Objective To investigate the association between long-term exposure to ambient fine particulate matter (PM_2.5) and the risk of liver diseases among middle-aged and elderly people in multiple provinces and cities in China.

Methods The data of this study were collected from China Health and Retirement Longitudinal Study (CHARLS) conducted by National School of Development at Peking University. With the design of prospective cohort study, a baseline survey was performed in 2011 and follow-up surveys were performed in 2013, 2015, and 2018, with the onset of chronic liver diseases (except fatty liver disease, tumor, and cancer) as the primary health outcome. The level of long-term PM_2.5 exposure was assessed according to the mean annual global atmospheric PM_2.5 dataset established by Atmospheric Composition Analysis Group of Washington University in St. Louis. The Cox proportional-hazards regression model was used to investigate the association between long-term PM_2.5 exposure and the incidence rate of liver disease after adjustment for confounding factors, and the modification effect of smoking, drinking and other factors was also analyzed.

Results A total of 12 301 patients were included, with a mean follow-up time of 6.34 years and a total of 78 063 person-years of follow-up, and 409 patients developed liver diseases, resultsing in an incidence rate of 3.3%. The Cox regression analysis showed that for every 10 μg/m³ increase in PM_2.5 exposure, the hazard ratio (HR) for the risk of liver disease was 1.33 (95% confidence interval CI: 1.24-1.43), and the corresponding population attributable fraction was 12.72% (95% CI: 9.58-15.76). If the patients were divided into four groups according to the quartile of PM_2.5 exposure, with the lowest group (Q1) as reference, the highest group (Q4) had an HR of 2.11 (95% CI: 1.59-2.81) for the risk of liver disease. Stratified analysis showed that for every 10 μg/m³ increase in PM_2.5 exposure in the population, the HR for liver disease was 1.40 (95% CI: 1.22-1.60) in smokers and 1.29 (95% CI: 1.18-1.41) in non-smokers, and the HR for liver disease was 1.38 (95% CI: 1.20-1.57) in drinkers and 1.32 (95% CI: 1.21-1.44) in non-drinkers.

Conclusion Long-term exposure to high-concentration PM_2.5 can increase the risk of liver disease among middle-aged and elderly people, with a higher risk in smokers and drinkers.