Background: In Ghana, priority-setting for reproductive health service interventions is known to be rudimentary with little wider stakeholder involvement. In recognizing the need for broad stakeholder engagement to advance reproductive care provision and utilization, it is necessary to jointly study the varied stakeholder views on reproductive care services. Methods: We applied an ethnographic study approach where field data was collected between March-May 2015 in three rural districts of northern Ghana. Data was collected among women with recent births experiences (n = 90), health care providers (n = 16) and policy actors (n = 6). In-depth interviews and focus group discussions was applied to collect all data. Each stakeholder participant’s audio file was transcribed, and repeatedly read through to identify similar and divergent views in data. A coding scheme guided coding processes. All transcripts were then imported into QSR NVivo 11 for further analysis. Results: Four themes emerged. Women participants accentuated that sex and sexuality values of men have changed over time, and drives gender roles, parity levels and decision making on reproductive care needs at community levels. Sexual stigma on reproductive care reduces the willingness of women to voice poor experiences related to their previous reproductive experiences. All stakeholders’ highlighted clinical treatments for post-abortion care are minimally covered under the fee exemption policy for antenatal and postnatal care. Policy processes on service delivery protocols still is top-down in Ghana. Conclusions: Health teams working to improve sexual and reproductive health care must find suitable context strategies that effectively work to improve women reproductive care needs at their operational levels. Private sector participation and informal community support clutches are encouraged to advance the delivery of reproductive care services.
This study employed an ethnographic design. In ethnographic study designs, the voice of participants is an important source of data gathered throughout the investigation process [18, 19]. We choose to apply ethnography because we were interested in explaining the “culture” of the phenomenon using participant voice, behavior and observations from the field [20]. Ethnography allows for understanding the phenomenon from both the insider and outsider perspective [21, 22]. The design ensured that all stakeholder concerns surrounding the study aim were given a “voice”[23]. It provided the opportunity to examine the complexity of reproductive care needs and how the interactions of women, health staff and policymakers influence the provision and utilization of reproductive care services. Data were collected between March and May 2015 alongside another study that investigated stakeholder’s perspectives on shortcomings for maternity care in rural Ghana. Three districts in addition to the capital of the Upper East region of Ghana form the study setting. The Upper East region is one of the 10 administrative regions in Ghana. The Upper East region currently has 13 administrative divisions under Ghana’s decentralization structure. Major ethnic groups in the region include Bimoba, Bissa, Buli, Frafra, Kantosi, Kasem, and Kusasi. The health system in the Upper East region is organized into a 4-tier system typical of most regions in Ghana; regional, district, sub-district and community levels often called CHPS centres. Three districts; Bongo, Talensi, and Nabdam were purposively selected for this study. The 3 districts were selected due to poor indices of maternal and reproductive health outcomes among the 13 administrative divisions in the region. Proximity and resource availability for data collection also influenced choice of districts. In each of these districts, three remote facilities providing reproductive health services in each district were selected (9 facilities in total). We also included Bolgatanga, the capital of the Upper East region as one of the settings because some policy interviewees resided in this area. Data were collected among three groups of stakeholders: women with records of recent births (2 years prior to study), health staff, and policymakers. Focus group discussions (FGDs) were conducted with women and health staff while in-depth interviews (IDIs) were held with policymakers (public and private). To select women participants, midwives in the included facilities provided information on 15 women per facility with recent birth experiences using data from antenatal and postnatal facility registers. In total 135 women were selected as potential participants in all 3 districts. At each district level community, meetings were held across all 9 facility settings to select final women participants for the study. A final inclusion checklist with the following criteria was used to rank final participants at each facility: woman should be physically present at meetings, consent to participate, interested in the study and able to share prior knowledge related to the study aim. At each facility meeting, 10 participants with high rating towards study were selected. A convenience sample of 30 eligible women per district took part in the FGDs (3 FGDs per district with 10 women in each FGD). Thus, in total 90 women took part in the study. Each woman participant was provided a participant consent form and was asked to complete this form through signing or thumbprint before all interviews started. Across the three facilities per district, health staff in one of the facilities was selected. In these district facilities, one midwife, a senior staff in-charge of facility, and facility nurses (on average 3–7 persons per district) responsible for providing reproductive health services at these facility centers were recruited as district health staff participants. Policymakers were later recruited and included 2 private health policy program managers and 4 public sector policymakers. The two private health policy managers worked in health facilities under the umbrella Christian Health Association of Ghana (CHAG) across our study settings. All four public sector policymakers worked in public health facilities across study districts. All four policy actors read and signed a consent information sheet before the start of each interview. FGDs with women were conducted in one main district local language; Talensi (Tali), Nabdam (Nabit) and Bongo (Grune). All IDIs were conducted in English. FGDs and IDIs were conducted using sample structured guides (see Table 1). In conducting the interviews, the ethnographic approach allowed for continuous inductions and verification of all stakeholder views. FGD and IDI audio files were transcribed verbatim into English by two research team members (the principle researcher and an assistant). We used two research team members to minimize single level biases in our analysis. Transcripts from all stakeholder groups were repeatedly read by the principle researcher and the assistant to identify similar and divergent views in data. All individual stakeholder transcripts were finally checked for accuracy and consistency with the original audio files. The research team then developed a coding scheme to guide the coding for each stakeholder group response based on the aim of the study. To ensure consistency among the two coders, we developed coding rules to facilitate the process. An initial coding for transcripts was undertaken by the principle researcher and the assistant using the principles for open coding [24]. All transcripts were then imported into QSR NVivo 11. In NVivo, one research team member undertook a further induction (open coding) for stakeholder concepts while another team member audited the first coder work to ensure consistency and reliability of coding in NVivo. Each imported transcript was analyzed in NVivo separately. Constant comparison enabled a further reduction of concepts, codes and nodes into themes [25]. To show how individual stakeholder views relate to each other and our study aim, we used the model explorer tool to map out how each stakeholder themes relate to each other [26]. All codes were further refined until we reached saturation. To guarantee the reliability for our results, we estimated interrater coding reliability for all coded data. Women, health staff and policymakers recorded r = 0.98, r = 0.96 and r = 0.94 respectively. Mapped out results for all stakeholder themes are tabulated and further illustrated in the results section. All participants were provided with detailed information about the study and how data are collected, used and stored in the study. During the data collection, we ensured that accurate and adequate data on the study aim was captured. We also ensured strict compliance by all research team members to prevent unauthorized access to participant’s data, loss, destruction or damage to any data collected. Study approval was granted by the Institutional Review Board of the Navrongo Health research Centre, Ghana with ID number: NHRCIRB202.
N/A