Menu
Menu
Menu
Menu
Between 1990 and 2015, Kenya had a 0.9% annual reduction in maternal mortality, one of the lowest reductions globally. This slow decline was linked to the relatively low utilization of delivery services. We designed a mobile phone–enhanced 24-hour transport navigation system coupled with personalized and interactive gestation-based text messages (MAccess) to address maternal child health service utilization. The primary purpose of this analysis is to explore the ways in which pregnant and postnatal women made decisions regarding care-seeking for pregnancy and childbirth services, the processes of getting care from home to the hospital as well their perceptions on how the MAccess intervention affected their pregnancy and childbirth care-seeking and utilization experience. We conducted semistructured, individual interviews with 18 postpartum women. Participants were purposively sampled. Interviews were audiotaped, transcribed, and analyzed using thematic analysis. For participants in this study, all three delays interacted in a complex manner to affect women’s utilization of pregnancy and childbirth services. Even though women were aware of the benefits of skilled birth attendance, other health system factors such as opening hours, or health workers’ attitudes still deterred women from delivering in health facilities. The MAccess innovation was highly acceptable to women throughout pregnancy and childbirth and helped them navigate the complex and layered individual, infrastructural, and health system factors that put them at risk of adverse maternal and newborn outcomes. These findings emphasize that an integrated approach, which addresses all delays simultaneously, is important for reducing perinatal morbidity and mortality.
We conducted a qualitative study using in-depth interviews (IDIs) as our primary data collection method. In this study, because we were interested in the personal experiences of women receiving the MAccess intervention while navigating pre- and postnatal maternal health services, in-depth interviews allowed us to obtain the lived experience of the individual women from their own perspective. The study was carried out in east Rachuonyo within Homa Bay County, which is located in rural western Kenya. Homa Bay County has one of the highest maternal and neonatal mortality rates in the country. Homa Bay’s maternal mortality rate is estimated as 583 and is very high compared with the national average of 488 and Kenya’s least-deprived counties that have rates below 200. The MAccess intervention comprised a two-way SMS messaging system that coordinated the sending of messages between a pregnant woman and staff. Community health workers enrolled pregnant women into the MAcess system during their routine monthly visits the homesteads as part of community maternal newborn health services. Once enrolled, each woman would automatically receive weekly messages describing what she can expect during her specific week of pregnancy. After each message, the woman was prompted to chat via text with a trained health-care worker. This was known as m-convo. The m-convo allowed the health-care worker to follow up with the woman through SMS text messaging until the woman’s questions were resolved. The woman was also able to SMS the word “mHelp” to “call for help” if she had questions, felt ill, had concerns over a danger sign, or was in labor. In the case of illness, a danger sign or labor, a central dispatcher would send a brief SMS survey to all trained motorcycle riders in the woman’s area asking a series of questions about their location and availability. Once a rider had been identified and was on his way, the dispatcher would send an SMS to the woman reassuring her that the transport is on the way and to the woman’s preferred health facility notifying them to prepare for the arrival of the pregnant women. This transport system was innovative in that, unlike the regular “Uber,” it was firmly embedded within the national and Homa Bay county government community health strategy. Within the community health strategy, CHWs are vetted for service to the communities and linked to specific community units. In the same way, motorcycle riders with a valid riders license and resident within a specific community unit applied to the community health committee in charge of community units and underwent a vetting exercise after which they are trained on their role in reducing the second delay, the importance of a quick response regardless of weather or terrain, and on community maternal newborn health. Transport fares were agreed on a community level, and each community unit had 2–3 motorcycle riders. The riders were, thus, people who the community within a specific community unit and health facility workers within the link health facility knew and who continued to participate in the monthly community dialogs. Figure 1 summarizes the transport navigation system diagram. Transport navigation system starting from when the dispatch is notified. This figure appears in color at www.ajtmh.org. We used purposive sampling to recruit women in the study. The trained interviewers approached the women independently to explain the study, answer questions about the study, and guide the eligible and willing participants through the informed consent process. Women were eligible to participate in the study if they were 18 years or older and had received the standard of care and the MAccess intervention. The qualitative methodology precludes a priori sample size estimation; however, for planning of time and finances, we estimated that we would conduct IDIs with approximately 20 women who received the intervention. Emphasis was placed on ensuring that there were participants across a range of sociodemographic characteristics and parity. Trained female field workers interviewed participants in a private location at the discretion of the participant. All interviews were done in English, Dholuo, or Kiswahili. We used a semistructured IDI guide with open-ended questions to elicit reflections on what the women saw as barriers and enablers in accessing the maternal health services, perceived negative or positive impact of the intervention, and comparisons between their previous lived experiences and the current experience while receiving the intervention. The interview guide was developed using practical knowledge of the topic and existing literature. Follow-up and probe questions were, however, guided as much as possible by the participants themselves, and we allowed participants to determine the pace and content of the interview. This approach allowed for new questions to emerge during the course of the interviews, thus giving us a more thorough view and rich narratives into the women’s experiences and perspectives. These interviews lasted approximately 90 minutes and were digitally recorded and transcribed. Informed consent was gained from all respondents before commencement of interviews. This involved an explanation of the purpose of the study, a guarantee of confidentiality relating to the information to be given, and assurance that participation would not have any negative bearing on the availability and provision of health care to them or their families. The interviews were translated into English and transcribed verbatim. Thematic analysis was chosen as our method of choice owing to its flexibility. Thematic analysis is characteristically independent of theory and epistemology and can be applied across a range of theoretical and epistemological approaches as compared with other approaches such as interpretive phenomenology.21 The complete data set was included in the analysis (i.e., the entire transcripts of all 18 interviews). The primary purpose of the analysis was to explore the ways in which pregnant and postnatal women made decisions regarding care-seeking for pregnancy and childbirth services, the processes of getting care from home to the hospital as well as their perceptions on how the MAccess intervention affected their pregnancy and childbirth care-seeking and utilization experience. We used the six steps prescribed by Braun and Clarke22 as a guideline to carry out thematic analysis of the data. Data analysis was performed manually and continuously during the data collection period. We first familiarized ourselves with the data by reading and reading the transcripts and noting the initial ideas. Once familiar with the data, we identified preliminary codes from data that appeared meaningful and interesting. We acknowledge that the codes identified were influenced by background literature and the researchers’ experiences and values.23 We used comparing and contrasting techniques24 to identify and define codes, assign data to different codes, and search for atypical data that did not fit a particular code. This process led to the identification of broad themes from the data. Transcripts were then coded a second time and phrases that represented similar themes were further refined and clustered together into specific themes, which were then defined. Complete text from the identified themes was analyzed. Both reflective discussions and narratives of positive, difficult, and meaningful aspects of care-seeking were analyzed. Finally, relevant data were extracted according to the defined themes, and typical statements were used for citation. Two coders independently did the coding. Discrepancies were discussed with other researchers for feedback until consensus was established.
N/A
