Talking health: Trusted health messengers and effective ways of delivering health messages for rural mothers in Southwest Ethiopia

Summary: Background: Access to trusted health information has contribution to improve maternal and child health outcomes. However, limited research to date has explored the perceptions of communities regarding credible messenger and messaging in rural Ethiopia. Therefore, this study aimed to explore sources of trusted maternal health information and preferences for the mode of delivery of health information in Jimma Zone, Ethiopia; to inform safe motherhood implementation research project interventions. Method: An exploratory qualitative study was conducted in three districts of Jimma Zone, southwest of Ethiopia, in 2016. Twelve focus group discussions (FGDs) and twenty-four in-depth interviews (IDIs) were conducted among purposively selected study participants. FGDs and IDIs were conducted in the local language, and digital voice recordings were transcribed into English. All transcripts were read comprehensively, and a code book was developed to guide thematic analysis. Data were analyzed using Atlas.7.0.71 software. Result: Study Participants identified as Health Extension Workers (HEWs) and Health Development Army (HDA) as trusted health messengers. Regarding communication channels, participants primarily favored face-to-face/interpersonal communication channels, followed by mass media and traditional approaches like community conversation, traditional songs and role play. In particular, the HEW home-to-home outreach program for health communication helped them to build trusting relationships with community members; However, HEWs felt the program was not adequately supported by the government. Conclusion: Health knowledge transfer success depends on trusted messengers and adaptable modes. The findings of this study suggest that HEWs are a credible messenger for health messaging in rural Ethiopia, especially when using an interpersonal message delivery approach. Therefore, government initiatives should strengthen the existing health extension packages by providing in-service and refresher training to health extension workers. The study was conducted in Jimma Zone, located in the southwest region of Ethiopia. Jimma town is 346 km from Addis Ababa, the capital of Ethiopia. The Zone is known for its production of “Coffee Arabica” which is the back bone of the country’s economy. Jimma town has located a latitude and longitude of 7°40′N 36°50′E. Jimma Zone has 21 districts. Among these, three districts were purposively selected for this project (Gomma, Seka Chekorsa and Kersa) by considering of high population size and low health service utilization. The number of health centers, health posts, HEWs and Health development army leaders in the selected three districts were 28, 110,231 and 3384 respectively [8]. The study was conducted in May and June 2016. This research used an exploratory qualitative case study to better understand participants’ understanding of trusted maternal health information sources and their preferred communication channels. For purposes of this study, trusted messengers were defined as persons regarded by community members as credible sources of health information for pregnant women in making informed decisions regarding health maintenance and health seeking behavior. Preferred communication channels described the different means or venues in which community members liked to receive IEC messaging. These preferences were assessed by in-depth interviews (IDI) and focus group discussion (FGDs). The FGDs and IDIs were considered to create an appropriate context for the researchers to explore community member feelings, perceptions, and understanding of the maternal/child health (MCH) topics that would form part of the IEC intervention. Data collection involved IDI and FGDs. Six individual depth-interview were conducted for each of the following stakeholder groups: religious leaders, health extension workers (HEWs), and members of the Women Development Army (WDA) and Male Development Army (MDA)1(total N = 24). Six FGDs were held among female community members and another six with male community members (see Appendices 1–6 for a detailed sampling overview). Researchers developed IDI and FGD guides based on a review of literatures in the areas of health communication, message development, and message delivery approaches, with a focus on improving maternal and child health outcomes associated with pregnancy and childbirth [9–11]. These instruments were tailored to explore the different experiences, perceptions, and roles of each group of study participants (HEW, WDA, MDA, religious leaders, male community members, and female community members). Each guide contains questions on basic socio-demographic variables like age, sex, educational status and role in the community, as well as questions and probes on existing and preferred IEC programs related to improving MCH outcomes in rural contexts. Written and/or oral consent was obtained from all participants. IDIs and FGDs were conducted in convenient, quiet and private locations in order to ensure confidentiality. Data were collected using digital audio recorders and field note memos were taken to document non-verbal or other behaviors observed during the data collection. Audiotapes and notes were transcribed following data collection. Atlas.ti 7.0.71 software was used for data analysis. A rough outline of thematic categories was drafted based on the themes that were specified in the FGDs and IDI guidelines. This outline was further developed into a preliminary code guide. Next, all FGD and IDI transcripts were read multiple times by the research team to produce a final code guide. To enhance inter-coder reliability, the coders independently applied the guide to all transcripts; discrepancies were reviewed and resolved. This exercise ensured that the coders had a common understanding of the code guide and its application. Exemplary quotations for selected codes were generated in Atlas.ti software using the Code Manager/Output feature. The summaries presented below reflect both widely-expressed ideas, as well as novel ideas that were mentioned less frequently. To promote credibility, data collection tools were pretested in similar contexts to maximize the validity of the tool. The IDI and FGD questions were open-ended and participants were encouraged to discuss the questions in an uninhibited manner while being guided to remain focused on the topic of interest. To promote transferability, appropriate probes were used to obtain detailed information on responses. Detailed field notes were taken, and all interviews were digitally recorded (thick description). To address confirmability, this study employed reflectivity and bracketing methods. These methods helped to minimize respondent bias and the risk of reactivity whereby participants could deny information due to the presence of data collectors and researcher, while “bracketing out” daily debriefing sessions provided an opportunity for the data collectors and researchers to explore how their own preconceived ideas might be affecting the study, and to increase their reflexivity in later interpreting the findings. To promote dependability, all data collectors were bilingual (fluent in both English and the local languages) and trained at the post-graduate level. They also had prior experience in qualitative data collection, and had participated in an intensive, week-long training program prior to undertaking field research. Interviews were conducted until data saturation was reached, within the limits imposed by geography and time-frame for the study. The duration of the interviews and discussions ranged from 45 to 90 min (prolonged engagement).