Accounting for overdispersion in skilled antenatal care: Identifying determinants using Bangladesh Demographic and
Md Muddasir Hossain Akib1, Farhia Azrin2, Bikash Pal1
1Department of Statistics, University of Dhaka, Dhaka, Bangladesh.
Abstract:
One of the major components required to ensure safe motherhood is taking a sufficient number of skilled antenatal care (SANC) visits by women during pregnancy. Various social, demographic, and economic factors have a strong influence on it. According to the World Health Organization (WHO), four or more antenatal care (ANC) visits are required for uncomplicated pregnancies and delivery of a healthy baby. SANC serves this purpose efficiently. As a result, modeling the number of SANC visits and identifying its determinants is crucial for the Bangladesh government in adopting appropriate policies and programs. In this study, we used score tests to assess dispersion in the count data and selected the Generalized Poisson Regression Model (GPRM) based on the lowest AIC (15,839.2) and BIC (16,001.7) values. Analysis of BDHS 2022 data (n = 3,839) revealed that only 38.6% of women received the recommended four or more SANC visits, with an overall mean of 3.06 visits. Key predictors of higher SANC utilization included urban residence (IRR = 1.09, p < 0.01), higher women's education (IRR = 1.57, p < 0.001), exposure to media (IRR = 1.16, p < 0.001), and being in the richest wealth quintile (IRR = 1.22, p < 0.001). In contrast, women in the Sylhet region (IRR = 0.89, p = 0.009) and those with higher birth order had lower SANC visit rates.
More Related Videos
05:35Author Spotlight: Developing a Point-of-Care Hemoglobin Estimation Method for Anemia Management
Published on: January 19, 2024
06:55Inverse Probability of Treatment Weighting Propensity Score using the Military Health System Data Repository and National Death Index
Published on: January 8, 2020
Related Concept Videos
Regression Toward the Mean
Statistical Methods for Analyzing Epidemiological Data
Bias in Epidemiological Studies
Confounding in Epidemiological Studies
Steps in Outbreak Investigation
Analysis of Population Pharmacokinetic Data
