Analyzing Overdispersed Skilled Antenatal Care Visits of Pregnant Women in Bangladesh Using Generalized Poisson Regression Model

Abstract

It is common to use the Poisson regression model (PRM) for analyzing the count data. The major limitation of this PRM is that the mean and variance of the response variable need to be equal. However, in the actual dataset, the response variable’s variance may exceed the mean, introducing overdispersion in the dataset. In this paper, the generalized Poisson regression model (GPRM) has been used in the modeling and analysis of the skilled antenatal care (ANC) count data extracted from the Bangladesh Demographic and Health Survey (BDHS) 2017-18. The findings of this study have revealed several socioeconomic and demographic variables that significantly impact the skilled ANC visits. Dhaka Univ. J. Sci. 71(1): 69-75, 2023 (Jan)