Statistical analysis of air pollution panel studies: an illustration
Holly Janes1, Lianne Sheppard, Kristen Shepherd
1Division of Public Health Sciences, Fred Hutchinson Cancer Research Center, University of Washington, Seattle, USA. hjanes@scharp.org
Purpose:
The panel study design is commonly used to evaluate the short-term health effects of air pollution. Standard statistical methods are available for analyzing longitudinal data, but the literature reveals that these methods are poorly understood by practitioners.
Methods:
We review standard statistical methods for modeling longitudinal data. Marginal, conditional, and transitional approaches are reviewed and contrasted with respect to their parameter interpretation and methods for accounting for correlation and dealing with missing data. We also discuss techniques for controlling for time-dependent and time-independent confounding and for exploring and summarizing panel study data. Notes on available software are provided.
Results:
These methods are illustrated by using data from the 1999 to 2002 Seattle Panel Study.
Conclusions:
The quality of statistical analyses and presentation of results of panel studies could be improved if the methods we present were followed.
Related Concept Videos
Sampling Plans
Random sampling is a method where each member of the population has an equal chance of being selected for the sample. It involves selecting individuals randomly, often using random number generators or lottery-type methods. For example, when analyzing the properties of a...
Statistical Methods for Analyzing Epidemiological Data
Linear Approximations
Mechanistic Models: Compartment Models in Individual and Population Analysis
Statistical Methods to Analyze Parametric Data: ANOVA
One-way ANOVA is applied when a single independent variable or factor is scrutinized. It compares the...
Data Collection by Observations
An astronomer viewing the motion and brightness of stars in the sky and recording the data is an example of observational data collection. A botanist recording...

