Exploring underutilization of skilled maternal healthcare in rural Edo, Nigeria: A qualitative study

Introduction Existing studies have acknowledged the underutilization of skilled maternal healthcare services among women in rural Nigeria. Consequently, women in rural areas face a disproportionate risk of poor health outcomes including maternal morbidity and mortality. Addressing the challenge of non-use of skilled maternal healthcare in rural areas necessitates the involvement of multi-stakeholders across different sectors who have vital roles to play in improving maternal health. This study explores the factors contributing to the non-use of maternal healthcare services in rural areas of Edo, Nigeria from the perspectives of community elders and policymakers. Methods In this qualitative study, data were collected through 10 community conversations (group discussions) with community elders each consisting of 12 to 21 participants, and six key informant interviews with policymakers in rural areas of Edo State, Nigeria. Participants were purposefully selected. Conversations and interviews occurred in English, Pidgin English and the local language; lasted for an average of 9 minutes; were audio-recorded and transcribed to English. Data were manually coded, and data analysis followed the analytical strategies for qualitative description including an iterative process of inductive and deductive approaches. Results Policymakers and community elders attributed the non-use of maternal health services to poor quality of care. Notions of poor quality of care included shortages in skilled healthcare workers, apathy and abusive behaviours from healthcare providers, lack of life-saving equipment, and lack of safe skilled pregnancy care. Non-use was also attributed to women’s complex utilization patterns which involved a combination of different types of healthcare services, including traditional care. Participants also identified affordability and accessibility factors as deterrents to women’s use of skilled maternal healthcare. Conclusion The emerging findings on pregnant women’s combined use of different types of care highlight the need to improve the quality, availability, accessibility, and affordability of skilled maternal care for rural women in Nigeria. This study uses qualitative description as a qualitative research approach. Qualitative description allows health researchers to examine a phenomenon from a naturalistic perspective and gives a straight description of a phenomenon [22, 23]. In using this method researchers acknowledge existing knowledge of a phenomenon, are able to be flexible in commitment to a theory during the study design and aim for a low-inference interpretation of study findings with an emphasis on describing participants’ views as close to the data as possible. This study presents information from community conversations and key informant interviews with community elders and policymakers respectively, in rural areas of Edo State, Nigeria. Key informants are individuals possessing particular knowledge, status and skill in and understanding of a subject matter. They provide information in various ways including formal interviews, informal conversations, manuscripts, artifacts, or other forms [24]. For this study, policymakers were considered to be in the best position to offer insights from a community and systems-level on non-use of maternal health services and proffer solutions [25]. Policymakers are crucial to the functioning of the health system, and they make decisions that influence service delivery and uptake of skilled maternal health care. In this study, face-to-face in-depth interviews were conducted with various policymakers. Community conversations are interactive processes whereby members of the community are engaged around a common problem and arrive at a solution [26, 27]. One key principle of community conversations is that they create spaces for interaction, change and transfer. This form of qualitative research employs different tools for data collection including conversational dialogues with individuals or reviewing artifacts or group discussions. The common theme in these procedures is an orientation towards problem-solving and the role of the individuals in arriving at solutions. Mainstream approaches to dealing with health issues in African communities have involved assembling people for sensitization sessions or awareness-raising activities [27]. The trade-off has been limited opportunities for dialogue and reflection with and among community members. Community conversations create opportunities for people to discuss health issues away from mainstream social environments thereby rejecting the status quo and enhancing new ways of thinking and questioning [28]. In this study, community conversations occurred as group discussions guided by trained facilitators to support critical thinking and problem solving around the low use of skilled maternal healthcare in their communities. In using this approach for this study, the authors recognize that elders of the various communities possess knowledge and capability to bring about social change individually and collectively around the issues of non-use of maternal health care services. This qualitative study is a part of a larger project by the Women’s Health Action Research Centre in Nigeria and the University of Ottawa, funded under the Innovating for Maternal and Child Health Africa initiative (a partnership of Global Affairs Canada, Canada’s International Development Research Centre, and Canadian Institutes of Health Research). The larger project is a community-based, multi-site, and multi-disciplinary cluster-randomized trial using a mixed methods approach. Details about the larger project have been reported elsewhere [15, 29]. Nigeria’s current population of 206 million makes it the sixth most populous country in the world [30]. Nigeria’s total fertility rate of 5.11 (live births per woman) is projected to drive a population boom and make Nigeria the second largest country in the world by 2100 with a population of 790 million [31]. Nigeria has 36 states and a Federal Capital Territory within which are a total of 774 local government areas (LGAs). It is grouped into six geopolitical zones namely: North West, North East, North Central, South East, South West and South. About 50% of Nigeria’s population reside in rural areas [32]. This study was conducted in Edo State, one of Nigeria’s 36 States and has a population of approximately 3 million people and a land area of 17, 802 square miles [33]. Specifically, this study was conducted in Esan South East (ESE) and Etsako East (ETE), both of which are local government areas (LGA) of Edo state. They are located in the rural and riverine areas of the State. Each LGA comprises 10 political wards within which are several communities, ESE has 100 communities and a total population of 313,717, and ETE has 42 communities and a total population of 145,996 [29]. These study sites were chosen because preliminary baseline assessments revealed high maternal mortality rates and low use of primary healthcare facilities. Primary healthcare centres (PHCs) are government-funded facilities and constitute the main source of skilled maternal healthcare in the two LGAs. There are 25 PHCs in ESE and 28 in ETE. Esan South East has one general hospital in the local government’s headquarters (Ubiaja) and Etsako East has two general hospitals; one in the local government’s headquarters (Agenebode) and another in nearby Fugar City. They are used in addition to existing PHCs for referral for maternal health services. For the community conversations, a total of 151 men elders and 7 women elders aged 50–101 years of age participated in the community conversations. Most of them attained post-primary education, whereas a few had no education. The majority were farmers and artisans. The majority of the participants were Christians, and a few declared no religious affiliation. For the key informant interviews, a total of 6 participants included: one senior official within the State Ministry of Health, one senior official within the State Primary Healthcare Development Agency (SPHCDA), two senior officials responsible for PHCs, with one from ETE and the other from ESE LGAs, two senior Local Government officials, one from ETE and the other from ESE. Community elders were purposefully recruited using locally accepted methods of establishing contact [34]. The lead investigators for the project (FO, LN) identified indigenous guides in ETE and ESE who then introduced them to the traditional ruler of the communities. The project leaders met with the traditional rulers, explained the purpose of the study, and obtained consent to conduct research with community elders. Community elders within a traditional age-based hierarchy are considered influential and agents of change [35]. Target participants were community elders who were often over 50 years old and who were recognized as influential among their communities. Consultations with the traditional ruler identified 10 communities with traditional age-based hierarchies across the two local governments. Traditional rulers in the communities worked with the research team to schedule meetings and communicated with the elders who gathered for the various group discussions. The lead investigators (FO, LN) identified key informants among known policymakers who held various policy-related positions at the State and Local government levels. Using a purposeful criterion sampling technique [36], participants’ eligibility to participate were determined based on the following criteria; 1) participants were in key policy positions 2) participants had experience within the PHC system. These criteria were necessary to enhance opportunities to obtain rich insights into the non-use of maternal healthcare services in rural areas. The lead investigators contacted each participant via phone or email with information about the study. Face-to-face in-depth interviews were conducted with 6 policymakers from different institutions in ETE and ESE LGAs and at the state level. A total of 10 community conversations were conducted, six were conducted in ESE and four in ETE between July 26 to August 16, 2017. Community conversations were conducted as group discussions, each consisting of 12 to 21 participants. Groups were small enough for open dialogue and freedom of expression yet large enough to maximize discussions from a diverse group of elders [37]. Conversations occurred outdoors and were conducted in Pidgin English and a few in the local languages (Ishan or Etsako). The lead investigators trained local researchers to conduct community conversations. The local researchers were conversant with the traditions of the community and spoke English, Pidgin English, Ishan and Etsako. The lead investigators developed a community conversation topic guide which was piloted in a neighbouring village with 12 elders aged 50 years and older. The guide was designed to engage elders in problem-solving. In keeping with cultural practices and values that emphasize oral modes of agreement, all participants provided verbal consent to participate in the study. They all subsequently provided written informed consent. Community conversations proceeded as follows: first, the researchers raised awareness about maternal mortality and preventative strategies. Second, they discussed challenges to women’s access to skilled pregnancy care and finally, elders proffered solutions and mentioned specific ways they will take action to address issues. Conversations lasted about 90 minutes and ended when no further issues arose. Resolutions generated from the discussions were itemized and read back to the elders at the end of the discussion. The elders provided feedback where necessary. Data collection using a key informant interview guide took place from July 16 to August 30, 2017. The principal investigators provided a three-day training session to research assistants who conducted in-depth interviews with participants. In keeping with a qualitative descriptive study, the interview guide was moderately structured to allow for the free description of opinions and experiences. Trained research assistants ran a pilot test of the guide in a community with similar characteristics as the study location. A total of 6 policymakers participated in the study. Quite frequently and for ambivalent reasons, guidelines adopt a sample size of multiples of 10 for interviews. However, studies recommend that in choosing sample size, researchers focus on what has the best opportunity to reach data saturation as that constitutes the gold standard by which purposeful sample sizes are determined in health research [37, 38]. For in-depth interviews, data saturation can be attained in as little as 6 interviews depending on the diversity of data and the sample population [37]. However, the concept of data saturation is also contested within research designs such as qualitative descriptions that stress the uniqueness of each individual’s experience [39]. The authors acknowledge that information obtained from six participants may never truly reach data saturation, the key, however, was to strive to attain thick and rich data. Based on the diverse policymakers interviewed for this study, the authors believe that the data obtained is detailed, nuanced and intricate. Participants signed a consent form prior to participating in the interviews. The interviews lasted for 45 minutes on average and ended when no further issues arose. See S2 Appendix for relevant interview questions. The ethical clearance needed for the larger study was granted by the National Health Research Ethics Committee (NHREC) on April 18, 2017 (reference number NHREC/01/01/2007–18/04/2017). Participants gave their free and informed consent to be enrolled in the study. Participants provided written informed consent prior to participating in this study. They were also informed that once they chose to participate, they could withdraw at any time or chose not to answer any questions, to which there would be no negative consequences. To ensure confidentiality, all personal information were not included in transcripts and quoted texts. The community conversations and key informant interviews were audio-recorded and transcribed by the paid research assistants who were conversant with the spoken languages. Data were transcribed verbatim if in English or translated if in Pidgin English or the local language. Direct translations (word-by-word) were carried out in order to portray the mentality of the participants and present their message as accurately as possible [40]. In cases where syntactical and grammatical structures precluded literal translations, free translations were used to enhance the readability of a text. The primary author (OU) and corresponding author (SY) analyzed the data independently and checked for consistency during frequent discussions. Data analysis was conducted manually and followed the analytical strategies for qualitative description [22]. After immersing themselves in the data, OU and SY read the data line by line, recorded insights, and proceeded to code the data. Next, coded information was sorted to identify patterns and themes from which similarities and differences were identified and extracted for further consideration and analysis. Similar themes generated sub-categories which gave a more general description of the content. Participants’ views of the non-use of maternal health care services in rural Nigeria are presented in the following overarching themes: quality of care, utilization patterns, affordability, and accessibility. The rigour of a study lies in the degree of confidence in methods used, data obtained and interpretation of data. Trustworthiness of a study is necessary for establishing confidence based on various criteria including credibility, dependability, confirmability and transferability [41, 42]. To enhance the credibility of the study, the authors used investigator triangulation; the coding process involved two coders working independently to code the data and working collaboratively to generate themes. Furthermore, this study used method triangulation by having different methods of data collection namely, in-depth interviews with policymakers and community conversations with influential community elders. After data collection, the lead investigators conducted member checks by feeding back data to the participants from whom data was obtained. To enhance confirmability, the primary author (OU) provided thick descriptions of participants’ responses, alongside relevant quotes to confirm interpretations. Quotes were also chosen to represent a typical response relative to the theme.