The Relationship Between Central Site, Home Outdoor and Indoor Concentrations of Fine and Ultrafine Particles: RUPIOH Study

Abstract

P-598 Introduction: Epidemiological time series studies typically assess exposure by measurements of air pollution at a fixed central site. Various studies have documented that this is a reasonable approach for fine particles (PM2.5), but very little information is available of the validity for ultrafine particles. We therefore conducted a study to improve exposure assessment for ultrafine particles. The study consisted of determination of the spatial variation and indoor/outdoor relationship for PM10,, PM2.5 and total particle number counts. Methods: The study was conducted in Amsterdam, Athens, Birmingham and Helsinki. At a central site, PM2.5, PM10 and total particle number counts was measured in outdoor air continuously. In each city, between 30 and 37 non-working, non-smoking patients with asthma or chronic obstructive pulmonary disease (COPD) were recruited. For each subject, measurements of PM2.5, PM10 and particle number counts inside the home and directly outside the home were conducted during one week. The air exchange rate of each home was measured. A time activity diary was kept to collect information about ventilation patterns and indoor particle sources. Results: Concentrations measured at the central site were highly correlated with concentrations measured near the homes for all pollutants, including particle numbers. Thus, a central site is a good estimate for the temporal variation. However, absolute concentration levels of particle number differed substantially between homes and central site. This suggests that it is virtually impossible to characterize the city-average concentration of PNC with one site. Considerably smaller spatial variability was found for PM2.5 and PM10. Daily average particle number concentrations measured at a central site were poorly to moderately correlated with indoor concentrations. In three of the cities, the correlation of PM2.5 at the central site and indoors was substantially higher. Very high indoor-outdoor correlations were found for fine particle sulfate and soot. Hourly average particle number concentrations during nighttime hours (‘non-source period“), showed moderately high correlations between indoor and outdoor concentrations, suggesting that indoor sources were partly responsible for the difference in correlation between particle number and particle mass / sulfur /soot. Regression slopes, describing the quantitative relationship between indoor and outdoor concentrations, were lower for particle number than for PM2.5 and particularly sulfate and ‘soot’. Conclusions: Concentrations measured at a central site were less related to indoor concentrations of ultrafine particles, than for (specific components of) PM2.5, suggesting that measurement error in epidemiological studies may be higher for ultrafine particles.