Impact of short-term temperature variability on emergency hospital admissions for schizophrenia stratified by season of birth.

Abstract

Diurnal temperature range (DTR) and temperature change between neighboring days (TCN) are important meteorological indicators closely associated with global climate change. However, up to date, there have been no studies addressing the impacts of both DTR and TCN on emergency hospital admissions for schizophrenia. We conducted a time-series analysis to assess the relationship between temperature variability and daily schizophrenia onset in Hefei, an inland city in southeast China. Daily meteorological data and emergency hospital admissions for schizophrenia from 2005 to 2014 in Hefei were collected. After stratifying by season of birth, Poisson generalized linear regression combined with distributed lag nonlinear model (DLNM) was used to examine the relationship between temperature variability and schizophrenia, adjusting for long-term trend and seasonality, mean temperature, and relative humidity. Our analysis revealed that extreme temperature variability may increase the risk for schizophrenia onset among patients born in spring, while no such association was found in patients born in summer and autumn. In patients born in spring, the relative risks of extremely high DTR comparing the 95th and 99th percentiles with the reference (50th, 10°C) at 3-day lag were 1.078 (95% confidence interval (CI) 1.025-1.135) and 1.159 (95% CI 1.050-1.279), respectively. For TCN effects, only comparing 99th percentile with reference (50th, 0.7°C) was significantly associated with emergency hospital admissions for schizophrenia (relative risk (RR) 1.111, 95% CI 1.002-1.231). This study suggested that exposure to extreme temperature variability in short-term may trigger later days of schizophrenia onset for patients born in spring, which may have important implications for developing intervention strategies to prevent large temperature variability exposure.