Ambient temperature and nervous system diseases-related mortality in Japan from 2010 to 2019: A time-stratified case-crossover analysis.

BACKGROUND:Despite the substantial disease burden caused by nervous system diseases, few studies have examined the association between ambient temperature and nervous system diseases, especially during cold seasons. Evidence for specific disease subgroups such as Alzheimer's and Parkinson's is also lacking. OBJECTIVES:This study examined the association between short-term changes in ambient temperature and nervous system diseases-related mortality in ten Japanese prefectures from 1 January 2010 to 31 December 2019. METHODS:A two-stage analysis based on a time-stratified case-crossover study design was conducted. A conditional quasi-Poisson regression model with a distributed lag non-linear model for temperature was applied followed by a multivariate random-effects meta-analysis to obtain average associations. Mortality due to all and cause-specific nervous system diseases (major neurodegenerative diseases, Alzheimer's, and Parkinson's) were analyzed, with consideration for sex and age (0-84 and 85+ years old) subgroups. RESULTS:We analyzed 162,315 death cases due to nervous system diseases. Cold was associated with all categories of nervous system diseases-related mortality, but not heat. The estimated relative risks for cold, obtained by comparing the 2.5th percentile of daily mean temperature distribution to the minimum mortality temperature (MMT), were 1.43 (95 % CI: 1.28-1.60), 1.37 (95 % CI: 1.17-1.59), 1.53 (95 % CI: 1.07-2.01), and 1.54 (95 % CI: 1.13-2.10) for all nervous system diseases, major neurodegenerative diseases, Alzheimer's, and Parkinson's, respectively. These associations were also observed in the sex and age subgroups in general but without evidence of effect modification. Heat (the 97.5th percentile of daily mean temperature distribution vs. MMT) was associated with Parkinson's disease-related mortality among those aged 85 years and over. DISCUSSION:Cold was associated with an increased risk of nervous system disease-related mortality. The effect of heat was not observed, except for mortality among the elderly aged 85 years and older with Parkinson's disease.