Advancing Implementation in Maternal and Newborn Health: Two Decades of Experience

International Journal of Gynecology and Obstetrics 130 (2015) S1–S2

Contents lists available at ScienceDirect

International Journal of Gynecology and Obstetrics

journal homepage: www.elsevier.com/locate/ijgo

EDITORIAL Advancing implementation in maternal and newborn health: Two decades of experience

In the past 15 years, Jhpiego has had the privilege of working with who work in MNH must truly embrace task-shifting and community in- governments and other partners to implement interventions for mater- terventions, especially those delivered by community health workers, to nal and newborn health (MNH) in nearly 40 countries. This Supplement reach the unreachable and end preventable deaths in the last mile. to the International Journal of Gynecology and Obstetrics (IJGO) is our at- MNH champions, including the readers of IJGO, play an essential role tempt to share our analysis and learning from those experiences as the not only in modeling and echoing technical excellence but also in promot- global health community looks toward 2030 and beyond, and to ensure ing the responsible task-shifting and community-based delivery that is that implementation challenges, as well as the resources required to ad- required to facilitate sustainable impact at scale. dress them, are an important part of the post-Millennium Development The third and final section explores specific interventions that Goals conversation. have or should be considered for scale: core examples of the The new global paradigm for MNH envisioned by the UN Secretary “what” of MNH implementation. Jhpiego’s experiences in setting up General’s Sustainable Development Goals and the Global Strategy for programs for emergency obstetric and newborn care, postpartum Women’s, Children’s, and Adolescents’ Health aims to bring evidence- family planning, malaria in pregnancy, and prevention of maternal- based MNH interventions to national scale, setting ambitious targets to-child transmission of HIV show the wide range of implementation to reach the unreached in every country and end all preventable deaths factors and the reality that there is no single successful implementa- among women and children. Scale is not only about what works; it is tion strategy for all interventions. MNH interventions will go farther about how to make that which has been proven to work in small, con- with a focus on pre-service education, engaging learners as early as trolled areas apply equally well across all contexts. So we must ask our- possible. selves: What does it take for health systems to reach every woman, Going forward, we need to harness the power of unlike minds and every child, everywhere, every time? those of our beneficiaries to approach implementation with the kind The answer to this classic implementation science question is diffi- of design thinking and crowd-sourced inputs that has benefitted the cult to quantify, but we do have some evidence, and as with any scien- fields of business and engineering. MNH is at a critical crossroads as tific inquiry, we should start with the evidence that we have. The papers the era of the Millennium Development Goals ends and new goals are in the Supplement have been written by and for implementers, describ- set. It is time to set goals that are more precise than “skilled care at ing in detail what has been accomplished and highlighting lessons birth” so we can hold ourselves accountable to coverage at scale, such about what did and did not work. The lessons themselves will not be as use of uterotonics and immediate postpartum contraception. If we new to anyone who has worked in MNH as long as we have, but we be- are committed to eliminating preventable deaths among women and lieve that their thoughtful analysis and collective and inductive presen- children, we must convince nations to invest in strengthening systems tation is a unique illustration of the complexity of achieving—and then and in improving quality of care for mothers and newborns, and achieve reinforcing—implementation results. Implementation science for MNH quality at scale. Just as we found the resources to reach high immuniza- has a long road ahead. tion rates, we must find the resources and political will to implement The papers in the first section directly confront the “how” of im- other MNH interventions and measure that implementation to achieve plementation. Even if we have all the right interventions, we need impact at scale. In doing so, we will not only achieve greater maternal political commitment and partnerships to bring those interventions and newborn survival but start to reduce the morbidities that will re- to populations in need. Sometimes these elements are beyond the main an obstacle far beyond 2030. If we learn to implement better, control of implementers owing to contextual factors and prioritiza- women and children will receive more effective care and be more likely tions inherent in real-world public health environments, but there to return to their families safe and healthy. are common principles that can help navigate politics and partnerships more strategically and systematically to reach goals more rap- Harshad Sanghvi, Jeffrey Michael Smith, Koki Agarwal, Blami Dao, idly and with longer-lasting efforts. We should not, however, Ronald Magarick Jhpiego sacrifice quality for speed; quality improvement, particularly the linkages between quality and health outcomes, remains a critical Acknowledgments but under-funded area of our work. The second section examines the “who” of implementation, The Guest Editors of this supplement thank Rehana Gubin and Judith documenting ways to empower all cadres of frontline health workers Fullerton for their editorial support during the preparation of manu- with appropriate competencies to deliver evidence-based interventions, scripts and acknowledge additional editorial contributions from Karen wherever women and children need them. It is our belief that those Kirk, Sarah Ju, and members of the Jhpiego Publications Department. http://dx.doi.org/10.1016/j.ijgo.2015.04.018 0020-7292/© 2015 Published by Elsevier Ireland Ltd. on behalf of International Federation of Gynecology and Obstetrics. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). S2 Editorial

The Guest Editors also thank the donors that funded the many programs Jeffrey Michael Smith described in this supplement, including the United States Agency for Jhpiego, Baltimore, MD, USA International Development. Koki Agarwal Conﬂict of interest Jhpiego, Washington, DC, USA

The authors have no conﬂicts of interest. Blami Dao Jhpiego, Baltimore, MD, USA Harshad Sanghvi⁎ Jhpiego, Baltimore, MD, USA Ronald Magarick Corresponding author at: 1615 Thames Street, Baltimore, Jhpiego, Baltimore, MD, USA MD 21231-3492, USA. Tel.: +1 410 537 1800. E-mail address: [email protected] (H. Sanghvi). International Journal of Gynecology and Obstetrics 130 (2015) S3

Contents lists available at ScienceDirect

International Journal of Gynecology and Obstetrics

journal homepage: www.elsevier.com/locate/ijgo

INVITED EDITORIAL Jhpiego: Model supporter of implementation science

As Editor of the International Journal of Gynecology and Obstetrics show that policy and national and local factors must be considered in (IJGO) it was my pleasure to shepherd this Supplement and, as Editor the transfer of scientific information to practice. Emeritus, it is my pleasure to thank the authors and guest editors for Jhpiego has long focused on “training the trainers” and, as an early their contributions. Jhpiego, an affiliate of Johns Hopkins University, adopter of pre-service education at the medical and nursing school was founded in 1974 by Doctor Theodore M. King, Director of the levels, is developing lifelong learners engaged in important work that Department of Gynecology and Obstetrics at Hopkins, realizing the remains in maternal and neonatal health. The readers of IJGO will find need to introduce reproductive health breakthroughs to low-income much to learn from the examples and lessons in this Supplement: the countries to save the lives of women and improve the health of “how,” the “who,” and the “what.” I thank Jhpiego not only for its families worldwide. Since its inception, Jhpiego has been providing many contributions to our shared vision and goals, but for the partner- education and capacity building for physicians, nurses, administrators, ship that it, as a nongovernmental organization, has shown to FIGO and and other pre-service providers, and supporting programs to make other professional organizations in achieving our mutual goals of a bet- these lifesaving skills and services available and accessible. In this ter world through improved women’s health worldwide. Supplement, the experience of Jhpiego in implementing maternal and I also thank the guest editors, Harshad Sanghvi, Jeffrey M. Smith, newborn health is described in order to provide a way forward to ad- Koki Agarwal, and Blami Dao, and especially the managing editors of dress the Sustainable Development Goals (SDGs) and achieve and attain the Supplement, Rehana Gubin and Judith Fullerton, for their efforts the maternal and newborn development goals of the post-Millennium and contributions in making it possible. Development Goals (MDGs) era. Implementation science promotes the uptake of research findings Conflict of interest into routine health care in both clinical and policy contexts [1,2].Itisa new concept and discipline that is the natural extension of translational research, clinical research, outcomes research, and health services re- Dr Johnson is the Editor Emeritus of IJGO. search, and melds translational research and health technologies development with health services science, public health, and policy. It is an References interdisciplinary science that is of the times and whose time has come. This is complex interdisciplinary work and Jhpiego has consistently [1] Peterson HB, Haidar J, Merialdi M, Say L, Gülmezoglu AM, Fajans PJ, et al. Preventing ma- evidenced, through its educational programs and demonstration pro- ternal and newborn deaths globally: using innovation and science to address challenges in implementing life-saving interventions. Obstet Gynecol 2012;120(3):636–42. jects, an understanding of the clinical, public health, and policy actions [2] Johnson TR. Implementing evidence-based science to improve women’s health required for the transformation of health care in low-income areas globally. Int J Gynecol Obstet 2013;122(2):91–3. around the world through scientific innovation. Jhpiego has partnered with FIGO, the International Confederation of Timothy R.B. Johnson Midwives, and national professional societies in projects on emergency University of Michigan, Ann Arbor, USA obstetrics and newborn care, postpartum hemorrhage, sexually trans- 1500 East Medical Center Drive, L-4000 Von Voigtlander Women's mitted infections and HIV, family planning (especially endoscopic and Hospital, Ann Arbor, MI 48109–5276, USA. Tel.: +1 734 764 8123. postpartum family planning), malaria, and preventing mother-to-child E-mail address: [email protected]. transmission of HIV, to implement science for the public good and

http://dx.doi.org/10.1016/j.ijgo.2015.04.019 0020-7292/© 2015 International Federation of Gynecology and Obstetrics. Published by Elsevier Ireland Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). International Journal of Gynecology and Obstetrics 130 (2015) S4–S10

Contents lists available at ScienceDirect

International Journal of Gynecology and Obstetrics

journal homepage: www.elsevier.com/locate/ijgo

SUPPLEMENT ARTICLE Scaling up high-impact interventions: How is it done?

Jeffrey Michael Smith a,⁎, Joseph de Graft-Johnson b,PashtoonZyaeec, Jim Ricca a, Judith Fullerton d a Jhpiego, Baltimore, MD, USA b Save the Children/MCHIP, Washington, DC, USA c International Confederation of Midwives, The Hague, Netherlands d Independent consultant, San Diego, CA, USA article info abstract

Keywords: Building upon the World Health Organization’s ExpandNet framework, 12 key principles of scale-up have Helping Babies Breathe emerged from the implementation of maternal and newborn health interventions. These principles are illustrat- Maternal and newborn health ed by three case studies of scale up of high-impact interventions: the Helping Babies Breathe initiative; pre- Midwifery service midwifery education in Afghanistan; and advanced distribution of misoprostol for self-administration Misoprostol at home births to prevent postpartum hemorrhage. Program planners who seek to scale a maternal and/or new- Scale-up born health intervention must ensure that: the necessary evidence and mechanisms for local ownership for the intervention are well-established; the intervention is as simple and cost-effective as possible; and the implementers and beneﬁciaries of the intervention are working in tandem to build institutional capacity at all levels and in consideration of all perspectives. © 2015 Published by Elsevier Ireland Ltd. on behalf of International Federation of Gynecology and Obstetrics. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/3.0/).

1. Introduction Services International, Broad Branch Associates, and the Johns Hopkins University’s Institute for International Programs. While the specifics of As countries push to achieve Millennium Development Goals Jhpiego’s scale-up approach have differed depending on the local (MDGs) 4 and 5, increasing attention is being paid to the equitable context, experience has allowed it to distill common, practice-based scale-up of proven high-impact interventions for reproductive, mater- principles. As a result, Jhpiego now has a more refined and articulable nal, newborn, and child health [1]. Experience has demonstrated that approach to scale-up that we, as authors representing Jhpiego and part- technical interventions that are known to be effective at a small scale ners who have worked closely with Jhpiego, aim to share for the benefit under tightly controlled conditions cannot naturally be assumed to be of other public health practitioners. widely adopted and scaled up to cover large segments of the population, For the purposes of the present article, we derive our definition of despite scale being essential for population-level impact. Scale-up is scale-up from ExpandNet, a community of practice for global public challenging, and it is not always successful. health practitioners that is focused on developing and promoting best Over the past 40 years, Jhpiego and its partners have assisted various practices at scale [3]. ExpandNet’sdefinition encompasses both the countries, through local efforts and global alliances, to achieve some process and the objective of increasing coverage of an intervention: level of scale-up of maternal and newborn health (MNH) interventions “deliberate efforts to increase the impact of health service innovations across the global development spectrum. Substantial scale-up activity successfully tested in pilot or experimental projects so as to benefit has occurred over the last decade, particularly with support from more people and to foster policy and programme development on a USAID via its flagship Maternal Child Health Integrated Program lasting basis” [3]. Progress toward expanding levels of coverage for an (MCHIP), which supported programmatic efforts in MNH in more than intervention is sometimes termed “horizontal scale-up.” The process 40 countries from 2008 to 2014 [2] and which was led by Jhpiego in of institutionalizing an intervention at all levels of a local implementing partnership with Save the Children, John Snow, Inc., Program for organization (usually the ministry of health), so that it can manage and Appropriate Technology in Health, ICF International, Population sustain an intervention at a high level of horizontal scale-up, is sometimes termed “vertical scale-up.” After stating the principles of scale- up that have emerged from our work, we will describe three illustrative cases in which Jhpiego’s and its partners’ deliberate efforts to assist ⁎ Corresponding author at: Jhpiego, 1615 Thames Street, Baltimore, MD 21231, USA. Tel.: +1 410 537 1896. vertical scale-up using these principles have led to successful horizontal E-mail address: [email protected] (J.M. Smith). scale-up. http://dx.doi.org/10.1016/j.ijgo.2015.03.010 0020-7292/© 2015 Published by Elsevier Ireland Ltd. on behalf of International Federation of Gynecology and Obstetrics. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/3.0/). J.M. Smith et al. / International Journal of Gynecology and Obstetrics 130 (2015) S4–S10 S5

Box 1 team and health system managers learn the factors necessary for its The 12 principles of scale-up used by Jhpiego and partners. successful local application and contextualization. As ministries of health increase geographic coverage of the intervention, scale-up A. Actions related to the STRATEGY for scale-up moves to an early expansion phase. In this phase, the resource team 1. Build evidence and engage in evidence-based advocacy works with the ministry of health to identify needs for capacity- fi (globally and nationally) building, and to continue to re ne the intervention based on evidence 2. Coordinate/partner with other donors and technical agen- from pilot experiences. As the ministry of health attains coverage of cies national scope, or enters the mature expansion phase, issues of institu- 3. Mobilize resources tionalizing the intervention and maintaining quality and fidelity 4. Promote country ownership by integrating and harmoniz- become the most crucial considerations. ing intervention with current systems A common understanding in the process of scale-up, which is implied B. Actions related to the INTERVENTION to be scaled up in the 12 scale-up principles, is that no organization or agency works in- 5. Simplify and standardize intervention dependently or in isolation. In fact, Jhpiego has always worked in an envi- 6. Make intervention cheaper (i.e. more cost-effective) ronment characterized by the leadership of relevant ministries, alliances C. Actions related to the IMPLEMENTING ORGANIZATION with partners, and multilateral networks that address the specific actions AND BENEFICIARIES of the scaled intervention required for expansion of coverage within complex health systems. 7. Identify and work with champions The three recent illustrative case examples of the scale-up principles 8. Advocate for and develop needed policy/guideline are outlined in Table 1 and described in further detail below. Each case changes involves a key MNH problem, and each is in a different phase of scale-up. 9. Build capacity of implementing organization(s) for training, management, and logistics 2. Managing newborn asphyxia: Ensuring newborn resuscitation 10. Use data for management, including strengthening through the Helping Babies Breathe approach monitoring and evaluation 11. Support institutions, such as pre-service education sites and Almost three million newborns die each year, and the majority of professional organizations, that are agents of scale-up these deaths occur within 24 hours of birth. Globally, 23% of newborn 12. Engage and empower clients and communities deaths, or 700 000 annually, are due to birth asphyxia [12]. Thus, birth asphyxia has been targeted by the global public health community as a priority issue. As described in Box 2, Helping Babies Breathe (HBB), a methodology developed by the American Academy of Pediatrics (AAP) 1.1. The principles of scale-up in 2009 to address birth asphyxia [13], expands upon time-tested clinical guidelines using simple, focused learning tools and approaches for new- There is a large body of literature discussing the variety of frameworks born resuscitation. This new methodology and training package has prov- that have been developed to guide the effort to scale-up health interven- en instrumental in moving toward effective implementation because of – tions [4 6]. The scale-up principles that Jhpiego and partners have its emphasis on mastery and consistent application of the initial steps of crystalized for the purposes of their collective implementation efforts in resuscitation, which are respiratory support through stimulation and global MNH are presented in Box 1. These principles are based on the use of a self-inflating bag and mask when necessary. HBB focuses on fi logic of the ExpandNet framework shown in Fig. 1 [7] but contain modi - prompt resuscitation during the Golden Minute (APP, Elk Grove Village, – cations that have been adapted from insights in other frameworks [8 10]. IL,USA)afterchildbirthasanintegralcomponentofessentialnewborn fi The ExpandNet framework links ve interacting pieces: the care (ENC). The HBB approach adheres to a simple assessment and inter- “ ” resource team (in this case, a group of technical assistance organiza- vention pathway, and development and maintenance of skills through “ tions led by Jhpiego or a partner), works in concert with the user organi- simulation, using a low-cost, lifelike anatomic model, both in the learning ” zation(s) (i.e. the ultimate implementer, usually a ministry of health) to venue and at the workplace. The HBB training package can be used alone “ ” “ help scale-up an innovation, or intervention, through a scaling-up or integrated into existing national training materials for ENC, emergency ” “ ” strategy within the relevant environment, or context. obstetric and newborn care, or integrated management of newborn and Scale-up tends to happen in phases similar to that of product intro- childhood illness. The desired output is to have at least one person who duction [11], as shown in Fig. 2, beginning with the introduction of an is skilled in newborn resuscitation at every birth [14]. intervention, during which the intervention is piloted by the resource MCHIP embraced the HBB approach for the management of asphyx- iated newborns and included it in all of its neonatal programs. Both directly and as a member of the HBB Global Development Alliance (GDA), which was established in June 2010, MCHIP worked with USAID missions, other technical agencies (most significantly AAP and Laerdal Global Health), host country governments, professional associations, and nongovernmental organizations to facilitate the scale-up of HBB in 54 countries as of April 2013 [15]. The MCHIP strategy for scaling up proven interventions was to galvanize action at both the global and the country level. MCHIP’s direct support for implementation included: regional training of trainers in Asia and Africa; partnership with AAP for development of an HBB training video and implementation guide; in-country mentorship through a variety of GDA members; and HBB website maintenance. MCHIP’s breadth allowed cost efficiencies, such as the waiver of copyright fees and access to training models at cost. Facilitating collaboration among partners, the GDA achieved progress to scale more rapidly than any partner alone because each partner fi Fig. 1. ExpandNet framework for scale-up. Reproduced with permission from WHO, could take on speci c roles. For example, AAP developed the technical ExpandNet [7]. evidence base, Laerdal provided initial training materials for national S6 J.M. Smith et al. / International Journal of Gynecology and Obstetrics 130 (2015) S4–S10

Fig. 2. Phases of the scale-up process. Adapted with permission from PATH [11]. and regional training events, and MCHIP supported implementation in generated through HBB implementation. Annually, approximately 30 some countries, and policy changes in others. A large trial in Tanzania, 000 newborn deaths in Bangladesh are due to birth asphyxia. To reduce led by AAP, demonstrated the mortality impact of HBB [16].Inparallel, this burden, MCHIP, in partnership with AAP, committed modest MCHIP used implementation science approaches in Malawi and funding to study the feasibility and impact of introducing HBB in Bangladesh to document processes for scale-up at the country level, Bangladesh. In collaboration with the Ministry of Health and Family evidence that was used to support training, mentorship, and quality Welfare (MOHFW), Bangabandhu Sheikh Mujib Medical University assurance in other countries, including Kenya, Mozambique, and (BSMMU), the International Centre for Diarrheal Disease Research, Zambia. Bangladesh, and Save the Children’s Saving Newborn Lives project, At the country level, a key scale-up principle for MCHIP was to sup- MCHIP supported the introduction of HBB in selected public sector port governments and their in-country partners to build on existing hospitals and clinics. An HBB champion emerged at BSMMU to lead platforms to introduce or expand HBB. In line with this principle, the study, which showed that HBB, within the context of ENC, was consultative advocacy and planning meetings were held with relevant feasible for all skilled birth attendants (SBAs) at both the facility and divisional heads within ministries of health regarding the potential community levels. The study data and demonstration of learning role that HBB could play in reducing newborn deaths due to birth materials allowed the HBB champion to foster local ownership of the asphyxia, focusing on the ministries’ decisions to endorse the introduc- intervention by engaging other stakeholders to discuss national scale- tion/expansion of HBB. This initial planning served to ensure country up. At a national conference in September 2010, the Minister for Health ownership, coordinate approaches, and mobilize multiple resource instructed his directorate heads to scale-up HBB across the country. This channels. During the planning process, decisions were made about: leadership and commitment, in the presence of stakeholders and (1) the level of the healthcare system at which the intervention would donors, became a pivotal point for mobilization of resources for a be delivered; (2) the cadre of healthcare worker to train, especially mid- nationwide HBB scale-up. The scale-up process is ongoing, led by the wives, and the selection of master trainers, particularly those working in MOHFW, with BSMMU as the technical implementation leader. the maternity or neonatal unit; (3) service delivery strategies to meet Through this coordinated and locally championed effort, HBB services each country’s needs, be it vertical (as in Bangladesh and Malawi) or have been rolled out to 55 districts in Bangladesh, involving over 21 integrated (as in Ethiopia); (4) a resource mobilization strategy for 000 health workers, since 2013 [17]. training, supervision, and service delivery of supplies; (5) a strategy for wide geographic coverage, phased in based on resource availability; 3. Addressing maternal mortality in Afghanistan: Rapidly expanding and (6) an ongoing monitoring and evaluation component, using midwifery education to overcome human resource shortages selected indicators. Most countries mobilized adequate resources to initiate HBB-related activities in selected regions, provinces, or The expansion of midwifery education in Afghanistan exempliﬁes districts while advocating for additional resources to scale nation- application of the principles of scale-up to a larger health system wide. A few countries—for example, Bangladesh—were able to component—human resource development—at a national level. mobilize the needed resources for a nationwide scale-up at the start. Among the biggest challenges facing the Ministry of Public Health Bangladesh is an example of a country in which the decision to (MoPH) in Afghanistan in 2002, after 23 years of conﬂict and isola- pursue national scale-up was reached because of use of local evidence tion, was the alarmingly high maternal mortality ratio, estimated at

Table 1 Illustrative case examples.

Intervention Health problem Location Phase of scale-up Lead technical organization

Helping Babies Breathe (targeted Birth asphyxia Multiple countries Early to mature expansion, in Save the Children newborn resuscitation package) various countries Training and deployment of midwives Low rates of skilled attendance at birth Afghanistan Mature expansion Jhpiego and limited human resources Advanced distribution of misoprostol for Postpartum hemorrhage Multiple countries Introduction to early Multiple, sometimes Jhpiego self-administration at home births expansion J.M. Smith et al. / International Journal of Gynecology and Obstetrics 130 (2015) S4–S10 S7

Box 2 attained certain competencies prior to graduation. The process of Helping Babies Breathe. developing educational standards that are now in alignment with current international guidelines [27], as well as a standards-based Helping Babies Breathe (HBB) is an evidence-based educational management and accreditation methodology, allowed common and program, based on the International Liaison Committee on Resus- transparent conversation among all implementing partners and clear citation’s (ILCOR) consensus on scientific conclusions that have leadership by the government [28,29]. Two educational pathways undergone a WHO scientific technical review. HBB has the follow- were identified, one through the existing Institutes of Health Sciences ing elements: for students with secondary school credentials, and the other through newly established provincial-level community midwifery schools, • Culturally sensitive, pictorial-based learning materials, which initially admitted students with a minimum ninth class educa- including a learner workbook, action plan wall poster, and tional achievement until 2009 when the curriculum was updated and facilitator flip chart. the admission requirement was advanced to a tenth class education • Realistic newborn simulator (NeoNatalie, developed by minimum [30]. Laerdal Medical) with the ability to imitate an umbilical Despite the two pathways, educators and MoPH human resource pulse, bag-mask ventilators, and penguin suction bulb that managers agreed on a common core educational curriculum and similar can be disinfected. All equipment has been tested for durabil- expected outcomes of learning (e.g. core competencies demonstrated ity in a variety of climates and teaching conditions and is prior to entry into the workforce). Standardization and systematization made available at cost to the Millennium Development Goal made the approach conceptually simple and programmatically achiev- countries. able. Initially, for example, quarterly meetings called by the MoPH • An ongoing mentorship program to provide expert assistance, were held in Kabul for all organizations implementing a midwifery implementation guidance, knowledge exchange, integration education program in order to benchmark progress and collectively and evaluation support, and continuous quality improvement address challenges. for sustained practice outcomes and decreased infant The formal cultivation of champions of midwifery was a critical part mortality. of the process that ultimately scaled up the intervention to 33 out of 34 provinces in the country. In 2005, the Afghan Midwifery Association (AMA) was formed not only for continuing professional support to midwives in the country, but also to promote the professionalization of the 1600–2200 per 100 000 live births [18].Atthattime,nationalprena- concept of midwifery [21]. Success of the revived concept of midwifery tal care and SBA rates were 16% and 14%, respectively, and in rural in Afghanistan, and indeed the system of service delivery of maternal areas the rates were even lower, at 8% and 7% [19].TheAfghan health services, was known to be intricately linked to the demand for MoPH recognized the urgent need to comprehensively revitalize services. Increasing demand was necessary to help new, relatively the health system for the provision of maternal health services young graduates gain experience and demonstrate to their communi- but also recognized the limitations inherent in a postconflict envi- ties that their skill-focused, professional education made them more ronment; therefore, it focused early and intensively on building valuable than traditional providers. Formation of the AMA, with human resource capacity. branches in almost every province of the country, promoted respect According to the MoPH, in 2002 only 467 midwives were available in for the profession and an understanding of the responsibilities that the country to serve an estimated population of 24.5 million people [20]. come from being a member of a profession. New graduates were Midwifery education in the country’s five schools had been suspended asked to recite the midwifery oath upon graduation, which formally by the Taliban, and no new graduates had emerged in more than conferred on them a mantel of responsibility. For many young women seven years. Nevertheless, midwives were acknowledged as the pre- in Afghanistan, membership in the AMA was the first time that they ferred providers of maternity services, given cultural preferences for had been members of any group outside their own family networks. female providers [21,22]. All of these components created numerous champions not only at a na- The MoPH and its international development partners quickly tional level, but also at the community level, which promoted respect identified the education of new midwives, who would be preferentially for the educational programs and the graduates, fostered demand for recruited from and immediately deployed back to the refurbished admission, and, theoretically, promoted sustainability. health centers in their home communities in rural areas, as a logical The results of adherence to these scale-up principles were re- response to the need to sustainably increase access to maternal health markable. Between 2002 and 2004, the five government midwifery services, especially among the rural poor [23]. Global evidence [24,25] schools reopened, and 17 new community midwifery schools were guided decision makers away from training traditional birth attendants established at the provincial level. By 2012 the country’saccredited and toward the competency-based education of women as health midwifery education programs had graduated 3827 new midwives. workers who would fulfill the criteria of SBAs [26]. This created a clear More recently, in response to the results of targeted evaluation and compelling narrative for building partnerships and securing studies and government strategic planning for the midwifery pro- donor support. The clarity of this vision united groups in the devel- fession, the nationally approved curriculum has been revised and opment of an integrated and harmonized strategy in line with program length expanded [30–33]. Some schools have also ceased other components of the health system, and across all areas of the operation in response to changing community characteristics and country. Initially, in 2002, funding for midwifery education was needs. In 2013, there were 970 students enrolled in midwifery edu- provided by USAID and UNICEF; by 2012, funding came from at cation in the remaining 24 schools in the country [33].Anumberof least 12 different sources, including the embassies of Denmark, private midwifery schools have also opened, although their adher- the Netherlands, Norway, Sweden, and Turkey; the European Com- ence to the guiding principles is less clear. Graduates have entered mission; the World Bank; USAID; WHO; UNFPA; UNICEF; and the both the public and private sectors, and while there are several Government of Afghanistan. constraining factors [34], retention of graduates in the workforce Drawing on lessons learned in other Asian countries (specifically, remains high [35].Recentdataindicatethat96%ofwomeneducat- Indonesia and Nepal), where Jhpiego supported the retraining of large ed through the community midwifery program remained in their numbers of midwives who had a limited skill set, an effort was made home communities, and 74% of women educated through the early on to develop a structured educational process that would government midwifery programs in the larger cities continued in ensure student skill progression was central and that they had their urban workplace [36]. S8 J.M. Smith et al. / International Journal of Gynecology and Obstetrics 130 (2015) S4–S10

4. Preventing postpartum hemorrhage with misoprostol: Three pills evidence fostered greater global enthusiasm and accelerated program- that can save a life matic activity. Even more evidence was needed; however, to generate additional Postpartum hemorrhage (PPH) remains the leading cause of mater- resources and willingness by countries to explore whether prevention nal mortality globally [37]. It is a major public health problem, and of PPH at home birth using misoprostol was applicable to their context. prevention has emerged as a priority under MDG 5 [38]. Misoprostol Jhpiego and partners therefore adapted the program from Indonesia for was first studied by El-Refaey in 1997 for its clinical ability to prevent two operations research studies—one in Afghanistan [49] and the other PPH [39];itsclinicalefficacy has been further refined in subsequent in Nepal [45]—to refine the model and focus on ADMSA immediately trials [40–45]. after birth. With the support of multiple funders, numerous research Advance distribution of misoprostol for self-administration and programmatic organizations took up the issue of implementation (ADMSA) has the appeal of a classic public health intervention: it to expand the use of misoprostol for PPH prevention. Between 2006 can be implemented at the moment of need, regardless of whether and 2012, an additional 15 programs were started in 11 countries the practitioner is present; and it has the potential to reach a large [50]. The sheer level of activity in the area may have influenced other segment of the vulnerable population. Misoprostol for PPH preven- countries to consider registering misoprostol for the PPH indication in tion is empowering, effective, and pro-poor, preferentially impacting order to establish similar programs for their own populations (Fig. 3) those with limited resources [45]. The attractiveness of a simple [51]. message that a pill could prevent a woman from dying in childbirth Despite enthusiasm within the global health community, WHO felt at home helped to attract the attention and interest of the global that the evidence for ADMSA remained incomplete [52]. The PPH community as a potentially scalable intervention. In addition, the prevention guidelines issued in 2007 and 2009 specifically restricted intervention addresses a classic public health reality: the clinical use of misoprostol to facility settings [52,53], which tempered progress practitioner cares for the patient who reaches the facility, while the at the country level. National ministries of health and other donors were public health practitioner must care for the client who does not. reserved in their adoption of the approach, awaiting WHO’s firm The first programmatic effort to use misoprostol for the prevention endorsement. In a 2012 survey, although 16 countries had piloted the of PPH at homebirth was made by Jhpiego in 1999 in Indonesia [46]. ADMSA approach, only five were scaling it up [54]. This experience was notable because it focused on a comprehensive In light of this, or perhaps because of this, champions and a global approach to PPH reduction by addressing births in a facility as well as community of practice have emerged to advocate for the ADMSA births at home. approach and bring supporting evidence within the reach of key Only a single program, in the Gambia, had adopted this approach stakeholders. Since 2005, when the Nepal and Afghanistan programs [47] until a randomized controlled trial in India demonstrated that the were initiated, Jhpiego has assisted in the initiation or expansion of administration of misoprostol reduced the rate of acute PPH at home PPH prevention programs using misoprostol in an additional 12 coun- birth from 12.0% to 6.4% compared with the placebo [48]. This additional tries (Bangladesh, Ethiopia, Guinea, Liberia, Madagascar, Malawi,

Fig. 3. Global regulatory approvals for misoprostol as of May 2013. Reprinted with permission from Venture Strategies Innovations [51]. J.M. Smith et al. / International Journal of Gynecology and Obstetrics 130 (2015) S4–S10 S9

Mozambique, Nigeria, Pakistan, Philippines, Rwanda, South Sudan) and experienced professionals, organizations, donors, and imple- [55,56], and other partners have done the same in another eight menters who mobilize and direct resources toward global impact countries (Burma, Ghana, Kenya, Senegal, Somaliland, Tanzania, for newborn resuscitation. It has helped increase and direct donor Uganda, Zambia) [50,57]. Jhpiego has supported widespread dissem- funds, ensured resources for evaluation, and untangled the logistics ination of its findings from these programs to ensure that discussion of commodity security. Donors in Afghanistan generously support- of ADMSA has remained prominent at numerous regional and global ed the midwifery education environment, in part due to the MoPH’s meetings, including the International Federation of Gynecology and demonstrated commitment to its vision and the structured path- Obstetrics’ (FIGO) conferences in Cape Town and Rome in 2009 and way pursued by Jhpiego and its partners. 2012, the Postpartum Hemorrhage Prevention meeting in Bangkok Constraints abound, however [63–65]. Continued expansion rein 2005, and MCHIP regional conferences in Addis Ababa and quires that all the principles of scale-up are adhered to for a substantial Dhaka in 2011 and 2012. These strategic engagement opportunities period of time. Moving from pilots to expansion to scale requires the de- have contributed to calls for providers from Asia and Sub-Saharan velopment of systems and the collection of data. Institutionalization of Africa to scale up the intervention [58]. Misoprostol is now recog- the intervention requires drafting and adhering to new policies and nized by the global health community as an essential commodity, human resource procedures, modifying pre-service education curricula, and its availability is being enhanced through the UN Commission and updating faculty. Data on the extent of coverage and achievement of on Life-Saving Commodities for Women’s and Children’s Health [59]. scale can come only after the development of appropriate indicators and the modification of health information systems. Achievement of scale 5. Discussion requires the development of systems, often in places where the systems themselves are barriers to scale. Sustaining scale-up and maximizing Although the ultimate goal of scale-up (i.e. high, effective coverage) equity at scale are critical challenges going forward. is similar across all programs, there is no single route to scale-up. Basic research and practical experimentation with several pathways to scale- 6. Conclusion up have, nevertheless, generated a number of lessons learned, particularly in low- and middle-income countries [2,60]. Country ownership, Achievement of coverage at scale of certain fundamental interven- including the engagement of both program-level champions and the tions is both necessary and achievable. By identifying and adhering to community, is key [10,61]. principles and processes, Jhpiego, over its 40-year history and with its — The rapid expansion of a technical intervention be it a clinical or ed- numerous partners, has laid a path for moving from innovation to — ucational one is facilitated by simplicity and standardization, clear ev- scale. As demonstrated by the three case examples, programs are ad- idence of effectiveness, dedicated champions, local capacity, adequate vised to focus on the essence of the intervention to be scaled; to plot a resources, and national or global ownership. Compelling messages, willful strategy built on evidence and informed through local owner- fi whether about simpli ed techniques for the resuscitation of newborns ship; and to engage and mobilize both implementers and beneficiaries or medications to prevent PPH in home births, make the intervention in the capacity-building and institutionalization processes necessary to understandable. Better comprehension likely encourages partners to ac- achieve national coverage. cept and promote the intervention and push it beyond the trial or pilot stage. Conflict of interest As seen with the use of misoprostol for PPH prevention, clinical and programmatic evidence is necessary to build technical consensus for an fl intervention. Clinicians, professional associations, governments, and The authors have no con icts of interest. global advocates need data on feasibility and effectiveness to justify the promotion of one intervention over another or the use of a selected References intervention where previously none was available. There will be coun- fi tries who are early adopters, willing to initiate programs based on clin- [1] UNICEF, World Health Organization. Ful lling the Health Agenda for Women and Children. The 2014 Report. Countdown to 2015: maternal, newborn and child sur- ical evidence alone, while other countries need programmatic evidence, vival. http://www.countdown2015mnch.org/reports-and-articles/2014-report. cost-effectiveness data, or population impact data to adopt new and in- Published 2014. novative approaches. For this reason, new programs must publish their [2] Larson A, Raney L, Ricca J. Lessons learned from a preliminary analysis of the scale- fi up experience of six high-impact reproductive, maternal, newborn, and child health ndings, and ongoing programs must continue to monitor and evaluate (RMNCH) interventions. Baltimore, MD: Jphiego; 2014. and share their lessons learned. Furthermore, new research and model- [3] World Health Organization, ExpandNet. Practical guidance for scaling up health ing approaches, such as that used to quantify the impact of scaling up service innovations. Geneva: WHO; 2009. http://www.expandnet.net/PDFs/WHO_ ExpandNet_Practical_Guide_published.pdf. midwifery, can help to develop new interest or to sustain commitment [4] Cooley L, Kohl R. Scaling up—from vision to large-scale change: a management to scale up strategies [62]. framework for practitioners. Washington, DC: Management Systems International; Yet even with data, programs sometimes cannot gain the necessary 2006. [5] Mangham L, Hanson K. Scaling up in international health: what are the key issues? foothold without dedicated champions who present the case time and Health Policy Plan 2010;25(2):85–96. again to the cautious and the skeptical. The success of the Afghanistan [6] Subramanian S, Naimoli J, Matsubayashi T, Peters D. Do we have the right models for midwifery experience required champions at all levels: in the MoPH scaling up health services to achieve the Millennium Development Goals? BMC Health Serv Res 2011;11:336. and its multiple departments; among the development partners, both [7] World Health Organization, ExpandNet. Nine steps for developing a scaling-up strategy. as leaders and implementers; within professional associations, such as Geneva: WHO; 2010. http://whqlibdoc.who.int/publications/2010/9789241500319_ the AMA; and among the population of young women being educated eng.pdf. and mothers being served. The same can be said for the HBB resuscita- [8] Levine R. Case studies in global health: millions saved. 2nd ed. Burlington, MA: Jones and Bartlett Learning; 2007. tion approach and the use of misoprostol for prevention of PPH. The [9] Yamey G. Scaling up global health interventions: a proposed framework for success. role of a champion must not be undervalued, for champions are the PLoS Med 2011;8(6):e1001049. ones who repeat the message, craft the approach, navigate the barriers, [10] Bryce J, Victora CG, Boerma T, Peters DH, Black RE. Evaluating the scale-up for maternal and child survival: a common framework. Int Health 2011;3(3):139–46. and motivate the masses. [11] Program for Appropriate Technology in Health. PATH’s framework for product intro- Finally, adequate resources, combined with national ownership and duction. http://www.path.org/publications/files/TS_product_intro_framework.pdf. global technical consensus, have enabled the interventions profiled to Published 2007. Accessed April 30, 2014. [12] Save the Children. Surviving the first day: state of the world’s mothers 2013. http:// be taken further and deeper into health systems, achieving institution- www.savethechildrenweb.org/SOWM-2013/. Published 2013. Accessed April 19, alization and sustainability. The HBB GDA is a collection of dedicated 2014. S10 J.M. Smith et al. / International Journal of Gynecology and Obstetrics 130 (2015) S4–S10

[13] American Academy of Pediatrics. Helping Babies Breathe training significantly re- [41] León W, Durocher J, Barrera G, Pinto E, Winikoff B. Dose and side effects of sublingual duces newborn deaths. http://www.aap.org/en-us/about-the-aap/aap-press-room/ misoprostol for treatment of postpartum hemorrhage: what difference do they pages/Helping-Babies-Breathe-Training-Significantly-Reduces-Newborn-Deaths. make? BMC Pregnancy Childbirth 2012;12:65. aspx. Published 2012. Accessed April 19, 2014. [42] Soltan MH, El-Gendi E, Imam HH, Fathi O. Different doses of sublingual misoprostol [14] Van Heerden C, Cur B, Cur M. An introduction to Helping Babies Breathe: the “Gold- versus methylergometrine for the prevention of atonic postpartum haemorrhage. en Minute” is here for South African newborn babies. Prof Nurs Today 2012;16(3): Int J Health Sci (Qassim) 2007;1(2):229–36. 6–7. [43] El-Refaey H, Nooh R, O’ Brien P, Abdalla M, Geary M, Walder J, et al. The misoprostol [15] Helping Babies Breathe: a global public-private alliance status report 2013. http:// third stage of labour study: a randomized controlled comparison between orally www.helpingbabiesbreathe.org/docs/HBB%20broch%20low.pdf. Published 2013. administered misoprostol and standard management. BJOG 2000;107(9):1104–10. Accessed April 19, 2014. [44] Hofmeyr GJ, Nikodem VC, de Jager M, Gelbart BR. A randomized placebo controlled [16] Msemo G, Massawe A, Mmbando D, Rusibamayila N, Manji K, Kidanto H, et al. New- trial of oral misoprostol in the third stage of labour. Br J Obstet Gynaecol 1998; born mortality and fresh stillbirth rates in Tanzania after Helping Babies Breathe 105(9):971–5. training. Pediatrics 2013;131(2):e353–60. [45] Rajbhandari S, Hodgins S, Sanghvi H, McPherson R, Pradhan YV, Baqui AH. [17] United Nations, Office of the United Nations Secretary-General Special Envoy Expanding uterotonic protection following childbirth through community-based for Financing the Health Millennium Development Goals and for Malaria. distribution of misoprostol: operations research study in Nepal. Int J Gynecol Obstet MDG success stories: Save The Children’s “Helping Babies Breathe.”. http:// 2010;108(3):282–8. www.mdghealthenvoy.org/mdg-success-stories-save-the-childrens-helping- [46] Sanghvi H, Wiknjosastro G, Chanpong G, Fishel J, Ahmed S, Zulkarnain M. Prevention babies-breathe. Published 2014. Accessed April 19, 2014. of postpartum hemorrhage study: West Java, Indonesia. https://www.k4health.org/ [18] Bartlett L, Mawji S, Whitehead S, Crouse C, Dalil S, Ionete D, et al. Where giving birth sites/default/files/10%20MNH%20PPH%20study%20West%20Java%202004_0.pdf. is a forecast of death: maternal mortality in four districts of Afghanistan, 1999–2002. Published 2004. Accessed April 21, 2014. Lancet 2005;365(9462):864–70. [47] Walraven G, Blum J, Dampha Y, Sowe M, Morison L, Winikoff B, et al. Misoprostol in [19] Afghanistan MoPH, Johns Hopkins School of Public Health, Indian Institute of Health the management of the third stage of labour in the home delivery setting in rural Management Research. Afghanistan Health Survey 2006: estimates of priority health Gambia: a randomised controlled trial. BJOG 2005;112(9):1277–83. indicators for rural, Afghanistan. http://s3.amazonaws.com/zanran_storage/www. [48] Derman RJ, Kodkany BS, Goudar SS, Geller SE, Naik VA, Bellad MB, et al. Oral miso- moph.gov.af/ContentPages/171062486.pdf#page=61.AccessedApril19,2014. prostol in preventing postpartum haemorrhage in resource-poor communities: a [20] Islamic Republic of Afghanistan, Ministry of Public Health. Afghanistan National randomized controlled trial. Lancet 2006;368(9543):1248–53. Health Resources Assessment. Kabul: MoH; 2002. [49] Sanghvi H, Ansari N, Prata NJ, Gibson H, Ehsan AT, Smith JM. Prevention of [21] Currie S, Azfar P, Fowler RC. A bold new beginning for midwifery in Afghanistan. postpartum hemorrhage at home birth in Afghanistan. Int J Gynecol Obstet 2010; Midwifery 2007;23(3):226–34. 108(3):276–81. [22] Mayhew M, Hansen PM, Peters DH, Edward A, Singh LP, Dwivedi V, et al. Determi- [50] Smith JM, Gubin R, Holston MM, Fullerton J, Prata N. Misoprostol for postpartum nants of skilled birth attendant utilization in Afghanistan: a cross-sectional study. hemorrhage prevention at home birth: an integrative review of global implementa- Am J Public Health 2008;98(10):1849–56. tion experience to date. BMC Pregnancy Childbirth 2013;13:44. [23] Herberg P. Nursing, midwifery and allied health education programmes in [51] Venture Strategies Innovations. Global misoprostol registration by indication. http:// Afghanistan. Int Nurs Rev 2005;52(2):123–33. www.vsinnovations.org/assets/files/Resources/VSI%20Global%20Miso%20Reg% [24] Koblinsky M, Matthews Z, Hussein J, Mavalankar D, Mridha MK, Anwar I, et al. Going 20Map%202013%2005%2021F.pdf. Published 2013. Accessed April 29, 2014. to scale with professional skilled care. Lancet 2006;368(9544):1377–86 Erratum in [52] World health Organization. WHO statement regarding the use of Misoprostol for 2006;368(9554):2210. postpartum haemorrhage prevention and treatment. Geneva: WHO; 2009. http:// [25] Koblinsky MA, Campbell O, Heichelheim J. Organizing delivery care: what works for whqlibdoc.who.int/hq/2009/WHO_RHR_09.22_eng.pdf?ua=1. safe motherhood? Bull World Health Organ 1999;77(5):399–406. [53] World Health Organization. WHO recommendations for the prevention of [26] World Health Organization. Making pregnancy safer: the critical role of the skilled postpartum hemorrhage. Geneva: WHO; 2007Accessed April 22, 2014. attendant. A joint statement by WHO, ICM and FIGO. Geneva: WHO; 2004. http:// [54] Smith JM, Currie S, Cannon T, Armbruster D, Perri J. Are national policies and whqlibdoc.who.int/publications/2004/9241591692.pdf?ua=1. programs for prevention and management of postpartum hemorrhage and [27] International Confederation of Midwives. Essential competencies for basic midwife- preeclampsia adequate? A key informant survey in 37 countries. Glob Health Sci ry practice. http://www.internationalmidwives.org/assets/uploads/documents/ Pract 2014;2(3):275–84. CoreDocuments/ICM%20Essential%20Competencies%20for%20Basic%20Midwifery% [55] Smith JM, Baawo SD, Subah M, Sirtor-Gbassie V, Howe CJ, Ishola G, et al. Advance 20Practice%202010,%20revised%202013.pdf. Published 2013. Accessed April 19, distribution of misoprostol for prevention of postpartum hemorrhage (PPH) at 2014. home births in two districts of Liberia. BMC Pregnancy Childbirth 2014;14:189. [28] Smith JM, Currie S, Azfar P, Rahmanzai AJ. Establishment of an accreditation system [56] Smith JM, Dimiti A, Dwived V, Ochiegn I, Dalaka M, Currie S, et al. Advance distribu- for midwifery education in Afghanistan: maintaining quality during national expan- tion of misoprostol for the prevention of postpartum hemorrhage in South Sudan. sion. Public Health 2008;122(6):558–67. Int J Gynecol Obstet 2014;127(2):183–8. [29] Bogren MU, Wiseman A, Berg M. Midwifery education, regulation and association in [57] Fujioka A, Smith JM. Prevention and management of postpartum hemorrhage six South Asian countries—a descriptive report. Sex Reprod Healthc 2012;3(2): and pre-eclampsia/eclampsia: national programs in selected USAID program- 67–72. supported countries. http://www.mchip.net/sites/default/files/mchipfiles/PPH_ [30] Islamic Republic of Afghanistan, Ministry of Public Health. National Policy and PEE%20Program%20Status%20Report.pdf. Published 2011. Accessed April 22, 2014. Strategy for nursing and midwifery services 2011–2015. http://moph.gov.af/en/ [58] Karanja J, Muganyizi P, Rwamushaija E, Hodoglugil N, Holm E, Regional Experts’ documents/category/strategies-and-policies?page=2.AccessedJanuary28,2015. Summit Group. Confronting maternal mortality due to postpartum hemorrhage [31] Turkmani S, Currie S, Mungia J, Assefi N, Javed Rahmanzai A, Azfar P, et al. “Midwives and unsafe abortion: a call for commitment. Afr J Reprod Health 2013;17(2):18–22. are the backbone of our health system”: lessons from Afghanistan to guide expan- [59] United Nations Commission on Life-Saving Commodities for Women’sand sion of midwifery in challenging settings. Midwifery 2013;29(10):1166–72. Children’s Health. Commissioners’ Report. http://www.unfpa.org/publications/un- [32] Zainullah P, Ansari N, Yari K, Azimi M, Turkmani S, Azfar P, et al. Establishing mid- commission-life-saving-commodities-women-and-children. Published 2012. wifery in low-resource settings: guidance from a mixed-methods evaluation of the Accessed April 22, 2014. Afghanistan midwifery education program. Midwifery Oct 2014;30(10):1056–62. [60] Knippenberg R, Lawn JE, Darmstadt GL, Begkoyian G, Fogstad H, Walelign N, et al. [33] Afghan Midwives Association. http://iawg.net/2011/Slides/Session%209/Sabera% Systematic scaling up of neonatal care in countries. Lancet 2005;365(9464): 28AMA%29%20Presentation-ST-final.pdf. Published 2011. Accessed July 23, 2014. 1087–98. [34] Rosskam E, Pariyo G, Hounton S, Aiga H. Increasing skilled birth attendance through [61] Prata N, Passano P, Sreenivas A, Gerdts CE. Maternal mortality in developing midwifery workforce management. Int J Health Plann Manage 2013;28(1):e62–71. countries: challenges in scaling-up priority interventions. Womens Health (Lond [35] Wood M, Mansoor GF, Hashemy P, Namey E, Gohar F, Ayoubi S, et al. Factors Engl) 2010;6(2):311–27. influencing the retention of midwives in the public sector in Afghanistan: a qualita- [62] Homer CS, Friberg IK, Dias MA, Ten-Hoope-Bender P, Sandall J, Speciale AM, et al. tive assessment of midwives in eight provinces. Midwifery 2013;29(10):1137–44. The projected effect of scaling up midwifery. Lancet 2014;384(9948):1146–57. [36] Mansoor G, Hill PS, Barss P. Midwifery training in post-conflict Afghanistan: tensions [63] Hanson K, Ranson MK, Oliveira-Cruz V, Mills A. Expanding access to priority health between educational standards and rural community needs. Health Policy Plan interventions: a framework for understanding the constraints to scaling-up. J Int 2012;27(1):60–8. Dev 2003;15(1):1–14. [37] Khan KS, Wojdyla D, Say L, Gülmezoglu AM, Van Look PF. WHO analysis of causes of [64] Hofmeyr G, Haws R, Bergström S, Lee AC, Okong P, Darmstadt GL, et al. Obstetric care maternal death: a systematic review. Lancet 2006;367(9516):1066–74. in low-resource settings: what, who and how to overcome challenges to scale up? [38] United Nations. The Millennium Development Goals Report 2012. http://www.un. Int J Gynecol Obstet 2009;107(Suppl. 1):S21–45. org/en/development/desa/publications/mdg-report-2012.html. Published 2012. [65] Moran AC, Kerber K, Pfitzer A, Morrissey CS, Marsh DR, Oot DA, et al. Benchmarks to Accessed April 21, 2014. measure readiness to integrate and sale up newborn survival interventions. Health [39] El-Refaey H, O’Brien P, Morafa W, Walder J, Rodeck C. Use of oral misoprostol in the Policy Plan 2012;27(Suppl. 3):iii29–39. prevention of postpartum haemorrhage. Br J Obstet Gynaecol 1997;104(3):336–9. [40] Hundley VA, Avan BI, Sullivan CJ, Graham WJ. Should oral misoprostol be used to prevent postpartum haemorrhage in home-birth settings in low resource countries? A systematic review of the evidence. BJOG 2013;120(3):277–85. International Journal of Gynecology and Obstetrics 130 (2015) S11–S16

Contents lists available at ScienceDirect

International Journal of Gynecology and Obstetrics

journal homepage: www.elsevier.com/locate/ijgo

COMMENTARY Best practices for a successful MNCH partnership that an external evaluation could never ﬁnd: Experiences from the Maternal and Child Health Integrated Program

Koki Agarwal a,⁎,NancyCaiolab, Anita Gibson c a Jhpiego, Washington, DC, USA b Jhpiego, Baltimore, MD, USA c Save the Children, Washington, DC, USA article info abstract

Keywords: Partnerships for maternal, newborn, and child health (MNCH) are increasingly prevalent, yet little has been pub- Child health lished about the possible reasons for their success or failure. In this commentary, we assess the presence of four Maternal health principles for a successful collaborative partnership—clear goals, clear roles, trust, and commitment—within the Newborn health Maternal and Child Health Integrated Program (MCHIP), an MNCH partnership among eight implementing orga- Organizational theory nizations that was funded by USAID from 2008 to 2014. MCHIP made substantial strides in developing clear goals Partner and partner roles, and despite external constraints, to develop the trust and commitment needed to work in an Partnership interdependent manner. Future collaborative MNCH partnerships should pursue a shared understanding of these four principles as early and often as possible to ensure success. © 2015 Published by Elsevier Ireland Ltd. on behalf of International Federation of Gynecology and Obstetrics. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

1. Background emerge. Some development partnerships remain, intentionally or not, at a low level of coordination, ensuring that constituents’ activities are Partnerships have been described as “the development approach of synchronized but not necessarily linked. Others proceed to a higher our time” [1]. Indeed, the sheer number of partnerships undertaken in level of cooperation, or integrated planning to achieve mutually bene- the development field in recent decades signifies overwhelming confi- ficial objectives. Still others strive for full collaboration, pooling re- dence in the comparative advantage of these joint ventures. In global sources to undertake common activities, with joint problem-solving health, the Global Alliance for Vaccines and Immunization, the U.S. and decision-making at every turn. President’s Emergency Plan for AIDS Relief, and the Global Fund to In the global maternal, newborn, and child health (MNCH) field, the Fight AIDS, Tuberculosis and Malaria are just a few of the most well- Partnership for Maternal, Newborn and Child Health (PMNCH) most known examples of development-focused partnerships [2–6]. closely resembles a coordinated partnership, harmonizing a research These “purposive strategic relationships” among governments, insti- and advocacy platform to guide independent endeavors by its constitu- tutions, private entities, and individuals have understandable appeal in ents in these interlocking technical areas [8–12]. The Health 4+ part- the pursuit of global health goals; many of these goals, such as the re- nership, by contrast, appears to function as a cooperative partnership, duction of maternal and newborn mortality, cannot be achieved with- aligning into one work plan the country-specific efforts of all relevant out comprehensive technical, contextual, and administrative expertise United Nations agencies in pursuit of Millennium Development Goals [7]. Partnerships are known by a range of labels—alliances, networks, (MDGs) 4 and 5 [13]. This partnership stops short of actually executing coalitions, and associations—but in almost all cases involve multiple in- activities under a single umbrella agency. dependent organizations seeking to accomplish complex initiatives that In this commentary, we assess the third type of partnership iden- would otherwise be unattainable by a single entity [7]. tified from the literature, a collaborative partnership funded by While there is ample literature, especially from the business com- USAID that we have managed: the Maternal and Child Health Integrated munity, on partnerships, we found no clear attempt to distinguish Program (MCHIP). There are many published external evaluations among their different structures, even though some distinctions readily of partnerships that determine whether such entities were able to meet their objectives, but these evaluations are often not intended—or able—to offer insight into how or why objectives were or were not met [14,15]. Given the scarce resources and ambitious agenda in ⁎ Corresponding author at: Maternal and Child Survival Program, 1776 Massachusetts Avenue, Suite 300, Washington, DC 20036, USA. Tel.: +1 202 835 3100. MNCH in particular, we aim to contribute our own assessment of our E-mail address: [email protected] (K. Agarwal). partnership’s internal dynamics and provide guidance to future http://dx.doi.org/10.1016/j.ijgo.2015.04.001 0020-7292/© 2015 Published by Elsevier Ireland Ltd. on behalf of International Federation of Gynecology and Obstetrics. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). S12 K. Agarwal et al. / International Journal of Gynecology and Obstetrics 130 (2015) S11–S16 partnerships on ways to build better collaborative partnerships for clear center of oversight and decision-making authority, the “convener,” MNCH. While this paper may be focused on how a USAID-funded to allocate roles and diffuse—or ideally prevent—tensions [18]. project brought together so many partners who worked well togeth- “Trust” and mutual interdependence rarely exist at the outset of a er, the issues and recommendations should be valid for most collabora- partnership; instead, they must be built over time through high- tive partnerships. quality communication among partners [1,3,17].Ataminimum,apart- nership should develop a set of communication norms that encourages widespread sharing and dissemination of accomplishments, obstacles, 2. A conceptual framework for evaluating MNCH partnerships next steps, and special acknowledgement of individual partner contributions [16,18]. Trust often arises when the convener is “capable of The large body of social science literature about partnerships reveals stepping back to allow others to come forward to fill needed roles” certain fundamental characteristics of a solid partnership: strong [17]. Trust can also be built through transparency in decision-making, management; well-defined goals; carefully-considered membership with the convener communicating clearly across the partnership. and representation; open lines of communication; core processes for “Commitment” is generally presumed when a partner elects to join a monitoring and evaluation; and strategies to overcome obstacles and partnership, but partners inevitably differ in their allocations of time, re- adjust policies and tactics when necessitated by the external environ- sources, influence, and priorities [19]. Recognizing these differences and ment [3,7,16,17]. Conversely, there are some commonly cited reasons continually reaffirming partner interest and investment allows other for partnership failure: among others, poor planning, inadequate orga- partners to adjust their expectations of the partnership over time [2]. nization, lack of mutual dependence, competing interests, battles over Every partnership also needs a well-coordinated exit strategy, a plan authority, and micromanagement or lack of management, especially of for its dissolution and eventual transfer of responsibilities to other resources [2,7]. stakeholders [17]. Drawing from this literature but seeking to distill a few easily- discernible measures of collaborative partnership success, we isolated 3. Application of successful collaborative partnership principles the following four principles: clear goals; clear roles; trust; and commit- to MCHIP ment (Fig. 1). The four identified principles loosely correspond to USAID’sowndefinition of “partnership” as “an association between MCHIP began in 2008 as USAID’s flagship program for MNCH, a US USAID, its partners and customers based on mutual respect, comple- $600 million Leader with Associates Cooperative Agreement “designed mentary strengths, and shared commitment to achieve mutually agreed to support the introduction, scale-up and further development of upon objectives” [8]. high-impact MNCH interventions” within interested countries in “Clear goals” refers to partners’ collective and explicit agreement on which USAID works [20]. MCHIP also subsumed the functions of five shared objectives, often through the creation of formal partnership pre-existing USAID projects relating to maternal and neonatal health: agreements, or unified strategic and operating plans [1,17,18]. Main- ACCESS, BASICS III, Immunization BASICS, POPPHI, and CSTS+ [20].In taining clear goals for the duration of the partnership requires proce- doing so, it joined teams from the following implementing organiza- dures to monitor progress and a mechanism for renegotiation and tions, each of which had its own extensive international reach: Jhpiego mid-course adjustment [17]. Corporation (Jhpiego), as the prime partner; John Snow, Inc.; Johns “Clear roles” emphasizes the place of each individual partner in rela- Hopkins University/Institute for International Programs; ICF Macro, tion to the whole and creates a shared vision of success. The most Inc.; Program for Appropriate Technology in Health (PATH); Save the successful partnerships articulate the strategic intent behind each part- Children; Broad Branch Associates; and Population Services Internation- ner’s participation in the alliance and consider the unique competencies al (PSI) [20]. As far as we know, MCHIP was the largest single financial that each partner brings [2,3,17]. While multiple partners can assume commitment ever made by USAID for maternal and newborn health leadership positions for different, selective aspects of work, it is difficult and child survival and a key investment in USAID’s expanding portfolio to entirely avoid some overlap and duplication of scope. There must be a in those areas, which had increased in size from US $361 million in 2001

Fig. 1. Principles of successful collaborative partnerships. K. Agarwal et al. / International Journal of Gynecology and Obstetrics 130 (2015) S11–S16 S13 to US $522 million in 2008 [21]. Because USAID provided the impetus central level and by each of the partnership’scountryoffices, some of- and funding for MCHIP, we do not consider USAID itself to be a partner fices were more conformant than others based on their history and and instead analyze the dynamics among the eight constituent organi- leadership as well as their USAID country mission’s understanding of zations. To analyze the role of USAID—or any donor—in the effectiveness MCHIP’sgoals. of a partnership that it funds would need a separate commentary. MCHIP’s management structure developed slowly over the first year, as partners merged their unique technical work streams, of fice cultures, 3.1. Clear goals and administrative systems into one unified program. Jhpiego, as the prime partner, assumed responsibility for strategic leadership but dele- MCHIP used “integration” in its name, but the meaning of that gated most other authority to an Executive Management Team (EMT) term—and, by extension, MCHIP’s overarching goals—was not clear at composed of staff with demonstrated experience managing complex the partnership’s outset. Some thought that MCHIP merely integrated programs from several “core” partners. For example, decisions about the five pre-existing projects to provide a “one-stop” source for which partners would be involved in a given country program were USAID-funded technical assistance across MNCH technical areas, while made by the EMT based on the operating principles and the technical others expected MCHIP to introduce integrated programming ap- focus for each country. Strategic considerations such as cost efficiency proaches that would bridge the typical technical area silos within and the strength of each partner’s in-country presence and mission MNCH [22]. The partnership therefore employed a management con- preference were considered as well. The project director decided the sultant who was not affiliated with any partner to facilitate extensive partner lead when the EMT could not reach a consensus. Box 2 summa- discussions among the partners and with the donor about expectations rizes the routine practices for MCHIP’s EMT meetings and demonstrates and translate those discussions into a strategic plan and an initial set of the way in which the joint management undertaken by the EMT proved “partnership principles.” While conflicting opinions about the part- particularly useful in identifying issues that needed to be addressed nership’s goals persisted throughout its duration, these partnership among the partnership as a whole and reviewing whether the partner- principles provided a strong foundation for future discussions. The re- ship’s collective efforts were meeting its goals. Initially, the partnership quirements of the program as outlined by the donor reflected the had also planned a “Partnership Management Team” that would help need for one organization to serve as the “prime” partner, or contractual execute administrative and financial decisions made by the EMT, but counterparty. Jhpiego, as the prime partner, therefore established this team did not meet consistently to support implementation of “teaming agreements” with each of the other partners that provided these decisions. platforms for negotiation and commitment. In addition, during the As a check on the EMT’s internal assessments, MCHIP continued to start-up phase, the partners debated the pros and cons of various employ the same management consultant to track the partnership’s partnership models and absorbed as many best practices as possible progress toward its strategic objectives and advise all partners during from these models into the teaming agreements. Based on these external evaluations. MCHIP also commissioned an internal mid-term teaming agreements and discussions about the partnership principles, review by the management consultant to focus on areas of implementa- anumberofmorepragmatic“operating principles” emerged, the most tion that could be strengthened and solicit opinions widely from all fundamental of which are listed in Box 1. While these operating prin- partners as well as from the donor, USAID [22]. The early introduction ciples were supposed to be adopted and implemented both at the and continuity of this trusted independent consultant might have contributed to the productive relationship between the partners and its evaluator. As a result of the mid-term review, MCHIP modified its operating principles to align with the review’s findings that MCHIP needed better internal communication and knowledge sharing. For example, Box 1 the review highlighted the low reliance by partners on MCHIP’s external Operating principles of the Maternal and Child Health Integrated website or on the internal, web-based “Sharepoint” site that was Program (MCHIP). established to serve as a central repository for documents and as a space for collaborative discussion. Improvements were made to re- 1. Decentralize management of country activities to the greatest spond to this shortcoming as this represents an important platform to extent possible to build in-country capacity and champions, communicate and reinforce clear goals; however, access remains prob- maximize efficiency, ensure rapid start-up, and contain costs. lematic for country offices. Still, the review provided the opportunity to 2. Maintain maximum responsiveness to the donor while being isolate common problems and devise solutions. sensitive to the planning needs of each partner. 3. Assign lead responsibility to one partner per country, based primarily on the scope of work and the technical areas identified, but with consideration of in-country capacity and cost Box 2 efficiency. 4. To the extent practicable, establish an MCHIP country office Routine practices for Maternal and Child Health Integrated Program where all MCHIP staff, regardless of partner affiliation, will be (MCHIP) Executive Management Team meetings. co-located. The lead partner in each country will be responsible for overall management of the office. • Occur weekly with few exceptions. 5. Create and follow branding guidelines. • Chaired by the Deputy Director, who is not affiliated with prime 6. Allow each partner to manage and budget its own in-country partner. activities if it has sufficient staff and capability, except when • Review and respond to questions elevated for management there are administrative requirements for or substantial efficien- decision. cies achieved by centralization. • Focus on project-wide issues, including managing the relation- 7. Encourage partners to hire their own technical staff to support ship with the donor. their area of expertise, but recruit cross-cutting technical • Invite presentations by partnership staff of major successes and staff, such as monitoring and evaluation officers, through the challenges. lead partner in each country, balancing the hires equitably • Notes prepared with follow-up actions that are revisited at future among the partners. meetings. S14 K. Agarwal et al. / International Journal of Gynecology and Obstetrics 130 (2015) S11–S16

3.2. Clear roles 3.3. Trust

MCHIP united eight implementing organizations that are all leaders Most of the MCHIP partners had never formally collaborated before, in the field of MNCH and possess a number of similar competencies. and some had directly competed against each other for prior USAID Fig. 2 represents the working relationship among these partner organi- awards, creating understandable apprehension about working together zations that resulted from internal deliberation about organizational and with the common donor. The EMT—and the diplomatic personali- strengths and technical leadership capabilities. Recognizing that issues ties appointed to it—was instrumental in providing a forum to build of “role creep” and role confusion would be a recurring problem, trust among partners, sharing best practices and expressing concerns MCHIP established a Corporate Representative Team (CRT) at its incep- about competing organizational priorities and management strategies. tion that was comprised of one senior corporate representative for each One comment made during MCHIP’s internal mid-term review captured partner. The CRT met quarterly for the first year and then, as roles be- this surprising openness: came better defined, twice a year thereafter. Jhpiego’s senior leadership also met separately with each other partner once a year to ensure that “I think that the culture of communication that we've established all were satisfied with their scope of work. In these meetings, and in between partners on this project, while it still has some occasionally other informal opportunities to comment, partners gave generally pos- rocky places, is actually pretty amazing in that none of the main core itive reactions, even those whose roles became narrower than originally partners seem to be shy about voicing any concerns that they have, conceived and were expected to have concerns. A more routine partner and the group commits to resolving the issues” [22]. assessment, through the CRT or otherwise, would have been useful. The EMT served as the partnership’s center of oversight and “con- The culture of open communication within the EMT and at the part- vener,” and although it did not include all partners, representation ner leadership level proved much easier to foster among partnership reflected those with cross-cutting functions across the program. While staff who were co-located in the same office as the EMT than among members were employed by different organizations, they could only other staff who were located remotely or in country offices. Face-to- perform their functions successfully if partners came together in a co- face meetings, including weekly EMT meetings and monthly staff herent, meaningful way. Accordingly, it was felt that this group was rep- meetings in which the EMT reported major decisions and lead partners resentative enough to make decisions fairly and inclusively. In fact, the presented their successes, were extremely conducive to generating donor sometimes engaged in direct communications with the EMT, in- this solidarity, but it was also important, and often difficult, to have stead of with the partner leading the relevant scope of work, because face-to-face interactions outside of formal meeting structures. High- it was a more efficient way to reach all partners about management quality communication requires an understanding of each other’s issues. There was one role, financial control, that Jhpiego was not able motivations and typical reactions that cannot easily be attained to share or delegate, largely because the partnership was not an without reading body language and gaining familiarity with per- independent legal entity. Some partners expressed frustration with sonalities in informal settings. Even in countries where there was an this outcome because it reduced budgetary transparency and created MCHIP office and co-location was encouraged, understanding how to significant administrative hurdles caused by inserting one partner’s fi- reconcile new ways of doing business under MCHIP within a broader nancial procedures into another partner’s operational decisions. Jhpiego competitive bilateral funding environment posed a challenge to build- addressed these challenges by developing templates to share budget ing trust. inputs between partners, leaving budgetary decisions to individual To anticipate its communications challenges, MCHIP made the project teams, and making funds available to partners before final ap- creation of a shared identity, or MCHIP “brand,” a top priority for the proval for projects has been obtained. Although the perceived lack of partnership. MCHIP hired a dedicated communications team leader transparency was a consistent challenge for the partnership, the roles who was selected by a recruitment team comprised of multiple themselves were shared to the extent possible under the terms of the partners, who was experienced in US government relations, and who program’s agreement. could provide support to each of MCHIP’s technical areas and country

Fig. 2. Partner roles within the Maternal and Child Health Integrated Program (MCHIP). K. Agarwal et al. / International Journal of Gynecology and Obstetrics 130 (2015) S11–S16 S15 programs. The communications leader reported directly to the EMT 4. Challenges and best practices for consideration by future and drew support from the communication teams within each of MNCH partnerships the partner organizations and respective country offices. Through the creation of branded communications tools such as web pages, This commentary offers an insider’s perspective by three senior document templates, and publication services, MCHIP enabled staff leaders of a collaborative MNCH partnership on four key dimensions to present themselves as affiliated with MCHIP. The communications (clear goals, clear roles, trust, and commitment), by examining the au- leader also worked with USAID to appropriately emphasize the donor’s thors’ own experience as managers of MCHIP. This perspective is neces- role and its connection with the partnership, as well as with each part- sarily subjective and relative to the management positions held, and is ner organization. Coordinating the partnership’s branding among all not intended to be representative of all those involved in the partner- partners and with USAID, and having the EMT reinforce use of the ship, especially country office colleagues who, as described above, brand through active outreach to technical and field teams, helped en- were inevitably distanced from many decisions. At the same time, we sure the quality of MCHIP’s work and recognition of each partner’scon- expect that this internal assessment of our partnership has revealed tribution, as though MCHIP were its own organization. It was also more, and different, reasons for the partnership’s strengths and weak- important to hire new staff full-time for the project itself, instead of nesses than any external evaluation might find. sharing time with the home organization, to build acceptance of the By its design, and through early and intentional discussion about op- MCHIP brand. erating principles and technical capabilities, MCHIP was well-equipped to set clear goals and roles; it took much more time and effort to develop trust and ensure commitment among all partners. Inherently, a global 3.4. Commitment program has a time-bound dimension, clearly different from the organizations that come together to implement such programs. Given the in- Because MCHIP was an unprecedented global program in scope and trinsically transitional nature of MCHIP, the program enjoyed a great size, partners with broader competencies and depth of experience in a reputation and recognition for the technical excellence it provided to particular country ultimately received a greater proportion of award over 50 countries. Even then, the partnership sometimes struggled in funding as there were existing technical and administrative platforms ensuring that its operating principles were implemented similarly in on which to build. Partners with more defined, narrower scopes were each country office, and in managing the “role creep” that prevented always less involved in day-to-day operations. In some cases, roles some staff from having clear and reasonable expectations for their narrowed over time, leading to gradual changes in levels of commit- work. While the EMT tried to maintain transparency in its decision- ment and resources. However, all partners continued to participate in making and institute a common identity throughout the program, it the CRT, and technical staff and program staff across the program was challenging to disseminate information to country offices and remained fully committed to the success of MCHIP, as demonstrated gain widespread understanding within a short time frame of a global by their continued and active engagement in program design and re- program. Because of the size of the partnership, its administrative as- views, internal and external meetings, and responses to USAID requests. pects were immense, and in retrospect, stronger investment in a Part- Partners also showed their shared sense of credit and responsibility by nership Management Team to carry out such functions and improve copying each other on correspondence and involving them in discus- collaboration might have been worthwhile. sions when information that was important to the partnership had On the other hand, in its relatively short, six-year duration, MCHIP been directed to only one organization. was able to combine and administer an unprecedented set of technical The significant amount of funding received by the project in resources for integrated MNCH programming that had previously part reflected the donor’sconfidence in and tremendous support been dispersed among multiple implementing organizations. In the of the partnership to take on an ambitious scope of work. The breadth spirit of our discipline’s commitment to evidence-based learning, we and reach of the program also meant there were a large number have summarized from this experience a set of best practices (Box 3) of USAID managers involved in different aspects of project activities. and a checklist of questions that we recommend all MNCH partnerships Inevitably, this sometimes led to lack of clarity about expectations consider at their inception (Box 4). and overburdened staff with competing demands. These situations We believe that one of the best things that can come out of ultimately reinforced the need for staff to come together as a part- our MCHIP partnership is for each partner to be viewed as a good nership to seek clarity when these situations arose. A clear and shared partner and as a good leader of future partnerships. Partnerships in commitment was particularly critical among the country-level staff who were executing project activities. The most successful country Box 3 programs co-located staff in the same office and hired new staff specif- Best practices for maternal, newborn, and child health (MNCH) ically for MCHIP, and they were managed by strong leadership that partnerships. could operate with autonomy as well as accountability within the wider project. • Like MCHIP’s starting goals, MCHIP’s exit strategy was hardwired Engage an independent management consultant to establish into the program’s original design. In its final year, MCHIP wound goals at start-up, and consider keeping the same consultant for down its operations with a common “close-out” reporting mecha- ongoing goal recalibration. • nism that documented accomplishments as overall project accom- Establish written operating principles, if not a full partnering plishments, not as those of any individual partner. Because technical agreement, that each lead individual continually reaffirms and teams led by one partner often reached into the partnership to engage acts upon. • staff from multiple organizations to achieve project goals, the close- Distribute ownership of partnership components among core out reports reflect the commitment of multiple organizations working partners and encourage intrapartnership evaluations. • together to achieve those goals. MCHIP’s partner organizations have af- Create a multipartner management team that meets regularly firmed their continued commitment to working with one another by and makes decisions that are accessible to all partnership staff. • seeking to implement together —and with several new partners—the Conduct periodic check-in evaluations through surveys to identi- next global USAID program for MNCH: the Maternal and Child Survival fy major bottlenecks preventing effective functioning. • Program. As a result of the partnership’s history, this new program To the extent possible, encourage co-location and common com- should be in a better position to clarify commitments and expectations munication norms that engender a shared partnership identity. at the beginning. S16 K. Agarwal et al. / International Journal of Gynecology and Obstetrics 130 (2015) S11–S16

Box 4 Conﬂict of interest Checklist for a successful maternal, newborn, and child health (MNCH) partnership. The authors have no conﬂicts of interest.

1. Are clear goals and objectives for the partnership established? 2. Is it clear how the work of the partnership will be evaluated? References 3. Is there a clear role for each partner? [1] Rein M, Stottsource L. Working together: critical perspectives on six cross-sector 4. Does each partner understand the role of the other partners? partnership in Southern Africa. J Bus Ethics 2009;90(1):79–89. 5. Are there written principles that guide relationships within the [2] Koza M, Lewin A. Managing partnerships and strategic alliances: raising the odds of – partnership? success. 2000;18(2):146 51. [3] Lob-Levyt J. Contribution of the GAVI Alliance to improving health and reducing 6. Are there written operating guidelines that outline how the poverty. Philos Trans R Soc Lond B Biol Sci 2011;366(1579):2743–7. work of the partnership will be implemented? [4] Sturchio J, Cohen G. Improving labs to strengthening supply chains: how PEPFAR's 7. Is there a clear process for referring problems that cannot public-private partnerships achieved ambitious goals. Health Aff 2012;31(7): 1450–8. swiftly be resolved by the central members of the partnership? [5] Simonds J, Cohen G. How PEPFAR's public-private partnerships achieved ambitious 8. Has the partnership considered hiring an external manage- goals, from improving labs to strengthening supply chains. Health Aff 2012;31(7): ment consultant to advise on different aspects of partnership 1450–8. [6] The Global Fund. The Global Fund’s Partnership Strategy. Washington DC: The work? Global Fund to Fight AIDS, Tuberculosis and Malaria. http://www.theglobalfund. 9. Is it clear where the “buck stops” within the partnership? org/documents/core/strategies/Core_Partnership_Strategy_en. Accessed May 14, 10. Is there a forum where partners can be heard on a regular 2014. basis? [7] Mohr J, Spekman R. Characteristics of partnership success: partnership attributes, communication behavior, and conflict resolution techniques. Strateg Manag J 11. Is there an information sharing system that enables all part- 1994;15(2):135–52. ners to have equitable access to information? [8] USAID. Automated Directives System (ADS) Chapter 101: Agency Programs and fi 12. Does the partnership have its own identity or brand that is dis- Functions. Published 2013. http://www.usaid.gov/sites/default/ les/documents/ 1868/101.pdf. Accessed February 12, 2015. tinguishable from the identity or brand of each individual [9] Bustreo F, Requejo JH, Merialdi M, Presem C, Songane F. From safe motherhood, partner? newborn, and child survival partnerships to the continuum of care and accountabil- – 13. How does the partnership ensure that partners universally feel ity: moving fast forward to 2015. Int J Gynecol Obstet 2012;119(1):S6 8. [10] Shiffman J, Smith S. Generation of political priority for global health initiatives: a apartofthisidentity? framework and case study of maternal mortality. Lancet 2007;370(9595):1370–9. 14. Is there a communication function within the partnership that [11] Freedman LP, Graham WJ, Brazier E, Smith JM, Ensor T, Fauveau V, et al. Practical les- is viewed as truly representing the partnership and not favor- sons from global safe motherhood initiatives: time for a new focus on implementation. Lancet 2007;370(9595):1383–91. ing any one member? [12] Requejo JH, Toure K, Bhutta Z, Katz I, Zaidi S, de Francisco A. Regional Collaborations 15. Is there a way for individual partners to represent the partner- as a Way Forward for Maternal, Newborn and Child Health: The South Asian ship while at the same time maintaining their organizational Healthcare Professional Workshop. J Health Popul Nutr 2010;5:417–23. [13] Sines E, Tinker A, Ruben J. The Maternal–Newborn–Child Health Continuum of Care: identity? A Collective Effort to Save Lives. Washington, DC: Population Reference Bureau; 16. Does the convener of the partnership demonstrate trust in 2006. www.prb.org/pdf06/SNL-ContOfCare_Eng.pdf. Accessed September 5, 2014. other partners, as shown by: [14] Health 4+: working with countries to improve maternal and newborn health. a. a transparent problem-solving mechanism? http://wcaro.unfpa.org/webdav/site/wcaro/users/wcaroadmin/public/H4%20Pamphlet. pdf; 2010. Accessed March 26, 2015. b. a transparent decision-making process? [15] HLSP. External evaluation of the Partnership for Maternal, Newborn and Child c. sharing management and implementation responsibility? Health. London: HLSP; 2008.http://www.who.int/pmnch/members/meetings/ d. allowing partners to speak freely about shortcomings of doc10.pdf. Accessed September 5, 2014. [16] Bill and Melinda Gates Foundation. Global Health Partnerships: assessing country other partners, including the convener? consequences. McKinsey and Partnership. Published 2005. http://www.who.int/ 17. Does each partner feel that being a member of the partnership healthsystems/gf16.pdf. will enhance its own organizational goals as well as the goals [17] Rosenberg ML, Hayes ES, McIntyre MH, Neill N. Real Collaboration: What it Takes for Global Health to Succeed. Berkeley, CA: University of California Press; 2010 New of the partnership? York: Milbank Memorial Fund. 18. Are the strengths of each partner fully utilized? [18] Aylward RB, Acharya A, England S, Agocs M, Linkins J. Global health goals: lessons from the worldwide effort to eradicate poliomyelitis. Lancet 2003;362(9387): 909–14. [19] Powell DL, Powell DL, Gilliss CL, Hewitt HH, Flint EP. Application of a partnership model for transformative and sustainable international development. Public Health MNCH appear to be here to stay, so we should seek to improve their Nurs 2010;27(1):54–70. chances of success. [20] USAID. Users Guide to USAID/Washington Health Programs. Washington, DC: United States Agency for International Development; 2013. http://www.usaid.gov/ sites/default/files/documents/1864/UG2013.pdf. Accessed September 5, 2014. Acknowledgments [21] Salaam-Blyther T. USAID Global Health Programs: FY2001-FY2012 Request. Washington DC: Congressional Research Service; 2011. [22] Leah M. MCHIP mid-term self-review. Washington, DC: Management Assistance The authors acknowledge all partner organizations involved in Group; 2011. MCHIP as well as MCHIP’s funding source, USAID. The authors thank Vandana Tripathi for her literature review and Rehana Gubin and Karen Kirk for their editorial support. International Journal of Gynecology and Obstetrics 130 (2015) S17–S24

Contents lists available at ScienceDirect

International Journal of Gynecology and Obstetrics

journal homepage: www.elsevier.com/locate/ijgo

SUPPLEMENT ARTICLE Implementation of the Standards-Based Management and Recognition approach to quality improvement in maternal, newborn, and child health programs in low-resource countries

Edgar Necochea a,⁎,VandanaTripathib,Young-MiKima,NabeelAkrama, Yolande Hyjazi c,MariadaLuzVazd, Emmanuel Otolorin e,TsiguePleaha, Tambudzai Rashidi f, Dustan Bishanga g a Jhpiego, Baltimore, MD, USA b Johns Hopkins Bloomberg School of Public Health, Baltimore, MD, USA c Jhpiego, Conakry, Guinea d Jhpiego, Maputo, Mozambique e Jhpiego, Abuja, Nigeria f Jhpiego, Lilongwe, Malawi g Jhpiego, Dar es Salaam, Tanzania article info abstract

Keywords: The Standards-Based Management and Recognition (SBM-R) approach to quality improvement was developed by Health systems strengthening Jhpiego to respond to common challenges faced by health systems in low-resource settings, including poor pre- Maternal, newborn, and child health service education, lack of resources for conventional supervisory models, and weak health information systems. Performance assessments Since its introduction in Brazil in 1997, SBM-R has been implemented in approximately 30 countries and continues Quality improvement expanding to new places and service delivery areas. The present article: (1) describes key steps in the SBM-R Standards-Based Management and Recognition methodology focusing on provider performance assessment using evidence-based standards; and (2) presents examples of improvements in provider performance in maternal, newborn, and child health care following SBM-R implementation derived from routine program data, quasi-experimental evaluations, and in-depth case studies. SBM-R incorporates evidence-based methods that are known to have positive effects on healthcare quality, including audit and feedback, educational outreach visits, and checklist usage; however, further rigorous research is needed to document the population-level impacts of the SBM-R approach. © 2015 Published by Elsevier Ireland Ltd. on behalf of International Federation of Gynecology and Obstetrics. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

1. Background has been adopted by facilities and facility networks with large catchment populations, creating the potential to affect quality of care on a rel- Since its establishment 40 years ago, Jhpiego has worked to atively large scale. strengthen educational approaches and systems in reproductive, maternal, and primary care in low-resource countries. In the last 15 years, 2. The SBM-R methodology Jhpiego’s scope has expanded beyond the preparation of competent health providers to improving the performance of these workers to pro- SBM-R pertains to the group of quality approaches that pursues basic duce effective, quality services for clients. To contribute to this quality standardization of services as an initial step in quality. It starts with the improvement aim, Jhpiego developed an intuitive but innovative ap- provision of recommended standards of care (including inputs and pro- proach called Standards-Based Management and Recognition (SBM-R) cesses) as a point of reference for facility staff and managers to identify that was first implemented in Brazil in 1997. Here, we describe key gaps in performance. Although most quality improvement approaches steps in the SBM-R methodology and provide illustrative examples of share common elements (e.g. the Deming improvement cycle: plan- improvements in provider performance in maternal, newborn, and do-check-act), they also have differences. Some approaches focus on child health (MNCH) care that have been documented to help demon- problem-solving and continuous improvement, whereas others pro- strate the impact of the SBM-R approach on service delivery. SBM-R mote the standardization of processes to achieve regularity in service provision [1]. SBM-R, as well as other commonly used approaches such as accreditation, pertains to the latter group of quality approaches. ⁎ Corresponding author at: Jhpiego, 1615 Thames Street, Baltimore, MD 21231, USA. Tel.: +1 410 537 1895. SBM-R consists of systematic utilization of detailed performance stan- E-mail address: [email protected] (E. Necochea). dards for rapid and repeated assessments of health facilities, including http://dx.doi.org/10.1016/j.ijgo.2015.04.003 0020-7292/© 2015 Published by Elsevier Ireland Ltd. on behalf of International Federation of Gynecology and Obstetrics. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). S18 E. Necochea et al. / International Journal of Gynecology and Obstetrics 130 (2015) S17–S24 both clinical and support systems; identification of gaps in compliance with these standards; implementation of corrective interventions; and rewarding of achievements through recognition mechanisms. The SBM-R model was designed to confront key realities in the health systems of many resource-constrained settings. In these settings, pre- service education curricula may not cover the full range of MNCH services, so quality assurance and supervision approaches must simultaneously function as in-service training, not only telling health workers what to do, but how to do it [2–5]. Conventional supervisory models using external supervisors also operate intermittently at best, owing to challenges such as understaffing in management cadres and lack of transportation resources [3,6]. Likewise, a “push” approach to the allocation and distribution of the critical resources needed for the efficient and effective provision of services is often inadequate, so a “pull” effect is also needed from health workers who are already knowledgeable about their particular needs [7,8]. Health management information systems (HMIS) in these settings are frequently too weak or disjointed to collect meaningful indicators with the periodicity or specificity needed for targeted quality improvement activities [9,10]. Finally, as in all settings, health workers are more likely to accomplish their tasks satisfactorily if they are engaged Fig. 1. The Standards-Based Management and Recognition (SBM-R process). Reprinted and motivated; however, many conventional supervisory approaches with permission from Necochea and Bossemeyer [14]. are focused on the attainment of service targets without addressing or supportingthemoraleofhealthprovidersthemselves[11,12]. at the facility level (sometimes including representative users of health Box 1 and Fig. 1 describe the key steps of the SBM-R methodology, services) to enable the identification of performance gaps, i.e. lack of which begins with the development of evidence-based and informative compliance with standards. Because SBM-R assessment tools distill performance standards. Unlike many other quality improvement ap- and summarize standards for health services, they can also serve as proaches, a cornerstone of SBM-R is not only providing a methodology job aids and support for on-the-job learning. Following an assessment, but also generating the content that guides the improvement process, local health teams and managers can analyze the causes of gaps and de- embodied in the performance standards. These standards incorporate velop action plans to implement appropriate interventions to correct evidence-based practices from a technical perspective as well as an em- them, considering potential contributing factors such as: lack of knowl- phasis on care that respects clients’ rights, dignity, and cultural prefer- edge and skills; inadequate existence of equipment, supplies, and other ences. SBM-R standards address “pathways of care,” or the key steps for resources or policies; and lack of motivation. While each local initiative the provision of distinct types of health services, in a holistic manner in- will have different improvement targets, the SBM-R methodology gen- stead of focusing on interventions in isolation. As a result, the standards erally encourages facilities to achieve 80% or better compliance with the encourage support systems that aim to create an organizational culture locality’s established performance standards. of quality and ultimately make improvements more sustainable. Typical- The motivational element of SBM-R, or recognition of achievements, ly, programs implementing SBM-R identify between 5–25 performance is the final but defining attribute of SBM-R, adapted from business man- standards per health service area, each with specific tasks that translate agement theory [13]. Local health teams are encouraged to begin with into verification criteria. The number of verification criteria per standard simple interventions in order to achieve early results that improve mo- hasrangedfromasfewastwotomorethan35.Box 2 provides a sample rale and create momentum for change, gradually acquiring or strength- SBM-R performance standard and its verification criteria. ening change management skills to address more complex gaps. These SBM-R employs a systematic change management process to achieve desired improvements in quality of care. Once performance standards are identified, they are then used in performance assessment Box 2 tools that can be scored to express performance in quantitative terms. These tools can be used for self, peer, internal, and external assessments Sample Standards-Based Management and Recognition (SBM-R) standard and its verification criteria.

Box 1 Area: Pregnancy care Key steps in the Standards-Based Management and Recognition (SBM- Performance standard: The facility conducts a routine rapid R) methodology. assessment of pregnant women Verification criteria: • Set standards of performance. Observe in the reception area or waiting room if the person who • Organize these standards into a checklist tool that can be receives the pregnant woman: used for performance assessment. • Implement the standards through an initial assessment, mea- • Asks if she has or has had: suring performance and pinpointing gaps in compliance with ○ Vaginal bleeding standards. ○ Headache or visual changes • Develop action plans to correct these gaps by addressing all ○ Breathing difficulty performance factors in a comprehensive way. ○ Severe abdominal pain • Measure progress through repeated periodic assessments, ○ Fever mainly internal, at each participating facility using the same • Immediately notifies the health provider if any of these checklist tool. conditions is present. • Recognize achievements, both fulfilling the action plans and improving compliance with performance standards. Source: Necochea and Bossemeyer [14]. E. Necochea et al. / International Journal of Gynecology and Obstetrics 130 (2015) S17–S24 S19 teams are also oriented to work in networks and share experiences with health system levels across a wide range of technical areas, including other participating facilities. The sense of personal achievement, MNCH, HIV/AIDS, and primary care. The remainder of this section illus- growth, and responsibility in a job as well as gaining recognition from trates specific MNCH performance assessment results produced in a stakeholders (e.g. clients, communities, institutional authorities) for sample of countries and technical areas and evaluates the evidence perceptible improvements are characteristic of the SBM-R process and that SBM-R contributes to improving quality of care at scale. SBM-R as- help make the process more acceptable to providers. SBM-R empower- sessments and routine program data (e.g. health service statistics) con- ment is achieved by the perception of improvements and also by having sistently show marked improvements over relatively short time periods standards that reassure providers/managers on evidence-based operat- of six months to one year. ing procedures. Knowing and having more control and responsibility for changes enhances empowerment and thus motivation. Partial improve- 3.2. Improving adherence to performance standards ments are rewarded during the process through positive feedback and social recognition (e.g. ceremonies or symbolic rewards). The achieve- In Guatemala, a quality improvement initiative applied the SBM-R ap- ment of compliance with standards by a facility is acknowledged proach to improve maternal and newborn health services in 20 district through broader recognition mechanisms that involve institutional au- hospitals, 42 health centers, and 52 health posts (personal correspon- thorities and the community and can lead to certification as a high dence, Oscar Cordon, Maternal and Neonatal Health Project Director, performing unit or center of excellence in health care. Even financial re- 2003). Fig. 2 shows substantial improvements in average facility perfor- wards have been included in some SBM-R initiatives. Each facility keeps mance relative to baseline across a range of SBM-R standards at 11 health graphs and other documentation about its performance at each point of centers that participated in an SBM-R evaluation. While no standard assessment, enabling the rapid identification of additional areas that reached the 80% target, performance increased from 2.5-fold to five- need attention to be addressed in the next action plan. A SBM-R “field fold within a year in basic maternal care; infection prevention; informa- guide” provides further details about the many tools that have been de- tion, education, and communication; material resources; and manage- veloped to implement this methodology [14]. ment capacity. In these facilities, the SBM-R performance assessment results were used to identify the technical areas needing more intensive 2.1. SBM-R focal area: Performance data collection support. For instance, despite a five-fold increase from baseline, the provision of basic maternal care remained furthest from performance tar- While the SBM-R process involves service data collection as well as gets, leading to greater investment of resources in provider training and performance assessments, the present article focuses on results and other knowledge and skills interventions. While Jhpiego no longer has a published findings related to SBM-R performance assessments. Service presence in Guatemala, the SBM-R approach has continued to be imple- data can be analyzed to determine the impact of SBM-R and to identify mented through the support of cooperating agencies. causes of performance gaps, as described in more detail in another publication [14]. As a part of the SBM-R process, all participating facilities conduct periodic rapid assessments (usually every three to six months) 3.3. Improving coverage with high-impact interventions and using tools based on the SBM-R standards. These are rapid assessments patient outcomes that present only an approximate image of true facility performance on any given occasion. But, because they can be conducted several times a Improvements in patient coverage of high-impact MNCH interven- year at a relatively low cost, over time they provide a comprehensive tions can begin to be associated with achievement of performance stan- picture of provider performance and are useful to guide ongoing dards measured through periodic assessments when coverage is decision-making by facility and service managers. These assessments measured through facility routine service data collection. Fig. 3 com- can be performed internally by facility staff, external supervisors, or pares achievement of defined child health performance standards peers from other facilities, including national and subnational officials with major child health indicators at two time points, February and from ministries of health, management personnel from health organiza- November 2010, in 11 facilities participating in an SBM-R improvement tions, or other authorized individuals from the community [14].Assess- initiative in Yanaoca, Peru (personal correspondence, Eva Miranda, ments are primarily based on direct observation of provider practices or Technical Advisor, Quality of Healthcare Project, 2013). A 33% overall of the facility’s infrastructure and operations. Interviews of health improvement in adherence to defined SBM-R standards was concurrent workers and record reviews can also augment information obtained with an increase in the percentage of children at these facilities docu- through observations. Assessments of performance for procedures de- mented as receiving growth monitoring (33% to 83%) and vaccination livered by more than one provider are based on a small convenience (41% to 85%), as well as with a modest increase in proportion of children sample, covering just one to three procedures. While the time required assessed as well-nourished (58% to 66%). Because of these improve- to conduct an assessment varies depending on the number of health ments and those obtained in other areas of the country using the service areas covered, performance assessments at a facility using SBM-R approach, Peru’s Ministry of Health, in an official resolution, for- SBM-R tools typically take from one half-day to two days. In several mally adopted SBM-R as a national approach to improving performance countries, governments have tracked key indicators related to SBM-R in health care [15]. objectives at the population level. In these cases, Jhpiego staff have pro- Similarly, in Honduras, the SBM-R process provided the opportunity vided support for the strengthening of the local information systems to introduce and track implementation of high-impact interventions and/or significant quality control of the data in countries such as such as active management of the third stage of labor (AMTSL) and a Mozambique, Guinea, Honduras, Peru, and Tanzania. shift away from routine episiotomy at six hospitals in two regions (personal correspondence, Gloria Fajardo, Technical Advisor, Maternal and 3. Results Neonatal Health Project, 2003). Baselines were conducted between August 2001 and January 2003 because the SBM-R process was phased 3.1. Implementation of SBM-R in at different times in different hospitals. Fig. 4 provides data for the last four months of 2003, illustrating the rapid adoption of appropriate prac- SBM-R interventions have been implemented in approximately 30 tices following SBM-R assessments and action plan development. Before countries. Table 1 presents a summary of completed or ongoing SBM- the SBM-R process was implemented, no deliveries were covered with R interventions, showing the number and type of facilities involved, ser- AMTSL, and episiotomies were performed in all cases. By the end of vices targeted, and progress in the achievement of specified standards. 2003, AMTSL was nearly universal (89.7% of deliveries), and episiot- This table shows that SBM-R has been used in facilities at different omies were being performed in less than half of deliveries (41.4%). S20 E. Necochea et al. / International Journal of Gynecology and Obstetrics 130 (2015) S17–S24

Table 1 Country experiences applying the Standards-Based Management and Recognition (SBM-R) methodology.a

Country Years Technical areas Number and type of facilities Progress on Standardsb

Afghanistan 2007–2012 DM, EPI, TB, FM, FP, IMCI, IPC, ANC, 281 BHC, 195 CHC, 28 DH Baseline: 26% PNC, L&D, LC, SNB, BCC Final: 80% Angola 2008–2011 ANC/MiP 2 MH, 40 HC, 14 HP Baseline: 6–47% Final: 82–100% 2006–2011 ANC, L&D, Immunizations, IMCI, general Luanda: 47 HC Baseline: 25–30% Huambo: 33 HC, Final: N65% for all facilities 66 HP Benin 2006–2012 Integrated RH and HIV 14 HC Baseline: 4–38% Final: 67–95% Brazil 1996–1999 RH-FP 9 HC Baseline: 18–57% Final: 62–95% Côte d’Ivoire 2008–2013 PMTCT, ARV 35 DH & HC Baseline: Not available (urban and rural) Final: 83–89% Equatorial Guinea 2011–2012 MNH 3 DH Baseline: 10–32% First: Not available Second: 39–44% Ethiopia 2011–2014 MNH 12 Hospitals, Baseline: 28% 104 HC First: 52% Second: 71% Ghana 2011–2014 Community-based health services 61 CHPS 30% improvement over baseline in 1 year for 18 CHPS assessed (2013 data) Guatemala 2001–2003 MNH 20 DH, 42 HC, Baseline: 14–20% 52 HP Final: 43–88% Guinea 2009–2014 EmONC, FP, IPC 27 DH, 15 HC Baselines: 9–65% Latest (2013): 46–96% Honduras 2001–2004 MNH 2 RH, 4 DH Baseline: 11–27% Final: 62–75% India 2011–2013 PSE 60 NS Baseline: 17–45% Final: 47–89% PPFP/ PPIUCD 38 DH&HC Baseline average: 42% Final assessment average: 79% Indonesia 2012–2013 MNH, IP, clinical governance, 23 DH, 94 HC, Baseline (DH): 24–36% referral systems 10 district referral systems Final (DH): 51–84% Baseline (HC): 23–39% Final (HC): 72–76% Jamaica 2002–2004 HIV-CT 14 HC Baseline: 15–60% Final: 33–98% Kenya 2008–2012 PMTCT 8 PH, 23 DH, 4 MiH, 7 HC Baseline: Not available Final: 54–80% 2012 FP 11 DH, 2 MiH, 3HC Baseline: Not available Final: 40–80% Malawi 2006–2014 IPC, M&RH (ANC, L&D, PNC, FP, STI, 28 DH, 32 HC Baseline: Data to come PAC, CECAP, PMTCT) Latest (2013):18 DH achieved N80% in IPC; 4 DH achieved N80% in M&RH Mozambique 2004–2014 IPC 46 CH/GH/PH/DH, 102 HC Baseline: 12–42% Latest (2013): 26–92% 2008–2014 Patient safety 68 CH/GH/PH/DH Baseline: 42–46%; Latest (2013): 76–84% 2008–2014 PSE 15 PSE Baseline: 36–55%; Latest (2012): 49–94% 2009–2014 MNH 3 CH, 7 PH, 5 GH, 33 DH, 47 HC Baseline: 34% Latest (2013) assessments: 60% Nigeria 2007–2009 MNH 6 GH Baseline: 0–25% Final: 77–100% 2010–2012 Sokoto State: MNH, FP, CH 17 GH Baseline: 4–33% Final: 22–56% Bauchi State: MNH, FP, CH 23 GH Baseline: 6–23% Final: 45–98% Pakistan 2004–2006 FP 48 Private sector providers Baseline: 15% Final: 64% 2006–2010 PHC 122 PHCF Baseline: Not available Final: Average of 30 PHCFs improved from 14% to 56% in 20 months; average of 59 PHCF improved from 7% to 30% in 8 months Paraguay 2011–2012 MNH 2 RH, 4 DH Baseline: 21–46% Latest: 26–65% Peru 2009–2012 BEmONC, CH 117 BEmONC, 239 CH Baseline (BEmONC): 36–92% Final (BEmONC): 49–98% Baseline (CH): 38–72% Final (CH): 59–87% Tanzania 2008–2014 BEmONC 21 RH, 233 HC Baseline: 20–68% Latest (2013): 53–76% E. Necochea et al. / International Journal of Gynecology and Obstetrics 130 (2015) S17–S24 S21

Table 1 (continued)

Country Years Technical areas Number and type of facilities Progress on Standardsb

Zambia 2007–2011 ART & PMTCT 8 HC (ART), 14 HC (PMTCT) Baseline (ART): 21–80% Final (ART): 52–92% Baseline (PMTCT): 5–86% Final (PMTCT): 43–86% Zimbabwe 2010–2012 MNH 17 OH, RuH, RHC, PC Baseline: 9–54% Latest (2012): 26–87%

Technical area abbreviations: ANC: prenatal care; ARV/ART: anti-retroviral therapy; BCC: behavior change and communication; BEmONC: basic emergency obstetric and neonatal care; CECAP: cervical cancer and prevention; CH: child health; DM: drug management; EmONC: emergency obstetric and neonatal care; EPI: expanded program of immunizations; FM: facility management; FP: family planning; HIV-CT: HIV counseling and testing; IMCI: integrated management of childhood illnesses; IP: infection prevention; IPC: infection prevention and control; LC: labor complications; L&D: labor and delivery; MiP: malaria in pregnancy; MNH: maternal and newborn health; M&RH: maternal and reproductive health; PAC: postabortion care ; PHC: pri- maryhealthcare;PMTCT:preventionof mother-to-child transmission (of HIV); PNC: postnatal care; PPFP/PPIUCD: postpartum family planning/postpartum intrauterine contraceptive device; PSE: pre-service education; RH: reproductive health; SNB: sick newborn; STI: sexually-transmitted infection; TB: tuberculosis. Facility type abbreviations: BHC: basic health center; CH: central hospital; CHC: comprehensive health center; CHPS: community-based health planning and services; DH: district hospital; GH: general hospital; HC: health center; HP: health posts; MH: municipal hospital; MiH: mission hospitals; NS: nursing school; PC: polyclinic; PH: provincial hospital; PHCF: primary healthcare facilities; RH: regional hospital; RHC: rural health center; RuH: rural hospital. a SBM-R has also been implemented in Guyana (CECAP) and Liberia (PSE); however, assessment data are not available. SBM-R has recently been implemented in Bangladesh, Uganda, and South Sudan but these are new programs with data that are not yet available. b SBM-R performance scores are represented as a percentage of the total possible performance. If a single value is reported, it is an average across facilities and standards/topics unless otherwise speciﬁed. “Final” is deﬁned as the end of Jhpiego data collection but not necessarily the end of SBM-R data collection.

Finally, Fig. 5 shows that improvements in adherence to a basic pack- Zambia, Malawi, and Afghanistan using quasi-experimental designs age of SBM-R maternal health standards that were related to prenatal with intervention and comparison groups of facilities. The selection of care, deliveries, and complications were concurrent with reductions in countries for these studies was based on feasibility, including the avail- the number of maternal complications and maternal deaths at a region- ability of suitable comparison facilities and funding sources. al hospital in Tanzania (personal correspondence, Dustan Bishanga, One systematic evaluation focused on the use of SBM-R for the October, 2013). As performance on maternal health standards increased prevention of mother-to-child transmission of HIV (PMTCT) and prenatal from 43% at baseline in 2009 to 83% to an interim assessment in 2011, care in Zambian defense force facilities between August 2010 and the number of maternal complications (postpartum hemorrhage, pre- December 2011. The evaluation matched four intervention and four com- eclampsia/eclampsia, and ruptured uterus) declined from 14 in 2010 parison facilities and found statistically signiﬁcant improvements in pro- to 10 in 2011, and the number of maternal deaths declined from 14 in vider performance at intervention sites on the following PMTCT and 2009 to 4 in 2010. Although these numbers are too small to make statis- prenatal care performance standards, when compared with changes at tical inferences, they circumstantially suggest a reduction in adverse comparison sites: family planning counseling (34% to 75%, P = 0.026); health outcomes. HIV testing during return prenatal care visits (13% to 48%, P =0.034); and HIV/AIDS management for HIV-positive mothers during prenatal 3.4. Systematic evaluations and implementation case studies—facility care that did not include an HIV test (1% to 34%, P = 0.004) [16].Improve- capacity and health worker performance ments in provider performance at intervention sites on overall prenatal care skills were not signiﬁcant. Because the results presented above do not include a comparison Another systematic evaluation that took place in Malawi in 2009 group and thus could be related to factors external to the SBM-R inter- compared the quality of reproductive health services and found that vention, Jhpiego has implemented more rigorous evaluations in the eight intervention facilities implementing SBM-R were more likely than the eight comparison facilities to have the needed infrastructure,

100 equipment, and supplies to offer evidence-based prenatal care and fam- 2001 (baseline) ily planning services. The evaluation also showed that achievement of 90 performance standards was significantly higher in the intervention fa- 2002 (1st external cilities than comparison facilities for postnatal care and family planning 80 evaluation) services [17]. Of the 120 family planning verification criteria developed 70 specifically for this comprehensive study, the mean number achieved by intervention facilities was 89.0 (74%), compared with 70.5 (58%) for 60 comparison facilities (P b 0.01). Of 200 postnatal care verification criteria, the mean number achieved by intervention facilities was 50 144.2 (72%), compared with 135.2 (68%) for comparison facilities Percentage 40 (P b 0.01). There were no significant differences between intervention and comparison facilities achievement of standards for prenatal care 30 or labor and delivery care. 20 Finally, a systematic evaluation of an SBM-R intervention to improve maternal health services in 31 facilities in Afghanistan showed that, in 10 the areas of prenatal care and family planning, provider performance in intervention facilities improved with longer exposure to the intervention 0 Basic maternal Infection Information, Resources Management as well as in relation to the comparison group [18]. Mean scores for pre- care prevention education, & natal care performance, measured by the percentage of tasks achieved communication Assessment area during clinical observations, were 47% for the comparison group, 56% for the beginning SBM-R group, and 71% for the advanced SBM-R group b Fig. 2. Achievement of Standards-Based Management and Recognition (SBM-R) maternal (P 0.05). Mean scores for family planning performance were 57%, 68%, and newborn care standards in 11 health centers: Guatemala, 2001–2002. and 77%, respectively, for these same groups ( P b 0.05). There was no S22 E. Necochea et al. / International Journal of Gynecology and Obstetrics 130 (2015) S17–S24

100

90 Performance on 80 child health standards 70 Growth monitoring 60

50 Vaccines <1 yr Percentage 40

30 ≤ 20 Well-nourished 5 yr 10

0 Feb-06 Nov-06 Month-Day

Fig. 3. Achievement of Standards-Based Management and Recognition (SBM-R) child health standards and coverage of interventions at 11 facilities: Peru, 2010.

significant improvement in performance scores for labor and delivery colleagues,” [20] they were able to improve educational performance care. The evaluation showed that client-provider communication and cli- and achieve accreditation status. ent satisfaction significantly improved with increased duration of the SBM-R intervention and relative to comparison groups. 4. Discussion Jhpiego has also conducted implementation case studies in a few countries to further explore the qualitative effects of SBM-R. One of SBM-R provides a rapid and feasible methodology to enable facilities these examined the PROQUALI reproductive health project in Brazil in to understand their service quality and implement solutions to address 1999, which was the first introduction of SBM-R in that country in five problems. Through this aggregate collection of program data, quasi- pilot clinics [19]. The case study observed that although SBM-R was experimental evaluations, and implementation case studies from a labor-intensive, it improved compliance with standards of care through number of countries, it is evident that SBM-R leads to improvements its self-assessments. In particular, the SBM-R tools identified perfor- in provider compliance with performance standards and might be asso- mance gaps and contributing factors not previously perceived in the ciated with improved coverage of high-impact MNCH interventions for clinics and overcame them through teamwork and improved allocation clients of facility services. Routine SBM-R assessments, though partially of resources such as staff, commodities, and educational materials. self-reported, generally document improvements in provider adher- A second case study assessed the use of SBM-R for accreditation of ence to best practices and improvements in facility infrastructure and community midwifery schools in Afghanistan and showed that SBM-R operations. The results of the quasi-experimental studies that compare assessments were systematically applied to improve adoption of recom- changes between intervention and control groups or between baseline mended processes and practices at participating schools [20].Three and post-intervention data provide even stronger evidence of the posi- schools received essentially no external technical assistance owing to tive effects of SBM-R. Finally, in-depth case studies reveal how these insecurity or funding limitations but, “because of the detailed nature improvements are achieved, through repeated assessments, better of the assessment tools and the interaction with other schools and communication among providers, and development and application of action plans. By triangulating the patterns that emerge from these multiple methodologies, it is possible to assert a causal relationship that 100 holds for real-world settings [21]. While the full details of the quasi-experimental studies are reported 90 elsewhere, it is important to note that these studies used representative 80 random sampling and well-trained, unbiased assessors. The results of 70 such studies generally showed less dramatic changes in overall skills improvement than the assessment-based data; however, improvements 60 September were statistically significant for many skills. The difference in magnitude 50 October between routine SBM-R assessments and quasi-experimental studies of 40 November SBM-R effects could be due to differences in measurement methods, Percentage December quality of assessors, and sampling methods, or they could be due to the 30 different contexts in which each were performed. 20 Program data from Peru, Honduras, and Tanzania further suggest that, 10 in addition to improving provider practices and quality of care, SBM-R could be associated with improvements in intermediate and long-term 0 Epiostomies (average) AMTSL (average) health outcomes, such as lower episiotomy rates, improved child nutrition, and fewer maternal complications and deaths. The Afghanistan fi Fig. 4. Changes in episiotomies and coverage of active management of the third stage of quasi-experimental study builds upon these ndings by identifying a labor (AMTSL) at six hospitals: Honduras, 2003. linkage between implementation of SBM-R and improvements in client E. Necochea et al. / International Journal of Gynecology and Obstetrics 130 (2015) S17–S24 S23

18 100%

16 90% Maternal complications 14 80% 70% 12 60% 10 50% Maternal 8 deaths Number 40% 6 Percetnage 30% 4 20% Performance 2 10% on SBM-R 0 0% standards 2009 2010 2011 Year

Fig. 5. Achievement of Standards-Based Management and Recognition (SBM-R) maternal healthcare standards and adverse maternal outcomes at a national hospital: Tanzania, 2010–2011.

satisfaction, an important outcome of care and a meaningful indicator of 4.1. Strengths and limitations care quality [22]. The fact that SBM-R was adopted by external stakeholders in Afghanistan, and also in Peru and Guatemala, appears to indi- The results in the present article have some important limitations. cate sustainability, yet sustainability remains an ongoing challenge in Most of the sources from which these data were drawn were able to doc- other countries. ument changes in adherence to SBM-R standards, but few were able to This collective presentation of the results of SBM-R interventions, track changes in patient outcomes. The suggested associations between using a wide variety of analytical approaches, is particularly meaningful improved SBM-R assessment results and improved health outcome re- because of the inherent difficulties in demonstrating the direct impact sults based on health service statistics may not be completely reliable be- of quality improvement initiatives on health outcomes. Raven et al. [23] cause of the shortcomings of routine information systems; however, analyzed the methodologies and tools used in 34 articles and reports on those programs made special data quality assurance efforts to overcome quality improvement in maternal and newborn care in low-income coun- these issues. In some countries, the national government, Jhpiego, and tries and concluded that, while there is emerging evidence to suggest that donor partners have worked to revise data registries. In others, program quality improvement interventions lead to demonstrable changes in care staff have performed data quality checks or supported sentinel sites for practices, most included reports did not offer detailed information on in-depth data collection. While the quasi-experimental studies are de- how to implement quality improvement methods and did not articulate signed with careful data collection practices from the start, their small how documented improvements occurred. Other recent studies on the ef- sample sizes might limit the generalizability of reported results. fectiveness of several quality-related approaches such as accreditation The many strengths of the SBM-R approach mitigate these limitations [24], supervision [25], and quality improvement collaboratives [26,27] and support further evidence generation. SBM-R does not focus on a sin- have also shown limited or disappointing results. The lack of positive at- gle intervention, but instead on a set of services and a systems approach tribution does not, however, mean that such attribution cannot be to facility readiness. While complex improvements across an array of made, just that common experimental methods might not suffice. Quality topics are more difficult to implement than single-intervention pro- improvement programs are evolving in a changing economic, social, and grams, such coordinated efforts can help reduce artificial, unsustainable political climate and use interventions that focus on complex adaptive so- improvements in performance; it is easier to temporarily comply with a cial systems, so their activities might have a synergistic effect rather than single practice than with a complex set of interventions. The improve- an effect on individual clients, and their outcomes might therefore need to ments reported in SBM-R assessments are affirmed by systematic studies be studied from the perspectives of different parties [28]. that used quasi-experimental designs and external data collection. Even Although the experimental evidence regarding the impact of quality though most of the results presented relate to provider adherence to improvement interventions on practice improvement is limited, other best practices, and not directly to patient outcomes, they begin to show systematically-collected evidence regarding the effects of SBM-R can be that performance of those best practices result in specific health out- found in research on component approaches that are embedded within comes when reliably implemented. the SBM-R methodology. The audit and feedback process, for example, which is routinely practiced in SBM-R after assessments are implemented, has shown small improvements in practice [29]. Similarly, educational 5. Conclusions outreach visits, which are part of the SBM-R external assessments, have resulted in small but consistent improvements in service delivery [30]. This broad survey of SBM-R initiatives and results demonstrates that Checklists, an essential feature for tracking SBM-R standards and verifica- the SBM-R approach has consistently contributed to improvements in tion criteria, have been associated with improved practices in maternal provider performance of MNCH service delivery. The SBM-R methodol- and newborn health [31]. The one approach that has not yet yielded evi- ogy facilitates rapid action to correct service gaps and has shown to be dence on improved outcomes is self-certification of the use of process helpful in promoting a culture of standardization, continuous measure- standards [32]. Every effort should be made to track patient outcomes ment, and recognition of achievements. The spread of SBM-R to and as part of quality improvement program evaluations, including evalua- within many countries suggests that it has a high level of system-wide tions of SBM-R, recognizing that accurate measurement of such outcomes acceptability, though this reach has been influenced by donor invest- is challenging in low-resource settings. ment. The large number of facilities that are now successfully applying S24 E. Necochea et al. / International Journal of Gynecology and Obstetrics 130 (2015) S17–S24

SBM-R shows its potential as a method that can be used to improve [13] Herzberg F. One more time: how do you motivate employees? Harv Bus Rev 1968;1: quality at scale. As a result, SBM-R should be part of the array of quality 53–62 Reprinted in Harvard Business Review, Business Classics: Fifteen Key Classic for Managerial Success, 1996. improvement options considered in low-resource settings, particularly [14] Necochea E, Bossemeyer D. Standards-based management and recognition: a where the adoption of evidence-based practices is inconsistent and field guide. Published 2005. http://www.jhpiego.org/files/SBMR%20FieldGuide. health systems are weak. Further effort is needed to strengthen docu- pdf. Accessed May 3, 2014. – mentation of the results of SBM-R implementation, including the effect [15] Government of Peru, Ministry of Health. Ministerial Resolution No. 556 2012; Published 2012. of SBM-R on key indicators of patient outcomes and population-level [16] Kim YM, Chilila M, Shasulwe H, Banda J, Kanjipite W, Sarkar S, et al. Evaluation of a changes. As with other quality improvement approaches, SBM-R will quality improvement intervention to prevent mother-to-child transmission of HIV benefit from the growing trend toward more robust implementation re- (PMTCT) at Zambia defense force facilities. BMC Health Serv Res 2013;13:345. [17] Rawlins BJ, Kim YM, Rozario AM, Bazant E, Rashidi T, Bandazi SN, et al. Reproductive search and evaluation methods. health services in Malawi: an evaluation of a quality improvement intervention. Midwifery 2012;29(1):53–9. [18] Zainullah P, Kim Young-Mi, Bazant E, Yari K. Raising the quality of care: an impact fl evaluation in Afghanistan. Presented at the International Society for Quality in Con ict of interest Health Care. 30th International Conference: Quality and Safety in Population Health; October 2013. The authors have no conflicts of interest. [19] Blake SM, Necochea E, Bossemeyer D, Brechin SJG, da Silva BL, da Silveira DM. PROQUALI: Development and dissemination of a primary care center accreditation model for performance and quality improvement in reproductive health services in northern Brazil. http://reprolineplus.org/resources/proquali-development-and- References dissemination-primary-care-center-accreditation-model-performance. Accessed May 10, 2014. [1] Hacker S, Jouslin de Noray B, Johnston C. Standardization versus improvement: ap- [20] Smith JM, Currie S, Azfar P, Rahmanzai AJ. Establishment of an accreditation system proaches for changing work process performance. London: European Quality, Publi- for midwifery education in Afghanistan: maintaining quality during national expan- cations, Ltd; 2001. sion. Public Health 2008;122(6):558–67. [2] Adegoke AA, Mani S, Abubakar A, van den Broek N. Capacity building of skilled birth [21] Shelton J. Evidence-based public health: not only whether it works, but how it attendants: a review of pre-service education curricula. Midwifery 2013;29(7):e64–72. can be made to work practicably at scale. Glob Health Sci Pract 2014;2(3): [3] Goga AE, Muhe LM. Global challenges with scale-up of the integrated management 253–8. of childhood illness strategy: results of a multi-country survey. BMC Public Health [22] Pittrof R, Campbell OM, Filippi VG. What is quality in maternity care? an internation- 2011;11:503. al perspective. Acta Obstet Gynecol Scand 2002;81(4):277–83. [4] Murila F, Obimbo MM, Musoke R. Assessment of knowledge on neonatal resuscita- [23] Raven J, Hofman J, Adegoke A, van den Broek N. Methodology and tools for quality tion amongst health care providers in Kenya. Pan Afr Med J 2012;11:78. improvement in maternal and newborn health care. Int J Gynecol Obstet 2011; [5] Sipsma HL, Curry LA, Kakoma JB, Linnander EL, Bradley EH. Identifying characteristics 114(1):4–9. associated with performing recommended practices in maternal and newborn care [24] Flodgren G, Pomey MP, Taber SA, Eccles MP. Effectiveness of external inspection of among health facilities in Rwanda: a cross-sectional study. Hum Resour Health compliance with standards in improving healthcare organization behaviour, 2012;10(1):13. healthcare professional behaviour or patient outcomes. Cochrane Database Syst [6] Manzi F, Schellenberg JA, Hutton G, Wyss K, Mbuya C, Shirima K, et al. Human re- Rev 2011(11):CD008992. sources for health care delivery in Tanzania: a multifaceted problem. Hum Resour [25] Bosch-Capblanch X, Liaqat S, Garner P. Managerial supervision to improve primary Health 2012;10:3. health care in low- and middle-income countries. Cochrane Database Syst Rev [7] Bukuluki P, Byansi PK, Sengendo J, Ddumba NI, Banoba P, Kaawa-Mafigiri D. Chang- 2011;9:CD006413. ing from the “Pull” to the “Push” system of distributing essential medicines and [26] Landon BE, Wilson IB, McInnes K, Landrum MB, Hirschhorn L, Marsden PV, et al. Ef- health supplies in Uganda: implications for efficient allocation of medicines and fects of a quality improvement collaborative on the outcome of care of patients with meeting the localized needs of health facilities. Glob Health Gov 2013;6(2):74–84. HIV infection: the EQHIV study. Ann Intern Med 2004;140(11):887–96. [8] USAID. Nepal family health program technical brief #13: pull system. http://www.jsi. [27] Mittman BS. Creating the evidence base for quality improvement collaboratives. Ann com/JSIInternet/Inc/Common/_download_pub.cfm?id=12152&lid=3.Accessed Intern Med 2004;140(11):897–901. May 3, 2014. [28] Øvretveit J, Gustafson D. Evaluation of quality improvement programmes. Qual Saf [9] Bosch-Capblanch X, Ronveaux O, Doyle V, Remedios V, Bchir A. Accuracy and quality Health Care 2002;11(3):270–5. of immunization information systems in forty-one low income countries. Trop Med [29] Ivers N, Jamtvedt G, Flottorp S, Young JM, Odgaard-Jensen J, French SD, et al. Audit Int Health 2009;14(1):2–10. and feedback: effects on professional practice and healthcare outcomes. Cochrane [10] Chan M, Kazatchkine M, Lob-Levyt J, Obaid T, Schweizer J, Sidibe M, et al. Meeting Database Syst Rev 2012;6:CD000259. the demand for results and accountability: a call for action on health data from [30] O’Brien MA, Rogers S, Jamtvedt G, Oxman AD, Odgaard-Jensen J, Kristoffersen DT, eight global health agencies. PLoS Med 2010;7(1):e1000223. et al. Educational outreach visits: effects on professional practice and health care [11] Bradley S, Kamwendo F, Masanja H, de Pinho H, Waxman R, Boostrom C, et al. Dis- outcomes. Cochrane Database Syst Rev 2007(4):CD000409. trict health managers’ perceptions of supervision in Malawi and Tanzania. Hum [31] Spector JM, Agrawal P, Kodkany B, Lipsitz S, Lashoher A, Dziekan G, et al. Improving Resour Health 2013;11:43. quality of care for maternal and newborn health: prospective pilot study of the WHO [12] Willis-Shattuck M, Bidwell P, Thomas S, Wyness L, Blaauw D, Ditlopo P. Motivation safe childbirth checklist program. PLoS One 2012;7(5):e35151. and retention of health workers in developing countries: a systematic review. BMC [32] Draycott T, Sibanda T, Laxton C, Winter C, Mahmood T, Fox R. Quality improvement Health Serv Res 2008;8:247. demands quality measurement. BJOG 2010;117(13):1571–4. International Journal of Gynecology and Obstetrics 130 (2015) S25–S31

Contents lists available at ScienceDirect

International Journal of Gynecology and Obstetrics

journal homepage: www.elsevier.com/locate/ijgo

SUPPLEMENT ARTICLE Task shifting in maternal and newborn health care: Key components from policy to implementation

Barbara Deller a,1,⁎, Vandana Tripathi b, Stacie Stender c, Emmanuel Otolorin d,PeterJohnsona, Catherine Carr a a Jhpiego, Baltimore, MD, USA b Johns Hopkins Bloomberg School of Public Health, Baltimore, MD, USA c Jhpiego, Pretoria, South Africa d Jhpiego, Abuja, Nigeria article info abstract

Keywords: Task shifting in various forms has been adopted extensively around the world in an effort to expand the reach of Health worker lifesaving services to the women, newborns, and families who need them. The emerging global literature, as well Human resources for health as Jhpiego’s field experiences, supports the importance of addressing several key components that facili- Maternal and newborn health tate effective task shifting in maternal and newborn health care. These components include: (1) policy Task shifting and regulatory support; (2) definition of roles, functions, and limitations; (3) determination of requisite Task sharing skills and qualifications; (4) education and training; and (5) service delivery support, including management and supervision, incentives and/or remuneration, material support (e.g. commodities), and referral systems. Jhpiego’s experiences with task shifting also provide illustrations of the complex interplay of these key components at work in the field. Task shifting should be considered as a part of the larger health system that needs to be designed to equitably meet the needs of mothers, newborns, children, and families. © 2015 Published by Elsevier Ireland Ltd. on behalf of International Federation of Gynecology and Obstetrics. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

1. Background tasks to one provider cadre and from another. Various other terms are found in the literature to communicate the task shifting concept, includ- The WHO has estimated that 57 countries have a critical shortage of ing “task sharing,”“substitute health worker,”“skills substitution,”“task healthcare workers, and that 2.4 million doctors, nurses, and midwives delegation,” and “optimizing health worker roles” [4,5]. For the purposes are needed to achieve the Millennium Development Goals (MDGs) [1]. of this paper, the authors define “task shifting” as either: (1) developing The most acute shortfalls occur in Southeast Asia and Sub-Saharan a new provider cadre, such as lay health workers (LHWs) with compe- Africa. Thirty-six of the 57 countries are in Sub-Saharan Africa, which tencies to perform tasks normally performed by health professionals contains 24% of the global burden of disease but only 3% of the global with more education and higher qualifications; or (2) expanding the health workforce [2]. Inequalities in health workforce distribution scope of practice of an existing health professional cadre to accept addi- within countries are also common; it is estimated that only 24% of tional tasks and functions. The intent of both types of task shifting is to physicians and 38% of nurses work in rural areas, although half of the bring services closer to the population and increase health system effi- world’s population is rural [1]. ciencies. LHWs, also known as community health workers (CHWs), Difficult working conditions, high rates of attrition, maldistribution might take the form of lady health visitors, traditional birth attendants, and out-migration of staff, and the HIV/AIDS epidemic have all contrib- health educators, or other paid or volunteer community members who uted to the inadequate supply of skilled health workers [3]. The dramatic have some health-related training but lack the broader educational shortage of skilled providers creates challenges to the provision of both preparation typical of a healthcare professional. facility- and community-based healthcare services, including services Task shifting builds on the assumption that less specialized health for maternal and newborn health (MNH). Task shifting has been workers can take on some of the responsibilities of more specialized promoted as one response to this global health worker crisis, shifting workers in a cost-effective manner without sacrificing quality of care [6,7]. Multiple efficiencies may arise from task shifting, given that the cadre to which tasks are shifted often require shorter training periods ⁎ Corresponding author at: Jhpiego, 1615 Thames Street, Baltimore, MD 21231, USA. and lower educational qualifications, might have skills specifictotheir Tel.: +1 202 835 6069. E-mail address: [email protected] (B. Deller). local setting (e.g. language), and are not as likely to emigrate to other 1 (retired). countries [8]. Task shifting is not, however, an intervention that occurs http://dx.doi.org/10.1016/j.ijgo.2015.03.005 0020-7292/© 2015 Published by Elsevier Ireland Ltd. on behalf of International Federation of Gynecology and Obstetrics. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). S26 B. Deller et al. / International Journal of Gynecology and Obstetrics 130 (2015) S25–S31 in a vacuum; instead, it must be aligned with broader health systems evidence [12,13]. Box 1 provides some illustrations of MNH tasks that strengthening activities [9]. might be appropriate for various cadres of healthcare providers. Many The present paper reviews Jhpiego’s experiences in the use of task of these recommendations are in line with five key components that shifting as a strategy for increasing the scope and breadth of its MNH Jhpiego has identified and prioritized as the basis for its task-shifting programming in over a dozen countries around the world (Table 1). work: (1) policy and regulatory support; (2) determination of roles, These experiences were chosen to illustrate both types of task shifting functions, and limitations; (3) determination of requisite skills and and to closely examine the processes and key components underlying qualifications; (4) education and training; and (5) service delivery the task-shifting approach that maximize the potential of available support. human resources.

2.1. Policy and regulatory support 2. Key components of task shifting in maternal and newborn care In most contexts, there are clear delineations of the roles that differ- There is a growing body of research and grey literature evaluating ent health cadres can perform; these may be established by ministries of task shifting in low-resource country settings, much of which has health and/or professional and licensing councils and enacted through come from the HIV/AIDS ﬁeld [10,11]. The WHO has also issued recom- pre-service curricula, job descriptions of and procedure manuals for mendations for MNH task shifting based on clinical and research health personnel, or legislative authorization. Changes in practice

Table 1 Selected examples of task shifting in Jhpiego’s maternal and newborn health programs.a

Country/location Years Cadre(s) “shifted to” New/shifted skills Cadre(s) “shifted from”/shared with

Afghanistan/20 districts Pilot: CHWs Community-based distribution of Oxytocin and ergometrine by skilled in Badakshan, Bamyan, 2005–2007 misoprostol for PPH prevention birth attendants, primarily midwives, Faryab, Jawzjan, and Expansion: 2011–2012 in facilities Kabul Provinces Afghanistan/national Pilot: CHWs Community-based postpartum FP, including Facility-based providers, particularly 2007–2010 injectables midwives Expansion: 2010–present Bangladesh/Sylhet 2007–present CHWs Postpartum FP counseling and contraceptive Facility-based health providers District distribution Ghana/Accra metro area, 2000–2003 Nurses Cervical cancer screening including: VIA, Physicians, including specialists (with Amasaman, and clinical decision-making, and cryotherapy partial role for nurses) Kumasi Guinea/national 1998–present Nurses, midwives, and PAC Physicians (emergency PAC only) auxiliary nurses India/22 states Pilot: MBBS doctors (general CEmONC Obstetricians/gynecologists 2005–2006 medical doctors) Expansion: 2007–present Kenya/Embu District 2007–present Midwives Postpartum FP, including provision of Obstetricians/gynecologists postpartum intrauterine contraceptive devices, and PNC Malawi/National 2005–2009 Health surveillance Community-based MNH services, including: Community health nurses assistants prenatal care (e.g. education on danger signs, promotion of facility delivery), PNC, and neonatal assessment Malawi/national 2005–2009 Nurse midwifery BEmONC Registered nurse midwives technicians Mozambique/national 2008–present CHWs HIV counseling and rapid diagnostic testing Nurses, midwives Mozambique/national 2011–present Maternal health nurses Initiation and management of clients on Physicians antiretroviral therapy Nepal/31 districts Pilot: Female community Community-based distribution of Skilled birth attendants (oxytocin 2005–2008 health volunteers misoprostol for PPH prevention only) Expansion: 2009–present Nigeria/Kano, Katsina, 2006–2012 CHEWs Clean and safe delivery, BEmONC, and FP Physicians and nurse-midwives Zambfara states counseling and services Nigeria/Akwa Ibom state 2007–2012 Volunteer Malaria services, including prevention of Midwives and CHEWs community-directed malaria in pregnancy and bed net promotion distributors Rwanda/13 districts 2010–2012 CHWs Community-based PPH prevention and FP Physicians and nurses in facilities Rwanda/9 districts 2010–2012 Nurses and midwives BEmONC, including management of breech Specialist physicians births South Africa/national 2007–2010 Professional nurses and Initiation and management of clients on Physicians midwives antiretroviral therapy Thailand/national 2000–2003 Nurses Cervical cancer screening including: VIA, Physicians, including specialists (with clinical decision-making, and cryotherapy partial role for nurses) Zambia/Southern, 2010–2011 CHWs HIV counseling and rapid diagnostic testing Nurses, midwives, laboratory Eastern, and Western personnel Provinces

Abbreviations: BEmONC, basic emergency obstetric and newborn care; CEmONC, comprehensive emergency obstetric and newborn care; CHEWs, community health extension workers; CHWs, community health workers; FP, family planning; MBBS, Bachelor of Medicine, Bachelor of Surgery; MNH, maternal and newborn health; PAC, postabortion care; PNC, postnatal care; PPH, postpartum hemorrhage; VIA, visual inspection with acetic acid. a The term “physician” includes both specialists (such as obstetricians and pediatricians) and nonspecialists (primary care, generalists or family doctors). The term “midwife” includes both direct-entry midwives and nurse-midwives, unless otherwise speciﬁed. B. Deller et al. / International Journal of Gynecology and Obstetrics 130 (2015) S25–S31 S27

Box 1 antiretrovirals (ARVs) for HIV, another group, such as LHWs, might take Illustrative maternal and newborn care tasks that may be shifted to dif- on the counseling and testing previously done by nurses. The results of ferent cadres [12]. task analysis should guide decisions regarding both the cadres “shifted to” and “shifted from.” It is particularly important to ensure that the Lay health workers cadre taking on new services has a clear understanding of the limita- • Malaria prevention tions of its expanded role. Responsible task shifting requires clear • Labor support boundaries and processes for referral to higher-level cadres, as needed • Prevention of postpartum hemorrhage [19], to maintain positive and productive working relationships • Neonatal resuscitation among all cadres affected by the shift. While research suggests that • Management of newborn sepsis “shifted from” providers generally have positive attitudes toward task • Initial dose of antibiotics for maternal sepsis shifting [20,21], task shifting without clearly defined roles and processes • All short-acting contraceptives, plus injectables can undermine teamwork and result in lasting tension between health • HIV counseling and testing worker cadres, in part due to “professional protectionism” [17,19,20]. Auxiliary nurse midwives Clear roles and responsibilities serve not only to distinguish cadres but As above, plus: also to harmonize them within the broader health system. Such harmo- • Labor and birth nization prevents health services from becoming too fragmented and • Initial antibiotic for premature rupture of membranes difficult for users to negotiate [20]. • BEmONC • Intrauterine devices 2.3. Determination of requisite skills and qualifications Midwives/nurses As above, plus: Even if cadres to whom tasks are being shifted are already extant and • Manual vacuum aspiration widely deployed, their position requirements and, in turn, recruitment • Prenatal corticosteroids for preterm labor strategies, must be adapted to their new skills and qualifications [9, • Antiretrovirals for treatment/prevention of HIV 10,19,20]. For example, literacy might not have been an important • Contraceptive implants requirement for CHWs when their role was limited to the promotion • Cesarean deliveries of healthy behaviors and health service utilization, given that Other non-physician clinicians and non-specialist physicians numerous pictorial tools and job aids exist for such work. However, As above, plus: if task shifting programs intend to involve CHWs in service provision • CEmONC and recordkeeping, then literacy and numeracy might become • Tubal ligations important requirements [9]. Clarifying such skill requirements is crucial to avoid a mismatch between expected roles and available human resource capacity [19].

2.4. Education and training must be affirmed in national regulatory mechanisms such as cadre scopes of practice, licensing examinations, re-licensing requirements, Appropriate learning resources and skills building are required for and professional association or council responsibilities. Shifting tasks cadres to whom tasks are being shifted. Successful task-shifting efforts within or between professional cadres, or from a professional cadre to have involved competency-based education and training for “shifted LHWs, therefore requires engagement with policymakers and regulatory to” health workers and include certification processes and clear perfor- bodies [13,14]. Formal revision of policy and regulatory frameworks mance standards when possible [10,14,22,23]. The involvement of often requires evidence regarding the effectiveness of shifting a particu- professional associations and educational and training institutions in lar task; pilot projects can provide this evidence but might involve secur- the development and use of teaching materials might also be important ing informal permission or a conditional waiver. Obtaining appropriate [24]. Many task-shifting programs begin with in-service education if policy and regulatory support may necessitate advocacy with national they are changing the roles of existing cadres [7]; however, sustained stakeholders; without such support, task shifting programs face numer- and scaled-up task shifting requires the revision of pre-service ous barriers (e.g. health cadres may be reluctant to take part owing to education to reach new providers [15,25]. Education can ensure the cru- liability concerns) [5,15]. Absence of explicit policy leadership or regula- cial harmonization of task shifting with the broader health system by tory support can severely limit the degree to which task shifting is providing a recognized platform for coordination among the “shifted scaled-up or sustained [11,16]. to” cadres and their supervisors and clinical counterparts, ensuring mutual understanding of roles and responsibilities and consistent guidelines for practice. This coordination is particularly important if 2.2. Determination of roles, functions, and limitations services are being shifted to non-clinical cadres and if clinicians are involved in supervisory and quality assurance roles. A task-shifting program should be preceded by task analysis to understand the specific services contributing to healthcare delivery 2.5. Service delivery support bottlenecks that undermine productive efficiency [17,18]. Task analysis provides information about the care actually provided by a specific While the preceding four components address key actions that cadre and often uncovers that the actual tasks performed by a cadre are necessary for successful task shifting, an enabling environment do not match the official job description, regulatory statutes, or pre- comprised of the following service delivery support elements service education curriculum. Task analysis can guide evaluation and encourages “shifted to” cadres and their supervisors to provide revision of the cadre’s scope of practice, the content of its pre-service services most effectively. and in-service curricula, and regulatory and deployment decisions. Task analysis must also consider the responsibilities that a cadre already 2.5.1. Management and supervision carries out and whether additional services will overburden that cadre. Maintaining the quality of services provided by lesser-educated Sometimes a second cadre might be engaged to make task shifting cadres is a primary challenge in task shifting. Virtually all task-shifting feasible. For example, when busy nurses take on the task of prescribing reviews and evaluations single out the importance of supportive S28 B. Deller et al. / International Journal of Gynecology and Obstetrics 130 (2015) S25–S31 supervision—a concept that includes performance assessment, remedial Box 2 education, mentoring, and motivation [7,10,18,20,26]. Such supervision Defining emergency obstetric and newborn care: Basic and comprehen- must be resourced with financial, logistical, and educational support for sive [33]. supervisors. Additionally, task shifting should be embedded in a comprehensive quality assurance framework including regular, standard- BEmONC: ized assessment [20]. Supportive supervision can help prevent the • Administration of parenteral antibiotics sense of isolation and frustration that has been implicated in high • Administration of parenteral anticonvulsants attrition rates, particularly among CHWs [27]. • Administration of parenteral uterotonics • Removal of retained products (manual vacuum aspiration) 2.5.2. Incentives and/or remuneration • Assisted vaginal delivery Task shifting frequently requires existing health workers to take on • Manual removal of the placenta more work, which can add burdens particularly to volunteer cadres. • Resuscitation of the newborn Thus, it is essential to devise appropriate incentive packages and recog- CEmONC: nition systems to reward cadres taking on new tasks to reduce their at- All above functions, plus: trition [7]. While some reviews have recommended financial incentives • Blood transfusion [27], other research on CHWs has suggested that effective incentives • Cesarean delivery include social prestige, support for income-generating activities, and recognition from the community and health system, and that in-kind incentives can contribute to motivation and retention [28–30].

2.5.3. Material support such as commodities, supplies, and job aids Cadres implementing newly shifted tasks require the same material can provide basic emergency obstetric and newborn care BEmONC support as the original cadre. However, because task shifting might (Box 2) [33]. The 2003 Demographic and Health Survey showed that require workers to practice new competencies or take services outside in the North-West Zone of Nigeria, where Jhpiego was working at the the health facility context, it is important to ensure that “shifted to” time, there was only a 12.3% rate of SBA-attended births [34].Mostpri- cadres have the material tools to maintain their skills and fulﬁll their mary healthcare centers did not have a doctor or midwife and were new roles. Logistics systems are especially important to maintain a staffed with Community Health Extension Workers (CHEWs). As a re- consistent and quality assured supply of drugs, equipment, supplies, sult, although Jhpiego was working to increase CEmONC capacity and commodities, as are up-to-date and simple job aids and service among midwives and physicians, rural primary healthcare centers still protocols, ideally tested for comprehensibility and ease of use [9,31,32]. lacked even BEmONC. In 2006, in collaboration with the Federal Minis- try of Health (FMOH) and Save the Children, Jhpiego began to equip 2.5.4. Referral systems CHEWs with the competencies and support needed to provide quality Task shifting often results in service provision outside the facilities BEmONC—previously provided by physicians, nurses, and midwives where the services have been customarily performed, e.g. into peripher- based in hospitals. al health facilities and the community. This dislocation increases the im- First, the FMOH agreed to modify the training package for CHEWs to portance of functioning referral systems to ensure that women and prepare them as SBAs, including the skills needed to provide BEmONC. newborns can be referred for further care, whether for needs that al- Next, the partners sought subnational government support to prepare ready exist upon presentation but are not performed by the shifted the health system to accept these new services. Magnesium sulfate, de- cadre, or for complications or adverse effects that might arise during livery kits, and other supplies and equipment had to be procured, and service provision [10,19]. For example, cadres newly capacitated as some maternity units required renovation. For the CHEWs who were al- skilled birth attendants (SBAs) need to be able to refer women for com- ready in practice, in-service training courses were established with pre- prehensive emergency obstetric care and newborn care (CEmONC), service preceptors and nursing faculty. Post-training supervision and a which includes cesarean delivery. The absence of such referral mecha- performance improvement system were also initiated to support nisms can heighten the risks of a fragmented health service delivery CHEWs in their new tasks. system. The training of CHEWs deployed to primary healthcare centers led to increased utilization of prenatal and delivery services. For example, in 3. Case studies: Applications of key components to Jhpiego’s ﬁeld programs

2,500 Global experience in task shifting is extensive. Jhpiego’s task-shifting 2144 experience began with the transition of intrauterine device services from physicians to nurses and midwives in Morocco and Egypt in 2,000 1983. In the last ﬁfteen years, Jhpiego has continued to implement task-shifting efforts in its programs to increase access to and improve 1,500 1285 the coverage of services to women and newborns. Table 1 summarizes 1177 Jhpiego’s more recent programmatic experiences with task shifting, most of which were implemented in partnership with governments 1,000 657

and other agencies. The following case studies provide more detail deliveries Number of about four of these experiences that best highlight the role of the key 500 components described above. 0 3.1. Nigeria: Increasing access to basic emergency obstetric and newborn 0 2007 2008 2009 2010 2011 care Year

As in many countries in Africa, Nigeria faces high maternal and new- Fig. 1. Nigeria case study. Number of deliveries conducted by community health extension born mortality rates, in large part due to the lack of access to SBAs who workers (CHEWs). B. Deller et al. / International Journal of Gynecology and Obstetrics 130 (2015) S25–S31 S29

Kano, Katsina, and Zamfara States, the Jhpiego-led program trained through recognition mechanisms [38].Certification of teaching sites CHEWs working in 23 primary healthcare centers. The total number of by FOGSI became mandatory to ensure sustained educational quality. women receiving SBA services from these CHEWs increased several- An expansive discussion of SBM-R is offered in another paper in this fold between 2007–2008 and 2009–2010 (Fig. 1). supplement [38]. The main challenges were the systemic problems often found in rural healthcare settings. There was an irregular supply of medications 3.3. Nepal: Expanding uterotonic coverage to prevent postpartum and consumable supplies, infrastructural deficiencies (power outages, hemorrhage in the community lack of potable water, poor communication), and frequent transfers of trained staff out of their service stations to other departments or Uterotonics (e.g. oxytocin and misoprostol) can prevent postpartum facilities where they could not practice their newly acquired skills. Pol- hemorrhage (PPH) and are customarily administered after delivery by a icy and regulatory support continued long after the tasks had been skilled birth attendant [39]. In 2006 in Nepal, less than 20% of births shifted to better formalize the BEmONC function of CHEWs and ensure were attended by an SBA, and hemorrhage remained the leading sustainability into the future. cause of maternal mortality [40]. From 2005 to 2009, under the leadership of its Family Health Division, the Nepal Family Health Program, and 3.2. India: Reaching rural areas with CEmONC services the Banke District Public Health Office, the Government of Nepal piloted community-based distribution of misoprostol for PPH prevention in In India, 69% of the population lives in rural areas with limited avail- Banke District, targeting home deliveries without SBAs. This interven- ability of providers who can deliver BEmONC and CEmONC [35].In tion shifted uterotonic provision from facility-based providers to female 2006, the Government of India identified this shortfall as an important community health volunteers (FCHVs)—an existing cadre that is factor in the urban/rural disparity in maternal mortality (267 versus deployed across Nepal’s 75 districts. Jhpiego provided technical 619 maternal deaths per 100 000 births, respectively) [36]. Until 2004, assistance to design, monitor, and evaluate the intervention, including only obstetric and gynecologic (ob/gyn) specialist physicians were the adaptation of educational materials and methods. trained and widely authorized to perform cesarean deliveries and sever- National and district level support and ownership of this project al other emergency obstetric skills. In addition, half of ob/gyn specialists helped ensure necessary authorization and funding for the project. in India work in the private sector, leaving a wide gap in public sector Intervention components included: building support among stake- human resources, especially at the primary healthcare level. Although holders; obtaining necessary government approvals; conducting base- more than 25 000 Bachelor of Medicine, Bachelor of Surgery line assessments; establishing monitoring systems; orienting district/ (MBBS) physicians, who function as general medical doctors, are health facility staff; and training and supervision of FCHVs. FCHVs iden- employed in rural areas, India continues to face a severe shortage tified pregnant women, provided prenatal health education, dispensed of ob/gyn physicians who can provide BEmONC or CEmONC in misoprostol tablets late in pregnancy, and made early postnatal home those areas. visits [41]. FCHVs reached nearly 19 000 women in the pilot area. An In 2006, the Government of India and the Federation of Obstetric and endline survey found that 73% of women received misoprostol from Gynaecological Societies of India (FOGSI), with technical assistance an FCHV during their pregnancy and that overall uterotonic coverage in- from Jhpiego, implemented a pilot program to capacitate MBBS physi- creased from 10.7% at baseline to 74.2% at endline (OR 25.0; 95% CI, cians to provide CEmONC. This pilot was implemented in three medical 15.6–40.1) [42]. The largest gains were among the poor, illiterate, and schools and their corresponding 12 district hospital practicum sites in those living in remote areas. The survey also indicated that FCHVs Gujarat and Rajasthan States. The pilot included the development and were positive about their expanded roles and that FCHV distribution establishment of competency-based educational courses for CEmONC of misoprostol was not associated with significant adverse events or in three medical schools, with a minimum of 16 graduates per year incorrect use [42].The maternal mortality ratio among misoprostol per site. Graduates from these courses were deployed to 21 first referral users in the pilot area was 72/100 000, significantly lower than among units to provide CEmONC services. High-level and sustained advocacy non-users (304/100 000) and across the nation (281/100 000) [42]. throughout the pilot resulted in widespread acceptance of the A program review indicated that the pilot’s success was due to competency-based approach by the Government of India, professional several interacting factors: high level of commitment and ownership associations, and medical school stakeholders. by the Banke District Public Health Office; community-level advocacy An independent evaluation in Gujarat State revealed a substantial and education; good teamwork and coordination between governmen- increase in access to CEmONC in four first referral units (from 0%–57% tal and nongovernmental partners; well-trained, carefully supervised to 94%–100%) after a period of only five months [37]. Another indepen- and supported FCHVs; and establishment of effective supply chains for dent study of 10 referral units to which graduates were posted found commodities [41]. In 2010, the Nepal Family Health Program II replicat- that CEmONC provision increased from two facilities to all 10 [36].Fol- ed this model in four mountainous districts with the most limited access lowing the pilot’s success, the Government of India funded the scale-up to SBAs. By 2011, the program was expanded to 21 districts through of this approach to additional states through program management government funding and partnership with international organizations. partners. Among the results of this scale-up were the development of By 2013, the program had been expanded to 31 districts. 34 sites for MBBS training and 235 district practicum sites; the capacitation of 333 district hospital teachers/tutors; and the capacitation of 3.4. Southern Africa: Task shifting for HIV care 1221 MBBS physicians across a total of 22 states for round-the-clock CEmONC services (Personal communication, Bhardwaj A, AVNI Founda- Three quarters of mortality in Sub-Saharan Africa is attributed to tion, Delhi, India, April 28, 2013). communicable, maternal, neonatal, and nutritional causes [43]. HIV is Task shifting of complex CEmONC skills from specialists who have the leading cause of life years lost, and HIV prevalence among adults traditionally provided these services to general practitioners raised un- 15 − 49 years of age is 5% [43,44]. The United Nations Declaration of derstandable questions and concerns about safety and quality. FOGSI, in Commitment on HIV/AIDS [45], coupled with the ongoing crisis in particular, was concerned that it might be held accountable if a graduate human resources for health, necessitated a new approach to service of the CEmONC program was found incompetent or caused harm during delivery when widespread scale-up of ARVs and prevention of service provision. In response, Jhpiego assisted FOGSI in implementing mother-to-child transmission (PMTCT). Since 2005, Jhpiego has con- the Standards-Based Management and Recognition (SBM-R;Jhpiego, tributed to expanding access to essential HIV services across Sub- Baltimore, MD, USA) methodology as the basis for quality assurance Saharan Africa, including PMTCT. Two distinct examples of HIV task and improvement, as well as for rewarding compliance with standards shifting include provision of ARVs by nurses and midwives and HIV S30 B. Deller et al. / International Journal of Gynecology and Obstetrics 130 (2015) S25–S31 testing and counseling by LHWs. In Zambia and Mozambique, LHWs 5. Conclusion were capacitated to competently test and counsel pregnant women and their families for HIV using rapid diagnostic testing, i.e. performing Expanding access to lifesaving MNH services requires innovative a finger prick and interpreting results. In Mozambique, Jhpiego first led methods to ensure that sufficient human resources are in place to a demonstration project with the government and other partners in one meet the needs of women and newborns. Task shifting is one way to site in 2006 to convince policymakers of the soundness of the approach; address the human resource crisis, but its sustainable implementation Jhpiego was then asked to assist the government to formalize the ap- requires a complex interplay of different components: a sound policy proach and roll it out nationally. As of 2013, more than one million and regulatory foundation, attention to qualifications and responsibili- Mozambicans have been tested and counseled in their communities ties, education and training, and service delivery support. Moreover, through these programs. task shifting is only one element of larger health system forces that For ARVs, task shifting progressed with less intentionality as the need to be structured equitably to meet the needs of all mothers, dominant, physician-driven approach to HIV treatment in district and newborns, children, and families. tertiary hospitals and became untenable owing to the large numbers in need of services driven in part by changing international guidelines Conflict of interest that expanded ARV eligibility criteria. Across Southern Africa, nurses and midwives were already managing common opportunistic infections fl and following up women receiving ARVs; widespread adoption of ARVs The authors have no con icts of interest. during pregnancy for all HIV-positive pregnant women in 2012–2013 only increased this demand for services. As a result, several countries References in the region had little choice than to accept and formalize the role of nurses and midwives in initiating and managing clients on ARVs, [1] World Health Organization. Working together for health The World Health Report whether for PMTCT or as treatment for the health of the mother. Jhpiego 2006. Geneva: WHO; 2006. [2] Anyangwe SC, Mtonga C. Inequities in the global health workforce: the greatest supported this expanded scope for nurses and midwives in ARV provi- impediment to health in sub-Saharan Africa. Int J Environ Res Public Health 2007; sion by contributing to the development and implementation of 4(2):93–100. competency-based pre- and in-service education in Mozambique and [3] Dovlo D. Using mid-level cadres as substitutes for internationally mobile health South Africa that enhanced competencies for clinical assessment and professionals in Africa. A desk review. Hum Resour Health 2004;2(1):7. [4] Sibbald S, Shen J, McBride A. Changing the skill-mix of the health care workforce. J management of common and serious opportunistic infections and in- Health Serv Res Policy 2004;9(Suppl. 1):28–38. troduced competencies for initiation and management of clients on [5] Nabudere H, Asiimwe D, Mijumbi R. Task shifting in maternal and child health care: – ARVs, including the management of adverse effects, as a comprehensive an evidence brief for Uganda. Int J Technol Assess Health Care 2011;27(2):173 9. [6] FIGO Committee for the Ethical Aspects of Human Reproduction and Women’s package to ensure high-quality HIV care and treatment service delivery. Health. Task shifting in obstetric care. Int J Gynecol Obstet 2013;120(2):206–7. Challenges to expanding HIV testing and counseling services through [7] Dawson AJ, Buchan J, Duffield C, Homer CS, Wijewardena K. Task shifting and the engagement of LHWs have included retention of volunteers in ser- sharing in maternal and reproductive health in low-income countries: a narrative synthesis of current evidence. Health Policy Plan 2014;29(3):396–408. vice, inadequate remuneration, and government hesitation to include [8] Kasper J, Bajunirwe F. Brain drain in sub-Saharan Africa: contributing factors, the cadre of LHWs as an official position in the public health system. potential remedies and the role of academic medical centers. Arch Dis Child 2012; HIV treatment with ARVs by nurses and midwives has been hindered 97(11):973–9. ’ [9] Hermann K, Van Damme W, Pariyo GW, Schouten E, Assefa Y, Cirera A, et al. by legislation that has been unsupportive of the cadres prescribing abil- Community health workers for ART in sub-Saharan Africa: learning from experience - ity and that has failed to address overall inadequate levels of human re- capitalizing on new opportunities. Hum Resour Health 2009;7:31. sources. Additionally, some providers have resisted, such as pharmacists [10] Celletti F, Wright A, Palen J, Frehywot S, Markus A, Greenberg A, et al. Can the fi deployment of community health workers for the delivery of HIV services represent not wanting to ll nurse prescriptions or nurses themselves not wanting an effective and sustainable response to health workforce shortages? Results of a to take on extra responsibilities. multicountry study. AIDS 2010;24(Suppl. 1):S45–57. [11] Lehmann U, Van Damme W, Barten F, Sanders D. Task shifting: the answer to the human resources crisis in Africa? Hum Resour Health 2009;7:49. [12] World Health Organization. WHO recommendations. Optimizing health worker 4. Discussion roles to improve access to key maternal and newborn health interventions through task shifting. Geneva: WHO; 2012. http://apps.who.int/iris/bitstream/10665/77764/ The four case studies detailed above illustrate the importance of 1/9789241504843_eng.pdf?ua = 1. fi [13] World Health Organization, PEPFAR, UNAIDS. Task shifting: rational redistribution of each of the ve key components of task shifting. Sometimes these tasks among health workforce teams: global recommendations and guidelines. components occur in a sequential manner; other times, they occur http://www.who.int/healthsystems/TTR-TaskShifting.pdf. Published 2008. Accessed simultaneously and continuously. Careful identification of the April 1, 2014. — [14] Joint WHO/OGAC Technical Consultation on Task Shifting. Key elements of a regula- missing task or service and often the geographic area most affected tory framework in support of in-country implementation of “Task Shifting”.Geneva: by its absence—provided a rationale for the shift, while a thorough World Health Organization; 12–13 February 2007. http://www.equinetafrica.org/ understanding of the essential skills needed to perform the task or bibl/docs/OGAhres270307.pdf. Published 2007. Accessed April 1, 2014. fi “ ” [15] Mdege ND, Chindove S, Ali S. The effectiveness and cost implications of task-shifting service justi ed the cadre(s) to which the task was shifted to. in the delivery of antiretroviral therapy to HIV-infected patients: a systematic Effective, competency-based in-service or pre-service education review. Health Policy Plan 2013;28(3):223–36. was foundational to build the requisite capacity. Health system [16] McCarthy CF, Voss J, Verani AR, Vidot P, Salmon ME, Riley PL. Nursing and midwifery regulation and HIV scale-up: establishing a baseline in East, Central and Southern inputs, such as strengthening logistic systems to ensure sufficient Africa. J Int AIDS Soc 2013;16:18051. supplies and the implementation of supportive supervision systems [17] Zachariah R, Ford N, Philips M, Lynch S, Massaquoi M, Janssens V, et al. Task shifting promoted the ability of the “shifted to” cadres as they assumed in HIV/AIDS: opportunities, challenges and proposed actions for sub-Saharan Africa. – responsibility for performance of their newly acquired skills. Trans R Soc Trop Med Hyg 2009;103(6):549 58. [18] Fulton BD, Scheffler RM, Sparkes SP, Auh EY, Vujicic M, Soucat A. Health workforce The critical role of policy and regulatory support cannot be skill mix and task shifting in low income countries: a review of recent evidence. overstated. In each case, the formulation of new policies with sub- Hum Resour Health 2011;9(1):1. stantial government involvement led to formal authorization for [19] Dambisya YM, Matinhure S. Policy and programmatic implications of task shifting in Uganda: a case study. BMC Health Serv Res 2012;12:61. the shifted cadres and created the mechanisms for ensuring the [20] Callaghan M, Ford N, Schneider H. A systematic review of task-shifting for HIV treat- quality of education and training for the tasks shifted. With these ment and care in Africa. Hum Resour Health 2010;8:8. policies in place, quality assurance mechanisms and monitoring [21] Jennings L, Yebadokpo AS, Affo J, Agbogbe M, Tankoano A. Task shifting in maternal and newborn care: a non-inferiority study examining delegation of antenatal and evaluation have been able to assess the strategic value of the counseling to lay nurse aides supported by job aids in Benin. Implement Sci 2011; task-shifting intervention over time. 6:2. B. Deller et al. / International Journal of Gynecology and Obstetrics 130 (2015) S25–S31 S31

[22] du Toit R, Palagyi A, Brian G. The development of competency-based education for [35] Population Reference Bureau. 2012 World Population Data Sheet. Washington, DC: mid-level eye care professionals: a process to foster an appropriate, widely accepted PRB; 2012. http://www.prb.org/pdf12/2012-population-data-sheet_eng.pdf. and socially accountable initiative. Educ Health (Abingdon) 2010;23(2):368. [36] Evans CL, Maine D, McCloskey L, Feeley, Sanghvi H. Where there is no obstetrician – [23] Mavalankar D, Sriram V. Provision of anaesthesia services for emergency obstetric increasing capacity for emergency obstetric care in rural India: an evaluation of a care through task shifting in South Asia. Reprod Health Matters 2009;17(33):21–31. pilot program to train general doctors. Int J Gynecol Obstet 2009;107(3):277–84. [24] FIGO Safe Motherhood and Newborn Health Committee. Human resources for [37] Jhpiego. Strengthening the provision of high-quality emergency obstetric and health in the low-resource world: collaborative practice and task shifting in newborn care services in India in collaboration with the Federation of Obstetric maternal and neonatal care. Int J Gynecol Obstet 2009;105(1):74–6. and Gynaecological Societies of India and the Indian College of Obstetrics and [25] Brentlinger PE, Assan A, Mudender F, Ghee AE, Vallejo Torres J, Martínez Martínez P, Gynaecology. Jhpiego Annual Report March 2007–February 2009. Baltimore, et al. Task shifting in Mozambique: cross-sectional evaluation of non-physician clini- MD: Jhpiego; 2009Published 2009. cians' performance in HIV/AIDS care. Hum Resour Health 2010;8:23. [38] Necochea E, Tripathi V, Kim Y-M, Akram N, Hyjazi Y, da Luz Vaz M. Implementation [26] Gilmore B, McAuliffe E. Effectiveness of community health workers delivering of the Standards-BasedManagement and Recognition approach to quality improve- preventive interventions for maternal and child health in low- and middle-income ment in maternal, newborn, and child health programs in low-resource countries. countries: a systematic review. BMC Public Health 2013;13:847. Int J Gynecol Obstet 2015;130(S2):S17–24. [27] Mwai GW, Mburu G, Torpey K, Frost P, Ford N, Seeley J. Role and outcomes of [39] Smith J, Gubin R, Holston M, Fullerton J, Prata N. Misoprostol for postpartum community health workers in HIV care in sub-Saharan Africa: a systematic review. hemorrhage prevention at home birth: an integrative review of global implementa- J Int AIDS Soc 2013;16:18586. tion experience to date. BMC Pregnancy Childbirth 2013;13:44. [28] Alam K, Tasneem S, Oliveras E. Performance of female volunteer community health [40] Ministry of Health and Population Nepal, New ERA, Macro International. Nepal workers in Dhaka urban slums. Soc Sci Med 2012;75(3):511–5. Demographic and Health Survey 2006. Kathmandu, Nepal: Ministry of Health and [29] Razee H, Whittaker M, Jayasuriya R, Yap L, Brentnall L. Listening to the rural health Population, New ERA, and Macro International Inc.; 2007. http://dhsprogram.com/ workers in Papua New Guinea—the social factors that inﬂuence their motivation pubs/pdf/fr191/fr191.pdf. to work. Soc Sci Med 2012;75(5):828–35. [41] Ejembi CL, Norick P, Starrs A, Thapa K. New global guidance supports community [30] Takasugi T, Lee AC. Why do community health workers volunteer? A qualitative and lay health workers in postpartum hemorrhage prevention. Int J Gynecol Obstet study in Kenya. Public Health 2012;126(10):839–45. 2013;122(3):187–9. [31] Georgeu D, Colvin CJ, Lewin S, Fairall L, Bachmann MO, Uebel K, et al. Implementing [42] Rajbhandari S, Hodgins S, Sanghvi H, McPherson R, Pradhan YV, Baqui AH, et al. nurse-initiated and managed antiretroviral treatment (NIMART) in South Africa: a Expanding uterotonic protection following childbirth through community-based qualitative process evaluation of the STRETCH trial. Implement Sci 2012;7:66. distribution of misoprostol: operations research study in Nepal. Int J Gynecol Obstet [32] Ledikwe JH, Kejelepula M, Maupo K, Sebetso S, Thekiso M, Smith M, et al. Evaluation 2010;108(3):282–8. of a well-established task-shifting initiative: the lay counselor cadre in Botswana. [43] Lozano R, Naghavi M, Foreman K, Lim S, Shibuya K, Aboyans V, et al. Global and re- PLoS One 2013;8(4):e61601. gional mortality from 235 causes of death for 20 age groups in 1990 and 2010: a sys- [33] World Health Organization, UNICEF, UNFPA. Averting Maternal Death and Disability. tematic analysis for the Global Burden of Disease Study 2010. Lancet 2012; Monitoring emergency obstetric care: A handbook. Geneva: WHO; 2009. http:// 380(9859):2095–128. whqlibdoc.who.int/publications/2009/9789241547734_eng.pdf. [44] UNAIDS. UNAIDS report on the global AIDS epidemic 2010. http://www.unaids.org/ [34] Nigeria National Population Commission, ORC Macro. Nigeria Demographic and globalreport/Global_report.htm. Published 2010. Accessed April 1, 2014. Health Survey 2003. Key ﬁndings. Calverton, Maryland: National Population [45] United Nations General Assembly. Declaration of Commitment on HIV/AIDS. A/RES/ Commission and ORC Macro; 2004. http://dhsprogram.com/pubs/pdf/SR101/ S-26/2 http://www.un.org/ga/aids/docs/aress262.pdf. Published 2001. Accessed SR101NG03.pdf. April 1, 2014. International Journal of Gynecology and Obstetrics 130 (2015) S32–S39

Contents lists available at ScienceDirect

International Journal of Gynecology and Obstetrics

journal homepage: www.elsevier.com/locate/ijgo

SUPPLEMENT ARTICLE Experiences engaging community health workers to provide maternal and newborn health services: Implementation of four programs

Jaime Haver a,⁎, William Brieger a, Jérémie Zoungrana b, Nasratullah Ansari c,JeanKagomad a Jhpiego, Baltimore, MD, USA b Jhpiego, Dar es Salaam, Tanzania c Jhpiego, Kabul, Afghanistan d Jhpiego, Kigali, Rwanda article info abstract

Keywords: A paucity of skilled health providers is a considerable impediment to reducing maternal, infant, and under-ﬁve Community health worker mortality for many low-resource countries. Although evidence supports the effectiveness of community health Human resources for health workers (CHWs) in delivering primary healthcare services, shifting tasks to this cadre from providers with ad- Primary health care vanced training has been pursued with overall caution—both because of difﬁculties determining an appropriate Maternal and newborn health package of CHW services and to avoid overburdening the cadre. We reviewed programs in Rwanda, Afghanistan, Nigeria, and Nepal where tasks in delivery of health promotion information and distribution of commodities were transitioned to CHWs to reach underserved populations. The community-based interventions were complementary to facility-based interventions as part of a comprehensive approach to increase access to basic health services. Drawing on these experiences, we illuminate commonalities, lessons learned, and factors contributing to the programs’ implementation strategies to help inform practical application in other settings. © 2015 Published by Elsevier Ireland Ltd. on behalf of International Federation of Gynecology and Obstetrics. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

1. Introduction health worker—a term often used interchangeably with CHW—as any health worker who: (1) performs functions related to healthcare deliv- The Alma Alta Declaration of 1978 incorporated volunteer commu- ery; (2) has received some form of training relevant to the given inter- nity health workers (CHWs) into the delivery of basic health services vention; and (3) does not hold a formal professional or paraprofessional at the village level [1] and paved the way for the proliferation of CHW certificate or tertiary education degree. Although the profile of CHWs programs [2]. More than 30 years later, many low-income countries varies across countries, common attributes such as being recruited are striving to increase the numbers of skilled providers as a way to im- and supported by the community served [8] may uniquely position prove coverage of and access to basic health services, including mater- the cadre to help address some of the health system challenges that nal and newborn health (MNH), family planning, and nutrition [3]. affect access to health facility services. Nevertheless, efforts to strengthen facility services have often not kept pace with the health system requirements needed to provide all citizens 1.1. Current community health worker situation access to care [4]. The ongoing crisis in human resources for health remains one of the most critical system challenges, resulting from a severe Globally, many women, particularly in rural areas, continue to give paucity in the number of providers, inappropriate distribution of birth at home without the presence of a skilled provider [9], and families existing providers, and insufficient capacity of providers due to lack of in such settings often seek treatment for child illnesses from informal training and education [5]. In this context, discussion around the appro- providers such as medicine vendors [10]. These challenges underscore priate role of CHWs in reducing maternal, infant, and under-five mortal- critical gaps in healthcare access. National MNH survival strategies ity has been revitalized [6,7]. must employ complementary facility- and community-based strategies, The term “community health worker” encompasses the roles and re- which in some cases might be accomplished by CHWs delivering health sponsibilities of various health cadres [6]. Lewin et al. [7] defines a lay promotion information and distributing commodities to communities to achieve greater coverage of priority interventions. Barros et al. [11] reviewed inequalities in maternal, newborn, and child health interven- ⁎ Corresponding author at: Jhpiego, 1615 Thames Street, Baltimore, MD 21231, USA. Tel.: +1 530 592 3270. tions and concluded that community-based interventions are more eq- E-mail address: [email protected] (J. Haver). uitable than static facility-based services. Thus, task shifting to CHWs to http://dx.doi.org/10.1016/j.ijgo.2015.03.006 0020-7292/© 2015 Published by Elsevier Ireland Ltd. on behalf of International Federation of Gynecology and Obstetrics. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). J. Haver et al. / International Journal of Gynecology and Obstetrics 130 (2015) S32–S39 S33 provide outreach services in rural communities can help increase over- 3. Case studies all access to basic interventions [12,13]. The role and need for CHWs has been examined in relation to To provide a general overview of the four selected CHW programs, programmatic opportunities for task shifting, and recommendations key information is summarized in Table 1. for program integration have been issued by both WHO [14] and USAID on this topic [15]. Community-level interventions in MNH 3.1. Case study 1: Improving the health of women and newborns in have focused on health promotion and distribution of commodities Rwandan communities such as iron, folic acid, and vitamin A supplementation [16]. CHWs have also been shown successful in strengthening management of Over the past decade, Rwanda has witnessed unprecedented im- uncomplicated child fever cases in community case management provement in many health outcomes, including those related to MNH, (CCM) of malaria [17], distributing and promoting contraceptives malaria, and HIV—due, in part, to the government’s pioneering of health [18], and promoting exclusive breastfeeding [7], although not always system reforms and decentralization of the healthcare delivery system. sustainably at scale. Community-based MNH care was built upon this strong existing foun- While there is evidence, at least in pilot settings, that CHWs can ef- dation and has been embedded within the Ministry of Health (MOH) fectively deliver interventions when adequately trained and supervised, through community health point persons at the central, district, health debate has persisted on how to best utilize the cadre. The global com- facility, and cell (i.e. group of villages) levels. munity has shifted responsibilities to CHWs with overall caution, focus- Of the three CHWs selected per village, two (one female and one ing largely on the intervention itself, rather than on populations in need male) managed integrated community case management of childhood who are not reached by the current health system [2]. Exploiting oppor- illness (iCCM), a strategy used to train, support, and supply CHWs capa- tunities to devolve select interventions to CHWs might help address the ble of diagnosing and treating childhood illnesses including pneumonia, human resources for health crisis by enabling skilled providers to focus diarrhea, and malaria [19]. A third (female) CHW, called an Animatrice on those services (e.g. diagnostic, therapeutic) that require a higher de santé Maternelles (ASM), focused specifically on pregnant women. level of education or training. In 2010, the Home-Based Maternal Neonatal Health Care program, supported by Jhpiego in partnership with Save the Children under the Ac- cess to Clinical and Community Maternal, Neonatal and Women’s 1.2. Jhpiego’s experience with community health workers Health (ACCESS) Program and later the Maternal and Child Health Inte- grated Program (MCHIP), was initiated to increase the capacity of the As part of a comprehensive approach to care, Jhpiego has worked ASMs. ASMs cover a catchment area of 100–150 households and are re- with CHWs in numerous countries as a complement to facility-based sponsible for: (1) registering all women of reproductive age and identi- care and a key strategy for reaching underserved women and their fam- fying those who are pregnant in the community to encourage prenatal ilies with lifesaving services with a focus on MNH. The objective of the care attendance and facility-based deliveries; (2) promoting healthy be- present article is to illuminate commonalities and lessons learned haviors during pregnancy and the postpartum period; (3) accompany- from four country programs in which tasks in health promotion and dis- ing women in labor to the health facility; (4) providing misoprostol (as tribution of commodities were intentionally shifted from skilled pro- a uterotonic) in advance to pregnant women for self-administration viders to CHWs to advance MNH strategies. These illustrative following birth if they deliver unexpectedly at home; and (5) making experiences, presented as case studies, are assessed—with a focus on early postpartum home visits to identify danger signs and refer women CHW scopes of work and factors contributing to the effective imple- to the health facility as needed. mentation of the programs—to help inform practical application in As part of the Home-Based Maternal Neonatal Health Care program, other settings. the ASMs were provided with an MNH care kit and training on how to use the included supplies. ASMs were selected according to specific criteria by community members from the village catchment area 2. Methods under the supervision of the village coordinator, in consultation with the facility health worker responsible for community health. A meeting Keeping in mind Lewin et al.’sdefinition of a lay health worker [7], with the village leaders and the community health supervisor was then which aligns with Jhpiego’sdefinition of a CHW, we reviewed program held to finalize the selection. Selected ASMs received six days of initial reports and other documents collected from Jhpiego’s internal reposito- education in community-based MNH services and refresher training ry and program staff’s inventories to identify those programs that met after six months. Previously, also under ACCESS, Jhpiego had worked the following criteria: (1) CHWs delivered health promotion informa- to improve the established ASM system through a wide range of tion and distributed commodities to the community for curative and capacity-building efforts. preventive interventions in support of MNH; (2) CHWs performed a The program was first implemented in a few districts; the MOH later task that had been shifted from a skilled provider; and (3) data on decided to take the program to national scale. To help facilitate scale-up, MNH services were available. innovative technologies were applied, such as the use of mobile phones We specifically looked for clear examples of “task shifting” because for rapid short message service, which allowed ASMs to transmit infor- of the potential for this practice to provide relief to overburdened mation into a computerized recording and response system. Each ASM health systems and to reach women and children who are often in the piloting zones received a mobile phone to report indicators—on overlooked. Programs that included components known to support community interventions (e.g. iCCM, community-based MNH services), CHW efforts (e.g. community involvement) were also given special deaths, and other information—directly to SISCom, Rwanda’s health consideration, in line with our focus on factors for effective management information system. In addition, the ASMs sent alarm implementation. messages to the health center to help refer pregnant women quickly. After the program documents were reviewed and discussed among CHWs were organized by the local government minister into coop- staff who were knowledgeable about the CHW programs that fit our in- eratives composed of approximately 120 CHWs within the facility clusion criteria, we selected four that best met our criteria and were catchment area, which provided a supportive framework of peer-to- “representative,” collectively, on a current and global level, in terms of peer support, motivation, and accountability. For example, under the setting/culture and range of challenges. Drawing from these programs, MOH’s coordination, the CHWs conducted income generation activities we developed the following case studies to highlight information most to provide a means of subsistence, while the range and extent of activi- pertinent to our objective. ties were driven by the cooperative. The cooperative program was also S34 J. Haver et al. / International Journal of Gynecology and Obstetrics 130 (2015) S32–S39

Table 1 Jhpiego community health worker programs selected for case studies.

Countries

Rwanda Afghanistan Nigeria Nepal

Focus of program MNH and iCCM PPFP Malaria in pregnancy PPH intervention Tasks shifted to 1. CHW: Treatment of diarrhea, Contraceptives to postpartum IPTp provision, insecticide-treated Advance distribution of misoprostol CHWs from malaria and pneumonia; RDT test; women, including injections net distribution for self-administration at home formally trained direct observation of tuberculosis births provider treatment; provision of family planning; IEC 2. ASM: Identification of pregnant women; birth preparedness; accompany women to deliver at facility; distribution of misoprostol for home deliveries; newborn care; screening for malnutrition for children and pregnant and breastfeeding women; IEC Program years 2010–2011 2007–2011 2007–2011 2005–2007 Total number of 16 447 out of the total 45 000 in the 14 389 out of the total 19 000 in the 735 At scale: 16 920 CHWs in program country country Geographic scale National: all 30 districts, among National: all 34 provinces of the 7 local government areas (districts) Pilot: 1 district (Banke) them 11 supported by MCHIP in country At scale: 29 districts 2010–2011 Government Female; 25–50 years old; primary Female; willing to work as a Willing to work as volunteer; Self-motivated; willing to work as selection criteria school education; literate; resident volunteer; resident in village where resident in village where working; volunteer; willing to act as a focal in village where working; available working; respected by community; perceived as honest and acceptable person to bridge health programs for home visits; respected by motivated to serve as a CHW by community with community community; volunteer; perceived as honest Government Initial 2 weeks initial; 6 days on the 9 weeks initial; 5 days on the 9 weeks initial; 5 days on the 18 days initial in two phases; 7 days Training and intervention; refresher after 6 intervention; intervention; refresher after 6 on the intervention; periodic program months refresher after 6 months months refresher training intervention training Linkages to health Community focal point at health Monthly meetings with supervisor Clinic provided trainers and Trainers of FCHVs were facility center level and district hospital and health facility provider; supervisors and managed government staff in health facilities level; CHWs call ambulances; referrals on LAPM to health facilities commodities for collection or district health offices; FCHVs access to prenatal care facility data reported to facilities; FCHVs shared for follow up; reference and progress in district semiannual counter-reference system clear and review meeting operational Reward/ Funds from income generation Community recognition, traditional Community recognition and sense Government organized FCHV day motivation/support activities, materials such as registers and creative non-monetary means of obligation to his/her clan; celebration and endowment fund; and counseling and screening cards, (e.g. home repair, free materials such as counseling flip information, education and and supplies including transportation) chart communication materials demonstration cloths for dry wrap and kangaroo mother care Impact on MNH 150 207 women in labor were Injectable contraceptives provided Proportion of pregnant women 840 postintervention survey accompanied by ASMs to deliver in at health posts increased from 49 taking at least two respondents, 73.2% received health facilities; 19 248 pregnant 922 (April 2006–March 2007) to sulfadoxine-pyrimethamine doses misoprostol, and uterotonic women were accompanied to the 103 898 (April 2009–March 2010); during pregnancy was 5 times in the coverage increased from 11.6% health center for pregnancy-related estimated contraceptive prevalence CDI communities compared with 3 before the intervention to 74.2% danger signs identified by the ASMs rate had reached 20%, compared times in the control group, for post intervention with 10% (2010) whom IPTp was available only at prenatal care (P b 0.001)

Abbreviations: ASM, animatrice de santé maternelles; CHW, community health worker; FCHV, female community health volunteer; iCCM, integrated community case management; IEC, information, education, communication; IPTp, intermittent preventive treatment in pregnancy; LAPM, long-acting and permanent methods; MNH, maternal and newborn health; MCHIP, Maternal and Child Health Integrated Program; PPFP, postpartum family planning; PPH, postpartum hemorrhage; RDT, rapid diagnostic test. linked to a national health performance-based system to incentivize MNH. In the same time period, in those six districts, 19 248 pregnant CHWs on an ongoing basis, based upon achievement and completeness women were accompanied to the health center for pregnancy-related of 26 indicators (e.g. number of women accompanied to the facility for danger signs identified by the ASMs, and 52.7% of women who delivered delivery). Each cooperative could receive money on a quarterly basis; in a facility were accompanied by ASMs [20]. on average, a cooperative received between US $1500 and $6000. Based upon the quarterly performance-based financing invoice, 70% of 3.2. Case study 2: Integration of postpartum family planning with existing the payment was directed to the cooperative, and 30% was directed to MNH services in Afghanistan CHWs as an individual incentive. To preclude adverse incentives, the system concentrated on preventive activities rather than curative care. Following the exit of the Taliban in 2001, health indicators in The cooperatives paradigm has been regarded as a solution for mo- Afghanistan were highly unfavorable. The Ministry of Public Health tivating and sustaining CHWs and supporting their integration into (MOPH) placed emphasis on community-based health care as part of the health system. Between July 2010 and June 2011, a total of 2433 its strategy to expand access to and availability of basic health services, ASMs in six of the 30 districts received training in community-based including family planning, as a pillar of maternal health for preventing J. Haver et al. / International Journal of Gynecology and Obstetrics 130 (2015) S32–S39 S35 unintended pregnancies [21]. Postpartum family planning (PPFP) had CHWs monetary and non-monetary incentives such as bicycles, literacy also been brought to the fore as a lifesaving intervention, having been training, or furniture for health posts [25]; however, the recent national repositioned within the last decade—in part by the ACCESS-FP Program Community-Based Health Care Strategy acknowledges that support (an associate award of the ACCESS Program) and proponents of the lac- from communities or shura to CHWs is variable [26]. The MOPH’s ex- tational amenorrhea method (a short-term contraceptive method based pansion of access to quality health care through the BPHS, which in- upon patterns of breastfeeding). cludes community-based services, has led to increased services for In 2007, ACCESS-FP, in collaboration with the Health Services Sup- women, newborns, and children, as well as improved health indicators. port Project (HSSP), partnered with the MOPH and the organization While the overall expansion of health services has undoubtedly played a Management Sciences for Health to introduce PPFP services to increase role in this increase, select indicators highlight the possible contribution contraceptive utilization, address unmet need for contraception, and of PPFP services delivered by CHWs. From 2007 to 2009, 13 provinces in promote healthy birth-to-pregnancy intervals of at least 24 months. the country piloted community-based PPFP services. Following the pilot The existing national community-based healthcare system positioned period, by the end of 2011, the program had rolled out to all 34 prov- these volunteers as frontline health workers within a basic package of inces. In total, 14,389 CHWs out of the 19,000 CHWs in the country re- health services (BPHS) [22] at the health post level, their primary func- ceived in-service training in PPFP [27]. In addition, National Health tions being health promotion and limited curative care. Management Information System data indicates that the number of in- CHWs were selected by community health shura (councils) after jectable contraceptives provided at health posts by CHWs increased meeting basic selection criteria and were given an eight-week pre- from 49,922 (April 2006–March 2007) to 103,898 (April 2009–March service training [23]. With the strong sociocultural preference for 2010). By 2010, the Afghanistan Mortality Survey estimated that the female workers, 50% of CHWs were women. Often the female CHW at contraceptive prevalence rate had reached 20%, compared with 10% es- the health post was a relative of the male CHW, which helped increase timated by the Multiple Indicator Cluster Survey in 2003 (Fig. 1) cultural acceptability of female CHW roles. [28–32]. The 2010 Afghanistan Mortality Survey also estimated the Before commencement of the PPFP program under HSSP, the country’s median birth interval at 27 months in the preceding 5 years. roles of a female CHW included conducting household visits to deliv- Although data on birth intervals before the survey were not available, er counseling in prenatal and postpartum care, as well as providing the estimate is promising, given that global evidence indicates that counseling on family planning and distributing oral contraceptives, birth-to-pregnancy intervals of less than 24 months are associated condoms, and follow-up doses of injectable contraceptives. Addi- with adverse maternal and neonatal outcomes. tionally, the Accelerating Contraceptive Use project, led by Manage- As a result of the progress observed from the PPFP services provided ment Sciences for Health, supported CHWs in initiating use of by CHWs, the MOPH integrated these services into policy documents, injectable contraceptives and administering the first injection on a including the national Reproductive Health Policy and Strategy, and pilot basis [24]. The PPFP program used the existing cadre, roles, into the BPHS. With the help of donors and the nongovernmental orga- and lessons learned from this pilot as a platform to build CHW capac- nizations delivering the BPHS, CHW-delivered PPFP services continue in ity to provide counseling on PPFP, including providing the first dose Afghanistan to date. Efforts are underway to integrate the intervention of injectable contraceptives to postpartum women. within the national curriculum for CHWs, as well as to incorporate indi- After identifying pregnant and postpartum women in their catch- cators into the National Health Management Information System. ment areas through community mapping, CHWs were expected to conducted five household visits at specified times: the first during the 3.3. Case study 3: Employing a community-directed intervention for malaria eighth month of pregnancy and the other four during the postpartum control in Nigeria period. The CHW: (1) provided counseling—assisted by pictorial flip charts—on healthy timing and spacing of pregnancies, exclusive In 1995, the African Program on Onchocerciasis Control (APOC) breastfeeding, the lactational amenorrhea method, the return to fertility managed treatment of onchocerciasis—a parasitic disease that can after pregnancy, transition from the lactational amenorrhea method to cause blindness—at the community level through a strategy entitled other modern methods of contraception, and adverse effects of various “community-directed treatment with ivermectin” (CDTI) [33]. CDTI methods; (2) distributed condoms and oral contraceptives; (3) adminis- harnesses community engagement through a community-clinic part- tered injectable contraceptives; and (4) made referrals to facilities for nership and empowers members to actively participate in delivery of long-acting methods. the drug. Under APOC, the “clinic” side of the partnership involved The strategy for initiating and expanding PPFP services by CHWs health facility staff: (1) reaching out to communities in their catchment was four-pronged: (1) mobilizing the community through the health area and encouraging them to take responsibility for the handling and shura and leaders; (2) building capacity of the CHWs through in- distribution of ivermectin; and (2) training community volunteers to service training on PPFP; (3) supporting the CHWs with supervision; distribute the drug, maintaining stocks of ivermectin at the facility, and (4) collecting information on contraception distributed by CHWs and collecting completed distribution records. The “community” side for monitoring. The expansion strategy was aided by a number of activities and implementing partners. For instance, HSSP supported the MOPH in training CHW trainers in PPFP; in turn, these trainers replicat- ed the CHW trainings at the provincial level with the assistance of nongovernmental organizations contracted by the MOPH to deliver the BPHS. The establishment of a pool of provincial trainers helped to both expand the package of services and promote sustainability. Because CHWs were accountable to health shura for performance and community satisfaction, community sensitization meetings were conducted with the shura and with community and religious leaders to provide messages on PPFP and galvanize support for the program. In these meetings, the community health supervisors provided information on and addressed myths and misconceptions about the purpose and benefits of PPFP, means of service delivery, and the role of the CHW. Operations research in Afghanistan on the national CHW program indicate that some, but not all, NGOs implementing the BPHS provided Fig. 1. Contraceptive prevalence rate in Afghanistan, 2003–2010 [24–28]. S36 J. Haver et al. / International Journal of Gynecology and Obstetrics 130 (2015) S32–S39 further engaged the community in planning ivermectin distribution, 3.4. Case study 4: Preventing postpartum hemorrhage at home birth in choosing volunteer community-directed distributors (CDDs) (CHWs Nepal in this context), and undertaking village censuses. Communities chose CDDs from among male and female neighbors who were reliable, The Nepalese Government created the female community health vol- respectable, and accountable [34]. Literacy was not a requirement, unteer program (FCHV) in 1988/89—a group of CHWs that has grown to but usually at least one CDD in a village was literate in order to 48 000 volunteers across the country. Selected by the communities, the maintain records. FCHVs receive 18 days of initial training in addition to a five-year cycle Given the success that APOC experienced in reaching 100 000 villages of five-day refresher training in maternal and child health based on the with CDTI, Jhpiego and other health programs realized the strategy’spo- government’sFCHVpolicy(2003)[41] and periodic refresher training. tential as a model for other health and social interventions [35]. CDDs be- Their responsibilities have included health promotion and distribution came involved in immunization outreach, control of other diseases, of commodities such as vitamin A, deworming tablets, and iron folate. agricultural extension, and general health promotion, which led the Unit- In 2006, the maternal mortality ratio in Nepal was 281 per 100 000 ed Nations Development Programme (UNDP)/World Bank/WHO/UNICEF live births [42], and nearly half of these deaths were attributable to Special Program on Research and Training in Tropical Diseases to design PPH [43]. The Nepal Family Health Program (NFHP), which aimed to im- intervention research to test CDTI with other health services. In addition prove MNH outcomes, explored the feasibility, acceptability, and safety to enhanced ivermectin coverage, this multicountry project found that of a program for advanced distribution of misoprostol to pregnant the approach, now termed “community-directed intervention” (CDI), women for self-administration after home births for prevention of increased coverage of malaria treatment, bed net use and prevention PPH. This intervention, which was implemented by John Snow, Inc., services, and vitamin A supplementation compared with traditional Jhpiego, Save the Children and partners, was piloted in Banke district, facility-based services, as well as assisted in directly observed treatment where there were 665 existing FCHVs in the government health system. for tuberculosis [36] and intestinal helminth control [37]. The FCHVs had received previous training on other aspects of their Jhpiego, recognizing the community involvement principles behind assigned tasks, worked 3–8 hours per week, and served for an average the successful implementation of CDI, adapted the approach to address of five years. Most of the FCHVs were illiterate [44]. For the purpose of malaria in pregnancy in Akwa Ibom State, a highly endemic area of this pilot on PPH prevention, the FCHVs received an additional seven Nigeria where uptake of intermittent preventive treatment in pregnan- days of training focused on the intervention, which involved identifying cy (IPTp) and insecticide-treated net distribution had been very low pregnant women in their catchment area, providing prenatal counsel- [38]. In the context of Jhpiego’s work, a pilot project was conceived on ing, and distributing misoprostol to women who were eight months community-directed delivery of malaria in pregnancy control interven- pregnant for self-administration at home births. tions. The clan (large group of families) served as the platform for com- Results showed that of the 840 post-intervention survey respon- munity engagement. Each clan included about 100 residents and dents, 73.2% received misoprostol, and uterotonic coverage increased was encouraged to select at least two CDDs to help manage interven- from 11.6% before the intervention to 74.2% after the intervention tions such as health promotion, IPTp provision, recordkeeping, and [44]. The most extensive improvements in uterotonic coverage were insecticide-treated net distribution. Primary healthcare facility staff re- observed in the two lowest wealth strata. This successful pilot program ceived in-service training in malaria in pregnancy and malaria; they added to the increasing body of evidence demonstrating that trained then reached out to the leadership and social structures at the village CHWs could effectively deliver misoprostol for self-administration by and clan levels. Every catchment area consisted of four to six villages, women at home deliveries [45], although scale-up has been challenged and each village contained approximately a dozen clans. Primary in areas such as consistent supply of misoprostol [46]. However, it is a healthcare facility staff trained the chosen volunteers, but the clan as a compelling example of the contribution that CHWs can make to im- whole kept the volunteers accountable and helped with tasks like proving equity of health interventions. census-taking and recordkeeping. Importantly, the involvement of CDDs in delivering IPTp was endorsed through a meeting of stake- 4. Discussion holders in the state to overcome skepticism about community capabilities and gain government commitment. The Alma Alta Declaration of 1978 trumpeted the provision of pri- Although CDDs frequently asked about monetary motivation and in- mary health care for all by 2000. Although universal coverage of MNH centives, the idea of having the smallest community unit, the clan, services has not yet been achieved, the CHW programs reviewed in choose at least two CDDs helped ensure that the workload did not over- this article have shown that CHWs can contribute to increased coverage whelm volunteers; the arrangement also capitalized on the likelihood of interventions in pilot and early scale-up efforts in different settings. that the volunteer would have a sense of obligation to his/her own Scale-up presents many challenges, as reported by Smith et al. [47] in clan. This intrinsic motivation enabled volunteers to see CDI as a this supplement. Successful national MNH survival strategies will re- way to help relatives and neighbors, while also gaining recognition quire not only facility-based strategies to provide care during pregnan- for providing a valuable service to their communities. CDD attrition cy, childbirth, and the postpartum period, but also community-based was not an issue because maintaining these human resources was strategies that bring services directly to the household level. regarded as the responsibility of the whole community (with the In each of the case studies presented, the CHWs undertook a variety support of the primary healthcare facility), which was always ready of MNH tasks that had been shifted to them from providers with more to identify new CDDs. advanced training. This shift was made possible through mutual con- The pilot that used the CDI process for malaria in pregnancy sensus between the health system, which approved the tasks, and the achieved significantly higher coverage of IPTp and insecticide-treated community, which had clear needs for the services. The health promo- nets and prenatal care attendance in intervention communities than tion and distribution of commodities afforded by these community- control communities without CDI. The effects of the CDI program were based strategies yielded greater uptake of interventions than would largest for IPTp adherence, increasing the proportion of pregnant have been achieved through facility-based services alone. women taking at least two sulfadoxine-pyrimethamine doses during pregnancy by five times in the CDI communities compared with three 4.1. Supportive program components times in the control group, for whom IPTp was available only at prenatal care (P b 0.001) [39]. Based on this pilot CDI program, the communities Drawing on these experiences, we identified cross-cutting health were willing to expand their efforts into iCCM, on a trial basis in system and community components that facilitated the programs’ two communities [40]. achievements through the use of CHWs, illustrated in Fig. 2. J. Haver et al. / International Journal of Gynecology and Obstetrics 130 (2015) S32–S39 S37

Fig. 2. Key components of successful community health worker (CHW) programs.

All programs built upon an existing platform for CHW delivery of in- messages within their Friday prayers, helping to achieve broader promo- formation and services, such as family planning in Afghanistan and pre- tion of PPFP within communities. natal care in Rwanda. Programs facilitated better linkages with health While community-based interventions have been found to be more facility services so that a complementary facility-community approach equitable than facility-based services, programs must still ensure that could be employed, often at the local level. In Afghanistan, however, underserved or disadvantaged areas receive access. In Rwanda, for ex- the intervention was introduced as a vertical intervention; later, once ample, despite a larger, more developed CHW presence than in many the program had demonstrated feasibility, the intervention was inte- comparable settings, many women and children are still not being grated horizontally at the national level within the larger purview of reached. To address this gap, the government is initiating a mapping ex- CHW responsibilities. ercise by CHWs to create a visual representation of catchment areas, in- Capacity building of CHWs was achieved initially through refresher cluding the location of all households. The “map” will be used by CHWs training, followed by supportive supervision. Refresher training rein- as an aid in monitoring delivery of interventions. forced the initial training received by CHWs as implementation Absence of political commitment is another category of challenge progressed. As the intervention became more ingrained within the that, alone, might circumvent CHWs from reaching their full potential, roles of CHWs, the knowledge and skills needed to perform it could be arising from a persisting notion that CHWs are only a temporary solu- integrated into their initial education. tion within the health system [48]. Although the 2012 WHO guidelines Community engagement was integral throughout implementation, for the prevention and management of PPH recommend the adminis- from community members selecting the CHWs to community members tration of misoprostol by CHWs [49], polarized discussion continues acting as equal and active partners in providing primary healthcare ser- about whether members of this cadre can provide misoprostol to vices. This engagement, in turn, helped to influence intervention uptake, women in advance of childbirth for self-administration after a home as well as CHW accountability and dedication. Whether taking the form birth, to prevent PPH. Despite mounting implementation evidence in of clan members in Nigeria or religious leaders in Afghanistan, the com- favor of this practice to prevent PPH [45],manycountriesarguably munity proved to be a potent force in programmatic success. lack the political will to put this lifesaving measure into the hands of A motivational element for CHWs was also shown to be important. CHWs —their most prevalent, far-reaching providers. In contrast, Globally, discussion about how to reward CHW performance and community-based distribution of misoprostol in Nepal was done whether CHWs should be paid is ongoing. Although we do not answer through the government health system, illustrating ample political that question, we have observed that all of these programs used ap- support. In fact, all four of the programs described in the present ar- proaches that targeted benefits to the CHW. These include both extrin- ticle operated within the government health system, and to date the sic benefits, such as contributions to a CHW savings fund in Rwanda, governments have continued to sustain their respective interven- and intrinsic benefits, such as community accountability to a clan as tions, integrating them into national policies, health management the primary impetus that fueled CHW motivation in Nigeria. and information systems, and logistics systems. Political support is Sufficient political commitment, including a steady supply of com- therefore paramount for CHW programs to continue expanding ac- modities, was integral for the approval of the tasks performed by cess to basic health services. CHWs. Moreover, government support of CHWs as a permanent component of a comprehensive health system also played a role in success. 5. Conclusion

CHWs can take an active role in the delivery of community-based 4.2. Addressing challenges primary healthcare interventions linked to the health facility, as posited by Alma Alta. As illustrated by four Jhpiego-led programs, CHWs have In these four country experiences, Jhpiego worked to address context- demonstrated that they can effectively deliver MNH and family plan- specific challenges to facilitate CHW program success. When the ning information and distribute commodities that were once regarded community-based PPFP initiative commenced in Afghanistan, for exam- as functions of formally-trained health workers. Other studies should ple, efforts were initially challenged by misconceptions and restrictive so- be conducted to explore the potential for other services to be ciocultural beliefs about contraception. One approach to address this transitioned to CHWs, such as screening for pre-eclampsia/eclampsia issue was to develop PPFP messages consistent with the teachings of and providing a loading dose of the anticonvulsant magnesium sulfate Islam. For example, the lactational amenorrhea method—in which fully to women with eclampsia. or nearly fully breastfeeding is one of three criteria for successful use of Many factors must be weighed when considering the devolution of the method—was linked with Islam’s promotion of breastfeeding. These tasks to CHWs and effective implementation of related programs. How- messages resonated with target groups and created common ground ever, these case studies demonstrate that there are MNH/family plan- whereCHWswereabletoconnectwithreligiousleadersonthesubject. ning interventions that can be facilitated by CHWs to complement Consequently, some mullahs integrated lactational amenorrhea method facility services. Through improved access to MNH/family planning S38 J. Haver et al. / International Journal of Gynecology and Obstetrics 130 (2015) S32–S39 interventions, the scale-up of community-based interventions has the [18] Prata N, Vahidnia F, Potts M, Dries-Daffner I. Revisiting community-based distribu- – fi tion programs: are they still needed? Contraception 2005;72(6):402 7. potential to reduce maternal, infant, and under- ve mortality. Other [19] Marsh DR, Hamer DH, Pagnoni F, Petersen S. Introduction to a special supplement: countries and CHW programs can apply these lessons learned to in- Evidence for the implementation, effects, and impact of the integrated community crease uptake of MNH/family planning interventions in similar settings. case management strategy to treat childhood infection. Am J Trop Med Hyg 2012; 87(Suppl. 5):2–5. [20] Republic of Rwanda Ministry of Health. Annual Report: July 2010–June 2011. Kigali: Acknowledgments Republic of Rwanda Ministry of Health; 2011. [21] Cleland J, Bernstein S, Ezeh A, Faundes A, Glasier A, Innis J. Family planning: the fi – The authors thank Jeffrey Michael Smith, Partamin Zainullah, un nished agenda. Lancet 2006;268(9549):1810 27. [22] Islamic Republic of Afghanistan Ministry of Public Health. A Basic Package of Health Emmanuel O. Otolorin, Bright Clement Orji, and Chandra Rai for their Services for Afghanistan. Kabul: Islamic Republic of Afghanistan Ministry of Public contributions to this article. Health; 2005. [23] Islamic Transitional Government of Afghanistan Ministry of Health. Policy on community based health worker. Kabul: Islamic Transitional Government of Conflict of interest Afghanistan Ministry of Health; 2003. [24] Huber D, Saeedi N, Samadi AK. Achieving success with family planning in rural fl Afghanistan. Bull World Health Organ 2010;88(3):227–31. The authors have no con icts of interest. [25] Islamic Republic of Afghanistan Ministry of Public Health, General Directorate of Pol- icy and Planning, Monitoring and Evaluation Department. Operations Research References Study on Community Health Worker Performance in Afghanistan; Findings 2006 − 2007; Afghanistan National Health Services Performance Assessment. [1] World Health Organization. Primary Health Care. Report of the International Confer- Kabul: Islamic Government of Afghanistan Ministry of Public Health; 2007. ence on Primary Health Care, Alma-Ata, USSR, 6 − 12 September, 1978. Geneva: [26] Islamic Republic of Afghanistan Ministry of Public Health. Community-Based Health WHO; 1978. http://www.searo.who.int/entity/primary_health_care/documents/ Care Policy and Strategy, 2009 − 2013. Kabul: Islamic Government of Afghanistan hfa_s_1.pdf. Ministry of Public Health; 2009. [2] Perry H, Zulliger R. How effective are community health workers? An overview of [27] Jhpiego. Health Services Support Project Final Report: July 1, 2006 − October 31, current evidence with recommendations for strengthening community health 2012. Baltimore: Jhpiego; 2012. worker programs to accelerate progress in achieving the health-related Millennium [28] Afghan Public Health Institute, Ministry of Public Health (APHI/MoPH) Development Goals. http://www.coregroup.org/storage/Program_Learning/ [Afghanistan], Central Statistics Organization (CSO) [Afghanistan], ICF Macro, Community_Health_Workers/review%20of%20chw%20effectiveness%20for% Indian Institute of Health Management Research (IIHMR) [India], World Health Or- 20mdgs-sept2012.pdf. Published 2012. Accessed April 12, 2014. ganization Regional Office for the Eastern Mediterranean (WHO/EMRO) [Egypt]. [3] Campbell OM, Graham WJ. Lancet Maternal Survival Series steering group. Strategies Afghanistan Mortality Survey 2010. Calverton, Maryland: APHI/MoPH, CSO, ICF for reducing maternal mortality: getting on with what works. Lancet 2006; Macro, IIHMR, and WHO/EMRO; 2011. http://dhsprogram.com/pubs/pdf/FR248/ 368(9543):1284–99. FR248.pdf. [4] Costello A, Azad K, Barnett S. An alternative strategy to reduce maternal mortality. [29] Central Statistics Office Afghanistan Transitional Authority, UNICEF. Moving beyond Lancet 2006;368(9546):1477–9. 2 decades of war: progress of provinces. Multiple Indicator Cluster Survey 2003. [5] Chen L, Evans T, Anand S, Boufford J, Brown H, Chowdhury M, et al. Human re- http://www.childinfo.org/files/AfghanistanMICS2003.pdf. Published 2004. Accessed sources for health: overcoming the crisis. Lancet 2004;364(9449):1984–90. April 15, 2014. [6] Christopher JB, Le May A, Lewin S, Ross D. Thirty years after Alma-Ata: A systematic [30] Ministry of Rehabilitation and Development, Central Statistics Office. The National review of the impact of community health workers delivering curative interventions Risk and Vulnerability Assessment (NRVA) 2005: Afghanistan. Kabul, Afghanistan: against malaria, pneumonia and diarrhea on child mortality and morbidity in sub- CSO; 2007. Saharan Africa. Hum Resour Health 2011;9(1):27. [31] Ministry of Public Health. Afghanistan Health Survey 2006: Estimates of priority health [7] Lewin S, Munabi-Babigumira S, Glenton C, Daniels K, Bosch-Capblanch X, van Wyk indicators for rural Afghanistan. http://moph.gov.af/Content/Media/Documents/ BE, et al. Lay health workers in primary and community health care for maternal Afghanistan-Health-Survey-2006-Report-English512011102721682.pdf. Accessed and child health and the management of infectious diseases. Cochrane Database April 15, 2014. Syst Rev 2010;3:CD004015. [32] MRD, CSO. National Risk and Vulnerability Assessment 2007/8: a profile of [8] Prasad BM, Muraleedharan VR. Community health workers: A review of concepts, Afghanistan. http://www.unodc.org/documents/afghanistan/EUGovtofAFG_NRVA_ practice and policy concerns. London: CREHS; 2007http://r4d.dfid.gov.uk/pdf/out- 2007-8_report.pdf. Published 2009. puts/equitable_rpc/chw_prasad_muraleedharan.pdf. Accessed April 14, 2014. [33] Akogun OB, Audu Z, Weiss MG, Adelakun AO, Akoh JI, Akogun MK, et al. [9] Crowe S, Utley M, Costello A, Pagel C. How many births in sub-Saharan Africa and Community-directed treatment of onchocerciasis with ivermectin in Takum, South Asia will not be attended by a skilled birth attendant between 2011 and Nigeria. Trop Med Int Health 2001;6(3):232–43. 2015? BMC Pregnancy Childbirth 2012;12:4. [34] Amazigo U. Community selection of ivermectin distributors. Community Eye Health [10] Goodman C, Brieger W, Unwin A, Mills A, Meek S, Greer G. Medicine sellers and ma- 1999;12(31):39–40. laria treatment in sub-Saharan Africa: what do they do and how can their practice be [35] Okeibunor JC, Ogungbemi MK, Sama M, Gbeleou SC, Oyene U, Remme JH. Additional improved? Am J Trop Med Hyg 2007;77(6 Suppl.):203–18. health and development activities for community directed distributors of ivermectin: [11] Barros AJ, Ronsmans C, Axelson H, Loaiza E, Bertoldi AD, França GV, et al. Equity in threat or opportunity for onchocerciasis control? Trop Med Int Health 2004;9(8):887–96. maternal, newborn, and child health interventions in Countdown to 2015: a retro- [36] CDI Study Group. Community-directed interventions for priority health prob- spective review of survey data from 54 countries. Lancet 2012;379(9822):1225–33. lems in Africa: results of a multicountry study. Bull World Health Organ 2010; [12] Sabo S, Ingram M, Reinschmidt KM, Schachter K, Jacobs L, Guernsey de Zapien J, et al. 88(7):509–18. Predictors and a framework for fostering community advocacy as a community [37] Ndyomugyenyi R, Kabatereine N. Integrated community-directed treatment for the health worker core function to eliminate health disparities. Am J Public Health control of onchocerciasis, schistosomiasis and intestinal helminths infections in 2013;103(7):e67–73. Uganda: advantages and disadvantages. Trop Med Int Health 2003;8(11):997–1004. [13] Talukder MN, Rob U. Equity in access to maternal and child health services in five [38] National Population Commission (NPC) [Nigeria], ICF Macro. Nigeria Demographic and developing countries: what works. Int Q Community Health Educ 2010;32(2): Health Survey 2008. Abuja, Nigeria: NPC, ICF Macro; 2009. http://dhsprogram.com/ 119–31. pubs/pdf/FR222/FR222.pdf. Accessed April 15, 2014. [14] Global Health Workforce Alliance, World Health Organization. Global experience of [39] Okeibunor JC, Orji BC, Brieger W, Ishola G, Otolorin EO, Rawlins B, et al. Preventing community health workers for delivery of health related Millennium Development malaria in pregnancy through community-directed interventions: evidence from Goals: a systematic review, country case studies, and recommendations for integra- Akwa Ibom State, Nigeria. Malar J 2011;10:227. tion into national health systems. Geneva: WHO; 2010. http://www.who.int/ [40] Brieger W, Orji B, Otolorin E, Ndekhedehe E, Jones Nwadike J. Establishing integrated workforcealliance/knowledge/resources/chwreport/en/. community management of malaria, pneumonia and diarrhea in two selected local [15] Crigler L, Hill K, Furth R, Bierregaard D. Community health worker assessment and government areas, Akwa Ibom State, Nigeria. http://www.jhpiego.org/files/malaria/ improvement matrix (CHW AIM): a toolkit for improving community health worker CDI%20Training/CDI-Readings/Establishing%20Integrated%20Community% programs and services. Revised version. Bethesda, MD: University Research Co., LLC; 20Management%20of%20Malaria%20Handout%20v3.pdf.AccessedApril15, 2013https://www.usaidassist.org/resources/community-health-worker-assess- 2014. ment-and-improvement-matrix-chw-aim-toolkit-improving-chw. Accessed August [41] New ERA, USAID. Government of Nepal. An analytical report on female community 28, 2014. health volunteers (FCHVs) of Nepal. Kathmandu: New ERA; 2007. [16] The Partnership for Maternal, Newborn and Child Health. Essential interventions, [42] Ministry of Health and Population (MOHP) [Nepal], New ERA, Macro International commodities and guidelines for reproductive health, maternal, newborn and child Inc. Nepal Demographic and Health Survey 2006. Kathmandu: Ministry of Health health: a global review of the key interventions related to reproductive, maternal, and Population, New ERA, and Macro International Inc.; 2007. newborn and child health (RMNCH). http://www.who.int/pmnch/knowledge/ [43] Pathak LR, Malla DS, Pradhan A, Rajlawat R, Campbell BB, Kwast B. Maternal Mortal- publications/201112_essential_interventions/en/. Published 2011. Accessed ity and Morbidity Study (MMMS) 1998. Kathmandu: Family Health Division, De- April 7, 2014. partment of Health Services, Ministry of Health, His Majesty’s Government of [17] Kisia J, Nelima F, Otieno DO, Kiilu K, Emmanuel W, Sohani S, et al. Factors associated Nepal; 1998. with utilization of community health workers in improving access to malaria treat- [44] Rajbhandari S, Hodgins S, Sanghvi H, McPherson R, Pradhan YV, Baqui AH, et al. ment among children in Kenya. Malar J 2012;11:248. Expanding uterotonic protection following childbirth through community-based J. Haver et al. / International Journal of Gynecology and Obstetrics 130 (2015) S32–S39 S39

distribution of misoprostol: operations research study in Nepal. Int J Gynecol Obstet [48] Perry HB, Zulliger R, Rogers MM. Community health workers in low-, middle-, and 2010;108(3):282–8. high-income countries: an overview of their history, recent evolution, and current [45] Smith JM, Gubin R, Holston MM, Fullerton J, Prata N. Misoprostol for postpartum effectiveness. Annu Rev Public Health 2014;35:399–421. hemorrhage prevention at home birth: an integrative review of global implementa- [49] World Health Organization. WHO recommendations for the prevention and treat- tion experience to date. BMC Pregnancy Childbirth 2013;13:44. ment of postpartum haemorrhage. Geneva: WHO; 2012. http://apps.who.int/iris/ [46] New ERA, USAID. Government of Nepal. Annual Report 2013. New ERA: Kathmandu; bitstream/10665/75411/1/9789241548502_eng.pdf. 2014. [47] Smith JM, de Graft-Johnson J, Zyaee P, Ricca J, Fullerton J. Scaling up high-impact interventions: How is it done? Int J Gynecol Obstet 2015;130(S2):S4–S10. International Journal of Gynecology and Obstetrics 130 (2015) S40–S45

Contents lists available at ScienceDirect

International Journal of Gynecology and Obstetrics

journal homepage: www.elsevier.com/locate/ijgo

SUPPLEMENT ARTICLE Preparing the next generation of maternal and newborn health leaders: The maternal and newborn health champions initiatives

Blami Dao a,⁎, Emmanuel Otolorin b,PatriciaP.Gomeza, Catherine Carr a, Harshad Sanghvi a a Jhpiego, Baltimore, MD, USA b Jhpiego, Abuja, Nigeria article info abstract

Keywords: A champion in health care can be deﬁned as any health professional who has the requisite knowledge and skills in Capacity building arelevanthealthﬁeld, who is respected by his/her peers and supported by his/her supervisors, and who takes the Champions lead to promote or introduce evidence-based interventions to improve the quality of care. Jhpiego used a com- Leadership mon approach during two distinct initiatives to identify individuals in Africa, Asia, and Latin America and the Maternal health Caribbean whose expertise in their clinical service area and whose leadership capacity could be strengthened Newborn health to enable them to serve as champions for maternal and newborn health (MNH). These champions have gone on to contribute to the improvement of MNH in their respective countries and regions. The lessons learned from this approach are shared so they can be used by other organizations to design leadership development strategies for MNH in low-resource countries. © 2015 Published by Elsevier Ireland Ltd. on behalf of International Federation of Gynecology and Obstetrics. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

1. Background importance. For example, in a multicountry survey conducted by the Maternal and Child Health Integrated Program (MCHIP)—USAID’s Recent reports from UN agencies regarding Millennium Develop- flagship maternal, neonatal, and child health (MNCH) program from ment Goals (MDGs) 4 and 5, which set targets for reductions in mater- 2008 to 2014—the mean knowledge score among frontline healthcare nal and child mortality, have shown that progress has been made in workers for the management of pre-eclampsia/eclampsia was just 25% both areas. At the global level, according to WHO data, maternal mortal- in Kenya [6]. In Tanzania, 95% of partographs, which are intended to ity ratios have declined from 380 to 210 maternal deaths per 100 000 be used during labor to monitor and anticipate complications, were live births between 1990 and 2013 [1].Similarly,UNICEFreportedare- completed after delivery [7]. Furthermore, the mean knowledge score duction in under-5 child mortality rates from 88 to 57 per 1000 live for actions to be taken for a retained placenta was 33% in Ethiopia [8], births for approximately the same period [2]. These global trends hide and the mean performance score for woman-friendly care was 58% in regional disparities, however, with most of the progress being made in Rwanda [9].Thesefindings suggest that significant knowledge and Asian and Latin American countries. In Sub-Saharan Africa, for example, practice gaps need to be addressed to promote high-quality care and maternal mortality ratios remain as high as 510 per 100 000 live births the use of essential interventions that have proven valuable in the re- [1], and Sub-Saharan Africa is the only region in which under-5 child ductionofmaternalandnewbornmortality[10,11]. mortality has not been cut by half since 1990 [3]. There are also examples from some low-resource countries showing Several challenges have impeded significant progress in the reduc- that besides competent health workers, strong political will along with tion of maternal and newborn mortality in Sub-Saharan Africa [4], in- good leadership in reproductive health are essential elements for the cluding shortages of human resources, poor infrastructure, scarcity of reduction of maternal mortality [6]. As a result, there is an apparent equipment and commodities, and insufficient implementation of need for providers who possess not only up-to-date clinical skills but evidence-based interventions [5]. Although the number of health also strong leadership skills, and who are well-established and workers plays a critical role, the quality of services delivered (resulting respected within their field of practice. Such healthcare workers should from those workers’ knowledge, skills, and attitudes) is of equal be able to advocate for the use and scale-up of evidence-based interventions [9], engage with decision makers to commit them to update national and subnational policies where necessary, and advocate for ⁎ Corresponding author at: Jhpiego, 1615 Thames Street, Baltimore, MD 21231, USA. increased funding for MNCH. In addition, they should be able to act Tel.: +1 410 537 1862. as role models, demonstrating quality care that is evidence-based E-mail address: [email protected] (B. Dao). and woman-friendly while guiding and inspiring others to do the http://dx.doi.org/10.1016/j.ijgo.2015.03.004 0020-7292/© 2015 Published by Elsevier Ireland Ltd. on behalf of International Federation of Gynecology and Obstetrics. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). B. Dao et al. / International Journal of Gynecology and Obstetrics 130 (2015) S40–S45 S41 same [12]. Health workers possessing such leadership and clinical skills Based on the successes and accomplishments of the MNH cham- could be called “champions.” pions who participated in the Regional Expert Development Initiative, and faced with the need for strong local leadership for MNCH in pro- 1.1. Champions in health care grams across more than 40 countries, particularly in Africa, the next Jhpiego-led and USAID-sponsored MNCH program, MCHIP, aimed to The concept of champions originated in the management field with replicate and improve upon the first initiative and focus on the African the notion of product champions. Champions and the promotion of in- region. This new leadership initiative, entitled the “Africa MNH Cham- novation are intimately linked. Champions can be defined as individuals pions Program,” was designed in collaboration with the WHO’s Regional who contribute decisively to the innovation process by overcoming re- Office for Africa, the UNFPA, the West African Health Organization sistance to innovations, building support for them, and making sure that (WAHO), and other regional institutions. As a result of this partnership, they are implemented [13]. WAHO was able to assist with institutional, logistic, and financial sup- There is a distinct role for champions in bringing changes and port for participants to complement the funding provided by USAID improvementinhealthcare[14,15]. Various types of healthcare cham- through MCHIP. Funding was provided to the champions for travel ex- pions have emerged, including executive champions who hold leader- penses to the courses, but they did not receive any other compensation ship positions within organizations; managerial champions who are for their participation in the initiative. responsible for managing clinical departments, wards or units; and clin- The Africa MNH Champions Program was an 18-month initiative ical champions who are frontline clinicians. All of these champions, that used a modern, evidence-based, “blended learning” (online and however, perform at least the following five functions: they advocate, face-to-face) approach to build the technical, training, and advocacy ca- disseminate knowledge, navigate boundaries between professional pacity of MNH experts from 10 African countries, five Anglophone and groups, build relationships, and achieve consensus. five Francophone countries. Champions from each country were Based on descriptions from the literature [16–18], a clinical champi- grouped into interdisciplinary teams comprised of midwives, obstetri- on can be further defined as a physician, nurse, midwife, physician assis- cians, and pediatricians, for a total of 30 champions. Due to original se- tant, or other healthcare professional who has the requisite knowledge lection and attrition, the final composition was 26 and more heavily and skills in a given discipline, who is respected by his or her peers and weighted toward midwives and obstetricians (Table 1). All champions, is supported by the system hierarchy, and who takes the lead in at the time of selection, were in leadership positions either in their facil- introducing a new practice or innovation to improve the quality of ities as chief of services and/or as pre-service education faculty. Since care. We define a maternal and newborn health (MNH) champion as a commencement of the program, the champions have been putting the specific kind of clinical champion: a health professional (doctor, mid- knowledge and skills learned into practice at all levels of the healthcare wife, nurse) with up-to-date knowledge, practices, and attitudes in delivery system in their respective countries. All have conducted step- MNH (i.e. an expert in his or her respective field of practice), who, down technical updates for colleagues in their own institutions, and through advocacy and action, promotes policies, practices, and all have been involved in national training workshops and revisions of programs that will help achieve MDGs 4 and 5 in his or her country pre-service and in-service curricula—designed by Jhpiego and and region. partners—either during or following their participation in the program. The present article describes two Jhpiego-led initiatives to develop MNH champions in countries with some of the most challenging MNH 3. Development process for both champions initiatives indicators in the world. We describe the MNH champion development process, identify ways in which the process evolved over time (including Some common elements of and key differences between the two differences between the two initiatives), highlight MNH champion con- MNH champions initiatives are described below. tributions through a series of case studies, and discuss lessons learned and challenges encountered. 3.1. Planning and design

2. Context and evolution for the champions initiatives Both initiatives were planned during a series of consultation meetings among Jhpiego, its partner organizations, and USAID. Their design One of the objectives of USAID’s flagship Maternal and Neonatal was guided by the following principles: (1) involvement of all parties Health Program, which was implemented from 1998–2004 by Jhpiego in the selection of countries and candidates; (2) competitive selection and partners, was to increase MNH providers’ use of evidence-based of candidates; and (3) ensuring local participation by involving regional standards of care, tools, and approaches. To achieve this goal as rapidly organizations and ministries of health as much as possible. To move im- as possible, the Maternal and Neonatal Health Program developed a net- plementation forward, a “facilitator team” was formed that included se- work of experts from Africa, Asia, and Latin America and the Caribbean nior midwives, obstetricians, and pediatricians, serving as Jhpiego staff who could “champion” the inclusion of essential interventions in na- or consultants. tional policies and in pre-service and in-service education curricula for doctors, midwives, and nurses at all levels of service delivery. This effort, 3.2. Selection criteria Jhpiego’s first MNH champions initiative, was called the “Regional Expert Development Initiative.” Forty-three midwives and obstetricians The facilitator team agreed on selection criteria for the candidates from 18 countries participated over a period of 12 months, receiving that were based on clinical capacity as well as perceived motivation continuing professional development activities to ensure that they and leadership skills (Box 1). For the first initiative, an announcement were proficient in life-saving best practices and advanced clinical training techniques, and that they understood the principles of advocacy for Table 1 MNH [19]. As their expertise and experience was recognized, many of Composition of the Africa MNH Champions Program. fi these champions subsequently ful lled the Maternal and Neonatal Region Original composition Attrition Final composition Health Program’s expectations for improving MNH in their regions, Anglophone Africa Midwives: 7 Midwives: 1 Midwives: 6 and they were later called upon to collaborate with the ACCESS Obstetricians: 5 Obstetricians: 5 Program, the follow-on USAID project that was also awarded to a Pediatricians: 3 Pediatricians: 2 Pediatricians: 1 Jhpiego-led consortium and implemented in 2004–2009. In addition, Francophone Midwives: 5 Midwives: 5 the champions were involved in MNH projects implemented by other Africa Obstetricians: 5 Obstetricians: 5 Pediatricians: 5 Pediatricians: 1 Pediatricians: 4 organizations in their respective countries. S42 B. Dao et al. / International Journal of Gynecology and Obstetrics 130 (2015) S40–S45

Box 1 3.4. Monitoring and evaluation Selection criteria for MNH champions initiatives. Several monitoring and evaluation tools were used to follow up and • Midcareer healthcare professionals (midwives, nurses, or track the outcomes of the various activities and action plans. These tools physicians) included reviews of activity logs, site visits between activities, and—in • Clinically proficient in provision of maternal and newborn the second initiative, with its blended learning approach—analysis of health services the e-learning courses; however, improvement of knowledge and skills • Currently active in clinical work of individuals trained by the champions was not measured. Because the • Committed to remaining in clinical practice majority of champions were already leaders in their respective facilities • Involved in in-service training or a pre-service education system or institutions when they were selected for the initiatives, they were • Able and motivated to do self-paced, independent learning well-positioned to carry out all activities and use the periods in between • Recognized as being, or having the potential to be, leaders in activities to implement action plans, organizing training sessions on the MNH field specific themes and supervising implementation of best MNH practices. In addition, because many of the facilities in which champions practiced were large-volume clinical practice sites for medical and midwifery pre- service education programs, students at these sites were exposed to the champions’ new knowledge and skills, and the sites themselves for applications was sent through Jhpiego’s networks alone; for the therefore benefitted early and consistently from the latest evidence- second initiative, an announcement was also sent through MCHIP part- based care. ner country offices as the “MNH Africa Champions Program Concept Note” to WHO, ministries of health, and USAID’s and UN agencies’ coun- 3.5. Key differences between the initiatives try offices. Once received, candidates’ applications were compared against the selection criteria to determine eligibility, and the final Although both MNH champions initiatives contained the same es- decision on champions from those eligible was made by a multiagency sential elements, there were some key differences, outlined in Table 3, selection committee. that were largely due to the regional focus of the Africa MNH Champions Because both initiatives involved multiple activities taking place Program and the increased use of technology and rapid emergence of over a period of 12–18 months, and also because of the high expecta- online learning platforms by the time that initiative was implemented. tions from the facilitator teams, each selected champion was asked to Before each of the face-to-face workshops held for the Africa MNH sign an agreement confirming that they would complete the entire Champions Program, online content was delivered to champions in program. Champions’ supervisors or institutional representatives were modules, including one on basic emergency obstetric and newborn also asked to sign agreements allowing the champions to commit to care (BEmONC) and one on effective teaching skills, via the Qstream full participation. platform, which uses a self-paced, spaced, learning approach [22].By giving champions online courses to cover this content prior to the workshops, the workshops were able to cover in-depth technical information 3.3. Representative activities and clinical skills more rapidly. For the Africa MNH Champions Program, email communication was also possible to conduct follow-up monitor- Table 2 shows the types of activities included in both initiatives to ing and evaluation, and a Community of Practice, or online site to share bolster champions’ clinical, training, and leadership skills. The ultimate resources and public discussions, was established to facilitate continued goal of these activities was to create a group of proficient clinical exchange of experiences and resources. Some participants even trainers with standardized clinical skills who could educate others and exchanged communications through social media. Neither of the advocate for improved MNH in their respective countries and regions. initiatives provided additional funding to increase services or improve Training materials included those developed by Jhpiego and its partners, upon those services that were currently provided in each of the such as the WHO manual, “Managing complications in pregnancy and champions’ facilities. childbirth: a guide for midwives and doctors” [20] and the “Emergency Obstetric Care for Doctors and Midwives Learning Resource Package” by 4. Case studies from the first MNH champions initiative the Maternal and Neonatal Health Program and the Averting Maternal Death and Disability (AMDD) Program of Columbia University’s Mail- The MNH champions participating in both initiatives have accom- man School of Public Health [21]. plished a great deal in their workplaces, countries, and regions. A survey of the champions from the Regional Expert Development Initiative indicated that many were able to practice the skills learned in the trainings and thus felt more competent and confident in their use. Some went on to become particularly notable and successful leaders. While cham- Table 2 pions’ facilities did not receive equipment, commodities, or funding, Representative activities for MNH champions initiatives. champions became advocates to their ministries of health and facilities

Activity Objective to improve MNH services. The case studies below are drawn from that ﬁrst initiative and highlight the contributions of these champions in Knowledge and clinical skills Ensure that champions are familiar with the latest their respective countries or regions. standardization course evidence in MNH (including malaria in pregnancy, PMTCT, and quality improvement) and have strengthened their clinical practice skills 4.1. Case study 1: Implementing best practices locally, educating regionally, Clinical training skills course Enable champions to serve as master trainers who advocating globally can effectively train other MNH providers Leadership and advocacy course Empower champions to be change agents and advocates for maternal and newborn health One obstetrician/gynecologist from Africa who was the head of an Follow-up visits Mentor the champions as they transfer new obstetrics and gynecology unit at a large teaching hospital joined the knowledge and skills to their workplaces and champions initiative with a keen interest in learning and implementing implement an action plan for training and advocacy new evidence-based approaches to improve health outcomes at her Abbreviation: PMTCT, prevention of mother-to-child transmission. hospital. She immediately began using the knowledge that she gained B. Dao et al. / International Journal of Gynecology and Obstetrics 130 (2015) S40–S45 S43

Table 3 Key differences between the two MNH champions initiatives.

MNH Regional Experts Initiative Africa MNH Champions Program

Selection criteria No technology requirements Possess a computer and have computer skills Composition Midwives and obstetricians Teams of midwives, obstetricians, and pediatricians (ideally 1:1:1) Geographic distribution of participants Anglophone, Francophone, and Spanish-speaking Limited to Anglophone and Francophone Africa participants from 3 continents: Africa, Asia, and Latin America and the Caribbean Training strategy Face-to-face only “Blended learning” approach (online and face-to-face) Communication among the participants during No formal mechanisms Community of Practice, email, social media and after participation in the program

to make several changes that are still in place in the hospital today. For today. Through the many talks that he has given throughout the region, one, she reorganized the “factory assembly-line approach” in the prena- he has educated other providers about the respectful maternity care tal clinic to create a more personalized, women-friendly service, allowing principles that were highlighted for him during the champions initia- each woman to be seen by one nurse in a screened-off area instead of by tive. Another of his ongoing advocacy efforts is the promotion of a mod- many providers in a public area. Second, she introduced an emergency ified insertion technique used to reduce expulsions of the postpartum triage system that was especially beneficial for her busy service because intrauterine contraceptive device by straightening the uterine–cervical it acted as a referral site for a large catchment area and treated angle postpartum implantation. This idea originated during the time many women with complications on a daily basis. Finally, she worked that he was involved in the initiative. He has collaborated with pro- with her hospital’s administration to advocate for a more consistent grams in India and the Philippines to develop training materials and supply of drugs to treat pre-eclampsia/eclampsia and postpartum train providers using his approach. hemorrhage—the greatest causes of maternal mortality in her country. Although the hospital’s maternity unit remains small and crowded, 4.4. Case study 4: Demonstrating the broad applicability of leadership it is still used as a clinical training site because of the evidence-based training practices that were supported by the champion and implemented there, in areas such as infection prevention, active management of the The first champions initiative reached Southeast Asia as well. third stage of labor, magnesium sulfate for pre-eclampsia/eclampsia, One obstetrician/gynecologist from Jakarta, Indonesia, was the dep- and newborn resuscitation. The champion has gone on to facilitate re- uty chief of medical services at a private hospital when he participat- gional trainings in these, and additional, best practices, and she has ed in the initiative and has gone on to apply what he learned, also worked at the global level as a contributor to a WHO technical particularly the training and leadership skills, in a variety of other re- working group and a member of the editorial review committee of the gional and global positions in MNH. For example, he has led trainings African Journal of Reproductive Health. for several international nongovernmental organizations in Indonesia and Afghanistan; served as a principal researcher for a World Bank 4.2. Case study 2: Leading the way, bringing hope to underserved areas Project; and worked with the University of Aberdeen’sprojecton maternal and newborn near misses in Indonesia. He is now the director Another Africa champion, a nurse-midwife who participated in the of his private hospital, which is a partner on a USAID-supported project initiative as she began leading an organization for private midwives, has for MNH. influenced the expansion of her organization from approximately 30 members working in private maternity practices in seven of the country’s 5. Lessons learned from the second champions initiative 30 regions to 75 members across 18 regions. Since the organization’s founding, its members have assisted at nearly 10 000 births, targeting The second initiative, the Africa MNH Champions Program, had sim- under-served communities where other health professionals often do ilar accomplishments to the first (Table 4) and improved upon the first not want to work. These committed midwives give women the all- by involving multiple partners in soliciting candidate applications, orga- important option of a skilled provider for their pregnancy and birth at nizing champions as interdisciplinary teams of midwives, obstetricians, an affordable cost. As the reach of the organization’s members has and pediatricians, and utilizing technology to increase opportunities to grown, the champion continues conducting in-service trainings for mem- transfer learning. Still, the initiative encountered a number of challenges bers to build skills in infection prevention, use of the partograph, and that might reflect universal issues to be considered when designing any newborn resuscitation. She also carries out supportive supervision as MNH champions program. funds allow. 5.1. Country selection 4.3. Case study 3: Sharing a passion for evidence-based practice, respectful care, and innovation The selection of countries from which the initiative could solicit candidates was necessarily driven by stakeholder preferences. The terms One active champion from Latin America, an obstetrician/gynecolo- for the Africa MNH Champions Program required selection of five An- gist who has worked throughout the region with several international glophone and five Francophone countries, and all of the partners nongovernmental organizations to strengthen MNH, is now an assistant supporting USAID in designing the program—the WHO’s Regional Office professor of obstetrics and gynecology at the national medical school. for Africa, UNFPA, WAHO, and other regional institutions—had different Since his participation in the champions initiative, he has advocated to and competing suggestions for those 10 countries. Ultimately, USAID the ministry of health in his country to adopt the WHO Integrated Man- made the final selection from a list of its own identified priority coun- agement of Pregnancy and Childbirth (IMPAC) guidelines for use in all tries. Individuals from a number of countries not included in that final health facilities to curb unnecessary “routine” practices such as episiot- selection expressed a strong desire for their country to be part of a fu- omy and cesarean delivery. His own facility has been able to drastically ture champions initiative, indicating the limited number and geograph- decrease the rates of these practices, from 90% to 16% for episiotomy ic diversity of candidates because of country selection as well as the high and from 50% to 24% for cesarean delivery—rates that still hold true perceived value of the program. S44 B. Dao et al. / International Journal of Gynecology and Obstetrics 130 (2015) S40–S45

Table 4 Selected accomplishments of the Africa MNH Champions Program.

Accomplishment Role of champion Recipient of champion’s support

Participated in national reproductive health meetings Invited representative National ministry of health Involved in establishing midwifery regulations Consultant National ministry of health Developed MNH guidelines or training materials Facilitators/consultants National ministry of health, local and international nongovernmental organizations Contributed to drafting of reproductive health law Team Leader National ministry of health Facilitated trainings on emergency obstetric care, pre-eclampsia/eclampsia Trainers Teaching hospitals Facilitated trainings on the Helping Babies Breathe approach Course organizers, trainers Teaching hospitals Coordinated regional trainings for basic emergency obstetric and newborn care Regional coordinator, lead trainer District facilities Introduced evidence-based practices Organizer, facilitator Teaching hospitals

5.2. Participant selection because of personal circumstances, but it is possible that they did so because of logistic difficulties and challenges working with their team. The Participant selection occurred on a tight timeline, likely affecting the most successful teams were able to leverage their identity as a cohort of number and quality of candidates from selected countries. The pool of champions to influence national health policies, initiate regional candidates from the Francophone region was larger and more balanced mentoring programs, and introduce evidence-based care in teaching in cadre than that of the Anglophone region because of the stronger hospitals. Participation in the initiative garnered respect for and lent existing networks in the Francophone region that facilitated outreach credibility to each champion’s opinions, but their collective endorse- to more contacts. Even though there was a requirement that candidates ment appeared to build even greater momentum for broad changes in possess and be able to use a computer, the candidates chosen were not national health systems. necessarily ideal for the online learning methods that were central to this initiative. For many, internet connectivity was poor and often 5.5. Adaptation to online activities irregular, and comfort with online learning varied from a beginning level of competence to expert, with similar ranges of competence expe- Particularly given the connectivity and other technology challenges fi rienced within each cadre. As a result, those who would have bene tted experienced by some of the champions, completion of the Qstream most from the program might not have been reached. modules was high, with completion rates of 80% and 97% on the BEmONC and Effective Teaching Skills modules, respectively. As expect- 5.3. Involvement from partners and national ministries of health ed, modules that were translated into French had higher completion rates from the Francophone group. All 26 champions in the initiative Partner involvement was strongest in West Africa, with support successfully enrolled in the Community of Practice, but few made use from WAHO that even included travel funds. Champions’ activity logs of it, citing barriers that included lack of time, logistical issues with reg- fi document a high level of participation by UNFPA, UNICEF, WHO, and a istering and obtaining new passwords, and ongoing connectivity dif - number of USAID-funded programs. Early country and regional engage- culties. Redundancy with other resource sites and discussion boards ment seemed to facilitate a higher level of partner participation. In some may also have contributed to lack of use. For many champions, internet countries, champions teams included those employed by the national connectivity was irregular even at work sites, and few had internet ministry of health or other government ministries, or the facilitators access at home. Approximately 25% of participants funded their own team was able to introduce teams to the relevant ministries and discuss internet use. For online activities to be successful in a champions initia- champions’ potential as leaders and advocates. Those countries experi- tive, they must be readily available and at least as accessible as in- enced strong early successes during and after the initiative as cham- person activities. pions were asked with increasing frequency to participate in national or subnational strategies, trainings, assessments, and programs, as evi- 6. Conclusion denced by the champions’ activity logs. Engagement of partners and national ministries at the start of the initiative, including selection These two Jhpiego-led MNH champions initiatives have used similar of champions with connections to the ministries, therefore appeared approaches that can provide a foundation for future efforts in capacity- to lead to greater potential for immediate activity and use of the building and leadership development in MNH. Despite the challenges champions’ skills. that come with multicountry initiatives, the investment was worth it in the light of what the experts/champions have accomplished collectively and individually. However, a more rigorous and long-term 5.4. Individual and team participation evaluation of the impact of these initiatives should take place before expanding the approach. Despite agreements signed at the beginning of the initiative commit- ting champions to full participation, participation was varied and Con flict of interest appeared to depend on previous experience with similar programs, geographic proximity to other team members and activities, and overall fl team cohesion. Teams with champions located in the same city general- The authors have no con icts of interest. ly had an easier time coordinating their work and calling on each other for support. Additional on-site follow-up between activities would have References been useful to increase participation and improve team integration but [1] WHO, UNICEF, UNFPA, World Bank. Trends in Maternal Mortality: 1990 to 2010. was not possible because of budget limitations. In addition, the initiative WHO, UNICEF, UNFPA and The World bank Estimates. https://www.unfpa.org/ was not designed to measure improvements in knowledge and skills of webdav/site/global/shared/documents/publications/2012/Trends_in_maternal_ those trained by champions—evaluations that would have been useful mortality_A4-1.pdf. Published 2010. Accessed May 15, 2014. [2] UNICEF. State of the World’s Children 2012. Children in an Urban World. http:// to determine the impact these initiatives had on MNH services. The www.unicef.org/sowc2012/pdfs/SOWC%202012-Main%20Report_EN_13Mar2012. four champions who dropped out of the initiative ostensibly did so pdf. Published 2012. Accessed May 15, 2014. B. Dao et al. / International Journal of Gynecology and Obstetrics 130 (2015) S40–S45 S45

[3] UNICEF. Statistics by Area/Child Survival and Health. http://www.childinfo.org/ [12] Flodgren G, Parmelli E, Doumit G, Gattellari M, O'Brien MA, Grimshaw J, et al. Local mortality_underfive.php. Published 2013. Accessed May 15, 2014. opinion leaders: effects on professional practice and health care outcomes. Cochrane [4] World Health Organization Regional Office for Africa. Addressing the Challenge of Database Syst Rev 2011;8:CD000125. Women’s Health in Africa. http://www.afro.who.int/en/clusters-a-programmes/ [13] Howell JM, Higgins CA. Champions of change: identifying, understanding, and frh/gender-womens-health-a-ageing/highlights/3741-addressing-the-challenge-of- supporting champions of technological innovations. Organ Dyn 1990;19(1):40–55. womens-health-in-africa.html. Published 2012. Accessed May 15, 2014. [14] Shaw EK, Howard J, West DR, Crabtree BF, Nease DE, Tutt B, et al. The role of the [5] Nyamtema AS, Urassa DP, van Roosmalen J. Maternal health interventions in re- champion in primary care change efforts: from the State Networks of Colorado source limited countries: a systematic review of packages, impacts and factors for Ambulatory Practices and Partners (SNOCAP). J Am Board Fam Med 2012;25(5): change. BMC Pregnancy Childbirth 2011;11:30. 676–85. [6] KagemaF,RiccaJ,RawlinsB,RosenH,LynamP,KidulaN.QualityofCareforPrevention [15] Soo S, Berta W, Baker GR. Role of champions in the implementation of patient safety and Management of Common Maternal and Newborn Complications: Findings from a practice change. Healthc Q 2009;12:123–8. National Health Facility Survey in Kenya. http://www.k4health.org/sites/default/files/ [16] Lewis T, Edwards C. How clinical champions can improve quality. Nurs Manag MCHIP_USAID%20KENYA%20QOC%20FINAL%20REPORT.pdf. Published 2010. Accessed (Harrow) 2008;14(10):24–7. May 15, 2014. [17] Cohn KH. Changing physician behavior through involvement and collaboration. J [7] Plotkin M, Tibaijuka G, Makene CL, Currie S, Lacoste M. Quality of Care for Prevention Healthc Manag 2009;54(2):80–6. and Management of Common Maternal and Newborn Complications: A Study of 12 [18] Pizarro-Duhart R. RCGP Clinical Champions Role Description. http://www. Regions in Tanzania. http://www.mchip.net/sites/default/files/mchipfiles/Tanzania_% rcgp.org.uk/clinical-and-research/~/media/Files/CIRC/Clinical%20Priorities/ 20QoC_StudyReport_FINAL_0.pdf. Published 2010. Accessed May 15, 2014. Clinical%20Champion%20Role%20Description.ashx. Published 2011. Accessed [8] Getachew A, Ricca J, Cantor D, Rawlins B, Rosen H, Tekleberhan A, et al. Quality of May 15, 2014. Care for Prevention and Management of Common Maternal and Newborn Complica- [19] Blouse A, Kinzie B, Sanghvi H, Hines K. Developing Regional Experts in Essential tions: A Study of Ethiopia’s Hospitals. http://www.healthynewbornnetwork.org/ Maternal and Newborn Care: The MNH Program Experience. http://reprolineplus.org/ sites/default/files/resources/Ethiopia_QoC_formatted_final.pdf. Published 2011. resources/developing-regional-experts-essential-maternal-and-newborn-care-mnh- Accessed May 15, 2014. program-experience; 2004. Published 2004. Accessed May 15, 2014. [9] Ngabo F, Zoungrana J, Faye O, Rawlins B, Rosen H, Levine R, et al. Quality of Care for [20] World Health Organization, UNFPA, UNICEF, World Bank. Managing complications Prevention and Management of Common Maternal and Newborn Complications: in pregnancy and childbirth: a guide for midwives and doctors. http://whqlibdoc. Findings from a National Health Facility Survey in Rwanda. http://www.mchip.net/ who.int/publications/2007/9241545879_eng.pdf. Published 2000 (reprint 2007). sites/default/files/Rwanda_QoC.PDF. Published 2012. Accessed May 15, 2014. Accessed May 15, 2014. [10] World Health Organization, Aga Khan University, Partnership for Maternal, Newborn [21] Jhpiego. Emergency Obstetric Care for Doctors and Midwives Learning Resource & Child Health. A Global Review of the Key Interventions Related to Reproductive, Package. http://reprolineplus.org/resources/emergency-obstetric-care-doctors-and- Maternal, Newborn and Child Health (RMNCH). http://www.who.int/pmnch/ midwives-learning-resource-package.AccessedMay15,2014. topics/part_publications/essential_interventions_18_01_2012.pdf. Published 2011. [22] Kerfoot BP, DeWolf WC, Masser BA, Church PA, Federman DD. Spaced education Accessed May 15, 2014. improves the retention of clinical knowledge by medical students: a randomised [11] Kerber K, de Graft-Johnson J, Bhutta Z, Okong P, Starrs A, Lawn J. Continuum of care controlled trial. Med Educ 2007;41(1):23–31. for maternal, newborn, and child health: from slogan to service delivery. Lancet 2007;370(9595):1358–69. International Journal of Gynecology and Obstetrics 130 (2015) S46–S53

Contents lists available at ScienceDirect

International Journal of Gynecology and Obstetrics

journal homepage: www.elsevier.com/locate/ijgo

SUPPLEMENT ARTICLE Essential basic and emergency obstetric and newborn care: From education and training to service delivery and quality of care

Emmanuel Otolorin a,⁎, Patricia Gomez b, Sheena Currie c, Kusum Thapa d, Blami Dao b a Jhpiego, Abuja, Nigeria b Jhpiego, Baltimore, Maryland, USA c Jhpiego, Washington, DC, USA d Jhpiego, Lalitpur, Nepal article info abstract

Keywords: Approximately 15% of expected births worldwide will result in life-threatening complications during pregnancy, Emergency obstetric and newborn care delivery, or the postpartum period. Providers skilled in emergency obstetric and newborn care (EmONC) services (EmONC) are essential, particularly in countries with a high burden of maternal and newborn mortality. Jhpiego and its Emergency obstetric care consortia partners have implemented three global programs to build provider capacity to provide comprehensive Essential newborn care EmONC services to women and newborns in these resource-poor settings. Providers have been educated to Signal functions deliver high-impact maternal and newborn health interventions, such as prevention and treatment of postpar- Skilled birth attendants tum hemorrhage and pre-eclampsia/eclampsia and management of birth asphyxia, within the broader context of quality health services. This article describes Jhpiego’s programming efforts within the framework of the basic and expanded signal functions that serve as indicators of high-quality basic and emergency care services. Lessons learned include the importance of health facility strengthening, competency-based provider education, global leadership, and strong government ownership and coordination as essential precursors to scale-up of high impact evidence-based maternal and newborn interventions in low-resource settings. © 2015 Published by Elsevier Ireland Ltd. on behalf of International Federation of Gynecology and Obstetrics. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

1. Introduction parenteral anticonvulsants; administration of parenteral uterotonics; removal of retained products (manual vacuum aspiration); assisted Approximately 15% of expected births worldwide will result in life- vaginal delivery; manual removal of the placenta; and resuscitation of threatening complications during pregnancy, delivery, or the postpartum the newborn [5]. Comprehensive emergency obstetric and newborn period [1]. The concept of emergency obstetric and newborn care care (CEmONC) includes all BEmONC services and adds surgical capac- (EmONC) was introduced by WHO, UNICEF, and UNFPA in 1997 as an ity and blood transfusion. This set of life-saving services defines a health organizing framework for the delivery of evidence-based clinical services, facility with regard to its capacity to treat obstetric and newborn emer- as a critical component of any program to reduce maternal and newborn gencies [4]. The decision to include these functions in a package of mortality [2]. Skilled birth attendants (SBAs) [3] provide EmONC services emergency obstetric and newborn care services was based on evidence within the context of community-focused and facility-based health from numerous quasi-experimental or experimental studies and is systems, enabling timely prevention of and intervention for these compli- summarized in various systematic reviews [6–9]. Recent global discus- cations and saving the lives of mothers and newborns. Universal access to sions have centered on expansion of the original seven signal functions EmONC is considered essential to reduce maternal mortality and requires to encompass activities related to routine care for mothers and new- that all pregnant women and newborns with complications have rapid borns because they enable prediction, prevention, and early interven- access to well-functioning facilities that include a broad range of service tion to mitigate life-threatening complications [10]. These expanded delivery types and settings [4]. functions include such services as: infection prevention and manage- A set of seven key obstetric services, or “signal functions,” has been ment for both mothers and infants; monitoring and management of identified as critical to basic emergency obstetric and newborn care labor using the partograph; active management of the third stage of (BEmONC): administration of parenteral antibiotics; administration of labor; and infant thermal protection, feeding, and HIV prevention. Table 1 shows the proposed expanded signal functions for obstetrics and newborn care alongside existing functions. ⁎ Corresponding author at: Jhpiego, Plot No. 3685 Erie Close, Maitama District, Abuja, fl FCT, Nigeria. Tel.: +234 803 478 3549. Over the last 15 years, Jhpiego has led consortia for USAID agship E-mail address: [email protected] (E. Otolorin). maternal and newborn health (MNH) programs—Maternal and Neonatal http://dx.doi.org/10.1016/j.ijgo.2015.03.007 0020-7292/© 2015 Published by Elsevier Ireland Ltd. on behalf of International Federation of Gynecology and Obstetrics. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). E. Otolorin et al. / International Journal of Gynecology and Obstetrics 130 (2015) S46–S53 S47

Table 1 Proposed obstetric and newborn signal functions (existing EmONC functions are in italics).a

Dimensions of facility care Obstetric Newborn

General requirements for health facility Service availability 24/7 Skilled providers in sufﬁcient numbers Referral service to higher-level care, communications tools Reliable electricity and water supply, heating in cold climates, clean toilets A. Routine care (for all mothers and Monitoring and management of labor Thermal protectionb babies) using partograph Infection prevention measures Immediate and exclusive breastfeeding (hand-washing, gloves) Active management of the third stage Infection prevention including hygienic of labor (AMTSL)c cord cared B. Basic emergency care (for mothers and Parenteral magnesium sulfate for (pre-) Antibiotics for preterm or prolonged babies with complications eclampsia PROM to prevent infection Assisted vaginal delivery Corticosteroids in preterm labor Parenteral antibiotics for maternal infection Resuscitation with bag and mask of non-breathing baby Parenteral oxytocic drugs for hemorrhage KMC for premature/very small babies Manual removal of placenta for retained Alternative feedinge if baby unable to placenta breastfeed Removal of retained products of conception Injectable antibiotics for neonatal sepsis (PMTCT if HIV-positive mother)f C. Comprehensive emergency care Surgery (e.g. cesarean) including anesthesia Intravenous ﬂuids (functions in addition to Basic) Blood transfusion Safe administration of oxygen

Abbreviations: KMC, kangaroo mother care; PMTCT, prevention of mother-to-child transmission; PROM, premature rupture of membranes; 24/7, 24 hours a day 7 days a week. a Reprinted from Gabrysch et al. [10]. b Thermal protection: drying baby immediately after birth, skin-to-skin with mother, wrapping, no bath in ﬁrst six hours. c AMTSL: oxytocin injection in thigh within 1 minute of delivery of baby, controlled cord traction, uterine massage after delivery of the placenta. d Hygenic cord care: cutting with sterile blade, application of 4% chlorhexidine on tip of cord and stump and no application of harmful substances (or clean and dry care in settings with low neonatal mortality and infection risk). e Breastmilk expression and cup/spoon feeding. f PMTCT: in brackets as not strictly a “newborn” function, but included for continuum of care; situational depending on HIV prevalence.

Health (1998–2004), the Access to Clinical and Community Maternal, necessary, special studies were conducted with approval from the Johns Neonatal and Women’s Health (ACCESS) Program (2004–2009), Hopkins University Human Subjects Review Board. Data were extracted and the Maternal and Child Health Integrated Program (MCHIP) from these documents to determine how evidence-based interventions (2008–2014). These three programs have focused on maternal and new- were incorporated at global and national levels into standards of practice, born survival through the implementation of evidence-based interven- guidelines and protocols, and how healthcare workers were prepared to tions in low-resource countries designed to accelerate the reduction of use these performance standards for quality improvement in the scale-up maternal, newborn, and child mortality [11]. MCHIP has continued to of access to MNH care. We also looked at how frontline healthcare build upon the successes of the Maternal and Neonatal Health and the workers were prepared to become SBAs to provide the most critical ACCESS programs while focusing on scale-up and integration into other evidence-based interventions. health areas in the 30 USAID priority countries facing the highest disease The activities and interventions undertaken by each of these programs burden [11]. These programs have identified and addressed many critical are summarized in Table 2. The table looks at the goals and strategic ob- gaps in EmONC service provision by taking research evidence to practice jectives of each program, the major EmONC interventions and notable in dozens of low-resource countries. Some of these identified gaps in- outcomes related to provider education and training, service delivery, clude: (1) poor knowledge and limited coverage of signal functions, espe- and quality of care in the interest of sharing lessons learned from program cially for the most vulnerable women in hard-to-reach communities; experience. We reflect on both achievements from and challenges in the (2) inadequate emphasis on the development of critical clinical skills implementation of programs addressing these EmONC services. for EmONC during pre- and in-service education, and poor retention of As many low-resource countries have worked to increase access to those skills over time; (3) limited use of birth planning and complication the quality and availability of BEmONC and CEmONC services, Jhpiego readiness, which aims to reduce or eliminate delays in seeking care and/ has been at the forefront of providing evidence-based technical support or reaching care facilities; and (4) poor quality of EmONC services at to ensure that the health workforce—one of the six “building blocks” of a health facilities in some settings, largely due to weak health systems. health system [12,13]—is competent, responsive, and productive in pro- This article aims to share lessons learned from the implementation viding these services. Addressing the learning needs of health workers of the three flagship USAID-funded programs relating to the expansion remains a critical component in improving health systems. of coverage and quality of EmONC signal functions. The article also com- The interventions and strategies for improving maternal and neonatal ments on Jhpiego’s corollary efforts in advancing the provision of addi- health and survival closely track the household-to-hospital continuum of tional high-impact interventions for the provision of high-quality care (HHCC) approach [14]. The ACCESS program applied the HHCC EmONC services, including the expanded signal functions presently framework by recognizing the importance of systematically and simulta- under consideration. neously addressing maternal and newborn issues at both the community and facility levels using evidence-based interventions. Supporting com- 2. Methods munity health workers (CHWs) and frontline health workers (physicians, nurses, and midwives) to promote community awareness and care- The information provided in this article has been extracted from pro- seeking behaviors in areas with poor utilization of maternal and newborn ject monitoring documents (service statistics), external evaluation re- services increases the potential to reduce the adverse impact of the three ports, and project publications from the three flagship programs. When delays—(1) delay in the decision to seek care; (2) delay in arrival at a S48 E. Otolorin et al. / International Journal of Gynecology and Obstetrics 130 (2015) S46–S53 health facility; and (3) delay in the provision of adequate care—as providers then form small groups and move to the clinical setting for su- discussed by Thaddeus and Maine [15]. This manner of community pervised practice. engagement is also known to impact maternal and newborn deaths. Jhpiego has developed a continuum of activities to ensure that Improving the capacity of providers and the health system specifically providers: (1) learn the evidence basis for EmONC interventions, in- reduces the “third delay”: receiving the appropriate care in a timely cluding the preventive signal functions presently under consideration; fashion. Community involvement is a critical element in helping families (2) practice critical life-saving skills; and (3) transfer their knowledge understand the benefit of using skilled providers, particularly those who and skills to their clinical sites. To prepare for the practical component, have had prior negative experiences and who are reluctant to engage in at least one clinical site is chosen and prepared as a demonstration site care-seeking behaviors for a current pregnancy—a situation often for “best practice.” These clinical sites must have a high volume of cli- acknowledged as the “Phase Four” delay [16]. ents and the providers often require targeted educational updates to enhance their ability to provide essential standardized interventions. 3. Implementation experience A dedicated manual was developed to ensure rapid site strengthening and optimal clinical experience during education and training [21]. The main lesson learned from all of these programs is that a combina- Supplemental materials have also been developed to support tion of well-coordinated demand creation and service improvement ac- standardized competency-based education and include a variety of tivities is essential to any successful scale-up of evidence-based teacher/tutor and participant teaching and learning resources. For interventions. Factors that promote such successful scale-up include example, the knowledge domain has been addressed using interactive increased awareness of the availability of global low cost solutions to presentations; the decision-making tools use case studies; and psycho- the major causes of maternal and newborn mortality; strong government motor skills and promotion of professional attitudes use demonstra- ownership of national EmONC efforts that results in the domestication tions and practice with checklists, first with anatomic models and and implementation of global MNH standards; competency-based pro- then with clients. A specific example of this process is Jhpiego’s work vider education leading to skilled birth attendance; and respectful mater- with the Averting Maternal Death and Disability (AMDD) program to nity care in the context of health systems strengthening. In the sections prepare healthcare workers in obstetric anesthesia for use during that follow, information is provided on the settings and process used for EmONC service delivery [22–25]. dissemination of generic international standards for essential MNH care, Because repetitive teaching interventions achieve better learning provider education, and quality improvement. outcomes than single interventions [26], Jhpiego has structured its educational programming with a “low-dose, high-frequency” educational 3.1. International standards for essential MNH care innovation that iteratively prepares lower-level health providers to improve maternal and newborn outcomes [27]. In order for providers to Collaborating with WHO, UNFPA, UNICEF, International Federation manage complications, it is essential that they understand and support of Gynecology and Obstetrics (FIGO), International Confederation of normal birth and labor management, including use of the partograph Midwives (ICM), and other donors and technical bodies, the Jhpiego- and active management of the third stage of labor (AMTSL). A number led consortia have helped facilitate the development of a number of of resource materials and manuals were developed that address these guidelines and manuals for provider use. Two manuals were developed supportive processes and EmONC skills [27]. in collaboration with WHO as part of The Integrated Management of Follow-up with providers occurs within three to six months to Pregnancy and Childbirth Series: Managing Complications of Pregnancy promote the retention of knowledge and skills. Providers undergo test- and Childbirth (MCPC) [17] and Managing Newborn Problems (MNP) ing for knowledge competencies and are observed performing skills on [18]. Both manuals used a symptom-based approach rather than a clients and/or models. Discussions are then held with providers’ super- traditional disease-based approach to clinical decision-making. MCPC visors and facility administrators to assess any barriers to use of best was originally published in English and has been translated into several practices. A guide for this process has been developed that can be languages including Spanish, French, Dari, Laotian, Bahasa, and Manda- used both for ongoing assessments and for supportive supervision rin. MNP is available in English, French, and Arabic. These documents (Box 1) [28]. have informed the development and adaptation of maternal and The lessons learned from provider education for MNH are that newborn health service delivery guidelines and associated pre- and competency-based clinical skills standardizations with the use of anatom- in-service education and training packages in many low-resource coun- ic models help to build confidence of providers to provide quality services tries including Afghanistan, Bolivia, Burkina Faso, Guatemala, Haiti, while protecting clients from harm (see Table 2 for illustrative notable Honduras, Indonesia, Nepal, Nigeria, Peru, Tanzania, and Zambia. More outcomes). A holistic approach to this transfer of learning that prescribes recently MCHIP has supported new guidelines from WHO on preventing tasks for the learner, teacher, supervisor, and co-workers before, during, postpartum hemorrhage (PPH) and pre-eclampsia/eclampsia (PE/E) by and after the teaching intervention is essential, and program managers making the information more readily available [19]. must learn to adequately invest resources for this implementation. The key lesson learned is that the provision of global generic guidance on low-cost effective interventions for the prevention and/or 3.3. Quality improvement treatment of the common causes of maternal and newborn mortality provides a template for countries who wish to set performance stan- Paxton et al. [29] reviewed 24 national-level needs assessments con- dards or protocols for MNH care and a roadmap for the program design. ducted in Asia, Africa, and Latin America to determine the availability of BEmONC and CEmONC services. The authors noted that while CEmONC 3.2. Provider education facilities were available in adequate number, BEmONC services were not as well distributed within geographic areas. There were also concerns The education of providers in EmONC always includes a knowledge about the quality of care, equity of service delivery, and financial accessi- domain, either in a classroom setting or using online resources, and a bility of these emergency obstetric care facilities. The authors therefore competency update enabling providers to practice decision-making recommended prioritizing upgrades to infrastructure for BEmONC facili- and new clinical skills. Jhpiego uses a “humanistic” approach to educa- ties, in-service education of staff on essential EmONC skills, and providing tion and training, in which providers first practice skills on anatomic supportive supervision and improved management systems. models, treating them as though they are actual clients. This approach To provide a platform for a root cause analysis of quality issues in is consonant with Jhpiego’s commitment to the philosophy and practice BEmONC and CEmONC facilities, Jhpiego introduced its quality im- of respectful maternity care [20]. Once skills are mastered on models, provement (QI) approach, known as Standards Based Management Table 2 Summary of EmONC and maternal/neonatal care services provided by Jhpiego and partners through USAID flagship programs (1998–2014).

Program/date/partners Countries Goals and strategic objectives Major EmONC activities and interventions Notable EmONC outcomes (education and training, service delivery, quality of care)

Program: Afghanistan To promote maternal and newborn survival in • Increased collaboration among organizations pro- • International evidence-based standards for MNH Maternal and Neonatal Health Bangladesh low-resource countries by increasing the use of moting maternal and newborn care, e.g. with were incorporated into national guidelines in Date: Bhutan appropriate MNH and nutrition interventions White Ribbon Alliance. more than 15 countries. 1998–2004 Bolivia • Establishment and promotion of international • Through competency-based education and training, Partners: Jhpiego, CEDPA, JHU-CCP Burkina Faso evidence-based standards for maternal and new- the program increased skilled attendance at birth in and PATH Egypt bornhealth,e.g.MCPCandMNPmanuals,andin- Burkina Faso, Guatemala, Indonesia, and Nepal. Guatemala corporation into national guidelines and protocols. • Burkina Faso: ANC coverage increased from 21% to Honduras • Facility assessments and strengthening. 58% while births attended by skilled providers in- .Oooi ta./ItrainlJunlo yeooyadOserc 3 21)S46 (2015) 130 Obstetrics and Gynecology of Journal International / al. et Otolorin E. India • Competency-based provider education and train- creased from 39% in 2001 to 58% in 2004. Indonesia ing in EmOC/EmONC, e.g. AMTSL, use of magne- • Nepal: Skilled attendance at birth increased from Nepal sium sulfate for eclampsia treatment, use of the 15% at baseline to 33% at endline. Pakistan partograph to monitor labor etc. • Afghanistan: Over 50 health providers trained in Peru • Promotion of birth planning and complication EmOC expanded services to all 32 provincial hospi- Tanzania readiness. tals. Healthcare workers were also trained in spinal Yemen • Performance and quality improvement programs in anesthesia for cesarean delivery. Zambia Guatemala, Burkina Faso, Indonesia, and Honduras. • Guatemala: Promoted development of Emergency Plans with better emergency funding and planning mechanisms for EmOC. • India: Established emergency response teams and education and training in EmOC. • Indonesia: Included PAC services and management of PPH as part of EmOC. • Nepal:ExpandedEmOCthroughskilledmidwifery and education and training activities and established 15-year plan aiming to provide BEmOC and CEmOC. • Pakistan: Practiced AMTSL for all deliveries and strengthened existing BEmONC sites (e.g. educating and training providers; improving infection prevention practices, donating basic obstetric equipment). • Yemen: Improved district hospitals' capacity to provide EmOC. Program: Afghanistan To improve the health and survival of mothers • Increased international and national attention and • Afghanistan, Ethiopia, Ghana, India, Malawi, ACCESS Cameroon and their newborns through expansion of commitment to improve MNH through global and Nigeria, Tanzania: Education and training in Date; Ethiopia coverage, access and use of MNH services and local alliances including the WHO, Roll Back BEmONC was implemented coupled with work – ’

2004 2009 Ghana through improving household health behaviors Malaria s Malaria in Pregnancy Working Group, to improve skills of midwifery tutors and pre- – Partners: Guinea and practices UNICEF, UNFPA, Partnership Maternal, Newborn and ceptors in 126 pre-service education institutions. S53 Jhpiego, Save the Children, Constella-Futures, Haiti Child Health, White Ribbon Alliance, International • Afghanistan: 96% of women in project interven- AED, ACNM, IMA World Health India Confederation of Midwives, International Federation tion areas received uterotonics for AMTSL com- Malawi of Gynecology and Obstetrics, Healthy Newborn pared with only 26% in the comparison area. 96% Mauritania Partnerships, etc. of women reached in the community by CHWs Nepal • Introduced and/or expanded focused antenatal care accepted to use misoprostol to prevent PPH. Niger services in 15 countries. • Nepal: Misoprostol was dispensed to 18 761 preg- Nigeria • Introduced and/or expanded PPH and PE/E pre- nant women by female CHWs, thereby increasing Rwanda vention programs in 14 countries. the proportion of deliveries protected by Tanzania • Introduced and/or expanded skilled attendance at uterotonics to 72.5% from 10.4% at baseline. Skilled Togo birthin15countries. attendance at birth increased from 9.9% to16%. • Introduced and or expanded newborn care in 16 • Ethiopia, Ghana, Malawi, Tanzania: 132 pre-- countries. service tutors/preceptors participated in technical and clinical skills standardization courses in BEmONC. • Malawi, Nigeria: National performance standards for EmONC were ﬁnalized and implemented. S49 (continued on next page) S50 Table 2 (continued)

Program/date/partners Countries Goals and strategic objectives Major EmONC activities and interventions Notable EmONC outcomes (education and training, service delivery, quality of care)

• Rwanda: Over 150 healthcare providers and CHWs were trained in BEmONC, leading to 100% of births in targeted facilities occurring with SBAs using a partograph and AMTSL. Program: MCHIP SBA: DRC, Ethiopia, To accelerate the reduction of maternal, • Conducted QoC assessments/introduced PQI/SBM-R. • Bangladesh: Through the MaMoni Program, CHWs Date: Ghana, India, Indonesia, newborn, and child mortality in 30 priority • Updated national standards of care. were trained to provide counseling on prevention 2008–2014 Kenya, Malawi, countries through the introduction, • Built capacity and competency of midwives and SBAs of PPH and distribution of misoprostol. Partners: Mozambique, Nigeria development, and scale-up of high-impact in prevention and treatment of PPH and PE/E. • Bolivia: Finalized 7 areas of SBM-R performance Jhpiego, Save the Children, John Snow, Inc., PPH prevention and maternal, newborn, and child health • Undertook multicountry assessments of PPH and standards while training 9 local consultants on Program for Appropriate Technology in Health, treatment:DRC, interventions PE/E programming in 2011 and 2012. AMTSL; advanced implementation of SBM-R for QI ICF International, Population Services Ghana, Guinea, India, • Operationalized updated WHO Guidelines on PPH in maternal health.

International, Broad Branch Associates, Johns Indonesia, Kenya, and PE/E. • DRC: expanded AMTSL activities, including S46 (2015) 130 Obstetrics and Gynecology of Journal International / al. et Otolorin E. Hopkins University’s Institute for International Liberia, Madagascar, conducting assessment of combined AMTSL/essential Programs Malawi, Mozambique, newborn care education and training packages; im- Nigeria, Rwanda, South proved performance of SBAs and supported update of Sudan, Zimbabwe national standards. PE/E prevention and • Ethiopia: The project is increasing the capacity of treatment: Ghana, local nongovernmental organizations to provide India, Indonesia, Kenya, BEmONC education and training, e.g. Ethiopian Malawi, Mozambique, Midwifery Association. Nigeria • Ethiopia, Kenya, Madagascar, Mozambique, Rwanda, Tanzania: Implemented the maternal and newborn QoC assessments with focus on management of PE/E, PPH, and birth asphyxia. • Ghana: Supported midwifery pre-service education. • India: Initiated SBM-R approach; increased capacity to provide anesthesia for EmONC. • Indonesia: Completed health facility survey; introduced SBM-R process with QI efforts; improved competency of midwives in AMTSL. • Madagascar: Adoption of national PPH prevention strategy through uterotonic coverage with emphasis on education and training on AMTSL using oxytocin and misoprostol; implemented QoC facility survey; developed capacity in BEmONC. • Malawi: Trained health providers on BEmONC; introduced PQI. • Mozambique: Institutionalized standardized QI –

approach in all 34 Model Maternity facilities S53 (country's largest EmONC facilities), including use of AMTSL and partograph. • Nepal: Conducted cross-over trial study about PE/E detection; provided in-service education. • Nigeria :ScaledupEmONCservicesto57health facilities; supported strengthening of Essential and EmONC with pre-service midwifery education. • Paraguay: Strengthened capacity for emergency care, especially capacity of SBAs. • Rwanda: Implemented maternal and newborn care QoC study. • Sierra Leone: Supported capacity building of SBAs. • Zimbabwe: Supported oxytocin potency testing.

Abbreviations: AMTSL, active management of the third stage of labor; ANC, antenatal care; BEmOC, basic emergency obstetric care; BEmONC, basic emergency obstetric and newborn care; CEmOC, comprehensive emergency obstetric care; CHWs, community health workers; EmOC, emergency obstetric care; EmONC, emergency obstetric and newborn care; MNH, maternal and newborn health; PAC, postabortion care; PE/E, pre-eclampsia/eclampsia; PPH, postpartum hemorrhage; PQI, performance and quality improvement; QoC, quality of care; QI, quality improvement; SBAs, skilled birth attendants; SBM-R, Standards-Based Management and Recognition (Jhpiego, Baltimore, USA). E. Otolorin et al. / International Journal of Gynecology and Obstetrics 130 (2015) S46–S53 S51

Box 1 [37,38]. However, current opinion is that further trial evidence is required Improving AMTSL service provision in Tanzania. to establish the efﬁcacy of partogram use [34].

In Tanzania, Jhpiego implemented the Mothers and Infants, Safe, 4.2. Scaling up use of magnesium sulfate for severe pre-eclampsia/eclampsia Healthy and Alive program to improve the quality of basic emergency obstetric and newborn care (BEmONC). Programmatic ap- Hypertensive disorders related to pregnancy affect about 10% of all proaches included: competency-based education of healthcare pregnant women around the world [39–41]. WHO has identified mag- providers on BEmONC; the development and implementation of nesium sulfate as the most effective and low-cost anticonvulsant for complementary national BEmONC standards as a tool for support- women with severe pre-eclampsia/eclampsia [42]. In compliance with ive follow-up of course participants in their workplace; procurement WHO guidelines, Jhpiego supported the use of magnesium sulfate in and distribution of equipment and supplies to fill identified gaps; each of the countries in which it worked with mixed success. In Nepal, and promotion of a supportive policy and advocacy environment. for instance, the ACCESS Program worked in collaboration with the These interventions have led to notable improvements in care. For Nepal Society for Obstetricians and Gynaecologists (NESOG) and the example, 60% of women received active management of the third Family Health Division of the Department of Health Services to stage of labor (AMTSL) according to the current guidelines and strengthen the use of magnesium sulfate for the treatment of severe standards in 2012—an increase of 19% from 2010. There was in- pre-eclampsia/eclampsia in 22 health facilities across Nepal, where creased use of oxytocin at all health facilities, and the overall use of pre-eclampsia/eclampsia accounts for 21% of all maternal deaths, 99% oxytocin in the performance of AMTSL reflected excellent using the SBM-R approach [43]. The drug was incorporated into pre- availability. In 2012, only four cases of postpartum hemorrhage and in-service education for Nepali SBAs starting in 2006 and was in- (PPH) (0.8% of 500 births) were observed compared with 10 cases cluded in the national medical standards and essential medicines list. in 2010, which is a testament to the impact of preventing PPH Evaluations of Nepali providers, following continuing professional edu- through AMTSL. All PPH cases were managed successfully. cation activities that were focused on appropriate use of the drug, indicated that the average facility score increased from 26% to 60% [43]. Eight out of 22 sites scored 80% or higher and another 12 sites showed substantial improvements [43]. By the end of the ACCESS program in and Recognition (SBM-R) (Jhpiego, Baltimore, USA) [30],inmanyofits 2009, 11 of the 22 facilities were performing at 80% or higher in achiev- EmONC projects as a complement to its education and training efforts. ing evidence-based standards [43]. During site visits, staff identified Assessment of facilities and providers according to EmONC performance gaps and created an action plan to address them. NESOG members con- standards helped draw attention to gaps in the quality of care at health ducted clinical updates, disseminated job aids, and supported staff with facilities and implementation of solutions that could lead to quality im- on-site coaching. This (and similar) experience(s) generated the lesson provement. The use of the SBM-R approach has been documented as a learned that national professional organizations can be engaged in ca- strategy that can help to improve the quality of maternal and newborn pacity building of healthcare workers to champion the scale-up of criti- care in many countries and under a variety of thematic areas, as a key cal evidence-based interventions for MNH. lesson learned. A full discussion of SBM-R is offered in another paper in this supplement [31]. 4.3. Essential newborn care 4. Translating research to practice Based on the evidence that outcomes for the mother and the new- 4.1. Use of the partograph born are intimately linked, ACCESS worked in 16 countries to improve outcomes for newborns with a particular focus on promotion of thermal The partograph, a tool used to provide a continuous pictorial overview protection, early and exclusive breastfeeding, newborn resuscitation, of labor on a preprinted single sheet of paper, is an obstetric care practice and infection prevention and treatment. Following the global introduc- that has been proposed as a “routine care” signal function. WHO has cited tion of the Helping Babies Breathe (HBB) program [44], MCHIP prepared fi the partograph as one of its essential interventions [32] because it helps hundreds of healthcare workers in essential newborn care and speci - identify obstructed labor [33] and thus helps to prevent uterine rupture, cally in HBB. HBB has been integrated into BEmONC education and one of the five main causes of maternal death, and birth asphyxia, one training in many countries such as Tanzania and Zimbabwe and is of the three main causes of newborn mortality. The partograph can discussed further in this Supplement [45]. In Nepal, for example, prompt a healthcare provider to take action, especially in low-resource under the ACCESS program, community health volunteers were trained settings where prolonged labor and delay in decision-making are signifi- in kangaroo mother care, resulting in a 60% adoption of the practice. The cant causes of adverse obstetric outcomes [34]. ICM includes use of the fact that low-cost, evidence-based, essential newborn interventions can partograph as a basic competency for midwives because of its value as be rapidly scaled-up in low-resource settings provided there is strong a clinical management tool in low-resource settings [35]. Jhpiego and government ownership and coordination of relevant stakeholders, other global implementing organizations emphasize its use even though was the important lesson learned. adoption is often slow. In Nigeria, for example, despite repeated educational sessions and supportive follow-up visits, use of the partograph at 5. Opportunities and challenges endline was only documented in 44.4% of 81 437 deliveries. Reasons frequently given for the suboptimal use of the partograph include late pre- While remarkable progress has been made toward the reduction of sentation of the women in labor (often in second stage) and staff maternal and child mortality in many low-resource countries, critical shortage in very busy labor wards [36]. Based on lessons learned from challenges remain in provision of high-quality EmONC services, partic- program experience, Jhpiego has adapted the text and pictures used in ularly in Sub-Saharan Africa and Southeast Asia. The global community its educational materials to reflect the specific context in which the must focus on reaching the poorest and most vulnerable populations to partograph will be used, to make it as simple as possible for providers address persistent inequities [46]. These inequities include, among to learn and integrate into their daily practice. Jhpiego has also other things, a shortage of SBAs in the most vulnerable communities recently introduced “E-Partograms”—electronic handheld devices—based that is driven by lack of targeted workforce planning strategies, for ex- on WHO’s paper partograph, designed to increase accessibility to ample matching deployment with the competencies of providers and patient data between different levels of healthcare providers and facilities addressing well-known factors that discourage workforce retention. S52 E. Otolorin et al. / International Journal of Gynecology and Obstetrics 130 (2015) S46–S53

At the very least, we can ensure that the SBAs, who are available, have management of life-saving commodities and consumable supplies, the knowledge and skills needed to provide the EmONC services that save record-keeping and use of data for decision-making. Countries, in col- maternal and newborn lives. Jhpiego has worked to achieve this objective laboration with their relevant professional organizations, should be in many low-resource countries over the past decade by influencing in- supported to initiate and own vital registrations of births and deaths ternational standards for EmONC; developing uniform, evidence-based, and institutional or community maternal and perinatal death reviews. educational approaches tailored for low-resource settings; and continual- Such confidential enquiries should embrace a “no name, no blame” pol- ly updating practices based on new research findings. icy to encourage honest and productive dialogue that will lead to quality Jhpiego’s programming has been cross-cutting along the continuum improvement of MNH services. of care. Responsive to identified country needs, Jhpiego has attempted to address not only elements of the proposed expansion of EmONC signal functions but also complementary health services that strengthen Conflict of interest the quality of MNH service delivery. Issues still persist, however, that adversely affect the delivery of high- The authors have no conflicts of interest. quality maternal, newborn, and EmONC services in health facilities. These include stock-outs of essential maternal, newborn, and child health com- References modities due to poor logistic management systems; limited opportunities for providers to access supportive supervision and engage in continuing [1]WHO,UNICEF,UNFPA,WorldBank.Trendsinmaternalmortality:1990to professional development activities; difficulties in keeping up-to-date 2010. Geneva: WHO; 2012. http://www.unfpa.org/public/home/publications/ clinical guidelines based on emerging scientific evidence; weak record- pid/10728. Accessed April 3, 2014. [2] Paxton A, Maine D, Freedman L, Fry D, Lobis S. The evidence for emergency obstetric keeping and monitoring and evaluation systems that limit opportunities care. Int J Gynecol Obstet 2005;88(2):181–93. to document both successes and challenges to service delivery; and [3] World Health Organization, International Confederation of Midwives, International inadequate financing of country programs, or conversely, excessive federation of Gynecology and Obstetrics. Making pregnancy safer: the critical role of the skilled attendant. A joint statement by WHO, ICM and FIGO. Geneva: WHO; donor dependency that stalls sustained progress. 2004. http://www.who.int/maternal_child_adolescent/documents/9241591692/en/. Jhpiego and MCHIP have worked in alliance with other existing global Accessed April 3, 2014. programs, such as the Global Fund for AIDS, Tuberculosis and Malaria, the [4] Campbell OM, Graham WJ, Lancet Maternal Survival Series steering group. Strategies President's Emergency Plan for AIDS Relief, the Partnership for Maternal, for reducing maternal mortality: getting on with what works. Lancet 2006; 368(9543):1284–99. Newborn and Child Health, and the Child Survival Call to Action, to pro- [5] World Health Organization, UNFPA, UNICEF, AMDD. Monitoring emergency obstet- vide opportunities for addressing issues that overarch the provision of ric care: a handbook. Geneva: WHO; 2009. http://whqlibdoc.who.int/publications/ quality health systems, such as the need for health systems strengthening 2009/9789241547734_eng.pdf?ua=1. Accessed April 3, 2014. [6] Hofmeyr GJ, Haws RA, Bergström S, Lee A, Okong P, Darmstadt G, et al. Obstetric care and strategies for management of communicable diseases that affect in low-resource settings: what, who, and how to overcome challenges to scale up? women and their families. Jhpiego and MCHIP have also worked with ini- Int J Gynecol Obstet 2009;107(Suppl. 1):S21–45. tiatives to address community- and facility-level challenges, such as the [7] Hussein J, Kanguru L, Astin M, Munjanja S. The effectiveness of emergency obstetric referral interventions in developing country settings: a systematic review. PLoS Med United Nations Commission for Life-Saving Commodities for Women 2012;9(7):e1001264. and Children’s Health, which aims to address the problem of stock-outs [8] van Lonkhuijzen L, Dijkman A, van Roosmalen J, Zeeman G, Scherpbier A. A system- of key commodities at the community level, and HBB, which focuses on atic review of the effectiveness of training in emergency obstetric care in low- – fi resource environments. BJOG 2010;117(7):777 87. having at least one quali ed provider in each delivery setting ready to [9] Dogba M, Fournier P. Human resources and the quality of emergency obstetric care provide newborn transition services. New opportunities are emerging in developing countries: a systematic review of the literature. Hum Resour Health in the post Millennium Development Goal era, such as: the Every 2009;7:7. [10] Gabrysch S, Civitelli G, Edmond KM, Mathai M, Ali M, Bhutta ZA, et al. New signal Woman, Every Child; Every Newborn Action Plan; the Commission on In- functions to measure the ability of health facilities to provide routine and emergen- formation and Accountability for Women's and Children's Health; A cy newborn care. PLoS Med 2012;9(11):e1001340. Promise Renewed-Call to Action; Ending Preventable Maternal and [11] Jhpiego. Jhpiego: a history. http://www.jhpiego.org/content/jhpiego-history. Child Deaths and Ending Preventable Maternal Mortality. Countries Accessed June 18, 2014. [12] World Health Organization. Monitoring the building blocks of health systems: a hand- need to take advantage of these global initiatives to leverage resources book of indicators and their measurementstrategies.Geneva:WHO;2010.www.who. for their efforts to achieve reductions in maternal and neonatal morality. int/healthinfo/systems/monitoring/en/index.html. Accessed April 3, 2014. ’ These challenges and potential opportunities need to be addressed [13] World Health Organization. Everybody s Business: Strengthening health systems to improve health outcomes. WHO’s framework for action. Geneva: WHO; 2007. in order to achieve a rights-based approach to healthcare delivery http://www.who.int/healthsystems/strategy/everybodys_business.pdf. Accessed [47]. Innovative approaches need to be implemented at scale to extend April 3, 2014. universal coverage of high-impact interventions and reduce or elimi- [14] Kerber KJ, de Graft-Johnson JE, Bhutta ZA, Okong P, Starrs A, Lawn JE. Continuum of fi care for maternal, newborn, and child health: from slogan to service delivery. Lancet nate inequity while fostering ef ciency of the services delivered. Coun- 2007;370(9595):1358–69. tries should prioritize workforce planning to increase the number of [15] Thaddeus S, Maine D. Too far to walk: maternal mortality in context. Soc Sci Med SBAs who will provide these high-impact interventions at the health fa- 1994;38(8):1091–110. [16] Pacagnella RC, Cecatti JG, Osis MJ, Souza JP. The role of delays in severe maternal cility or at home in countries with high homebirth rates. In the interim, morbidity and mortality: expanding the conceptual framework. Reprod Health Mat- provision should be made for task-shifting or task-sharing policies that ters 2012;20(39):155–63. will allow for low-dose high-frequency capacity building of lower level [17] World Health Organization, UNFPA, UNICEF, World Bank. Managing complications in pregnancy and childbirth: a guide for midwives and doctors. http:// cadres of staff to provide selected basic EmONC services. For example, whqlibdoc.who.int/publications/2007/9241545879_eng.pdf?ua=1 Geneva: such staff could be trained in community distribution of misoprostol WHO; 2000 (reprint 2007)Accessed April 3, 2014. for AMTSL or chlorhexidine for prevention of umbilical cord sepsis. [18] World Health Organization, UNFPA, UNICEF, World Bank. Managing newborn They can also be trained to administer the loading dose of magnesium problems: a guide for doctors, nurses, and midwives. http://whqlibdoc.who. int/publications/2003/9241546220.pdf Geneva: WHO; 2003Accessed April 3, 2014. sulfate before referral of eclamptic patients to higher level facilities. In [19] USAID, MCHIP, PRE-EMPT, World Health Organization. WHO recommendations for addition to health workforce interventions, countries need to identify prevention and treatment of pre-eclampsia and eclampsia: implications and actions. fi and mobilize existing community structures to educate pregnant http://www.mchip.net/sites/default/ les/PEE%20Briefer.pdf. Accessed April 3, 2014. [20] Bowser D, Hill K. Exploring evidence for disrespect and abuse in facility-based child- women and their families on danger signs in pregnancy, during and birth: report of a landscape analysis. http://www.mhtf.org/wp-content/uploads/ after childbirth, and to advocate for institutional deliveries. Countries sites/17/2013/02/Respectful_Care_at_Birth_9-20-101_Final.pdf. Published 2010. and development partners should also invest in health systems Accessed April 3, 2014. [21] Jhpiego. Site assessment and strengthening for maternal and newborn health pro- strengthening to improve basic infrastructure in maternity units includ- grams. http://www.jhpiego.org/files/SiteAssessMNH.pdf Baltimore, USA: JHPIEGO; ing electricity and water supply, blood transfusion services, logistic 2004Accessed April 3, 2014. E. Otolorin et al. / International Journal of Gynecology and Obstetrics 130 (2015) S46–S53 S53

[22] Islam MT, Haque YA, Waxman R, Bhuiyan A. Implementation of emergency obstetric [35] Fullerton J, Thompson J. 2013 Amendments to international conference of midwives’ care training in Bangladesh: lessons learned. Reprod Health Matters 2006;14(27): essential competencies and education standards core documents: clarification and 61–72. rationale. Int J Childbirth 2013;3(4):184–94. [23] Mavalankar D, Sriram V. Provision of anaesthesia services for emergency obstetric [36] Airede LR, Ishola G, Segun T, Isiugo-Abanine U, Rwalins B, Otolorin E. Strengthening care through task shifting in South Asia. Reprod Health Matters 2009;17(33):21–31. Emergency Obstetric and Newborn Care and Family Planning Services in Northwest [24] Jhpiego. Emergency Obstetric Care for Doctors and Midwives. http://reprolineplus. Nigeria (poster). http://www.mchip.net/sites/default/files/mchipfiles/FPService_ org/resources/emergency-obstetric-care-doctors-and-midwives-learning-resource- GHC.pdf. Published 2011. Accessed April 7, 2014. package. Published in 2003. [37] Jhpiego. Jhpiego Innovations Program: development, delivery and impact at scale. [25] Jhpiego. Anesthesia for Emergency Obstetric Care for Doctors and Midwives. http:// http://www.jhpiego.org/en/content/innovations. Accessed July 1, 2014. reprolineplus.org/resources/anesthesia-emergency-obstetric-care-doctors-and- [38] USAID. The Catalog: version 1.0. http://www.jhpiego.org/files/innovations/USAID_ midwives-learning-resource-package. Published 2003. Accessed April 7, 2014. The_Catalog_Version_1.pdf. Published 2013. Accessed July 1, 2014. [26] Bluestone J, Johnson P, Fullerton J, Carr C, Alderman J, BonTempo J. Effective in- [39] Duley L. The global impact of pre-eclampsia and eclampsia. Semin Perinatol 2009; service training design and delivery: evidence from an integrative literature review. 33(3):130–7. Hum Resour Health 2013;11:51. [40] Steegers EA, von Dadelszen P, Duvekot JJ, Pijnenborg R. Pre-eclampsia. Lancet 2010; [27] Jhpiego. Guidelines for assessment of skilled providers after training in maternal and 376(9741):631–44. newborn healthcare. http://www.jhpiego.org/files/GdlnsSkillProvEN.pdf Baltimore, [41] Roberts CL, Ford JB, Algert CS, Antonsen S, Chalmers J, Cnattingius S, et al. USA: JHPIEGO; 2004Accessed April 3, 2014. Population-based trends in pregnancy hypertension and pre-eclampsia: an interna- [28] Jhpiego. Maternal and newborn health. http://reprolineplus.org/resources/technical/ tional comparative study. BMJ Open 2011;1(1):e000101. maternal-newborn-health. Accessed April 3, 2014. [42] World Health Organization. WHO recommendations for prevention and treatment [29] Paxton A, Bailey P, Lobis S, Fry D. Global patterns in availability of emergency obstet- of pre-eclampsia and eclampsia: evidence base. http://whqlibdoc.who.int/hq/ ric care. Int J Gynecol Obstet 2006;93(3):300–7. 2011/WHO_RHR_11.25_eng.pdf. Accessed April 3, 2014. [30] Necochea E, Bossemeyer D. Field guide: Standards-based management and recognition [43] USAID, ACCESS. Strengthening the use of magnesium sulphate for management of se- learning resource package. http://reprolineplus.org/resources/field-guide-standards- vere pre-eclampsia and eclampsia. http://reprolineplus.org/resources/strengthening- based-management-and-recognition-learning-resource-package. Published 2005. use-magnesium-sulphate-management-severe-pre-eclampsia-and-eclampsia.Pub- Accessed April 3, 2014. lished 2010. Accessed April 9, 2014. [31] Necochea E, Tripathi V, Kim Y-M, Akram N, Hyjazi Y, Rashidi T, et al. Improving qual- [44] Van Heerden C, Cur B, Cur M. An introduction to Helping Babies Breathe: the “Golden ity at scale: how quality improvement can foster the implementation of high impact Minute” is here for South African newborn babies. Prof Nurs Today 2012;16(3):1–2. evidence-based practices in maternal and newborn health. Int J Gynecol Obstet 2015. [45] Smith J, de Graft-Johnson, Azfar P, Ricca J, Fullerton J. Scaling up high-impact inter- [32] The Partnership for Maternal, Newborn and Child Health. Essential interventions, com- ventions: how is it done? Int J Gynecol Obstet 2015. modities and guidelines for reproductive health, maternal, newborn and child health: [46] Souza J, Gulmezoqlu AM, Vogel J, Carroli G, Lumbiganon, Qureshi Z, et al. Moving be- a global review of the key interventions related to reproductive, maternal, newborn yond essential interventions for reduction of maternal mortality (the WHO and child health (RMNCH). http://www.who.int/pmnch/knowldeges/publications/ multicountry survey on maternal and newborn health): a cross-sectional study. 201112_essential_interventions/en/ Geneva: PMNCH; 2011Accessed April 7, 2014. Lancet 2013;381(9879):1747–55. [33] Javed I, Bhutta S, Shoaib T. Role of partogram in preventing prolonged labour. J Pak [47] Hammonds R, Ooms G. The emergence of a global right to health norm – the unre- Med Assoc 2007;57(8):408–11. solved case of universal access to quality emergency obstetric care. BMC Int Health [34] Lavender T, Hart A, Smyth RM. Effect of partogram use on outcomes for women in Hum Rights 2014;14:4. spontaneous labour at term. Cochrane Database Syst Rev 2013;7:CD005461. International Journal of Gynecology and Obstetrics 130 (2015) S54–S61

Contents lists available at ScienceDirect

International Journal of Gynecology and Obstetrics

journal homepage: www.elsevier.com/locate/ijgo

SUPPLEMENT ARTICLE A facility birth can be the time to start family planning: Postpartum intrauterine device experiences from six countries

Anne Pﬁtzer a,⁎, Devon Mackenzie a, Holly Blanchard a, Yolande Hyjazi b, Somesh Kumar c, Serawit Lisanework Kassa d, Bernabe Marinduque e, Marie Grace Mateo e, Beata Mukarugwiro f, Fidele Ngabo g, Shabana Zaeem h,1, Zonobia Zafar h, Jeffrey Michael Smith i a Jhpiego/MCHIP, Washington, DC, USA b Jhpiego, Conakry, Guinea c Jhpiego, New Delhi, India d Jhpiego, Addis Ababa, Ethiopia e Jhpiego, Manila, Philippines f Jhpiego, Kigali, Rwanda g Ministry of Health, Kigali, Rwanda h Jhpiego, Islamabad, Pakistan i Jhpiego, Baltimore, MD, USA article info abstract

Keywords: Initiation of family planning at the time of birth is opportune, since few women in low-resource settings who give Intrauterine device birth in a facility return for further care. Postpartum family planning (PPFP) and postpartum intrauterine device Postpartum family planning (PPIUD) services were integrated into maternal care in six low- and middle-income countries, applying an inser- Postpartum care tion technique developed in Paraguay. Facilities with high delivery volume were selected to integrate PPFP/ Prenatal care PPIUD services into routine care. Effective PPFP/PPIUD integration requires training and mentoring those pro- Program implementation viders assisting women at the time of birth. Ongoing monitoring generated data for advocacy. The percentages of PPIUD acceptors ranged from 2.3% of women counseled in Pakistan to 5.8% in the Philippines. Rates of complications among women returning for follow-up were low. Expulsion rates were 3.7% in Pakistan, 3.6% in Ethiopia, and 1.7% in Guinea and the Philippines. Infection rates did not exceed 1.3%, and three countries recorded no cases. Offering PPFP/PPIUD at birth improves access to contraception. © 2015 Published by Elsevier Ireland Ltd. on behalf of International Federation of Gynecology and Obstetrics. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

1. Background for integrating these services are limited. In many countries, institutional births are on the rise, and there is strong policy support for the use of In 2012, an estimated 222 million women in low-resource countries skilled birth attendants. Thus, initiation of family planning during a wanted to avoid pregnancy but were not using modern contraception facility stay at the time of birth is particularly opportune, especially [1]. For many of these women, childbearing begins at an early age, inter- since few women who give birth in a facility return for further postnatal vals between pregnancies are too short, and lifetime fertility is high [2]. care [6–8]. The resulting fertility patterns lead to excess mortality and morbidity for According to the WHO, “postpartum family planning (PPFP) focuses both mothers and offspring [3–5]. Although family planning services are on the prevention of unintended and closely spaced pregnancies intended to address desires to space and limit births, typically they are through the first 12 months following childbirth” [9].Operationalizing offered separately from maternity services. The providers who work in PPFP requires integration of family planning with maternal, newborn, family planning units frequently are not the same individuals who and child health services (see Fig. 1). In the present paper, we define care for women prenatally, at birth, and postnatally, so opportunities “immediate postpartum” as the first 48 hours after birth, “early postpartum” as the six weeks after a birth, and “extended postpartum” as the 12 months after a birth. The provision of a contraceptive method ⁎ Corresponding author at: Jhpiego/Maternal and Child Health Integrated Program, before discharge ensures that women are protected against pregnancy 1776 Massachusetts Avenue NW, Suite 300, Washington, DC 20036, USA. Tel.: +1 202 before they resume sexual activity or return to fecundity. Family plan- 835 6072. E-mail address: Anne.Pfi[email protected] (A. Pfitzer). ning programs with a wide range of contraceptive choices are associat- 1 (retired). ed with greater use and lower costs [10]. Yet, according to current WHO http://dx.doi.org/10.1016/j.ijgo.2015.03.008 0020-7292/© 2015 Published by Elsevier Ireland Ltd. on behalf of International Federation of Gynecology and Obstetrics. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). A. Pfitzer et al. / International Journal of Gynecology and Obstetrics 130 (2015) S54–S61 S55

Fig. 1. Health-related contact points from pregnancy to the extended postpartum period and opportunities for family planning integration. Adapted with permission from World Health Organization [9]. recommendations, contraceptive choices are more limited during postpartum insertion. These studies did not describe the PPIUD the early postpartum period, in particular for hormonal methods, if a insertion technique used other than to indicate hand or instrumental woman follows recommendations to exclusively or predominantly insertion [14–17]. breastfeed her baby [11,12]. Some countries allow progestin-only hor- In the present article, we present program experiences from six monal methods in the immediate postpartum for lactating women, countries where Jhpiego or the Jhpiego-led and USAID-supported Ma- but many follow the WHO’s recommendation to delay initiation of ternal and Child Health Integrated Program (MCHIP) integrated PPFP these methods for at least six weeks. Exclusive breastfeeding for the into maternal care, with PPIUD services offered. We describe the rollout ﬁrst six months, without resumption of menses—the lactational amen- and implementation of the programs, present service data on uptake orrhea method (LAM)—is a highly effective method of family planning and follow-up, and discuss operational challenges and solutions to sup- in the short term [13]. Other immediate contraceptive options include port the scale-up and replication of PPIUD services in other countries. All the postpartum intrauterine device (PPIUD), which is an intrauterine six of the country programs used the Copper T 380A IUD. The Copper T is device (IUD) inserted soon after delivery, as opposed to the interval reversible and effective for 12 years, requires very little routine follow- IUD inserted later, or postpartum tubal ligation. PPFP includes any con- up, and can be inserted within 10 minutes of placental expulsion, during traception used during the extended postpartum, regardless of timing cesarean deliveries, or within 48 hours after childbirth [18]. of initiation, of which PPIUD is one option and is limited to the 48 hours after a birth. Quality programs always counsel women on all 2. Methods and context their PPFP options. Box 1 describes modalities for PPIUD insertion. As a result of this convenient timing, PPFP/PPIUD services can be The authors reviewed program documentation, country-level mon- organized to take advantage of prenatal care and labor and delivery as itoring and evaluation databases, and monthly summary reports from prime opportunities to address postpartum contraceptive needs. In participating health facilities to assess commonalities and differences many countries, PPFP/PPIUD services also align with national efforts in program implementation. In Pakistan, Jhpiego staff obtained data to promote facility-based births. However, early studies of PPIUDs compiled by facilities from the district or regional headquarters, while that have examined different types and timings of insertion (up to in Ethiopia, India, and the Philippines, data were obtained from monthly seven days after birth) have found high expulsion rates, from 3.7% to facility reports to program staff. In Guinea, monthly reports were sent to over 30%, with mixed conclusions concerning post-placental or early program staff at the same time as to the ministry of health (MOH). The S56 A. Pﬁtzer et al. / International Journal of Gynecology and Obstetrics 130 (2015) S54–S61

Box 1 Types of PPIUD insertions, timings, and modalities.

Insertion type Timing Provider Description Post-placental 0 to 10 minutes Physician, midwives, Insertion, either manual or instrumental, performed after a nurses, and others vaginal birth, immediately after expulsion of placenta. Assumes counseling and choice made prenatally. Immediate postpartum 10 minutes up Physician, midwives, Insertion performed later than 10 minutes after expulsion of or pre-discharge to 48 hours nurses, and others placenta, typically the morning following the birth, by same or different provider. Sometimes called “morning after” insertion. Can be performed following postpartum counseling and choice in cases where prenatal counseling was not possible. Intra-cesarean During the cesarean Physicians and cadres Insertion done during cesarean delivery, after expulsion of delivery authorized to perform placenta and through the uterine incision, using either instruments cesarean deliveries or by hand. Uterine incision is closed once the intrauterine device is in place. This type of insertion assumes prenatal counseling and choice.

MOH of Rwanda provided the Rwanda data. The timing of interventions 3.1. PPIUD service delivery models varied across countries, as did the availability of program data. Where country programs collected data with a lesser level of detail, we have in- Ideally, women would receive PPFP counseling during prenatal care dicated in our results that data are missing. and decide which family planning method to use—a choice that would Secondary analysis of Demographic and Health Survey (DHS) data then be recorded in the prenatal care record and communicated to her for prospective unmet need among women in the first year postpartum provider when she arrives at the facility in early labor. However, if a cli- in the six countries shows high unmet need for family planning during ent has not been counseled before birth, she can still be counseled dur- the first year postpartum (Table 1) [19]. “Prospective unmet need” is ing latent labor (if she is receptive) or in the immediate postpartum based on DHS questions about desires for another child within the period. Active labor is not considered an appropriate time to provide next two years (asked of postpartum women as of other nonpregnant PPFP counseling. PPIUDs are provided to clients who want them and women), whereas the standard unmet need definition is based on who do not have exclusion factors, i.e. unresolved postpartum hemor- questions about whether a current or recent pregnancy was intended, rhage or chorioamnionitis. Every woman’s request for a PPIUD is con- for those women who are either pregnant or had a birth within the firmed immediately prior to insertion. past two years. Total prospective unmet need for spacing and limiting All six countries have adopted and trained providers on the insertion births during the first 12 months after delivery ranged from 60.5% in technique pioneered by Dr Vicente Battaglia Araujo, which focuses the Philippines to 81.2% in Ethiopia. Overall, the use of IUDs ranged on achieving high fundal placement of the IUD. This is accomplished from a fraction of a percent of married women in Ethiopia, Guinea, by elongating the cervical-uterine angle, elevating the uterus and the and Rwanda to 3.7% of married Filipino women currently using a con- use of a long placental, or “Kelly,” forceps (Fig. 2) [20]. Placement of traceptive [19]. Currently-married women in Pakistan and India report- the IUD high in the fundus minimizes the risk of expulsion [21,22]. ed IUD use of 1.7% and 2.3%, respectively. Jhpiego has also worked with an anatomic model manufacturer to devise a new uterine model, the “Mama-U,” that simulates the problem of the cervical-uterine angle to practice PPIUD insertion competencies. 3. PPIUD program elements: Process and variations

Table 2 shows the locations of PPFP/PPIUD services in the six coun- 3.2. Demand creation tries and the numbers and cadres of providers trained by Jhpiego and MCHIP from 2010–2013. India and the Philippines trained more Approaches to creating demand for PPIUDs have evolved over time. physicians than nurses and midwives, whereas the other country pro- Initially, demand creation materials, such as counseling pamphlets and grams did the opposite. From our experiences in these countries, we posters, were developed for use in both individual and group counseling have synthesized common steps in the introduction of PPIUDs as part settings to recruit patients attending prenatal care at facilities where of a PPFP program. PPIUD services were available. Women who were missed in prenatal care were counseled in early labor or postpartum. In all programs, this Table 1 facility-based approach alone was enough to maintain a steady level a Prospective unmet need for family planning among women in the ﬁrst year postpartum. of PPIUD acceptors. The workload associated with PPFP counseling led Country Percentage (%) Percentage (%) Total percentage the India program to develop and hire a new cadre of dedicated coun- (DHS year) of women with of women with (%) of women selors for this purpose: social workers assigned to facilities providing unmet need for unmet need for with unmet need PPIUDs. In 2011 and 2012, providers and obstetrics and gynecology fac- spacing limiting ulty successfully advocated to the Government of India to hire these Ethiopia (2011) 50 31 81 counselors as facility staff. As a result, in 2012, more than 1300 new po- Guinea (2005) 58 17 74 sitions were created for dedicated counselors at high-volume facilities India (2005–2006) 33 33 65 Pakistan (2006–2007) 35 30 65 across the country through the National Rural Health Mission. The ded- Philippines (2008) 24 36 61 icated counselors are typically stationed adjacent to the prenatal care Rwanda (2010) 33 29 62 ward in facilities, and all women who access prenatal care are routed Abbreviation: DHS, Demographic and Health Survey. through the counselor. The counselor also performs rounds in the ma- a Source: USAID [19]. ternity and postnatal wards to counsel women on PPFP. A. Pﬁtzer et al. / International Journal of Gynecology and Obstetrics 130 (2015) S54–S61 S57

Table 2 Country program postpartum intrauterine device site and training proﬁles.

Country Dates of reporting Catchment area population Number of Jhpiego-assisted facilities Number of providers trained

Health centers Hospitals Doctors Nurses/midwives

Ethiopia March 2012–July 2013 33 255 949 3 15 7 54 Guinea April 2011–April 2013 2 550 393 4 14 19 22 India February 2010–April 2013 1 141 158 597 0 330 1385 705 Pakistan September 2012–June 2013 1 466 465+ (estimated) 0 20 84 413 Philippines January–June 2013 12 904 871 1 9 16 7 Rwanda May 2010–February 2013 unknown 8 5 12 32

Over time, strategies to extend demand generation beyond the facil- program strategically introduced PPIUD services in one medical college ity have emerged. Guinea relies on community health workers to spread and expanded them to other states after carefully documenting that the the word about PPFP services, while Pakistan supplies PPFP leaflets and expulsion rate remained low. Experts in Rwanda conducted an opera- counseling cards to lady health workers operating at the community tions research study to assess feasibility, and the results were used to in- level. Indian states now involve community-based, accredited social form guidelines for PPIUD insertion. In both Pakistan and Ethiopia, health activists (ASHAs) in demand generation. Future efforts might intensive policy engagement focused first on the provincial and regional involve mass media. levels before moving to national scale. In Pakistan, a technical working group with members from the public and private sectors and nongov- 3.3. Policy and advocacy ernmental organizations developed a PPFP strategy and implementation plan. The group held several advocacy meetings and presented A policy to introduce PPIUDs was common across the six countries, the plan at meetings and seminars organized by professional obstetrics as governments and ministries of health aimed to address unmet need and gynecology bodies and a medical university. In Ethiopia, early en- for contraception and reduce the proportion of short birth intervals. In gagement and advocacy to support initiating PPFP/PPIUD service deliv- India, the Ministry of Health and Family Welfare launched a national ery in select facilities focused on regional health bureaus as gatekeepers strategy in 2009 to reinvigorate PPIUD use as part of a maternal, new- for the individual facilities. The FMOH focal person was aware of the born, and child health initiative. These revitalization efforts dovetailed proposed PPIUD introduction at the regional level but not the details with the government-initiated Janani Suraksha Yojna (safe mother- of the program. At a national family planning symposium in November hood) scheme, which encouraged childbirth with a skilled birth atten- 2012, an update on the program’s progress attracted national attention dant and has resulted in a more than 15-fold increase in institutional and resulted in significant policy support at the federal level. In all coun- deliveries since 2005 [23]. In Guinea, the Philippines, and Rwanda, na- tries, the PPFP/PPIUD programs would not have been possible without tional strategies to revitalize family planning emphasized long-acting ministry leadership and ownership. methods, including PPIUDs. In the Philippines, the Department of Health Engagement with government decision-makers, medical, or aca- approved in late 2013 a separate guideline for PPFP, which includes the demic gatekeepers to advocate regarding the safety and benefits of PPIUD [24]. Similarly, the Federal Ministry of Health (FMOH) of Ethiopia PPFP/PPIUD services has been a key and ongoing component of the supported expanding access to services by incorporating Jhpiego’s programs. In preparation for initiating PPIUD services, Rwandan PPIUD training package into the national training package for family stakeholders visited a demonstration program in Kenya [25],while planning. In Pakistan, Jhpiego collaborated with the health and popula- project staff in Pakistan looked for opportunities at professional tion departments at the subnational level to establish PPFP services at meetings to marshal interest in and overcome resistance to PPIUDs. public health facilities. All six countries needed continued advocacy after initiation of The decision to initiate a PPFP program that includes PPIUDs in- PPIUD services and have shared their progress along the way, includ- volves engagement of international and national experts with relevant ing service data on expulsions and infections, to help address resis- ministry officials. The favorable change in WHO medical eligibility tance. For example, in the Philippines, the program worked to criteria for IUDs in the immediate postpartum to category 1 has influ- incorporate PPIUD into new clinical practice guidelines for cesarean enced this decision in many countries [11]. In India, leading voices delivery and family planning. In India, whenever the program ex- from reputable universities were instrumental in reviewing the litera- pands to new states, it begins with a state-level workshop for the ture on PPIUDs and addressing concerns about expulsion rates. The heads of all district health teams.

Fig. 2. Illustration of uterine extension during placement of postpartum intrauterine device. Adapted with permission from Jhpiego [20]. S58 A. Pﬁtzer et al. / International Journal of Gynecology and Obstetrics 130 (2015) S54–S61

3.4. Site selection and training coordinated with government, begin immediately after training to reinforce provider performance, promote quality services, and troubleshoot. Site selection for PPIUD services has focused on facilities with: (1) a The programs in Ethiopia, Guinea, India, Pakistan, and Rwanda also con- high volume of deliveries; (2) interested providers; and (3) adequate duct quarterly supportive supervision visits. In some cases, like Guinea, staff to integrate PPIUD service delivery into routine care, along with program staff conduct these visits, while elsewhere, like Pakistan, these the ability to conduct supportive supervision. Through needs assess- are conducted jointly with government supervisors. Structured tools are ments, programs identified sites with high delivery caseloads and gaps used to support supervision in India, Pakistan, and the Philippines. in equipment and training. In the Philippines, the program specifically Because routine health management information systems (HMIS) selected “centers of excellence” in 10 geographical areas to serve as are not yet adapted for tracking PPIUDs, as they are for interval IUD model service delivery sites for training, with the intent of increasing insertions and removals, programs have established a variety of sup- sustainability. In Pakistan, the program conducted qualitative formative plementary records and data collection tools and systems, including research with new mothers, husbands, and grandmothers (mothers-in- modified or separate registers in prenatal and labor and delivery law) to help craft better counseling messages and educational ma- wards, client cards, or other similar files kept in facilities. The need to terials. In Guinea, the program started in six high-volume facilities in capture country-specific data and evidence of the feasibility and safety the capital city, Conakry. The facilities served as clinical practicum of PPFP/PPIUD services justified the request for supplemental data sites for medical and midwifery school students. The program then ex- reporting. Guinea and the Philippines have included PPFP on their panded to sites in the interior of the country. In all six programs, instru- national HMIS forms. The Guinea National HMIS and National Safe ment kits—containing, in particular, long placental forceps such as Kelly Motherhood Program approved a modified labor and delivery register forceps—were distributed when training was initiated. In some coun- with a column added for PPFP counseling and LAM use, as well as a tries, such as Ethiopia and Pakistan, infection prevention materials separate register for recording PPIUD insertions. The Philippines ap- were also provided. In both the Philippines and India, the government proved a new version of a family planning form that includes PPIUD, has taken over the procurement of instruments at the state level. In but this form has yet to be widely disseminated. In Guinea, India, India, a country-approved manufacturer produces modified long pla- Pakistan, and the Philippines, facilities prepare monthly summary re- cental forceps. ports to share with relevant health authorities. Because only a subset All six programs also developed a pool of national trainers who could of clients typically return for a postnatal visit, during which a PPIUD support program expansion and sustainability. In Ethiopia, Guinea, can be checked, three programs—Guinea, India (selected facilities), Pakistan, and Rwanda, training of healthcare providers began with a and the Philippines—monitor use after discharge by having providers training event on general PPFP counseling, which targeted providers call women to ask about complications or expulsions. from the prenatal care, labor and delivery, and postnatal wards and In Guinea, India, Pakistan, and Rwanda, programs use a special covers all contraceptive options, including LAM. The PPFP counseling stamp to indicate that a woman has received PPFP counseling in event was followed by a course on PPIUD insertion (covering all types prenatal care and to specify the method, if any, that she has elected to of insertion) for labor and delivery staff. During the time between train- use postpartum. The stamp is placed on the client’s prenatal care card, ing events (an interval that ranged from two weeks to several months which she then brings with her to the facility at the time of birth. In across the programs), clients were counseled on PPFP, including PPIUDs, the Philippines, where facilities are developing their own means to and facilities increased their caseloads of PPIUD acceptors. In Pakistan track clients from prenatal care to delivery, one center is using the and the Philippines, general information on PPFP was incorporated woman’s birth plan to record PPFP counseling and her choice of method. into a PPIUD course for prenatal care and maternity clinicians. Where PPIUD services were still a novelty, failure to communicate 4. Results with facility staff about the introduction of the services sometimes resulted in instances of nontrained staff influencing clients to reverse Table 3 presents the proportion of eligible women in the six pro- their decision about PPIUD. As a result, whole-site orientations for all grams who were counseled on PPFP and accepted a method. The facility staff members, not just those who attended training, have been Pakistan and Philippines programs counseled more women at prenatal conducted in all programs. These orientations have even included care than at the time of birth. Ideally, all women delivering in facilities cleaning and support staff so all staff would be informed about the ser- would have been counseled at prenatal care and made their choice at vice integration and respond accurately to client questions about the that time, but tracking counseling offered in facilities other than that availability of PPIUD services. of the birth facility poses a challenge. Percentages of PPIUD acceptors The timing of the orientations varied across programs. In Guinea, for ranged from 2.3% of women counseled in Pakistan to 5.8% in the example, the orientation takes place immediately after training and be- Philippines. When accounting for all facility births, India had the highest fore the start-up of services; in Pakistan and the Philippines it is held percentage of acceptors, with 6.9% of women who delivered in Jhpiego- after training as part of onsite mentoring or supportive supervision assisted facilities electing to have a PPIUD. The India program was the visits by technical staff. In both Guinea and Ethiopia, participants in most extensive, reaching more than 99 000 women over the course of the training carry out this orientation as part of their post-training about three years (February 2010 to April 2013). Timing of PPIUD inser- action plan, which is subsequently monitored by program staff in tion varied among countries (Fig. 3), with Pakistan achieving the follow-up visits. highest proportion of postplacental IUD insertions (68%) and Rwanda the highest proportion of insertions during cesarean deliveries (43%). 3.5. Post-training and ongoing monitoring Ethiopia had the highest proportion of insertions during the immediate postpartum period (53%) and the lowest percentage during cesarean Ongoing supportive supervision visits and post-training follow-up deliveries (3%). Only three countries systematically track alternative visits to support the transfer of learning enable programs to reinforce methods of immediate postpartum contraception. LAM uptake is higher provider performance, promote quality of services, and troubleshoot than PPIUDs in Ethiopia (8% of facility births) and Guinea (29% of issues with service delivery or equipment. The visits enable newly births), whereas documented LAM uptake is roughly equivalent to trained providers to reorganize processes in a way that optimizes a con- PPIUD acceptance in Pakistan. sistent supply of instruments and contraceptives in the delivery rooms Rates of complications noted during six-week postpartum follow-up and initiate PPFP/PPIUD services with help from trainers. Post-training visits were collected from facility registers in five of the six country transfer-of-learning visits are conducted in all but the Philippines programs (Table 4). Although the proportion of women who returned program. In the Philippines, structured supportive supervision visits, to the hospital or health center for follow-up varied greatly, data on A. Pfitzer et al. / International Journal of Gynecology and Obstetrics 130 (2015) S54–S61 S59

Table 3 Postpartum family planning counseling and method acceptors by country.

Country No. of facility Number of women counseled on PPFP by Number of women accepting PPFP by method births timing

Prenatal Labor Postpartum Totala LAM PPTL PPIUD PPIUD as percentage PPIUD as percentage care (%) of women counseled (%) of facility births

Ethiopia 28 226 6164 4549 5676 16 389 2180 309 620 3.8 2.2 Guinea 61 814 ------20 699b 18 048 213 2438 11.8b 3.9 India 1 450 084 1 135 280 -- 632 600 1 767 880 -- -- 99 470 5.6 6.9 Pakistan 182 930 24 618 -- 9884 34 502 1062 1172 796 2.3 0.4 (estimated) Philippines -- 19 578 -- 14 322 33 900 -- -- 1957 5.8 -- Rwanda ------1147 -- -- Total 1 723 054 ------106 428 -- --

Abbreviations: LAM, lactational amenorrhea method; PPFP, postpartum family planning; PPIUD, postpartum intrauterine device; PPTL, postpartum tubal ligation. a It is likely that some women were double-counted in the total number of women counseled on PPFP since women may have been counseled at multiple visits. b For Guinea, the total number of women counseled is an estimate constructed as a sum of all LAM, PPTL, and PPIUD acceptors (all of whom received counseling). Women who were counseled but did not opt for one of these methods prior to discharge are thus not captured in this total, artificially inflating the percentage of PPIUD acceptors among women counseled on PPFP. returning PPIUD acceptors showed low rates of complications. The proportion of postplacental PPIUD insertions, the insertion timing that highest expulsion rates were 3.7% in Pakistan and 3.6% in Ethiopia, minimizes discomfort for the client. All countries could improve the while Guinea and the Philippines had the lowest rates at just 1.7%. coverage of prenatal counseling for PPFP, or the proportion of prenatal Rates of infection did not reach higher than the 1.3% recorded in care clients counseled, although it is difficult to determine individual Guinea, with two countries—Ethiopia and the Philippines—recording prenatal care coverage given that clients are expected to have repeat no cases of infection during follow-up visits. prenatal care visits. The country programs in India and the Philippines have sought 5. Discussion to saturate facilities with trained PPIUD providers, whereas the other programs have struggled with staff turnover. In India, this saturation The overarching lesson from these programs is the importance of has enabled widespread institutionalization of the service. In the engaging government leadership and building local ownership for Philippines, capacitating providers as trainers has helped to ensure PPFP/PPIUD programs. As the Pakistan program found, building owner- that more providers, including residents, are involved in service ship at multiple levels of the healthcare system and sustaining that delivery, and that the program is prepared to expand from the 10 ownership at the district level can be challenging. Providers’—including centers of excellence outward, including reaching out to private sector specialists’—myths and misconceptions about IUDs, and specifically midwives. Post-training support to facilities, either through transfer- PPIUDs, must be addressed as part of stakeholder engagement and the of-learning follow-up visits, supportive supervision visits, or other establishment of services at each new facility. In Rwanda and Guinea, strategies for assistance has been crucial to address quality gaps in the programs have identified champion providers who are instrumental in medium term. ensuring success. Ensuring that all staff in a given facility are aware of and understand Supplying facilities with sufficient instruments for service delivery the messages around the introduction of PPIUDs has been essential, par- and ensuring that they have a regular source of commodities, positioned ticularly during expansion to new sites. This review did not detect any in maternity, is a basic minimum for continued service delivery. Inte- implications of the variation in timing in programs’ whole-site orienta- gration of counseling at every possible contact with a client, whether tions, but failing to hold such orientations appears to contribute to low prenatal, at birth, or in the immediate postpartum period, means uptake of the service. integration at multiple points and by multiple providers to ensure Record-keeping systems for monitoring PPIUD services help show consistent access. At the same time, emphasizing universal prenatal where improvement is needed. Transparent sharing of monitoring counseling for PPFP that includes PPIUDs is key to increasing the results allows for continued advocacy for expansion to government

Rwanda 27 43 30

Guinea 37 18 45

Postplacental India 40 32 28 (10 minutes after delivery of placenta)

Cesarean delivery Ethiopia 44 3 53 (during cesarean section)

Philippines 55 21 23 Postpartum (within 48 hours after delivery)

Pakistan 68 17 15

0% 100%

Fig. 3. Percentage of postpartum intrauterine device insertions by timing and country. S60 A. Pﬁtzer et al. / International Journal of Gynecology and Obstetrics 130 (2015) S54–S61

Table 4 Postpartum intrauterine device follow-up and complications by country.a

Country PPIUD acceptors who PPIUD acceptors who Complications returned for follow-up returned for follow-up Expulsions Infections Removals with a complication

Ethiopia 329 (53.1) 19 (5.8) 12 (3.6) 0 (0.0) 7 (2.1) Guinea 2008 (82.4) 74 (3.7) 35 (1.7) 27 (1.3) 12 (0.6) India 37 252 (37.5) 2937 (7.9) 989 (2.7) 361 (1.0) 1587 (4.3) Pakistan 269 (33.8) 41 (15.2) 10 (3.7) 2 (0.7) 29 (10.8) Philippines 641 (32.8) 15 (2.3) 11 (1.7) 0 (0.0) 4 (0.6) Total 40 499 (38.1) 3086 (7.6) 1057 (2.6) 390 (1.0) 1639 (4.0)

Abbreviations: PPIUD, postpartum intrauterine device. a Values are given as number (percentage). stakeholders and better informs those who expect to see high expulsion most low-resource contexts, midwives and obstetric nurses are the rates or other adverse consequences. Keeping records of counseling cadres with the most contact with women during birth. To succeed, from one contact to the next in the continuum of care has proven chal- PPFP programs require effective counseling in the prenatal period, avail- lenging, as has capturing the adoption of methods other than the PPIUD ability of trained and competent staff at the time of birth, and adequate (e.g. LAM or condom use). Innovations are still needed to perfect those follow-up of trained providers. IUDs, including PPIUDs, are among the processes, perhaps with the use of electronic medical records, and more safest and most effective methods of contraception. They offer the addi- experience is needed to inform how HMIS can and should track PPFP tional advantage of being “forgettable” contraception, in the sense that uptake. Nevertheless, ongoing monitoring of program data offers man- little or no follow-up is required, other than for removal of the device. agers and policymakers essential information for decision-making. But ensuring widespread access to this convenient and cost-effective There is a dearth of detail on PPFP/PPIUD implementation in the method requires greater acceptance and uptake among maternal published program literature. The findings and results presented here healthcare teams in clinical settings and the larger maternal health are all from introductory phases of programming, except in India, community. Data and evidence on acceptability, feasibility, and compli- where the program is in an expansion phase and has now reached 20 cation rates of PPIUDs can play a part in ongoing advocacy for expanding states and hundreds of facilities. We strongly believe that the focus on access to this underutilized PPFP method. PPIUD insertion technique with IUD placement high in the fundus is what has enabled these programs to demonstrate low expulsion rates Acknowledgments that are equivalent to those from interval IUD insertion [26]. Further- more, attention to quality and counseling, along with carefully planned The authors thank Zhuzhi Moore of MEASURE/DHS for her analysis and coordinated program implementation, has brought promising re- of the DHS data presented in Table 1. We also acknowledge Elizabeth sults in terms of acceptance of PPIUDs. The absence of reports of uterine Sasser for her help preparing and compiling the other tables, and Koki perforation in these data is consistent with a global review by Kapp et al. Agarwal for her careful review of the manuscript. Lastly, we are grateful [27]. While we did not report on client satisfaction, the India program to program staff from all six countries who contributed to this article has published results from a special study of clients [28]. and to the programs described herein. There are limitations to our findings, as is expected with data collected in program settings rather than rigorously controlled research envi- Conflict of interest ronments. We report only on continuation and complications seen at the six-week postnatal visit, but additional expulsions and removals The authors have no conflicts of interest to declare. likely will occur throughout the first year postpartum. We could not analyze client data through multiple contacts, so we do not know how References many women who opted for a PPIUD during prenatal care did not actually obtain the method, either because they could not reach the [1] Singh S, Darroch JE. Adding it up: costs and benefits of contraceptive services. Esti- facility for the birth, the service was not available in the immediate mates for 2012. New York: Guttmacher Institute, United Nations Population Fund; postpartum period, or they changed their mind after further discussion 2012. http://www.guttmacher.org/pubs/AIU-2012-estimates.pdf. Accessed April with family or peers. These data do not include any observation of the 27, 2014. [2] Williamson N. Motherhood in childhood: facing the challenge of adolescent quality of counseling. Nevertheless, with vast numbers of women ex- pregnancy. State of world population 2013. New York: United Nations Population pressing unmet need for both birth limiting and birth spacing, particu- Fund; 2013. http://www.unfpa.org/webdav/site/global/shared/swp2013/EN- larly in the year following childbirth, the robust uptake among women SWOP2013-final.pdf. Accessed April 27, 2014. [3] Cleland J, Conde-Agudelo A, Peterson H, Ross J, Tsui A. Contraception and Health. counseled in six diverse low- and middle-income countries shows the Lancet 2012;380(9837):149–56. viability of making PPIUDs more widely available to prevent unintended [4] Kozuki N, Lee A, Silveira MF, Victora CG, Adair L, Humphrey J, et al. The associations pregnancies in low-resource settings. of birth intervals with small-for gestational-age, preterm, and neonatal and infant mortality: a meta-analysis. BMC Public Health 2013;13(Suppl. 3):S3. [5] Kozuki N, Walker N. Exploring the association between short/long preceding birth 6. Conclusion and recommendations intervals and child mortality: using reference birth interval children of the same mother as comparison. BMC Public Health 2013;13(Suppl. 3):S6. Women around the world are at risk of unintended pregnancy in the [6] The Partnership for Maternal, Newborn and Child Health. Opportunities for Africa's newborns: practical data, policy and programmatic support for newborn care in postpartum period and need improved access to effective contraception Africa. http://www.who.int/pmnch/media/publications/africanewborns/en. Pub- methods before they resume sexual activity and their fertility returns. lished 2006. Accessed April 27, 2014. Equipping and motivating maternal health providers to offer PPFP/ [7] Singh A, Padmadas SS, Mishra US, Pallikadavath S, Johnson FA, Matthews Z. Socio- economic inequalities in the use of postnatal care in India. PLoS One 2012;7(5): PPIUD counseling in prenatal care and at the time of birth improves ac- e37037. cess to contraception. For integration of PPFP/PPIUD services to be effec- [8] Regassa N. Antenatal and postnatal care service utilization in southern Ethiopia: a tive, the target of training and mentoring must be different than it is for population-based study. Afr Health Sci 2011;13(3):390–7. [9] World Health Organization. Programming strategies for postpartum family plan- traditional family planning programs: the focus must be on prenatal ning. Geneva: WHO; 2013. http://www.mchip.net/sites/default/files/Postpartum- care and maternity staff, those assisting women at the time of birth. In family-planning.pdf. Accessed April 24, 2014. A. Pfitzer et al. / International Journal of Gynecology and Obstetrics 130 (2015) S54–S61 S61

[10] Pariani S, Heer DM, Van Arsdol Jr MD. Does choice make a difference to contracep- [20] Jhpiego. Postpartum intrauterine contraceptive device (PPIUD) services: a reference tive use? Evidence from East Java. Stud Fam Plann 1991;22(6):384–90. manual for providers. Baltimore, MD: Jhpiego Corporation; 2010. http://reprolineplus. [11] World Health Organization. Medical eligibility criteria for contraceptive use. 4th ed. org/resources/postpartum-intrauterine-contraceptive-device-ppiud-services-learning- Geneva: WHO; 2010. http://whqlibdoc.who.int/publications/2010/9789241563888_ resource-package. Accessed November 11, 2014. eng.pdf. Accessed April 27, 2014. [21] Araujo VB, Ortiz L, Smith J. Postpartum IUD in Paraguay. A case series of 3000 cases. [12] World Health Organization. Guidelines on maternal, newborn, child and adoles- Contraception 2012;86(2):173–4. cent health: recommendations on maternal and perinatal health. Geneva: WHO; [22] Cordero CF, Barone MA, Calderon V. A postpartum IUD program in the Dominican 2013. http://www.who.int/maternal_child_adolescent/documents/guidelines- Republic. Int J Gynecol Obstet 1996;55(2):181–2. recommendations-maternal-health.pdf. Accessed April 27, 2014. [23] Ministry of Health and Family Welfare. NHM Health Management Information Sys- [13] Peterson AE, Perez-Escamilla R, Labbok MH, Hight V, von Hertzen H, Van Look P. tem (HMIS). https://nrhm-mis.nic.in. Accessed April 27, 2014. Multicenter study of the lactational amenorrhea method (LAM) III: effectiveness, [24] Department of Health. Postpartum family planning: Supplement to the Philippines duration, and satisfaction with reduced client–provider contact. Contraception clinical standards manual on family planning. Manila, Philippines: Department of 2000;62(5):221–30. Health; 2013. [14] Chi IC, Farr G. Postpartum IUD contraception: a review of international experience. [25] Charurat E, Ayuyo CM, Muthoni J, Kamunya R, Archer L, Koskei N, et al. An assess- Adv Contracept 1989;5(3):127–46. ment of postpartum intrauterine contraceptive device services in Embu, Kenya. [15] Celen S, Möröy P, Sucak A, Aktulay A, Danişman N. Clinical outcomes of early https://www.k4health.org/sites/default/files/Kenya%20PPIUCD%20Assessment% postplacental insertion of intrauterine contraceptive devices. Contraception 2004; 20Report%204Feb11_Final.pdf. Published 2011. Accessed May 2, 2014. 69(4):279–82. [26] Whiteman MK, Tyler CP, Folger SG, Gaffield ME, Curtis KM. When can a woman have [16] Eroglu K, Akkuzu G, Vural G, Dilbaz B, Akin A, Taskin L, et al. Comparison of efficacy an intrauterine device inserted? A systematic review. Contraception 2013;87(5): and complications of IUD insertion in immediate postplacental/early postpartum pe- 666–73. riod with interval period: 1 year follow-up. Contraception 2006;74(5):376–81. [27] Kapp N, Curtis KM. Intrauterine device insertion during the postpartum period: a [17] Grimes DA, Lopez LM, Schulz KF, Van Vliet HA, Stanwood NL. Immediate post- systematic review. Contraception 2009;80(4):237-36. partum insertion of intrauterine devices. Cochrane Database Syst Rev 2010;5: [28] Kumar S, Sethi R, Balasubramaniam S, Charurat E, Lalchandani K, Semba R, et al. CD003036. Women’s experience with postpartum intrauterine contraceptive device use in [18] Long-term reversible contraception: Twelve years of experience with the TCu380A India. Reprod Health 2014;11:32. and TCu220C. Contraception 1997;56:341–52. [19] USAID. Measure DHS statcompiler. http://dhsprogram.com/data/STATcompiler.cfm. Accessed April 27, 2014. International Journal of Gynecology and Obstetrics 130 (2015) S62–S67

Contents lists available at ScienceDirect

International Journal of Gynecology and Obstetrics

journal homepage: www.elsevier.com/locate/ijgo

SUPPLEMENT ARTICLE Evolution of malaria in pregnancy control: Jhpiego’s 10-year contribution

Elaine Roman a,⁎, Augustine Ngindu b,BrightOrjic, Jérémie Zoungrana d, Sarah Robbins a, William Brieger a,e a Jhpiego, Baltimore, MD, USA b Jhpiego, Nairobi, Kenya c Jhpiego, Abuja, Nigeria d Jhpiego, Kigali, Rwanda e Johns Hopkins Bloomberg School of Public Health, Baltimore, MD, USA article info abstract

Keywords: Malaria continues to be a life-threatening illness throughout Sub-Saharan Africa, with pregnant women and Focused antenatal care children being particularly vulnerable and an estimated 10 000 women and 200 000 newborns dying each Intermittent preventive treatment year as a result of malaria in pregnancy (MIP). Since 2004, WHO has supported a three-pronged MIP approach: Malaria control programs (1) intermittent preventive treatment with sulfadoxine-pyrimethamine; (2) use of insecticide-treated bed nets; Malaria in pregnancy and (3) effective case management. The present article identiﬁes benchmarks in Jhpiego’s 10-plus years of MIP Sub-Saharan Africa experience at the regional and national levels that have contributed to its global MIP leadership and aligned programs and policies with global approaches toward malaria elimination. As countries continue to develop and expand MIP programming, support will continue to be essential in the following eight MIP program areas: integration, policy, capacity development, community engagement, quality assurance, commodities, monitoring and evaluation, and ﬁnancing. © 2015 Published by Elsevier Ireland Ltd. on behalf of International Federation of Gynecology and Obstetrics. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/3.0/).

1. Introduction In 2000, in Abuja, Nigeria, the Roll Back Malaria (RBM) Partnership set a target coverage rate for these three interventions of 80% by 2010 The 2013 World Malaria Report estimated that malaria led to [6]. Since then, the RBM partnership has set targets for IPTp uptake 207 million cases of malaria illness and an estimated 627 000 deaths and universal coverage for LLINs at 100% [7]. The US President’sMalaria in 2012 worldwide [1]. Ninety percent of all malaria deaths occur in Initiative (PMI) targets are 85% for both IPTp and ITN coverage for those Sub-Saharan Africa due to infections caused by the Plasmodium countries with PMI support [8]. While 39 countries in Sub-Saharan falciparum parasite. Pregnant women and young children are among Africa have MIP policies in place, most countries are far from achieving the most vulnerable populations. Each year, approximately 25 million target coverage goals for these interventions [9]. African women become pregnant in malaria-endemic areas of Africa For more than a decade, Jhpiego has been a committed partner to with intense transmission of P. falciparum [2]. An estimated 10 000 of the RBM partnership, which includes governments of endemic coun- these women and 200 000 of their newborns die as a result of malaria tries, nongovernmental organizations, donor organizations, and in pregnancy (MIP) [3,4]. corporate members. Jhpiego has been involved in advancing the In areas of stable malaria transmission, WHO supports a three- MIP global dialogue through the RBM MIP Working Group as an pronged approach to addressing MIP, ideally delivered through antena- active technical participant since the group’s inception and has tal care: (1) intermittent preventive treatment (IPTp), with sulfadoxine- served as co-chair for seven years. In addition, Jhpiego has provided pyrimethamine (SP) (IPTp-SP); (2) use of insecticide-treated bed nets technical assistance to more than 20 countries to help accelerate (ITNs), speciﬁcally long-lasting insecticide-treated nets (LLINs) and malaria prevention and control. The present article presents an (3) effective case management of malaria illness [5].WHO’sthree- historical overview of Jhpiego’s MIP programs starting around pronged MIP approach, ofﬁcially made policy in 2004, marked a pivotal 2002, reviewing the methods by which Jhpiego has achieved leader- change in MIP programming from the use of weekly chemoprophylaxis. ship in MIP control. It presents as benchmarks the seminal contributions of Jhpiego in three countries (Burkina Faso, Kenya, and Nigeria) as well as regional MIP efforts in Sub-Saharan Africa. We examine gaps in implementation and key actions required to ensure ⁎ Corresponding author at: Jhpiego, 1615 Thames Street, Baltimore, MD 21231, USA. Tel.: +1 303 482 2852. scale-up of MIP interventions across the region to guide other E-mail address: [email protected] (E. Roman). implementing organizations. http://dx.doi.org/10.1016/j.ijgo.2015.03.009 0020-7292/© 2015 Published by Elsevier Ireland Ltd. on behalf of International Federation of Gynecology and Obstetrics. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/3.0/). E. Roman et al. / International Journal of Gynecology and Obstetrics 130 (2015) S62–S67 S63

2. Methods intervention (CDI) approach to increase MIP service coverage and overcome service delivery bottlenecks (Nigeria). Jhpiego’s MIP programming is represented in the present paper Countries and programs were chosen for further analysis based on using a case study approach documenting Jhpiego’s contribution to their contribution to MIP programming both at the country level and major benchmarks in MIP program evolution. Benchmarks were regionally. An in-depth review of publications related to target countries defined as either technical or managerial innovations that moved MIP in Sub-Saharan Africa, including MIP country program reports, ministry efforts to the next level of comprehensiveness and expanded coverage of health (MOH) reports, strategy documents, and peer reviewed arti- of MIP interventions. A start date of around 2002 was chosen because cles, was performed to determine what countries did to implement that was when both the MIP working group and the US Agency for Inter- MIP programs, what the results were, and what implementation national Development’s (USAID) Malaria Action Coalition, comprised of challenges were faced. USAID’s ACCESS and RPM Plus programs, the US Centers for Disease Control and Prevention (CDC), and the WHO/AFRO, began in earnest. 3. Findings Jhpiego reviewed program activity reports and other publications over a 10-year period to identify and subsequently describe major MIP The five benchmark contributions of Jhpiego are shown in Table 1. programming benchmarks. These benchmarks included the following: (1) testing the efficacy of IPTp-SP on a platform of antenatal care 3.1. Benchmark 1 (technical): Testing the efficacy of IPTp-SP (Burkina Faso) (Burkina Faso); (2) integration of focused antenatal care services (FANC), in which providers focus on assessment and actions needed In 2001, the Burkina Faso MOH, in collaboration with Jhpiego and to make decisions and provide care for each woman’s individual situa- the CDC, designed and implemented one of the first pilot programs in tion, with MIP nationwide (Kenya); (3) support to regional networks West Africa testing FANC as a platform for malaria in pregnancy inter- to accelerate MIP implementation (regional); (4) documentation of ventions [10]. A total of 23 health facilities were included in the pilot MIP program implementation processes to learn best practices and study, and eight sites were selected for the baseline and follow-up bottlenecks (regional); and (5) adaptation of the community-directed assessment. A total of 2014 women were enrolled in the assessment.

Table 1 Benchmark contributions of Jhpeigo to malaria in pregnancy control.

Benchmark 1 Benchmark 2 Benchmark 3 Benchmark 4 Benchmark 5 Testing the efﬁcacy of IPTp-SP Integration of FANC services Accelerate MIP implementation Documentation of MIP Community-directed (Burkina Faso) with MIP nationwide (Kenya) (East, Southern and West Africa) implementation interventions practices (Regional) (Nigeria)

Type Technical Technical Managerial Technical Technical Issue(s) Bottleneck of IPTp-SP, health Integration of Kenya’s MOH MIP Accelerate MIP implementation Low national coverage Low ANC utilization, poor workers not following uptake policy with WHO’s FANC model. in East, Southern and West of IPTp and ITNs in community attitude toward procedures, and health facilities Africa. Sub-Saharan Africa. ANC, and low IPTp and ITN charging for SP treatments. coverage in Nigeria. Solution(s) Between 2001 and 2004, eight • Between 2002 and 2004, • In East and Southern Africa, Documented best Between 2007 and 2011, used sites adopted comprehensive, developed a two-phase five countries developed practices and remaining two-pronged approach to system-wide approach fostering service package called MIPESA coalition between challenges with MIP increase MIP service coverage in partnerships between “Focused Antenatal Care and 2002 and 2005 focusing on programs in Zambia, Akwa Ibom State to increase reproductive health and malaria Malaria in Pregnancy” that best practices learned; Malawi, and Senegal access to MIP services through control programs; trained and was expanded to 19 malaria fostering relationships be- community-directed supervised district health staff endemic districts. tween national reproductive interventions; trained more on quality improvements; • Fostered partnership between health and malaria control than 600 frontline health community mobilization; and the MOH’s Division of Repro- programs; and increasing ca- facilities and 800 improved record keeping. ductive Health and the pacity to support MIP pro- community-directed Division of Malaria Control. grams at all levels. distributors on basic FANC, • Trained frontline healthcare • West Africa established record keeping and data use for providers on FANC. RAOPAG expanding to 10 treatment; provided basic • Provided supportive supervi- countries and international supplies to all clinics; and sion reinforcing knowledge, partners. updated ANC registers reflective skills, and addressing gaps in • Between 2003 and 2005, of MIP indicators. service delivery facilitating an information • Delivered community sensiti- exchange between member zation on ANC changes to countries on advocacy, create awareness research, MIP prevention and treatment. • Supporting regional expertise through planning and documentation. Results • Attendance of four or more • One intervention and one Repositioned MIP from Analyzed and developed • Five-times increase in inter- ANC visits increased from 21% control district. innovation to a major ANC eight key MIP program vention arm and three-times to 44%. • IPTp1 and IPTp2 increased program component; areas essential for increase in control arm with • Receipt of two doses of IPTp from 20.3% to 61.7% (inter- demonstrated importance of health systems when IPTp uptake. increased from 0% to 75%. vention) and 9.3% to 28.3% peer influence. addressing MIP • ANC attendance increased • Peripheral parasitemia (control). morbidity and mortality using community volunteer decreased from 22% to 15%. • Attendance at four ANC visits rates. referrals. • Increased ownership of ITN 17.0% (intervention) com- • Elderly community-directed from 22% to 46%. pared with 6.5% (control). distributors helped mobilize younger women to clinics.

Abbreviations: IPTp-SP, intermittent preventive treatment with sulfadoxine-pyrimethamine; MIP, malaria in pregnancy; ITN, insecticide-treated bed nets; SP, sulfadoxine-pyrimethamine; MOH, ministry of health; FANC, focused antenatal care; ANC, antenatal care; MIPESA, Malaria in Pregnancy East and Southern Africa; RAOPAG, Réseau d’Afriquedel’OuestcontrelePaludisme pendant la Grossesse [Roll Back Malaria’s West African Network for the Prevention and Treatment of Malaria in Pregnancy]. S64 E. Roman et al. / International Journal of Gynecology and Obstetrics 130 (2015) S62–S67

The initial assessment revealed practical bottlenecks in the provision could be done in four pilot districts with simple, practical education of IPTp-SP that appeared to limit efficacy, namely: health workers were and orientation for health workers and communities—the expansion not following directly observed therapy to ensure IPTp uptake and were phase reached all 19 malaria-endemic districts nationwide [12]. providing IPTp-SP on an empty stomach; and health facilities were Key elements of the initiative were: (1) fostering partnerships charging women for SP despite an MOH instruction to deliver it free of between the MOH’s Division of Reproductive Health (DRH) and the charge. As a result, the pilot intervention adopted a comprehensive, Division of Malaria Control (DMOC), where the DRH led implementation system-wide approach, including fostering partnerships between and the DMOC provided technical oversight; (2) educating frontline reproductive health and malaria control programs; education and su- healthcare providers on FANC emphasizing comprehensive care, includ- pervision of district health staff; quality improvement at facilities; com- ing IPTp uptake and promotion of ITN use; (3) supportive supervision munity mobilization; and improved recordkeeping [11]. to reinforce knowledge and skills learned in training courses and Between baseline (2001) and follow-up (2004), results of the pilot address gaps in service delivery through coaching and mentoring; and study revealed statistically significant improvements in health out- (4) community sensitization to create awareness about the new changes comes. Women attending four or more ANC visits increased from 21% in provision of ANC services (e.g. pregnant women were only expected to 44% (P = 0.01); women receiving two doses of IPTp increased from to visit the clinics for four comprehensive visits). 0% to 75% (P = 0.02); and peripheral parasitemia decreased significant- The CDC evaluated the Jhpiego/MOH FANC activities by conducting ly from 22% to 15% (P b 0.0001) (Fig. 1). Other notable outcomes includ- pre- and post-intervention evaluations in one of the intervention dis- ed owning an ITN, which increased from 22% to 46% [12]. ITN ownership tricts (Asembo) and one of the control districts (Gem), and it included might have increased further if ITN stock-outs had not persisted at ANC a cross-sectional household survey of women who recently delivered sites. at baseline (2002) and follow-up (2005) [13]. Based in part on the results of this pilot, Burkina Faso adopted an up- Results in Fig. 2 illustrate that uptake of IPTp1 (phase 1) and IPTp2 dated MIP policy in 2006 that reflected WHO’s three-pronged approach. (phase 2) in the intervention district increased threefold, from 20.3% From this work, the MOH, CDC, and Jhpiego gathered evidence not only to 61.7% and 9.3% to 28.3%, respectively (P b 0.05), suggesting that edu- about the efficacy of IPTp, but also about the system-wide strategies cation combined with supportive supervision and community sensitiza- needed to manage IPTp delivery through ANC. tion contributed to increased uptake of IPTp among pregnant women. In addition, women from the intervention district were more likely than 3.2. Benchmark 2 (technical): Integration of FANC with MIP nationwide women from the control district to attend four ANC visits (17.0% versus (Kenya) 6.5%) and were significantly more likely to state that SP is helpful and safe during pregnancy [13]. In 1998, Kenya was one of the early adopters of WHO’sthree- pronged approach. In 2003, Kenya operationalized its MIP policy and 3.3. Benchmark 3 (managerial): Support to regional networks to accelerate adopted the platform of FANC to deliver IPTp and promote ITNs. The MIP implementation (regional) delivery of MIP preventive care through ANC services was pivotal, especially at that time, since Kenya was one of the first countries in Africa to In an effort to augment country-level implementation of MIP on a adopt this integrated approach. platform of FANC, Jhpiego worked closely with countries in East, With support from the UK Department for International Develop- Southern, and West Africa to support the development of regional ment (DFID), Jhpiego played a facilitative role of integrating the bodies that could increase momentum, advocacy, and support for MIP MOH’s inputs with WHO’s original FANC model, resulting in a service programming. package called “Focused Antenatal Care and Malaria in Pregnancy.” The In East and Southern Africa, five countries (Kenya, Malawi, Tanzania, program was launched in two phases (demonstration and expansion) Uganda, and Zambia) were early adopters of WHO’s three-pronged MIP between 2002 and 2004. The demonstration phase showed what approach. In 2002, representatives from these countries came together

Fig. 1. Antenatal care and malaria in pregnancy-related outcomes in Burkina Faso: baseline (2001) and follow-up (2004). Abbreviations: ANC, antenatal care; IPTp2, intermittent preventive treatment phase 2; ITN, insecticide-treated bed net; LBW, low birth weight. Reprinted with permission from Jhpiego [12]. E. Roman et al. / International Journal of Gynecology and Obstetrics 130 (2015) S62–S67 S65

Fig. 2. Intermittent preventive treatment (IPTp phase 1 and 2) coverage among recent mothers who attended antenatal care in intervention and control districts in Kenya: baseline (2002) and follow-up (2005). Adapted with permission from Ouma et al. [13]. with international partners to form the Malaria in Pregnancy East and These two regional networks helped reposition MIP programming Southern Africa (MIPESA) coalition. Jhpiego played a key role as a from a mere innovation to a major program component of ANC. MIPESA partner [14]. Between 2002 and 2005, MIPESA transferred best They also demonstrated the importance of peer influence inherent in practices and lessons learned across countries, fostered partnerships South-to-South dialogue and planning. between national reproductive health and malaria control programs, and increased regional capacity to support MIP programming at the 3.4. Benchmark 4 (technical): Documentation of MIP program implemen- country level [15]. In particular, MIPESA contributed to: (1) harmonized tation practices (regional) and appropriately integrated national policies, strategies, and education materials in Zambia and Kenya; (2) adaptation of education materials While MIP program experiences from Burkina Faso and Kenya as from Kenya to other countries (i.e. adapting training for FANC as a well as regional efforts through MIPESA and RAOPAG contributed to platform to deliver MIP services in Tanzania, Uganda, and Malawi); acceleration of MIP programming across Sub-Saharan Africa, national and (3) increased commitments from national reproductive health and coverage of IPTp uptake and ITNs remains low. Based on the most recent malaria control programs across all countries to work together to Demographic and Health Surveys, Malaria Indicator Surveys, and Malaria address MIP programming. Indicator Cluster Surveys, IPTp uptake (two doses) across PMI countries Similarly, in March 2002, directors of reproductive health and malaria averaged 33%, ranging from 1% (Benin) to 70% (Zambia), and LLIN use control programs in West African countries came together to consider among pregnant women averaged 43%, ranging from 10% (Zimbabwe) the way forward for addressing MIP in their region. Among their recom- to 72% (Rwanda) [17]. To better understand the features of MIP mendations was the creation of the Roll Back Malaria’s West African programs that work well and those that need improvement, Jhpiego, Network for the Prevention and Treatment of Malaria in Pregnancy with support from PMI, documented best practices and remaining (Réseau d’Afrique de l’Ouest contre le Paludisme pendant la Grossesse, challenges in MIP programming in three relatively successful countries: or RAOPAG), which expanded to 10 countries between 2003 and 2005 Zambia, Malawi, and Senegal. The results of this program assessment (Benin, Burkina Faso, Côte d’Ivoire, Guinea, Mali, Mauritania, Niger, have been published elsewhere and are summarized in Box 1 [18]. Nigeria, Senegal, and Togo) and multiple international partners [16]. RAOPAG’s mission was to facilitate information exchange between member countries with respect to advocacy, research, and implementa- 3.5. Benchmark 5 (technical): Community-directed interventions tion of MIP prevention and treatment interventions, and to support regional expertise through planning and documentation. RAOPAG was Starting in 2006, Jhpiego received funding from Exxon Mobil to the first and only network in West Africa to bring together reproductive address one of the more visible and challenging MIP program areas in health and malaria control experts regionally to share knowledge and Nigeria: community involvement. Of primary concern was low ANC lessons learned in MIP prevention and control [15,16]. utilization, poor community attitudes toward ANC because of low Other RAOPAG contributions included: (1) fostering the creation of perceived service quality, and very low coverage of IPTp and ITNs country-specific action plans for policy change and acting as a catalyst among pregnant women. for IPTp-SP adoption in member countries; (2) creating a database of From 2007 to 2011, Jhpiego organized a community-clinic partner- online MIP resources, including country-specific policies, guidelines, ship, illustrated in Fig. 3, which used a two-pronged approach to and action plans by country focal persons; (3) facilitating the adaptation increase coverage of MIP services: (1) improving the quality of FANC of FANC/MIP clinical education materials at the regional level and services; and (2) increasing access to MIP services through CDIs [19]. organizing education of trainers at the country level; and (4) assisting CDIs enable communities to lead health interventions in collaboration member countries in developing advocacy plans to solicit support with health facilities [20]. Health facility staff train, supervise, and pro- for their MIP action plans and marketing MIP strategies to potential vide communities with needed commodities. Communities direct donors. their own health promotion activities, maintain registers, decide when S66 E. Roman et al. / International Journal of Gynecology and Obstetrics 130 (2015) S62–S67

Box 1 attendance and develop action plans to solve them. More than 300 Malaria in pregnancy program areas. FLHFs in both arms of the study were provided with basic education on FANC including MIP, record-keeping, and use of data for decision- Integration: Integrated MIP services include collaboration between making purposes. Basic supplies that included SP were also provided national reproductive health and malaria control programs and other to all clinics since nothing was available at baseline. ANC registers comprehensive maternal and child health services programs to were updated to reflect MIP indicators. ensure harmonized policies, guidelines, and education materials, FLHFs in the intervention arm mobilized their catchment communi- and a coordinated effective program implementation. ties for CDI and encouraged the communities to select the volunteer Policy: MIP policies are based on the latest scientific evidence, CDDs, conduct a community census, and collect MIP supplies on a regu- defined country goals, and national guidelines that are harmonized lar basis. CDDs provided a first contact with MIP services, including and effectively integrated across different health sectors. distributing SP and ITNs, maintaining records and submitting monthly Commodities: Commodities include correct medicines and medical tally sheets, giving health education, and referring pregnant women to products, systems that ensure commodity availability that include ANC, thereby strengthening community–clinic links. By using CDDs, SP, LLINs, diagnostic tools, rapid diagnostic tests and/or microsco- the FLHFs were able to extend the reach of their ANC services. py, medicines to provide effective MIP case management. In addition to technical supervision by the FLHFs, CDDs were held Quality assurance: Quality assurance is a concept that covers accountable by their communities. Communities had selected CDDs individually or collectively efficacy, safety, appropriateness, and whom they trusted, who were able to read and write, who were resi- acceptability of MIP services measuring quality of care, supervision dent in the community, and (perhaps most importantly) who were will- support, and self-assessment by healthcare providers to monitor ing to volunteer their services. Two CDDs were even fined and replaced delivery of services. by their communities for diverting MIP commodities, demonstrating Capacity Building: MIP capacity building is strengthening human their local accountability. resources by improving knowledge and skills that includes compe- Comparing baseline and endline surveys showed a statistically tency-based pre- and in-service education based on WHO guide- significant (P b 0.001) five-times increase in coverage of IPTp in the lines and national policies. CDI communities as compared with a three-times increase in the con- Community Awareness and Involvement: MIP community trol arm where IPTp was available only at ANC. ANC attendance was engagement involves the promotion and/or delivery of health also enhanced with community volunteers making referrals, showing services including raising awareness of MIP prevention and control; that the community–clinic partnership was working; elderly CDDs promotion of IPTp and ITNs; and effective case management. were particularly helpful in mobilizing younger women to attend clinics Monitoring and Evaluation: Monitoring and evaluation captures and utilize available services [21]. MIP service delivery indicators and provide data to the national health management information system to be used for decision 4. The way forward making at all points of care and national policy formulation. Financing: MIP financing is a combination of national government This analysis of Jhpiego’s role in global MIP control highlights key and donor funding that guarantee that programs receive the steps and important achievements in MIP programming that helped ac- necessary support and resources to reach all pregnant women. celerate program implementation throughout Sub-Saharan Africa. From piloting MIP on a platform of FANC to expansion nationally and adapting Abbreviations: MIP, malaria in pregnancy; SP, sulfadoxine-pyrimethamine; LLIN, ascientifically proven intervention strategy of CDI, as well as supporting long-lasting insecticide-treated nets. the role of regional networks to further accelerate implementation, Jhpiego’s commitment and technical expertise has contributed essential program learning about MIP interventions, both in terms of successful and where to distribute commodities (ITNs and IPTp), and promote practices and remaining challenges. This learning has subsequently service utilization (referral to comprehensive ANC services). been shared with and scaled up by RBM partner organizations to The two-pronged intervention was tested in three Local Govern- improve outcomes for women and their newborns. ment Areas (LGAs) of Akwa Ibom State, Nigeria, using another three Despite these programmatic advances, much work remains to ad- LGAs as control areas. Overall, 25 health facilities and catchment com- here to WHO guidelines and reduce MIP prevalence. Very few countries munities containing more than 820 000 people were involved. In the have reached either the targets set at the Abuja meeting in 2000 or their three intervention LGAs, frontline health facilities (FLHFs) were trained own policy ambition, and countries are even further away from the to mobilize their communities to select their own community-directed more recent RBM targets set for 2010 [4,7]. The new WHO IPTp recom- distributors (CDDs). After a five-day education session, the 800 selected mendation, disseminated in October 2012, which promotes IPTp at CDDs were supervised by the FLHFs and monitored using Jhpiego’s every scheduled ANC visit beginning in the second trimester of preg- quality improvement approach, Standards-Based Management and nancy [22], presents an opportunity for countries to review and update Recognition (SBM-R), to detect performance gaps that deterred ANC their national policies and guidelines as well as explore new MIP program possibilities—making strategic decisions to accelerate scale-up and further improve coverage for pregnant women. As countries continue to expand MIP programming, comprehensive support for the eight MIP program areas will be essential. These program areas are the core components of any health system; if one is weak, it impacts the others. This underscores the importance of addressing MIP care comprehensively with attention to health system strengthening to achieve long-lasting results. Jhpiego’s approach has been to address malaria prevention and control comprehensively across the health “continuum of care,” from the community to the health facility to the national level on a platform of maternal, child, and newborn care, addressing each of the MIP program areas. Fig. 3. Nigeria malaria in pregnancy (MIP) community–clinic partnership intervention. Underpinning the acceleration of MIP programming in each of the Reprinted with permission from Brieger et al. [19]. country and regional case studies was the strong partnership between E. Roman et al. / International Journal of Gynecology and Obstetrics 130 (2015) S62–S67 S67 the national reproductive health program and the national malaria con- References trol program, as well as the national HIV program. Both Burkina Faso [1] World Health Organization. World Malaria Report 2013. http://www.who.int/ and Kenya faced multiple bottlenecks as MIP programs were introduced malaria/publications/world_malaria_report_2013/report/en/. Published 2013. and scaled up respectively. For example, addressing issues of commod- Accessed April 25, 2014. ities and supplies in collaboration with all stakeholders was key to [2] Dellicour S, Tatem AJ, Guerra CA, Snow RW, ter Kuile FO. Quantifying the number of pregnancies at risk of malaria in 2007: a demographic study. PLoS Med 2010;7(1): success of the program. In both countries, the dynamic partnership e1000221. between and leadership from the national reproductive health program [3] World Health Organization. Malaria in pregnant women. http://www.who.int/ and the national malaria control program meant that gaps in provider malaria/areas/high_risk_groups/pregnancy/en. Published 2013. Accessed April 25, performance (i.e. understanding of guidelines, education, supervision) 2014. [4] Guyatt HL, Snow RW. Malaria in pregnancy as an indirect cause of infant mortality in could be addressed efficiently. These successful program initiatives in sub-Saharan Africa. Trans R Soc Trop Med Hyg 2001;95(6):569–76. Burkina Faso and Kenya catalyzed the introduction and expansion of [5] World Health Organization. A Strategic Framework for Malaria Prevention and FANC services with MIP in other East, Southern, and West African coun- Control During Pregnancy in the African Region. http://whqlibdoc.who.int/afro/ 2004/AFR_MAL_04.01.pdf. Published 2004. Accessed April 25, 2014. tries using the same elements of partnerships, education, supportive [6] Roll Back Malaria Partnership Secretariat, World Health Organization. The Abuja supervision, and community sensitization. The pace of expansion to Declaration and the Plan of Action: an extract from the African Summit on Roll other countries was hastened through Jhpiego’s participation in regional Back Malaria. http://www.rbm.who.int/docs/abuja_declaration.pdf.Published 2000. Accessed April 25, 2014. MIP networks. [7] Roll Back Malaria Partnership. Global malaria action plan for a malaria-free world. Recognizing that MIP is both a maternal and newborn health issue, http://www.rollbackmalaria.org/gmap/index.html. Published 2008. Accessed April reproductive health programs should lead efforts to manage program 25, 2014. [8] The President’s Malaria Initiative. Eighth Annual Report to Congress. http://www. implementation on a platform of FANC, and national malaria control pmi.gov/docs/default-source/default-document-library/pmi-reports/pmireport_ programs should provide technical leadership including procurement final.pdf?sfvrsn=12. Published 2014. Accessed April 25, 2014. and distribution of needed MIP supplies to ANC clinics (i.e. SP, LLINs, [9] van Eijk AM, Hill J, Alegana VA, Kirui V, Gething PW, ter Kuile FO, et al. Coverage of malaria protection in pregnant women in sub-Saharan Africa: a synthesis and and treatment drugs). This critical partnership starts at the national analysis of national survey data. Lancet Infect Dis 2011;11(3):190–207. level, where both programs can harmonize national level MIP materials, [10] USAID, ACCESS. Focused antenatal care: providing integrated, individualized care including policies, guidelines, and education materials, and coordinate during pregnancy. http://www.accesstohealth.org/toolres/pdfs/accesstechbrief_ effective implementation. fanc.pdf. Published 2007. Accessed April 25, 2014. [11] Roman E, Rawlins B, Gomez P, Dineen R, Dickerson A, Brieger W. Malaria in pregnan- To date, IPTp uptake has been the main focus of country programs; cy: the dynamic relationship between policy and program implementation. Harv however, ITN use and correct case management are also recognized as Health Policy Rev 2008;9(1):198–209. essential components of a comprehensive MIP program. Engagement [12] Jhpiego. Scaling up malaria in pregnancy programs: What it takes! http://www. jhpiego.org/files/MalariaStrategyPaper_2008.pdf. Published 2008. Accessed April with ITN partners can help to increase awareness among pregnant 25, 2014. women about the importance of ITN use, and the role of ANC in routine [13] Ouma PO, Van Eijk AM, Hamel MJ, Sikuku E, Odhiambo F, Munguti K, et al. The effect distribution of ITNs among the partners. The eight program areas of MIP of health care worker training on the use of intermittent preventive treatment for malaria in pregnancy in rural western Kenya. Trop Med Int Health 2007;12(8): are essential health systems components to address when countries are 953–61. focusing on improving MIP morbidity and mortality rates [18].Ascoun- [14] Robb-McCord J, Voet W. Scaling up practices, tools, and approaches in the Maternal tries consider application of community-based distribution and/or and Neonatal Health program. http://www.jhpiego.org/files/scaleupMNH.pdf. Published 2003. Accessed April 25, 2014. social mobilization models to augment their existing strategies, it will [15] Rational Pharmaceutical Management (RPM) Plus Program, Jhpiego, U.S. Centers for be essential to document outcomes related to MIP prevention and Disease Control and Prevention, World Health Organization. Malaria Action case management to better understand the feasibility and acceptability Coalition Final Report: October 2002–September 2007. http://pdf.usaid.gov/pdf_ docs/Pdacl147.pdf. Published 2007. Accessed April 25, 2014. of these approaches and guide application. [16] Newman RD, Moran AC, Kayentao K, Benga-De E, Yameogo M, Gaye O, et al. Preven- In the future, countries will need to develop or review their national tion of malaria during pregnancy in West Africa: policy change and the power of malaria control policy, strategies, and guidelines for MIP to ensure align- subregional action. Trop Med Int Health 2006;11(4):462–9. ment with global approaches toward malaria elimination. The WHO’s [17] Centers for Disease Control and Prevention, USAID, Presidential Malaria Initiative, Maternal and Child Health Integrated Program. A health systems approach to three-pronged approach should be an integral part of national malaria implementing malaria in pregnancy programs (presentation). http://www.mchip. prevention and control policies and should also be reflected in national net/sites/default/files/08.%20%20MIP%20health%20systems%20approach%20presen- reproductive health policies. In particular, the third prong—malaria case tation%201.30.2013.pdf. Published 2013. Accessed April 25, 2014. — [18] Roman E, Wallon M, Brieger W, Dickerson A, Rawlins B, Agarwal K. Moving malaria management with improved diagnostics and treatment is needed as in pregnancy programs from neglect to priority: experience from Malawi, Senegal, part of case detection and surveillance among pregnant women to and Zambia. Glob Health Sci Pract 2014;2(1):55–71. better serve both women and their children. [19] Brieger WR, Orji BC, Okeibunor J, Ishola G, Otolorin E, Rawlins B, et al. Improving quality, involving community: a two-pronged approach to preventing malaria in Jhpiego continues to test new approaches that will support MIP pregnancy in Nigeria (poster). American Society of Tropical Medicine and Hygiene scale-up and help countries achieve their target goals. Ongoing collabo- 59th Annual Meeting, Atlanta, Georgia, November 3–7, 2010; 2010. ration and support are needed to achieve Millennium Development [20] CDI Study Group. Community-directed interventions for priority health problems in Africa: results of a multicountry study. Bull World Health Organ 2010;88(7):509–18. Goals 4 and 5 and move beyond these goals to eliminate malaria. [21] Okeibunor JC, Orji BC, Brieger W, Ishola G, Otolorin E, Rawlins B, et al. Preventing malaria in pregnancy through community-directed interventions: evidence from Akwa Ibom State, Nigeria. Malar J 2011;10:227. Conflict of interest [22] World Health Organization. Updated WHO policy recommendation (October 2012). Intermittent preventive treatment of malaria in pregnancy using sulfadoxine- pyrimethamine (IPTp-SP). http://www.who.int/malaria/iptp_sp_updated_policy_ The authors have no conflicts of interest. recommendation_en_102012.pdf?ua=1. Published 2012. Accessed April 25, 2014. International Journal of Gynecology and Obstetrics 130 (2015) S68–S73

Contents lists available at ScienceDirect

International Journal of Gynecology and Obstetrics

journal homepage: www.elsevier.com/locate/ijgo

SUPPLEMENT ARTICLE RED for PMTCT: An adaptation of immunization’s Reaching Every District approach increases coverage, access, and utilization of PMTCT care in Bondo District, Kenya☆

Lynn Kanyuuru a,⁎,MarkKabueb, Tigistu A. Ashengo c, Chandrakant Ruparelia b, Evans Mokaya d, Isaac Malonza a a Jhpiego/MCHIP, Nairobi, Kenya b Jhpiego, Baltimore, MD, USA c Jhpiego /MCHIP, Washington, DC, USA d JSI/MCHIP, Nairobi, Kenya article info abstract

Keywords: Gaps exist in coverage, early access, and utilization of prevention of mother-to-child transmission of HIV Access (PMTCT) services in Kenya. The Maternal and Child Health Integrated Program, led by Jhpiego, piloted an Coverage adaptation of immunization’s Reaching Every District (RED) approach in Bondo District as a way of improving Maternal and child health PMTCT care. Routine district-level monthly summary service delivery pre- and post-implementation data were Prevention of mother-to-child transmission analyzed. Marked improvements resulted in the proportion of HIV-infected and non-infected pregnant women (PMTCT) completing four focused prenatal care visits, from 25% to 41%, and the proportion of HIV-exposed infants Reaching Every District approach (HEIs) tested at six weeks, from 27% to 78% (P b 0.001). The proportion of HEIs tested for HIV infection at Utilization 12 months was 52%, while 77% of HEIs were issued antiretroviral prophylaxis by the end of the pilot. Implementation of RED for PMTCT demonstrated that PMTCT services can be delivered effectively in the context of the existing community strategy and resulted in increased coverage, access, and utilization of care for HIV-positive pregnant women and their children. © 2015 Published by Elsevier Ireland Ltd. on behalf of International Federation of Gynecology and Obstetrics. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

1. Background [4]. The number of HIV-positive infants from those HIV-positive pregnancies is estimated at 22 000 [3,4]. An estimated 35.3 million people are living with HIV globally, The Government of Kenya’s Ministry of Health (MOH) PMTCT around 70% of whom are in Sub-Saharan Africa [1]. Although new HIV program was launched in 2000 and has achieved nearly universal infections among children have dropped by 52% since 2001, and access facility-based coverage, with 4,000 of the 4,400 (90%) health facilities to services for the prevention of mother-to-child transmission (PMTCT) offering maternal, newborn, and child health (MNCH) services includ- of HIV has increased over the years, children are still disproportionately ing PMTCT services [5]. However, there have been challenges to the affected, with an estimated 260 000 (230 000–320 000) of new HIV in- full scale-up of PMTCT in Kenya that included: late prenatal care fections in 2012 occurring among children [1]. In Kenya, HIV transmis- attendance; low utilization of prenatal care services and facility-based sion from mother to child is considered one of the biggest health and births; lack of integration of PMTCT services with reproductive health development challenges. Out of Kenya’s estimated population of 38.6 and family planning services; and lack of integration of early infant million in 2009 with 1.55 million births [2], the national HIV prevalence diagnosis in the MNCH continuum, resulting in missed opportunities among pregnant women in 2013 was 6.0% [3], with an estimated 81 000 for pediatric diagnosis, care, and treatment [6,7]. children exposed to the virus through maternal-to-child transmission At the time of the present study, the relevant WHO PMTCT recommendations (2010) called for initiating antiretroviral (ARV) prophylaxis as early as 14 weeks of gestation and continuing through to delivery ☆ Part of this work was presented as a poster abstract at the 19th Conference on (also known as “Option A”) [8]. To study the programmatic possibilities Retroviruses and Opportunistic Infections, held March 5–8, 2012, in Seattle, WA, USA. for and challenges of implementing this recommendation in Kenya’s ⁎ Corresponding author at: Jhpiego/Maternal and Child Health Integrated Program, 14 national PMTCT strategy, the USAID-funded Maternal and Child Health Riverside, off Riverside Drive, Arlington Block, 3rd floor, Nairobi, Kenya. Tel.: +254 731 ’ 780 903. Integrated Program (MCHIP), led by Jhpiego, adapted the WHO s suc- E-mail address: [email protected] (L. Kanyuuru). cessful Reaching Every District (RED) approach for scaling up http://dx.doi.org/10.1016/j.ijgo.2015.04.002 0020-7292/© 2015 Published by Elsevier Ireland Ltd. on behalf of International Federation of Gynecology and Obstetrics. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). L. Kanyuuru et al. / International Journal of Gynecology and Obstetrics 130 (2015) S68–S73 S69 immunization services in several African countries [9] to PMTCT service Resources from other programs were leveraged to maximize the delivery and piloted this RED for PMTCT model in Bondo District, Kenya. services provided to the end users. The MCHIP activities in the district The RED approach emphasizes five operational components that are were harmonized with other programs conducted by Jhpiego and specifically aimed at improving programmatic coverage: (1) better other partners in the district that focused on the development of facility planning and management of resources; (2) reaching all target popula- infrastructure, strengthening HIV commodities and supplies manage- tions through outreach services; (3) supportive supervision for service ment, and increasing the number of health workers knowledgeable providers; (4) linking communities with service delivery; and (5) mon- about HIV through in-service education. To prevent unintended itoring for action [9]. pregnancies among women living with HIV, for example, Jhpiego’s RED for PMTCT, also known as the Reaching Every Pregnant Woman ACCESS Uzima program (preceded MCHIP) trained service providers approach, was piloted in Bondo District with the aim of demonstrating in- on long-term family planning methods, ensuring availability of family creased uptake and utilization of PMTCT services. Bondo District is located planning commodities and supplies, and disseminating health messages in Nyanza Province (administrative structures changed from provinces to through radio announcements and community health workers (CHWs) counties in 2011), in the western part of Kenya. At the time the RED for to educate the community on family planning with the aim of increasing PMTCT pilot started in 2010, the district had some of the worst health in- family planning uptake in the district. dicators in Kenya, including high maternal mortality (640 deaths per 100 Other activities capitalized on the community-based approach that 000 live births), infant mortality (110 infant deaths per 1000 live births), was set out in the MOH’s community strategy [12], which empowered and child mortality (208 deaths of children under five years per 1000 live households and communities to strengthen their role in health and births) rates due to HIV/AIDS, tuberculosis (TB), and malaria (Bondo Dis- health-related development by increasing their knowledge, skills, and trict Management Team, unpublished data, 2010). Nyanza Province also participation. Under this strategy, communities are organized into had some of the highest health disparities in the country, with an estimat- community units each comprising approximately 1000 households or ed population HIV prevalence of 20.2% (17.1% in men and 22.8% in 5000 people living in the same geographical area and sharing resources women) and under-five mortality rates of 101 per 1000 live births, as and challenges. The community units are organized around villages and compared with 60 per 1000 live births in Nairobi [10]. This burden of dis- other interest groups that are responsible for identifying and supporting ease, coupled with rampant poverty and underdevelopment, contributed volunteer CHWs. The CHWs report to a community health committee greatly to the poor health status of its populace [10].Furthermore,widow (CHC) through a community health extension worker (CHEW), who is inheritance, a common cultural practice where a designated male as- an employee of the MOH and a secretary for the CHC. The health gover- sumes social and/or economic responsibilities for the widow upon the nance structure closest to the community is the CHC, whose members death of her husband, has also been associated with an increased risk of are elected in such a way that all the villages in the community unit acquisition of HIV infection in this region [11]. are represented [12]. The present article reports the results and program experience of RED Because CHWs were key personnel in this RED for PMTCT pilot, for PMTCT and evaluates its attempts to increase coverage, access, and uti- facility-based healthcare workers were encouraged to partner with lization (proxies from the immunization field for uptake, access, and re- communities through the CHWs to ensure utilization of their health tention) relating to PMTCT services through focused prenatal care in services. CHWs were recruited from all villages; one CHW served one pilot district. We also report challenges and lessons learned to help approximately 500 persons (about 100 households). The CHWs were fosterreplicationofthisapproachinotherresource-limitedsettings. trained on the community strategy, community-based health information systems, and the RED for PMTCT, or Reaching Every Pregnant 2. Program experience Woman, approach. Two educational models were used: (1) training of CHEWs as master trainers who subsequently trained CHWs in their The RED for PMTCT pilot intervention was conducted between July respective community units; and (2) training of some CHWs who 2010 and June 2012 in Bondo District, Nyanza Province, Kenya. Before then cascaded the training to fellow CHWs. the intervention began, advocacy, consultative, and planning meetings CHW exit desks were established at all health facilities to schedule were held at the national and regional levels to promote the active appointments for prenatal care, postnatal care, and HIV-exposed infant engagement of health authorities. Catchment areas were mapped, (HEI) visits. The CHWs thus served as links with the community. with average distances to health facilities determined. With the help Women who missed a scheduled appointment were contacted by of community representatives and service providers, villages with phone to schedule new appointments. If necessary, women were phys- high numbers of unreached pregnant women were identified. A situa- ically traced by their respective CHW. Regardless of HIV status, all tional analysis done to identify barriers to access and/or utilization of MNCH clients were also connected with the CHW from their respective health services revealed that of the 329 villages mapped, only seven village to be followed up in the community. In addition, the CHWs (2%) were more than five kilometers from the nearest health facility. actively engaged the male partners of the pregnant women to educate Although most communities lived within walking distance to a health them on MNCH issues and encourage their participation in prenatal facility, consistent access to health facilities remained difficult. Some care. Pelzer et al. [13] reported that the involvement of the male partner of the challenges identified included: cost; lack of knowledge (majority is likely to increase the chance of successful PMTCT interventions. didn’t know whether it was necessary to go to attend prenatal care Integrated community outreaches in hard-to-reach areas were con- early or complete at least four visits); proximity of the nearest clinic; ducted monthly. Prioritization was done based on distance to the nearest husbands (or religions) prohibiting wives (or pregnant followers) facility and the number of unreached pregnant women. Communities from attending prenatal care; and that they were too busy with eco- participated in service delivery by providing the venue and community nomic activities (usually fishing). Reaching communities living on the resources for the outreach. Services offered at these outreaches included: Islands of Lake Victoria presented unique challenges for monitoring prenatal care; postnatal care; early infant diagnosis (EID); HIV testing treatment utilization owing to transportation issues and the fact that and counseling; family planning counseling; distribution of basic family these communities frequently migrated to different fishing locations, planning commodities and referrals to health facilities; immunization; some of which were in neighboring districts and countries. Service de- defaulter tracing; growth monitoring of children, deworming, treat- livery strategies therefore included those for fixed centers (established ment and referral for the sick; and health education. health facilities) and those for outreach to hard-to-reach areas, as Quarterly supportive supervision visits to health facilities were de fined by distance from a health facility and terrain. A mobile service undertaken by the District Health Management Team (DHMT). Skills was not adopted because most villages were within five kilometers’ gaps among health service providers were identified and addressed walking distance to a health facility. through on-site mentorship, and in-service courses were conducted S70 L. Kanyuuru et al. / International Journal of Gynecology and Obstetrics 130 (2015) S68–S73 for service providers when necessary. These courses covered new to June 2012 (the last six months of the intervention). We analyzed PMTCT [14] and Infant and Young Child Feeding (IYCF) guidelines data collected using MOH tools that were available at the community [15], data collection tools and registers, the Reaching Every Pregnant level, namely the household register, the CHW logbook, and the Woman approach, and Standards-Based Management and Recognition monthly CHEW summary. Two additional summary monthly forms (SBM-R, Jhpiego, Baltimore, USA) for PMTCT. SBM-R for PMTCT was (711A and 731) were used by the MOH to capture PMTCT data. Data the quality improvement strategy used to ensure that high-quality from 33 health facilities and all the 26 community units in Bondo PMTCT services were provided as demand for services increased. This District were summarized and routinely entered in the national District quality improvement approach involves facility self-assessments on Health Information System (DHIS2) [18] by the district health records established performance standards for given health areas and is officer. The findings for the indicators of interest from the two time described in Necochea et al. [16] in the present supplement. In this periods were compared through a two-sample test of proportions, and project, health facilities were supported to assess their performance MOH summary data and program reports were also reviewed to against Kenya’s national PMTCT performance standards [17], a set of triangulate the information obtained from the DHIS2. P b 0.05 was 110 standards across the following intervention areas: (1) information considered statistically significant. No institutional review board education and counselling (IEC); (2) focused prenatal care; (3) labor determination was sought for the study because the Kenya DHIS2 data and delivery; (4) postnatal care; (5) infection prevention and control are publically available [18], and the use of program reports in aggregate (IPC); (6) human and physical resources; and (7) management systems. form was not human subjects research. The IEC standards mainly assessed whether clients received all relevant information related to PMTCT. The focused prenatal care, labor and 4. Findings delivery, and postnatal care standards assessed whether clients received all services as per the national guidelines in a respectful manner. 4.1. Access to and utilization of PMTCT-related services though prenatal The IPC standards assessed whether all protocols of infection prevention care and labor and delivery were adhered to. Human and physical resources standards assessed whether there was adequate infrastructure, commodities, and supplies The findings of the situational analysis showed that there was high for provision of PMTCT services. In addition, the standards also assessed coverage of one prenatal care visit services ( N95% of pregnant women whether the service providers were adequately trained in the provision attended at least one prenatal care visit); however, utilization of prena- of these services. Whether facilities were ready to provide the services tal care services overall was low, more than 65% of the pregnant women was assessed under the management systems’ standards. Client flow, avail- dropping out of care between their first and fourth visits. Although ability and display of appropriate signage, and the use of data for decision most pregnant women attended at least one prenatal care visit, more making were also assessed. The PMTCT−MNCH integrated services pack- than 75% attended in the third trimester only, with just 3% attending age was defined and disseminated to health workers as a job aid (http:// in the first trimester. Furthermore, the majority of women did not reprolineplus.org/resources/pmtctmnch-integrated-service-job-aid). know about the importance of attending prenatal care early and Monitoring and evaluation was done jointly by MCHIP, national and attending at least four visits. Other identified barriers included: regional MOH staff. MCHIP facilitated the availability of MOH tools and husbands prohibiting their wives attending prenatal care; religious registers for use at the community level. Health facility administrators, leaders prohibiting their followers to seek medical services; the percep- MOH partners, and the DHMT held monthly meetings to review health fa- tion that services were too costly (although in many areas traditional cility performance, and the community units held monthly data review birth attendants charged significantly more); and conflicts with the meetings. Strategies were re-prioritized based on the outcomes of those timing of economic activities such as fishing. meetings. Technical assistance was provided to the health records team Between 2010 and 2012, the proportion of pregnant women who to ensure timely, complete, and accurate reporting of selected prenatal completed four focused prenatal care visits improved significantly care indicators that were introduced as proxy measurements for access from 25% to 41% (P b 0.001), and delivery with skilled birth attendants to prenatal care, early prenatal care attendance, and service utilization increased from 23% to 47% (P ≤ 0.001). Uptake of prenatal care partner both before and after the RED for PMTCT intervention (Table 1). testing (a proxy for male involvement) increased from 1.8% in 2010 to 19.3% in 2012 (P b 0.001) as more men sought testing. At the same 3. Program assessment time, the proportion of partners (men) who tested positive for HIV decreased (22.6% in 2010 vs 7.3% in 2012; P b 0.001). All clients were To document the utility of the RED for PMTCT intervention, we offered HIV testing and were given the option of opting out. Table 2 collected and analyzed cross-sectional data at two similar time points summarizes the findings of these comparisons between the pre- two years apart: January to June 2010 (pre-intervention) and January intervention and end-of-pilot periods.

4.2. Diagnostics for HIV-positive pregnant women and HIV-exposed infants Table 1 Proxy measurements for access to prenatal care, early prenatal care attendance, and service utilization. Over the course of the intervention, the proportion of pregnant women attending prenatal care who tested positive for HIV decreased Indicator Intended measurement only slightly from 21% to 18% (P = 0.002). The proportion of women Proportion of pregnant women who Proxy indicator for access (the possibility at maternity who tested positive for HIV did not change significantly attend prenatal care 1 visit of a pregnant woman reaching a health (25.8% in 2010 vs 27.3% in 2012) but the proportion of HEIs tested for (prenatal care 1 coverage) facility easily and getting the HIV infection at six weeks increased dramatically from 27% to 78% service required) b Proportion of pregnant women who Proxy indicator for early prenatal care (P 0.001) (Fig. 1). By the end of the pilot, the proportion of HEIs tested attend prenatal care 4 visit attendance (only those who attend at 12 months remained high, at 52%, while 77% of HEIs at 12 months prenatal care attendance early can were issued ARV prophylaxis. In addition, 77% of HEIs aged six months complete the four scheduled visits) a were exclusively breastfed, as shown in Table 3. Only 35 of the 765 Dropout rate between prenatal care 1 Proxy indicator for utilization fi and prenatal care 4 coverage HEIs younger than 12 months, or 4.6%, were identi ed as testing HEI testing and ARV infant prophylaxis Proxy indicators for retention in care positive for HIV. Overall, 62% of the HIV-positive pregnant women were assessed for antiretroviral therapy (WHO staging and CD4 counts) a The project adapted the dropout rate indicator usually used by immunization teams and defined it as the difference in coverage between the first and forth prenatal care visits. at prenatal care clinics. For these endline results, baseline data were not All other indicators are standard indicators. available for comparison. L. Kanyuuru et al. / International Journal of Gynecology and Obstetrics 130 (2015) S68–S73 S71

Table 2 Prenatal care and labor and delivery results compared between pre-intervention (2010) and end-of-pilot (2012).a

Prenatal care attendance and HIV testing Jan − Jun 2010 Jan − Jun 2012 P value (n = 3600) (n = 3633)

Proportion of prenatal care users returning for the fourth visit 25.0 41.0 b0.001 Proportion of women delivering with assistance of skilled attendants 23.0 47.0 ≤0.001 Proportion of prenatal care users tested for HIV (testing uptake) 79.2 75.7 NS Proportion of prenatal care users who tested positive for HIV 21.4 18.3 0.004 Estimated LLITN coverage among prenatal care users 80.9 81.8 NS Proportion of HIV-positive mothers referred for follow-up 78.5 81.5 NS Proportion of prenatal care partners tested (couple testing) 1.5 14.6 b0.001 Sub-group analyses Proportion of women testing positive given preventive ARVsb 93.1 108.0 Not done Proportion of HIV-positive mothers given infant prophylaxisb 97.3 124.7 Not done Proportion of new prenatal care partners who tested positive for HIVc 22.6 7.3 b0.001 Proportion of HEIs tested for HIV at 6 weeksd 27.5 77.5 b0.001 Maternity, labor and delivery HIV testing Proportion of women in maternity tested for HIV (testing uptake)e 88.8 87.8 NS Proportion of women in maternity who tested positive for HIVe 25.8 27.3 NS

Abbreviations: ARV, antiretroviral; HEIs, HIV-exposed infants; LLITN, long-lasting insecticide-treated net; PMTCT, prevention of mother-to-child transmission of HIV; NS, not signiﬁcant. a Values are given as percentage unless otherwise indicated. b The denominators for these analyses are 770 for 2010 and 665 for 2012. Some HIV-positive pregnant women were counted more than once after testing and receiving ARV prophylaxis multiple times, so the results are greater than 100% in 2012. c The denominators for this analysis are 53 for 2010 and 532 for 2012. There was a 10-fold increase in partner/couple testing. d The denominators for this analysis are 622 for 2010 and 503 for 2012. e The denominators for this analysis are 632 for 2010 and 800 for 2012.

4.3. Performance and quality improvement through SBM-R with virtually all pregnant women during the implementation period reached. The establishment of a CHW desk in health facilities, use of An overall upward trend was observed in SBM-R performance scores CHWs to actively identify those lost from care in their communities on PMTCT standards between baseline and two assessment points. and link them back to services, and an improved community-to- Overall, performance scores for all intervention areas (IEC, labor and health-facility referral system contributed in reducing the dropout delivery, infection prevention, and management systems) improved rate between one prenatal care visit and four prenatal care visits from from a mean of 28% at baseline to a mean of 52% by the second assess- 71% in 2010 to 57% in 2012 (P b 0.001). ment (Fig. 2). The largest change in performance occurred at labor and delivery, from 17% to 56%, although that baseline score was the lowest. 5. Discussion and lessons learned The highest second (and final) assessment score was in focused prenatal care, at 70%, although it had the highest baseline value of 49%. The RED for PMTCT pilot intervention was evaluated using routine program monitoring data collected electronically through a government-sponsored health information system where data are 4.4. Other findings constantly updated, publically available, and can be used in real-time for decision making. Implementation of the RED for PMTCT approach CHW coverage of the catchment area of Bondo District increased included general adherence to RED’s operational components and a from 38% in June 2010 to 100% in June 2012—a total of 26 community particular emphasis on the lowest level of service delivery, the commu- units. The average number of women eligible for prenatal care atten- nity. Overall, the findings are suggestive that RED for PMTCT was dance was approximately 6000 annually between 2010 and 2012, successful at increasing access to and utilization of PMTCT services among those who attended prenatal care in Bondo District, as demonstrated by the increase in early attendance, delivery under skilled care, 100.0% HIV positive prenatal care users and HEI testing at six months by the end of the pilot. The quality of PMTCT service delivery also improved, particularly in labor and deliv- 80.0% HEI who tested for HIV at 6 weeks ery, where other services might otherwise have been prioritized.

60.0% Table 3 HIV-exposed infants and early infant diagnosis services indicators at end-of pilot 40.0% (January − June 2012). Percentage Percentage Indicator No. Percentage

20.0% Proportion of known HIV-positive women 438/1829 23.9 at entry at prenatal care Proportion of women with unknown HIV 334/1829 18.3 0.0% status at entry at prenatal care 2010 2011 2012 Proportion of HIV-positive mothers assessed 485/772 62.0 for ART at prenatal care Proportion of HIV-positive mothers started 82/772 10.6 Year 2010 2011 2012 on ART at prenatal care HIV-positive prenatal care users 622/2903 697/3253 503/2750 Proportion of HEIs tested by 12 months 765/1480 51.7 (Num/Den) Proportion of HEIs aged 6 months who are 749/973 77.0 HEI who tested for HIV at 6 weeks 171/622 278/697 390/503 (Num/Den) exclusively breastfed Proportion of HEIs aged 12 months issued 691/897 77.0 with ARV prophylaxis Fig. 1. Trends in proportion of HIV-positive women at prenatal care and HIV-exposed infants at six weeks. Abbreviation: HEI, HIV-exposed infants. Abbreviations: ART, antiretroviral therapy; HEIs, HIV-exposed infants. S72 L. Kanyuuru et al. / International Journal of Gynecology and Obstetrics 130 (2015) S68–S73

80 70 70 65 60 60 57 56 52 50 49 49 42 38 36 40 37 36 33 32 31 27 30 27 22 21 PERCENTAGE 20 17 10 0 IEC FANC LABOUR & PNC INFECTION HUMAN & MANAGEMENT DELIVERY PREVENTION PHYSICAL SYSTEMS RESOURCES INTERVENTION AREAS

Baseline Assessment 1st Assessment 2nd Assessment

Fig. 2. Standards-Based Management and Recognition (SBM-R) performance scores by intervention areas at three assessment points. Abbreviations: IEC, information, education, and counseling; FANC, focused prenatal care; PNC, postnatal care.

Despite relatively higher prenatal care attendance at the end of the pilot Defaulter tracing of the HEIs is often difficult, a finding reported else- than at the beginning, low early prenatal care attendance and high pre- where [21–23]. The present study also documented this challenge. In natal care dropout rate continue to be a major barrier to early and con- addition, HIV-positive infants had to be referred out of their primary fa- tinued initiation of needed services. cilities when they were 18 months old. On top of this, there were occa- From the findings, it is apparent that improvement in the sional regional stock outs of HIV commodities that influenced PMTCT community-based PMTCT initiative may have been largely due to the service delivery, especially EID. sustained engagement of the key stakeholders and collaboration with Bondo District had a number of partners with frequently competing the various partners working in the area through all the stages of the priorities. CHWs who were part-time volunteers often supported on program. Stakeholder participation was ensured in the process of many intervention areas including MNCH, HIV, tuberculosis, water and identifying challenges, problem-solving, prioritizing targeted solutions, sanitation, and malaria. Frequent engagement of facility-based health and mobilizing resources for solutions that were already well packaged workers in offsite meetings and trainings by other partners sometimes in the RED approach. These same factors have been shown as catalysts caused artificial staff shortages and adversely affected service delivery. for systematic changes at institutional and national levels within Barron et al. [24] identified the need to ensure good coordination with PMTCT programs in 34 global countries [19]. technical partners, such as international health agencies and international Regarding capacity building of CHWs, the master trainer model and local nongovernmental organizations, to improve program outcomes. using CHEWs was found to be an efficient, inexpensive, and expedient way to train CHWs in their local setting. However, it was necessary to 6. Conclusion complement the training approach with regular and structured supportive supervision; requests to share challenges and successes; Implementation of RED for PMTCT resulted in improved early access and a mechanism for more regular and timely data collection. Uwinma to and increased utilization of PMTCT services by HIV-positive pregnant et al. [20] identified the need for systematic skill-building, enhanced women and their children in the low-resource setting of Bondo District, scopes of practice, and consistent supervision of CHWs as critical to Kenya. It also enabled PMTCT services to be delivered effectively and in the success of PMTCT integrated programming, with the corollary a sustainable manner in the context of Kenya’s national community need to equip these training and supervision programs with a reliable strategy, which is an important effort to bring quality services closer referral and monitoring and evaluation system. to the people who need them. RED for PMTCT integrates a known pack- The service delivery component of the RED for PMTCT approach faced age of best practices and underscores the importance of community- other challenges, including difficulties tracking the women during the driven approaches to improve the uptake of and retention in PMTCT prenatal and postnatal periods, given that Kenya uses cross-sectional pre- services. With funding from USAID, the RED for PMTCT model will be natal and postnatal care registers from which patients’ histories were not rolled out in two other Kenyan districts, Igembe North and East Pokot, documented in a manner that allowed for follow-up. To overcome this providing an opportunity to test solutions aimed at addressing the challenge, the health prenatal and postnatal care services provided were challenges observed in Bondo and collect additional data. Further recorded in mother − child booklets that the mothers were required to scale-up of this model will be largely dependent on interest from the present during all clinic visits. However, some mothers routinely did not Government of Kenya and its implementing partners. carry the booklets thus obtained new ones each visit while others, in order to avoid stigmatization, altered the entries in the booklets, especially with respect to their HIV status. In addition, some community members Acknowledgments did not want their HIV status known to their CHWs and therefore did not disclose their HIV status to them. As a result of this missing information, This project was funded by USAID through MCHIP with additional service providers at health facilities could not reliably generate defaulter funding from PEPFAR. The authors acknowledge Judith Fullerton lists for the HIV-positive population. A number of studies, including a sys- (independent consultant, Jhpiego), Julius Siguda Oliech (district medi- tematic review of PMTCT services in 12 countries in Sub-Saharan Africa, cal officer of health, Bondo District, Kenya), Oniyire Adetiloye (Jhpiego, confirmed the persistence of stigma as a major barrier to the uptake of Nigeria), Kelly Curran (Jhpiego, Baltimore), and Allan Gohole (Jhpiego, ARV drugs for PMTCT [19,21]. Kenya) for their valuable input and careful review of the manuscript. L. Kanyuuru et al. / International Journal of Gynecology and Obstetrics 130 (2015) S68–S73 S73

We are also grateful to Kenya program staff for their support during [12] MOH. Taking the Kenya essential package for health to the community: a strategy for the delivery of level one services. Nairobi, Kenya: Ministry of Health, Afya implementation of this project. House; Published 2006. [13] Peltzer K, Jones D, Wess SM, Shikwane E. Promoting male involvement to improve Conflict of interest PMTCT uptake and reduce antenatal HIV infection: a cluster randomized controlled trial protocol. BMC Public Health 2011;11:778. [14] Kenya Ministry of Health. Guidelines for Prevention of Mother to Child Transmission The authors have no conflicts of interest. (PMTCT) of HIV/AIDS in Kenya. Nairobi, Kenya: NASCOP; 2009. [15] Kenya Ministry of Health. Kenya National Infant and Young Child Feeding Guidelines References in the Context of HIV AIDS. Nairobi, Kenya: Kenya Ministry of Health; 2010. [16] Necochea E, Tripathi V, Kim YM, Akram N, Hyjazi Y, Luz Vaz M, et al. Implementation of the Standards-Based Management and Recognition approach to quality improve- [1] UNAIDS. Global Report: UNAIDS report on global AIDS epidemic 2013. Published 2013 ment in maternal, newborn, and child health programs in low-resource countries. http://www.unaids.org/en/media/unaids/contentassets/documents/epidemiology/ Int J Gynecol Obstet 2015;130(Suppl. 2):S17–24. 2013/gr2013/UNAIDS_Global_Report_2013_en.pdf. Accessed May 25, 2014. [17] Kenya Ministry of Health. Performance, Standards and Assessment Tool for PMTCT. [2] Kenya National Bureau of Statistics. Kenya 2009 Population and Housing Census Nairobi, Kenya: MOH; 2009. Highlights. Nairobi, Kenya: KNBS; 2010. [18] Kenya Ministry of Health, DHIS2. Kenya Health Information System. https://hiskenya. [3] UNAIDS. HIV and AIDS estimates. http://www.unaids.org/en/regionscountries/ org/dhis-webcommons/security/ login. action. Accessed June 8, 2014. countries/kenya; 2013. Accessed March 26, 2015. [19] Ladner J, Besson M, Rodrigues M, Sams K, Audureau E, Saba J. Prevention of mother- [4] Mbori-Ngacha D, Shaffer N. Prevention of mother-to-child transmission of HIV to-child HIV transmission in resource-limited settings: assessment of 99 Viramune (PMTCT) and pediatric HIV/AIDS care and treatment, 2010. Joint IATT Technical Re- Donation Programmes in 34 countries, 2000–2011. BMC Public Health 2013;13:470. view Mission Report; Published 2010 http://nascop.or.ke/library/pmtct/JRM%20 [20] Uwinma J, Zarowsky C, Hausler H, Jackson D. Engagement of non-government report%20final.pdf. Accessed July 11, 2014. organisations and community care workers in collaborative TB/HIV activities [5] Turan JM, Steinfeld RL, Onono M, Bukusi EA, Woods M, Shade SB, et al. The study of including prevention of mother to child transmission in South Africa: opportunities HIV and antenatal care integration in pregnancy in Kenya: design, methods and and challenges. BMC Health Serv Res 2012;12:233. baseline results of a cluster-randomized controlled trial. PLoS One 2012;7(9): [21] Gourlay A, Birdthistle I, Mburu G, Iorpenda K, Wringe A. Barriers and facilitating e44181. factors to the uptake of antiretroviral drugs for prevention of mother-to-child [6] Stinson K, Jennings K, Myer L. Integration of antiretroviral therapy services into transmission of HIV in sub-Saharan Africa: a systematic review. J Int AIDS Soc antenatal care increases treatment initiation during pregnancy: a cohort study. 2013;16(1):18588. PLoS One 2013;8(5):e63328. [22] Kalembo FW, Zgambo M. Loss to follow up: a major challenge to successful [7] Nutman S, McKee, Khoshnood K. Externalities of prevention of mother-to-child implementation of prevention of mother-to-child transmission of HIV-1 programs transmission programs: a systematic review. AIDS Behav 2013;17(2):445–60. in sub-Saharan Africa. ISRN AIDS 2012;2012:589817. [8] World Health Organization. Antiretroviral drugs for treating pregnant women and [23] Thompson KA, Cheti EO, Reid T. Implementation and outcomes of an active defaulter preventing HIV infections in infants. Geneva: WHO; 2010 http://whqlibdoc.who. tracing system for HIV, prevention of mother to child transmission of HIV (PMTCT) int/publications/2010/9789241599818_eng.pdf. Accessed July 11, 2014. and TB patients in Kibera, Nairobi, Kenya. Trans R Soc Trop Med Hyg 2011;105(6): [9] Ryman T, Macauley R, Nshimirimana D, Taylor P, Shimp L, Wilkins K. Reaching every 320–6. district (RED) approach to strengthen routine immunization services: evaluation in [24] Barron P, Pillay Y, Doherty T, Sherman G, Jackson D, Bhadwaj S, et al. Eliminating the African region, 2005. J Public Health (Oxf) 2010;32(1):18–25. mother-to-child HIV transmission in South Africa. Bull World Health Organ 2013; [10] KNBS, ICF Macro. Kenya Demographic and Health Survey 2008–2009. Calverton, 91:70–4. Maryland, USA: Kenya National Bureau of Statistics and ICF Macro; Published 2010. [11] Agot KE, Vander Stoep A, Tracy M, Obare BA, Bukusi EA, Ndinya-Achola JO, et al. Widow inheritance and HIV prevalence in Bondo District, Kenya: baseline results from a prospective cohort study. PLoS One 2010;5(11):e14028. International Journal of Gynecology and Obstetrics 130 (2015) S74–S76

Contents lists available at ScienceDirect

International Journal of Gynecology and Obstetrics

journal homepage: www.elsevier.com/locate/ijgo

COMMENTARY Eyes on the prize: Linking pre-service education to outcomes

Catherine Carr ⁎, Peter Johnson

Jhpiego, Baltimore, Maryland, USA article info

Keywords: Competency-based education Conceptual model Education outcomes Health worker Health workforce Pre-service education

Crown Copyright © 2015 Published by Elsevier Ireland Ltd. on behalf of International Federation of Gynecology and Obstetrics. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

1. Introduction making the wisest investment possible in a complex system that is influenced by a host of factors, sometimes beyond educators’ control. The pre-service or initial education of a health worker prior to de- We aim to convince readers that future efforts to improve PSE must ployment into the healthcare system is critically important. It ensures begin with careful consideration of the performance outcomes that that all workers begin their careers with a foundation built on compe- graduates need in order to improve outcomes in their community and tence and are prepared to be lifelong learners and potential leaders. the larger health systems to which they will dedicate their careers. There has been long-standing global attention to developing and strengthening pre-service education (PSE) pathways for health workers 2. A case for starting on the right side of the model in low-resource settings as a means to address an ongoing, severe shortage of human resources for health (HRH). Unfortunately, many efforts Optimal health and health system outcomes cannot be achieved have failed to achieve the sustainable results intended [1,2]. Minimal without a fully capable health workforce prepared to provide the educational and infrastructure resources, poorly constructed systems range of services needed to achieve them. While we must trust that for ensuring educational quality, and other daunting influencing factors this assumption is valid, we can also measure the performance out- have combined to impede improvements in community and health sys- comes of students and, to some extent, the impact that they have on tems outcomes through PSE. their community. During the planning or at the outset of projects aimed at improving 1.1. Pre-service education conceptual model PSE, leaders are advised to closely examine health workforce requirements at both the national and community levels. They should consider The pervasive lack of documented results despite a steady stream of government packages of health services outlining service expectations well-intended innovations suggests a need for a new approach to de- at varying levels of the health system, job descriptions for the cadres signing and strengthening PSE systems. The conceptual model depicted being educated, and regulatory documents outlining professional in Fig. 1 [3] provides a visual link between desired pre-service outcomes, scopes of practice. Task analysis, a method of surveying graduates the inputs needed to achieve them (for example, students, teachers, about the frequency and importance of tasks they perform in their – clinical practice sites), and the influencing factors affecting the educa- work, can provide valuable evidence guiding PSE inputs [4 6]. tion, deployment, and assimilation of graduates into the healthcare sys- Stakeholders should ask whether essential resources and materials tem. The authors hope that this commentary will result in readers required for graduate performance are in place before adding competencies to a curriculum. Stakeholders might consider whether communities are ready to accept the services that graduates are prepared to offer and should consider whether the health system authorizes the ⁎ Corresponding author at: Jhpiego, 1615 Thames Street, Baltimore, MD 21231, USA. Tel.: +1 202 834 9014. graduates to perform those services. Educating students to provide E-mail address: [email protected] (C. Carr). care that they are unable to deliver is a completely wasted effort. http://dx.doi.org/10.1016/j.ijgo.2015.03.003 0020-7292/Crown Copyright © 2015 Published by Elsevier Ireland Ltd. on behalf of International Federation of Gynecology and Obstetrics. This is an open access article under the CC BY- NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). C. Carr, P. Johnson / International Journal of Gynecology and Obstetrics 130 (2015) S74–S76 S75

Fig. 1. Conceptual model: the health impacts of pre-service education. This ﬁgure was adapted from Johnson et al. [3]. It is slightly different from the original.

Education and development experts must understand these outcomes 3.2. Teachers/tutors/preceptors and prepare for them with a measurable set of indicators as a first step in any project aimed at improving PSE systems. With measurable Although including investment in teacher preparation in improve- outcomes, direct linkage of results with pre-service education is feasi- ments to PSE is intuitive, it does not always happen [2]. Effective teach- ble. For example, the community outcome of greater utilization of ing in competency-based education requires competent teachers [15]. local health centers for healthy timing and spacing of pregnancy links Preparation of teachers and maintenance of teacher competency should with PSE clinical experience in the community, emphasis on counseling be outcome-focused and integrated with the educational program, in- skills, and development of community relationships with the health cluding preceptors and clinical sites. Preceptors should be a formal facility and the educational institution. part of the faculty and have student assessment responsibilities and authority. When educational integration is extended to clinical sites, classroom tutors become part of the facility, reducing artificial academic– 3. Linking the left side of the model to the right side clinical barriers and connecting teachers and preceptors to the larger educational system. Teachers who are closely connected to clinical set- Given that the purpose of PSE is to graduate competent new service tings are aware of the parameters of facilities and are more able to de- providers who can be absorbed into the health system with relative velop lesson plans that leverage resources and mitigate challenges. ease, both sides of the model must be examined and considered in The outcome focus of graduating competent providers should be implementing any plan. grounded in reality, and all teaching staff (teachers and preceptors) must work with students to solve problems and provide the best care using available resources. 3.1. Students Retention of teachers and preceptors is improved when they are respected in the community. This is a reason for targeting faculty re- Student selection criteria should be evaluated and possibly cruitment to areas where they are needed and more likely to share revised in relationship to the outcomes desired. The traditional crite- the language and culture of the community members. It is important rion of academic grades or scores on admission exams, though predic- that professional leaders demonstrate both academic and clinical tive of academic success, limits the pool of students. With limited competence. Overwhelming workloads in both classroom and clinical resources for HRH, it is important to select students who will not only settings are disincentives for teachers and preceptors to be effective matriculate and graduate, but will also remain in the profession and educators, much less take on additional professional work. Teacher/ contribute to the health workforce. There are numerous examples preceptor incentives must include manageable workloads, time for of targeted recruitment that positively affect student selection and professional activities, and recognition of the need for ongoing prep- subsequent retention [7–11]. aration. Teachers may have great difficulty maintaining their own Some programs have found it imperative to include a cultural lens in clinical competency, given their educational responsibilities. Providing student selection. For example, in Afghanistan, candidates for education teachers with opportunities for clinical practice should be encouraged as community midwives are selected by their communities. In a culture but may be a challenge to implement with current teacher shortages. where a woman is required to obtain permission from her father or husband, at a minimum, to work or go to school, community support of the student is imperative [8]. Although some countries assign students to 3.3. Infrastructure and management cadres without considering the students’ interest in the profession, this is likely to be a detriment to successful education and deployment. There is good evidence from the in-service training literature [16] Fowler and Norrie [12] note that absence of expressed interest in the that learners’ practice in simulated settings increases their engagement profession is a predictive factor for attrition. Enlisting community and in their education and enhances their skill competency before they see professional stakeholders in efforts to identify qualified candidates patients. An appropriately equipped and organized simulation lab facil- may lead to recruitment of qualified students who would otherwise itates practicing in a situation that reflects the clinical setting as closely be overlooked. Students who enter with the encouragement or sponsor- as possible. Use of standardized patients, cases, role plays, and simula- ship of professionals in their chosen field may have the advantage of tions should replicate what is likely to be seen in the community. being vetted by their future colleagues. Targeting recruitment to specific An effective learning environment requires a sufficient number of geographic or population groups may increase the likelihood that grad- teaching and learning tools for the student cohort, physical space for uates will return to their communities, particularly if recruitment is classrooms, and an area where students can study, whether it is a library paired with support such as graduate internships or other bridge or computer lab or a combination. Modern education practices require a programs, academic mentoring, and peer tutoring [10,11,13,14]. computer lab. In areas of unavailable or irregular connectivity, there are S76 C. Carr, P. Johnson / International Journal of Gynecology and Obstetrics 130 (2015) S74–S76 offline resources such as the eGranary Digital Library from the WiderNet can assess student learning. Implementers have an obligation to ensure Project that work to reduce digital divide barriers [17]. that an intervention does not do damage to other parts of the model. Providing education models and resources to clinical facilities Investment in PSE is a long-term commitment—and it is a crucial invest- strengthens the PSE program and offers additional opportunities for ment in preparing leaders for the future. staff to practice. Brief, on-site continuing education programs benefit the facility as well as the students. Conflict of interest

3.4. Curriculum The authors have no conﬂicts of interest.

The curriculum should reflect national health needs balanced with international competencies. International competencies for practice References [18–21] can be appropriately integrated with the work that graduates [1] Frenk J, Chen L, Bhutta ZA, Cohen J, Crisp N, Evans T, et al. Health professionals for a must be able to do to affect outcomes in their country. Assignments new century: transforming education to strengthen health systems in an interde- should be consistent with local practices and culture; learning materials pendent world. Lancet 2010;376(9756):1923–58. should be reflective of national health needs rather than classic texts. An [2] Fullerton JT, Johnson PG, Thompson JB, Vivo D. Quality considerations in midwifery pre-service education: exemplars from Africa. Midwifery 2011;27(3):308–15. example would be placing greater emphasis on kangaroo care for [3] Johnson P, Fogarty L, Fullerton J, Bluestone J, Drake M. An integrative review and pre-term and low birth weight newborns and less emphasis on use of evidence-based conceptual model of the essential components of pre-service educa- neonatal exchange transfusion. Support from regulators of educational tion. Hum Resour Health 2013;11:42. [4] Garbin M, Chmielewski C. Job analysis and role delineation: LPN/LVNs and hemodi- programs such as relevant ministries and professional councils is alysis technicians. Nephrol Nurs J 2013;40(3):225–39. needed for policy-level buy-in. Administrators and practitioners at [5] Oshio S, Johnson P, Fullerton J. The 1999–2000 task analysis of American nurse- sites associated with the school should understand the content and midwifery practice. J Midwifery Womens Health 2002;47(1):35–41. fl [6] Prekeges J. Components of preparedness statements to accompany 2003 task analy- ow of the curriculum as well as the role of teachers. sis. J Nucl Med Technol 2005;33(1):48–60. [7] Currie S, Azfar P, Fowler RC. A bold new beginning for midwifery in Afghanistan. 3.5. Clinical sites Midwifery 2007;23(3):226–34. [8] Glasser M, Hunsaker M, Sweet K, MacDowell M, Meurer M. A comprehensive medical education program response to rural primary care needs. Acad Med 2008; Clinical sites must reflect a balance that considers the needs of the 83(10):952–61. learner, the needs of individuals seeking care, and the needs of the facil- [9] Kaye DK, Mwanika A, Sewankambo N. Influence of the training experience of ity and providers. For an education program to be outcome-focused, the Makerere University medical and nursing graduates on willingness and competence to work in rural health facilities. Rural Remote Health 2010;10(1):1372. clinical sites must have the resources they need for education as well as [10] Mathews M, Rourke JTB, Park A. The contribution of Memorial University’smedical the resources to achieve the desired outcomes. Educational institutions school to rural physician supply. Can J Rural Med 2008;13(1):15–21. [11] Matsumoto M, Inoue K, Kajii E. Characteristics of medical students with rural origin: have to be part of a resource stream; health professional students re- – fi Implications for selective admission policies. Health Policy 2008;87(2):194 202. quire a signi cant amount of relevant, high-quality clinical practice to [12] Fowler J, Norrie P. Development of an attrition risk prediction tool. Br J Nurs 2009; become competent providers. Competent providers working at a clini- 18(19):1194–200. cal practice site increase opportunities for engagement with the com- [13] Mulholland J, Anionwu EN, Atkins R, Tappern M, Franks PJ. Diversity, attrition and transition into nursing. J Adv Nurs 2008;64(1):49–59. munity that receives the services and therefore improve community [14] Pariyo GW, Kiwanuka SN, Rutebemberwa E, Okui O, Ssengooba F. Effects of changes health outcomes. High-quality clinical sites should be assessed by stan- in the pre-licensure education of health workers on health-worker supply. Cochrane dard criteria for quality, maintain quality improvement systems, and Database Syst Rev 2009(2):CD007018. [15] Fullerton J, Thompson J, Johnson P. Competency-based education: The essential place a high value on demonstration of professional behavior and ethics. basis of pre-service education for the professional midwifery workforce. Midwifery 2013;29(10):1129–36. 4. Conclusion [16] Bluestone J, Johnson P, Fullerton J, Carr C, Alderman J, BonTempo J. Effective in- service training design and delivery: evidence from an integrative literature review. Hum Resour Health 2013;11:51. The parts of the PSE conceptual model are inextricably linked, and [17] eGranary Digital Library: http://www.widernet.org/eGranary/. successful investment in PSE requires consideration of the whole [18] Affara F. ICN Framework of Competencies for the Nurse Specialist. Geneva: Interna- model. Nongovernmental organizations that invest in PSE have differ- tional Council of Nurses (ICN); 2009. [19] International Confederation of Midwives. ICM Essential Competencies for Basic Midwife- ent scopes of interest and implementation budgets. However, at a min- ry Practice. Published 2013. Accessed April 16, 2014 http://www.internationalmidwives. imum, those interested in investing in, establishing, or improving org/assets/uploads/documents/CoreDocuments/ICM%20Essential%20Competencies% education systems should go through the exercise of figuring out 20for%20Basic%20Midwifery%20Practice%202010,%20revised%202013.pdf. fi [20] World Health Organization. Strengthening Midwifery Toolkit: Module 4, Competencies where their efforts t with each part of the model. Factors on both for Midwifery Practice. Published 2011. Accessed April 16, 2014 http://whqlibdoc.who. sides of the model inevitably influence any investment in PSE. Consider- int/publications/2011/9789241501965_module4_eng.pdf. ation may need to be given to advocacy for alternative investments. For [21] World Health Organization. Global Standards for the initial education of professional nurses and midwives. Published 2009. Accessed April 16, 2014 http://www.who.int/ example, instead of a new curriculum, more critical needs might include hrh/nursing_midwifery/hrh_global_standards_education.pdf. stronger clinical sites, better-prepared teachers, and preceptors who