Background: Having a birth attendant with midwifery skills during childbirth is an effective intervention to reduce maternal and early neonatal morbidity and mortality. Nevertheless, many women in Ethiopia still deliver a baby at home. The current study aimed at exploring and describing reasons why women do not use skilled delivery care in North West Ethiopia. Methods: This descriptive explorative qualitative research was done in two districts of West Gojjam Zone in North West Ethiopia. Fourteen focus group discussions (FGDs) were conducted with pregnant women and mothers who delivered within one year. An inductive thematic analysis approach was employed to analyse the qualitative data. The data analysis adhered to reading, coding, displaying, reducing, and interpreting data analysis steps. Results: Two major themes client-related factors and health system-related factors emerged. Factors that emerged within the major theme of client-related were socio-cultural factors, fear of health facility childbirth, the nature of labour, lack of antenatal care (ANC) during pregnancy, lack of health facility childbirth experience, low knowledge and poor early care-seeking behaviour. Under the major theme of health system-related factors, the sub-themes that emerged were low quality of service, lack of respectful care, and inaccessibility of health facility. Conclusions: This study identified a myriad of supply-side and client-related factors as reasons given by pregnant women, for not giving birth in health institution. These factors should be redressed by considering the specific supply-side and community perspectives. The results of this study provide evidence that could help policymakers to develop strategies to address barriers identified, and improve utilisation of skilled delivery service.
This study was part of a bigger research project by the researchers (Biruhtesfa Bekele and LM Modiba) and details of the research methods have been published elsewhere [22]. The research was done in two rural districts (Womberema and Burie Zuria) of the Amhara region in North West Ethiopia. Primarily, the Amhara regional state was selected for this study because of the low coverage of skilled delivery service (27.1%) [2]. By selecting this region, which was identified as having low performance in skilled delivery service, the study intended to inform designing of strategies that will help in improving utilization of skilled delivery care. By doing so, an in-depth understanding of reasons for poor or no utilization of skilled delivery care, was critical. As data acquired from regional routine health management information system (HMIS) reports evidenced, there was a variation in coverage of skilled delivery performance among health institutions and district health offices in the study region. Though several districts had low performance, a few of them were performing well. Our study focused on the districts with low performance in skilled delivery care. Well-performing districts in skilled delivery care were excluded from the study, because this research was not a comparison study. Each of the study districts comprised of 20 rural kebeles, 4 health centers, and 20 health posts. We purposively selected 7 kebeles for this study, 4 of them were from Burie Zuria District and the remaining three kebeles were from Womberema district. The detail of the sampling of the research sites is portrayed in Fig. 1. Sampling procedure of the study sites A qualitative descriptive explorative study design was employed to explore and describe why pregnant women in North West Ethiopia do not use skilled delivery service. A descriptive qualitative research was done because the phenomenon of interest in the current study has been well-defined, and because of the need to describe the subject of study accurately and present a detailed picture or accounts of the phenomenon of interest [23]. In view of this, the coverage of utilization of skilled delivery care in the study districts and region has been well known and many quantitative researches have been conducted on the subject under study. Therefore, this descriptive qualitative research was done aimed at describing and presenting a detailed accounts or pictures on why mothers do not use skilled delivery care in the study areas. To capture reasons for home childbirth, data were collected from pregnant women, and mothers who delivered in the past one year. The research participants were purposively included in the study if they had previously given birth once or more at home, in health institution or both. The study subjects were identified through health extension workers (HEWs), who work in the selected kebeles and serve the community by providing basic promotive, preventive, and selected curative services. We also corroborated whether the selected participants fulfilled the inclusion criteria or not. Before embarking on the actual data collection, we established a good rapport with the study subjects, and this enabled us to win their trust which helped to obtain the information needed. The researchers also asked the participants a series of follow-up and probing questions, after posing the main question. This helped to take the discussions to a deeper level and obtain required information. Two experienced qualitative data collectors who were graduates of health science and social science conducted the data collection. A focus group discussion was used to capture the data. The researchers prepared and used a semi-structured FGD guide to guide the group discussions. We conducted pretesting of the preliminary FGD guides with each of the research participants who were excluded from the actual data collection. This helped in estimating time required to conduct the interviews and FGDs, to refine the interview guides and questions, to check appropriateness of the data capturing procedures and to familiarise the researcher with the data recording equipment (audiotape recorder). All the FGDs were conducted in Amharic, which is the official business language of Ethiopia and is widely spoken in the region. The data collectors conducted the FGDs in the compounds of the health posts because of health posts’ accessibility to research participants and avoid any disturbance from non-participants because the health posts had a fence. We made sure that the HEWs were not around while the FGDs were underway. The FGDs lasted between 60 to 90 min. The data collectors took field notes during FGDs, expanded notes after each FGD sessions, and shared with researchers and data collectors. All the FGDs were digitally recorded with participants’ consent using two audiotape recorders, one was used as a backup, in case the other audio tape recorder failed. We conducted FGDs separately with pregnant women (7 FGDs) and mothers who delivered within 12 months (7 FGDs). The researcher recognised that no new idea or insight emerged after conducting five FGDs with pregnant women and five with mothers who delivered a child within one year. This revealed data saturation and was confirmed as such, with a final additional of two FGDs with pregnant women and two with mothers who delivered a child within one year. Those two final FGDs were added only for sake of confirmation of point of saturation; otherwise, it had no any relevance. An inductive thematic analysis approach was employed to analyse the qualitative data. The translated data were exported onto Atlas ti version 7 software to efficiently store, organise, manage and reconfigure the data to enable human analytic reflection. The current study adhered to data analysis steps which included reading, coding, displaying, reducing, and interpreting. The data analysis was initiated in the field before completion of data collection. The researcher listened to the audio files and read the expanded field notes and transcripts after each FGD session is completed and the transcribed data were ready to use. The audiotape records of the FGDs were transcribed and the data collectors prepared the interview transcripts for analysis expanded the field notes. The researcher translated the Amharic transcripts directly into English. To ensure accuracy of the translation, a colleague of the researcher, who has high level of proficiency in both English and Amharic language, checked the consistency of the translation. The engagement of the researcher in the translation and partly in the transcription of the interviews helped to familiarise and acquaint himself with concepts. Moreover, the whole process of analysing the data in advance helped to make necessary revisions and refinements before subsequent FGD sessions took place. The transcribed data (transcripts) were imported to the Atlas ti version 7 data analysis software as a primary document, using the assign command in the main menu of the Atlas ti. The researcher labelled the coded texts with words that explain the text description. The techniques employed for coding were open coding, quick coding and coding by list. The researcher used the open coding command when coding for the first time. Coding by list helped to assign existing codes to a selection and quick coding was employed to assign currently selected codes to consecutive text segments. The codes were then examined to form categories and sub-categories. The study explored patterns of categories to discover emerging themes. The study credibility, where the results of the research are closely related to reality, was achieved through prolonged engagement, triangulation, peer debriefing, and member check. Prolonged engagement involves establishing adequate contact with the participants and the context with the objective of acquiring data the researchers needs. We spent adequate period of time in the study areas with the research participants in FDGs and this enabled the researchers to acquire adequate understanding of the contexts and to establish rapport with the participants. We also collected separate data from pregnant women and mothers who delivered a child recently, who have different childbirth experiences, to conduct data triangulation. Besides, peer debriefing was held with colleagues from universities who were experienced in qualitative research, and presentation of the results and interpretations of the data were made. This helped to avoid bias and misinterpretation of the data and unfolded aspects of the research that remained covert. Member check was also used to ensure the study credibility in which study participants were allowed to validate whether the researchers’ interpretations were a good representation of the participants’ realities. To conduct this study, ethical clearance was secured from the UNISA Department of Health Studies Higher Degrees Committee and Amhara Regional Health Bureau Research and Laboratory Department. The research got letters of support from regional health bureaus and zonal health departments to get access to the study sites. Written informed consent was acquired from participants who could read and write and fingerprints were used for participants unable to read and write. Confidentiality was ensured by removing all names and addresses of participants from the data collecting tools. The information that the participants provided was also kept confidential and used only for the research. Only codes were used to identify participants. Data collected were kept in the strictest confidence; they were not made public. The audio files, saved on memory cards, were also erased after the completion of the research. Only aggregated demographic information was reported to maintain anonymity.
N/A