大气粗颗粒物与中国多城市抑郁症每日住院人数、费用和天数的关联

Association of ambient coarse particulate matter with hospital admissions, related expenditures and stays for depression in China: a multicity study

目的基于中国城镇职工基本医疗保险和城镇居民基本医疗保险数据库,分析大气粗颗粒物(coarse particulate matter, PM2.5-10)短期暴露与中国部分城市人群抑郁症住院人数、费用与天数的关联。方法基于2013—2017年中国56个地级及以上城市大气PM2.5-10数据及抑郁症住院信息,进行两阶段时间序列分析。使用广义相加模型控制星期几效应、节假日效应、长期趋势、气象因素等混杂因素后,分别分析各城市大气PM2.5-10短期暴露对抑郁症住院人数影响,并采用随机效应模型综合不同城市结果,进而计算住院人数、住院费用和住院天数的归因数及归因百分比。结果研究共纳入56个城市82 708例抑郁症住院患者信息。大气PM2.5-10短期暴露与抑郁症住院人数、费用与天数之间存在正向关联。在lag0、lag02和lag03时间窗上,大气PM2.5-10短期暴露浓度每升高10μg/m^(3),抑郁症住院人数分别增加0.58%(95%CI:0.04%~1.12%)、1.20%(95%CI:0.10%~2.29%)和1.20%(95%CI:0.01%~2.39%)。其中在lag03时间窗,住院人数、费用和天数归因于PM2.5-10短期暴露的百分比达到最大,分别为4.14%(95%CI:0.04%~8.03%)、3.67%(95%CI:0.03%~7.14%)和3.89%(95%CI:0.04%~7.54%);平均每个患者因PM2.5-10短期暴露导致的归因住院费用和天数分别为483.77(95%CI:4.49~940.63)元和1.03(95%CI:0.01~1.99) d。结论短期暴露于大气粗颗粒物与中国城市人群抑郁症住院人数、费用和天数的增加有关。

Objective Based on China’s urban employee basic medical insurance and urban residents’basic medical insurance database,this study aims to investigate the associations between short-term exposure to ambient coarse particulate matter(PM2.5-10)and risk of hospital admissions,related expenditures and stays(counted as days in hospital)for depression.Methods A two-stage time-series study was conducted using the daily data in ambient PM2.5-10 and hospital admissions for depression in 56 prefecture-level or above cities in China from 2013 to 2017.The generalized additive model(GAM)with adjustment for day-of-week effect,holiday effect,secular trend,and meteorological variables was used to investigate the city-specific associations between short-term exposure to PM2.5-10 and risk of hospital admissions for depression,and a random-effects model was used to pool the city-specific associations to generate an overall association over different cities.Attributable numbers and fractions of hospital admissions,related expenditures and stays associated with short-term exposure to ambient PM2.5-10 were also calculated based on the results of the two-stage time-series analyses.Results A total of 82708 hospital admissions for depression were identified in the urban medical insurance schemes in 56 Chinese cities during the study period.Short-term exposure to ambient PM2.5-10 was associated with increased hospital admissions,related expenditures and stays.For each 10μg/m^(3) increase in lag0,lag02,and lag03,there were significant increases of 0.58%(95%CI:0.04%-1.12%),1.20%(95%CI:0.10%-2.29%)and 1.20%(95%CI:0.01%-2.39%)in hospital admissions for depression,respectively.The attributable fractions of hospital admissions,related expenditures and stays associated with ambient PM2.5-10 reached maximum values at lag03,and were 4.14%(95%CI:0.04%-8.03%),3.67%(95%CI:0.03%-7.14%)and 3.89%(95%CI:0.04%-7.54%),respectively;and the average hospitalization expenditure and stay per patient due to short-term exposure to PM2.5-10 were 483.77(95%CI:4.49-940.63)yuan and 1.03(95%CI:0.01-1.99)d,respectively.Conclusion Short-term exposure to ambient PM2.5-10 is associated with increased risk of hospital admissions,expenditures and stays for depression in the general urban population in China.