Association between floods and hospital admissions for schizophrenia in Hefei, China: The lag effects of degrees of floods and time variation

Abstract

Background: Schizophrenia is a serious mental disorder, endangering 7.5 million patients in China. Floods, as the adverse consequence of temperature-rise, have a negative influence on mental health. However, the relationship between floods and schizophrenia is still insufficient. This study aimed to quantitative the relationship between floods and the admissions for schizophrenia in Hefei, China.

Methods: A Poisson generalized linear model (GLM) combining a distributed lag non-linear model (DLNM) was used to quantify the lag effects of floods on schizophrenia and subgroups (male, female; ≤40 y, >40 y; the married, the unmarried) from 2005 to 2014, Hefei, China. We further explored the effects of different degrees (moderate and severe) of floods and their temporal changes on schizophrenia.

Results: There was a significant association between floods and admissions risk for schizophrenia. And the lag effects for schizophrenia lasted ten days (lag 5-lag 14), with the greatest effect on lag 9 (RR = 1.036, 95% confidence interval (CI): 1.014-1.058). The married, ≤40 y were sensitive to floods. The significant difference wasn't found for genders. The effects of the severe flood were higher than moderate floods, with the largest RR of 1.073 (95%CI: 1.029-1.119). The adverse effects were found in the middle and late period with a decreasing trend in the later period.

Conclusions: This study suggests a significant association between floods and schizophrenia with ten days of lag effects in Hefei, China. Male, female, <40 y and the married are vulnerable to both moderate and severe floods. The findings might be used to allocate medical resources of mental health after floods.

Keywords: Distributed lag non-linear model; Flood; Schizophrenia; Vulnerable populations.