Globally, Nigeria is the second most unsafe country to be pregnant, with Lagos, its economic nerve center having disproportionately higher maternal deaths than the national average. Emergency obstetric care (EmOC) is effective in reducing pregnancyrelated morbidities and mortalities. This mixed-methods study quantitatively assessed women’s satisfaction with EmOC received and qualitatively engaged multiple key stakeholders to better understand issues around EmOC access, availability and utilization in Lagos. Qualitative interviews revealed that regarding access, while government opined that EmOC facilities have been strategically built across Lagos, women flagged issues with difficulty in access, compounded by perceived high EmOC cost. For availability, though health workers were judged competent, they appeared insufficient, overworked and felt poorly remunerated. Infrastructure was considered inadequate and paucity of blood and blood products remained commonplace. Although pregnant women positively rated the clinical aspects of care, as confirmed by the survey, satisfaction gaps remained in the areas of service delivery, care organization and responsiveness. These areas of discordance offer insight to opportunities for improvements, which would ensure that every woman can access and use quality EmOC that is sufficiently available.
This study utilized mixed methods consisting of quantitative and qualitative components to assess perceptions related to availability, access and utilisation of EmOC in Lagos. Using a sequential explanatory mixed methods design9 that comprised two distinct phases – quantitative phase and qualitative phase subsequently. Both methods were used for the purposes of triangulation and complementarity.10 The quantitative satisfaction survey serves as a baseline for future EmOC service satisfaction surveys in the state and the qualitative interviews and focus group discussions (FGDs) helped to elaborate on some of the quantitative results, as it provided further explanation regarding the situation of the service and perspectives for improvement. We therefore gave ‘priority’9 to the qualitative findings. The study was conducted in Lagos state, South-West, Nigeria with an estimated population of approximately 17 million, of which women constitute about half of the total population.11 At the time of this study, Lagos had 22 functional public secondary healthcare facilities that provided CEmOC (Figure 1). Map of Lagos showing CEmOC facilities within the local government areas. For the quantitative phase, a cross-sectional survey was conducted to assess satisfaction of women with the service received from all 22 CEmOC facilities in Lagos, using a modified version of a standard questionnaire – the Six Simple Questionnaire (SSQ)12 – for data collection. The questionnaire was administered to consenting women 18 years or older, who had received at least one signal function between February 2015 and April 2015 from any of the facilities and had been admitted in the post-natal wards. The SSQ assessed the following care features: feeling of control over care, problems dealt with effectively, adequacy of care organization, compassion of health worker, needs of patient addressed and make same choice of facility for next pregnancy. The survey also collected the respondents’ socio-economic and demographic background, duration of travel to facility, waiting time to see an HCP, reasons for choosing the facility and previous facility deliveries. Cluster randomized (facility-randomized) sampling was used to recruit women to the study. Sample size was estimated using G*Power 3.1®, targeting a power of 80% and an α-error of 0.05. As the population size of our target group was unknown, we used the conservative assumption for proportion of 50%, such that the constant proportion of women who give positive rating would be 50%. The sample size was estimated to be able to describe a 25% significant difference in rating. These parameters were used to generate sample size with a design effect of 1.5 to adjust for the methodological consequences of using a cluster randomized sampling method. Assumption made to estimate “ρ” clusters was similar within each sample, to bring value to “1”. The computation under these assumptions prescribed a sample size of 30 participants in each CEmOC unit, which adds to 660. Subsequently, factoring the design effect led to an estimated sample size of 990. STATA SE 13.0® was used for data analysis. Demographic data was presented using summary tables. Frequency and percentages was used to describe research findings. In addition, specific information about the pregnancy were summarized as frequencies and percentages. Median satisfaction scores across the six satisfaction dimensions were estimated along with inter-quartile ranges and variances. The results were presented using a box plot. From the pool of women who had received EmOC recruited for the quantitative phase, some were purposively sampled to ensure a representation across different socio-economic groups and signal functions received. We conducted FGDs with the recruited women while excluding any woman who was emotionally fragile following delivery. FGDs were viewed as the most suitable method for this study as such forums enable interaction amongst respondents. 13 Each FGD session with 6-8 participants lasted approximately 45 minutes and was conducted within the post-natal ward of the facilities at the convenience of the women. The FGDs were facilitated by a moderator while a note-taker captured any subliminal non–verbal events. Key Informant Interviews (KIIs) were conducted with purposively selected stakeholders who could share rich information related to EmOC in Lagos. They included relatives of women, HCPs, hospital managers, and government representatives. Topic guides that focused on access, availability and utilization of EmOC in Lagos were used. Each topic guide for the different stakeholder groups was pilot-tested with non-included respondents. The topic guides used for women and their relatives were initially developed in English and subsequently translated and back-translated to Yoruba and Creole (Broken English), languages commonly spoken in the region. The KII topic guides for other stakeholders were only in English. All FGDs and KIIs were audio-recorded using dictaphones after informed consent was obtained from respondents. Those present during the FGDs were the moderator, notetaker and respondents. However, only the interviewer and interviewee were present for the KIIs. Data collection continued until theoretical saturation was attained.13 Before closing the sessions, the moderator/interviewer double-checked accuracy of gathered information with the respondent(s). All audio recordings were transcribed verbatim, with the resulting transcripts reviewed independently for accuracy by the moderators/interviewers. For data reduction, the thematic approach, which focuses on detecting and describing both implicit and explicit ideas (themes) within the transcript was applied.14 We followed Braun and Clarke’s six-step approach: Becoming familiar with the data, generating initial codes, searching for themes, reviewing themes, defining and naming themes and producing the report.14 All qualitative analyses were conducted with the aid of NVivo 10TM. Ethics approval was obtained from the Health Research and Ethics Committee of the Lagos State University Teaching Hospital, Ikeja, Lagos. Permission to conduct the study was granted by the Lagos State Ministry of Health. For all respondents, details of purpose, format, risks and benefits of participating in the research were verbally explained. Respondents were informed that their participation was voluntary and given sufficient time to decide on their participation. Written informed consent was obtained from each respondent. No financial incentive was given to respondents.
N/A