Sauna science, opened up

The Study.

Big questions.
Room for doubt.

Heat clearly changes how the body responds. Its long-term health benefits are harder to pin down. Cold water can ease soreness, but recovery comes with trade-offs. Here’s what the research can tell us so far.

56 studies · 34 claims · Original sources linked

The research, in perspective

The picture so far.

A few useful findings to start with. Open one to see what it means, what is uncertain, and the research behind it.

01Sauna & heart healthA promising link. Not yet proof of protection.Frequent sauna users had fewer fatal heart events in a long Finnish study. We cannot tell how much of that difference came from sauna itself.Low certaintyOne group, followed over time20.7 yearsSauna habits
recorded
Heart outcomes
compared
Median follow-up · 2,315 Finnish men
Explore this findingClose this finding

What this means

This is a reason to keep investigating sauna and heart health. It is not a way to calculate your own benefit, or a reason to aim for more sessions.

Where the uncertainty is

People chose their own sauna habits. Other differences in their lives could help explain the result. Several familiar sauna headlines also come from this same Finnish group, rather than independent confirmations.

Why this level of confidence?

Low certainty. Could well change. Treat it as a lead, not a fact.

  1. Starts at low: an observational study, where people chose their own habits
  2. One group of Finnish adults, asked about sauna once
  3. More sauna went with steadily lower rates, and the difference was large
  4. A Japanese hot-bath study saw no link with sudden cardiac death, though it had few such deaths to study
What was measured
Illness and death
Who and where
Middle-aged men in eastern Finland; one analysis added women. Close to a normal session.
Independent evidence
1 study group (KIHD); a Japanese hot-bath study saw no link for sudden death

KUUME’s editorial assessment · Reviewed 2026-09-27 · How we read research

How big was the reported effect?

Large effect. Sudden cardiac death 0.37 (0.18–0.75); heart-disease death in men and women 0.23 (0.08–0.65). Rates for 4–7 vs 1 session a week.

Size and certainty answer different questions. This describes the research result, not a prediction for you.

See the full claim and its context →
Read the research 4 papers
Link to this finding
02Cold water & recoveryLess sore can be a real benefit.After hard exercise, people who used cold water generally reported less soreness than people who rested.Moderate certaintyLess soreness≠Better performanceExplore this findingClose this finding

What this means

Feeling less sore over the next few days is a useful outcome in its own right. It does not automatically mean better performance or bigger training gains.

Where the uncertainty is

Most trials were small and involved young, active men. People knew whether they were in cold water, which can influence how they rate soreness. Reviews share trials, so more reviews do not always mean more independent evidence.

Why this level of confidence?

Moderate certainty. Probably right. New research could still shift it.

  1. Starts at high: randomized trials
  2. Could not be blinded, and soreness is self-rated
  3. Reviews generally point the same way, but share trials; the soccer review’s prediction interval includes no benefit
What was measured
How you feel or perform
Who and where
Mostly young, active men. A lab, clinic or sports version.
Independent evidence
Overlapping reviews of small trials, including 10 post-match soccer studies

KUUME’s editorial assessment · Reviewed 2026-09-27 · How we read research

How big was the reported effect?

Medium effect. Self-rated soreness: standardized difference −0.55 at 24 hours to −0.93 at 72 hours (Cochrane). A review of 9 trials called it slightly better than rest.

Size and certainty answer different questions. This describes the research result, not a prediction for you.

See the full claim and its context →
Read the research 6 papers
Link to this finding
03The recovery trade-offAfter lifting, colder is not always better.Regular cold-water immersion after strength training was linked to smaller strength gains in controlled studies.Moderate certaintyRelief now≠Gains laterExplore this findingClose this finding

What this means

Short-term comfort and long-term adaptation are different goals. A recovery habit can help with one without helping the other.

Where the uncertainty is

This finding concerns repeated cold-water immersion alongside a training program. It does not show that one occasional dip undoes a workout. The strength findings should not be applied to endurance training.

Why this level of confidence?

Moderate certainty. Probably right. New research could still shift it.

  1. Starts at high: controlled training studies
  2. Eight small studies
  3. Consistent across strength and power measures; endurance gains were unaffected
What was measured
How you feel or perform
Who and where
People following strength or endurance training programs. A lab, clinic or sports version.
Independent evidence
8 studies in the strength/endurance review; a separate 16-man interval trial adds context

KUUME’s editorial assessment · Reviewed 2026-09-27 · How we read research

How big was the reported effect?

Medium effect. Strength gains: standardized difference −0.60 (−0.87 to −0.33). Endurance gains: no difference (−0.07 and 0.00).

Size and certainty answer different questions. This describes the research result, not a prediction for you.

See the full claim and its context →
Read the research 2 papers
Link to this finding
04What heat does nowYour heart responds to heat. That is not the same as exercise.A rise in heart rate is well established. Matching one body response does not establish the same long-term benefits as a workout.High certaintyA body response≠A health outcomeExplore this findingClose this finding

What this means

An immediate body response and a lasting health benefit answer different questions. We can be confident about the first while still investigating the second.

Where the uncertainty is

The lab experiment measured a short exposure in healthy young adults. It could not tell us whether repeated sauna sessions prevent illness or replace exercise.

Why this level of confidence?

High certainty. Well established. New research is unlikely to change the picture.

  1. Starts at high: measured directly, each person their own comparison
  2. Large, immediate and in line with basic physiology
  3. Only 25 people, but nobody disputes the direction
What was measured
Body signals
Who and where
Healthy young adults. A lab, clinic or sports version.
Independent evidence
1 lab study (25 people) and two reviews

KUUME’s editorial assessment · Reviewed 2026-09-27 · How we read research

How big was the reported effect?

Large effect. Heart rate up to about 131 beats a minute; systolic blood pressure down about 16 mmHg during 30 minutes at 73°C.

Size and certainty answer different questions. This describes the research result, not a prediction for you.

See the full claim and its context →
Read the research 3 papers
Link to this finding
05Sweat & detoxSome metals leave in sweat. Health benefits remain uncertain.Studies have measured arsenic, cadmium, lead and mercury in sweat, sometimes in substantial amounts in exposed workers. How much ordinary sauna lowers body stores is less clear.Low certaintyIn a sweat sample≠Change in body storesExplore this findingClose this finding

What this means

Excretion is a real finding worth investigating. To establish a detox benefit, research also needs to measure lasting changes in body levels and health.

Where the uncertainty is

Collection methods and exposures vary. The historical worker mercury case combined chelation, sweating and physiotherapy. Several chemical-excretion papers reuse the same 20 volunteers, so they are not independent confirmations.

Why this level of confidence?

Low certainty. Could well change. Treat it as a lead, not a fact.

  1. Starts at low: measured sweat composition, mostly small observational studies
  2. Independent studies find metals, but collection methods, skin contamination and exposures vary
What was measured
Body signals
Who and where
Exposed workers, healthy volunteers and adults with varied health conditions. A lab, clinic or sports version.
Independent evidence
Several groups; the 2012 review contains overlapping older reports

KUUME’s editorial assessment · Reviewed 2026-09-27 · How we read research

How big was the reported effect?

Size not reported. No single treatment effect to size. Concentration in a sweat sample is not the total amount removed or the change in body stores.

Size and certainty answer different questions. This describes the research result, not a prediction for you.

See the full claim and its context →
Read the research 3 papers
Link to this finding
06Where evidence is missingCold showers for depression? Still a hypothesis.The paper behind this claim proposed an idea. It did not test a treatment in a clinical trial.No clinical trial in this paperAn idea≠A tested treatmentExplore this findingClose this finding

What this means

A plausible explanation is a starting point for research. It does not tell us whether a treatment works, for whom, or how it compares with existing care.

Where the uncertainty is

This claim in our library rests on a theory paper with informal reports. It does not establish cold showers as a treatment for depression. Evidence about a brief mood change would answer a different question.

Why this level of confidence?

Not tested. Not tested yet: an idea without human data.

  1. No human data to score: a theory paper with informal reports
What was measured
Nothing measured yet
Who and where
No clinical trial participants
Independent evidence
None

KUUME’s editorial assessment · Reviewed 2026-09-27 · How we read research

How big was the reported effect?

Size not reported. Nothing measured.

Size and certainty answer different questions. This describes the research result, not a prediction for you.

See the full claim and its context →
Read the research 1 paper
Link to this finding

A closer look

What does it add up to?

We keep two questions separate: how sure are we that it’s true, and if it is, how big is the effect?

A large effect can still be uncertain. Several papers can also come from the same group of people.

Explore the evidence map34 claims, compared by certainty and effect size

34 claims, placed honestly.

Further right means more certainty. Higher up means a larger effect, if it is real. Select a claim to read what supports it.

34 claims about heat and cold, scored

  • 5Well supported
  • 13Some support
  • 9Early signal
  • 6Not supported
  • 1No evidence yet

If it’s true, how big is the effect?Compared with not doing it, in the studies.

Big claims, shaky proof
Big and solid, but only during the session
Large
Medium
Small
No effect foundNot supported
Size not reported
Not testedNo evidence yet
Very lowEarly signal
LowSome support
ModerateWell supported
HighWell supported

How sure are we that it’s true?Would better studies change the answer?

Grouped by how sure we are that each claim is true. The bars on each claim show how big the effect is, if it’s true.

  • Illness and death
  • How you feel or perform
  • Body signals
  • Comes with a safety note
Read every claimThe findings, their limits, and the studies behind them

Every claim

The full list, scored.

The same 34 claims as a list, grouped by what the evidence supports. Each one links to the studies behind it.

Well supported 5

High or moderate certainty. What we would stand behind today.

Some support 13

Low certainty. A real lead, and could well change with better research.

Early signal 9

Very low certainty. One small study, or a thin margin. A reason to keep looking.

Not supported 6

Studies looked for an effect and did not find one.

No evidence yet 1

An idea without human data. Nothing to score.

Scores last reviewed 27 September 2026, from 56 studies.

One study, many headlinesCompare the findings from the same Finnish study

One study, many headlines

Eight claims, one Finnish study.

Many of sauna’s biggest headlines come from the Kuopio Ischaemic Heart Disease study (KIHD): middle-aged adults in eastern Finland, asked once about their sauna habits and followed for 15 to 26 years, 1,600 to 2,300 people per analysis. Here are its findings side by side.

How many fewer cases frequent sauna users had in the KIHD study, with the likely range for each. The same numbers are in the table below.

4 to 7 sauna sessions a week, compared with 1

Psychotic disorders

79% fewerlikely 48% to 92%

Heart-disease death, men and women

77% fewerlikely 35% to 92%

Dementia

66% fewerlikely 29% to 84%

Sudden cardiac death

63% fewerlikely 25% to 82%

Stroke

62% fewerlikely 19% to 82%

High blood pressureCould also be no effect

47% fewerlikely 2% to 72%

4 or more sauna sessions a week, compared with 1 or fewer

Respiratory disease

41% fewerlikely 6% to 63%

PneumoniaCould also be no effect

37% fewerlikely 0% to 61%

Death from any cause, compared with men low on both sauna and fitness

Frequent sauna with low fitness

22% fewerlikely 4% to 36%

High fitness with little sauna

37% fewerlikely 26% to 46%

  • Best estimate for frequent sauna users
  • Likely range
  • Fitness, for comparison

How to read it. Each bar shows how many fewer cases the frequent sauna users had than the comparison group. The outlined band is where the true figure likely sits (the 95% confidence interval). When the band reaches back to “no difference”, sauna may have made none. These are shares of the comparison group’s cases, not anyone’s personal odds. A large share of a rare illness is still few people.

Consistent, not independent.

Every bar points the same way, and every one comes from the same study. Nine results from one study are not nine confirmations. The finding needs other countries, other habits, other groups.

Wide bands mean uncertainty.

The bands for high blood pressure and pneumonia reach almost back to no difference. The bar is the headline; the width of the band is the honesty.

Fitness, for comparison.

In the same men, being fit went with even fewer deaths than frequent sauna did. Sauna is no stand-in for exercise, and the people who did both did best.

Show these results as a table
OutcomeComparisonFewer casesLikely rangeRate ratio (95% CI)
Psychotic disorders4–7 vs 1 session a week79% fewer48% to 92%0.21 (0.08–0.52)
Heart-disease death, men and women4–7 vs 1 session a week77% fewer35% to 92%0.23 (0.08–0.65)
Dementia4–7 vs 1 session a week66% fewer29% to 84%0.34 (0.16–0.71)
Sudden cardiac death4–7 vs 1 session a week63% fewer25% to 82%0.37 (0.18–0.75)
Stroke4–7 vs 1 session a week62% fewer19% to 82%0.38 (0.18–0.81)
High blood pressure4–7 vs 1 session a week47% fewer2% to 72%0.53 (0.28–0.98)
Respiratory disease4+ vs 1 or fewer sessions a week41% fewer6% to 63%0.59 (0.37–0.94)
Pneumonia4+ vs 1 or fewer sessions a week37% fewer0% to 61%0.63 (0.39–1.00)
Death from any cause, frequent sauna with low fitnessvs men low on both sauna and fitness22% fewer4% to 36%0.78 (0.64–0.96)
Death from any cause, high fitness with little saunavs men low on both sauna and fitness37% fewer26% to 46%0.63 (0.54–0.74)

All articles

Find a question that interests you. Every summary includes what the study found, its limits and the original source.

56 articles

  • Heart & circulation Systematic review · 2018

    What 40 sauna studies can tell us

    A systematic review brings together 40 studies of regular dry sauna bathing. The picture is promising, varied and still incomplete.

  • Heart & circulation Observational study · 2015

    Sauna and heart health: what a 20-year study found

    A Finnish study followed 2,315 men and found an association between more frequent sauna use and fewer fatal heart events. Here is what that finding can and cannot tell us.

  • Cold water Systematic review · 2012

    Cold water after exercise: soreness, recovery and uncertainty

    A review found some evidence of less muscle soreness after cold-water immersion. Small trials and inconsistent methods leave practical questions open.

  • Brain & mood Observational study · 2017

    Sauna and brain health: a promising link, with limits

    Researchers found a link between sauna habits and later dementia diagnoses in Finnish men. The study raises an interesting question; it does not establish prevention.

  • Cold water Narrative review · 2017

    Cold water immersion: hazard, treatment, or both?

    Cold water can be dangerous and, in some contexts, medically useful. This review from safety researchers weighs both sides and finds the evidence varies by claim.

  • Cold water Randomized trial · 2016

    Cold showers and sick days: 3,000 people, one winter

    In a trial of 3,018 Dutch adults, ending a daily shower with 30 to 90 seconds of cold went with fewer self-reported sick days from work. People were ill just as many days, and everyone knew their group.

How we read research

The finding, the context, the limits. Every summary links to its original source.

These are KUUME’s editorial readings, not a medical review or a promise about a KUUME ritual.

How we score the evidenceOur scales, a worked example, and the full scoring rules

How we score

Two questions for every claim.

We score claims, not single studies. A claim like “sauna means fewer heart deaths” can rest on several studies, and one study can feed several claims. Each claim gets two separate answers. Think of a weather forecast: how sure is it, and how much rain? A forecast can be sure of a drizzle, or unsure of a storm. That’s why we never merge the two into one score.

01

How sure are we that it’s true?

How likely is it that better studies would change the answer?

  1. HighWell established. New research is unlikely to change the picture.
  2. ModerateProbably right. New research could still shift it.
  3. LowCould well change. Treat it as a lead, not a fact.
  4. Very lowA first hint. Very uncertain.
  5. Not testedNot tested yet: an idea without human data.

02

If it’s true, how big is the effect?

How much changed, compared with not doing it, in the people studied. A big effect can still be one we’re unsure of.

  1. LargeLess than half the rate, or a 2-fold change or more.
  2. MediumHalf to three quarters of the rate, or a clear change.
  3. SmallAn effect that showed up, but a modest one.
  4. No effect foundNo clear effect was found. This does not rule out a benefit or harm.
  5. Size not reportedAn effect was reported, but not how big, or nothing has been measured yet.

Effects are relative, in the people studied. “Less than half the rate” is not your personal odds, and not a drop of 50 percentage points.

Same size. Very different certainty.

Heart deaths

Large effectLow certainty

One Finnish study, where people chose their own habits. A link, not proof that sauna caused it.

Some support

Heart rate in the heat

Large effectHigh certainty

Measured directly while people sat in the heat, each compared with themselves. It fits basic physiology.

Well supported

What was measured

A change in a blood marker is not the same as living longer.

  • Illness and death
  • How you feel or perform
  • Body signals

How close to a normal session

Hot baths, infrared therapy, a heating device and an hour in 14°C water are not your Tuesday sauna. We say which it was.

How many separate groups

Eight claims lean on the same Finnish study. Eight results from one study are not eight confirmations, so we count groups of people, not papers.

Worked example: “Regular sauna means fewer deaths from heart disease.”

Every score shows its steps, so anyone can see why a claim sits where it does, and argue with a step rather than with a star rating.

  1. StartStarts at low: an observational study, where people chose their own habitsLow
  2. −1One group of Finnish adults, asked about sauna onceVery low
  3. +1More sauna went with steadily lower rates, and the difference was largeLow
  4. ±0A Japanese hot-bath study saw no link with sudden cardiac death, though it had few such deaths to studyLow
  5. = LowHow sure: low. How big: large effect. Verdict: Some support.Low
The full scoring rules

Where the method comes from

We adapted GRADE, the approach Cochrane and the World Health Organization use to rate evidence. Randomized trials start at high and observational studies start at low. A score goes down for small numbers, a single group of people, habits asked about only once, a different kind of heat or cold, or a range that reaches “no effect”. It goes up when the effect is large, or when more sauna goes with a steadily bigger effect.

What the verdict means

  • Well supported: High or moderate certainty. What we would stand behind today.
  • Some support: Low certainty. A real lead, and could well change with better research.
  • Early signal: Very low certainty. One small study, or a thin margin. A reason to keep looking.
  • Not supported: Studies looked for an effect and did not find one.
  • No evidence yet: An idea without human data. Nothing to score.

How big is big

  • Rates between groups: large under 0.5, medium 0.5 to 0.74, small 0.75 to 0.89.
  • Trial scores (standardized effect size): large 0.8 and up, medium 0.5 to 0.79, small 0.2 to 0.49.
  • Body signals: large at 2-fold or more, medium 1.25- to 2-fold.

Who scores, and when

Scores are an editorial reading by KUUME, not a medical review. Every score shows its steps, links every study and carries the date it was last reviewed. When we add a study, the claims it feeds are scored again, and the change is dated.