Determinants of serum aflatoxin B1 exposure among pregnant women in coastal Kenya: a cross-sectional study in Kwale County

Abstract

Aflatoxin B1 (AFB1) exposure is a major environmental health concern among vulnerable groups such as pregnant women, yet its determinants in coastal settings remain poorly understood. We aimed to quantify AFB1 exposure and examine its environmental, behavioral, and sociodemographic determinants among pregnant women in Kwale County, a predominantly rural area on Kenya’s southern Indian Ocean coast. In a facility-based cross-sectional study, serum AFB1–albumin adduct concentrations were measured by enzyme-linked immunosorbent assay in 298 pregnant women. Determinants were examined with a pre-specified multivariable linear regression of log-transformed concentrations using heteroscedasticity-consistent (HC3) standard errors, and a secondary Firth penalized logistic model of detectable versus non-detectable exposure. Adducts were detectable in 251 participants (84.2%); the median concentration was 28.1 pg/mL (IQR 13.7–68.7). Exposure was determined primarily by where women lived: it was lowest in the community closest to the coast and progressively higher in two drier, more inland communities (2.02-fold, 95% CI 1.44–2.85; and 3.02-fold, 95% CI 1.89–4.82). Single marital status was associated with modestly higher exposure (1.89-fold; 95% CI 1.17–3.05). Food source, flour storage, maternal education, and age did not predict exposure, and the penalized logistic model reproduced the same geographical pattern. Despite near-universal exposure, only 1.7% of participants had ever heard of aflatoxin. AFB1 exposure was widespread and geographically structured, pointing to place-based rather than individual-level drivers and supporting area-targeted food-safety interventions to protect maternal and fetal health. These findings have implications for food-safety strategies in other tropical regions where climate and subsistence agriculture contribute to chronic aflatoxin exposure.