Community-based health programmes: role perceptions and experiences of female peer facilitators in Mumbai's urban slums

Abstract

Community-based initiatives have become a popular approach to addressing the health needs of underserved populations, in both low- and higher-income countries. This article presents findings from a study of female peer facilitators involved in a community-based maternal and newborn health intervention in urban slum areas of Mumbai. Using qualitative methods we explore their role perceptions and experiences. Our findings focus on how the facilitators understand and enact their role in the community setting, how they negotiate relationships and health issues with peer groups, and the influence of credibility. We contextualize this within broader conceptualizations of peer-led health interventions and offer recommendations for similar community-based health initiatives.

Introduction

In recent decades the popularity of community-based initiatives to address the health needs of underserved populations has grown [1, 2]. Community health programmes have been implemented in a variety of settings and have provided a range of services in both low- and higher-income countries. Whereas community health worker (CHW) programmes include disease prevention, treatment and health promotion activities, peer worker interventions tend to focus on providing emotional and moral support, advice and information, and targeted health education [3–5].

A common aim of peer health interventions is behaviour change and increased uptake of health services [6–8]. Health behaviour—and behaviour change—is influenced by social and peer group norms [9]. Peer interventions have, therefore, frequently emphasized the shared characteristics, similar life experiences and mutual understanding of peers and their groups, claiming that these aid communication and increase participation in programme activities [10]. It is the peers’ experiential knowledge and target group representation—rather than formal training—that determine their suitability for peer-led programmes [11].

The results of peer-led health interventions have been mixed. A recent Cochrane review of 43 ‘lay health worker’ programmes concluded that, while they seemed to be effective in increasing child and adult immunization, improving health in sufferers of lung infections and malaria, increasing breastfeeding and decreasing deaths in the elderly, the evidence for other successes was inconclusive [1].

Most reports of peer-led health programmes focus on their feasibility, cost-effectiveness and impact on health outcomes. Relatively few have considered how peer workers conceptualize, experience and give meaning to their role. Since a better understanding of programme implementation can help to explain outcomes [12], we conducted a qualitative study with female facilitators involved in a community-based urban health intervention in slum areas of Mumbai. We wanted to explore their role perceptions and experiences as change agents in the community setting, and the broader question of how much the concepts behind peer-led interventions are manifest at community level.

Method

Context

The Society for Nutrition, Education and Health Action (SNEHA) is an Indian nongovernmental organization whose goal is to improve the health of women and children in Mumbai's urban slums. SNEHA collaborates in the City Initiative for Newborn Health (CINH), which works with public health care providers and users to improve maternal and newborn health in vulnerable communities. One component involves mobilizing community groups to stimulate improvements in maternity care in the home and to encourage appropriate use of health care services. The project's conceptual framework developed from action research conducted in Bolivia [13], experiences in community mobilization in Nepal [14] and appreciative inquiry [15]. Appreciative inquiry builds on the existing strengths and achievements of individuals, groups and communities to encourage positive change in attitudes and behaviour. Its emphasis on participants’ own perceptions, values and understanding of their situation may make it effective among underserved populations [16].

SNEHA's community mobilization intervention provided the context in which we conducted our research. The intervention was initiated in 2006 in 24 vulnerable slum areas across six municipal wards of Mumbai. Its effects on maternal and newborn health outcomes were evaluated in a cluster-randomized controlled trial. Twenty-four locally recruited female facilitators (called sakhis—literally ‘female friend’) were assigned to randomly allocated intervention clusters. The name was chosen by the facilitators to identify themselves to the local population. Their role was to establish and strengthen community groups and to facilitate participatory discussions and experience sharing on pregnancy, childbirth, postpartum care and newborn health (Figure 1). In this context, they were conceived of as both brokers of information and agents of change [17]. The conceptual basis of the intervention was a familiar one: an action research cycle or spiral in which women's groups begin by sharing their experiences, move on to an analysis of problems and achievements, augment their understanding of health issues through discussion and strategic involvement of ‘experts’, prioritize the issues they would like to address, design strategies to address them at a local level, share potential strategies with other community members, roll out their planned activities and evaluate their success (Figure 2).

Fig. 1.

Women's group meeting with Mary Rao, sakhi facilitator. ©SNEHA/Rebecca Sherman 2008.

Fig. 2.

Conceptual basis of the intervention. Note. Meeting numbers are rough advance estimates.

Sakhis were employed on a full-time basis and received a small fixed monthly salary. Recruitment criteria prioritized experiential knowledge and personal qualities over specialized skills and qualifications. Although sakhis had to be literate—preferably with 10 years of schooling—a willingness to learn and build leadership skills was also considered important. Only women were selected, with a preference for those who were married with children. As local residents, sakhis were expected to be familiar with the slum context and have rapport with community women. The nature of their work required acceptance by family members and a degree of autonomy to allow them to travel outside their communities.

Sakhis attended a 12-day induction programme that included the basics of maternal and newborn health, talks by public health professionals, information gathering on the organization of slum communities (bastis), and discussions on local beliefs and practices during pregnancy, delivery and the postpartum period. It also incorporated knowledge-building sessions on group formation and dynamics, rapport-building and facilitation skills, reporting and documentation. Over the course of the intervention, they met weekly with project managers and each other in a rolling process of planning and review.

Data collection and analysis

The basic research design, literature review and analysis were guided by a medical anthropologist of UK origin (G.A.A.), who was based in Mumbai as a long-term member of the SNEHA project team. A local female translator/transcriber was integral to the research, participating in discussions with the SNEHA community mobilization team during data collection and analysis, and acting as a cultural broker between researcher and participants [18]. SNEHA programme staff and University College London academics reviewed and contributed to early drafts of this article.

We conducted a series of focus groups with 23 sakhis (one sakhi left the project before data were collected), each comprising seven or eight participants. We then conducted semi-structured interviews with 12 sakhis purposively selected to ensure a range of sociodemographic characteristics and experience. We observed three women's group meetings and several sakhi weekly meetings, reviewed programme documentation and interviewed programme staff. Verbal consent was obtained after explaining our objectives and giving the sakhis assurances on confidentiality. The average duration of focus groups and interviews was just over an hour. After each activity, a verbatim transcription was made in English. We followed an iterative process, using findings from initial focus groups and interviews to inform subsequent lines of inquiry. A framework approach was used for analysis, whereby emerging themes were identified and categorized. Findings were validated through a feedback session with the sakhis and discussions with programme managers.

Results

We compiled a sakhi profile using selected sociodemographic indicators to compare with women from slum neighbourhoods. Twenty-one sakhis had between one and four children, and two were childless. Their mean age was 34 years. Only the two childless sakhis were not married. Half were Hindu, 6 Buddhist, 4 Muslim and 2 Catholic. All the sakhis lived in slum areas; 7 worked in their own neighbourhoods, whereas 16 worked in nearby slums. Sixteen owned their own homes. Seventeen had metered electricity, whereas 13 had their own water supply. Twenty only had access to common toilet facilities. Eighteen had completed up to 10 years of education, the remainder up to 13 years. With the exception of educational level (programme selection criteria required sakhis to have completed 10 years), these data were representative of the female slum population.

We categorized the qualitative data from focus groups and interviews according to three emergent themes: routine activities, sakhi role perceptions, and role negotiation and credibility.

Routine activities

While the sakhis understood that the aim of the intervention was to improve ‘the health of pregnant mothers and newborn infants’, their role descriptions tended to centre on their routine activities. These included forming women's groups and holding group meetings, which many perceived as a way of ‘bringing women together’ and enabling residents to ‘know each other better’. All sakhis agreed, however, that the main purpose of meetings was to share local knowledge and experiences of pregnancy, childbirth and care-seeking behaviours:

We have to take meetings and explain … for the pregnant woman … what care should be taken. If she is ill, what she should do. For this illness, that illness … [we ask them] ‘What care do you take at home? Any home-based treatment?’ If you do delivery at home, then what is the risk you can face, and if you go to the hospital then what advantage do you get? All this we explain to her in the meeting. (Sakhi interview)

Sakhis also detailed their informal interaction with other community members. They described visiting residents’ homes, establishing and maintaining good relationships with women, and offering them advice on issues that arose. They also described giving practical help, the most common of which were accompanying women to health facilities for treatment or to register for childbirth, and helping to obtain birth certificates and ration cards.

Role perceptions

While none of the sakhis considered themselves to be health educators, a minority noted that they were ‘mostly required to give information’. Although most acknowledged that their level of health knowledge was limited to maternal and newborn health, we noted that community women approached them about a range of health problems:

… we work for newborn infants and pregnant women …. If any other problem arises … malaria … some other illnesses … the women ask about those too! Well, we won't have full knowledge about those diseases … so we cannot provide any information about them. (Sakhi interview)

Sakhis described their role as a resource for accessing information rather than a source of knowledge. When questions could not be resolved within the group, their responsibility was to seek information and relay it back to the community. As a result of the range of health issues that arose, additional training was conducted on mental health, domestic violence, malaria and HIV/AIDS.

Central to the sakhis’ role perception was the notion of ‘being a friend’ to community women. The name and its meaning (female friend) provided them with a collective and social identity. Conceptual and theoretical understandings of identity vary across academic disciplines, most significantly on whether it is a fixed internal phenomenon or is flexible and socially ascribed [19]. Here, we use ‘identity’ in the general sense that sakhis identified themselves through their collective name and that they associated certain cultural characteristics with that name [20]. Friendship was variously translated as treating women well, considering them as equals, listening to them and understanding their problems, and sharing their experiences:

A sakhi is like a friend …. We try to understand their issues, and we also explain to them our own …. So, that's how it is, like a friend. (Sakhi interview)

The sakhis perceived their friendships and interaction with community women on different levels. While a minority saw these more as a function of their role, others likened their relationships to ‘that of a relative’. Some described how women often confided in them rather than their own family, community or other group members. For most sakhis, meaningful relationships with women were formed through friendship and by demonstrating a genuine concern for their well-being. This contrasted with the approach of workers from other organizations:

They stay at the door, and do not engage in friendly conversation with the basti residents. We are not like that. We show an interest in their welfare and wellbeing and show willingness to give them the time to listen to their problems. That's how we've developed good friendships with the group members. (Sakhi interview)

Sakhis made clear comparisons between themselves and women in their target groups. While they acknowledged individual differences in terms of caste, language, ethnicity and religion, they emphasized characteristics and experiences they had in common. For some, the most important characteristic was being a woman. As one pointed out: ‘Whether Hindu or Muslim, these do not make any difference. Women are all alike, no?’

Role negotiation and credibility

We noted different ways in which sakhis negotiated their role from persuading women to join groups and participate in meetings to sharing experiences and discussing health issues. They frequently dealt with reservations from women who, for example, expected payment or medicines in return for attendance at meetings. Some mothers with older children, childless women or those who were not pregnant saw little benefit in attending meetings. Negotiation with these women involved explaining the value of their experience for others or that the knowledge acquired during group discussions could be used to advise on other family members’ health:

I said, ‘You have a daughter. [She] will get married, get pregnant. If she faces any problems then, if you have more knowledge, you can help her …. You can explain [matters] …. Your daughter-in-law will arrive [one day] … you can explain to her too’. (Sakhi interview)

Most sakhis thought that the likelihood that women would accept their advice depended on a level of familiarity and trust. For some, the formation of friendship and strong relationships helped to convince women that they were acting in their best interests:

Friendship comes first. If I try to talk to an unfamiliar person, they will not be able to accept what I say. Once the friendship becomes strong, then whatever we say, the women will think, ‘No, she is right. Whatever she is saying is the truth, it is to my benefit’. (Sakhi interview)

Sakhis used a variety of tools and strategies to negotiate health information. They often guided discussions using examples of their own experiences in combination with ‘new’ knowledge acquired through training. Many sakhis felt that picture cards (a set of locally adapted visual aids depicting common health problems affecting pregnant women, mothers and newborns) helped to illustrate health problems and appropriate care-seeking behaviours. However, they sometimes had difficulty persuading women about unsafe practices during pregnancy or after childbirth. This was especially true when the woman had not experienced problems despite practising the unsafe behaviour. In these cases, negotiation usually included acknowledging the woman's own understanding but also giving alternative explanations. For example, in the following extract, a sakhi recalled her experience with a woman who had associated heavy work during her pregnancy with a normal delivery:

So, I even said to this woman, ‘Fine, things went well with you. It is God's gift that all went well for you. But it is not that all will have the same good fortune’. I explained to her in the same way, ‘That's fine, that you did more work, that's why your delivery was normal but at that time, you must also have eaten well, [taken] fruits and so on. You must have taken good care of yourself, or you must also have rested as much. That's why your delivery was normal’. (Sakhi interview)

We asked the sakhis about their perceptions of credibility. Some felt that age was relevant as older women in India generally command greater respect. Others said that having children allowed them to discuss their personal experiences. Individual qualities like friendship and perseverance were also important so that, ‘if any sakhi comes along and builds a good relationship, then she too will get a good response’. Knowledge of health issues was another possibility, although younger sakhis placed more value on the knowledge of experienced women:

In that group there is a woman. She has 3 children, 2 sons and a daughter, so she has a lot of knowledge! She told us in full detail, that, ‘In my first delivery such-and-such happened. Then, I had gone to the village’. She narrates all her experiences in full detail. Then full 2 hours are spent in just that! She is the one who gives more information to the women! And then the discussions too go on well. (Sakhi interview)

We noted other factors that may have contributed to the sakhis’ credibility. Those who lived in different communities pointed out that new group members often thought they worked at health facilities. Picture cards also seemed to have a symbolic value, as they backed up the sakhis’ health knowledge and training, and suggested that they belonged to a credible organization:

Because of the cards, we have gained a lot! On seeing the cards, the people feel more …. If they initially had no interest, their interest grows in discussions. They feel that ‘Yes, these sakhis must be coming from some credible place’. (Sakhi interview)

Finally, sakhis reported a number of positive changes in their self-image as a result of participating in the intervention. Many described having initial doubts about their ability to learn new knowledge, gain the trust and respect of other women, and communicate clearly and confidently. In contrast, their present self-perceptions included feelings of increased confidence and self-esteem, greater autonomy in decision making and mobility, improved communication skills, greater knowledge of health issues and, to an extent, improved status in the home.

Discussion

We have explored perceptions of women health facilitators involved in a community-based peer health intervention in urban slum communities of Mumbai. These included routine activities, role perceptions, role negotiation and credibility. Central to the sakhis’ role perceptions were friendship and familiar relationships with women. Negotiation was identified as a complex but integral aspect of their role, which also seemed to be influenced by their credibility.

This study was conducted towards the end of the community mobilization intervention. We investigated the sakhis’ present role perceptions and did not document how these may have developed with the evolution of the intervention. Although the sakhis participated in the design of intervention strategies, our research focused on their role in the community setting. Limitations during data collection included the possibility that the sakhis’ responses were influenced by their unfamiliarity with members of the research team or from the desire to give us a good impression of their work (which could have been influenced by the fact that they were salaried). Neither can we fully discount the possibility that some responses were biased by recounting the contents of training rather than describing their personal experiences. To minimize this, we validated responses by cross-checking interview data and by observing the sakhis in their work. It is also important to acknowledge the existence of cultural differences between researchers, translator/transcriber, programme staff and sakhis, and their potential impact on the collection, interpretation and reporting of data. Translation and transcription are susceptible to misinterpretation, loss of meaning and the influence of the translator's own assumptions and values [18, 21, 22]. Qualitative researchers are also vulnerable to ‘selective reporting’ [23]. Therefore, ensuring the relevance of lines of inquiry prior to data collection and discussing findings with the different actors were key components of our research.

Our study highlighted the multiple ways that peer facilitators perceived and enacted their roles: as community mobilizers, women's group facilitators, brokers of health information and friends. The sakhis’ descriptions initially focused on their routine activities (forming groups, holding meetings and interacting with the community). However, role perceptions were influenced by notions of friendship and the importance of establishing positive relationships with community members. We found that sakhis negotiated their role on a number of levels, from forming friendships, through encouraging participation, to discussing health issues. Perceptions of peer credibility were ambiguous; sakhis were unsure of what form this might take, or whether it had any relevance. Finally, they said that their self-confidence, autonomy, communication, knowledge of health issues and social status had increased as a result of their participation in the intervention. Similar findings have been reported by other peer programmes [5, 24].

Research suggests that successful peer interventions involve meaningful relationships between peers and their groups [5] and that some peer health workers consider friendship an important aspect of their role [25]. Our findings support these assertions. Friendship was a central part of the sakhis’ role perceptions in a number of ways. It informed their sense of identity, it was considered an important personal quality that was reflected in their attitude and behaviour, and sakhis defined their relationships with women in terms of friendship and familiarity. This was important, not least because they relied on relationships of trust to negotiate health issues: ‘Only when they trust us, have faith in us, only then they will listen to us’. However, in some contexts, relationships and familiarity between peers and their groups might act as barriers to communication. For example, a peer-led HIV intervention in the UK reported that gay men were often unwilling to discuss sex with peer educators they already knew [26].

Understanding how peers negotiate new or innovative ideas is important for programme implementation. Few studies allow us to compare our experiences with those of other interventions. Patel et al. [27] identified several strategies used by adolescents during peer group discussions of HIV-related risks and decision making. These included the justification of one's own opinions coupled with the elaboration of alternative opinions, the use of concrete examples from shared knowledge to relate these alternative opinions to the real world and the clarification of knowledge through further negotiation [27]. Our findings showed that similar strategies were used between sakhis and community women, although successful negotiation might also depend on peer relationships, direct experience of pregnancy and childbirth, and credibility.

A number of theories have claimed that credible peers can influence health behaviour change. For example, social learning theory and diffusion of innovations theory assert that higher status and competence make peers credible, which, in turn, enhances their ability to influence other members of their peer group [24]. Others emphasize the socio-cultural similarities and common life experiences of peers [11, 28]. In our study, sakhis identified themselves with other slum women, implying that they did not have a higher status. However, our analysis suggested factors that may have differentiated them. One was the misperception by new group members that the sakhi worked at a health facility; others included being full-time salaried workers, their comparatively greater mobility, autonomy, confidence and communication skills. Finally, there seemed to be some value attached to their use of picture cards and visits by programme staff, which may have led community members to associate the sakhis with a credible programme or organization. The positive influence of such associations on peer credibility has been reported [29]. In addition, we propose that picture cards may carry a symbolic value that enhances the sakhis’ source credibility, a concept developed by Hovland et al., among others, in which two primary components—the level of ‘expertise’ one is perceived to posses about a subject and the ‘trustworthiness’ and ‘reliability’ of an information source—influence listeners’ perceptions of a communicator [30–32]. We propose that, in a context such as this, successful programme implementation requires peers to strike a balance between factors that align them with their peer groups and those that differentiate them. Clearly, this is an area that merits further research.

Programme design requires the translation of broad conceptual understandings into locally implementable strategies. Within the health systems literature, the modification of programmes between plan and deed has been called ‘street-level bureaucracy’ [33]. Our findings suggest that, during the transition from planning to implementation, a localization of these broader understandings and long-term goals takes place, whereby community workers translate them into more concrete short-term strategies and activities that are relatively concept free. Within a programme that was heavily conceptually driven and whose implementers had been extensively involved in discussions about the nature of the intervention, it was noteworthy how much the sakhis focused on local exigencies over programme aims. We speculate that this tendency would be amplified under conditions of a large-scale rollout because of the reduced opportunity for charismatic leadership and intimacy among sakhis.

We think that our findings have five major implications for health programming that involves peer workers. (i) Programme managers need to reflect on the nature of relationships between peer workers and their groups and how these may enhance (or hinder) the implementation of programme strategies. (ii) Programme staff can benefit from understanding the role perceptions of peer leaders and should also consider the perceptions and expectations of intervention recipients. Clarification of misperceptions—especially in situations of changing peer group membership—may lead to greater acceptance and participation by target groups. (iii) The conceptual frameworks for community-based programmes should be based on empirical (street level) evidence as well as theory. (iv) Organizers of peer training programmes should emphasize rapport-building, communication and negotiation skills. (v) Programme managers should consider the ways in which credibility may be constructed, in the recruitment, training and supervision of peer workers.

Funding

ICICI Centre for Child Health and Nutrition; the Wellcome Trust (081052).

Conflict of interest statement

None declared.