Statistical Methods

This page has a list of publications and news articles related to Innovative Strategies - Statistical Methods. Find more information on Innovation in HEI's research programs.

Research Report 123
Francesca Dominici
2004

This report describes a study funded under the Walter A. Rosenblith New Investigator Award. Dr Francesca Dominici and colleagues at Johns Hopkins University developed more flexible methods and statistical models for the National Morbidity, Mortality, and Air Pollution Study database.

Special Report
Health Effects Institute
2003

Over the past decade, time-series studies conducted in many cities have contributed information about the association between daily changes in concentrations of airborne particulate matter (PM) and daily morbidity and mortality. In 2002, however, investigators at Johns Hopkins University and at Health Canada identified issues in the statistical model used in the majority of time-series studies. This HEI Special Report details attempts to address several questions raised by these discoveries.

Research Report 94-I
Jonathan M Samet
Francesca Dominici
Scott L Zeger
Joel Schwartz
Douglas W. Dockery
2000

In an effort to address the uncertainties regarding the association between PM and daily mortality, and to determine the effects of other pollutants on this association, HEI funded the National Morbidity, Mortality, and Air Pollution Study (NMMAPS). Dr Jonathan Samet and his colleagues at Johns Hopkins University, in collaboration with investigators at Harvard University, conducted this time-series study in large cities across the US where levels of PM and gaseous pollutants were varied.

Research Report 86
William Navidi
Duncan Thomas
Bryan Langholz
Daniel Stram
1999

Dr. Navidi and colleagues at the University of Southern California discussed the development of three sophisticated statistical methods that would improve the estimates of the health effects of air pollution obtained from epidemiologic studies. First, they took a standard case-crossover design and introduced a bidirectional element where control data were obtained both before and after the health event of interest.