THE ASSESSMENT OF PERSONAL EXPOSURE TO NITROGEN-DIOXIDE IN EPIDEMIOLOGIC STUDIES

Personal and indoor exposure to nitrogen dioxide (NO2) were studied in a population of housewives and children. Personal exposure, indoor concentrations and peak exposures were measured using Palmes' diffusion tubes during three measurement periods of 1 week within 1 year. Very high peak concentrations were found in the kitchens (up to 2000-mu-g m-3). Three different estimates of personal exposure were constructed using indoor concentrations and time budget data. All three were found to be closely associated with measured personal exposure. The variability over time of personal and indoor concentrations, and of the three exposure estimates was investigated in a one-way analysis of variance. The within-location or within-subject variances of indoor concentrations and estimated personal exposures were found to be smaller, relative to the between-location or between-subject variances, than the within-subject variance of measured personal concentrations. The close association between measured indoor concentrations (or exposure estimates based on indoor concentrations) and measured personal exposure suggests that measuring indoor concentrations of nitrogen dioxide in the home is sufficient to estimate personal exposure accurately. Short term peak concentrations were found to be only weakly associated with indoor NO2 sources and weekly average indoor NO2 concentrations. Repeated exposure to short term peak concentrations has been suggested to be more harmful than long term exposure to elevated NO2 concentrations at a lower level. This suggests that diffusion type measurements of personal and indoor exposure to NO2, which are widely used in epidemiologic studies, may not adequately reflect the most biologically relevant exposure.