Association of ambient particulate matter with hospital admissions, length of hospital stay, and hospital costs due to cardiovascular disease: time-series analysis based on data from the Shanghai Medical Insurance System from 2016 to 2019

Background There is limited evidence supporting a relationship of ambient particulate matter (PM), especially PM 1 , with hospital admissions, hospital costs, and length of hospital stay (LOS) due to cardiovascular disease (CVD). We used a generalized additive model (GAM) to estimate the associations of these indicators due to CVD for each 10 μg/m 3 increase in the level of PM 1 , PM 2.5 , and PM 10 , and the attributable risk caused by PM on CVD was determined using the WHO air quality guidelines from 2005 and 2021. Results For each 10 μg/m 3 increase in the level of each PM and for a 0-day lag time, there were significant increases in daily hospital admissions for CVD (PM 1 : 1.006% [95% CI 0.859, 1.153]; PM 2.5 : 0.454% [95% CI 0.377, 0.530]; PM 10 : 0.263% [95% CI 0.206, 0.320]) and greater daily hospital costs for CVD (PM 1 : 523.135 thousand CNY [95% CI 253.111, 793.158]; PM 2.5 : 247.051 thousand CNY [95% CI 106.766, 387.336]; PM 10 : 141.284 thousand CNY [95% CI 36.195, 246.373]). There were no significant associations between PM and daily LOS. Stratified analyses demonstrated stronger effects in young people and males for daily hospital admissions, and stronger effects in the elderly and males for daily hospital costs. Daily hospital admissions increased linearly with PM concentration up to about 30 µg/m 3 (PM 1 ), 60 µg/m 3 (PM 2.5 ), and 90 µg/m 3 (PM 10 ), with slower increases at higher concentrations. Daily hospital costs had an approximately linear increase with PM concentration at all tested concentrations. In general, hospital admissions, hospital costs, and LOS due to CVD were greater for PM 2.5 than PM 10 , and the more stringent 2021 WHO guidelines indicated greater admissions, costs, and LOS due to CVD. Conclusions Short-term elevation of PM of different sizes was associated with an increased risk of hospital admissions and hospital costs due to CVD. The relationship with hospital admissions was strongest for men and young individuals, and the relationship with hospital costs was strongest for men and the elderly. Smaller PM is associated with greater risk.