Association between dust storm occurrence and risk of suicide: Case-crossover analysis of the Korean national death database

Abstract

Background: Asian dust storms (ADSs) have been associated with adverse health outcomes, including respiratory and cardiovascular diseases. Considering the increasing global desertification driven by climate change, it is necessary to assess dust storm-related adverse health effects for establishing appropriate public health interventions. Recent studies have found that ambient air pollution has negative effects on mental health including cognitive disorders, depression, and suicide. However, these studies mostly focused on traditional anthropogenic pollutants from traffic exhaust or fossil fuel power plants; the association between dust storms and suicidal death is yet to be determined. Objective: To assess the association between ADSs and suicide risk in Seoul, South Korea from 2002 to 2015. Methods: To determine whether increased risk of suicide is associated with occurrence of ADSs, we performed a time-stratified case-crossover study that linked the national death statistics database with ADS occurrence data from the Korea Meteorology Administration. Exposure to ADSs was compared between the day of suicide and control days, matched to the day of the week, month, and year. We further examined whether the effect of ADSs on suicide risk differed according to ADS duration and intensity. Results: Over the 14-year period, 30,704 people died by suicide and 133 ADSs were reported. Of these, 55 ADSs lasted over 2 days (long-duration ADSs), and 67 ADSs had higher levels of particulate matter < 10 μm in diameter (PM10) that exceeded the 50th percentile value over the total 133 ADS days (high-intensity ADSs). Exposure to ADS was associated with a 13.1% (95% confidence interval [CI], 4.5–22.4; P = .002) increase in suicide risk on the day of ADS occurrence. Long-duration and high-intensity ADSs were associated with a 19.8% (95% CI, 6.5–34.7; P = .003) and 17.0% (95% CI, 5.2–30.0; P = .004) increase in suicide risk, respectively. These associations remained robust after adjusting for local air pollution levels and meteorological factors. However, this association was not replicated in the unconstrained distributed lag model which revealed inferior goodness-of-fit to our data. Conclusions: Exposure to ADSs was associated with an increased risk of suicide, especially on the same day. This study provides novel evidence of a relationship between ADSs and suicide. These findings could help in establishing public health interventions for suicide prevention as well as in establishing dust storm warning systems. Future studies are warranted to confirm if our findings are replicable and to elucidate the underlying mechanisms. © 2019 The Authors