Menu
Menu
Menu
Menu
Background: In Senegal, sub-Saharan Africa, many women continue to die from pregnancy and childbirth complications. Even though health facility delivery is a key intervention to reducing maternal death, utilization is low. There is a dearth of evidence on determinants of health facility delivery in Senegal. Therefore, this study investigated the predictors of health facility-based delivery utilization in Senegal. Methods: Data from the 2017 Senegal Continuous Survey were extracted for this study, and approximately 11,487 ever-married women aged 15–49 years participated. Chi-square test was used to select significant variables and multivariable logistic regression analysis was performed to identify statistically significant predictors at a 95% confidence interval with a 0.05 p-value using Stata version 14 software. Results: Facility-based delivery utilization was 77.7% and the main predictors were maternal educational status (primary school Adjusted Odds Ratio [aOR] = 1.44, 95% CI; 1.14–1.83; secondary school aOR = 1.62, 95% CI; 1.17–2.25), husband’s educational status (primary school aOR = 1.65, 95% CI; 1.24–2.20, secondary school aOR = 2.17, 95% CI; 1.52–3.10), maternal occupation (agricultural-self-employed aOR = 0.77, 95% CI; 0.62–0.96), ethnicity (Poular aOR = 0.74, 95% CI; 0.56–0.97), place of residence (rural aOR = 0.57, 95% CI; 0.43, 0.74), media exposure (yes aOR = 1.26, 95% CI; 1.02–1.57), economic status (richest aOR = 5.27, 95% CI; 2.85–9.73), parity (seven and above aOR =0.46, 95% CI; 0.34–0.62), wife beating attitude (refuse aOR =1.23, 95% CI; 1.05–1.44) and skilled antenatal care (ANC) (yes aOR = 4.34, 95% CI; 3.10–6.08). Conclusion: Uptake of health facility delivery services was seen among women who were educated, exposed to media, wealthy, against wife-beating, attended ANC by skilled attendants and had educated husbands. On the other hand, women from ethnic groups like Poular, those working in agricultural activities, living in rural setting, and those who had more delivery history were less likely to deliver at a health facility. Therefore, there is the need to empower women by encouraging them to use skilled ANC services in order for them to gain the requisite knowledge they need to enhance their utilization of health facility delivery, whiles at the same time, removing socio-economic barriers to access to health facility delivery that occur from low education, poverty and rural dwelling.
Senegal, located in West Africa, is well-known as the “Entry to Africa” [32, 33]. Up to half of its 15.4 million people (as of 2016) live in and around Dakar and other urban areas [33]. Since 1960, three very non-violent political changes have taken place, ensuring its stability [34]. The nation’s economic growth, reported at 6% growth rate in 2018, looks promising for the future [34]. According to available data in 2011 by the World Bank, 38% of Senegal’s population lives on less than $1.90 per day [34]. Senegal’s health system is a hierarchical structure, with each of the 14 regional medical offices in charge of the provision and supervision of healthcare within the regions. There are also health districts which usually consist of one health center linked to rural health posts, some of which supervise the allied health huts [35]. There are also community-level facilities known as health huts, usually operated by a community health worker employed by community health committees [36]. We used the most recent (2017) Senegal Continuous Survey (SCS) for this analysis [37]. Sampling for the 2017 SCS was done using a stratified, two-stage cluster sampling design to provide estimates for essential population and health indicators for the country. Large geographic settings known as enumeration areas (EAs) were selected in the first stage through Probability Proportional to Size (PPS). The survey included a total of 8800 (4092 in urban areas and 4708 in rural) households and a total of 16,787 women (15–49 years of age) and 6977 Men (15–59 years of age) were interviewed [37]. Household listing was completed in each EA to ready the sampling frame. Selected participants were questioned using standard and country-specific questionnaire modules covering a wide range of health topics. For this study, we included 11,487 currently married women aged 15–49 years with a birth, for the most recent live births in the 5 years preceding the survey [37] from the kids (children) recode file (KR). The survey is publicly available on the DHS website (www.dhsprogram.com). Place of delivery was the outcome variable in this study and was grouped into health facility delivery (deliveries that occurred in a government hospital, government health center/maternity, government health post, mobile government clinic, government field worker, other public sector, private hospital/clinic and other private sectors) and non-health facility delivery (deliveries that occurred at respondents’ or relatives’ homes, or in other places like on the road). Births with missing information were added to the denominator for both the distribution of place of delivery and percentage of all births that occurred in a health facility. The percentage distribution of place of delivery included a separate category for missing values. Despite the fact that data were available for all live births to questioned women in the 5 years preceding the survey, we calculated for only the most recent birth as recommended by DHS guideline. Several individual and community level explanatory variables were incorporated from previous studies [17, 21, 23, 24, 38–43] due to their role in contributing to increase or decrease in the use of facility delivery. The independent variables were maternal age (15–19, 20–24, 25–29, 30–34, 35–39, 40–44, 45–49), maternal educational status (no formal education, primary school, secondary school, higher), maternal occupation (not working, sales and services, agricultural-self-employed, other), husband education (no formal education, primary school, secondary school, higher), husband occupation (not working, professional/technical or managerial, sales and services, agricultural-self-employed, skilled manual, unskilled manual, other), religion (Muslim, Christian), ethnicity (Wolof, Poular, Serer, Mandingue/Soce, Diola, Soninke, other Senegalese, other), region (Dakar, Ziguinchor, Diourbel, Saint-Louis, Tambacounda, Kaolack, Thios, Louga, Fatick, Kolda, Matam, Kaffrine, Kedougou, Sedhiou), and wealth index (poorest, poorer, middle, richer, richest). We looked at media exposure (if the respondent was exposed to any of the three types; read newspaper, listened to radio or watched television for at least less than once a week it was coded as yes, and otherwise, no), place of residence (urban, rural), and parity (<=2, 3–4, 5–6, 7+). Decision making power was also included; we looked to see if the respondent had no decision-making power, she alone made decisions, or if she made decisions together with her husband. There were three decision making parameters; decision making about her health, to purchase large household items, to visit family/relatives. We coded “no decision making” if only the husband or other family members made decisions; we coded “decision making one” if the respondent had decision making power either alone or together with her husband on two of the above decision-making parameters; and we coded “decision making power two” if the respondent made decisions alone or together with her husband on all three decision making power parameters. Attitude toward wife beating was assessed as “refused” if the respondent disagreed with all five of the wife beating circumstances presented (burning the food while cooking, arguing with husband, going to visit family without husband permission, neglecting children, refusing to have sex with her husband), and “accepted” if she agreed to any of the five wife beating parameters. We included the use of skilled antenatal care (ANC); if the women had ANC follow up by a skilled attendant (i.e. doctor, midwife, nurse) we coded as yes, if not we coded as no. The participants’ socio-demographic characteristics were computed. Chi-square test was performed to identify variables that showed significant associations with the outcome variable at p-value less than 0.05 cut point. These variables were entered into the multivariable logistic regression model. Results of the multivariable logistic regression were reported using adjusted odds ratios (aORs) at a 95% confidence interval. Data was analyzed using Stata version 14 software (Stata Corp, College Station, Texas, USA). Weighting was applied using the guidelines provided in the user manual (https://www.dhsprogram.com/pubs/pdf/DHSG4/Recode7_DHS_10Sep2018_DHSG4.pdf), while the ‘SVY’ command was used to account for the complex sampling design. Since we used secondary data from SCS dataset which is available publicly, we did not need further ethical approval to use the data. However, in addition to obtaining the participants consent prior to survey, the ICF international strictly followed the ethical standards collaborating with the concerned country’s Ethical Review Board to ensure the DHS data collection process was in line with the U.S. Department of Health and Human Services regulations for the respect of the right of human subjects.
N/A
