Effects of a liquefied petroleum gas stove and fuel intervention on head circumference and length at birth: A multi-country household air pollution intervention network (HAPIN) trial

Abstract

Background

Air pollution may impair child growth and cognitive development, with potential markers including birth length and head circumference.

Methods

The Household Air Pollution Intervention Network (HAPIN) trial was an open label multi-country-randomized controlled trial, with 3200 pregnant women aged 18–34 years (9–19 weeks of gestation) randomly assigned in a 1:1 ratio to receive liquefied petroleum gas (LPG) stove intervention compared to women continuing to cook with solid fuels for 18 months. Particulate matter ≤ 2.5 μm (PM2.5), black carbon (BC) and carbon monoxide (CO) 24-hour personal exposures were measured three times during pregnancy. Head circumference and length were measured < 24 h of birth. We conducted intention-to-treat and exposure–response analyses to determine the intervention effects and associations between household air pollution (HAP) exposure during pregnancy and head circumference, head circumference-for-gestational age Z-score, length, and length-for-gestational age Z-scores at birth. The study is registered with ClinicalTrials.gov (Identifier NCT02944682).

Results

Between May 2018, and Feb 2020, 3200 pregnant women were randomly assigned to intervention (n = 1593) and control groups (n = 1607) with 3060 births included in the analysis. There was a 71.9 % reduction in PM2.5 in the intervention group with similar reductions for BC and CO. Intention-to-treat analysis showed that the intervention did not affect head circumference (β = -0.01 cm, 95 %CI −0.11, 0.09), head circumference-for-gestational age Z-score (β = -0.01, 95 %CI −0.08, 0.07), or birth length (β = 0.14 cm, 95 %CI −0.01, 0.29) but did increase birth length-for-gestational age Z-score (β = 0.09, 95 %CI 0.01, 0.16). After covariate adjustment, exposure–response analysis revealed that each log-unit increase in BC was associated with a decrease in birth length-for-gestational age Z-score (β = -0.07, 95 %CI −0.13, −0.005). There was no evidence of hypothesized associations with PM2.5 or CO.

Conclusion

An LPG intervention reduced HAP exposure during pregnancy but had minor effects on birth length-for-gestational age Z-score. Birth length-for-gestational age was only associated with BC.Clinical Trial Registration: The study has been registered with ClinicalTrials.gov (Identifier NCT02944682).