Abstract
BACKGROUND: Aflatoxins are found in diverse foods widely consumed worldwide. This study investigated the association between aflatoxin exposure and (a) consumption of specific foods, (b) dietary diversity (DD), and (c) seasonality. METHODS: Women enrolled in the AflaCohort Study in Banke, Nepal (n = 1648) were asked how often they ate certain food items in the past 7 days and 24 h. Serum aflatoxin B(1)-lysine (AFB(1)-lys) adduct levels, measured during pregnancy, were determined using high-performance liquid chromatography. Multivariable ordinary least squares and quantile regression models were used to examine incremental increases in AFB(1)-lys adduct levels per frequency of food consumption and the relationship between DD, seasonality, and increases in AFB(1)-lys adduct. RESULTS: Roughly 94% of women were exposed to aflatoxin (geometric mean 1.37 pg/mg). Women in the 30th, 50th, and 70th quantiles of aflatoxin exposure who reported one more occasion of maize consumption in the past week showed increases in AFB(1)-lys adduct levels: 0.094, 0.112, and 0.109 pg/mg (p < 0.05, all). Women in the 30th, 50th, 70th, and 90th quantiles of exposure who reported one more occasion of groundnut consumption in the past week also showed increases in AFB(1)-lys adduct levels: 0.058 (p < 0.001), 0.085 (p < 0.01), 0.133 (p < 0.001), and 0.133 (p < 0.001) pg/mg. Winter month recruitment was positively associated with AFB(1)-lys adduct levels at all quantiles of aflatoxin exposure (range: 0.313-1.101 pg/mg, p < 0.001). DD was not predictive of aflatoxin exposure. CONCLUSIONS: Our findings justify integrated approaches to aflatoxin reduction, including regulatory, agricultural, and food safety interventions across the value chain and at the household level.