Abstract
BACKGROUND: Despite various international efforts initiated to improve maternal health, more than half a million women worldwide die each year as a result of complications arising from pregnancy and childbirth. This research was guided by the following questions: 1) How does women's autonomy influence the choice of place of delivery in resource-poor urban settings? 2) Does its effect vary by household wealth? and 3) To what extent does women's autonomy mediate the relationship between women's education and use of health facility for delivery? METHODS: The data used is from a maternal health study carried out in the slums of Nairobi, Kenya. A total of 1,927 women (out of 2,482) who had a pregnancy outcome in 2004-2005 were selected and interviewed. Seventeen variable items on autonomy were used to construct women's decision-making, freedom of movement, and overall autonomy. Further, all health facilities serving the study population were assessed with regard to the number, training and competency of obstetric staff; services offered; physical infrastructure; and availability, adequacy and functional status of supplies and other essential equipment for safe delivery, among others. A total of 25 facilities were surveyed. RESULTS: While household wealth, education and demographic and health covariates had strong relationships with place of delivery, the effects of women's overall autonomy, decision-making and freedom of movement were rather weak. Among middle to least poor households, all three measures of women's autonomy were associated with place of delivery, and in the expected direction; whereas among the poorest women, they were strong and counter-intuitive. Finally, the study showed that autonomy may not be a major mediator of the link between education and use of health services for delivery. CONCLUSION: The paper argues in favor of broad actions to increase women's autonomy both as an end and as a means to facilitate improved reproductive health outcomes. It also supports the call for more appropriate data that could further support this line of action. It highlights the need for efforts to improve households' livelihoods and increase girls' schooling to alter perceptions of the value of skilled maternal health care.