Interactive effects of specific fine particulate matter compositions and airborne pollen on frequency of clinic visits for pollinosis in Fukuoka, Japan.

BACKGROUND: Previous studies have revealed the interactive effects of airborne pollen and particulate matter on the daily consultations for pollinosis, but it is uncertain which compositions are responsible. This study aimed to investigate the interactive effects of specific PM2.5 compositions and airborne pollen on the daily number of clinic visits for pollinosis in Fukuoka.

METHODS: We obtained daily data on pollen concentrations, PM2.5 compositions, PM2.5 mass, gaseous pollutants (SO2, NO2, CO, and O3), and weather variables monitored in Fukuoka between February and April, 2002-2012. In total, 73,995 clinic visits for pollinosis were made at 10 clinics in Fukuoka Prefecture during the study period. A time-stratified case-crossover design was applied to examine the interactive effects. The concentrations of PM2.5 and its compositions were stratified into low (<15th percentile), moderate (15th-85th percentile), and high (>85th percentile) levels, and the association between airborne pollen and daily clinic visits for pollinosis was analyzed within each level.

RESULTS: We found a significant interaction between specific PM2.5 compositions and airborne pollen. Specifically, the odds ratio of daily clinic visits for pollinosis per interquartile increase in pollen concentration (39.8 grains/cm(2)) at the average cumulative lag of 0 and 2 days during high levels of non-sea-salt Ca(2+) was 1.446 (95% CI: 1.323-1.581), compared to 1.075 (95% CI: 1.067-1.083) when only moderate levels were observed. This result remained significant when other air pollutants were incorporated into the model and was fairly persistent even when different percentile cut-off points were used. A similar interaction was found when we stratified the data according to non-sea-salt SO4(2-) levels. This finding differed from estimates made according to PM2.5 and NO3(-) levels, which predicted that the effects of pollen were strongest in the lower levels.

CONCLUSIONS: Associations between airborne pollen and daily clinic visits for pollinosis could be enhanced by high levels of specific PM2.5 compositions, especially non-sea-salt Ca(2+).