Health Risk of Fine Particulate Matter Across China with a Pollutant-Specific Index from 2013 to 2023

Abstract

Fine particulate matter (PM2.5) remains a major public health challenge in China despite substantial improvements in air quality over the past decade. Assessments of PM2.5 distribution and its associated health risks are essential for pollution control and health protection. Thus, we developed a 1-km resolution daily PM2.5 estimation model using machine learning approaches from 2013 to 2023 across China. Meta-analyses were further conducted to estimate the short-term health effects of PM2.5 exposure on mortality and hospitalization outcomes. We integrated PM2.5 concentration and its health effect into a PM2.5-specific Air Quality Health Index (AQHIPM2.5) to quantify region-specific health risks. The model achieved high predictive accuracy (cross-validation R2 = 0.88 spatially and 0.82 temporally at daily level). PM2.5 concentrations declined substantially nationwide from 55.06 μg/m3 in 2013 to 26.50 μg/m3 in 2022, followed by a slight rebound to 27.46 μg/m3 in 2023. Each 10 μg/m3 increase in short-term PM2.5 exposure within 3 days was associated with increased risks of all-cause (0.42%), cardiovascular (0.61%), and respiratory (0.84%) mortality. AQHIPM2.5 indicated persistent moderate-to-high health risk days in northern and southwestern China, with some regions experiencing more than 80 medium- and high-risk days annually. Eliminating high-risk days identified by AQHIPM2.5 could avert an estimated 6,985 (95% CI: 6,419 to 7,588) deaths in 2023, accounting for 7.42% of PM2.5-attributable all-cause mortality burden. These findings indicate that substantial health risks from PM2.5 persist despite improved air quality and demonstrate that integrating high-resolution exposure estimation with health risk assessment can support more targeted pollution control and public health interventions.