During 2018–2020, the 24-hr average of PM2.5 levels in Bangkok were above both the recommendation from World Health Organization (WHO) at 15 µg/m3 and from Thai Pollution Control Department at 37.5 µg/m3. Annual mean of PM2.5 levels in Bangkok were also above both the recommendation from WHO at 5 µg/m3 and from Thai Pollution Control Department at 15 µg/m3.2, 12, 13
We found a significant association between weekly average of PM2.5 level and number of outpatients less than 2 years with URIs and acute bronchitis after adjusted for confounders including year, seasons, influenza and RSV infection. Especially if PM2.5 level higher than 55 µg/m3 in winter, we found 40% increase in outpatients with URIs and 80% increase in outpatients with acute bronchitis in children less than 2 years. Apart from PM2.5 level, influenza and RSV infection also have a significant association with number of many pediatric respiratory cases. RSV infection had strong association in children less than 5 years.
PM2.5 affects many respiratory diseases because it impairs the bronchial mucociliary system and decreases viral clearance. Moreover, PM2.5 and PM2.5-induced inflammatory cytokines trigger lung epithelial and fibroblast cells death and inhibit the intercellular communication between these cells. These increase the permeability of epithelial barrier and impair pulmonary innate immunity.14
PM2.5 levels were also significant associated with number of outpatient children less than 5 years with asthmatic exacerbation. However, subgroup analysis during winter didn’t show significant association because PM2.5 level was high along the whole winter period.
Most of previous studies have shown the association between PM2.5 level and number of respiratory cases. Xiao Q, et al. found that short term exposure to PM2.5 and other pollutants (ex. O3, CO, NO, SO2) was significantly associated with increased risk of emergency department visits due to URIs, acute bronchitis, pneumonia, and asthma exacerbation in Georgia.5 Kim KN et al. found the associations of short-term PM2.5 exposure with URIs, acute bronchitis, and acute bronchiolitis among children at 0–4 years of age.15 Davila Cordova JE et al. studied in Peru children less than 5 years of age and found a significant association between an increase of PM2.5 level and number of outpatient visits from acute lower respiratory infections including pneumonia, acute bronchiolitis and asthma.16 This result was in contrast to our study which revealed a significant association between an increase of PM2.5 level and number of outpatient visits from only URIs and acute bronchitis in children less than 2 years but not the others.
Previous studies found that for each interquartile range (IQR) increase in PM2.5, emergency room visits for pediatric respiratory diseases increased 3% (95% CI: 1.02–1.04%)17 and hospital visits for pediatric respiratory diseases in children younger than 5 years increased 6% (95% CI: 1.05–1.08%).16 Our estimate in children younger than 2 years was much higher. A recent meta-analysis of 16 time-series studies of hospital admissions in children younger than 5 years, found a respiratory disease increase of 2.7% (95% CI = 0.9–7.7%) per 10 µg/m3 increase in PM2.5.18
All countries in the previous studies located in different region, had different seasons, duration of each season and seasonal outbreak of respiratory viruses. Thailand locates in tropical zone and has only 3 seasons. We usually have seasonal outbreak of influenza and RSV in rainy season and winter. These factors affect the number of pediatric respiratory cases in addition to PM2.5 level. The different age group of children in each study also affect the degree of association. Most of previous studies used only data from young children but our study included data from all children and adolescent less than 18 years of age. Our study found that high PM2.5 level was associated with number of cases only in young children less than 2 years of age. Young children have faster respiratory rate than adolescents and adults, which results in inhalation of higher quantities of air and pollutant load.15 Moreover, they spend most of their time outdoor and can’t wear mask effectively.
In a time-series analysis using a negative binomial model, we included year as an adjusted variable because the number of pediatric respiratory cases in later year have decreased. This could be the effect from our national campaign which raise public awareness of air pollution and promote people to wear mask.
Our findings suggest that it is necessary for the government and independent health organizations to make an action plan for reduction of PM2.5 level. We should raise public awareness of the effects from these environmental pollutants, reduce sources that can produce PM2.5. We should closely monitor younger children who are in vulnerable age group. Because they have more effects from high PM2.5 level. Moreover, they can’t use protective measure such as mask wearing and activity restriction effectively. Because of younger age, they may also have long term effects and study of those in this age group is still needed.
For limitation of our study, we didn’t use other air quality indices besides PM2.5 for analysis because there was no available data of those from the government agencies.