Objective: To explore the effect of ambient fine particulate matters (PM(2.5)) and temperature interaction on schizophrenia admission. Methods: All admission data were retrieved from the Psychiatric Hospital and Municipal Hospital of TongLing from January 1, 2014 to December 31, 2017. Daily air pollution and meteorological data were collected from the Tongling Environmental Protection Agency and Meteorological Bureau, respectively. A distributed lag non-linear model combined with the generalized additive model were applied to explore the effects of PM(2.5), multi-pollutants, and the interaction between temperature and PM(2.5) on schizophrenia admission. The stratification of temperature was divided by three criteria. The low temperature layer was defined as <P(5) or <P(10) or <P(20); P(5)-P(95) or P(10)-P(90) or P(20)-P(80) was defined as the middle temperature layer; >P(95) or >P(90) or >P(80) was defined as the high temperature layer. Results: From 2014 to 2017, 6 642 patients were admitted for schizophrenia in Tongling, and the median of PM(2.5) and temperature were 47.0 μg/m(3) and 17.5 ℃, respectively. The median concentration of PM(2.5) (P(50)) was taken as a reference. When the exposure concentration of PM(2.5) was P(90), the lagged effect appeared in the first day with RR=1.03 (95%CI: 1.00-1.07) and reached the maximum in the fifth day with RR=1.16 (95%CI: 1.13-1.19). In the multi-pollutant models, it was found that the simultaneous inclusion of PM(2.5) and NO(2) had higher risk of schizophrenia admission, with the RR=1.18 (95%CI: 1.15-1.22), P<0.001. The risk of schizophrenia admission caused by PM(2.5) exposure at high temperature was greatest under the three temperature stratification standards, which were 12.1% (8.5%-15.7%), 9.7% (6.9%-12.6%) and 17.1% (11.6%-22.8%), all P values <0.001. Conclusion: With the increase of PM(2.5) concentration, the risk of schizophrenia admission is increased, and the risk effect of PM(2.5) is stronger at high temperature.
目的： 探讨大气PM(2.5)和温度交互作用对精神分裂症患者入院的影响。 方法： 收集2014年1月1日至2017年12月31日安徽省铜陵市第三人民医院和市立医院精神分裂症患者的入院资料、铜陵市环境保护局和气象局的环境污染物资料和气象资料。采用分布滞后非线性模型联合广义相加模型，探讨PM(2.5)、多种污染物以及PM(2.5)和温度的交互作用对精神分裂症患者入院的影响，温度的分层按照3种标准进行划分，分别为<P(5)、<P(10)、<P(20)分别为低温，P(5)~P(95)、P(10)~P(90)、P(20)~P(80)分别为中温，>P(95)、>P(90)、>P(8)0分别为高温。 结果： 2014—2017年，铜陵市精神分裂症患者入院共计6 642人次，PM(2.5)和温度的中位数分别为47.0 μg/m(3)和17.5 ℃。以PM(2.5)暴露浓度P(50)为参照，当PM(2.5)暴露浓度为P(90)时，在滞后第1天精神分裂症患者入院RR值为1.03（95%CI：1.00~1.07），第5天达到最大，RR值为1.16（95%CI：1.13~1.19）；多污染物模型中，同时纳入PM(2.5)和NO(2)时，PM(2.5)致精神分裂症患者入院风险较大，RR值为1.18（95%CI：1.15~1.22）（P<0.001）；3种温度分层标准下均为高温时PM(2.5)暴露致精神分裂症患者入院风险最大，超额风险度分别为12.1%（8.5%~15.7%）、9.7%（6.9%~12.6%）和17.1%（11.6%~22.8%），P值均<0.001。 结论： 随着PM(2.5)浓度的升高，精神分裂症的入院风险增加；并且在高温时PM(2.5)的危险效应更大。.
Keywords: Ecological study; Particulate Matter; Schizophrenia; Temperature; Time series analysis.