Quantifying the impacts of temperature variability on hospitalizations for schizophrenia: A time series analysis in Hefei, China

Abstract

ObjectivesNon-optimum temperature affected mental health seriously, but little was known about the relationship between intra- or inter-day temperature variability (TV) and schizophrenia. We aimed to quantify the impacts of TV on hospitalizations for schizophrenia, estimate attributable risk, and identify the vulnerable groups. MethodsData of hospitalizations and meteorological factors were collected in Hefei, China, 2005–2014. Defining TV0-t as the standard deviation of daily maximum and minimum temperature within 0-t days. First, distributed lag non-linear model was used to estimate the relative risk (RR) of TV0-t on hospitalizations for schizophrenia, controlling for seasonal and long-term trends, day of the week, mean temperature, and relative humidity. Individuals were stratified by sex, age and marital status. Second, attributable fraction (AF) and attributable number (AN) were calculated as a reflection of attributable risk. Third, we highlighted the extent to which TV0-t affected schizophrenia by comparing with diurnal temperature range (DTR) and the temperature change between neighboring days (TCN). Results36,607 patients included totally. RR of hospitalizations for schizophrenia per 1 °C rises in TV peaked at TV0–7 (RR = 1.023, 95% confidence interval (CI): 1.008–1.039). And 823 (95% CI: 291–1374) cases were attributable to TV0–7 in total, with the AF of 2.25% (95% CI: 0.79%–3.75%). The male patients presented higher AF and AN than the female at TV0–1-TV0–7. The married patients seemed more affected by prolonged TV exposure than the unmarried. At short TV exposure (0–1 days), patients aged ≥61 years old had higher RR, but greater AN was observed at 41–60 years old. Compared with DTR and TCN, TV0-t accounted for more impacts on hospitalizations. ConclusionsTV is significantly associated with increased risk of schizophrenia hospitalizations. Effective guidance such as focusing on the timing for post-exposure prevention for vulnerable groups is recommended, which has important implications for risk management of schizophrenia.