Spatial Variation and Predictors of Women's Sole Autonomy in Healthcare Decision-Making in Bangladesh: A Spatial and Multilevel Analysis.

Knowing the spatial variation and predictors of women having sole autonomy over their healthcare decisions is crucial to design site-specific interventions. This study examined how women's sole autonomy over their healthcare choices varies geographically and what factors influence this autonomy among Bangladeshi women of childbearing age. Data were obtained from the Bangladesh Demographic and Health Survey (BDHS) 2017-18. The final analysis included data from a total of 18,890 (weighted) women. Spatial distribution, hot spot analysis, ordinary Kriging interpolation, and multilevel multinomial regression analysis were employed. The study found that approximately one in ten women (9.62%) exercised complete autonomy in making decisions about their healthcare. Spatial analysis revealed a significant clustering pattern in this autonomy (Moran's I = 0.234, < 0.001). Notably, three divisions-Barisal, Chittagong, and Sylhet-emerged as hot spots where women were more likely to have sole autonomy over their healthcare choices. In contrast, the cold spots (poor level of sole healthcare autonomy by women) were mainly identified in Mymensingh and Rangpur divisions. Women in the age group of 25-49 years, who were highly educated, Muslim, urban residents, and had not given birth recently were more likely to have sole autonomy in making healthcare decisions for themselves. Conversely, women whose husbands were highly educated and employed, as well as those who were pregnant, were less likely to have sole autonomy over their healthcare choices. Since the spatial distribution was clustered, public health interventions should be planned to target the cold spot areas of women's sole healthcare autonomy. In addition, significant predictors contributing to women's sole healthcare autonomy must be emphasized while developing interventions to improve women's empowerment toward healthcare decision-making.