As Chuseok approaches, Korean safety notices lead with bee stings and snake bites. Open the fire service data those notices cite, though, and the same tables carry a much larger number that rarely makes the headline. Across five Chuseok holidays, 1,474 people were taken to hospital after a bee sting — and 12,038 after a road accident.
1. This piece re-divides the fire service figures by day. The Chuseok holiday runs anywhere from four to six days, so comparing yearly totals as published makes the longest holiday look like the most dangerous one.
2. It puts the three categories in order of size. By people transported, road accidents are 8.2 times bee stings; by cardiac arrests, 27.7 times.
3. It separates what is rising from what is falling. On a per-day basis, road accidents are down 9.1 per cent over the five years while snake bites have roughly doubled.
The data covers five holidays — twenty-five days
The source is a National Fire Agency press release dated 10 September 2025. It attaches three tables covering the 2020 to 2024 Chuseok holidays, recording what 119 ambulance crews actually responded to and transported, split into bee stings, snake bites and road accidents.
One thing to settle first: the tables keep “dispatches”, “transports” and “people transported” in separate columns.
- Dispatches counts times an ambulance went out, including calls that ended without a transport — for bee stings, 1,310 of 1,435 dispatches led to a transport (91.3 per cent).
- People transported counts individuals taken to hospital. One run can carry several people, so this exceeds the transport count — road accidents show 12,038 people across 10,566 transports, 1.14 people per run.
- This piece works from people, because counting people is the point.
It is also worth saying plainly that this is not a count of accidents — it is a count of people the 119 service moved. Anyone who drove themselves to hospital, or was hurt and did not go, is not in here. The real totals are larger than these. Whether a given injury calls for the emergency room is something the list of emergency symptoms writes down.
The holidays are different lengths, so the totals do not compare
Transcribed as published, the tables look like this. Read the period column alongside the counts.
| Year | Holiday | Bee stings | Road accidents |
|---|---|---|---|
| 2020 | 30 Sep–4 Oct, 5 days | 144 | 2,482 |
| 2021 | 18–22 Sep, 5 days | 348 | 2,647 |
| 2022 | 9–12 Sep, 4 days | 301 | 1,871 |
| 2023 | 28 Sep–3 Oct, 6 days | 373 | 2,782 |
| 2024 | 14–18 Sep, 5 days | 308 | 2,256 |
| Total | 25 days | 1,474 | 12,038 |
2022 was a four-day holiday and 2023 a six-day one. Reading that table as “2023 was the worst year” means counting two extra days. So we divided the same figures by the number of days.
| Year | Road, per day | Bee, per day | Snake, per day |
|---|---|---|---|
| 2020 | 496.4 | 28.8 | 5.0 |
| 2021 | 529.4 | 69.6 | 6.2 |
| 2022 | 467.8 | 75.2 | 9.0 |
| 2023 | 463.7 | 62.2 | 5.0 |
| 2024 | 451.2 | 61.6 | 10.4 |
| 5-year mean | 481.5 | 59.0 | 7.0 |
The press release itself gives 59 a day for bee stings and 482 a day for road accidents. Our recomputed figures (59.0 and 481.5) match, differing only in rounding. The year-by-year split is our own working.
The biggest number is the one not in the headline
Across the five holidays, 13,686 people were transported in these three categories. Their shares:
| Category | People | Share | Per day |
|---|---|---|---|
| Road accidents | 12,038 | 88.0% | 481.5 |
| Bee stings | 1,474 | 10.8% | 59.0 |
| Snake bites | 174 | 1.3% | 7.0 |
Road accidents run at 8.2 times bee stings. Yet holiday safety messaging tends to lead with bees and snakes. A plausible reason is that those are risks attached to distinctly holiday behaviour — visiting and clearing family graves. That is our reading, though, and it is not stated in the source.
The release does not explain the road figure either, but the sheer volume of travel over the holiday is worth keeping in view. How the expressways are opened up over the same days is set out, clause by clause, in our Chuseok toll exemption piece.
For cardiac arrests the gap widens to 27.7 times
The tables carry a separate cardiac arrest column — people who were in arrest when transported.
| Category | Arrests, 5 years | Per 1,000 transported |
|---|---|---|
| Road accidents | 83 | 6.9 |
| Bee stings | 3 | 2.0 |
| Snake bites | 0 | 0 |
In raw counts that is 83 against 3 — 27.7 times. As a rate it still favours the road figure, 6.9 per 1,000 transported against 2.0. And snake bites produced no cardiac arrest at all across the five years.
But the three bee-sting arrests should not be read as small. Bee stings generate one-eighth the transports of road accidents and still produced arrests. Anaphylaxis moves in minutes. What to do in those minutes is set out step by step in our CPR and choking guide.
What has risen over five years is snake bites
Once the day counts are matched, 2020 and 2024 point different ways depending on the category.
- Road accidents — 496.4 to 451.2 a day, down 9.1 per cent. The peak was 2021 (529.4) and it has fallen every year since.
- Snake bites — 5.0 to 10.4 a day, roughly double. The absolute numbers are small, but the highest of the five years is the last one.
- Bee stings — 28.8 to 61.6 a day, but not on a straight line. The peak was 2022 (75.2), and the figure has sat in the low sixties since. 2020 is unusually low.
Why, we cannot say from this source alone. Chuseok itself moves between early September and early October from year to year, which changes how far the holiday overlaps with bee and snake activity — Chuseok 2022 fell on 9–12 September, Chuseok 2020 on 30 September to 4 October. We did not bring temperature or population data alongside these figures, so we are not offering a cause.
What the fire agency published in 2026
The National Fire Agency issued a new release on 22 September 2026 (checked 28 September 2026). It gives January–August totals and monthly shares, not holiday transport figures — a different kind of number from the 2020–2024 holiday table above, so do not mix them.
- Bee-nest removal call-outs in January–August 2026: 202,593, up 37,718 (22.9%) on 164,875 a year earlier.
- Bee-sting patients transported: 3,634 → 4,079 (up 445, 12.2%).
- Over 2023–2025, September accounted for 6,121 bee-sting transports, 29.2% of the year — the September cells of the monthly table (2,119, 1,892, 2,110) add up to the same figure (our check).
- Fires from unattended cooking averaged 6.8 a day over the same years, but 132 over 18 Chuseok holiday days, 7.3 a day.
The release does not give bee, snake or road-accident transport figures for the 2025 holiday (3–9 October), so the five-year table above could not be extended to six.
What people asked 119 most over the 2025 holiday — which clinic is open
The table above counts people 119 transported. The agency’s release of 14 October 2025 after last year’s holiday counts the other side — what people phoned 119 to ask. Over the 2025 Chuseok holiday (3–9 October, seven days) there were 56,151 consultations, 8,022 a day, 73.8% above the usual 4,616.
| Type of call | Calls | Share |
|---|---|---|
| Which clinic or hospital is open | 33,574 | 59.8% |
| Illness advice | 9,278 | 16.5% |
| First-aid guidance | 7,394 | 13.2% |
| Which pharmacy is open | 2,292 | 4.1% |
Six calls in ten were “where is a clinic open right now?” The shares are our division of the counts and match the percentages in the release. Before heading for an emergency room over the holiday, see the order of checks in ER or not.
For fires the agency published two different bases. Its release of 18 August gives 2,077 fires over the last five Chuseok holidays (9 deaths, 100 injured, about 22 billion won of damage); its release of 23 September gives 2,412 over the last five special-alert periods (123 casualties, about 28.38 billion won). They count different days, so do not set them side by side. The later release says 2025 had 644 fires, up 30.1% on 495 the year before, led by carelessness (41.5%) and electrical causes (31.8%).
Where readers usually go wrong
So is the warning about bees wrong?
No. Bee stings run at 59 people a day over the holiday, with 2.0 cardiac arrests per 1,000 transported. The problem is only that leaving out the larger number in the same tables reorders what a reader prepares for. Both belong in the notice.
Does the holiday itself cause these numbers?
This source cannot answer that. The release reports transports during the holiday period only, with no comparison against an equivalent run of ordinary days. Reading it as “the holiday made this worse” would need that baseline, and we did not obtain it.
After a bee sting, is hospital always necessary?
We are not writing a new threshold here. Where to go is covered in emergency room versus clinic, and what to do first at home in our household emergencies guide. The thing that differs over a Korean holiday is simply that far fewer places are open.
What is worth having ready?
What to keep in a home kit is in our first aid kit piece. If the question is instead what a shift over the holiday pays, that is a separate calculation in the holiday work pay calculator.
Matched day for day across five holidays, the place people are most likely to be hurt is not the hillside but the road. And the road is already improving, while the hillside is not.
Back from tidying graves or walking in grass? Watch for fever and a black scab for 10 days — scrub typhus cases peak in October and November (scrub typhus in autumn).
Mosquitoes matter in autumn outings too — this year's Japanese encephalitis warning and free vaccination are in Japanese encephalitis.
If holiday plans change and a train ticket has to be cancelled, the fee differs from ordinary days — see Chuseok train refunds.
Sources
- National Fire Agency press release — primary source — “Bee stings, snake bites, brush cutter injuries: take care of safety ahead of the Chuseok holiday” (10 September 2025). It attaches three tables of 119 transport statistics for the 2020–2024 Chuseok holidays covering bee stings, snake bites and road accidents. The dispatch, transport, people, cardiac arrest and injury columns are transcribed from those tables. Released under Korea Open Government Licence type 1.
- The per-day conversion is our own working — the release’s yearly people-transported figures divided by the holiday length (4 to 6 days) given in the same tables. Checked against the totals: 1,474 ÷ 25 = 58.96 for bee stings and 12,038 ÷ 25 = 481.52 for road accidents, which matches the “59” and “482” stated in the release’s own text.
- Three National Fire Agency releases — primary sources — 119 emergency centres over Chuseok (14 Oct 2025 · 56,151 consultations over the 2025 holiday, by type), 2,077 Chuseok fires in five years (18 Aug 2026 · 2,077 fires over the holidays), special alert over Chuseok (23 Sep 2026 · 2,412 fires over the special-alert periods, 644 in 2025, causes). Checked 29 September 2026.
- Press coverage — not a primary source — News1, “Chuseok holiday: bee stings, snake bites and road accidents rise”. Coverage of the same release, which we looked at only to confirm how the three categories were ordered in the headline. All figures here come from the release itself, not the article.
Where to check further
- A baseline of ordinary days. Judging whether the holiday is unusually dangerous needs the same day count outside the holiday — the National Fire Agency statistics portal publishes ambulance data annually.
- Cause breakdowns. These tables record outcomes, not how people came to be hurt — for road accidents, the Korea Road Traffic Authority’s TAAS system publishes cause-level data.
- The 2025 and 2026 figures. The tables here stop at 2024 — the agency refreshes them ahead of each Chuseok.
Written as of August 2026. The underlying tables are the National Fire Agency release of 10 September 2025; the per-day conversion, ratios and shares are ours — checked back against the totals so they agree with the figures stated in the release. Causes and reasons for the trends are not stated here, because this source cannot support them. Chuseok 2026 runs from 24 to 27 September, and which of those days are actually toll-free is separated out clause by clause in our toll exemption piece.
