First global review links heat to children's mental health, and finds a data gap

A systematic review published 6 August in the International Journal of Epidemiology — described by its authors as the first global synthesis of its kind — found that rising temperatures and heat waves are associated with worse mental health outcomes in children and young people, while also showing how little research exists on the places likely to be hit hardest by heat in the coming decades [s1].

What the numbers actually say

The review pooled 23 studies: 19 examining ambient temperature and four examining heat waves specifically [s1]. Across the temperature studies, each 1°C rise in temperature was associated with roughly a 0.9% increase in poor mental health risk (relative risk 1.009, 95% CI 1.003–1.016) [s1]. Narrowing to children and young people specifically — ages 5 to 18 — that per-degree risk rose to about 1.5% (RR 1.016, 95% CI 1.005–1.027) [s1]. For heat waves specifically, the same 5–18 age group showed a 25% higher risk of poor mental health outcomes (RR 1.253, 95% CI 1.006–1.561) [s1].

Read those confidence intervals carefully: the heat wave estimate's interval runs from barely above 1.0 to well above 1.5, which is a wide range for a single pooled estimate — consistent with a real effect, but not precise enough to say exactly how large it is. The per-degree temperature estimates are tighter and more consistent across studies.

Where the evidence comes from — and where it doesn't

Ten of the 23 included studies were conducted in the United States and three in China; the review's authors state explicitly that no studies came from the Pacific region or Africa [s1], despite those regions facing some of the most severe and fastest-rising heat exposure globally. The authors also flagged that weaker associations observed in middle-income countries may reflect underreporting of mental health symptoms in those settings rather than genuinely lower vulnerability [s1] — in other words, an apparent gap in the data could be a measurement problem, not a health difference.

Flinders University researcher Jacqueline Stephens, a co-author, said children are more susceptible to climate-linked mental health effects in part because of "social and developmental stressors," while co-author Corey Bradshaw pointed to children's "physiological and behavioral susceptibility to heat" specifically [s1].

The limits the authors themselves flagged

The review reports substantial heterogeneity across the included studies (I² = 82.6%), meaning the studies did not agree closely with each other on effect size — a sign that results should be read as a directional signal rather than a precise, generalizable number [s1]. Only 32% of the temperature studies adequately controlled for confounding factors that could independently explain both heat exposure and mental health outcomes, such as socioeconomic status or air quality [s1]. The heat wave estimate rests on just four studies, a small enough base that a single differently-designed study could shift the pooled result meaningfully. The review was also restricted to English-language publications, which likely excludes research conducted in some of the hottest, least English-publishing regions — compounding the geographic gap the authors already noted.

The authors' own framing captures the balance: "High temperatures and heatwaves were associated with adverse mental health risks in children and young people. Data were lacking for some of the hottest and most socio-economically disadvantaged regions globally" [s1]. That is a study establishing a real association and simultaneously showing how much is still unknown about where it matters most.

What to watch next

The evidence gap the authors identified is itself now a research target: whether the same heat–mental-health relationship holds, and at what magnitude, in the Pacific islands, sub-Saharan Africa, and South Asia — regions absent from this review but facing among the highest projected heat exposure. Longer-term, higher-quality studies that control more rigorously for confounding would also narrow the current uncertainty, particularly around heat waves specifically, where only four studies currently exist worldwide.