Background: Despite the commitments of the government to minimize unintended pregnancy, abortion, and unmet need for contraceptives, as per our search of the literature, there is no study on the pooled prevalence and determinants of informed choice of contraceptive methods in sub-Saharan Africa. Therefore, this study aimed to assess the pooled prevalence and determinants of informed choice of contraceptive methods among reproductive-aged women in sub-Saharan Africa. Methods: This study was based on the 32 Sub-Saharan African countries Demographic and Health Survey data. A total weighted sample of 65,487 women aged 15–49 was included in the study. The data were cleaned, weighted, and analyzed using STATA Version 14 software. Multilevel logistic regression modeling was used to identify determinants of an informed choice of contraceptive methods. Adjusted odds ratio (AOR) with 95% Confidence Interval (CI) and p-value < 0.05 were used to declare the significant determinants. Result: The pooled prevalence of informed choice of contraceptive methods among reproductive age (15–49) women in sub-Saharan Africa was 49.47% (95%CI: 44.33, 54.62%) with I2 =99.5%, and variations in range of 19.42 to 78.42%. Women aged 25–34 years old (AOR = 1.26 95%CI: 1.21, 1.32) and 35–49 years (AOR = 1.33 95%CI: 1.27, 1.40), attending primary education (AOR = 1.26, 95% CI: 1.20, 1.32), secondary education (AOR = 1.50, 95% CI: 1.43, 1.58) and higher education (AOR = 2.01, 95% CI: 1.84, 2.19), having media exposure (AOR = 1.12, 95%CI: 1.07, 1.16), utilizing IUD (AOR = 1.98, 95%CI: 1.79, 2.19), injectable (AOR = 1.29, 95%CI: 1.23, 1.36) and implants (AOR = 1.70, 95%CI: 1.61, 1.79), survey year 2016–2020 (AOR = 1.38, 95%CI: 1.31, 1.44), women from lower middle (AOR = 1.25, 95%CI: 1.19, 1.31) and upper middle income level countries (AOR = 1.37, 95%CI: 1.23, 1.53) were associated with increased odds of informed choice of contraceptive methods. While, women who accessed contraceptives from private clinics (AOR = 0.64, 95%CI: 0.61, 0.67), pharmacies (AOR = 0.37, 95%CI: 0.35, 0.40), and others (AOR = 0.47, 95%CI: 0.43, 0.52), women in East Africa (AOR = 0.70, 95% CI: 0.67, 0.73), Central Africa (AOR = 0.52, 95% CI: 0.47, 0.57), and South Africa (AOR = 0.36, 95% CI: 0.32, 0.40) were associated with decreased odds of informed choice of contraceptive methods. Conclusion: The pooled prevalence of informed choice of contraceptive methods in Sub-Saharan Africa is low with high disparities among the countries. Enhancing maternal education and media exposure, providing greater concern for the source of contraceptive methods, and strengthening the economic status of the country are recommended to enhance informed choice of contraceptive methods.
The data used in this study were the most recent Demographic and Health Surveys (DHS) data compiled in the 32 SSA countries (Angola, Burkina-Faso, Benin, Burundi, Democratic Republic of Congo, Congo, Ivory Coast, Cameroon, Ethiopia, Gabon, Ghana, Gambia, Guinea, Kenya, Comoros, Liberia, Lesotho, Mali, Malawi, Nigeria, Niger, Namibia, Rwanda, Sierra Leone, Senegal, Chad, Togo, Tanzania, Uganda, South Africa, Zambia, and Zimbabwe) from 2010 to 2019/2020. The data were derived from the measure DHS program. These datasets were merged together to determine the pooled prevalence and determinants of informed choice of contraceptive methods across the SSA countries. The DHS data is a nationwide representative survey and it has different datasets (men, women, birth, children, and household datasets). For this study, we used the individual data set (IR file). The DHS used two stages of stratified sampling technique to select the study participants. We pooled the DHS surveys conducted in the 32 SSA countries and a total weighted sample of 65,487 women aged 15–49 who are currently using selected modern contraceptive methods (pill, IUD, injectable, female sterilization, and implants) was included in the study. The outcome variable for this study was informed choice of contraceptive methods, which was a binary outcome variable coded as “0” if a woman did not receive informed choice of contraceptive methods, and “1” if she did receive informed choice of contraceptive methods. The variable was generated using the three questions. The three questions were: (1) “Were you told about possible side effects or problems you might have with the method?” (2) “Were you told what to do if you experience any side effects?” and (3) “Were you told about other methods of family planning?” Answers were coded as 1 = Yes and 0 = No, then categorized as a woman received informed choice of contraceptive methods if they answered all three questions, and otherwise they did not receive an informed choice of contraceptive methods (18, 25). The independent variables were classified as individual level variables such as age, marital status, and maternal educational status, husband educational status, maternal occupation, husband occupation, media exposure status, household wealth index, internet use, visiting health facilities in the last 12 months, source of contraceptive method, and type of contraceptive use. Whereas, community level variables, such as residence, region in SSA countries, DHS survey year, and country income level. Media exposure was calculated from three variables; listening to the radio, reading newspapers, and watching television. If a woman were exposed to at least one type of media, she was considered as exposed to media (26). The country income level was categorized as low income, lower middle income, and upper-middle-income country based on the World Bank List of Economies since 2019 (27). After we accessed the data from the DHS program, we cleaned the data, carried out cross-tabulation, and calculated descriptive and summary statistics using STATA version 14 software. Before inferential analysis, we applied weighting using sampling weight, primary sampling unit, and strata to restore the representativeness of the survey and to get reliable statistical estimates. The pooled prevalence of informed choice of contraceptive method across countries from 2010 to 2019/2020 was done using a metan STATA command and presented in a forest plot with 95% Confidence Interval (CI). Besides, further subgroup analyses were done to minimize the heterogeneity between studies using regions in SSA countries and the DHS survey year. The multilevel binary logistic regression analysis were used to determine the association between the likelihood of informed choice of contraceptive methods and explanatory variables at both individual and community levels. After bivariable analysis, variables with a p-value < 0.2 were included in the multivariable multilevel logistic regression model. Adjusted Odds Ratios (AOR) with a 95% CI and a p-value < 0.05 in the multivariable multilevel logistic model were used to declare significant determinants of informed choice of contraceptive methods. The DHS data has a hierarchical nature. Hence, women are nested within a cluster, and we anticipate that women within the same cluster may be more similar to each other than women in the rest of the country. This implies that advanced models need to take into account the between cluster variability. Therefore, we used the mixed-effect logistic regression analysis method. Moreover, model comparison and fitness were assessed based on the Intra-class Correlation Coefficient (ICC), Median Odds Ratio (MOR), Proportional Change in Variance (PCV), and deviance (-2LLR) values. The calculation for MOR and PCV is as follows; Generally, in this study, four models were fitted: the first model was a null model (model without the independent variables), which was used to check the variability of informed choice of contraceptive methods in the cluster. The second (model I) and third (model II) models, which contain individual-level variables and community-level variables, respectively. The final model (model III), which contains both individual and community-level variables simultaneously. We used deviance to select the best fit model for the data. The final model (model III) was selected because of its lowest deviance value (Table 3). Ethical approval and participant consent were not necessary for this study since it was a secondary data analysis based on the publically available DHS data. We requested to access the data from the measure DHS program and permission was obtained to download and use the data for this study. There are no personal identifier in the data files.
N/A