Wind and cold weather. Woman wearing coat, scarf and knit hat outdoors

(Credit: © encierro - stock.adobe.com)

In a Nutshell

  • On the days with the most cardiac arrests, counts ran nearly a third above normal, about 19 extra cases nationwide, and those days tended to be cold and calm.
  • Colder temperatures and lower wind speeds were consistently associated with higher cardiac arrest counts, with weather patterns extending across a window of up to three days.
  • Winter months saw almost 18% more cardiac arrests than summer months, a pattern confirmed by both traditional statistics and machine learning.

On the days when cardiac arrests outside hospitals ran highest, counts climbed nearly a third above a typical day, an extra 19 or so nationwide, and those days shared a common signature: cold, still air. That’s the central finding from a five-year study that tracked every out-of-hospital cardiac arrest recorded by Hungary’s national ambulance service, the kind of sudden, deadly event where a person’s heart stops and paramedics race against the clock to save them.

Researchers compared more than 114,000 cardiac arrest cases against daily weather data collected from late 2018 through the end of 2023, publishing their results in the journal Public Health. Higher cardiac arrest counts were consistently associated with colder temperatures and lower wind speeds, and for every 1 degree Celsius the temperature dropped, daily cardiac arrests rose by 1.4%.

This is not a story about one bad flu season or a single storm. It’s a pattern that repeated across five years, and it held up even when researchers checked their work a second, completely different way, using a separate machine-learning model to test which weather factors were most useful for predicting daily cardiac arrest counts. That second check pointed to the same factors, which is why researchers say weather forecasts could someday give emergency services an early warning, enough lead time to prepare before a surge hits.

How Scientists Tracked Cardiac Arrest Cases in Hungary

Hungary turned out to be a nearly ideal place to study this question. Its ambulance service is centralized, meaning every cardiac arrest case in the country gets logged in one system, with no gaps caused by patients using different hospital networks or private providers. That gave researchers a full national picture instead of a patchwork drawn from a few cities.

After setting aside part of 2020 to avoid mixing in the disruption of the early COVID-19 pandemic, when ambulance protocols and public behavior changed dramatically, scientists were left with 1,584 days of data covering 114,830 cardiac arrest cases. On an average day, close to 61 people suffered a cardiac arrest outside a hospital. Those affected were typically older adults, around 70 on average, and more often men than women.

Weather information came from Hungary’s national weather service and included temperature, wind speed, air pressure, humidity, and pollution levels, all measured as daily national averages. Preliminary checks found strong agreement among weather stations for daily temperature, suggesting the national averages reasonably represented conditions across a country that is not especially small.

Crunching the Numbers on Weather and Heart Emergencies

To connect weather patterns to heart emergencies, the team used a statistical method built for counting daily events, adjusted to account for the fact that some days randomly see more ups and downs in the data than others. They also tested effects across a four-day window, since the body’s stress response to cold or bad air doesn’t always show up the same day something changes outside. Sometimes it takes a day or two, or even three, for the strain to catch up with someone’s heart.

Cold proved the dominant force. Even modest daily drops in temperature nudged the numbers upward. Calmer air mattered too. Wind actually appeared protective, likely because moving air helps clear out pollution and improves air quality overall, and the study found a real drop in cardiac arrests on breezier days.

Winter told the same story from a different angle. Daily cardiac arrest counts averaged close to 66 cases during winter compared with about 56 during summer, a gap of almost 18% that held up across multiple statistical checks.

pic of researcher showing weather data and emergency services on the road on a laptop screen
Weather data could help predict up to three days in advance when the number of out-of-hospital cardiac arrests is likely to rise above average, according to a nationwide study of more than 114,000 cases by researchers from Semmelweis University, the Budapest University of Technology and Economics, and the Hungarian National Ambulance Service. The study, published in Public Health, found that lower temperatures emerged as one of the most important weather-related risk factors examined. In the future, models based on weather forecasts could help people with cardiovascular disease and their families prepare for higher-risk periods, while also supporting ambulance services and hospitals in managing expected increases in demand. (Credit: Photo by Bálint Barta/Semmelweis University, Budapest, Hungary)

Confirming the Pattern With Artificial Intelligence

To make sure the effect wasn’t just noise in the numbers, researchers picked out the days with unusually high cardiac arrest counts across the five years, flagging cases that spiked well beyond normal fluctuation. Those flagged days lined up consistently with colder-than-average temperatures and calmer air, matching what the broader statistics had already shown.

Then they ran a second, independent test using machine learning, training a model on the weather data to see which factors were most useful for predicting cardiac arrest surges. It landed on the same two factors: temperature and wind speed. When researchers also fed the model information about how many cardiac arrests had already happened in the days before, its predictions improved considerably, suggesting that pairing recent case history with weather forecasts could sharpen any future warning system.

What Rising Cardiac Arrest Risk Means for Emergency Response

Weather forecasts already give people days of notice before a cold front arrives. That same notice could double as a warning for hospitals and ambulance crews about a likely rise in cardiac emergencies. Because those weather associations can take up to three days to surface, emergency services have a real window to prepare staffing and equipment before a crisis hits, rather than scrambling after it already has.

None of this means weather causes cardiac arrest on its own. Individual risk still depends on personal factors this study could not measure, from age to existing heart disease. What this research adds is a broader pattern across a whole population: when the temperature drops and the air goes still, more hearts stop, and it happens predictably enough that health systems could plan around it instead of being caught off guard.

Disclaimer: This article summarizes findings from a peer-reviewed study and is intended for general informational purposes only. It is not medical advice. The research is observational, meaning it identifies associations between weather and cardiac arrest rates across a population and does not prove that weather directly causes cardiac arrest. Individual risk depends on personal health factors this study did not measure. Anyone with concerns about heart health should consult a qualified healthcare professional, and anyone witnessing a suspected cardiac arrest should call emergency services immediately.


Paper Notes

Limitations

Researchers note that their approach looked at population-level data using national daily weather averages, which cannot capture how individual people experienced local weather conditions or account for personal risk factors. National averages may also miss regional differences during extreme events like fast-moving cold fronts, even though preliminary checks showed weather stations across the country tended to agree closely on daily temperature. The study also relied on a single country’s data, which the authors say may limit how well the findings apply to places with different climates, healthcare systems, or populations. Additionally, the combined weather model explained only about one-fifth of the day-to-day changes in cardiac arrest numbers, meaning other factors, such as behavior, healthcare access, and unmeasured variables, likely also play a role.

Funding and Disclosures

Semmelweis University provided funding for open-access publication, and one author received support through the university’s PhD Excellence Program. The research also received implementation support through Hungarian and European Union programs related to climate change research and a national cardiovascular artificial intelligence lab framework. Several authors disclosed unrelated institutional or personal funding and consulting relationships with pharmaceutical and medical device companies, which the authors stated were unconnected to this research. The funding organizations were not involved in designing the study, collecting or analyzing data, or deciding to publish the results. The authors disclosed using AI writing tools, including Anthropic’s Claude and OpenAI’s ChatGPT, along with Grammarly, to help refine the manuscript’s language, and stated they reviewed and take full responsibility for the final content.

Publication Details

Paper Title: “Meteorological associations with out-of-hospital cardiac arrest: A national population-based time-series analysis”

Authors: Ádám Pál-Jakab, Patrik Pesti, Zsuzsanna Horti-Maricza, Bettina Nagy, Boldizsár Kiss, Botond Biebel, György Pápai, Gábor Csató, Nora Boussoussou, Béla Merkely, András Gelencsér, Péter Sótonyi, Brigitta Szilágyi, and Endre Zima.

Journal: Public Health (Volume 252, 2026, article 106145)

DOI: 10.1016/j.puhe.2026.106145

About StudyFinds Analysis

Called "brilliant," "fantastic," and "spot on" by scientists and researchers, our acclaimed StudyFinds Analysis articles are created using an exclusive AI-based model with complete human oversight by the StudyFinds Editorial Team. For these articles, we use an unparalleled LLM process across multiple systems to analyze entire journal papers, extract data, and create accurate, accessible content. Our writing and editing team proofreads and polishes each and every article before publishing. With recent studies showing that artificial intelligence can interpret scientific research as well as (or even better) than field experts and specialists, StudyFinds was among the earliest to adopt and test this technology before approving its widespread use on our site. We stand by our practice and continuously update our processes to ensure the very highest level of accuracy. Read our AI Policy (link below) for more information.

Our Editorial Process

StudyFinds publishes digestible, agenda-free, transparent research summaries that are intended to inform the reader as well as stir civil, educated debate. We do not agree nor disagree with any of the studies we post, rather, we encourage our readers to debate the veracity of the findings themselves. All articles published on StudyFinds are vetted by our editors prior to publication and include links back to the source or corresponding journal article, if possible.

Our Editorial Team

Steve Fink

Editor-in-Chief

John Anderer

Associate Editor