Cover: 0.7% on Everest now. The death rate above Everest base camp halved from 1.4% to 0.7% after 2007, even as crowds grew.

Hiking and mountaineering deaths, and why Everest got safer

Key takeaways · 9 min read

  • US national parks average about 358 deaths a year, at about one per million visits; most are car crashes, drowning, falls and suicide.
  • Falls cause most traumatic hiking deaths. Sudden cardiac death is the main non-traumatic cause in men over 34.
  • Heart risk peaks on the first day of unaccustomed exertion, and with long gaps without food or drink.
  • Slow ascent and acetazolamide prevent altitude illness; ginkgo does not.

Mount Everest became four times more crowded during summit bids without becoming deadlier. A 2026 analysis of every recorded climber from 1921 to 2024 found the death rate above base camp halved, from 1.4% before 2007 to 0.7% since. The queues on the summit ridge in the famous photographs are real; the rise in deaths they seem to imply is not.

On ordinary hills the risk is lower still and looks quite different. Most people who die while hiking fall, and a large share of the rest are middle-aged men whose hearts stop on the first day of an unaccustomed climb. In US national parks, car crashes and drowning lead the causes of accidental death, ahead of wild animals or cliffs.

This article follows the series format: what hurts hikers and mountaineers, where the risk concentrates, what has been measured to reduce it, and why the answer is not to stay at the bottom.

What actually kills and injures

The US National Park Service publishes its own mortality data. From 2014 to 2019 an average of 358 people died in national parks each year, about seven a week, across 177 parks. Half the deaths were unintentional, and motor vehicle crashes, drowning and falls led among them. Half of medical deaths happened during physical activity. In 2019 the overall rate was 0.11 deaths per 100,000 recreational visits.

Deaths in US national parks, 2014–2019

Leading recorded causes, from National Park Service data. Categories overlap.

Suicide381
Motor vehicle370
Drowning314
Hiking-related255
Falls206

Source: National Park Service mortality data, as reported by National Parks Traveler (May 2024). Total 2,149 deaths; average 358 a year.

Falls dominate the traumatic deaths on the trail itself. In the Swiss canton of Bern, all of 77 fatal hiking accidents from 2003 to 2018 were falls, many of them off marked trails. Among non-traumatic deaths the pattern is as consistent: research from the Austrian Alps found sudden cardiac death the most common non-traumatic cause of death in men over 34 hiking in the mountains.

Higher up, the list lengthens. A review of mountain deaths lists trauma, high-altitude illness, cold injury, avalanche burial and sudden cardiac death, and notes that some forms of mountaineering carry death rates comparable to hang gliding or boxing. Trekking is much safer. Of 275,950 trekkers in Nepal from 1987 to 1991, 40 died, about 14 per 100,000: illness caused 14 deaths, trauma 12 and altitude sickness 10.

Deaths among Nepal trekkers, 1987–1991

Causes of 40 deaths among 275,950 trekkers, about 14 per 100,000.

Illness14
Trauma12
Altitude sickness10
Missing3–4

Source: Shlim and Gallie, International Journal of Sports Medicine (1992).

Ordinary illness killed slightly more trekkers than trauma did, and altitude sickness, the hazard most trekkers prepare for, came third. The mix changes with height and with the kind of trip, so advice that fits a day walk in the Alps does not fit a month in the Himalaya.

When the risk concentrates

An Austrian case-control study compared 247 men over 34 who suffered sudden cardiac death while hiking or skiing with 741 matched controls. Risk was highest on the first day at altitude and rose the longer people had gone without food or drink. The men who died were far more likely to have had a previous heart attack (17% against 0.9%) or known coronary disease (17% against 4%). Only 14% of the hikers received early resuscitation.

A pointillist illustration in a surreal style: a sphere of dots sitting at the foot of a steep diagonal slope, with a short, dark shadow on the first step of the slope only.
The first step up carries the weight.

The researchers concluded that unaccustomed physical activity seems more likely to trigger sudden cardiac death than altitude itself. The profile is a sedentary man who drives to a mountain and starts climbing hard, having skipped breakfast. After a gentle first day, eating and drinking as he goes, the same man has a much lower risk. Altitude illness depends in the same way on the speed of the climb rather than the height alone. In a classic study at 4,243 metres in Nepal, 53% of trekkers who had not acclimatised developed acute mountain sickness, and the rates were highest in those who flew in, climbed fast or spent fewer nights on the way up. Previous experience at altitude made no difference.

The danger on high peaks concentrates on the way down. Of 94 deaths above 8,000 metres on Everest from 1921 to 2006, 56% happened while descending from the summit. The newer analysis found 76.5% of climber deaths happened on summit day, while 82% of Sherpa deaths happened during route preparation: the workers who fix the ropes carry a different risk from the clients who use them.

Sudden cardiac death on the mountain

Share with each risk factor among men over 34 who died suddenly while hiking or skiing, against matched controls.

Previous heart attack, cases17%
Previous heart attack, controls0.9%
Known coronary disease, cases17%
Known coronary disease, controls4%
High cholesterol, cases54%
High cholesterol, controls20%

Source: Burtscher M, in Hypoxia and the Circulation, Advances in Experimental Medicine and Biology, doi:10.1007/978-0-387-75434-5_1 (2007).

What has been measured to work

Altitude illness has the best trial evidence of any mountain hazard. In a trial on the trek to Everest base camp, acetazolamide at 125 milligrams twice a day halved acute mountain sickness, from 24.7% to 12.2%; about eight people needed to take it to prevent one case. In a larger trial of 614 trekkers, 34% developed the illness on placebo and 12% on acetazolamide, while ginkgo did no better than placebo.

Preventing altitude illness

Randomised trials and current guidance.

Acetazolamide trials
Everest trek: 24.7% → 12.2% (NNT 8)
614 trekkers: 34% → 12%
Ginkgo: no better than placebo
Wilderness Medical Society 2024
Above 3,000 m: sleep no more than 500 m higher each day
Rest day every 3 to 4 days
Descent is mandatory for HAPE and HACE

Source: Basnyat et al., High Altitude Medicine and Biology 4(1):45–52 (2003); Gertsch et al., BMJ 328:797 (2004); Luks et al., Wilderness and Environmental Medicine 35(1 suppl):2S–19S (2024).

The 2024 guidelines of the Wilderness Medical Society turn that into rules. Above 3,000 metres, sleeping elevation should rise by no more than 500 metres a day, with a rest day every three to four days. Acetazolamide is recommended for moderate and high-risk ascents. High-altitude pulmonary oedema is the leading cause of death from altitude illness, and descent is mandatory for it and for high-altitude cerebral oedema.

A pointillist illustration in a surreal style: a staircase of low horizontal steps of dots climbing gently from left to right, each step the same small height, with a single steep dark rule beside them going straight up.
Small steps, or one steep line.

For the knees, trekking poles have measurable effects. In a laboratory study on a 25-degree downhill ramp, poles reduced knee joint forces and ground reaction forces by 12% to 25%, and on level ground they reduced the load on the knee extensors by 7% to 10%. The evidence is not all one way: a Nordic walking study found no reduction in knee compression, and in people with knee osteoarthritis poles slightly increased one measure of knee load. Search and rescue data explain where the rest of the effort goes. In US national parks from 2003 to 2006 there were 12,337 search and rescue operations, and day hiking accounted for more than any other activity. In Yosemite during the 1990s, most rescue fatalities occurred while hiking. For the heart risk, the Austrian findings translate directly: an easy first day, food and water on the way, and a medical check for anyone with known heart disease before a strenuous trip.

Heat is the recent case. In June 2026 the National Park Service reported three apparent heat-related deaths on inner-canyon trails at the Grand Canyon within five days: a 72-year-old man on the South Kaibab Trail, and a 67-year-old man and 68-year-old woman on the North Kaibab. Temperatures there can exceed 109°F in the shade, and the park asks visitors to stay off inner-canyon trails between 10 a.m. and 4 p.m.

Why the answer is not to stay at the bottom

Walking, which is most of hiking, has some of the strongest health evidence in this series. A meta-analysis of 15 cohorts with 47,471 adults found 40% to 53% lower mortality in the highest quarter of daily steps than in the lowest. The benefit levelled off at about 6,000 to 8,000 steps a day in adults over 60 and 8,000 to 10,000 in younger adults. An umbrella review of 40 reviews linked green space with lower mortality and better mental health, though most of those studies were cross-sectional and cannot show cause.

Everest became safer

Deaths per climber above base camp, and recent seasons.

1921–20061.4%
2007–20240.7%
Climber deaths on spring summit descent, 1921–20063.0%
Same, 2007–20240.8%
2025 spring season: deaths / summits5 / 722

Source: Dodge et al., Journal of Physiology, doi:10.1113/jp290840 (2026); Firth et al., BMJ 337:a2654 (2008); Lacrux (2025).

A pointillist illustration in a surreal style: a long line of small spheres of dots queued along a horizon line, with the line itself drawn thinner and lower at the right than at the left.
A longer queue, a lower line.

Everest offers the strongest argument that mountain risk is not fixed: crowds grew, yet the death rate halved. Another analysis found the rate fell from 1.6% in 1990 to 2005 to 1.0% in 2006 to 2019, with risk rising from about age 60. Better forecasting, fixed ropes, bottled oxygen and experienced guides are the likely reasons, though no study can attribute the fall to any one of them. Ordinary hikers can take the same lesson: most deaths come from a few predictable situations (a fall off the path, a first-day heart attack, a midday descent into heat), and each has a known countermeasure.

Against a national park death rate of about one per million visits, the choice for most people is not between safety and hiking. It is between a plan and no plan: stay on the trail, start gently, carry water and food, know the weather and turn back early.

What the official pages say about hiking deaths and preparedness

This section collects what park, health and mountain-rescue pages say about how deaths are counted in parks, what to carry, what to do when in trouble and the signs of altitude illness. It is general information, not safety training.

Source and what it says
SourceWhat it says
US National Park Service, Deaths in National Parks (updated June 2, 2026)An average of 358 deaths a year for 2014–2019. About half are unintentional: motor vehicle crashes come first, then drownings, then falls. Half of medical deaths happened during physical activity such as hiking, biking or swimming. There is no separate hiking category.
US National Park Service, What is the most dangerous park? (Feb. 12, 2025)Says there is no “most dangerous park” and that death data alone give limited insight, since many illnesses, injuries and near misses go unreported.
NPS, Yosemite park statistics (Nov. 2, 2026) and Grand Canyon park statistics (July 28, 2026)Yosemite lists 210 search and rescue operations and 14 fatalities for 2025, and 194 and 17 for 2024. The Grand Canyon lists 232 search and rescue incidents for 2025. The pages count things in their own ways.
Parks Canada, How to prepare for a hike (modified July 2, 2026), and NPS, About hiking safety (Jan. 26, 2021)Carry a map, first aid kit, whistle, layers, a light and food; Parks Canada suggests about half a litre of water per hour. Tell someone your location and expected return time, and do not rely on phone GPS alone. The NPS page lists the “Ten Essentials” (navigation, sun protection, insulation, light, first aid, fire, repair tools, food, shelter, water).
US CDC, Yellow Book: high-altitude travel and altitude illness (Apr. 23, 2025 on the page)Acute mountain sickness is a headache plus another symptom such as nausea or fatigue, starting 2–12 hours after ascent. Confusion or poor balance (cerebral oedema) and chest congestion with breathlessness at rest (pulmonary oedema) are emergencies, and descent is urgent for the latter. Acute mountain sickness improves with a descent of at least 300 m, and you should descend if symptoms worsen despite rest. Above 3,000 m increase sleeping altitude by no more than 500 m a day. It says about 25% of visitors sleeping above 2,450 m in Colorado develop it (no year given).
NHS, Altitude sickness (reviewed July 31, 2023; next review was due July 2026)Mild symptoms usually start 6–10 hours after arrival. Rest and do not go higher until they clear; if they worsen or last a day, descend about 300–1,000 m. Confusion, trouble with balance, breathlessness at rest, a frothy or bloody cough or blue lips mean descend straight away and get medical help.
Mountain Rescue England and Wales (a charity, not government), Stay safeIn the UK, dial 999 or 112, ask for Police and then Mountain Rescue, and stay where you are. Six long whistle blasts in a minute is the distress signal. It suggests registering for emergency SMS by texting “register” to 999 first.

What we could not confirm: we found no official page that ranks hiking deaths or gives a hiking death rate, because the NPS counts all park deaths and does not separate hiking, and the Yosemite and Grand Canyon figures count search operations and fatalities in their own ways, so they cannot be added or compared. We did not confirm whether the Yosemite “fatalities” column covers all park deaths. We found no official Everest or high-altitude mortality rate: a Nepal Department of Tourism document we read lists summiteers but has no fatality table. CDC and NHS set the altitude threshold slightly differently (2,450 m versus about 2,500 m) and differ on the one-day jump rule, and the CDC page has internal unit inconsistencies. NPS compares 0.11 deaths per 100,000 visits with a US population rate, which are different denominators. Emergency numbers differ by country, and an older NPS dashboard (2014–2016) is too old to use.

What to do

Steps sorted by when they matter. Each one says how strong the evidence is: Measured Official advice Untested

Before

  • Above 3,000 metres, raise your sleeping altitude by no more than 500 metres a day, with a rest day every three to four days. Official advice see Sources
  • Ask a doctor about acetazolamide before a high trip. In trials it halved acute mountain sickness; ginkgo did no better than placebo. Measured see Sources
  • If you have heart disease or several cardiovascular risk factors, get a medical check before a strenuous trip. Official advice see Sources

If it happens

  • Make the first day easy and carry food and water. Heart risk peaks on the first day of unaccustomed exertion and with long gaps without food or drink. Measured see Sources
  • Stay on the trail and use poles on steep descents. Falls cause most traumatic hiking deaths, and poles cut knee forces downhill by 12% to 25%. Measured see Sources
  • In hot canyons, stay off inner trails between 10 a.m. and 4 p.m. and turn back early. Official advice see Sources

Afterwards

  • With signs of high-altitude pulmonary or cerebral oedema, descend. Descent is mandatory for both. Official advice see Sources

Questions people ask

What is the most common cause of death while hiking?

Falls are the leading traumatic cause. Among non-traumatic causes, sudden cardiac death is the most common in men over 34.

How can I prevent altitude sickness?

Ascend slowly: above 3,000 metres, raise your sleeping altitude by no more than 500 metres a day. Acetazolamide halved the risk in trials; discuss it with a doctor before a high trip.

Is Everest getting more dangerous?

No. Despite crowding, the death rate above base camp halved, from 1.4% before 2007 to 0.7% since.

Do trekking poles protect your knees?

They reduced knee forces by 12% to 25% going downhill in a laboratory study, but not in every study or every group.

Who should see a doctor before a hard hike?

Anyone with known heart disease, a previous heart attack or several cardiovascular risk factors, especially before a strenuous or high-altitude trip.

The short version

  • US national parks average about 358 deaths a year, at about one per million visits; most are car crashes, drowning, falls and suicide.
  • Falls cause most traumatic hiking deaths. Sudden cardiac death is the main non-traumatic cause in men over 34.
  • Heart risk peaks on the first day of unaccustomed exertion, and with long gaps without food or drink.
  • Slow ascent and acetazolamide prevent altitude illness; ginkgo does not.
  • Everest’s death rate halved despite crowding, from 1.4% to 0.7%.
  • Walking more is linked to 40% to 53% lower mortality.

This is a summary of published research, not medical or mountaineering advice. It does not replace training, local guidance or a medical check. If you have heart disease or are planning a high-altitude trip, talk to your doctor first. In an emergency, call the local rescue service.

Further reading: Luks et al., Wilderness and Environmental Medicine 35(1 suppl):2S–19S (2024), the current altitude illness guidelines. Huey et al., PLoS One 15(8):e0236919 (2020), on crowding and death on Everest. The National Park Service’s mortality data page.

Sources

  1. National Park Service, Deaths in National Parks, mortality data (nps.gov).
  2. National Parks Traveler, No. 1 cause of death in national parks (May 2024).
  3. Zürcher S, Jackowski C, Schön C. doi:10.7892/boris.142504 (2020).
  4. Burtscher M, Ponchia A. Progress in Cardiovascular Diseases 52(6):507–511 (2010).
  5. Burtscher M. Heart, Lung and Circulation, doi:10.1016/j.hlc.2017.01.010 (2017).
  6. Burtscher M. Risk of cardiovascular events during mountain activities. In: Hypoxia and the Circulation, Springer, doi:10.1007/978-0-387-75434-5_1 (2007).
  7. Windsor JS, Firth PG, Grocott MPW, et al. Postgraduate Medical Journal, doi:10.1136/pgmj.2009.078824 (2009).
  8. Shlim DR, Gallie J. International Journal of Sports Medicine, doi:10.1055/s-2007-1024601 (1992).
  9. Hackett PH, Rennie D, Levine HD. Lancet 2(7996):1149–1155 (1976).
  10. Luks AM, et al. Wilderness and Environmental Medicine 35(1 suppl):2S–19S (2024).
  11. Basnyat B, et al. High Altitude Medicine and Biology 4(1):45–52 (2003).
  12. Gertsch JH, et al. BMJ 328:797 (2004).
  13. Firth PG, Zheng H, Windsor JS, et al. BMJ 337:a2654 (2008).
  14. Dodge RE, Zheng H, Windsor JS, et al. Journal of Physiology, doi:10.1113/jp290840 (2026).
  15. Huey RB, Carroll CL, Salisbury RA, Wang JL. PLoS One 15(8):e0236919 (2020).
  16. Lacrux, Everest season report (2025).
  17. National Park Service, Grand Canyon news release (19 June 2026).
  18. Schwameder H, et al. Journal of Sports Sciences 17(12):969–978 (1999).
  19. Willson J, et al. Medicine and Science in Sports and Exercise 33(1):142–147 (2001).
  20. Hansen L, et al. Scandinavian Journal of Medicine and Science in Sports (2008).
  21. Bechard DJ, et al. Osteoarthritis and Cartilage (2012).
  22. Heggie TW, Heggie TM. Journal of Travel Medicine 16(1):23–27 (2009).
  23. Hung EK, Townes DA. Wilderness and Environmental Medicine 18(2):111–116 (2007).
  24. Paluch AE, Bajpai S, Bassett DR, et al. Lancet Public Health 7(3):e219–e228 (2022).
  25. Yang BY, et al. The Innovation (2021).
  26. NPS mortality data; NPS; NPS Yosemite; NPS Grand Canyon; Parks Canada; NPS hiking safety; CDC Yellow Book; NHS; Mountain Rescue England and Wales.

Similar Posts