Short-term exposure to PM2.5 on influenza incidence: mechanisms underlying effect modification in a case-crossover study stratified by spatiotemporal factors across several cities in Xinjiang, China

PM 2.5 is an important and modifiable environmental risk factor for adverse health outcomes, including respiratory infections, cardiovascular diseases, and cancers. However, current evidence on the impact of air pollution on influenza risk remains inconsistent. This study aimed at examining the associations between short-term ambient PM 2.5 exposure and influenza incidence in Xinjiang China, and to explore effect modifications. In this regional case-crossover study, we linked prospectively collected influenza surveillance data for 13 cities in Xinjiang, China from Jan 1, 2023, to Dec 31, 2025. We associated these data with daily PM 2.5 , carbon monoxide (CO), nitrogen dioxide (NO 2 ), ozone (O 3 ), and sulphur dioxide (SO 2 ) exposure on the day of diagnosis (lag 0) or 1–3 days before diagnosis (lags 1–3), as well as the moving average across days 0–1 and days 0–3. Daily exposure was calculated by averaging measurements from multiple environmental monitoring stations within each city. The effect of PM2.5 on risk of influenza morbidity was estimated using a time-stratified case-crossover design coupled with conditional logistic regression analysis, while adjusting for daily temperature, relative humidity, and wind speed. Single-pollutant models were also conducted for individual gaseous pollutants (CO, NO 2 , O 3 , and SO 2 ), as well as two-pollutant models for PM 2.5 with these gaseous pollutants. Stratified analyses were conducted by sex, age group, influenza subtype, and epidemic period. This three-year study included 32,407 influenza cases, with the median daily PM 2.5 concentration of 21 µg/m³ (IQR 12–46). Each 100 µg/m³ increment in PM 2.5 was linked to elevated influenza risk on the same day (OR 1.051, 95% CI 1.025–1.077) and at all lag periods up to 3 days, with the strongest effect observed at lags 0 to 3 days (OR 1.067, 95% CI 1.032–1.103). PM₂.₅ was significantly associated with influenza risk in both winter and summer. Children aged 0–14 years and periods dominated by influenza A(H3) were more susceptible to PM 2.5 exposure. No significant association was found in adults aged over 65 years, and no sex differences were detected. CO and NO 2 were positively associated with influenza, whereas O 3 was inversely associated with influenza. In two-pollutant models, the positive association between PM 2.5 and influenza remained robust and independent of other pollutants. In the unique geographical and environmental context of Xinjiang, short-term PM₂.₅ exposure is associated with an increased risk of influenza incidence in both winter and summer. Air quality regulations and public health interventions should recognize and incorporate the ongoing health benefits associated with further reductions in PM 2.5 pollution, with particular emphasis on targeted protective measures for children and other vulnerable populations during the influenza A(H3) epidemic periods.