Three June papers count East Asia's heat burden in three different ways

Heat harms health in ways that do not share a single unit of measurement. Three papers published in Environment International this month, all using East Asian data, make that concrete: one counts deaths that already happened, one counts births that have not happened yet, and one counts ambulance calls in a country where the population is ageing faster than the climate is warming.

Heat and ozone arriving together

The first, published on 7 June, examines Hong Kong — dense, subtropical, and heavily monitored [s1]. The authors assembled daily mortality, meteorological and air pollution data from March to November for the years 2000 to 2021, and identified compound heat-ozone events, which they sorted into three categories: same-day events, overlapping events and successive events [s1].

Compared with non-event days, same-day events were associated with increased risk of non-external mortality (relative risk 1.05, 95% CI 1.01-1.10), respiratory mortality (RR 1.13, 95% CI 1.03-1.23) and pneumonia-related mortality (RR 1.12, 95% CI 1.01-1.24) [s1]. Overlapping and successive events were also linked to raised mortality risk, except for cardiovascular mortality [s1].

The effect was more pronounced among women (RR 1.07, 95% CI 1.01-1.13) and among older people (RR 1.11, 95% CI 1.06-1.17) [s1]. An additive interaction between heat and ozone exposure was notably observed among the elderly, particularly in overlapping-event scenarios [s1].

The finding the authors emphasise is methodological rather than epidemiological: the definition of a compound event determines the result. Successive events showed effects only for specific causes of death, while same-day and overlapping events showed broader associations [s1]. Studies that pick one definition without saying why are not measuring a stable quantity.

This is a time-series analysis using population-level exposure and administrative death records. It cannot follow individuals or establish causation.

Births that arrive too early

The second paper, published on 14 June, projects heatwave-related preterm birth across Chinese cities from 2015 to 2100 [s2].

A heatwave was defined as two or more consecutive days with daily maximum temperature above the 90th percentile for 1995-2014, and burdens were estimated at city level under three Shared Socioeconomic Pathway scenarios, in cases, deaths, disability-adjusted life years and the value of a statistical life [s2].

Heatwave-related preterm births showed a fluctuating upward trend under all three scenarios [s2]. By the 2090s, cases per million births are estimated at 2,314.64 (95% CI 1,112.30-3,518.23) under SSP1-2.6, 4,335.15 (2,651.81-6,020.31) under SSP3-7.0 and 5,217.16 (3,606.26-6,829.17) under SSP5-8.5 [s2].

One number moves the other way. Attributable deaths in infants under one year are projected to fall from 19.91 thousand (95% CI 11.37-28.32) in the 2020s to 6.58 thousand (4.61-8.51) in the 2090s under the highest-emissions scenario [s2]. That is not heat becoming less dangerous; it reflects the modelled trajectories of birth numbers and infant survival over eight decades. More heat-attributable preterm births, fewer heat-attributable infant deaths, is a coherent pair of projections — and a good illustration of why single summary figures mislead.

Attributable DALY rates are projected to increase after 2023, concentrated in the Sichuan Basin and the Beijing-Tianjin-Hebei urban agglomeration [s2].

These are projections built on emissions scenarios, demographic models and an exposure-response function estimated from historical data. Every one of those layers carries uncertainty that compounds across an 85-year horizon.

When the ambulance peak moves seasons

The third paper, published on 18 June, addresses a question specific to a country with the world's oldest population [s3].

The authors analysed 26,595,933 all-cause emergency ambulance dispatch cases among older adults in Japan between 2011 and 2019, and projected incidence to 2099 under combined climate and population scenarios [s3]. They used the peak-to-trough ratio and attributable fraction to measure the amplitude and burden of seasonality [s3].

Annual case counts are projected to rise by 15% until mid-century and then decline, as the ageing bulge passes through [s3]. The annual incidence rate is expected to increase by 2.95% to 4.25% by the 2090s under various scenarios [s3].

The structural finding is the seasonal one. The high-incidence peak is projected to shift from the cold season to the warm season, while high incidence in the cold season persists [s3]. Both the peak-to-trough ratio and the attributable fraction are projected to increase, with more pronounced changes under the extreme climate scenario [s3].

A health system that staffs for a winter peak and gets a summer one as well — rather than instead — has a different problem from one that simply gets busier.

Reading the three together

None of these studies is about the same outcome, and none can be summed with the others. What they share is a design pattern: link a long administrative record to a gridded climate variable, estimate an exposure-response relationship, then push it forward under scenarios.

The strength of that pattern is that the historical half is real data on very large numbers — 22 years of Hong Kong deaths, 26.6 million Japanese ambulance dispatches [s1][s3]. The weakness is that the projected half is arithmetic on assumptions, and the further out it runs the more the assumptions dominate.

What to watch

Whether compound-event definitions are standardised, which the Hong Kong authors argue is a precondition for targeted public health strategies [s1]. And whether emergency medical services in ageing, warming countries plan for two seasonal peaks rather than one [s3].

Sources

• [s1] Assessing the impact of compound heatwave-ozone event with varied definitions on mortality risk in Hong Kong. Environment International, 7 June 2025. https://doi.org/10.1016/j.envint.2025.109595
• [s2] Projecting future disease and economic burdens of heatwave-related preterm birth in China under different climate scenarios. Environment International, 14 June 2025. https://doi.org/10.1016/j.envint.2025.109608
• [s3] Projected compound effects of population aging and climate warming on emergency ambulance demand in Japan. Environment International, 18 June 2025. https://doi.org/10.1016/j.envint.2025.109619