|
Activity Number:
|
49
|
|
Type:
|
Invited
|
|
Date/Time:
|
Sunday, August 2, 2009 : 4:00 PM to 5:50 PM
|
|
Sponsor:
|
Section on Statistics in Epidemiology
|
| Abstract - #302919 |
|
Title:
|
Statistical Approaches to Quantifying and Checking the Health Effects of Particulate Air Pollution
|
|
Author(s):
|
Scott Zeger*+ and Francesca Dominici and Roger D. Peng
|
|
Companies:
|
Johns Hopkins University and Johns Hopkins University and Johns Hopkins Bloomberg School of Public Health
|
|
Address:
|
Bloomberg School of Public Health, Department of Biostatistics, Baltimore, MD, 21205,
|
|
Keywords:
|
log-linear regression ; regression diagnostics ; time series ; space time models ; decomposition ; relative risk
|
|
Abstract:
|
Evidence that particulate air pollution causes hospitalizations and premature death derives from associations of pollution and health outcomes at a variety of temporal and spatial scales. As in any observational study, confounding by unmeasured factors is a significant threat to the validity of inferences. In this talk, we review the major evidence in support of the hypothesis that particulate pollution increases the risk of disease and death and discuss the major statistical methods used to quantify this excess risk. We present novel methods for decomposing risk estimates into contributions from different temporal and spatial scales to check for consistency and the possible influence of confounders.
|
