Address:

    Room No.: Stat-3161, Ground Floor, Statistics Discipline, Kabi Jibananda Das Academic Building, Khulna University, Khulna-9208, Bangladesh.

    Email:

    office@stat.ku.ac.bd

    Contact:

    +880 1752385769 ; +880 1970385769

    Personal Webpage:
    click here

Exploring the determinants and prevalence of underweight and overweight/obesity among Reproductive-Aged Women in Bangladesh: A Multinomial Multilevel Analysis

The simultaneous occurrence of undernutrition and excessive weight gain poses a significant public health issue in many low- and middle-income countries, including Bangladesh. This research investigates the extent and associated socio-demographic factors of both underweight and overweight/obesity among Bangladeshi women of reproductive age (15–49 years), using data from the 2022 Bangladesh Demographic and Health Survey (BDHS). After excluding incomplete cases, the final sample comprised 9,946 women.

A multinomial multilevel logistic regression model was applied to evaluate the influence of individual and contextual-level variables on nutritional status, categorized by Body Mass Index (BMI) into underweight (<18.5 kg/m²), normal (18.5–22.9 kg/m²), and overweight/obese (≥23 kg/m²). Key determinants of underweight included younger age, lack of formal education, lower economic status, and being currently breastfed. Conversely, overweight and obesity were more prevalent among older, educated women, those with higher household wealth, and greater exposure to media. Rural residents had increased likelihood of undernutrition (OR = 1.31; 95% CI: 1.10–1.56), while urban residency was associated with a higher risk of overweight/obesity. Use of modern contraceptives was linked to a reduced chance of being underweight (OR = 0.71; 95% CI: 0.63–1.08). Notable geographic differences were observed, with women in Sylhet showing higher odds of underweight and lower odds of overweight/obesity compared to those in Barisal. The initial model revealed significant variation at the community level (ICC = 5.03%), which declined to 2.09% in the fully adjusted model. The final model exhibited a strong fit (AIC = 17158.10), with a 76% reduction in community-level variance (PCV = 0.76), highlighting the relevance of contextual influences.

In summary, the findings underscore the need for tailored interventions that address both under- and over-nutrition simultaneously. Multisector strategies that consider both individual behaviors and community-level disparities are essential to alleviate the dual nutritional challenges and advance progress toward achieving Sustainable Development Goal 2.

Details
Role Supervisor
Class / Degree Masters
Students

Suvra Ghosal 

Student ID: MS 242019

Start Date 1 july, 2024
End Date 31 July, 2025