Sauna: what the research actually says

What does this article cover?

We are not trying to prove sauna works. We read what the research says, including the studies that found nothing and the one that contradicts our own experience.

Every study we refer to is listed at the bottom with a PMID, and every reference in the text takes you straight to the study itself. Where something is reasoning rather than a measurement, the sentence says so.

Is sauna healthy?

Almost everything you read about sauna traces back to one study: 2,315 Finnish men between 42 and 60, followed for a median of 20.7 years ^[1].

What it found is striking. What it also says, and what almost nobody carries forward, is who the comparison is against.

Read that first line again. The comparison group is not "people who do not use a sauna". In this study everybody did. The lowest group went once a week.

That is not a detail. It means every article saying "sauna halves your risk" is describing a study that never had the control group the reader is picturing. What was measured is more often against less often, inside a population where the sauna is so ordinary that nobody skipped it.

The same holds for the rest of this cohort. Dementia: 0.34 at four to seven times against once a week ^[3]. Hypertension: 0.54, over almost 25 years ^[4]. Respiratory disease and pneumonia show the same pattern ^{[5] [6]}, and sauna and fitness appear to reinforce each other ^[7].

Later work extended it to women and found the same association ^[2], and the group pulled it together in a review ^[8].

Are the percentages you see online right?

Search for sauna and the same figures come up everywhere: 40% lower risk of dying from anything, 48% less heart disease at four to seven times a week, somewhere between 21% and 48% for dementia.

Those numbers come from the study above. They are not invented. But they are almost always presented as if they hold against people who do not use a sauna, and that is not what was measured. It is against people who went once a week.

That difference is not small. "Going more often than once a week is associated with less risk than going once a week" is a completely different sentence from "sauna lowers your risk by 40%", and only the first one is in the research.

We went through the two most widely shared sauna protocols on the internet to see what sits underneath them. Nothing does. No PMID, no author, no reference, while percentages are quoted to the decimal point. The figures are traceable, they are simply never attributed, and somewhere along the way they changed meaning.

That is not an attack on anybody. It is why every number in this article links to the study underneath it, so you can check whether we are doing the same thing.

Why does observational research need extra care?

This is all genuinely measured, and it is all observational. That means something specific and worth understanding.

People who spend half an hour in a hot room four times a week are different from people who do it once. They are healthier, they have more time, they are fit enough to tolerate it. Somebody too ill to sit in a sauna lands in the lowest group and then dies of the illness that kept him out.

Researchers adjust for this, and in these studies carefully: age, smoking, BMI, blood pressure, diabetes, cholesterol, previous heart attack, and in some analyses fitness and socioeconomic status too. But adjusting does not remove everything, and nobody claims it does.

Which is why nothing in this article says sauna causes anything. It says sauna is associated with things.

What does randomised research on sauna show?

This is where the evidence thins, and gets more interesting.

There is one well-designed randomised trial putting sauna on top of exercise in sedentary adults with at least one risk factor. Two papers came out of it, and together they tell the whole story.

Both are true. Same participants, same eight weeks, different outcome measure. Anybody quoting only the first is selling something. So is anybody quoting only the second.

There is also a randomised trial in people with coronary artery disease ^[12] and older work in untreated hypertension ^[13].

The systematic review of this whole field is more honest about itself than the summaries cut out of it.

Is an infrared sauna the same as a traditional one?

This is the most important sentence in this article for somebody about to buy something.

Almost every sauna sold in the Netherlands is an infrared cabin. Almost all the research above is traditional Finnish sauna: 80 degrees, hot dry air, water on the stones. They are different things doing different things to your body.

There is infrared research, but it looks nothing like the above. It comes almost entirely from Japan, largely from one research group, and it was done in people with heart failure rather than in healthy people. It is called Waon therapy: 60 degrees of far infrared for 15 minutes, then 30 minutes resting under a blanket.

A trial that misses its primary endpoint and hits its secondary ones is weaker evidence than it looks, because the primary outcome is the one chosen in advance. Earlier work from the same group found effects on prognosis and quality of life ^{[16] [17] [15]}, and one study looked at mild depression ^[18].

We could not find a direct comparison of infrared against traditional sauna. That absence is itself the finding: infrared has trials in sick people, traditional has cohorts in healthy people, and neither tells you much about the other.

Is sauna good for your skin?

This chapter is here because it is uncomfortable.

My own experience: my skin got worse with regular sauna use. That is one person, and it runs against the only study we could find on this.

So what happened to me?

Honest answer: I do not know, and one person cannot answer that question. What I do know is that during that period I was not the person that study was studying. Its participants were healthy volunteers. I was not in a good state at the time.

There is one possible bridge between the two, and it is in the same study: less sebum. Less oil on skin that is already dry is not obviously an improvement. That is a hypothesis, not a finding. Nobody has studied it.

The useful lesson is not "sauna is bad for your skin". It is that heat is a load, and a load lands differently on somebody who is already running on empty. Anyone sleeping badly, eating too little or recovering from something is adding heat to a system that is already struggling.

Is sauna bad for fertility?

This is a small study, ten men. It is also the only adverse effect that the systematic review across forty studies found ^[9].

What that means in practice: if you are trying to conceive, skip the sauna for a while. It is reversible, so that is enough.

On cooling your testicles with an ice pack. This happens in practice: it is in the best-known sauna protocols online, where people go in carrying an ice pack.

Mechanically it makes sense, since heat is the cause. But it has never been studied at the sauna. We found one study protocol for a scrotal cooling patch, and that is an announcement of research rather than a result. Nobody has measured whether it prevents the disruption to sperm production.

If you want to do it anyway: never ice directly on skin. And the simple version remains that skipping the sauna for a while provably works, because the effect fully reverses.

Does sauna help you sleep better?

The evidence here is remarkably thin.

Passive body heating improved sleep in older women with insomnia, in a study from 1996 ^[23]. And the Global Sauna Survey found that sauna users report sleeping better ^[24], which is self-report from sauna users about sauna.

That is all of it. For a healthy adult wanting to know whether a sauna will improve their sleep, there is no answer.

When should you not use a sauna?

Pregnancy. This is not a precautionary note. There is research linking maternal hyperthermia and sauna during pregnancy with cardiovascular and other malformations in the child ^[21]. Talk to your midwife or doctor, not to a blog.

Alcohol. People die in saunas ^[22]. It is rare, and it overwhelmingly involves drink. The combination is the risk factor, not the sauna on its own.

Heart problems, low blood pressure, medication affecting your fluid balance or blood pressure. See your doctor. The randomised trial in coronary artery disease was supervised ^[12], which is exactly why it could be done safely.

If you are already running on empty. See the chapter on skin. That is not medical advice, it is one person's observation.

How often and how long, according to the research?

The same recipe is everywhere online: 80 degrees, 20 minutes, four to seven times a week. That number comes from somewhere, so here is what was actually measured.

In the Finnish cohort the men reported at the start how often and how long they used the sauna ^[1]. The split was:

Duration was split at under 11 minutes, 11 to 19 minutes, and over 19 minutes. Only the last group showed a lower risk.

This is traditional Finnish sauna, in Finland, in the 1980s. Hot dry air around 80 to 100 degrees with water on the stones, not an infrared cabin.

Now the problems, because they come with the number.

It was asked once. The questionnaire was filled in between 1984 and 1989, and these men were then followed for more than twenty years. Nobody checked whether they kept it up. A man who went four times a week in 1986 and stopped in 1994 still counts in the analysis as somebody who goes four times a week.

Nobody measured a temperature. No thermometer was involved. "80 degrees" comes from descriptions of what a Finnish sauna is usually like, not from measurements on participants.

It is self-reported. People estimate this kind of thing badly, usually in their own favour. That sort of noise generally weakens an association rather than strengthening it, which argues in favour of the finding, but it makes the exact number useless as a prescription.

The groups are uneven. There were 201 men in the highest group against 1,513 in the middle one. Every dramatic hazard ratio comes from that smallest group, and small groups give wider confidence intervals and less stable estimates.

And the question nobody asks: why does somebody in Finland go to the sauna seven times a week? Because he has the time, is healthy enough, and probably has one at home. Those are not neutral characteristics.

What you can take from it: a reasonable indication that more often and longer is associated with better outcomes than shorter and rarer. What you cannot: turn it into a protocol and pretend something was tested.

What we honestly do not know

Whether it works for you. Almost all the strong evidence comes from one cohort of middle-aged Finnish men, in a country where the sauna is not a wellness purchase but simply something you do. Whether that transfers to a thirty-year-old Dutch person going twice a week has not been studied.

Whether it causes anything. There is no randomised trial with mortality or hard outcomes as an endpoint, and there is not going to be one.

How it would work. Heat shock proteins get named often, and they are real: Hsp72 measurably rises with heat and exercise acclimation in humans. That is demonstrated acclimation biology, and there is no line running from there to lower mortality. It is an explanation waiting in case the effect is real, not evidence that it is.

What we make of it

Sauna is pleasant, it is safe in healthy people, and in Finnish men it is associated with better outcomes in a way that is too strong and too consistent to wave away.

It is also not a treatment, it has not been proven in randomised research, the infrared version you are most likely to buy has barely been studied in healthy people, and it has one measured downside that disappears when you stop.

If you enjoy it: go. If you are doing it because a graph said you would live longer: read this article again.