Reproductive Autonomy and Modern Contraceptive Use at Last Sex Among Young Women in Ghana.

Abstract

CONTEXT Variability in the conceptualization and measurement of women's empowerment has resulted in inconsistent findings regarding the relationships between empowerment and sexual and reproductive health outcomes. Reproductive autonomy-a specific measure of empowerment-and its role in modern contraceptive use have rarely been assessed in Sub-Saharan contexts. METHODS Survey data were collected from a sample of 325 urban Ghanaian women aged 15-24 recruited from health facilities and schools in Kumasi and Accra in March 2015. Bivariate and multivariable logistic regression analyses were used to examine associations between two adapted reproductive autonomy subscales-decision making and communication-and women's use of modern contraceptives at last sex, controlling for demographic, reproductive and social context (i.e., approval of and stigma toward adolescent sexual and reproductive health) covariates. RESULTS In multivariable analyses, reproductive autonomy decision making-but not reproductive autonomy communication-was positively associated with women's modern contraceptive use at last sex (odds ratio, 1.1); age, having been employed in the last seven days and living in Kumasi were also positively associated with modern contraceptive use (1.1-9.8), whereas ever having had a previous pregnancy was negatively associated with the outcome (0.3). Reproductive autonomy decision making remained positively associated with contraceptive use in a subsequent model that included social approval of adolescent sexual and reproductive health (1.1), but not in models that included stigma toward adolescent sexual and reproductive health. CONCLUSIONS The reproductive autonomy construct, and the decision-making subscale in particular, demonstrated relevance for family planning outcomes among young women in Ghana and may have utility in global settings. Future research should explore reproductive autonomy communication and the potential confounding effects of social context.