Childhood exposure to PM10: relation between personal, classroom, and outdoor concentrations.

OBJECTIVES To investigate the validity of outdoor concentrations of particulate matter < 10 microns diameter (PM10) as a measure of exposure in time series studies, and to study the extent to which differences between personal and outdoor PM10 concentrations can be explained. METHODS Four to eight repeated measurements of personal and outdoor PM10 concentrations were conducted for 45 children, aged 10-12 years, from four schools in Wageningen and Amsterdam, The Netherlands. Repeated PM10 measurements in the classrooms were conducted in three of the schools. Averaging time was 24 hours for the personal and outdoor measurements, and eight hours (daytime) and 24 hours for the classroom measurements. For each child separately, personal exposures were related to outdoor concentrations in a regression analysis. The distribution of the individual correlation and regression coefficients was investigated. Information about factors that might influence personal exposures was obtained by questionnaire. RESULTS Median Pearson's correlations between personal and outdoor concentrations were 0.63 for children with parents who did not smoke and 0.59 for children with parents who smoked. For children with parents who did not smoke, excluding days with exposure to environmental tobacco smoke (ETS) improved the correlation to a median R of 0.73. The mean personal PM10 concentration was 105 micrograms/m3; on average 67 micrograms/m3 higher than the corresponding outdoor concentrations. The main part of this difference could be attributed to exposure to ETS, to high PM10 concentrations in the classrooms, and to (indoor) physical activity. CONCLUSIONS The results show a reasonably high correlation between repeated personal and outdoor PM10 measurements within children, providing support for the use of fixed site measurements as a measure of exposure to PM10 in epidemiological time series studies. The large differences between personal and outdoor PM10 concentrations probably result from a child's proximity to particle generating sources and particles resuspended by personal activities.