Table of Contents
The 30 Deaths Emergency Rooms Counted
Between May 15 and August 9, 2026, the Korea Disease Control and Prevention Agency’s (KDCA) emergency room surveillance system for heat-related illness recorded an estimated 30 deaths. Cumulative patients reached 3,320. On August 7 alone, 202 new patients and 3 deaths were reported, pushing the running totals to 3,089 patients and 26 deaths that day. Additional reports since then have brought the death toll to 30. This is the figure the government most often cites when it needs to give the public a fast read on how bad the heat wave has been.
But reading that number as the full scale of this summer’s heat-related deaths risks underestimating the actual toll. The KDCA figure is built almost entirely from cases reported by 516 hospitals nationwide that operate emergency rooms. People who died at the scene, were taken to facilities without an ER, or died after an existing condition worsened in the heat can fall outside this surveillance net. In other words, a statistic designed to track trends quickly and a statistic designed to confirm the final death count serve different purposes and cover different ground.
The Gap Between 298 and 104, Confirmed Over Five Years
The gap becomes clearer when you line up several years of data side by side. According to cause-of-death statistics that Rebuilding Korea Party lawmaker Kim Sun-min obtained from the National Data Agency, 298 people died of heat-related illness between 2020 and 2024. Over the same period, the KDCA’s emergency room surveillance system recorded only 104 heat-related deaths — just 34.9% of the cause-of-death total. The two statistics are measuring the same disaster, but capturing it through different scopes and different timelines.
In 2022, the emergency room surveillance system counted 9 deaths. When the National Data Agency’s cause-of-death statistics for that same year came out the following year, the number of heat-related deaths had risen to 34. National statistics that compile death certificates and causes of death can capture cases that never passed through an ER, but the tradeoff is a year-long delay before the results are released. The emergency room system, by contrast, shows in near real time where, at what age, and at what time of day patients are falling ill during a heat wave — but it was never designed to be the final count of the dead.
This isn’t simply a matter of numbers getting revised upward after the fact. To be counted in the emergency room system, a patient has to go through the ER of a participating hospital, and medical staff there have to identify the case as heat-related and report it. Cause-of-death statistics classify the cause of death using death certificates and vital statistics data, which means a wider set of cases and a more thorough process. Between the fast information available in the middle of a heat wave and the fuller picture confirmed afterward, there’s a structural time lag built into the system.
Why Heat-Related Deaths Are Hard to See in the Moment
The damage extreme heat does to the body doesn’t always show up with a label like heat stroke or heat exhaustion attached to it. High temperatures can worsen existing conditions — cardiovascular and cerebrovascular disease, respiratory illness, kidney disease. When an elderly person dies alone in a sweltering home, or an outdoor laborer collapses on site and never makes it to an emergency room, the link between that death and the heat isn’t something anyone can confirm right away. Cases where a patient’s condition deteriorates rapidly before medical staff can fully check body temperature and exposure history make a heat-related diagnosis even harder to pin down.
The KDCA itself says the emergency room system was never meant to investigate every heat-related death — its job is to track trends quickly. And it does that well: it shows in near real time which regions and age groups are seeing more patients, and where risk is concentrated, whether that’s outdoor worksites or farm fields. The problem arises when a provisional, breaking-news-stage number gets treated as if it represents the full scope of the damage. A limited statistic can end up hardening into the disaster’s final tally in the public mind.
What Excess Mortality Reveals About the Indirect Toll
To fill this gap, the KDCA introduced the concept of “excess mortality” in its first Climate and Health Impact Assessment, published in 2022. Excess mortality measures how many more people actually died during a given period than would normally be expected. Because it works statistically rather than case by case, it can estimate the rise in deaths during a heat wave even when heat-related illness was never written on a death certificate — capturing indirect effects, like the worsening of underlying conditions, that a case-by-case count would miss.
Using this method, the estimated excess mortality from the 2018 heat wave was 804. That same summer, the KDCA’s emergency room system counted an estimated 48 heat-related deaths, while cause-of-death statistics confirmed 160. The wide gap between these three numbers for the same summer doesn’t mean any one of them is simply wrong. Each measures something different: the ER system counts reported patients, cause-of-death statistics confirm a direct cause of death, and excess mortality captures deaths the heat may have influenced either directly or indirectly.
Excess mortality figures also need to be read with some caution. The estimate can shift depending on how the baseline expected death count is set and how the analysis period and temperature thresholds are defined. It’s not a method for determining that any individual person’s death was caused by heat. Even so, because it can capture population-level effects — like worsening chronic disease — that a death certificate alone can’t show, it works as a complement to both the ER count and the cause-of-death statistics rather than a replacement for either.
Accurate Numbers Decide the Size of the Policy Response
Which statistic policymakers look at shapes the priority and scale of the government’s response. The cooling support, outdoor work restrictions, home visits for elderly residents, and medical readiness needed if officials believe the toll is a few dozen deaths look very different from what’s needed if the indirect toll runs into the hundreds. Heat waves are a disaster whose risk varies sharply by housing conditions, occupation, income, and age, which means understanding the damage requires tracking not just the overall number but who is being hit hardest.
When these statistics are published in isolation, the public has no easy way to understand why the provisional number from midsummer differs from the confirmed death count that comes out the following year. The emergency room figures need to be presented alongside an explanation of their scope and limits, and once cause-of-death statistics and excess mortality analysis for the same period become available, the results should be connected and shown together. Only then can people tell whether a rising or falling number reflects an actual change in risk, or just a difference in how the data was collected and what it covers.
The answer isn’t to scrap the emergency room system or slow down its real-time reporting. What’s needed is an approach that connects the provisional statistics, the cause-of-death data, and the excess mortality estimates according to what each one is actually built to do — and makes the gaps between them public. The KDCA is now working on its second Climate and Health Impact Assessment, due in 2026. Before anyone declares how severe this summer’s heat wave really was, the work that remains is building a statistical system that can show who the current count is failing to see.
Source: Financial News and Newsis, Aug. 8, 2026; Yonhap News Agency, Aug. 10, 2026; The Dong-a Ilbo, Aug. 10, 2026; SisaIN, Aug. 1, 2026; KDCA Climate and Health Impact Assessment data.
