Heat and mortality by age group in Brazil, 2014–2024: a distributed lag non-linear time-series study of 11 million deaths
This study of 11 million deaths across 14 Brazilian federative units from 2014 to 2024 reveals that the acute mortality risk from extreme heat increases significantly with age, being approximately three times higher for those aged 85 and older compared to those under 65, with no statistically significant evidence of compensatory mortality reduction (mortality displacement) within a 21-day window, thereby supporting the prioritization of adults aged 75 and older in heat response plans.
Original paper licensed under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer
The air is getting hotter, and the world is getting older. These two trends are converging in a way that poses a serious challenge to public health. When temperatures spike, the human body struggles to cool itself, a process that becomes increasingly difficult as we age. Older bodies lose the ability to regulate temperature efficiently, and the chronic health conditions that often accompany aging make the stress of extreme heat even more dangerous. For decades, scientists have known that heatwaves kill more people than almost any other weather event, but a critical question has remained unanswered for many regions: exactly how much more dangerous is the heat for the very old compared to the middle-aged, and does the spike in deaths stop quickly once the heat passes?
For years, a prevailing theory in some circles suggested that heat might not be killing as many people as it seemed. The idea was that heatwaves might simply "bring forward" deaths that were already inevitable. In this view, a person with a fragile heart might die a few days earlier than they would have otherwise, meaning the total number of deaths over a month wouldn't change much; the heat just shifted the timing. If this were true, the immediate tragedy of a heatwave would be less severe in the long run. However, testing this idea requires looking at death rates not just during the heat, but for weeks afterward, to see if there is a "cancellation" period where fewer people die than usual.
A new study from Brazil has tackled these questions with unprecedented scale and precision. Researchers analyzed more than 11 million deaths recorded across 14 different states in the country between 2014 and 2024. They paired these death records with daily temperature data from hundreds of weather stations, creating a massive, detailed map of how heat affects people of different ages. The team focused on four distinct age groups: those under 65, those between 65 and 74, those between 75 and 84, and those 85 and older. By comparing days with extreme heat to cooler days, they could measure exactly how much the risk of dying increased for each group.
The results paint a stark picture of vulnerability. The study found that the risk of death rises steadily with age, creating a steep gradient. On a day of extreme heat, the risk of dying for someone aged 85 or older is nearly three times higher than for someone under 65. Specifically, in the week following an extreme heat day, the excess risk of death for the oldest group was about 33 percent, compared to roughly 12 percent for the youngest group. This gap widens even further when looking at a three-week window, with the oldest group facing a 44 percent increase in risk compared to 13 percent for the younger group. The danger is not just immediate; it lingers. For the oldest adults, the elevated risk of death persists for up to two weeks after the heatwave, whereas for younger people, the risk returns to normal much faster.
Perhaps most importantly, the study looked for the "cancellation" effect—the idea that deaths are merely brought forward. The researchers scanned the three weeks following a heat event to see if there was a period where fewer people died than expected, which would suggest that the heat had only shifted the timing of inevitable deaths. They found no statistically significant evidence of this compensatory reduction in deaths during the weeks after the heatwave. While the study design cannot rule out that a small number of deaths were brought forward, nor can it estimate the exact fraction of deaths that were merely shifted, it found no statistically significant evidence of a compensatory dip in mortality within the observed three-week window.
This lack of statistically detectable compensation changes how we should think about heatwaves. It suggests that the immediate spike in deaths is not just a reshuffling of the calendar, but likely represents a genuine increase in the total number of lives lost, at least within the timeframe observed. The findings are particularly clear for the oldest population, where the risk remains elevated for weeks. The study also confirmed that this pattern holds true regardless of how the heat was defined, whether by looking at official meteorological records of heat events or by using statistical thresholds of temperature. The consistency across different methods and regions strengthens the conclusion that the age gradient is real and the lack of statistically significant compensation is robust.
The implications for public health are direct and urgent. Because the risk is so much higher for the very old, and because the danger does not disappear quickly after the heat breaks, emergency response plans need to be tailored specifically for this group. The study suggests that efforts to protect the population should prioritize those aged 75 and older, with a focus not just on the day of the heatwave, but on the days and weeks that follow. It also highlights that the assumption that heat-related deaths are merely "brought forward" is not supported by the data within the observed window. In a warming world where the population is aging, the evidence points to a need for more robust, long-term strategies to protect the most vulnerable, rather than relying on the hope that the worst is over once the temperature drops.
Drowning in papers in your field?
Get daily digests of the most novel papers matching your research keywords — with technical summaries, in your language.