Seasonally modulated nonlinear effects of PM_2.5 on pediatric respiratory health: evidence from a time-series analysis in urban China

Introduction

Air pollutants are major health risks, driving global disease burdens, especially in low- and middle-income countries (GBD 2015 Risk Factors Collaborators, 2016). Particulate matter, especially that with an aerodynamic diameter equal to or less than 2.5 µm (PM_2.5), has been recognized as a major public health risk factor. Studies have shown that PM_2.5 has become the leading factor contributing to the global disease burden (GBD 2021 Risk Factors Collaborators, 2024). In China, for each 10 µg/m³ increase in PM_2.5, the death risk of respiratory diseases (RDs) increases by 0.10% (Liu et al., 2025). Although the impact of PM_2.5 exposure on health has attracted attention around the world, nearly 99% of the world’s population is still exposed to PM_2.5 levels higher than those of the 2021 update to the World Health Organization (WHO) Global Air Quality Guidelines (AQGs) (Faridi et al., 2022). It is estimated that the impacts of PM_2.5 on health will continue through 2030 (Chi et al., 2024).

Respiratory tract infection has become the disease with the highest morbidity in children due to the immature anatomical physiology and immune function of children’s respiratory systems (Ingham et al., 2019). Pneumonia is the sixth leading cause of death in children under 5 years of age in China (Faridi et al., 2022; He et al., 2017). Owing to its rapid urbanization and industrial and economic development, China has paid insufficient attention to environmental pollution; this has resulted in serious urban environmental pollution, with adverse environmental, economic, and social effects of high PM_2.5 concentrations that have attracted the attention of local governments (Wang et al., 2018a). Liu et al. (2022a) conducted a study on PM_2.5 and RDs among children in Zhengzhou, China and found that PM_2.5 exposure was significantly related to RDs in children. Another researcher (Liu et al., 2019) studied the association between daily air pollution and hospitalization for RDs among children aged 0–17 years in Jinan, China. They found that PM_2.5 was significantly correlated with RD-related hospitalization among children. Dong et al. (2021) conducted a study on PM_2.5 and RDs among children aged 0–14 years in Lanzhou, China and found that PM_2.5 was significantly correlated with RDs.

While PM_2.5 levels in Guangzhou meet national standards, they remain higher than the PM_2.5 guideline limits recommended by the WHO in 2021. In Guangzhou, it has been reported that outdoor PM_2.5 concentrations were positively correlated with neonatal bacterial infection (Liu et al., 2022b). However, there are relatively few reports regarding the relationship between PM_2.5 and respiratory diseases. Therefore, this study aims to analyze the relationship between PM_2.5 and RDs in children in Guangzhou.

Materials & Methods

Results

From 2018 to 2021, a total of 22,585 pediatric patients were hospitalized, among whom 18,291 were diagnosed with RDs, accounting for 80.99% of all hospitalized children. The daily respiratory disease hospitalizations ranged from 0 to 35, with the highest numbers observed in children aged ≤2 years and males. Air pollution concentrations varied, with PM_2.5 ranging from 3.71 to 145.79 µg/m³, O₃ from 3.88 to 175.67 µg/m³, and CO from 0.21 to 26.00 mg/m³. Meteorological factors showed a temperature range of 5.32–33.70 °C, relative humidity of 17.87%–97.25%, wind speed of 0.66–11.00 s/m, and dew point temperature from −9.77 °C to 26.65 °C (Table 1).

Table 2 presents the Spearman correlation coefficients between air pollutants and meteorological factors in Guangzhou. PM_2.5 exhibited significant negative correlations with temperature (ρ_s = −0.355), wind speed (ρ_s = −0.400), and dew point temperature (ρ_s = −0.443) (P < 0.05), O₃ showed a significant positive correlation with temperature (ρ_s = 0.239; P < 0.05) and a weak positive correlation with CO (ρ_s = −0.033).

Adjusted for O₃, CO, and meteorological parameters, the curve between PM_2.5 and RDs formed a zigzag shape. With increased concentration of PM_2.5, the risk of RDs also increased; the risk of RDs decreased gradually with the extension of lag time, and the RD risk was the lowest at lag day 5 (Fig. 1).

Table 3 shows the cumulative effect of PM_2.5 exposure on RDs in percentiles of lag 0–14 days. After adjustment for O₃, CO, and meteorological parameters, the cumulative effects of PM_2.5 exposure at the 25th, 50th, and 75th percentiles (over lag days 0–14) were associated with increased risks of RDs, with relative risk (RR) of 1.004 (95% confidence interval (CI) [1.000–1.007]), 1.034 (95% CI [1.004–1.065]), and 1.077 (95% CI [1.010–1.149]), respectively.

Additionally, each 10-µg/m³ increase in PM_2.5 concentration was associated with a RR of 1.018 (95% CI [1.002–1.034]) for RDs at lag day 4 and 1.016 (95% CI [1.000–1.032]) at lag day 7, indicating a significant increase in the risk of RDs (Fig. 2).

As shown in Fig. 3, a per 10-µg/m³ increase in PM_2.5 concentration at lag 0–14 days was associated with a cumulative RR for RDs of 1.074 (95% CI [1.009–1.143]), indicating a statistically significant increased risk. In males, the cumulative RR was 1.046 (95% CI [1.005–1.089]), also reflecting a significant increase in risk. By contrast, the association was not statistically significant in females.

After adjusting for confounding factors, each 10-µg/m³increase in the concentration of PM_2.5 with a lag of 0–14 days was associated with an 85.7% higher risk of RDs in summer (RR = 1.857; 95% CI [1.077–3.203]) and a 5.2% higher risk in winter (RR = 1.052; 95% CI [1.013–1.092]). In boys, each 10-µg/m³ increase in PM_2.5 exposure was associated with a 113% higher risk of RDs during summer (RR = 2.130; 95% CI [1.064–4.264]) and a 5.7% higher risk during winter (RR = 1.057; 95% CI [1.008–1.107]) (Table 4).

Discussion

In this study, we conducted a time-series analysis of the relationship between PM_2.5 and RDs among children in a children’s hospital in Guangzhou. Our findings indicate that PM_2.5 exposure substantially increases the risk of RDs, and the lagged effect reaches its peak on days 4 and 7. The increase in risk is more pronounced among male children. Additionally, this study found that the risk of RDs in children was higher in summer and winter. These findings highlight the necessity for targeted air quality management and public health interventions to reduce the risk of RDs.

Previous studies have demonstrated that exposure to PM_2.5 increases the susceptibility of the respiratory system to various pathogens (Yang, Li & Tang, 2020). Research indicates that, among environmental health risks, air pollution—particularly PM_2.5—presents the most significant threat concerning respiratory issues in infants and children (Jakubiak-Lasocka, Lasocki & Badyda, 2015). He et al. (2022) found that PM_2.5 not only affects emergency room visits for all respiratory diseases on the day of exposure but also exhibits a cumulative effect. In a study involving 332,337 children (aged 0–13 years) in Lanzhou, the results demonstrated a significant correlation between increased PM_2.5 levels and a rise in outpatient visits for overall respiratory diseases among children (Ma et al., 2020). A key strength of this study lies in its specific focus on children within the unique context of Guangzhou, a densely populated subtropical metropolis with distinct air pollution patterns and a large migrant population. Our findings align well with previous studies, indicating that the risk of respiratory diseases (RDs) increases with rising concentrations of PM_2.5. This consistency can be attributed to several factors, primarily the anatomical and physiological immaturity of children’s respiratory systems, rendering them more susceptible to external environmental factors, especially particulate matter. Additionally, children’s immune systems are still in the developmental stage, conferring a relatively weaker capability to fend off external pathogens and pollutants. Consequently, in environments with high PM_2.5 levels, children are particularly vulnerable, leading to an augmented risk of developing respiratory diseases. This comprehensive understanding underscores the necessity for targeted public health strategies aimed at mitigating the adverse effects of air pollution on pediatric respiratory health.

A study conducted in Yichang City, China, demonstrated that for every interquartile range increase in PM_2.5 levels with a lag of 0 days, there was increase in pediatric respiratory outpatient visits (Liu et al., 2017). A study involving 39,766 children in Taiwan found that the RR of acute upper respiratory infections due to PM_2.5 exposure peaked at a lag of 4 to 5 days (Lin et al., 2013). Research conducted in Shenzhen, China, indicated that the risk of hospitalization for respiratory conditions significantly increased due to PM_2.5 exposure after a lag of five days (Liang et al., 2021). However, a study in Poland found no significant increase in the risk of upper respiratory infections from short-term PM_2.5 exposure within a lag of 0 to 6 days (Ratajczak et al., 2021). This study also showed that there is a lag effect between PM_2.5 and RDs in children. PM_2.5 exposure had the lowest effect on RDs in children with a time lag of 5 days, which is inconsistent with earlier studies (Liang et al., 2021; Slama et al., 2019). Inconsistencies in findings may be attributed to several factors. Geographical and climatic differences, such as variations in temperature, humidity, and wind speed, influence the dispersion and retention of PM_2.5, thereby affecting its health impacts on respiratory diseases (Garsa et al., 2023; Li et al., 2019; Li, Liu & Zhao, 2022). Furthermore, differences in health status, lifestyle, and access to healthcare across populations may modify the association between PM_2.5 exposure and respiratory diseases. For example, Guangzhou differs demographically from cities such as Yichang, Taiwan, and Poland in its substantial population of migrant children, who often belong to lower socioeconomic groups and have severely limited access to healthcare services. This subgroup may be more susceptible to the adverse effects of air pollution due to underlying health inequities. Moreover, delayed medical care may alter the timing and severity of disease onset, thereby amplifying the observed health impacts of PM_2.5 exposure in this population. Collectively, these elements contribute to the variability observed in research outcomes.

A study in Changchun, China showed that when the concentration of PM_2.5 increased by 10 µg/m³, the risk of hospitalization for RDs increased by 0.31% (Jia et al., 2022). A study of the effects of air pollutants and meteorological conditions on hospitalization for RDs in Shenzhen, China found that when the concentration of PM_2.5 increased by 10%, the risk of hospitalization for RDs increased by 0.23% (Liang et al., 2021). The results of a time-series study conducted in Yichang City, China showed that with each interquartile range increase in PM_2.5 concentrations, there was a 1.91% increase in pediatric respiratory outpatient visits (Liu et al., 2017). The results of a meta-analysis indicated that for each 10 µg/m³ increase in PM_2.5, the risk of chronic obstructive pulmonary disease (COPD)-related emergency department (ED) visits and outpatient clinic visits increased by 2.5% (95% CI [1.6%–3.4%]) (DeVries, Kriebel & Sama, 2017). The results of the present study were significantly higher than those of previous studies, with the risk of RDs increasing by 7.4% per 10-µg/m³ increase in the concentration of PM_2.5, which is slightly different from other studies. The differences in results may be attributed to several factors. Firstly, the study populations varied: studies conducted in Changchun and Shenzhen focused on adults and children, whereas our study targeted children exclusively. Secondly, there were differences in the research methodologies employed. Our study incorporated a broader range of meteorological parameters, whereas previous studies often considered only one or two factors. The combined effects of these meteorological factors can exacerbate PM_2.5 pollution, thereby increasing the risk of respiratory and other related diseases. These differences in study populations and methodologies may have led to variations in results.

An increasing number of studies have shown a link between air pollution and RDs in both sexes, but the results are not completely consistent. A study conducted by Hu et al. (2024) in China showed that males are more vulnerable to air pollutants. Wang et al. (2018b). studied the effects of air pollutants on RDs in Shandong Province, China. The results showed that air pollutants had significant effects on both men and women. Our study results showed that male children exposed to PM_2.5 were more prone to RDs than female children. This may be owing to differences between male and female airways (Becklake & Kauffmann, 1999; Dong et al., 2011; Polgar & Weng, 1979), male children have more immature lungs and relatively narrow airways in childhood, making them more prone to RDs than female children.

Research has shown that PM_2.5 is closely linked to the exacerbation of respiratory diseases in winter (Romaszko-Wojtowicz et al., 2025). The findings from a study encompassing 372 cities across 19 countries and regions indicate that the impact of PM_2.5 on mortality due to respiratory diseases is significantly more pronounced during the cold season (Liu et al., 2023). However, this study reveals an increased risk of respiratory diseases among children during both the summer and winter seasons. The observed inconsistencies may be attributed to the study area being located in a subtropical zone, where hot and humid summers can impede the dispersion of PM_2.5. Elevated temperatures also enhance photochemical reactions, contributing to the formation of ozone and other secondary pollutants, thereby further deteriorating air quality. Furthermore, individuals across different regions adopt varying protective measures during different seasons, which can influence the study outcomes.

This study has several strengths. First, by focusing on the pediatric population in the subtropical megacity of Guangzhou, it provides critical evidence for developing strategies to mitigate the health impacts of PM_2.5 exposure on children with respiratory diseases in similar metropolitan settings. Second, time-series analysis effectively captures significant lagged effects of PM_2.5 exposure and reveals distinct seasonal patterns in associated health risks. However, this study has several limitations. First, the sample was derived exclusively from clinical records of a single children’s hospital in Guangzhou, restricting the population to pediatric patients without comparative data from other age groups; this limited representativeness may constrain the generalizability of findings. Second, although adjustments were made for confounders including pollutants and meteorological factors, residual confounding may persist due to the model’s inability to fully control for potential confounders such as indoor air pollution exposure, personal protective behaviors, and socioeconomic status. Furthermore, modeling season as a categorical variable in this study potentially oversimplifies its continuous and dynamic nature, failing to capture intra-seasonal fluctuations and complex interactions with other environmental factors, thereby compromising the accurate assessment of real-world health risks. Finally, individual PM_2.5 exposure was estimated using fixed-site monitoring concentrations, which neglects variations in personal activity patterns, microenvironmental differences, and spatial heterogeneity. This approach may introduce non-differential misclassification in exposure assessment, potentially biasing effect estimates. Despite the aforementioned limitations, these findings highlight the urgency of implementing effective measures to reduce air pollution levels. It is recommended to strengthen industrial emission controls, accelerate the transition to clean energy, and optimize urban public transportation systems to mitigate pollutant emissions at the source. Concurrently, widespread implementation of vaccination strategies against respiratory infections should be promoted to enhance immune protection in children (Cattaneo, 1994; Ding et al., 2025). Coordinated efforts in environmental and public health interventions are essential to mitigate the long-term adverse effects of air pollution on children’s respiratory health.

Conclusions

In summary, this study demonstrates a significant association between PM_2.5 exposure and an increased risk of respiratory conditions in children, particularly during the summer and winter seasons. These findings emphasize the urgent need for targeted public health interventions to mitigate the adverse effects of air pollution on children’s respiratory health.

Supplemental Information