Heat and air pollution appear to compound each other in two suicide studies

If you or someone you know is struggling, contact a local crisis line. This article describes research findings and is not clinical guidance.

Environmental exposures rarely arrive one at a time. Heat waves and poor air quality tend to share the same weather — stagnant air, strong sunlight, high pressure — yet epidemiological studies usually treat them as separate variables and adjust one away when estimating the other. Two studies published in February take the opposite approach, and both find that the combination is doing something the individual exposures do not fully explain.

The Utah analysis: an interaction, not just an addition

The first study, in Environment International, examined 7,551 suicide cases in Utah between 2000 and 2016 [s1]. Each decedent's residential address at the time of death was linked to daily maximum wet bulb globe temperature (WBGT) — a heat-stress index that combines temperature, humidity, wind and radiation, rather than air temperature alone — using data from the European Centre for Medium-Range Weather Forecasts, and to modelled PM2.5 and nitrogen dioxide concentrations from a national spatiotemporal ensemble [s1].

The design was a case-crossover with conditional logistic regression, in which each case effectively serves as its own control on nearby days [s1]. That structure removes confounding from anything about the person that does not change week to week. The researchers looked at single days preceding the death (lags 0 through 6) and at averages across preceding days (lags 0–1, 0–3 and 0–6), and stratified by season [s1].

Across all seasons, maximum WBGT was associated with suicide: an odds ratio of 1.05 (95% CI 1.01–1.10) per 5 °C increase over lag 0–3 days [s1]. In the warm season, defined as 22 March to 21 September, the associations were stronger, with odds ratios ranging from 1.08 (1.02–1.15) to 1.20 (1.10–1.30) per 5 °C increase depending on the lag period [s1].

The central finding is the interaction. In the warm season, the authors observed synergistic interactions between WBGT and both PM2.5 and NO2, associated with higher odds of suicide, with the WBGT association most pronounced at high NO2 levels [s1]. "Synergistic" here has a specific statistical meaning — the joint effect exceeded what the two exposures would produce independently — and it is the reason the authors argue that heat stress and air pollution should be considered together in prevention planning rather than one at a time [s1].

The Louisiana analysis: what the pathway might be

A study published in PLOS Mental Health ten days earlier approached the same relationship from the other direction, asking not how many deaths but what mechanism [s2]. It analysed 11,684 suicide hotline risk assessments in Louisiana between 2019 and 2023, focusing on night-time temperatures and using distributed lag nonlinear models, with natural language processing applied to counsellor notes to classify the crisis contexts callers described [s2].

Suicide-related help-seeking rose steeply with night-time temperature: 19% higher at the 90th percentile (prevalence ratio 1.19, 95% CI 1.10–1.29), 55% higher at the 95th (PR 1.55, 1.44–1.68) and 166% higher at the 99th (PR 2.66, 2.42–2.92) [s2]. In absolute terms the authors estimate an increase of 19 suicide-related calls per 100 crisis calls in the two days following hot nights [s2].

The counsellor-note analysis is what makes the paper more than a dose-response curve. At the 99th percentile of night-time temperature, mentions of sleep difficulties were 146% higher, expressed intent to die 163% higher, availability of lethal means 145% higher, and few future plans 224% higher [s2]. Crises related to mental health rose 20.9% and those related to basic needs 59.9% at the 90th percentile [s2]. Sleep-related crisis content ran in the opposite direction at the cold end, being 68.1% more prevalent at the 10th percentile and 31.9% less prevalent at the 90th [s2].

What these studies do not establish

Neither study can demonstrate causation, and both say so in their own terms. The Utah analysis is observational, geographically confined to one state, and covers deaths through 2016 [s1]. The Louisiana study describes its demographic stratifications as exploratory because of high missingness in demographic fields, and notes its results are most applicable to other humid subtropical climates with similar populations [s2].

The two designs also measure different outcomes — deaths in one, calls to a crisis line in the other — and a change in help-seeking is not the same as a change in risk, though it plausibly tracks it.

What the pair adds up to is a fairly specific hypothesis: that heat's association with suicide risk is partly mediated by sleep disruption and by pressure on basic needs [s2], and that ambient air pollution amplifies rather than merely accompanies it [s1]. Both are testable, and neither has been tested in a design that could rule out confounding by other things that co-vary with hot, polluted days.

What to watch

The Louisiana authors point toward operational responses that follow directly from their data — temperature-triggered staffing protocols at crisis lines, and integrating basic-needs support into suicide prevention during extreme weather [s2]. Those are cheap to trial and would generate the kind of prospective evidence neither of these retrospective designs can produce.

The Utah authors' recommendation is narrower and aimed at researchers: stop estimating heat and pollution effects in isolation when the exposures arrive together [s1].

Sources

• [s1] Independent and interactive effects of wet bulb globe temperature and air pollution exposures on suicide mortality. Environment International, 21 February 2026. https://doi.org/10.1016/j.envint.2026.110152
• [s2] Temperature extremes contribute to suicide-related help-seeking through multiple pathways: evidence from crisis hotline data (2019–2023). PLOS Mental Health, 11 February 2026. https://doi.org/10.1371/journal.pmen.0000501