DAILY MORTALITY AND MAJOR CRITERIA POLLUTANTS IN WUHAN CHINA

ABSTRACT

Funded by the US Health Effects Institute as one of the four time-series studies in Asia, this study is to determine the health effects of exposure to ambient air pollutants PM10, SO2, NO2, and O3) on daily mortality in Wuhan, China - a city with 4.3 million residents in its urban core districts, and multiple pollution sources during the period of July 1, 2000 to June 30, 2004. The methodology will include abstracting daily mean concentrations of the pollutants from the Wuhan Environmental Monitoring Center database and matching the data to the nonaccidental mortality and cause-specific mortality data provided by the Wuhan Centres for Disease Prevention and Control. We will perform analyses to examine associations of daily variations in concentrations of each pollutant with daily variations in nonaccidental mortality and cause-specific daily mortality, controlling for known potential confounders/covariates. The main multivariable modeling techniques will be Poisson regression to regress daily death counts. Under the assumption that daily death counts follow a Poisson variate with constant over- or underdispersion, we will use quasi-likelihood estimation within the context of the Generalized Additive Models (exact in S-plus or penalized or natural spline models in R) to model the natural logarithm of the expected daily death counts as a function of the predictor variables. We expect that the proposed study will provide, at high daily pollution levels and over a wide pollution concentration range, data to explore the shape of the exposure-response relationship. The results from the proposed study will provide important new information with profound implications for regulations by the Wuhan local government, Chinese government, and U.S. Environmental Protection Agency.