Determinants of maternity health care utilization in Tajikistan: learning from a national living standards survey.

This study examines the demographic, economic, social, and geographic factors that help explain maternity health care utilization in Tajikistan, a low-income transitional country in Central Asia, based on a newly available nationally representative living standard survey. Two regression models, namely, a binomial logit model and a zero-inflated negative binomial (ZINB) model are estimated. The estimations of the two models show strikingly consistent results--the year of last childbirth, educational attainment, and the availability of public infrastructure are important predictors of maternity health care utilization. The results also signify an urgent need to overturn the current negative trend in maternity health care utilization. The findings also suggest that many determinants of maternity health care are outside the direct sphere of health care policies, indicating the need of cross-sectional policies in addressing maternity health care utilization, for example, policies and strategies that include contributions from other sectors.