The association between PM2.5 level and respiratory tract infections among children: A cross-sectional study

Abstract

Background: PM2.5is a key air pollutant that contributes to respiratory morbidity, especially in children. In Jakarta, Indonesia, PM2.5levels often exceed safe thresholds. This study contributes local evidence from Indonesia, where research on the health effects of PM 2.5in children remains limited. To address this gap in the existing literature, particularly within the Indonesian context, this study offers novel insights by specifically investigating the association between ambient PM 2.5exposure and respiratory tract infections (RTIs) in school-aged children and further exploring this association within male and female subgroups, an aspect that has received limited attention in this setting. Objective: This study aims to assess the association between ambient PM2.5exposure and RTIs in school-aged children, and to explore this association within male and female subgroups. Methods: This cross-sectional study was conducted among 107 children aged 6–12 years from two elementary schools: one in Jakarta (high PM2.5exposure) and one in Bandung (low PM2.5exposure). Data on PM2.5levels were obtained from local air quality monitoring. RTI symptoms were assessed through structured interviews and physical examination. Participants were selected using random sampling. Chi-square tests and effect size calculations (phi coefficient) were used to compare groups. Potential confounders such as age, gender, and household smoke exposure were minimized through inclusion/exclusion criteria and the selection of demographically and environment ally similar school communities. Multiple binary logistic regression adjusting for confounders was also performed to assess the independent association between PM2.5exposure and RTIs. Results: The Chi-square test indicated a significant association between PM2.5levels and the occurrence of RTI (χ² = 22.154, df = 1, p < 0.001, φ = 0.475). Given the potential low expected counts in some cells, the statistical significance was further evaluated using Fisher’s Exact Test, which also showed a significant association (p < 0.001). The prevalence of RTI was significantly higher in the high exposure group (71.43%) compared to the low exposure group (25.86%) (p < 0.001). Further analysis did not reveal significant differences in the proportion of each age group between the high and low PM 2.5exposure groups [χ²(1) = 0.093, p = 0.761]. Similarly, no significant differences were found in the proportion of gender between the high and low PM2.5exposure groups [χ²(1) = 1.611, p = 0.204] in the total sample. Likewise, there were no significant differences in the proportion of RTI across different age groups [χ²(6) = 5.327, p = 0.503] or between genders [χ²(1) = 0.008, p = 0.928] in the total sample. However, further analysis examining the association between PM2.5exposure and RTI within gender subgroups revealed a significant association in both male [χ²(1) = 10.873, p = 0.001] and female [χ²(1) = 11.755, p = 0.001] children. The estimated prevalence ratio (PR) was 2.76 (95% CI: 1.68–4.54), indicating that children in the high PM2.5exposure area had approximately 2.76 times higher prevalence of RTI compared to those in the low exposure area. The absolute prevalence difference (PD) was 45.57% (95% CI: 25.9%–65.2%). Binary logistic regression analysis showed that children in the high PM2.5exposure group had significantly higher odds of having RTI (OR = 7.167, 95% CI: 3.050–16.837, p < 0.001). Further analysis examining the association between maternal socioeconomic factors and both PM 2.5exposure and RTI occurrence revealed no statistically significant relationships. Chisquare tests showed no significant association between maternal education level (low vs. medium) and PM 2.5exposure group [χ²(1) = 0.045, p = 0.833], nor between maternal occupation (blue collar vs. semiprofessional) and PM2.5exposure group [χ²(1) = 0.006, p = 0.937]. Similarly, no significant associations were found between maternal education level and RTI [χ²(1) = 0.233, p = 0.629] or between maternal occupation and RTI [χ²(1) = 0.447, p = 0.504]. Crucially, after adjusting for potential confounders including gender, age, maternal education, and maternal occupation in a multivariate logistic regression model, the odds of having RTI remained significantly higher in children with high PM2.5exposure (adjusted OR = 7.883, 95% CI: 3.228–19.250, p < 0.001). Conclusions: Children exposed to higher levels of PM2.5had significantly more respiratory tract infections. These findings highlight the need for targeted public health interventions in polluted urban areas. This study is among the first to quantify this association in the Indonesian context and provides a newly developed and validated instrument (RAAEC-C instrument) for future research. These findings should be interpreted as preliminary evidence and require replication in future longitudinal studies before firmconclusions can be drawn. Further research using longitudinal designs is needed to understand the long-term impacts of PM2.5exposure on children’s respiratory health and to inform appropriate mitigation strategies.