Exposure to PM2.5 through Cookstoves in Central Sri-Lanka

Abstract

Exposure to PM2.5 through Cookstoves in Central Sri-LankaAbstract Number:2619 Paul Mosquin*, Michael Phillips, Ryan Chartier, Katherine Bronstein, Myles Elledge, Vanessa Thornburg, Thornburg Jonathan, Sumal Nandasena, Charles Rodes Paul Mosquin* RTI International, United States, E-mail Address: [email protected] Search for more papers by this author , Michael Phillips RTI International, United States, E-mail Address: [email protected] Search for more papers by this author , Ryan Chartier RTI International, United States, E-mail Address: [email protected] Search for more papers by this author , Katherine Bronstein RTI International, United States, E-mail Address: [email protected] Search for more papers by this author , Myles Elledge RTI International, United States, E-mail Address: [email protected] Search for more papers by this author , Vanessa Thornburg RTI International, United States, E-mail Address: [email protected] Search for more papers by this author , Thornburg Jonathan RTI International, United States, E-mail Address: [email protected] Search for more papers by this author , Sumal Nandasena National Institute of Health Sciences, Sri Lanka, E-mail Address: [email protected] Search for more papers by this author , and Charles Rodes RTI International, United States, E-mail Address: [email protected] Search for more papers by this author AbstractWe report on a cookstove indoor air pollution study performed in rural Sri Lanka during the summer of 2012. A total of 53 houses were visited each over a period of two days, with data collected including real-time MicroPEM®TM personal and indoor exposures to PM2.5, real-time indoor exposures to carbon monoxide, time activity, heath outcomes, and stove and chimney types. Personal exposures were collected from the household member with primary responsibility for meal preparation. Primary covariates of interest for exposure were stove type (traditional or improved ‘Anagi’) and presence or absence of chimney. Medians of daily indoor exposures within these groups were 371 ug/m3 in the traditional no-chimney group, 183 ug/m3 in the improved no chimney group, 72 ug/m3 in the traditional chimney group and 66 ug/m3 in the improved chimney group. Tests of difference in mean on the log-scale using linear mixed models showed significant chimney effect (p<0.001) and borderline significant stove effect (p=0.051). When indoor exposure levels were low, personal exposures were of comparable level. However, when indoor exposure levels were high, personal exposures tended to be lower. See Figure 1. Additional results will be presented describing the CO and time activity data and further illustrating relationships among the various measures of exposure, and the importance of personal level data.