Evaluating the impacts of an advanced stove intervention on traditional cooking patterns and pollutant exposures in India using Stove Use Monitors

Abstract

Background/Aims. Household air pollution in 2010 was responsible for ~3.5 million premature deaths and 500,000 through contributions to outdoor air pollution. 900,000 of these deaths occurred in India, where ~66% of households rely on wood, dung, or crop residues as their primary cooking fuels. Since alternative fuels are unlikely to become available soon, cleaner ways of using these fuels need evaluation. Methods. Beginning in 2012 as part of ongoing work at the SOMAARTH surveillance area in Palwal district, Haryana, we distributed 200 forced-draft ('advanced') biomass stoves to pregnant women. These stoves have low laboratory emissions and good field evaluations. Using Stove Use Monitors (SUMs), portable data-logging thermometers, we monitored usage of traditional and intervention stoves. This revealed how use patterns changed after advanced stove introduction. In a subset of homes, we monitored PM2.5 and carbon monoxide (CO) in kitchens and personal CO exposures of participants. SUMs were placed on traditional stoves up to 2 weeks before SUMs-equipped advanced stove dissemination. Prior to and following advanced stove distribution, pollutant concentrations were assessed over 48-hour periods in a subset of homes. SUMs logged temperatures every 10 minutes and were downloaded twice monthly. Nearly 400 stoves continue to be monitored, yielding over 110,000 stove-days of data. Results/ Conclusions. Personal CO exposures decreased post intervention from 10.9 ppm to 4.8 ppm. Area PM and CO concentrations by 59.8% and 55.9% respectively. Prior to the intervention, households used traditional stoves twice daily on average - consistent with previous findings. A small proportion of households used advanced stoves only. The majority used both stoves, a phenomenon termed "stacking," which has profound implications for intervention strategies, as traditional stoves dominate household emissions and drive personal exposures. Mixed effects models were used to evaluate the contribution of stove type and household characteristics to area concentrations and personal exposures.