Effects of temperature and humidity on cerebrovascular disease hospitalization in a super-aging society

Abstract

Weather conditions influence the incidence of cardiovascular disease. However, few studies have investigated the association between weather temperature and humidity and cerebrovascular disease hospitalizations in a super-aging society. We included 606,807 consecutive patients with cerebrovascular disease admitted to Japanese acute-care hospitals between 2015 and 2019. The primary outcome was the number of cerebrovascular disease hospitalizations per day. Multilevel mixed-effects linear regression models were used to estimate the association of mean temperature and humidity, 1 day before hospital admission, with cerebrovascular disease hospitalizations, after adjusting for air pollution, hospital, and patient demographics. Lower mean temperatures and humidity < 70% or humidity ≧ 70% are associated with an increased incidence of cerebrovascular disease hospitalization (coefficient, - 1.442 [- 1.473 to - 1.411] per °C, p < 0.001, coefficient, - 0.084 [- 0.112 to - 0.056] per%, p < 0.001, and coefficient, 0.136 [0.103 to 0.168] per %, p < 0.001, respectively). Lower mean temperatures and extremely lower or higher humidity are associated with an increased incidence of cerebrovascular disease hospitalization in a super-aging society.

Figure 1

Association between mean temperature or humidity and cerebrovascular disease hospitalizations. (A) Linearity was checked for continuous and categorical variables using STATA’s multivariable regression splines (MVRS) command. MVRS indicated a non-linear relationship with mean temperature. The predicted number of cerebrovascular diseases per day was univariate. The x-axis represents temperature (°C) as a continuous variable. The solid and dashed lines indicate 95% confidence intervals. (B) MVRS indicated a non-linear relationship with humidity. The predicted number of cerebrovascular diseases per day was univariate. The x-axis represents humidity (%) as a continuous variable. The solid and dashed lines indicate 95% confidence intervals.

Figure 2

Association between mean temperature and subgroups of cerebrovascular disease hospitalizations. (A) Linearity was checked for continuous and categorical variables using STATA’s multivariable regression splines (MVRS) command. MVRS indicated a non-linear relationship with mean temperature. The predicted number of ischemic stroke hospitalizations per day was univariate. The x-axis represents temperature (°C) as a continuous variable. The solid and dashed lines indicate 95% confidence intervals. (B) MVRS indicated a non-linear relationship with temperature. The predicted number of cerebral hemorrhage hospitalizations per day was univariate. The x-axis represents temperature (°C) as a continuous variable. The solid and dashed lines indicate 95% confidence intervals. (C) MVRS indicated a non-linear relationship with temperature. The predicted number of cerebral hemorrhage hospitalizations per day was univariate. The x-axis represents temperature (°C) as a continuous variable. The solid and dashed lines indicate 95% confidence intervals.

Figure 3

Association between mean humidity and subgroups of cerebrovascular disease hospitalizations. (A) Linearity was checked for continuous and categorical variables using STATA’s multivariable regression splines (MVRS) command. MVRS indicated a non-linear relationship with mean humidity. The predicted number of ischemic stroke hospitalizations per day was univariate. The x-axis represents humidity (%) as a continuous variable. The solid and dashed lines indicate 95% confidence intervals. (B) MVRS indicated a non-linear relationship with humidity. The predicted number of cerebral hemorrhage hospitalizations per day was univariate. The x-axis represents humidity (%) as a continuous variable. The solid and dashed lines indicate 95% confidence intervals. (C) MVRS indicated a non-linear relationship with humidity. The predicted number of cerebral hemorrhage hospitalizations per day was univariate. The x-axis represents humidity (%) as a continuous variable. The solid and dashed lines indicate 95% confidence intervals.

Figure 4

Flowchart of patient disposition. A total of 606,807 patients who underwent cerebrovascular disease hospitalization from 715 hospitals were included. The patients with planned hospitalization were excluded because only patients with acute hospitalized cerebrovascular disease were included for analysis. The patients with cardiovascular disease were also excluded.