Quantitative metrics of exposure and health for indoor air pollution from household biomass fuels in Guatemala and India.

摘要

Approximately 3 billion people globally rely on solid fuels (e.g. wood, crop residue, animal dung, coal, etc.) for their everyday cooking and heating needs. Exposure to IAP translates into 1.6 millions deaths and over 38.5 million disability-adjusted life years attributable annually to indoor air pollution (IAP) resulting from indoor smoke from solid fuels. To facilitate more accurate IAP exposure assessment for better understanding health effects, I explore here (1) the quantitative implications of selecting different exposure assessment strategies within a framework called the Exposure Assessment Pyramid, (2) low-cost exposure assessment instruments and lung function metrics for use in large-scale IAP epidemiology studies, and (3) a novel methodology to translate particulate matter exposure reductions into corresponding health benefits.;The Exposure Assessment Pyramid is a conceptual framework that organizes the available exposure metrics for one particular hazard--here IAP in less-developed-countries countries--along a spectrum from imprecise but easy at the base to accurate but difficult at the top. Within this framework, it is hypothesized that the various exposure metrics are correlated and that the ones located in close proximity on the Pyramid are better correlated. Using five proxy personal carbon monoxide (CO) exposure metrics collected from repeated measurements of 58 woman in highland Guatemala, the first objective of this dissertation was to compare trade-offs in accuracy among different exposure assessment metrics of the Exposure Assessment Pyramid. Overall, the strength of the correlations between continuous personal CO measurements and simultaneously collected indirect and microenvironmental CO measurements ranged from R2 = 0.29 -- 0.42, adjusted for stove type, fuel type, roof type, available assets, and/or father's education.;In the second objective of this dissertation, 24-hour kitchen IAP concentrations and lung functions were measured in a cross-sectional study of 617 homes located in four geographically dispersed states in India with the primary objective of quantifying the impact of IAP exposure on pulmonary function and respiratory symptoms. Significant differences in the mean percent predicted forced expiratory volume in 1 second (FEV1) and forced vital capacity (FVC) were detected among males in solid-fuel-using and clean-fuel-using households in Tamil Nadu as well as among all male clean fuel users across the four states. The strongest effects on lung function were detected relative to high concentrations of PM2.5; in this population, peak PM2.5 exposures were inversely correlated with FEV1 and FVC in women. How much of the effect is acute and potentially reversible and how much of it is long-term reduction that cannot be recovered remains unclear.;The final objective of this dissertation was to determine if exposure--response data from both IAP and outdoor air pollution (OAP) epidemiology studies could be used to estimate a reasonable IAP exposure--response relationship so that the risks due to IAP exposures that are intermediate between full exposure from open fires and low exposures from clean fuels could be determined. The resulting estimates were assessed by determining (1) how well the estimated exposure--response relationships predicted the World Health Organization's (WHO) estimates of acute lower respiratory infections (ALRI) and chronic obstructive pulmonary disease (COPD) mortality in India and Guatemala and (2) the amount of variability between predictions of ALRI mortality in children five years and COPD mortality in women >30 years in study populations in India and Guatemala. Compared to the WHO's estimate for ALRI mortality, the OAP--derived exposure--response relationships overestimated the percent decrease in risk of ALRI mortality compared to the IAP--derived exposure--response relationships. Compared to the WHO's estimate for COPD mortality, however, the OAP--derived exposure--response relationships displayed no consistent relationship to the IAP--derived exposure--response relationship. Part of this discrepancy may be due to the way the OAP--derived exposure--response relationship was truncated.;I recommend that both researchers and promoters of improved stoves who desire to estimate the potential reduction in ALRI and COPD mortality associated with an improved biomass stove project(s) use both the IAP and OAP exposure--response relationships. (Abstract shortened by UMI.)