Comparison of personal portable and fixed-site PM2.5 monitoring strategies in predicting pulmonary function decline among asthmatic children in Seoul, Korea: a pilot study

초록

Traditional fixed-site monitoring networks cannot capture individual mobility patterns and microenvironmental PM2.5 variability, leading to exposure misclassification that attenuates observed health effects estimates. This pilot study compared personal portable monitoring device (PICO) with two fixed-site national automatic (Air Korea, AK) and regional monitoring (KTR) stations in relation to pulmonary function decline in asthmatic children. Twenty asthmatic children (mean age 10.8 ± 2.4 years) living in the Seoul metropolitan area were sequentially enrolled from November 2019 to January 2020, with simultaneous PM2.5 and peak expiratory flow rate (PEFR) monitoring conducted from November 2019 to March 2020 (1436 child-day observations; 367 PEFR decline events). Seventy-two-hour cumulative PM2.5 exposure was quantified using five indicators. Primary analysis used a generalized linear mixed model (GLMM) with a random intercept per child (ICC = 0.42), adjusted for age, sex, BMI, temperature, and humidity. Restricted cubic spline (RCS) regression and time-varying Cox regression were performed as supplementary analyses. Mean PM2.5 concentrations (29.4, 31.6 and 31.2 µg/m3 for PICO, AK, and KTR, respectively), substantially exceeded the WHO air quality guideline of 15 µg/m3. PICO-measured PM2.5 showed the strongest negative correlation with PEFR (r = − 0.481, p < 0.001) compared to AK (r = − 0.229, p < 0.001) and KTR monitors (r = − 0.315, p < 0.001). In the primary GLMM analysis, each 1 µg/m3 increase in PICO-measured PM2.5 level was significantly associated with β = 0.01013 (95% CI 0.00915–0.01110, p < 0.001; C-index = 0.837), compared with AK (β = 0.00294; C-index = 0.563) and KTR (β = 0.00289; C-index = 0.585). The RCS analysis identified an exploratory threshold level of 25 µg/m3, above which PEFR decline probability increased sharply (observed rate: 5.7% below versus 41.9% above the threshold). In this pilot study, PICO monitor showed stronger associations with PEFR decline in asthmatic children and higher predictive discrimination than two fixed-site monitoring approaches, consistent with reduced exposure misclassification. These findings support the potential utility of personal portable PM2.5 monitors for pediatric respiratory health protection and warrant confirmation in larger and multi-seasonal longitudinal cohort studies in the future.

키워드