Small Area Estimation of Antenatal Care (ANC) Coverage in Bangladesh: An Application of Multinomial Multilevel Modelling

Bernard Baffour, Australian National University
Sumonkanti Das, Australian National University
Ashis Taludker, Australian National University
Syed Abdul Basher, East West University
Alice Richardson, The Australian National University ACT 2601

This paper presents a small area estimation approach to assess antenatal care (ANC) coverage among reproductive-age women in Bangladesh. A multinomial multilevel logistic model is developed using district-level ANC coverage data, categorizing women into three groups: (i) those with no ANC visits (ANC0), (ii) those with 1-3 visits (ANC123), and (iii) those with four or more visits (ANC4+), as recommended by the World Health Organization (WHO). The model accounts for the distinct behavioral profiles of each group in relation to factors influencing maternal health coverage. In Bangladesh, as in many developing countries, the health system operates under a decentralized structure, making district-level statistics crucial for localized planning and decision-making to ensure health equity. Our proposed model-based estimators incorporate significant social, economic, and geographical disparities, identifying areas of improvement, stagnation, and decline. The model-based estimates demonstrate improved precision over direct estimates and provide valuable insights for designing targeted interventions to maximize impact. Additionally, a spatio-temporal extension of the multinomial multilevel logistic model is expected to yield further refined estimates, helping to uncover disparities in the trends of maternal health care access in Bangladesh.

Keywords: Small area estimation, Sexual and Reproductive Health and Rights, Multi-level modeling

See extended abstract.