Association between ambient temperature and intentional injuries: A case-crossover analysis using ambulance transport...

Epidemiological studies based on mortality and crime data have indicated that short-term exposure to higher temperature increases the risk of suicide and violent crimes. However, there are few studies on non-fatal intentional injury, especially on non-fatal self-harm which is much more common than suicide.

We aimed to clarify how short-term exposure to temperature is associated with emergency ambulance transport caused by intentional injuries including acts of self-harm and assault.

We applied a time-stratified case-crossover design using a conditional quasi-Poisson regression model for each of the 46 prefectures. All temperatures were converted to percentile value for each prefecture, to account for the varied climate across Japan. A Distributed Lag Non-Linear Model was used to explore the temperature percentile and lag pattern. The prefecture-specific results were combined using a meta-analysis with the random effects model.

Between 2012 and 2015, the number of acts of self-harm and assault across all 46 prefectures totaled 151,801 and 95,861, respectively. We found that as the temperature increased, the relative risk (RRs) for both self-harm and assault behaviors increased in a nearly linear manner. The pooled relative risk at the 99th percentile temperature for self-harm behavior was 1.11 (95% CI: 1.07, 1.15) compared with the risk at the 1st percentile temperature, and that for assault was 1.12 (95% CI: 1.08, 1.16) at lag 0. The RRs were highest at lag0 and less than 1 at lag7-20.

The present study found that short-term exposure to higher temperature promotes the risk of emergency ambulance transport due to acts of self-harm and assault. The lag pattern indicates a possible "displacement" effect. These results suggest that exposure to high temperatures may potentially function as a trigger for intentional injuries.