Correlations of personal exposure to particles with outdoor air measurements: A review of recent studies

Epidemiological studies have found a correlation between daily mortality and particle concentrations in outdoor air as measured at a central monitoring station. These studies have been the central reason for the U.S. EPA to propose new tighter particle standards. However, personal monitoring studies have generally found very poor correlations (often approaching 0) between personal exposures and outdoor air concentrations. This has led to questions regarding the scientific basis of the new EPA standard. One reason for the poor correlations noted in the personal monitoring studies is that the correlations were cross-sectional: that is, they included many persons (generally monitored for only 1 or a few days) with no differentiation between persons exposed to high indoor sources (such as smokers) and those with much smaller exposures. Some recent studies have improved on this situation by following persons for a number of days and performing longitudinal correlations on individuals over time. These longitudinal correlations are almost invariably much higher than cross-sectional correlations. A second possible reason for the low correlations of personal exposure with outdoor concentrations is that most studies have been done on healthy persons who may be much more active and therefore generate more particles than sick persons. Therefore a second set of studies has begun to be carried out looking at high-risk subpopulations, such as those with existing respiratory or cardiovascular disease. A common element in these studies is the existence of a "personal cloud," an increased personal exposure beyond what is calculated from a time-weighted average of indoor and outdoor concentrations. The personal cloud for healthy persons can be quite sizable, on the order of 50 mu g/m(3) during the day when people are active and may be a major reason for the poor correlations of personal exposure with outdoor air concentrations. If the personal cloud of high-risk subpopulations is considerably smaller than that of healthy populations and if indoor sources are reduced in the impaired populations, then better correlations of personal exposure with outdoor air could be expected. Recent studies that have quantitated the personal cloud for both healthy and sick populations are reviewed.