Methods: From 2005, 2,168 elementary school children were enrolled to a prospective 4-year follow-up survey. Individual exposure to PM10 was estimated by using a geometric information system from birth to 2004. Measuring urine cotinine levels and methacholine challenge test were conducted every 2 years from 2005.
Results: PM10 from birth to 2004 and urine cotinine at start year increase the risk of AHR at start year (aOR 1.48, 95% CI 1.18-1.87 and aOR 1.48, 95% CI 0.99-2.21, respectively). PM10 for 2years before start year and urine cotinine at start year increased the risk of AHR at start year additively (aOR 1.95, 95% CI 1.08-3.50, p for interaction 0.117). Higher urine cotinine levels during 4-year follow-up increased the risk of new AHR (positive conversion from PC20 > 16mg/dl) in children with higher PM10 from birth to 2004 (aOR 2.66, 95% CI 1.10-6.45, p for interaction 0.005). However, cotinine at start year/ PM10 from birth to 2004 and cotinine during 4- year follow-up period/ PM10 for 2years before start year did not increase the risk of new AHR.
Conclusions: Short-term exposures to PM10 and/or tobacco smoke increase the risk of AHR additively. Long-term combined exposure is more important in new development of AHR.