Warm season hourly temperature and psychiatric emergencies: A case-crossover study of a vulnerable population in Boston, 2005–2019

Background: Previous studies have demonstrated associations between extreme heat and mental health; however, the acute temporal dynamics of this relationship, including diurnal variation and lag, remain unexplored, particularly among high-risk groups such as those who are unstably housed. In this time-stratified case-crossover study, we investigated the association between hourly ambient temperature exposure and psychiatric emergency department (ED) visits in a vulnerable population in the United States. Methods: Using data from the Boston Emergency Services Team, we identified warm-season (May-September) psychiatric ED visits from 2005 through 2019, including publicly insured and uninsured individuals in Massachusetts. Hourly ambient temperature at 1 × 1 km spatial resolution was estimated for each individual's residential location for each of the 24 hours preceding each ED visit. Results: Among 63,691 psychiatric ED visits, cumulative hourly temperature increases were associated with elevated psychiatric emergency risk over 24 hours. At the 99th percentile (31.96°C), risk increased 1.13-fold (95% confidence interval: 1.03, 1.25) compared with the minimum morbidity temperature (8.57°C). Hyper-acute exposure (0-6 hours) showed imprecise protective associations, while delayed exposure (15-18 hours) demonstrated substantially increased risk. Notably, sustained extreme heat exposure was associated with stronger risk (e.g., 14 hours of extreme temperatures yielded a 1.43-fold increased risk [95% confidence interval: 1.05, 1.94]). Conclusions: Our results demonstrate that heat impacts are acute (i.e., within a single day) but not immediate, manifesting most strongly after several hours of sustained exposure. These findings suggest diurnal temporal patterns should be considered for heat-health interventions and provide insights for clinical resource allocation during extreme heat.