Environmental heavy metal exposure, socioeconomic inequalities, and childhood anaemia in Bangladesh: a nationally representative, survey-weighted Bayesian hierarchical analysis

Abstract

Background: Childhood anaemia remains a major public health challenge in low- and middle-income countries. Although nutritional deficiencies are well-established contributors, the population-level role of environmental heavy metal exposure remains poorly understood in South Asia. Objectives: To estimate the prevalence, socioeconomic inequalities, and factors associated with childhood anaemia among children aged 12-59 months in Bangladesh, with particular emphasis on environmental heavy metal exposure. Methods: We analysed nationally representative cross-sectional data from the Bangladesh Multiple Indicator Cluster Survey (MICS) 2025, including 10,343 children aged 12-59 months. Blood concentrations of lead, arsenic, cadmium, and mercury were transformed using base-2 logarithms (log2). The primary outcome was binary child anaemia status. Survey-weighted Bayesian hierarchical logistic regression using R-INLA, with primary sampling unit (PSU)-level random intercepts, was the primary analytical approach. Socioeconomic inequalities were evaluated using the Slope Index of Inequality (SII) and Relative Index of Inequality (RII). Population attributable fractions (PAFs) were estimated separately using counterfactual analyses. Results: Overall, 4779 (44.3%) children had anaemia. In multiple Bayesian models, each 2-fold increase in blood arsenic (posterior aOR 1.12, 95% CrI 1.06-1.18; P(aOR>1): >0.99) and blood cadmium (aOR 1.09, 95% CrI 1.04-1.13; P(aOR>1): >0.99) was independently associated with higher odds of childhood anaemia. Conversely, higher blood mercury was inversely associated (aOR 0.82, 95% CrI 0.78-0.86; P(aOR<1): >0.99). Older child age (aOR 0.69 per 13.2 months increase, 95% CrI 0.65-0.72), higher Height-for-Age Z-score (aOR 0.88, 95% CrI 0.84-0.91), maternal higher education (aOR 0.69, 95% CrI 0.56-0.86), and richest wealth status (aOR 0.80, 95% CrI 0.66-0.97) were associated with lower anaemia odds. Anaemia was more common among children from the poorest households (SII: -12.39%, 95% CI -17.62, -7.17), while anaemia prevalence was lower among wealthier households (RII: 0.76, 95% CI 0.67-0.85). Interpretation: Childhood anaemia in Bangladesh is associated with environmental toxicant exposures, nutritional deficits, and structural socioeconomic disadvantage. The findings highlight the potential importance of integrating nutritional, environmental, and socioeconomic considerations within strategies to address the high burden of child anaemia in Bangladesh.