Daily temperature variability in cities and mental health-related hospitalizations

Abstract

Abstract Despite plausible behavioral and physiological pathways, limited evidence exists on how temperature variability is associated with acute mental health-related episodes. This analysis explored associations between daily temperature range (DTR) and mental health-related hospitalizations in New York State during 1995-2014. A case-crossover design with distributed lag non-linear DTR terms (0-6 days) estimated associations between ZIP Code-level DTR and hospitalizations for mood, anxiety, adjustment, and schizophrenia disorders, adjusting for daily mean temperature, and how associations varied by age, sex, admission type and season. For all outcomes, across most DTR distribution, divergence from the study period daily mean (7.7 °C) yielded a positive association to a threshold (5.2 °C, 25th percentile; 12.2 °C, 90th percentile), beyond which negative associations were observed. For mood disorders, an increase in DTR from 7.7 °C to 12.2 °C was associated with a cumulative 4.8% (95%CI, 4.0%−5.6%) increase in daily hospitalization rates. This increase was highest during transition seasons (6.0%, 4.5%−7.4%) compared with summer (1.8%, 0.6%−3.0%) and winter (-0.6%, -0.7%−1.9%). Evidence for effect-modification by age group and admission type was inconsistent across outcomes, without evidence for effect modification by sex. This study highlights potential impacts of increased temperature variability in a changing climate on mental health-related outcomes.