Effects of Cooking with Liquefied Petroleum Gas or Biomass Fuels on Neurodevelopmental Outcomes in Preschool-Age Children Living in Puno, Peru

Abstract

BACKGROUND: Burning biomass fuels for cooking is a widespread environmental exposure that may adversely affect child health. We conducted an 18-month randomized field trial comparing a liquefied petroleum gas (LPG) stove, continuous fuel delivery, and behavioral messaging intervention starting in pregnancy through infancy with usual cooking practices using biomass fuels in Puno, Peru. A total of 800 pregnant women were enrolled. The intervention successfully lowered personal exposures to fine particulate matter with aerodynamic diameter ≤2:5 lm (PM2:5 ) during pregnancy and infancy. OBJECTIVE: We sought to evaluate the effects of the intervention on neurodevelopment among preschool-age offspring. METHODS: We conducted a longitudinal follow-up study in a subset of Peruvian children born during the trial and assessed neurodevelopment between 24 and 36 months of age. Trained staff measured neurodevelopment using the Bayley Scales of Infant and Toddler Development Third Edition (Bayley-III). The Bayley-III neurodevelopmental assessment test yields separate cognition, language, and motor scores. We measured personal exposures to PM2:5 three times during pregnancy and three times in infancy. We conducted modified intention-to-treat analyses of the intervention on Bayley-III scores and exposure–response analyses between prenatal and postnatal PM2:5 exposures and Bayley-III scores. RESULTS: A total of 301 children (mean age 29:0±3:5 months, 47.8% girls, and 54.8% in intervention) were visited. Mean cognitive, language, and motor scores were 90:8±11:0, 89:0±10:8, and 95:3±12:9, respectively. On intention-to-treat, the adjusted differences between intervention and control arms were −1:9 [98.3% confidence interval (CI): −4:94, 1.15], −2:94 (−5:98, 0.1), and −1:35 (−4:99, 2.29) for the composite cognitive, language, and motor scores, respectively. We did not identify associations between PM2:5 exposures during the 18-month intervention and any of the Bayley-III scores or when the personal exposures to PM2:5 were separated as either prenatal or postnatal. DISCUSSION: The LPG intervention did not improve neurodevelopment in Peruvian children. We also did not find an association between prenatal or postnatal PM2:5 exposures and neurodevelopment. https://doi.org/10.1289/EHP15500.