NATIONAL MORTALITY, MORBIDITY AND AIR POLLUTION STUDY: 
                     STATISTICAL CHALLENGES

                     Francesca Dominici

                     Department of Biostatistics,
                     Johns Hopkins University
                     
                      fdominic@jhsph.edu

Time series studies have shown associations between air pollution 
concentrations and morbidity
and mortality. These studies have largely been conducted within single cities 
with varying methods.
Key questions remain unaddressed concerning the findings, including 1) the 
extent of heterogeneity
of air pollution effects across locations and its sources; 
2) the public health significance of the short-term associations (harvesting); 
and
3) the effect of measurement error on the estimated effect of air pollution.
NMMAPS comprises the development of statistical methods to address these 
questions and their application
to national data sets on mortality and hospitalization among persons 65 years 
of age and older,
as an index of morbidity. 

For analyzing data from multiple locations, we presentlog-linear regression analyses of daily time-series data from the largest 20 
U.S. cities,
and we introduce hierarchical regression models 
for combining estimates of the pollution-mortality relationship.
For addressing  ``harvesting'' in air pollution studies, we propose a novel 
statistical strategy
based on frequency domain log-linear regression methods. Here we decompose the 
information about the pollution-mortality
association into distinct time scales and we then create harvesting-resistant
estimates by excluding the short-term information which is affected by 
harvesting.
Finally, for evaluating the effects of measurement error, we present 
hierarchical longitudinal
models with missing data, and we apply the method to five epidemiological 
studies on personal
and ambient concentrations of PM10.


Data bases have been assembled on mortality of the 100 largest U.S. cities. In 
the next phase of NMMAPS,
the morbidity and mortality analyses will be completed and the combined 
analyses of morbidity
and mortality will be undertaken. The methods of NMMAPS should prove useful 
for future surveillance
for the health effects of air pollution.

--- joint work with Jonathan Samet and Scott L. Zeger.