Patient and public involvement
Patient were not involved in this study.
Data source, sampling design and study population
In this study, secondary data of demographic and health survey (DHS) of Sub-Sahara Africa countries were used. DHS is a nationally representative household sample survey that evaluates population socio-demographics, maternal and child health, and a variety of health indicators including the use of contraceptives. The DHS is a valuable source of data for studying population health indicators because of its coverage, data quality, and comparability throughout the world. In addition, the sample used is generally representative at the national, regional, and residence (rural and urban) level. DHS sampling is based on a two-stage cluster design approach. In the first stage, there is stratification and proportional allocation of the sample frame. The second stage involves a selection of households per cluster with equal probabilities in a systematic approach. Details of the sampling design and sampling procedure can be found at the DHS program websites (www.dhsprogram.com/methodology/survey-types/DHS).
The available recent demographic and health survey (DHS) conducted in 37 Sub-Sahara Africa countries from 1995 to 2020 were included in this study. These countries include Angola, Benin, Burkina Faso, Burundi, Cameroon, Central African Republic, Comoros, Congo, Congo Democratic, Cote d'Ivoire, Eswatini, Ethiopia, Gabon, Gambia, Ghana, Guinea, Kenya, Lesotho, Liberia, Madagascar, Malawi, Mali, Mozambique, Namibia, Niger, Nigeria, Rwanda, Sao Tome and Principle, Senegal, Sierra Leone, South Africa, Sudan, Tanzania, Togo, Uganda, Zambia and, Zimbabwe. Data were downloaded from the DHS programme website (www.dhsprogram.com) after granting permission. The data archive at the DHS website had 38 Sub-Sahara Africa countries excluding Ondo State in Nigeria. Chad was not included because of its extremely old data (1990) and was missing most of the independent variables of interest as well as lack of stratification. The unit of analysis in this study was women of reproductive age (15-49 years).
Definition of variables
Outcome variable
The current use of MC by women of reproductive age (15–49 years) was the primary outcome of interest. This was dichotomized as “use of a modern method (coded as “1”) and non-use of modern contraceptive (coded as “0”). Modern contraceptive was described as the use of any of the following contraceptive methods: sterilization (female), intrauterine system (IUD), injectable, implant, tablets, condom (female), standard days method (SDM), emergency contraception, diaphragm, foam/jelly, diaphragm, country-specific modern methods, and other modern contraceptive methods respondent mentioned (including cervical cap, contraceptive sponge, and others) but does not include abortion, menstrual regulation as described in the DHS questionnaire. Traditional methods included periodic abstinence (rhythm, calendar method), withdrawal (coitus interruptus) and country-specific traditional methods of proven effectiveness, and folk methods including locally described methods and/or spiritual methods such as herbs, amulets, gris-gris, etc.
Independent variables
The independent variables considered in this study include socio-demographic characteristics such as the age of the respondent (“15-19”, “20-24, 25-29”, “30-34”, “35-39”, “40-44” and “45-49”), age at first birth, recoded (“no birth”, “<20”, “20-29” and 29+), education (“no education”, “Primary”, “secondary” and “higher”), husband/partners education (“no education”, “Primary”, “secondary” and “higher” “don’t know”) religion, recoded (“Christian”, “Islamic”, “Traditional” and “Other”), marital status, recoded (“Never married” “Married”, “Co-habiting” and “Other”), wealth index, recoded (“poorer/poorest”, “Middle” and “Richer/Richest”), Employment (“working” and “not working”). Others include the number of living children, recoded (“0”, “1-2”, “3-4”, “5-7” and “7+”), source of reproductive information, recoded as media (radio, television, newspaper, text messages, “Yes” and “No”), been told of family planning at a health facility (“Yes” and “No”), place of residence (“rural” or “urban”), the number of sex partners excluding the spouse, recoded (“none”, “1”, “2”, “3+” and “don't know”) and knowledge of modern contraceptive (“Yes” and “No”), field worker visited and talked about family planning (“Yes” and “No”) and visited health facility in the last 12 months (“Yes” and “No”). These variables were chosen based on previous studies (8, 14, 16, 22).
Data analyses
Data for this study were analyzed using STATA version 16 for windows. Data were cross-checked for missing data and no response or interviewer error (9 or 99) and were excluded in the analyses. Again, missing data associated with the outcome variable, use of modern contraceptives, were dropped from the analyses. Descriptive statistics were summarised for demographic characteristics and prevalence of the use of MC in each country. A bivariate analysis (Rao Scott’s X2) was done to determine the association of socio-demographic characteristics, questions relating to the use of contraceptives, and the outcome variable (use of modern contraceptive). Variables that showed significance in the bivariate analysis were used for the multiple logistic regression analyses. The independent variables were checked for multi-colinearity before the multiple logistic regression. Sample weight was adjusted by dividing the individual women's sample weight by 1000,000 (v005/106). In all analyses, clustering, stratification, and applied sampling weights were accounted for to reduce bias and to improve on the adjusted estimates and standard errors as recommended in complex survey design analysis. A p<0.05 was considered statistically significant.
Ethical Approval
This study required no ethical clearance as secondary data was used. However, written permission was sought and was granted from the DHS program before data access.