SO₂ levels and perturbations in mortality: a study in the New York - New Jersey metropolis

Analysis of daily mortality for 422 places in the United States from 1962 to 1966 provided a consistent set of mortality predictors of three classes: annual cycle, day of week, and Christmas holidays; influenza epidemics; and days or spells of extreme cold or heat. Analysis of deaths for New York metropolitan region, with daily SO2 measurements in Manhattan at the Davis Laboratory, showed the pollution measurement contributes significantly to prediction of deaths. Mortality was 1.5% less than expected on 232 days with SO2 levels below 30μg/cu m and 2% greater than expected on 260 days with SO2 levels above 500μg/cu m, after correcting for the other factors. Coefficient of haze (COHS) does as well as SO2 in predicting death. A parallel analysis in the Philadelphia area showed similar, but weaker, SO2 and COHS effect.

Duke Scholars

Author Victor Hasselblad Biostatistics & Bioinformatics, Division of Biostatistics