Understanding the local context and its possible influences on shaping, implementing and running social accountability initiatives for maternal health services in rural Democratic Republic of the Congo: a contextual factor analysis

Background: Social accountability has to be configured according to the context in which it operates. This paper aimed to identify local contextual factors in two health zones in the Democratic Republic of the Congo and discuss their possible influences on shaping, implementing and running social accountability initiatives. Methods: Data on local socio-cultural characteristics, the governance context, and socio-economic conditions related to social accountability enabling factors were collected in the two health zones using semi-structured interviews and document reviews, and were analyzed using thematic analysis. Results: The contexts of the two health zones were similar and characterized by the existence of several community groups, similarly structured and using similar decision-making processes. They were not involved in the health sector’s activities and had no link with the health committee, even though they acknowledged its existence. They were not networked as they focused on their own activities and did not have enough capacity in terms of social mobilization or exerting pressure on public authorities or providers. Women were not perceived as marginalized as they often occupied other positions in the community besides carrying out domestic tasks and participated in community groups. However, they were still subject to the local male dominance culture, which restrains their involvement in decision-making, as they tend to be less educated, unemployed and suffer from a lack of resources or specific skills. The socio-economic context is characterized by subsistence activities and a low employment rate, which limits the community members’ incomes and increases their dependence on external support. The governance context was characterized by imperfect implementation of political decentralization. Community groups advocating community rights are identified as “political” and are not welcomed. The community groups seemed not to be interested in the health center’s information and had no access to media as it is non-existent. Conclusions: The local contexts in the two health zones seemed not to be supportive of the operation of social accountability initiatives. However, they offer starting points for social accountability initiatives if better use is made of existing contextual factors, for instance by making community groups work together and improving their capacities in terms of knowledge and information. A multiple case-study approach was employed to identify local contextual factors and discuss their possible influences on shaping, implementing and running social accountability initiatives at local level using qualitative research methods. It was conducted from May to June 2013 in two health zones (HZ) of DRC, the Muanda HZ (Kongo Central) and the Bolenge HZ (Equateur). These HZ were purposively selected. The case study inclusion criteria were: 1) health zone in post-conflict situation currently involved in sustainable development activities; and 2) the presence of health sector partners implementing or planning to implement health interventions including social accountability components for more than 4 years, targeting amongst others the improvement of maternal health. Details of the selected HZ are described in Table 1. Essential contexts indicators of selected study health zones An initial exploratory discussion were held separately with HZ officers and main community leaders to map out key community actors involved in maternal health at the local level, from which a representative sample was purposively selected to participate in the interviews. Among these community actors included public officers such as health services providers, political and administrative authorities, HZ authorities, and community representatives such as community leaders, community group members, women groups members, health committee members, and community health workers. The project managers of the NGO projects in both HZ were also included in the sample. Participants were purposively selected using maximum variation and identified from the pool of actors listed above. Selection was based on gender, age, involvement at community level activities in relation to health or other administrative functions. The selected individual were then approached through community health workers (CHWs) or HZ officers in-charge of community activities to participate in the interviews. No contacted individual refused to participate. The interview guides were based on a conceptual model built on the framework and key concepts from Thindwa et al. (2003) enriched by those drawn from Marston et al. [33], McCoy et al. [37], Bukenya et al. [34], and Lodenstein et al. [32]. The framework from Thindwa et al. distinguishes four contextual factors that can enable or constrain the capacity of community members to engage in community development activities at the national and local levels in a sustained and effective manner. These factors are “the legal and regulatory framework; the political and governance context; socio-cultural characteristics; and economic conditions”. They in turn influence the “enabling elements” which are: “the freedom of citizens to associate (Association); their ability to mobilize resources to fulfill the objectives of their organizations (Resources); their ability to voice i.e. formulate, articulate and convey opinion collectively (Voice); their access to information, necessary for their ability to exercise voice, engage in negotiation and gain access to resources (Information); and the existence of spaces and rules of engagement for negotiation and public debate” (Negotiation). In this study, we put together the legal and regulatory framework with the political and governance context, and we extend the concept of resources beyond financial ones. We used this framework to explore if the context in the selected districts in DRC is enabling the shaping and implementation of social accountability interventions/mechanisms. Some variables related to community participation drawn from Marston et al. [33], McCoy et al. [37], Bukenya et al. [34], and Lodenstein et al. [32] were used to further operationalize the main factors in the framework, such as societal values, status of women, health committee recognition by the community and its interface role. The interview guides were adapted, pretested, and validated for the DRC local settings and for maternal health by the study team (see Table 2). Local contextual factors analysis conceptual model Data were collected through individual semi-structured interviews and a document review. At each study site the research team interviewed selected actors. Face-to-face interviews were held in a quiet place away from other people to optimize privacy, and lasted 35 min on average. They were conducted in French or Lingala, and tape-recorded with the participants’ permission. There were no follow-up interviews as these were single-round interview discussions. A documentary review was used to collect information on the health center’s activities, community groups’ activities, and socio-economic, political, and demographic data using a data collection form. Documents reviewed included the health center’s annual reports, health projects’ annual reports, health committee’s monthly reports, and some national policy documents. Recorded in-depth interviews were transcribed verbatim. The interviewers proofread the transcribed work to cross-check accuracy of content since the interview transcripts were not returned for participant check and comment. The interview transcripts and data extracted from the documents were analyzed using the thematic approach [38], based on our context analysis conceptual model. A coding plan was developed using data from the first three interview transcripts and the core concepts of the conceptual model. Two members of the research team read and re-read each transcript thoroughly and assigned codes to each section of the text. Data processing was performed using Atlas-ti 6.1.1© software (ATLAS-ti GmbH, Berlin). Thematic analysis was performed to build a common and comprehensive understanding of the local context with respect to themes expressed by community members, triangulated by those coming from providers and public officers and the document review. Four steps were taken to enhance the credibility of the study: the research team received training in interview techniques, the interview guides were pre-tested and adapted accordingly; the results and interpretations were critically discussed by the research team and shared with local health partners and participants. The interview guides were written in French, translated into Lingala, and translated back into French.