Provision of a liquefied petroleum gas cookstove and fuel during pregnancy and infancy and linear growth trajectories between birth and 12 months: Evidence from the multi-center Household Air Pollution Intervention Network (HAPIN) trial.

Exposure to particulate pollution from cooking with solid biomass fuels is associated with impaired child linear growth. We examined the effect of a liquefied petroleum gas (LPG) cookstove randomized controlled trial during pregnancy and infancy on linear growth trajectories among infants born to women enrolled during pregnancy. The Household Air Pollution Intervention Network (HAPIN) randomized controlled trial enrolled 3195 pregnant women (9 to <20 weeks' gestation) from rural areas in Guatemala, Peru, India, and Rwanda who relied primarily on biomass fuels for cooking. Women in the intervention group received an LPG cookstove and fuel for approximately eighteen months, while those in the control group continued to use biomass for cooking. We measured the infants' recumbent length at birth and 3, 6, 9 and 12 months of age and calculated length-for-age z-score (LAZ). We conducted a multiple group latent class growth analysis among the 2802 infants who finished the study and had ≥ 3 length measurements across the five timepoints to examine if latent classes differed by study arm. We identified three latent classes of linear growth, based on visual inspection of mean LAZ and model fit statistics, which represent higher, medium, and lower LAZ trajectories. Approximately 13.2% of infants belong to the high LAZ trajectory, 53.8% of infants belong to the medium LAZ trajectory and 33.0% belong to the low LAZ trajectory. The distribution of infants in each latent class did not differ by intervention assignment. Provision of an LPG cookstove and fuel during pregnancy and infancy did not alter linear growth trajectories among the offspring. Clinical Trials Registration Number ClinicalTrials.gov NCT02944682.