NO □1.YES 20 21 B23 Urinary catheters 2 1 0 □0.NO □1.YES 22 23 Uristix/Albustix (dip stick for □0.NO □1.YES B24 2 1 0 24 protein in urine) 25 26 Filled oxygen cylinder with □0.NO □1.YES 27 B25 cylinder carrier and key to 2 1 0 28 open valve 29 30 B26 Mouth gag 2 1 0 □0.NO □1.YES 31 Patient transport (wheelchair, □0.NO □1.YES 32 B27 2 1 0 33 gurney, hammock) 34 Incubator/warm Room □0.NO □1.YES 35 36 B28 2 1 0 37 38

39 http://bmjopen.bmj.com/ 40 Infection prevention (Kuzuia maambukizi) 41 42 B29 Soap 2 1 0 □0.NO □1.YES 43 B30 Antiseptics 2 1 0 □0.NO □1.YES 44 45 B31 Sterile gloves 2 1 0 □0.NO □1.YES 46

47 Non-sterile protective clothing □0.NO □1.YES on October 1, 2021 by guest. Protected copyright. B32 2 1 0 48 (Gauns/Apron) 49 50 B33 Decontamination container 2 1 0 □0.NO □1.YES 51 Bleach or bleaching powder □0.NO □1.YES 52 B34 2 1 0 53 (Jick/Chlorinated lime) 54 B35 Prepared disinfection solution 2 1 0 □0.NO □1.YES 55 56 B36 Covered contaminated trash bin 2 1 0 □0.NO □1.YES 57 58 B37 Puncture proof sharps container 2 1 0 □0.NO □1.YES 59 Mayo stand (or equivalent to □0.NO □1.YES 60 B38 2 1 0 establish sterile field)

18

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 46 of 72

1 2 3 ID A. Available B. Supply 4 BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 5 Are there enough At least 1 Available NOT 6 for the daily Essential Items available & but NONE Availa 7 caseload of Functional functional ble 8 deliveries? 9 10 B39 Sterilizer/autoclave 2 1 0 □0.NO □1.YES 11 12 B40 Placenta pit 2 1 0 □0.NO □1.YES 13 14 B41 Daily caseload of deliveries? 15 B41 □0.NO □1.YES 16 Artery forceps 2 1 0 17 a For peer review only 18 B41 Cord-cutting/blunt-ended □0.NO □1.YES 19 2 1 0 b scissors 20 21 B41 □0.NO □1.YES Cord ties 2 1 0 22 c 23 24 B41 □0.NO □1.YES Gloves 2 1 0 25 d 26 27 B41 □0.NO □1.YES Plastic sheets/ Macking tosh 2 1 0 28 e 29 B41 □0.NO □1.YES 30 Gauze swabs 2 1 0 31 f 32 B41 □0.NO □1.YES 33 Cloth (Green towels) 2 1 0 34 g 35 Perneal/Vaginal/Cervical 36 B42 37 repair pack 38 B42 39 Sponge forceps 2 1 0 □0.NO □1.YES http://bmjopen.bmj.com/ a 40 41 B42 Artery forceps large/small 2 1 0 □0.NO □1.YES 42 b 43 44 B42 □0.NO □1.YES Needle holder 2 1 0 45 c 46

47 B42 □0.NO □1.YES on October 1, 2021 by guest. Protected copyright. Stitch scissors 2 1 0 48 d 49 B42 □0.NO □1.YES 50 Dissecting forceps, toothed 2 1 0 51 e 52 B42 □0.NO □1.YES 53 Vaginal speculum, (Sims) 2 1 0 54 f 55 B42 □0.NO □1.YES 56 Vaginal speculum (Bivalve) 2 1 0 57 g 58 B43 Vacum extractior/ 59 60 B43 □0.NO □1.YES Vacuum extractor 2 1 0 a

19

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 47 of 72 BMJ Open

1 2 3 ID A. Available B. Supply 4 BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 5 Are there enough At least 1 Available NOT 6 for the daily Essential Items available & but NONE Availa 7 caseload of Functional functional ble 8 deliveries? 9 10 B43 □0.NO □1.YES forceps delivery 2 1 0 11 b 12 13 Uterine evacuation 14 B44 15 16 B44 17 Vaginal speculumFor (Sims) peer 2 review1 only0 □0.NO □1.YES 18 a 19 B44 Sponge (ring) forceps or uterine □0.NO □1.YES 20 2 1 0 b packing forceps 21 22 B44 □0.NO □1.YES Single tooth tenaculum forceps 2 1 0 23 c 24 25 B44 □0.NO □1.YES Long dressing forceps 2 1 0 26 c 27 28 B44 Uterine dilators, sizes 13-27 □0.NO □1.YES 2 1 0 29 d (French) 30 31 B44 Sharp uterine curettes, size 0 or □0.NO □1.YES 2 1 0 32 e 00 33 B44 Blunt uterine curettes, size 0 or 34 2 1 0 □0.NO □1.YES 35 f 00 36 B44 □0.NO □1.YES 37 Metal uterine sound 2 1 0 38 g

39 http://bmjopen.bmj.com/ B45 Manual Vacuum evacuation 40 41 □0.NO □1.YES 42 B45 Basic uterine evacuation 43 2 1 0 a instruments (B43a-e & h)PLUS: 44 45 46 B45 Vacuum syringes (single / □0.NO □1.YES 47 2 1 0 on October 1, 2021 by guest. Protected copyright. 48 b double valve) 49 B45 □0.NO □1.YES 50 Silicone lubricant 2 1 0 c 51 52 B45 □0.NO □1.YES Adapters 2 1 0 53 d 54 55 B45 □0.NO □1.YES Flexible cannulae, 4 - 12 mm 2 1 0 56 e 57 58 Kipo japo Kipo lakini Vinatolewa kiasi NOT 59 kimoja hakuna cha kutosha kwa B46 Neonatal resuscitation park Availa 60 kinafanya kinachofan siku? ble kazi ya kazi

20

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 48 of 72

1 2 3 ID A. Available B. Supply 4 BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 5 Are there enough At least 1 Available NOT 6 for the daily Essential Items available & but NONE Availa 7 caseload of Functional functional ble 8 deliveries? 9 10 B46 □0.NO □1.YES Mucus extractor 2 1 0 11 a 12 13 B46 □0.NO □1.YES Infant face mask 2 1 0 14 b 15 16 B46 □0.NO □1.YES Ventilatory bag 2 1 0 17 c For peer review only 18 B46 □0.NO □1.YES 19 Suction catheter Ch 12 2 1 0 20 d 21 B46 □0.NO □1.YES 22 Suction catheter Ch 10 2 1 0 23 e 24 B46 Infant laryngoscope with spare □0.NO □1.YES 25 2 1 0 26 f bulb & batteries 27 B46 □0.NO □1.YES 28 Endotracheal tubes 3.5 2 1 0 g 29 30 B46 □0.NO □1.YES Endotracheal tubes 3.0 2 1 0 31 h 32 33 B46 Suction apparatus: Foot- or □0.NO □1.YES 2 1 0 34 i electrically-operated 35 36 B46 □0.NO □1.YES Oxygen cylinders 2 1 0 37 j 38

39 http://bmjopen.bmj.com/ 40

41 42 Essential Drugs 43 44 Cod If yes, enough supply to last for up Drugs Available 45 e to (Check NUMBER OF WEEKS): 46

47 B47 Antibiotics on October 1, 2021 by guest. Protected copyright. 48 B47 □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 49 Amoxicillin 50 a 4+ 51 B47 □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 52 Ampicillin 53 b 4+ 54 □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 55 B47c Benzyl penicillin (X-Pen) 4+ 56 57 B47 □ <1 □ 1 □ 2 □ 3 □ Cloxacillin □0.NO □1.YES 58 d 4+ 59 60 B47 □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ Erythromicin e 4+

21

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 49 of 72 BMJ Open

1 2 3 Cod If yes, enough supply to last for up

Drugs Available BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 4 e to (Check NUMBER OF WEEKS): 5 6 □ <1 □ 1 □ 2 □ 3 □ B47f Gentamicin □0.NO □1.YES 7 4+ 8 9 B47 □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ Metronidazole (Flagyl) 10 g 4+ 11 12 B47 □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ Nitrofurantoin 13 h 4+ 14 □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 15 B47i Penicillin G 16 4+ 17 For peer review only Procaine penicillin G □ <1 □ 1 □ 2 □ 3 □ 18 B47j □0.NO □1.YES 19 (PPF) 4+ 20 Trimethoprim/Sulfamethoz □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 21 B47k 22 azole (Septrine) 4+ 23 □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ B47l Magnesium sulfate 24 4+ 25 26 B47 □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ Phenytoin 27 m 4+ 28 29 B48 Antihypertensives 30 B48 □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 31 Hydralazine 32 a 4+ 33 B48 □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 34 Nifedipine 35 b 4+ 36 Oxytocics, 37 B49 38 Prostaglandins & other

39 http://bmjopen.bmj.com/ B49 □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 40 Ergometrine a 4+ 41 42 B49 □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ Methylergometrine 43 b 4+ 44 45 □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ B49c Misoprostol 46 4+

47 on October 1, 2021 by guest. Protected copyright. 48 B49 □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ Oxytocin 49 d 4+ 50 B49 □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 51 Prostaglandin E2 52 e 4+ 53 Dawa zinazotumika 54 B50 55 wakati wa dharura 56 B50 □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 57 Adrenaline 58 a 4+ 59 B50 □ <1 □ 1 □ 2 □ 3 □ 60 Aminophylline □0.NO □1.YES b 4+

22

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 50 of 72

1 2 3 Cod If yes, enough supply to last for up

Drugs Available BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 4 e to (Check NUMBER OF WEEKS): 5 6 □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ B50c Atropine sulfate 7 4+ 8 9 B50 □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ Calcium gluconate 10 d 4+ 11 12 B50 □ <1 □ 1 □ 2 □ 3 □ Digoxin □0.NO □1.YES 13 e 4+ 14 □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 15 B50f Ephedrine 16 4+ 17 For peer review only B50 □ <1 □ 1 □ 2 □ 3 □ 18 Frusemide (Lasix) □0.NO □1.YES 19 g 4+ 20 B50 □ <1 □ 1 □ 2 □ 3 □ 21 Naloxone □0.NO □1.YES h 22 4+ 23 □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ B50i Nitroglycerine 24 4+ 25 26 □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ B50j Prednisolone 27 4+ 28 29 Promethazine □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ B50k 30 (Phenergan) 4+ 31 32 B51 Anethetics (Nusu kaputi) 33 B51 □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 34 Halothane 35 a 4+ 36 B51 □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 37 Ketamine 38 b 4+

39 http://bmjopen.bmj.com/ □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 40 B51c Lignocaine 2% or 1% 4+ 41 42 B52 Analgesics 43 44 B52 □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ Morphine 45 a 4+ 46

47 B52 □ <1 □ 1 □ 2 □ 3 □ on October 1, 2021 by guest. Protected copyright. Paracetamol □0.NO □1.YES 48 b 4+ 49 □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 50 B52c Pethidine 51 4+ 52 □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 53 B53 Sedatives 54 4+ 55 B53 □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 56 Diazepam (Valium) a 4+ 57 58 B53 □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ Phenobarbitone 59 b 4+ 60 B54 Tocolytics

23

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 51 of 72 BMJ Open

1 2 3 Cod If yes, enough supply to last for up

Drugs Available BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 4 e to (Check NUMBER OF WEEKS): 5 6 B54 □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ Nifedipine 7 a 4+ 8 9 B54 □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ Salbutamol 10 b 4+ 11 12 B55 Steroids 13 B55 □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 14 Betamethasone 15 a 4+ 16 B55 □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 17 DexamethasoneFor peer review only 18 b 4+ 19 □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 20 B55c Hydrocortisone 4+ 21 22 B56 IV Fluids 23 24 B56 □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ Dextrose 5% 25 a 4+ 26 27 B56 □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ Glucose (20%/50%) 28 b 4+ 29 □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 30 B56c Normal saline 31 4+ 32 B56 □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 33 Ringer’s lactate 34 d 4+ 35 B57 Antimalarial 36 37 B57 □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ ALU 38 a 4+ 39 http://bmjopen.bmj.com/ 40 B57 □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ Artesunate 41 b 4+ 42 43 □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ B57c Quinine 44 4+ 45 46 B57 Sulfadoxine/Pyrimethamin □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □

47 d e (SP) 4+ on October 1, 2021 by guest. Protected copyright. 48 B57 □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 49 Anti-tetanus serum 50 e 4+ 51 □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 52 B57f Ferrous sulfate 53 4+ 54 B57 □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 55 Folic acid g 4+ 56 57 B57 □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ Magnesium trisilicate 58 h 4+ 59 60 □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ B57i Tetanus antitoxin 4+

24

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 52 of 72

1 2 3 Cod If yes, enough supply to last for up

Drugs Available BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 4 e to (Check NUMBER OF WEEKS): 5 6 □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ B57j Tetanus toxoid 7 4+ 8 9 □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ B57k Anti-retrovirals – mother 10 4+ 11 12 □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ B57l Anti-retrovirals – newborn 13 4+ 14 15 B58 PMTCT / HIV care 16 B58 □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 17 ART For peer review only 18 a 4+ 19 B58 □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 20 Niverapine b 4+ 21 22 □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ B58c Rapid testing kit 23 4+ 24 25 B58 Post-HIV exposure □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 26 d prophylactic treatment 4+ 27 28 29 30 31

32 33 34 35 36 37 C. Operating theatre 38

39 Code Available http://bmjopen.bmj.com/ 40 41 C1 Does this facility have an operating theatre? □0.NO □1.YES 42 43 Kama jibu ni NO ruka nenda chumba kingine (NENDA 44 D1) 45 Miund 46 o mbinu

47 on October 1, 2021 by guest. Protected copyright. 48 49 Code Available 50 51 C2 Electricity □0.NO □1.YES 52 53 Sufficient light source to perform tasks during the □0.NO □1.YES C3 54 day 55 C4 Sufficient light source to perform tasks at night □0.NO □1.YES 56 □ □ 57 C5 Running water □0.NO □1.YES 58 59 C6 Means of ventilation (Fan) □0.NO □1.YES 60

25

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 53 of 72 BMJ Open

1 2 3 Vifaa

4 BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 5 Code A. Availability B. Supply 6 7 Are there enough At least 1 Available NOT 8 for the daily Essential Items available & but NONE Avail 9 caseload of Functional functional able 10 deliveries? 11 12 Infection prevention 13 14 C7 Soap 2 1 0 □0.NO □1.YES 15 C8 Antiseptics 2 1 0 □0.NO □1.YES 16 □ □ 17 C9 Sterile glovesFor peer 2review 1 only0 □0.NO □1.YES 18 19 C10 Non-sterile gloves 2 1 0 □0.NO □1.YES 20 21 C11 Non-sterile protective clothing 2 1 0 □0.NO □1.YES 22 23 C12 Decontamination container 2 1 0 □0.NO □1.YES 24 C13 Bleach or bleaching powder 2 1 0 □0.NO □1.YES 25 26 C14 Prepared disinfection solution 2 1 0 □0.NO □1.YES 27 28 C15 Regular trash bin 2 1 0 □0.NO □1.YES 29 30 Covered contaminated waste □0.NO □1.YES C16 2 1 0 31 trash bin 32 33 C17 Puncture proof sharps container 2 1 0 □0.NO □1.YES 34 Mayo stand (or equivalent to □0.NO □1.YES 35 C18 2 1 0 36 establish sterile field) 37 C19 Sterilizer/autoclave 2 1 0 □0.NO □1.YES 38

39 Obstetric laparotomy / http://bmjopen.bmj.com/ C20 40 cesarean section pack 41 42 Stainless steel instrument tray □0.NO □1.YES C20a 2 1 0 43 with cover 44 45 C20b Towel clips 2 1 0 □0.NO □1.YES 46 C20c Sponge forceps 2 1 0 □0.NO □1.YES 47 on October 1, 2021 by guest. Protected copyright. 48 C20d Straight artery forceps 2 1 0 □0.NO □1.YES 49 50 C20e Uterine haemostasis forceps 2 1 0 □0.NO □1.YES 51 52 C20f Needle holder 2 1 0 □0.NO □1.YES 53 54 C20g Surgical knife handle 2 1 0 □0.NO □1.YES 55 56 C20h Surgical knife blades 2 1 0 □0.NO □1.YES 57 Triangular point suture □0.NO □1.YES 58 C20i 2 1 0 needles/7.3 cm/size 6 59 60 C20j Round-bodied needles/No 2 1 0 □0.NO □1.YES

26

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 54 of 72

1 2 3 Code A. Availability B. Supply 4 BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 5 Are there enough At least 1 Available NOT 6 for the daily Essential Items available & but NONE Avail 7 caseload of Functional functional able 8 deliveries? 9 10 12/size 6 11 12 Abdominal retractors/double- □0.NO □1.YES C20k 2 1 0 13 ended (Richardson) 14 Curved operating scissors/blunt □0.NO □1.YES 15 C20l 2 1 0 16 pointed (Mayo)17cm 17 For peer review only Straight operating scissors/blunt □0.NO □1.YES 18 C20m 2 1 0 19 pointed (Mayo)17cm 20 21 C20n Scissors, straight, 23 cm 2 1 0 □0.NO □1.YES 22 C20o Suction nozzle 2 1 0 □0.NO □1.YES 23 24 Suction tube, 22.5 cm, 23 □0.NO □1.YES C20p 2 1 0 25 French gauge 26 27 C20r Intestinal clamps 2 1 0 □0.NO □1.YES 28 Dressing (non-toothed tissue) □0.NO □1.YES 29 C20s 2 1 0 30 forceps 31 Sutures (different sizes and □0.NO □1.YES 32 C20t 2 1 0 33 types) 34 C21 Anesthesia equipment 35 36 C21a Anesthetic face masks 2 1 0 □0.NO □1.YES 37 38 C21b Oropharyngeal airways 2 1 0 □0.NO □1.YES

39 http://bmjopen.bmj.com/ 40 Laryngoscopes (with spare □0.NO □1.YES C21c 2 1 0 41 bulbs and batteries) 42 43 C21d Intubating forceps (Magill) 2 1 0 □0.NO □1.YES 44 Endotracheal tube connectors: □0.NO □1.YES 45 □ □ 46 15 mm plastic (can be C21e connected directly to the 2 1 0 47 on October 1, 2021 by guest. Protected copyright. 48 breathing valve; three for each 49 tube size) 50 Spinal needles (18-gauge to 25- □0.NO □1.YES 51 C21f 2 1 0 52 gauge) 53 Suction apparatus: Foot- □0.NO □1.YES 54 C21g 2 1 0 operated 55 56 Suction apparatus: Electrically □0.NO □1.YES C21h 2 1 0 57 operated 58 59 Anesthesia apparatus □0.NO □1.YES C21i 2 1 0 60 (EMO/draw-over system)

27

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 55 of 72 BMJ Open

1 2 3 Code A. Availability B. Supply 4 BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 5 Are there enough At least 1 Available NOT 6 for the daily Essential Items available & but NONE Avail 7 caseload of Functional functional able 8 deliveries? 9 10 Oxygen cylinders c manometer 11 C21j and flowmeter tubes and 2 1 0 12 connectors/o2 concentrator □0.NO □1.YES 13 14 Perineal / vaginal / cervical C22 15 repair pack 16 17 C22a Sponge forcepsFor peer 2review 1 only0 □0.NO □1.YES 18 19 C22b Artery forceps large/small 2 1 0 □0.NO □1.YES 20 21 C22c Needle holder 2 1 0 □0.NO □1.YES 22 C22d Stitch scissors 2 1 0 □0.NO □1.YES 23 24 C22e Dissecting forceps, toothed 2 1 0 □0.NO □1.YES 25 26 C22f Vaginal speculum, large (Sims) 2 1 0 □0.NO □1.YES 27 28 C23 Uterine evacuation 29 C23a Vaginal speculum (Sims) 2 1 0 □0.NO □1.YES 30 □ □ 31 Sponge (ring) forceps or uterine C23b 2 1 0 □0.NO □1.YES 32 cpacking forceps 33 34 C23c Single tooth tenaculum forceps 2 1 0 □0.NO □1.YES 35 36 C23d Long dressing forceps 2 1 0 □0.NO □1.YES 37 Uterine dilators, sizes 13-27 38 C23e 2 1 0 □0.NO □1.YES 39 (French) http://bmjopen.bmj.com/ 40 41 C23f Sharp/blunt uterine curettes, 2 1 0 □0.NO □1.YES 42 C23g Malleable metal uterine sound 2 1 0 □0.NO □1.YES 43 44 Manual vacuum C23 45 evacuation(MVA) 46

47 Basic uterine evacuation on October 1, 2021 by guest. Protected copyright. C23a 2 1 0 □0.NO □1.YES 48 instruments PLUS: 49 50 Vacuum syringes (single / C23b 2 1 0 □0.NO □1.YES 51 double valve) 52 53 C23c Silicone lubricant 2 1 0 □0.NO □1.YES 54 55 C23d Adapters 2 1 0 □0.NO □1.YES 56 C23e Flexible cannulae, 4 - 12 mm 2 1 0 □0.NO □1.YES 57 58 C24 Neonatal resuscitation pack 59 60 C24a Mucus extractor 2 1 0 □0.NO □1.YES

28

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 56 of 72

1 2 3 Code A. Availability B. Supply 4 BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 5 Are there enough At least 1 Available NOT 6 for the daily Essential Items available & but NONE Avail 7 caseload of Functional functional able 8 deliveries? 9 10 C24b Infant face mask 2 1 0 □0.NO □1.YES 11 12 C24c Ventilatory bag 2 1 0 □0.NO □1.YES 13 14 C24d Suction catheter 2 1 0 □0.NO □1.YES 15 Infant laryngoscope with spare 16 C24e 2 1 0 □0.NO □1.YES 17 bulb& batteriesFor peer review only 18 C24f Endotracheal tubes 2 1 0 □0.NO □1.YES 19 20 C24g Suction apparatus: warmer 2 1 0 □0.NO □1.YES 21 22 Craniotomy equipment for C25 23 destructive operation 24 25 C25a Craniotomy set/kit 2 1 0 □0.NO □1.YES 26 27 28 Essential Drugs 29 30 Cod If yes, enough supply to last for up to Drugs Available 31 e (Check NUMBER OF WEEKS): 32 33 C26 Antibiotics 34 35 C26 □0.NO □1.YES Ampicillin □ <1 □ 1 □ 2 □ 3 □ 4+ 36 a 37 C26 □0.NO □1.YES 38 Benzyl penicillin □ <1 □ 1 □ 2 □ 3 □ 4+ 39 b http://bmjopen.bmj.com/ 40 C26 □0.NO □1.YES 41 Gentamicin □ <1 □ 1 □ 2 □ 3 □ 4+ 42 c 43 C26 □0.NO □1.YES 44 Metronidazole □ <1 □ 1 □ 2 □ 3 □ 4+ d 45 46 C26 □0.NO □1.YES

Penicillin G □ <1 □ 1 □ 2 □ 3 □ 4+ on October 1, 2021 by guest. Protected copyright. 47 e 48 49 C27 Anticonvulsants 50 51 C27 □0.NO □1.YES Magnesium sulfate □ <1 □ 1 □ 2 □ 3 □ 4+ 52 a 53 C27 □0.NO □1.YES 54 Phenytoin □ <1 □ 1 □ 2 □ 3 □ 4+ 55 b 56 57 C28 Antihypertensives 58 C28 □0.NO □1.YES 59 Hydralazine □ <1 □ 1 □ 2 □ 3 □ 4+ a 60 C28 Nifedipine □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+

29

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 57 of 72 BMJ Open

1 2 3 b

4 BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 5 Oxytocics, 6 C29 Prostaglandins & 7 other 8 9 C29 □0.NO □1.YES Ergometrine □ <1 □ 1 □ 2 □ 3 □ 4+ 10 a 11 12 C29 □0.NO □1.YES Misoprostol □ <1 □ 1 □ 2 □ 3 □ 4+ 13 b 14 C29 □0.NO □1.YES 15 Oxytocin □ <1 □ 1 □ 2 □ 3 □ 4+ 16 c 17 For peer review only Drugs used in 18 C30 19 Emergencies 20 C30 □0.NO □1.YES 21 Adrenaline □ <1 □ 1 □ 2 □ 3 □ 4+ 22 a 23 C30 □0.NO □1.YES Aminophylline □ <1 □ 1 □ 2 □ 3 □ 4+ 24 b 25 26 C30 □0.NO □1.YES Atropine sulfate □ <1 □ 1 □ 2 □ 3 □ 4+ 27 c 28 29 C30 □0.NO □1.YES Calcium gluconate □ <1 □ 1 □ 2 □ 3 □ 4+ 30 d 31 32 C30 □0.NO □1.YES Digoxin □ <1 □ 1 □ 2 □ 3 □ 4+ 33 e 34 35 C30f Ephedrine □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 36 C30 □0.NO □1.YES 37 Frusemide □ <1 □ 1 □ 2 □ 3 □ 4+ 38 g

39 http://bmjopen.bmj.com/ C30 □0.NO □1.YES 40 Naloxone □ <1 □ 1 □ 2 □ 3 □ 4+ h 41 42 C30i Nitroglycerine □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 43 44 C30j Prednisone □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 45 46 C30 □0.NO □1.YES Promethazine □ <1 □ 1 □ 2 □ 3 □ 4+

47 k on October 1, 2021 by guest. Protected copyright. 48 49 C31 Anesthetics 50 C31 □0.NO □1.YES 51 Halothane □ <1 □ 1 □ 2 □ 3 □ 4+ 52 a 53 C31 □0.NO □1.YES 54 Ketamine □ <1 □ 1 □ 2 □ 3 □ 4+ b 55 56 C31 □0.NO □1.YES Lignocaine 2% or 1% □ <1 □ 1 □ 2 □ 3 □ 4+ 57 c 58 59 C32 Analgesics 60 C32 Morphine □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+

30

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 58 of 72

1 2 3 a

4 BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 5 C32 □0.NO □1.YES Paracetamol □ <1 □ 1 □ 2 □ 3 □ 4+ 6 b 7 8 C32 □0.NO □1.YES Pethidine □ <1 □ 1 □ 2 □ 3 □ 4+ 9 c 10 11 C33 Sedatives 12 C33 □0.NO □1.YES 13 Diazepam □ <1 □ 1 □ 2 □ 3 □ 4+ 14 a 15 C33 □0.NO □1.YES 16 Phenobarbitone □ <1 □ 1 □ 2 □ 3 □ 4+ 17 b For peer review only 18 19 C34 Tocolytics 20 21 22 C34 □0.NO □1.YES Nifedipine □ <1 □ 1 □ 2 □ 3 □ 4+ 23 a 24 C34 □0.NO □1.YES 25 Salbutamol □ <1 □ 1 □ 2 □ 3 □ 4+ 26 b 27 C34 □0.NO □1.YES 28 Steroids □ <1 □ 1 □ 2 □ 3 □ 4+ 29 c 30 C34 □0.NO □1.YES 31 Betamethasone □ <1 □ 1 □ 2 □ 3 □ 4+ d 32 33 C34 □0.NO □1.YES Dexamethasone □ <1 □ 1 □ 2 □ 3 □ 4+ 34 e 35 36 C34f Hydrocortisone □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 37 38 C35 IV Fluids

39 http://bmjopen.bmj.com/ C35 □0.NO □1.YES 40 Dextrose □ <1 □ 1 □ 2 □ 3 □ 4+ 41 a 42 C35 □0.NO □1.YES 43 Normal saline □ <1 □ 1 □ 2 □ 3 □ 4+ 44 b 45 C35 □0.NO □1.YES 46 Ringer’s lactate □ <1 □ 1 □ 2 □ 3 □ 4+ c

47 on October 1, 2021 by guest. Protected copyright. 48 C36 Antimalarial 49 50 C36 □0.NO □1.YES Artemether □ <1 □ 1 □ 2 □ 3 □ 4+ 51 a 52 53 C36 □0.NO □1.YES Artesunate □ <1 □ 1 □ 2 □ 3 □ 4+ 54 b 55 C36 □0.NO □1.YES 56 Coartem (ALU) □ <1 □ 1 □ 2 □ 3 □ 4+ 57 c 58 C36 □0.NO □1.YES 59 Quinine dihydrochloride □ <1 □ 1 □ 2 □ 3 □ 4+ 60 d

31

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 59 of 72 BMJ Open

1 2 3 C36 □0.NO □1.YES

4 Quinine sulfate □ <1 □ 1 □ 2 □ 3 □ 4+ BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from e 5 6 C37 Other 7 8 C37 □0.NO □1.YES Anti-tetanus serum □ <1 □ 1 □ 2 □ 3 □ 4+ 9 a 10 11 C37 □0.NO □1.YES Magnesium trisilicate □ <1 □ 1 □ 2 □ 3 □ 4+ 12 b 13 C37 □0.NO □1.YES 14 Tetanus toxoid □ <1 □ 1 □ 2 □ 3 □ 4+ 15 c 16 C37 □0.NO □1.YES 17 Anti-retroviralsFor – Mother peer review□ <1 □only 1 □ 2 □ 3 □ 4+ 18 d 19 C37 Anti-retrovirals - □0.NO □1.YES 20 □ <1 □ 1 □ 2 □ 3 □ 4+ e Newborn 21 22 C37f HIV rapid testing kit □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 23 24 C37 Post-HIV exposure □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ 4+ 25 g prophylactic treatment 26 27 28 29 D. Obstetric /Maternity ward 30

31 ID Available 32 33 D1 Does this facility have an obstetric / maternity ward? □0.NO □1.YES 34 35 Kama jibu ni NO, ruka nenda chumba kingine(NENDA 36 E1) 37

38

39 http://bmjopen.bmj.com/ 40 41 Infrastructure 42 43 ID Miundo mbinu kwa ujumla Available 44 45 D2 Electricity □0.NO □1.YES 46

47 D3 Sufficient light source to perform tasks during the day □0.NO □1.YES on October 1, 2021 by guest. Protected copyright. 48 49 D4 Means of ventilation □0.NO □1.YES 50 51 D5 Running water □0.NO □1.YES 52 D6 Fan / air conditioning (if applicable) □0.NO □1.YES 53 54 D7 Curtains/means of providing patient privacy □0.NO □1.YES 55 56 D8 Waiting area for visitors (iliyo na viti na kivuli) □0.NO □1.YES 57 58 59 60 Vifaa

32

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 60 of 72

1 2 3 ID A. Availability B. Supply 4 BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 5 Availabl Are there enough 6 At least 1 e but NOT for the daily 7 Essential Items available & NONE Availabl caseload of 8 Functional function e 9 deliveries? al 10 11 D9 Beds 2 1 0 □0.NO □1.YES 12 13 D10 Linens 2 1 0 □0.NO □1.YES 14 15 D11 Blankets for cold weather 2 1 0 □0.NO □1.YES 16 17 D12 BP cuff For peer 2review 1 only0 □0.NO □1.YES 18 D13 Stethoscope 2 1 0 □0.NO □1.YES 19 20 D14 Baby weighing scale 2 1 0 □0.NO □1.YES 21 22 D15 Fetal stethoscope 2 1 0 □0.NO □1.YES 23 24 D16 Sponge bowls 2 1 0 □0.NO □1.YES 25 26 D17 Clinical oral thermometer 2 1 0 □0.NO □1.YES 27 D18 Clinical oral thermometer 2 1 0 □0.NO □1.YES 28 29 D19 Scissors 2 1 0 □0.NO □1.YES 30 31 D20 Low reading thermometer 2 1 0 □0.NO □1.YES 32 Surgeon’s handbrush w/ white □0.NO □1.YES 33 D21 2 1 0 34 nylon bristles 35 36 D22 Needles and Syringes (10-20cc) 2 1 0 □0.NO □1.YES 37 D23 Suture needles/suture materials 2 1 0 □0.NO □1.YES 38 39 D24 IV Stand(s) 2 1 0 □0.NO □1.YES http://bmjopen.bmj.com/ 40 41 Filled oxygen cylinder with cylinder □0.NO □1.YES 42 D25 carrier and key to open valve/ 02 2 1 0 43 concentrator 44 45 D26 Adult ventilator bag and mask 2 1 0 □0.NO □1.YES 46 D27 Mouth gag 2 1 0 □0.NO □1.YES 47 on October 1, 2021 by guest. Protected copyright. 48 Patient transport (wheelchair, □0.NO □1.YES 49 D28 2 1 0 50 trolley, hammock) 51 Availabl 52 Are there enough At least 1 e but NOT 53 for the daily Infection and Prevention available & NONE Availabl 54 caseload of Functional function e 55 deliveries? 56 al 57 D29 Soap 2 1 0 □0.NO □1.YES 58 59 D30 Antiseptics 2 1 0 □0.NO □1.YES 60 D31 Sterile gloves 2 1 0 □0.NO □1.YES

33

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 61 of 72 BMJ Open

1 2 3 ID A. Availability B. Supply 4 BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 5 D32 Non-sterile gloves 2 1 0 □0.NO □1.YES 6 7 D33 Decontamination container 2 1 0 □0.NO □1.YES 8 9 D34 Bleach or bleaching powder 2 1 0 □0.NO □1.YES 10 11 D35 Prepared disinfection solution 2 1 0 □0.NO □1.YES 12 Covered contaminated waste trash □0.NO □1.YES 13 D36 2 1 0 bin 14 15 D37 Puncture proof sharps container 2 1 0 □0.NO □1.YES 16 17 For peer review only 18 19 ID Instructions Jibu 20 21 D38 Is food provided by the hospital to patients? □0.NO □1.YES 22 D39 Are there empty beds for the next patients? □0.NO □1.YES 23 24 D40 If yes, are the empty beds clean and ready to receive new □0.NO □1.YES 25 patients? 26 27 28 29 Essential Drugs 30 31 ID Drugs Available 32 D41 Antibiotics 33 34 D41a Amoxicillin □0.NO □1.YES 35 36 D42 Ampicillin 37 38 D42a Benzathine penicillin □0.NO □1.YES

39 http://bmjopen.bmj.com/ 40 D42b Benzyl penicillin □0.NO □1.YES 41 D43c Cloxacillin □0.NO □1.YES 42 43 D43d Erythromicin □0.NO □1.YES 44 45 D43e Gentamicin □0.NO □1.YES 46 D43f Metronidazole □0.NO □1.YES 47 on October 1, 2021 by guest. Protected copyright. 48 D43g Nitrofurantoin □0.NO □1.YES 49 □ □ 50 D43h Penicillin G □0.NO □1.YES 51 52 D43i Procaine penicillin G □0.NO □1.YES 53 54 D43j Trimethoprim/Sulfamethozazole □0.NO □1.YES 55 56 D44 Anticonvulsants □0.NO □1.YES 57 D44a Magnesium sulfate □0.NO □1.YES 58 59 D44b Phenytoin □0.NO □1.YES 60 D45 Antihypertensives □0.NO □1.YES

34

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 62 of 72

1 2 3 ID Drugs Available

4 BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 5 D45a Hydralazine □0.NO □1.YES 6 7 D45b Nifedipine □0.NO □1.YES 8 9 D46 Drugs used in Emergencies □0.NO □1.YES 10 D46a Adrenaline □0.NO □1.YES 11 12 D46b Aminophylline □0.NO □1.YES 13 14 D46c Atropine sulfate □0.NO □1.YES 15 16 D46d Calcium gluconate □0.NO □1.YES 17 For peer review only 18 D46e Digoxin □0.NO □1.YES 19 D46f Ephedrine □0.NO □1.YES 20 21 D46g Frusemide □0.NO □1.YES 22 23 D46h Naloxone □0.NO □1.YES 24 25 D46i Nitroglycerine □0.NO □1.YES 26 27 D46j Prednisone □0.NO □1.YES 28 D46k Prednisolone □0.NO □1.YES 29 30 D46l Promethazine □0.NO □1.YES 31 32 D47 Analgesics 33 34 D47a Morphine □0.NO □1.YES 35 36 D47b Paracetamol □0.NO □1.YES 37 D47c Pethidine □0.NO □1.YES 38 39 D48 Sedatives http://bmjopen.bmj.com/ 40 41 D48a Diazepam □0.NO □1.YES 42 43 D48b Phenobarbitone □0.NO □1.YES 44 45 D49 IV Fluids 46 D49a Dextrose □0.NO □1.YES

47 on October 1, 2021 by guest. Protected copyright. 48 D49b Glucose □0.NO □1.YES 49 50 D49c Normal saline □0.NO □1.YES 51 52 D49d Ringer’s lactate □0.NO □1.YES 53

54 55 E. Laboratory 56 57 58 ID Available 59 60 E1 Does this facility have a laboratory? □0.NO □1.YES

35

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 63 of 72 BMJ Open

1 2 3

4 BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 5 Kama jibu ni NO, ruka nenda chumba kinachofuata (Nenda 6 F1) 7 8

9 10 11 12 13 14 Infrastructure 15 16 ID Infrastructure kwa ujumla Available 17 For peer review only 18 E2 Electricity □0.NO □1.YES 19 20 E3 Chanzo cha mwanga kinachotosheleza kufanya kazi mchana □0.NO □1.YES 21 22 E4 Njia za kuingizia hewa (ex. ceiling fan?) □0.NO □1.YES 23 E5 Maji ya bomba □0.NO □1.YES 24 □ □ 25 26 27 Vifaa 28 29 ID A. Availability B. Supply 30 At least 1 Available 31 Are there enough Provision of donor blood for available but NOT 32 for the daily 33 transfusion & NONE AVAILABL caseload of 34 Function functiona E deliveries? 35 al l 36 37 E6 Jokofu 2 1 0 □0.NO □1.YES 38

39 E7 Test tubes - various sizes 2 1 0 □0.NO □1.YES http://bmjopen.bmj.com/ 40 E8 Slides (microscope) 2 1 0 □0.NO □1.YES 41 42 E9 Compound microscope 2 1 0 □0.NO □1.YES 43 44 E10 Microscope illuminator 2 1 0 □0.NO □1.YES 45 46 E11 Blood lancets 2 1 0 □0.NO □1.YES

47 on October 1, 2021 by guest. Protected copyright. 48 E12 Cotton wool 2 1 0 □0.NO □1.YES 49 E13 Rack 2 1 0 □0.NO □1.YES 50 51 E14 8.5 g/l Sodium Chloride solution 2 1 0 □0.NO □1.YES 52 53 E15 20% Bovine albumin 2 1 0 □0.NO □1.YES 54 55 E16 Centrifuge 2 1 0 □0.NO □1.YES 56 E17 Blood typing and cross-marching □0.NO □1.YES 57 2 1 0 58 reagents 59 E18 Blood collection bags 2 1 0 □0.NO □1.YES 60 E19 Artery forceps 2 1 0 □0.NO □1.YES

36

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 64 of 72

1 2 3 ID A. Availability B. Supply 4 BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 5 At least 1 Available Are there enough 6 Provision of donor blood for available but NOT for the daily 7 transfusion & NONE AVAILABL caseload of 8 Function functiona E 9 deliveries? al l 10 11 E20 Pilot bottles (containing 1 ml □0.NO □1.YES 2 1 0 12 ACD solution) 13 14 E21 Hepatitis Test 2 1 0 □0.NO □1.YES 15 16 E22 HIV Test 2 1 0 □0.NO □1.YES 17 For peer review only 18 E23 Syphilis Test 2 1 0 □0.NO □1.YES 19 Laboratory supplies 20 21 E24 Microscope 2 1 0 □0.NO □1.YES 22 23 E25 Immersion oil 2 1 0 □0.NO □1.YES 24 25 E26 Glass rods 2 1 0 □0.NO □1.YES 26 27 E27 Sink or staining tank 2 1 0 □0.NO □1.YES 28 E28 Measuring cylinder (10-50 ml) 2 1 0 □0.NO □1.YES 29 30 E29 Wash bottle containing buffered □0.NO □1.YES 2 1 0 31 water 32 33 E30 Interval timer clock 2 1 0 □0.NO □1.YES 34 35 E31 Rack for drying slides 2 1 0 □0.NO □1.YES 36 37 E32 Leishman stain 2 1 0 □0.NO □1.YES 38 E33 Methanol 2 1 0 □0.NO □1.YES 39 http://bmjopen.bmj.com/ 40 E34 Refrigerator 2 1 0 □0.NO □1.YES 41 42 E35 Field stains A and B 2 1 0 □0.NO □1.YES 43 44 E36 Glass containers 2 1 0 □0.NO □1.YES 45 46 E37 Counting chamber (Neubauer) 2 1 0 □0.NO □1.YES

47 on October 1, 2021 by guest. Protected copyright. E38 Pipette (various sizes) 2 1 0 □0.NO □1.YES 48 49 E39 Tork diluting solution 2 1 0 □0.NO □1.YES 50 51 E40 Tally counter, differential if □0.NO □1.YES 2 1 0 52 possible 53 54 E41 Haemoglobinometer 2 1 0 □0.NO □1.YES 55 56 E42 Hydrochloric acid solution 2 1 0 □0.NO □1.YES 57 E43 Microhaematocrit centrifuge □0.NO □1.YES 58 2 1 0 (manual or electric) 59 60 E44 Scale for reading results 2 1 0 □0.NO □1.YES

37

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 65 of 72 BMJ Open

1 2 3 ID A. Availability B. Supply 4 BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 5 At least 1 Available Are there enough 6 Provision of donor blood for available but NOT for the daily 7 transfusion & NONE AVAILABL caseload of 8 Function functiona E 9 deliveries? al l 10 11 E45 Heparinized capillary tubes (75 □0.NO □1.YES 2 1 0 12 mm x 1.5 mm) 13 14 E46 Spirit lamp 2 1 0 □0.NO □1.YES 15 16 E47 Ethanol 2 1 0 □0.NO □1.YES 17 For peer review only 18 Laboratory Supplies 2 1 0 □0.NO □1.YES 19 E48 Indicator papers and tablets 2 1 0 □0.NO □1.YES 20 21 E49 Benedict solution 2 1 0 □0.NO □1.YES 22 23 E50 Test-tube holder 2 1 0 □0.NO □1.YES 24 25 E51 Beakers (various sizes) 2 1 0 □0.NO □1.YES 26 27 E52 Spirit lamp 2 1 0 □0.NO □1.YES 28 E53 Sodium nitroprusside 2 1 0 □0.NO □1.YES 29 30 E54 Glacial acetic acid 2 1 0 □0.NO □1.YES 31 32 E55 Ammonia 2 1 0 □0.NO □1.YES 33 Sulfosalicyclic acid (300 g/I □0.NO □1.YES 34 E56 2 1 0 35 aqueous solution) 36 37 E57 Lugol’s iodine solution 2 1 0 □0.NO □1.YES 38 E58 Ehrlich reagent 2 1 0 □0.NO □1.YES 39 http://bmjopen.bmj.com/ 40 E59 Uristix (dip stick for protein in □0.NO □1.YES 2 1 0 41 urine) 42 43 Infection Prevention 2 1 0 □0.NO □1.YES 44 45 E60 Soap 2 1 0 □0.NO □1.YES 46 E61 Antiseptics 2 1 0 □0.NO □1.YES 47 on October 1, 2021 by guest. Protected copyright. 48 E62 Sterile gloves 2 1 0 □0.NO □1.YES 49 □ □ 50 E63 Non-sterile gloves 2 1 0 □0.NO □1.YES 51 52 E64 Decontamination container 2 1 0 □0.NO □1.YES 53 54 E65 Bleach or bleaching powder 2 1 0 □0.NO □1.YES 55 56 E66 Prepared disinfection solution 2 1 0 □0.NO □1.YES 57 E67 Regular trash bin 2 1 0 □0.NO □1.YES 58 59 E68 Covered contaminated waste □0.NO □1.YES 2 1 0 60 trash bin

38

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 66 of 72

1 2 3 ID A. Availability B. Supply 4 BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 5 At least 1 Available Are there enough 6 Provision of donor blood for available but NOT for the daily 7 transfusion & NONE AVAILABL caseload of 8 Function functiona E 9 deliveries? al l 10 11 E69 Puncture proof sharps container 2 1 0 □0.NO □1.YES 12 13 14 15 F: Pharmacy 16

17 Code For peer review onlyAvailable 18 Ess 19 F1 Does this facility have a pharmacy? □0.NO □1.YES enti 20 al 21 Dru 22 23 F2 Is drug inventory register up to date? □0.NO □1.YES gs 24 F3 Are records on supply requests from wards up to date? □0.NO □1.YES 25 26 F4 Is ‘First-in-First-out’ system for supply management used? □0.NO □1.YES 27 28 F5 Is there a regularly used mechanism to ensure that expired drugs are □0.NO □1.YES 29 not distributed? 30 31 F6 Are drugs protected from moisture, heat or infestation (e.g., placed on □0.NO □1.YES 32 shelves or slats, ventilated)? 33 34 F7 Does it have a buffer stock? □0.NO □1.YES 35 36 F8 Is there a minimum stock level for ordering new drugs? □0.NO □1.YES 37 F9 Do you receive what you order (Accuracy)? □0.NO □1.YES 38 39 F10 Time taken from ordering till receiving the supply □0.---Wiki □1.----- http://bmjopen.bmj.com/ 40 Miezi 41 42 If yes, enough supply to last for up ID Drugs Available 43 to (Check NUMBER OF WEEKS): 44 45 F11 Antibiotics 46 □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 □ □ □ □ □ □ on October 1, 2021 by guest. Protected copyright. 47 F11a Amoxicillin 48 □ 4+ 49 □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 50 F11b Ampicillin 51 □ 4+ 52 □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 53 F11c Benzathine penicillin 54 □ 4+ 55 □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 56 F11d Benzyl penicillin □ 4+ 57 58 □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 F11e Ceftriaxone 59 □ 4+ 60 F11f Cloxacillin □0.NO □1.YES □ <1 □ 1 □ 2 □ 3

39

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 67 of 72 BMJ Open

1 2 3 If yes, enough supply to last for up

ID Drugs Available BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 4 to (Check NUMBER OF WEEKS): 5 6 □ 4+ 7 8 □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 F11g Erythromicin 9 □ 4+ 10 F11i □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 11 Gentamicin 12 h □ 4+ 13 □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 14 F11i Kanamycin 15 □ 4+ 16 □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 17 F11j MetronidazoleFor peer review only 18 □ 4+ 19 □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 20 F11k Nitrofurantoin □ 4+ 21 22 □ <1 □ 1 □ 2 □ 3 F12 Penicillin G □0.NO □1.YES 23 □ 4+ 24 25 □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 F12a Procaine penicillin G 26 □ 4+ 27 28 Trimethoprim / □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 F12b 29 Sulfamethozazole □ 4+ 30 31 F13 Anticonvulsants 32 □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 33 F13a Magnesium sulfate 34 □ 4+ 35 □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 36 F13b Phenytoin 37 □ 4+ 38 F14 Antihypertensives 39 http://bmjopen.bmj.com/ 40 □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 F14a Hydralazine 41 □ 4+ 42 43 □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 F14b Aldomet/Metheldopa 44 □ 4+ 45 □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 46 F14c Nifedipine

47 □ 4+ on October 1, 2021 by guest. Protected copyright. 48 Oxytocics, Prostaglandins & 49 F15 50 other 51 □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 52 F15a Ergometrine 53 □ 4+ 54 □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 55 F15b Methylergometrine □ 4+ 56 57 □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 F15c Misoprostol 58 □ 4+ 59 60 □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 F15d Oxytocin □ 4+

40

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 68 of 72

1 2 3 If yes, enough supply to last for up

ID Drugs Available BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 4 to (Check NUMBER OF WEEKS): 5 6 □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 F15e Prostaglandin E2 7 □ 4+ 8 9 □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 F16 Drugs used in Emergencies 10 □ 4+ 11 12 □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 F16a Adrenaline 13 □ 4+ 14 □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 15 F16b Aminophylline 16 □ 4+ 17 For peer review only □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 18 F16c Atropine sulfate 19 □ 4+ 20 □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 21 F16d Calcium gluconate 22 □ 4+ 23 □ <1 □ 1 □ 2 □ 3 24 F16e Digoxin □0.NO □1.YES □ 4+ 25 26 □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 F16f Ephedrine 27 □ 4+ 28 29 □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 F16g Frusemide 30 □ 4+ 31 32 F16 □ <1 □ 1 □ 2 □ 3 Naloxone □0.NO □1.YES 33 h □ 4+ 34 □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 35 F16i Nitroglycerine 36 □ 4+ 37 □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 38 F16j Prednisone

39 □ 4+ http://bmjopen.bmj.com/ 40 □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 41 F16k Prednisolone □ 4+ 42 □ 43 □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 44 F16l Promethazine □ 4+ 45 46 F17 Anaethetics

47 on October 1, 2021 by guest. Protected copyright. 48 F17a □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 Halothane 49 □ 4+ 50 F17b □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 51 Ketamine 52 □ 4+ 53 F17c □ <1 □ 1 □ 2 □ 3 54 Lignocaine 2% or 1% □0.NO □1.YES 55 □ 4+ 56 F18 Analgesics 57 58 F18a □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 59 Morphine □ 4+ 60 F18b Paracetamol □0.NO □1.YES □ <1 □ 1 □ 2 □ 3

41

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 69 of 72 BMJ Open

1 2 3 If yes, enough supply to last for up

ID Drugs Available BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 4 to (Check NUMBER OF WEEKS): 5 6 □ 4+ 7 8 □ <1 □ 1 □ 2 □ 3 F18c Pethidine □0.NO □1.YES 9 □ 4+ 10 F18d □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 11 Sedatives 12 □ 4+ 13 F18e □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 14 Diazepam 15 □ 4+ 16 F18f □ <1 □ 1 □ 2 □ 3 17 PhenobarbitoneFor peer□ 0.NOreview □1.YES only 18 □ 4+ 19 F19 Tocolytics 20 21 F19a □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 Nifedipine 22 □ 4+ 23 24 F19b □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 Salbutamol 25 □ 4+ 26 27 F20 Steroids 28 □ <1 □ 1 □ 2 □ 3 29 F20a Betamethasone □0.NO □1.YES 30 □ 4+ 31 □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 32 F20b Dexamethasone 33 □ 4+ 34 □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 35 F20c Hydrocortisone □ 4+ 36 37 F21 IV Fluids 38

39 F21a □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 http://bmjopen.bmj.com/ Dextrose 40 □ 4+ 41 42 F21b □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 Glucose 43 □ 4+ 44 F21c □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 45 Normal saline 46 □ 4+

47 on October 1, 2021 by guest. Protected copyright. F21d □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 48 Ringer’s lactate 49 □ 4+ 50 F22 Antimalarial 51 52 □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 F22a ALU 53 □ 4+ 54 55 □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 F22b Artesunate 56 □ 4+ 57 58 □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 F22c Quinine dihydrochloride 59 □ 4+ 60 F22d Quinine sulfate □0.NO □1.YES □ <1 □ 1 □ 2 □ 3

42

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 70 of 72

1 2 3 If yes, enough supply to last for up

ID Drugs Available BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 4 to (Check NUMBER OF WEEKS): 5 6 □ 4+ 7 8 □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 F22e Sulfadoxine/Pyrimethamine 9 □ 4+ 10 □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 11 F23 Other 12 □ 4+ 13 □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 14 F23a Anti-tetanus serum 15 □ 4+ 16 □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 17 F23b Ferrous sulfateFor peer review only 18 □ 4+ 19 □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 20 F23c Folic acid □ 4+ 21 22 □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 F23d Magnesium trisilicate 23 □ 4+ 24 25 □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 F23e Sodium citrate 26 □ 4+ 27 28 □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 F23f Tetanus antitoxin 29 □ 4+ 30 □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 31 F23g Tetanus toxoid 32 □ 4+ 33 □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 34 F23h Anti-retrovirals – Mothers 35 □ 4+ 36 □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 37 F23i Anti-retrovirals – Newborn 38 □ 4+

39 http://bmjopen.bmj.com/ □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 40 F23j HIV rapid testing kits □ 4+ 41 42 Post-HIV exposure prophylactic □0.NO □1.YES □ <1 □ 1 □ 2 □ 3 F23k 43 treatment □ 4+ 44 45 46

47 on October 1, 2021 by guest. Protected copyright. 48 49 G. Autoclave room 50 51 Code Available 52 Infr 53 G1 Does this facility have an autoclave room? □ 0.No □ 1. Yes ast 54 ruc Kama jibu ni NO, ruka nenda chumba (Nenda MWISHO) 55 tur 56 e 57 58 59 Code Available 60 G2 Electricity □0.NO □1.YES

43

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 71 of 72 BMJ Open

1 2 3 G3 Sufficient light source to perform tasks during the day □0.NO □1.YES 4 BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 5 G4 Sufficient light source to perform tasks at night □0.NO □1.YES 6 7 G5 Means of ventilation □0.NO □1.YES 8 9 Running water 10 11 12 Equipments 13 14 ID A. Availability B. Usambazwaji 15 16 At least 1 Are there enough Available NOT 17 For peeravailable review only for the daily Infection prevention but NONE Availab 18 & caseload of functional le 19 Functional deliveries? 20 21 Autoclave (with temperature □0.NO □1.YES G6 2 1 0 22 and pressure gauges) 23 24 G7 Soap 2 1 0 □0.NO □1.YES 25 26 G8 Antiseptics 2 1 0 □0.NO □1.YES 27 G9 Sterile gloves 2 1 0 □0.NO □1.YES 28 □ □ 29 G10 Non-sterile gloves 2 1 0 □0.NO □1.YES 30 31 Non-sterile protective □0.NO □1.YES G11 2 1 0 32 clothing 33 34 G12 Decontamination container 2 1 0 □0.NO □1.YES 35 36 G13 Bleach or bleaching powder 2 1 0 □0.NO □1.YES 37 Prepared disinfection 2 1 0 □0.NO □1.YES 38 G14

39 solution http://bmjopen.bmj.com/ 40 G15 Regular trash bin 2 1 0 □0.NO □1.YES 41 42 Covered contaminated waste 2 1 0 □0.NO □1.YES G16 43 trash bin 44 45 Puncture proof sharps 2 1 0 □0.NO □1.YES G17 46 container

47 on October 1, 2021 by guest. Protected copyright. 48 Mayo stand (or equivalent to 2 1 0 □0.NO □1.YES G18 49 establish sterile field) 50 51 52

53 54 55 56 57 58 59

60

44

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 72 of 72

1 2 3 REFERRAL

4 BMJ Open: first published as 10.1136/bmjopen-2015-008999 on 12 February 2016. Downloaded from 5 6 7 CODE ITEM Availability 8 9 R Not Available and Available 10 available functional but not 11 functional 12 13 R1 Land Telephones 0 1 2 14 R2 Mobile phones 0 1 2 15 16 R3 Radio communication set with repeater 0 1 2 17 stationFor peer review only 18 19 R4 Motor vehicle ambulance 0 1 2 20 21 R5 Ox carte 0 1 2 22 23 R6 Motor vehicle 0 1 2 24 R7 Motorcycle 0 1 2 25 26 R8 Bicycle 0 1 2 27 28 R9 Boat 0 1 2 29 30 R10 Who provides fuel for the 31 ambulance(s)? 32 33 34 35 36 Finish Time: Hour__/___ Minutes___/____ 37 38

39 http://bmjopen.bmj.com/ 40 41 42 43 End 44 45 46

47 on October 1, 2021 by guest. Protected copyright. 48 49 50 51 52 53 54 55 56 57 58 59 60

45

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml