Sauna and Longevity: What 20 Years of Finnish Data Show

What two decades of Finnish sauna research establish

Sauna bathing is associated with lower cardiovascular and all-cause mortality in Finnish population cohorts, and the association is graded by frequency, large in size, and consistent across several separate outcomes. The association is also entirely observational. The evidence base rests on one cohort in particular, the Kuopio Ischaemic Heart Disease Risk Factor Study, known as KIHD, which recruited 2,315 middle-aged men in eastern Finland between 1984 and 1989 and followed them for a median of 20.7 years.

No trial has randomized people to years of sauna use and counted deaths. Randomized sauna research runs for weeks and measures surrogate markers such as blood pressure and arterial stiffness. A 2025 systematic review in the American Journal of Preventive Cardiology pooled 20 randomized trials of passive heating and found no statistically significant pooled effect on most cardiometabolic and vascular markers. The distance between the cohort signal and the trial results defines what this literature can and cannot support.

How large is the mortality association in the KIHD cohort?

Men in the KIHD cohort who used a sauna 4 to 7 times per week had a 40% lower rate of death from any cause than men who used a sauna once per week, with a hazard ratio of 0.60 and a 95% confidence interval of 0.46 to 0.80. The 2015 analysis by Laukkanen and colleagues in JAMA Internal Medicine recorded 929 deaths from all causes, 407 fatal cardiovascular events, 281 fatal coronary heart disease events, and 190 sudden cardiac deaths across a median 20.7 years of follow-up.

Comparing 4 to 7 sessions per week against a single weekly session, the multivariable-adjusted hazard ratios were:

• Sudden cardiac death: 0.37 (95% CI 0.18 to 0.75)
• Fatal coronary heart disease: 0.52 (95% CI 0.31 to 0.88)
• Fatal cardiovascular disease: 0.50 (95% CI 0.33 to 0.77)
• All-cause mortality: 0.60 (95% CI 0.46 to 0.80)

Intermediate users sat between the two extremes, which is the pattern a causal effect would produce and equally the pattern a graded confounder would produce. Men using a sauna 2 to 3 times per week had an all-cause mortality hazard ratio of 0.76 (95% CI 0.66 to 0.88). Session length tracked in the same direction, with sessions longer than 19 minutes carrying a sudden cardiac death hazard ratio of 0.48 (95% CI 0.31 to 0.75) against sessions under 11 minutes.

A 2018 analysis in BMC Medicine extended the design to women. Among 1,688 participants with a mean age of 63 years, 51.4% of them women, 181 fatal cardiovascular events occurred over a median 15.0 years of follow-up. Participants bathing 4 to 7 times weekly had a cardiovascular mortality hazard ratio of 0.23 (95% CI 0.08 to 0.65) against once-weekly bathers. The confidence interval is wide because few events fell in the most frequent group.

What did the same cohort find for stroke, dementia, and blood pressure?

The KIHD cohort has produced associations between frequent sauna bathing and lower rates of stroke, dementia, and newly diagnosed hypertension. All three point the same way, all three are observational, and all three draw on overlapping participants rather than independent populations.

• Stroke: a 2018 analysis in Neurology followed 1,628 men and women aged 53 to 74 with no stroke history and recorded 155 incident strokes over a median 14.9 years. Participants bathing 4 to 7 times weekly had 61% lower stroke risk than once-weekly bathers, and those bathing 2 to 3 times weekly had 14% lower risk.
• Dementia: a 2017 analysis in Age and Ageing followed the 2,315 KIHD men for a median 20.7 years and recorded 204 dementia and 123 Alzheimer's disease diagnoses. The hazard ratio for dementia at 4 to 7 sessions per week was 0.34 (95% CI 0.16 to 0.71) and for Alzheimer's disease 0.35 (95% CI 0.14 to 0.90).
• Hypertension: a 2017 analysis in the American Journal of Hypertension followed 1,621 men for a median 24.7 years and reported a hazard ratio of 0.54 (95% CI 0.32 to 0.91) for newly diagnosed hypertension among the most frequent bathers.

Why the cohort findings cannot establish causation

Observational cohorts of this design cannot establish that sauna bathing causes lower mortality. Two statisticians consulted by the UK Science Media Centre on the 2018 BMC Medicine paper set out the specific problems. Professor Kausik Ray of Imperial College London noted that people who adopt one health behaviour tend to carry other favourable characteristics that go unmeasured, so residual confounding can generate the entire association.

Professor Kevin McConway of the Open University raised a structural limitation. Roughly 2% of the Finnish participants used no sauna at all, so the comparison is between frequent and infrequent bathers rather than between bathers and abstainers. Frequent bathers also differed from occasional bathers in alcohol consumption, physical activity, income, education, and housing type. Sauna habits were recorded once at baseline, leaving decades of change unmeasured, and the findings transfer poorly to countries where sauna use is uncommon.

Reverse causation is the third problem. People who are already ill sit in a hot room less often, which would produce the observed association with no protective effect present. Investigators have addressed this by re-running analyses with early follow-up years excluded, which reduces the concern without eliminating it.

What do randomized trials show?

Randomized trials of sauna and related passive heating have not reproduced the cohort signal on vascular endpoints. The 2025 meta-analysis in the American Journal of Preventive Cardiology pooled 20 randomized controlled trials with individual sample sizes from 15 to 188 participants and intervention lengths of 2 to 15 weeks. Pooled systolic blood pressure fell by 2.46 mmHg (95% CI -5.02 to 0.10) and diastolic pressure by 1.08 mmHg (95% CI -2.79 to 0.63), neither result reaching statistical significance.

Flow-mediated dilation and pulse wave velocity showed no significant pooled change, and the review found no significant effect on resting heart rate, fasting glucose, HbA1c, cholesterol fractions, triglycerides, or C-reactive protein. Heterogeneity for systolic blood pressure was substantial, with an I-squared of 60.3%. One subgroup did move, as whole-body heating in populations with coronary risk lowered systolic pressure by 4.11 mmHg (95% CI -7.36 to -0.86). The review authors judged most included trials at moderate to high risk of bias and flagged evidence of possible publication bias.

The most direct test in a cardiac population is a 2023 randomized trial published in the Journal of Applied Physiology. Forty-one adults with stable coronary artery disease, mean age 62 years, 33 men and 8 women, took four sauna sessions per week for 8 weeks. Brachial artery flow-mediated dilation changed by 0.15% in the sauna arm against 0.07% in the control arm, with no difference between groups, and no difference appeared in carotid-femoral pulse wave velocity or in blood pressure. Heat acclimation did occur, shown by lower resting core temperature and a greater sweat rate, so the intended exposure was delivered.

What exposure did these studies actually measure?

The exposure throughout this literature is a traditional dry Finnish sauna rather than heat exposure in general. A 2025 review in Frontiers in Cardiovascular Medicine describes the standard conditions as 80°C to 100°C at 10% to 20% relative humidity, with stays of 5 to 20 minutes. During heat exposure cardiac output rises by up to 60% to 70% with an elevated heart rate and stable stroke volume, and 50% to 70% of blood flow shifts from the core to the skin.

KIHD classified frequency as 1, 2 to 3, or 4 to 7 sessions per week, and duration as under 11 minutes, 11 to 19 minutes, or over 19 minutes. Those categories describe what a Finnish population already did rather than a prescribed dose. Infrared cabins, steam rooms, and hot tubs deliver different thermal loads and have not been evaluated against mortality in any comparable cohort.

Who should avoid sauna bathing?

Sauna bathing is contraindicated in several cardiovascular conditions and is unsafe in combination with alcohol. The Frontiers in Cardiovascular Medicine review and the 2018 Mayo Clinic Proceedings review identify the following:

• Unstable angina pectoris or recent myocardial infarction
• Decompensated heart failure
• Severe aortic stenosis
• Acute chest pain or acute infectious illness
• Alcohol consumed before or during bathing, which raises the risk of hypotension, arrhythmia, and sudden death
• Antihypertensive medication taken immediately before a session, which can precipitate orthostatic hypotension

Pregnancy calls for individual medical advice. Extreme heat exposure in early pregnancy has been linked to birth defects, while the Frontiers review notes that most Finnish women bathe weekly throughout pregnancy and that Finland records the lowest incidence of anencephaly in the world. Anyone with established cardiovascular disease, uncontrolled blood pressure, or a complicated pregnancy should consult a clinician before beginning regular heat exposure.

What the evidence supports today

The defensible summary is narrow. In Finnish populations with lifelong sauna habits, frequent bathing is associated with substantially lower cardiovascular and all-cause mortality across follow-up as long as 24.7 years, and no randomized trial has yet shown that adding sauna sessions improves vascular function or blood pressure over weeks to months.

Sauna bathing carries a plausible physiological mechanism, a consistent dose-response gradient across four outcomes, and a decades-long observational record that survives adjustment for conventional risk factors. The literature still lacks the trial evidence that would rule out confounding by the generally healthier lives of frequent bathers. The same distance between a strong observational gradient and a demonstrated cause runs through the physical capacity measures that predict healthspan. Treating sauna as a low-risk habit with a promising but unproven mortality benefit matches the evidence as it stands.