AIR-POLLUTION AND MORTALITY - THE IMPLICATIONS OF UNCERTAINTIES IN REGRESSION MODELING AND EXPOSURE MEASUREMENT

In a previous paper,(1) we showed that the mean effects on daily morta lity associated with air pollution are essentially the same for gases and particulate matter (PM) and are invariant with respect to particle size and composition, based on 27 statistical studies that had been p ublished at that time. Since then, a new analysis(2) reported stronger mortality associations for the fine fractions of PM obtained from dic hotomous samplers, relative to the coarse fractions. In this paper, we show that differential measurement errors known to be present in dich otomous sampler data preclude reliable determination of such statistic al relationships by particle size. Further, it is necessary to conside r gaseous pollutants simultaneously with particles to provide robust e stimates of the responsibilities for the implied daily mortality gradi ents. Finally, certain regression model specifications may be sensitiv e to differences in frequency distribution characteristics according t o particle size.