Heat waves impact what happens in Austria’s hospitals: overall, fewer people are admitted for inpatient care ‒ yet at the same time, in-hospital deaths are rising. Older adults are especially affected. That’s according to CSH researchers, who drew on historical health and weather data and model calculations, illustrated in an interactive visualization.
Austria is looking back on a summer of unprecedented heat. Over the past several decades, heat waves have become longer, more frequent, and more intense.
How do longer heat spells affect hospital admissions and deaths? Researchers at the Complexity Science Hub (CSH) set out to answer that question, teaming up with the Austrian Institute of Technology (AIT), the weather service UBIMET, and other project partners. The team examined historical weather data (2007–2019) — specifically, how many heat days (a daily high above 30°C, or 86°F) occurred in any given two-week window. They also analyzed health data covering every inpatient hospital stay in Austria between 2007 and 2019. The researchers then linked the two datasets to learn more about how rising heat affects both the number of hospital admissions and the risk of dying in the hospital.
The researchers then projected this relationship forward across the coming decades, calculating what increasing heat could mean through 2075 under two different climate scenarios — one with high and one with low emissions.
Across all 32 healthcare regions into which Austria’s health system is organized, two central findings emerged. During heat waves — defined as multiple days above 30°C occurring directly in a row or in close succession — fewer people are admitted for inpatient care. At the same time, heat waves raise the risk of dying in hospital care. “As the population ages, this vulnerability keeps growing — as heat days become more frequent, the health burden can intensify,” says Katharina Ledebur, a postdoctoral researcher at CSH. “Our calculations aren’t a forecast; they show how this burden could evolve under different assumptions about climate and population. This can help target heat protection and healthcare planning for the groups and regions most affected.”
Together with CSH data visualization specialist Liuhuaying Yang, the team built the interactive visualization “Heat and Health in Austria”, which illustrates these relationships along with projections for the future.
KEY FINDINGS
More heat, but sharp regional differences: The visualization maps the number of heat days (days with a high above 30°C/86°F) for every cell of a nationwide grid with roughly one-kilometer resolution, covering Austria from 2025 to 2075. Under a scenario of continued high greenhouse gas emissions, the projected number could more than double by 2070 compared with today.
More deaths in the hospital: Based on the different climate scenarios, the researchers calculated how the dynamics of hospital admissions and in-hospital deaths could evolve through 2075. In scenarios without meaningful climate action, the number of excess deaths in the hospital during heat waves could rise by an average of 50 percent.
Older adults are hit hardest: For the 2025–2035 period, about 83 percent of the modeled heat-associated excess deaths in hospitals occur among people age 65 and older. But people between 45 and 65 are already notably more at risk than younger adults.
Focus on urban centers and eastern Austria: The projected excess mortality is highest in the areas most affected by heat — eastern Austria and all of the country’s urban centers — which, according to population forecasts, is also where most older adults will be living in 2070. This pattern is already visible for 2025–2035 and would keep intensifying through 2070. The east already accounts for a disproportionate share of excess deaths: “For 2025–2035, our calculations show an annual average of roughly 39 heat-associated excess deaths in hospitals in Vienna, about 27 in Lower Austria, and about 14 in Upper Austria. Together, that’s roughly 82 percent of the modeled nationwide total,” Ledebur explains.
Fewer admissions: The calculations show that total inpatient admissions tend to decline more sharply the more heat days occur within the preceding 14 days. That doesn’t hold across all diagnostic groups, however — admissions actually rise for some conditions. “So a drop in the overall number doesn’t automatically mean a lower health burden,” Ledebur says. “Possible explanations include changes in behavior and in how people seek medical care during hot spells. At the same time, the mix of hospital cases shifts. But based on the data available to us, we can’t pin down exactly which mechanisms are driving the decline.”
The effect of longer heat spells: For two weeks of daily highs above 30°C (86°F), the calculations show 4.6 percent fewer hospital admissions nationwide — but 22.3 percent more in-hospital deaths. For five heat days within a 14-day window, admissions drop by 3.5 percent while deaths rise by 4.4 percent.
Sustained climate action could significantly cut heat-associated hospital deaths: Under the low-emissions future scenario, the modeled number of excess in-hospital deaths stays comparatively stable through 2070. For 2065–2075, it comes in roughly 65 percent lower than under the scenario with continued high emissions.
Increase in certain diagnoses during heat waves: The relationship between heat and in-hospital mortality also varies by diagnosis. It is particularly pronounced for certain infectious, respiratory, and neurological diseases. The analysis focuses on the diagnoses associated with the hospital stay, not the individual cause of death.
Greening makes a difference, too: For Vienna, the researchers also examined how greening scenarios affect heat and hospital dynamics. According to the calculations, even a moderate amount of additional greening — equivalent to de-sealing 11.6 square kilometers of surface area — would lower the number of heat days between 2055 and 2065 by an average of 1.2 days, and by as much as 14.5 days in individual districts, Ledebur says: “For 2065–2075, the modeled number of excess in-hospital deaths would then be about 4 percent lower.” The more de-sealing, the greater the effect.
The health projections come with an important caveat: the underlying data run from 2007 to 2019. Since then, extreme heat days reaching up to 40°C (104°F) have become considerably more common. “So it’s possible that our models substantially underestimate the true heat burden and the associated excess mortality in hospitals,” Ledebur says.
“In general, the observed effects grow stronger the more heat days occur within a 14-day window,” says Peter Klimek, who leads CSH’s Healthcare & Medicine research group. “Because climate change means we have to plan for longer heat spells more often, our findings underscore just how important adaptation measures are.”
On the question of air conditioning, Katharina Ledebur says: “Our data doesn’t cover which hospitals are air-conditioned. But we also see more deaths occurring on the very first day of admission, which suggests that part of the health burden likely builds up before a patient is even admitted. Air conditioning matters, but it should be part of a broader heat-protection strategy.”
About this research
You can find the Complexity Science Hub’s interactive visualization “Heat and Health in Austria” here.
Here you can find the study, currently available as a preprint under the title “Hospital Evidence on the Health Impacts of Cumulative Heat.”
More information on the “Heat Protect” project is available here. Heat Protect is supported by the Austrian Research Promotion Agency (FFG). Additional project partners include the Austrian Institute of Technology (AIT), the Medical University of Vienna, Caritas Vienna, Johanniter Austria Training and Research, UBIMET, The Austrian National Public Health Institute (Gesundheit Österreich GmbH, GÖG), and the Ludwig Boltzmann Institute for Digital Health and Patient Safety.
