Menu
Menu
Menu
Menu
Introduction Maternal mortality is an issue of global public health concern with over 300,000 women dying globally each year. In sub-Saharan Africa (SSA), these deaths mainly occur around childbirth and the first 24hours after delivery. The place of delivery is, therefore, important in reducing maternal deaths and accelerating progress towards attaining the 2030 sustainable development goals (SDGs) related to maternal health. In this study, we examined the prevalence and determinants of the place of delivery among reproductive age women in SSA. Materials and methods This was a cross-sectional study among women in their reproductive age using data from the most recent demographic and health surveys of 28 SSA countries. Frequency, percentage, chi-square, and logistic regression were used in analysing the data. All analyses were done using STATA. Results The overall prevalence of health facility delivery was 66%. This ranged from 23% in Chad to 94% in Gabon. More than half of the countries recorded a less than 70% prevalence of health facility delivery. The adjusted odds of health facility delivery were lowest in Chad. The probability of giving birth at a health facility also declined with increasing age but increased with the level of education and wealth status. Women from rural areas had a lower likelihood (AOR = 0.59, 95%CI = 0.57–0.61) of delivering at a health facility compared with urban women. Conclusions Our findings point to the inability of many SSA countries to meet the SDG targets concerning reductions in maternal mortality and improving the health of reproductive age women. The findings thus justify the need for peer learning among SSA countries for the adaption and integration into local contexts, of interventions that have proven to be successful in improving health facility delivery among reproductive age women.
The study made use of collective data from the most recent Demographic and Health Surveys (DHS) in 28 countries in SSA conducted between 2010 and 2018. The DHS is a nationwide study undertaken in five years intervals in several developing countries in Africa, parts of Asia and Latin America. The DHS follows consistent procedures in questionnaires design, sampling, data collection, data cleaning, coding, and analyses, which allows for comparability across countries [21, 22]. For this study, only women who had given birth in the five years preceding the survey were included, which is 167,763. The main outcome variable was the place of delivery. The outcome variable was coded as 0 = ‘home’ and 1 = ‘‘health facility’ [19]. Fourteen explanatory variables were used namely: age, residence, women and partner’s level of education, wealth status, marital status, number of ANC visits, skilled ANC provider, getting medical help for self: money needed for treatment, distance to a health facility and getting permission to go, listening to the radio and watching television. Age was classified in 5 –year grouping and categorized as 15–19 = 1, 20–24 = 2, 25–29 = 3, 30–34 = 4, 35–39 = 5, 40–44 = 6, and 45–49 = 7. Place of residence was captured as urban = 1 and rural = 2. Women and partner’s levels of education were captioned as no education = 1, primary = 2, secondary = 3, and higher education = 4. Wealth status was categorized as poorest = 1, poorer = 2, middle = 3, richer = 4, and richest = 5. Marital status was also categorized as married = 1, cohabitation = 2, widowed = 3, divorced = 4, and separated = 5. The number of Antenatal Care (ANC) visits was captured as less than four visits = 1 and four or more visits = 2. Skilled ANC provider was categorised as no = 0 and yes = 1. Getting medical help for self: money needed for treatment, distance to a health facility, and getting permission to go were captured as a big problem = 1 and not a big problem = 2. Listening to radio and watching television were recorded as not at all = 1, less than once a week = 2 and at least once a week = 3. Descriptive and inferential analyses were performed. The descriptive analysis reported results on background characteristics, country, and the prevalence of place of delivery. Two Inferential models were analysed using binary logistic regression. Model 1 explored the association between place of delivery and the country variable. Model 2 also explored the association between the outcome variables and all the explanatory variables. The results of Model 1 are presented as crude odds ratios (CORs) with 95% confidence intervals (CIs). Whereas Model 2 is presented as adjusted odds ratios (AOR) with 95% confidence intervals (CIs). Stata version 14 was used for the analysis. The multifaceted nature of the sampling structure of the DHS data was adjusted using the Stata Survey command ‘svyset v021 [pweight = wt], strata (v023)’, and the individual sample weight variable (v005). Questionnaires and procedures for the surveys were reviewed and approved by the Ethics Committee of Opinion Research Corporation Macro International Inc and ICF Institutional Review Board (IRB). As nationally representative surveys, the DHS survey protocols for the various countries were also reviewed and approved by the ICF IRB and the relevant IRBs of the various countries. All data were completely anonymized, de identified, and/or aggregated before access and analysis. Detailed information on the ethical procedures observed by the DHS program can be accessed via http://goo.gl/ny8T6X. As we used secondary data for our analysis, we did not require further ethical approval from our named institutional bodies as the national level ethical clearance was sufficient for our analysis to be carried out.
N/A
