Community and health system factors associated with facility delivery in rural Tanzania: a multilevel analysis

Abstract

Objectives: Tanzania, a country with high maternal mortality, has many primary health facilities yet has a low rate of facility deliveries. This study estimated the contribution of individual and community factors in explaining variation in the use of health facilities for childbirth in rural Tanzania.

Methods: A two-stage cluster population-based survey was conducted in Kasulu District, western Tanzania with women with a recent delivery. Random intercept multilevel logistic regression models were used to assess the association between individual- and village-level factors and likelihood of facility delivery.

Results: 1205 women participated in the study. In the fully adjusted two-level model, in addition to several individual factors, positive village perception of doctor and nurse skills (odds ratio (OR) 6.72, 95% confidence interval (CI): 2.47-18.31) and negative perception of traditional birth attendant skills (OR 0.13, 95% CI: 0.04-0.40) were associated with higher odds of facility delivery.

Conclusion: This study suggests that community perceptions of the quality of the local health system influence women's decisions to deliver in a clinic. Improving quality of care at first-level clinics and communicating this to communities may assist efforts to increase facility delivery in sub-Saharan Africa.