Comprehensive review of the evidence regarding the effectiveness of community-based primary health care in improving maternal, neonatal and child health: 6. strategies used by effective projects

Background As part of our review of the evidence of the effectiveness of community-based primary health care (CBPHC) in improving maternal, neonatal and child health (MNCH), we summarize here the common delivery strategies of projects, programs and field research studies (collectively referred to as projects) that have demonstrated effectiveness in improving child mortality. Other articles in this series address specifically the effects of CBPHC on improving MNCH, while this paper explores the specific strategies used. Methods We screened 12 166 published reports in PubMed of community- based approaches to improving maternal, neonatal and child health in high-mortality, resource-constrained settings from 1950- 2015. A total of 700 assessments, including 148 reports from other publicly available sources (mostly unpublished evaluation reports and books) met the criteria for inclusion and were reviewed using a data extraction form. Here we identify and categorize key strategies used in project implementation. Results Six categories of strategies for program implementation were identified, all of which required working in partnership with communities and health systems: (a) program design and evaluation, (b) community collaboration, (c) education for community-level staff, volunteers, beneficiaries and community members, (d) health systems strengthening, (e) use of community-level workers, and (f) intervention delivery. Four specific strategies for intervention delivery were identified: (a) recognition, referral, and (when possible) treatment of serious childhood illness by mothers and/or trained community agents, (b) routine systematic visitation of all homes, (c) facilitator-led participatory women’s groups, and (d) health service provision at outreach sites by mobile health teams. Conclusions The strategies identified here provide useful starting points for program design in strengthening the effectiveness of CBPHC for improving MNCH.

We conducted a comprehensive review of the effectiveness of community–based primary health care (CBPHC) in improving maternal, newborn and child health (MNCH) by reviewing 12 186 published reports of community–based programs for improving MNCH in low– and middle–income countries. 552 of these reports qualified. An additional 148 reports were identified from the “grey” literature (documents publicly available on the internet) and books. A total of 700 assessments were included in this review. A full description of the search strategy and creation of the database is available elsewhere [11]. Of particular importance for this paper is that a data extraction form was designed to capture as much information as possible in the document containing the project’s assessment that describes the project strategies and what role the community played. We did not attempt to force any strict definition of the term “community” in the analysis of the findings since there was no uniform definition used in the projects or by the reviewers. By strategies we mean the activities that these projects used to make the intervention effective – to plan the project, engage partners (including the community), implement the project, engage in associated activities not directly related to intervention delivery, and evaluate the project. The data extraction forms used to collect information from the assessments were designed to capture the available information regarding strategies used for project implementation. In particular, open–ended descriptions of project implementation were completed by reviewers. A copy of the data extraction form is contained in Online Supplementary Document of the above–mentioned paper [11]. The form allows for open–ended as well as close–ended responses related to strategies and community engagement. Data were extracted from each assessment by two independent reviews and a third reviewer resolved any differences between the first two reviews. The maternal, neonatal and child health database was searched carefully to identify all information that described the strategies that were used by projects. All available evidence in the database regarding strategies for project implementation was reviewed by reviewing all the open–responses individually and summarizing common themes as well as by adding up the number of responses to close–ended questions. We identified six categories of strategies used by the projects in our database: (a) program design and evaluation, (b) community collaboration, (c) education for community–level staff, volunteers, beneficiaries and community members, (d) health systems strengthening, (e) use of community–level volunteers and workers (hereafter referred to as community health workers, or CHWs), and (f) intervention delivery. Table 1 summarizes these strategies. The strategies were not mutually exclusive and most projects used at least several of these strategies and, in fact, some of the strategies fit into several categories (eg, participatory women’s groups). Summary of strategies used by CBPHC projects to improve child health Strategies for project design and evaluation shown in Table 1 often included baseline and endline knowledge, practice and coverage (KPC) population–based household surveys. These made it possible to measure changes in intervention coverage in the program population as well as changes in childhood nutritional status as determined by anthropometry. Oftentimes, community members served as interviewers or collaborators for these surveys. In some projects, Participatory Rural Appraisal (PRA), an approach that incorporates the viewpoints of local people in the planning and management of development projects, was used to guide project planning or evaluation. Various approaches were used to determine the beneficiary population (usually mothers, including pregnant women, and their young children) such as household censuses carried out by the project in collaboration with community members or the development of village rosters of beneficiaries. Sometimes projects included a disease–surveillance component using information provided by community–based workers and communities. Examples are surveillance for acute flaccid paralysis (to identify possible cases of polio) and for other vaccine–preventable diseases such as neonatal tetanus and measles. Some projects measured changes in mortality directly, either through prospective vital events registration as in Care Group projects [12] and in the pioneering CBPHC field project at Gadchiroli, India, conducted by SEARCH [13,14] or through retrospective measurements obtained from maternal birth histories [15,16]. Verbal autopsy methods have been used to assess the leading causes of child deaths in the project area and whether or not the cause of death “structure” has changed over time [17]. Finally, communities have been consulted during the project planning phase as well as at the time of project evaluation. In these circumstances, community members assist with data collection for structured surveys and participate as key informants or participants in focus group discussions. Community engagement takes many forms and is commonly mentioned in the assessments included in our database (Table 1). Village health committees are often formed if they were not previously in existence, and projects work with them in project design, implementation and evaluation. Community leaders, including local religious leaders, are commonly consulted. Communities are often mobilized to participate in health campaigns or to practice key healthy behaviors. Many projects have worked with existing community groups or formed new ones, often women’s groups. Activities that empower women are common forms of community engagement, including education and consciousness raising of women as well as formation and support of women’s microcredit and savings groups. Communities are commonly requested to participate in the selection of CHWs and to provide support to them and participate in their supervision. Finally, in some projects, special activities are geared toward engaging fathers, mothers–in–law, traditional healers and local drug sellers. Finally, though not commonly, projects have engaged communities by sharing surveillance and evaluation results. Noteworthy examples of projects with strong community engagement strategies include mobilization of churches in Mozambique [12] and Nigeria [18] and national mobilization of communities and short–term community workers for national health weeks in Sierra Leone [19]. Assessments of the effectiveness of projects included in our database have adopted many innovative approaches to educating CHWs, beneficiaries, and community members as a whole. Some have used social marketing channels such as radio and posters to convey key messages to the entire community. Others have conveyed health education messages through skits, puppet shows and games that engaged children, mothers, or the entire community. One noteworthy example of this approach is the World Relief child survival project in Cambodia [20,21]. Other approaches involved teaching health education messages to volunteer or paid community workers (who most often are mothers) who then conveyed them to their neighbors at the time of home visits or at meetings of small groups of neighbors. Sometimes projects targeted grandmothers for health education messages since they are respected and influential elders in the community. One particularly innovative educational strategy used in some projects is positive deviance inquiry, usually for addressing childhood undernutrition [22]. With this strategy, mothers of undernourished children in a village learn from the mothers of well–nourished children in the village how they care for their children – not just how they feed them but how they care for them more broadly. Another approach used by some projects is called Care Groups [23], which involves training a small number of master trainers in a project area with a set of health education messages. These trainers each then train another set of trainers who then train another set. Through this “cascade training” approach, large numbers of peer–to–peer counselors can be trained to convey key messages to every household. Many CBPHC projects carried out health system strengthening activities of various sorts. One of the most common was providing mothers and their families with educational messages about warning signs for serious childhood illness or about pregnancy and childbirth for which care should be sought at a health facility. A stronger health system is one in which people seek care appropriately and, when potentially serious conditions are present, prompt care is sought. This is core feature of the approach known as Community–based Integrated Management of Childhood Illness (C–IMCI), utilized in many child survival projects funded by the US Agency for International Development, often with marked expansions of geographic coverage of key child survival interventions. A publication highlighting a number of these projects has been published [24]. Another approach has been to work with communities to establish emergency transport systems to ensure that mothers and children can access the nearest health facility whenever a complication arises and also ensure that the family can obtain transport at a fixed, fair, and affordable price. These referral systems are sometimes linked to insurance schemes whereby families pay small amounts of money on a regular basis, usually during pregnancy, to cover all or most of the cost of such transport if needed. One such approach has been developed by Curamericas for isolated mountainous communities in Guatemala [25,26]. Many projects, while implementing community–based interventions, also engage in activities to strengthen the quality of care provided at primary health care centers or referral hospitals, including the capacity of facilities to accept and care for referrals. This often takes the form of training staff who work there or helping the facility to improve its own stock of drugs and supplies. Other approaches include improving the quality of the community–based health system itself by providing training to CHWs, by strengthening the supervision provided to CHWs, or strengthening the logistics/drug supply system for CHWs. Community–based programs often rely on various types of CHWs – trained volunteers or more formally trained and paid workers who can implement specific interventions aimed at improving MNCH. The projects in our database engaged a broad variety of CHWs. For some projects, the training lasted only a few hours or days while for others CHWs had one year or more of full–time formal training. Some CHWs received only a “per diem” payment for attending a training course or a certificate for their service, while others were formally paid government employees. Some CHWs were volunteers or workers who had been engaged for a specific local project or study while others were part of a national government–run program. Table 2 provides a listing and description of the types of CHWs described by reports in our database. Specific examples of community health workers (CHWs) utilized in community–based primary health care (CBPHC) projects with evidence of effectiveness in improving neonatal and child health Four types of strategies for implementing interventions were: (1) recognition, referral, and (in certain circumstances) treatment of serious childhood illness by mothers and/or CHWs; (2) routine systematic visitation of all homes, (3) facilitator–led participatory women’s groups; and (4) provision of health services at community outreach points by mobile teams from peripheral facilities. The review identified considerable evidence regarding the effectiveness of training and supervising CHWs to teach pregnant women and their families about danger signs during pregnancy and childbirth, during the newborn period, and among sick children [27–29]. CHWs can learn to recognize danger signs and they can teach these to mothers, other caregivers, and family members. Some projects that were effective in improving neonatal and child health also trained and supported CHWs to manage these conditions themselves (or in some cases these CHWs also taught mothers how to treat these conditions). This requires, in addition to proper training, appropriate supervision and logistical support for medications and other supplies [30–33]. The community–based treatment modalities included administration of oral (and in a few cases intramuscular) antibiotics [34], administration of oral rehydration fluids, provision of highly nutritious foods available locally or commercially prepared (known as ready–to–use therapeutic foods, or RUTF), and in some cases provision of micronutrients such as iron, vitamin A and zinc. When community–level workers did not have the capacity to treat children with acute illness, they informed mothers and caretakers that urgent treatment at a referral health facility was needed. A comprehensive manual for community–based diagnosis and treatment of serious childhood illness is available for general use [35]. Integrated community case management (iCCM) for childhood illness is now being scaled up in many countries [36]. Routine systematic visitation of homes makes it possible to identify those in need of basic services and to provide everyone in the program population with essential health education and selected key services, particularly during pregnancy and the early neonatal period. Community–level workers who make home visits are generally able to identify pregnant women and mothers of young children, provide education to them and other family members (especially husbands and mothers–in–law), recognize danger signs during pregnancy and childhood illness, encourage referral when danger signs are present, and provide treatment for certain conditions that can be identified at the time of home visits such as growth faltering, diarrhea, pneumonia, and malaria. Based on current evidence, the World Health Organization and UNICEF recommend that all pregnant women receive two home visits during the prenatal period, one home visit during the first 24 hours after birth, and at least one visit as soon as possible after delivery [37]. Activities that should take place during these visits include the following: education about proper nutrition, promotion of antenatal care, education about danger signs during pregnancy and childbirth, promotion of breastfeeding immediately after birth, prevention of hypothermia, and measurement of the weight of newborns to identify low–birth–weight newborns who need additional home visits. A number of studies have highlighted the difficulties many women face in accessing health facilities due to distance and cost [38]. Home visitation provides an alternative for those without ready access to health facilities. Home visitation is also an effective means of providing counseling about breastfeeding and appropriate complementary feeding, hand washing, prevention and treatment of diarrhea, detection and treatment of childhood pneumonia, and family planning services. There are a number of variations of home visitation strategies using community–level workers, from weekly home visits for providing micronutrients to children [39] to regular monthly visitation of all homes in a program population as part of a more comprehensive approach to delivering basic services to the entire population [40]. Finally, an ongoing program of home visitation provides a foundation of trust and awareness. When children develop signs of serious illness that can be managed by CHWs (such as for pneumonia, diarrhea or malaria), families will be more predisposed to contact the CHW for early and prompt treatment. Participatory women’s groups are led by facilitators with less than two weeks of training who provide the opportunity for further empowerment and education about healthy behaviors, danger signs of serious illness, and proper care of the newborn. These groups may also address issues outside of the health domain that are a priority to the community and that may also have an indirect effect on health (such as income generation activities). These groups may also provide a vehicle for counseling about breastfeeding, birth spacing, infant feeding, hand washing, prevention and treatment of diarrhea, signs of childhood pneumonia, and danger signs during pregnancy and childbirth. Participatory women’s groups also can be effective for assisting mothers to rehabilitate malnourished children detected through growth monitoring. The literature illustrates several effective approaches to facilitating participatory women’s groups, including the use of a participatory action–learning cycle [41,42], formation of Care Groups (10–15 women volunteers who meet with a facilitator (promoter/animator) once a month to learn a key health education message to disseminate to each of the mothers in the 10–15 households surrounding each volunteer) [43,44], and education sessions led by community mobilizers [45]. Provision of services at satellite clinics, including holding outreach immunization sessions, by mobile teams based at health centers is a common means of community–based outreach. These mobile teams may have a vehicle or more likely a motorcycle, bicycle, horse or donkey, or they may even travel by foot. The provision of immunization services by mobile health teams at points beyond a peripheral health facility is now well–developed in many low–income countries [46]. Other examples of services that can be provided through outreach include promotion of and provision of family planning services, basic antenatal care, testing for HIV and syphilis, distribution of insecticide–treated bed nets, distribution of medications to prevent or treat malaria, and growth monitoring to detect cases of childhood malnutrition. One widely implemented variation of this strategy is Child Health Days (or sometimes called Child Health Weeks). Generally occurring twice a year, they usually include some combination of immunization administration, vitamin A supplementation, nutritional monitoring (and referral of malnourished children), and distribution of oral rehydration packets, water–purification tablets, or de–worming tablets [47,48]. Services are provided at peripheral outreach points separate from a health center such as at a school or community building or even under a tree, and home visits are often carried out in addition to reach those mothers and children who did not come to the outreach points. These children are often identified on the basis of previously developed household registers. Table 3 demonstrates which evidence–based child survival interventions can be implemented by which implementation modality. The interventions shown in Table 3 are those which have been identified by the Lives Saved Tool (LiST) for inclusion in program plans for reducing under–5 mortality [49]. A more detailed discussion of these four intervention delivery strategies has been reported elsewhere [50]. Child health interventions with strong evidence of effectiveness through community–based implementation *Outreach of health facility staff includes holding mobile clinics and/or immunization sessions at specified locations outside of health facilities in outlying communities on a regular basis. When program assessments that qualified for the review underwent data extraction, reviewers were asked to describe the degree to which communities were involved in various aspects of the project. Some of the findings are contained in Table 4. These findings demonstrate a high degree of community engagement, both in the maternal as well as the neonatal/child health CBPHC projects. More than three–fourths of the projects trained CHWs and more than one–third engaged communities in the formation or support of community groups as well as in the planning of project activities. 81% of the projects engaged communities in project implementation, and more than half promoted partnerships between the community and the health program, promoted the use of local resources, or promoted community empowerment. Almost half promoted women’s empowerment, one–third promoted leadership in the community, and one–quarter promoted equity. 40% of the projects involved the community in the project evaluation. These findings are highly likely to underestimate the true situation since a large portion of the assessments did not go into this level of detail in describing the community engagement component of the project. Information provided in the assessment was rarely sufficient to provide any deeper understanding of the quality of community engagement or details of how community engagement was actually carried out. Community involvement in the implementation of maternal, neonatal and child health CBPHC projects included in the database

N/A

• Dima Health