Promoting and delivering antenatal care in rural Jimma Zone, Ethiopia: A qualitative analysis of midwives’ perceptions

Background: Despite improvements in recent years, Ethiopia faces a high burden of maternal morbidity and mortality. Antenatal care (ANC) may reduce maternal morbidity and mortality through the detection of pregnancy-related complications, and increased health facility-based deliveries. Midwives and community-based Health Extension Workers (HEWs) collaborate to promote and deliver ANC to women in these communities, but little research has been conducted on the professional working relationships between these two health providers. This study aims to generate a better understanding of the strength and quality of professional interaction between these two key actors, which is instrumental in improving healthcare performance, and thereby community health outcomes. Methods: We conducted eleven in-depth interviews with midwives from three rural districts within Jimma Zone, Ethiopia (Gomma, Kersa, and Seka Chekorsa) as a part of the larger Safe Motherhood Project. Interviews explored midwives’ perceptions of strengths and weaknesses in ANC provision, with a focus as well on their engagement with HEWs. Thematic content analysis using Atlas.ti software was used to analyse the data using an inductive approach. Results: Midwives interacted with HEWs throughout three key aspects of ANC promotion and delivery: health promotion, community outreach, and provision of ANC services to women at the health centre and health posts. While HEWs had a larger role in promoting ANC services in the community, midwives functioned in a supervisory capacity and provided more clinical aspects of care. Midwives’ ability to work with HEWs was hindered by shortages in human, material and financial resources, as well as infrastructure and training deficits. Nevertheless, midwives felt that closer collaboration with HEWs was worthwhile to enhance service provision. Improved communication channels, more professional training opportunities and better-defined roles and responsibilities were identified as ways to strengthen midwives’ working relationships with HEWs. Conclusion: Enhancing the collaborative interactions between midwives and HEWs is important to increase the reach and impact of ANC services and improve maternal, newborn and child health outcomes more broadly. Steps to recognize and support this working relationship require multipronged approaches to address imminent training, resource and infrastructure deficits, as well as broader health system strengthening. Evidence for this article is derived from the larger Safe Motherhood Project, a mixed-methods, randomized cluster intervention trial testing the roll out and scale up of interventions to improve access to health facilities for pregnant women and reduce preventable maternal and neonatal morbidity and mortality. The study focuses on two interventions: the delivery of information, education, and communication workshops in conjunction with key community leaders; and improving the functionality of maternal waiting areas (residences nearby health facilities for women nearing their delivery date, allowing for close monitoring and rapid birth attendance by a skilled midwife [25]). This research is being conducted in Jimma Zone, Ethiopia, across 24 PHCUs in three rural districts: Gomma, Kersa, and Seka Chekorsa. There are 112 kebeles in the study area (41 in Gomma, 32 in Kersa and 38 in Seka Chekorsa). As a part of the larger Safe Motherhood Project baseline qualitative assessment, we conducted semi-structured, in-depth interviews with midwives, HEWs, PHCU directors, male and female community leaders, and religious leaders to explore various aspects of MNCH in the community. The interviews included questions about: MNCH service delivery and use; stakeholder roles in promoting MNCH; community participation in MNCH activities; problems related to MNCH; and maternity waiting areas. This paper focuses on data collected from 11 midwives, employed at PHCUs across Gomma (n = 5), Kersa (n = 2) and Seka Chekorsa (n = 4) districts. Findings derived from the HEW interviews (n = 31) have been reported elsewhere [27]. The interview guide used in this study was developed for the Safe Motherhood Project (Additional file 2). For interviews with midwives, six random PHCUs for each intervention arm were selected using the Wolfram Alpha software. This software generated sets of six random and different integers, which were matched with PHCU listing numbers to randomly select PHCUs in each arm for midwife interviews. In total, 18 interviews were to be conducted with PHCU midwives in accordance with the sampling frame. Data collectors were able to conduct 11 of these interviews, achieving 61% of their target. The interviews took place in selected areas in each respective PHCU where the midwives worked. The baseline qualitative data were collected over a four-month period (November 2016 – February 2017), working with data collectors from Jimma University. (A parallel baseline quantitative survey was also administered, but is not reported on in this article.) The data collectors were fluent in both English and the local language (Afan Oromo), had graduate-level university training, and had previous experience doing qualitative research. The data collectors participated in a one-week training and induction workshop prior to the field work, which focused on techniques to conduct interviews and keep field notes. At least two researchers from the Safe Motherhood Project were present in a supervisory capacity during the field work to address any issues that arose. A member of the research team obtained informed written or oral consent from all study participants. With the permission of the participants, interviews were digitally recorded. The data collectors transcribed and translated the interviews. Detailed field notes were taken by the data collectors and their supervisors to contextualize and supplement the interview data. Following detailed readings of the transcripts, a code guide was developed for analysis of all the qualitative baseline data, and continually refined to allow for emergent themes. Data analysis was undertaken by members of the research team, using Atlas.ti software. After summarizing descriptive findings according to the coding categories, major themes and findings from the midwives’ transcripts were identified and discussed by the research team. This inductive approach to analysis facilitated new insights into the data, such as the highly significant and focal aspect of midwife-HEW interaction and engagement surrounding the provision of ANC services (reported in this article). We adopt a simple health systems analysis framing construct to present the results of the analysis (context – inputs – process – outputs) [28]. First, highlighting context, we begin with midwives’ descriptions of their and HEWs’ extant roles across the three stages of ANC promotion and delivery: health promotion, community outreach and ANC visits. Then, turning to midwife-HEW interactions, we present midwives’ perceptions of the strengths and weaknesses related to the inputs, processes and outputs of ANC delivery. Ethical clearance was obtained from the authors’ respective institutes: Jimma University College of Health Sciences Institutional Review Board, and University of Ottawa Health Sciences and Science Research Ethics Board.