Search

Epidemiological studies have associated chronic exposure to arsenic (As) from drinking water with increased risk of hypertension. However, evidence of an association between As exposure from food and hypertension risks is sparse. To quantify the association between daily As intake from both food (rice, wheat and potatoes) and drinking water (As water) along with total exposure (As total) and hypertension risks in a study population in Bihar, India, we conducted an individual level cross-sectional analysis between 2017 and 2019 involving 150 participants. Arsenic intake variables and three indicators of hypertension risks (general hypertension, low-density lipoprotein (LDL) and high-density lipoprotein (HDL)) were derived, and any relationship was quantified using a series of crude and multivariable log-linear or logistic regression models. The prevalence of general hypertension was 40% for the studied population. The median level of HDL was 45 mg/dL while median value of LDL was 114 mg/dL. Apart from a marginally significant positive relationship between As intake from rice and the changes of LDL (p-value = 0.032), no significant positive association between As intake and hypertension risks could be ascertained. In fact, As total was found to be associated with lower risks of general hypertension and higher levels of HDL (p-value = 0.020 and 0.010 respectively) whilst general hypertension was marginally associated with lower As water (p-value = 0.043). Due to limitations regarding study design and residual confounding, all observed marginal associations should be treated with caution. • Cross-sectional study of hypertension risks associated with arsenic exposure from food and drinking water. • Total median arsenic intake of 36 µg/day for the study population in Bihar, India. • Total arsenic intake marginally associated with lower risks of general hypertension and higher levels of HDL. • Marginal positive relationship between arsenic intake from rice and level of LDL. • Limitations regarding study design and residual confounding problems should not be neglected. [ABSTRACT FROM AUTHOR]