In 2017, a Finnish research team published a finding dramatic enough to rewrite the conversation about sauna and brain health: men who used the sauna four to seven times per week had a 66% lower risk of dementia and a 65% lower risk of Alzheimer’s disease compared with those who used it once a week. The study followed 2,315 men for over twenty years, and the numbers, even after adjusting for lifestyle factors, were striking.
But a second, larger Finnish study tells a more complicated story. And that complication is where the real understanding begins.
What the Laukkanen study found
Behind the headline number is Dr Jari Antero Laukkanen, a cardiologist at the University of Eastern Finland who has spent more than a decade studying the health effects of regular sauna use. His team’s work draws on the Kuopio Ischaemic Heart Disease Risk Factor Study, a prospective cohort that began enrolling middle-aged Finnish men in the 1980s.
Published in Age and Ageing, the 2017 dementia paper divided participants into three groups by self-reported sauna frequency: once per week, two to three times, and four to seven times. Over a median follow-up of 20.7 years, the dose-response pattern was consistent. Compared with men who used the sauna once weekly, those in the highest frequency group showed a 66% reduction in dementia risk and a 65% reduction in Alzheimer’s risk specifically. The middle group showed a more modest but still meaningful reduction of around 22%.
These associations held after adjusting for age, alcohol consumption, BMI, smoking, systolic blood pressure, physical activity, and several other confounders. The researchers controlled for baseline health status, so the effect was not simply explained by healthier men being more likely to use the sauna.
Laukkanen himself has been careful not to overstate the finding. “Sauna bathing may protect both the heart and memory to some extent via similar, still poorly known mechanisms,” he said after the paper was published. That phrase — still poorly known mechanisms — is worth holding onto. Even the scientist most responsible for the finding maintains a disciplined uncertainty about the causal chain.
Its power lies in scale and follow-up duration. Its limits are also clear: the cohort was exclusively male, exclusively Finnish, and sauna frequency was measured only at baseline. If a man reported using the sauna four times a week in 1986, the study assumed that frequency continued for two decades. In Finland, that assumption is reasonable. Whether it would hold elsewhere is an open question.
The study almost nobody covers
Three years later, a separate Finnish research group led by Paul Knekt at the Finnish Institute for Health and Welfare published a study in Preventive Medicine Reports that was, in several respects, the larger and more demanding test.
Knekt’s team followed 13,994 men and women — nearly six times Laukkanen’s sample — over a 39-year period, tracking not just sauna frequency but also temperature and duration. The study confirmed that moderate sauna use was associated with reduced dementia risk. Participants who bathed nine to twelve times per month showed a hazard ratio of 0.47, roughly a 53% reduction, directionally consistent with Laukkanen’s findings.
But the data also revealed something the “more is better” narrative cannot accommodate. Participants who reported thirteen to thirty sessions per month showed no statistically significant benefit. Past that threshold, the dose-response curve levelled off.
More striking still was the temperature finding. Sauna sessions conducted at 80–99°C were associated with the strongest protective signal. Sessions above 100°C were associated with an elevated risk of dementia. The difference was not marginal; it inverted the direction of the association entirely.
This finding has been almost entirely absent from popular health coverage. What it suggests is a ceiling at roughly three sessions per week and a temperature sweet spot that falls well below the extremes many enthusiasts pursue. The two studies agree on the broad signal: regular sauna use is associated with reduced dementia risk. They disagree, meaningfully, on how much and how hot.
What the two studies say together
But what neither study can prove deserves stating plainly. Both are observational. Neither randomly assigned people to sauna groups. The participants were Finnish, living in a culture where sauna use is lifelong, social, and woven into daily routine. The relaxation, social connection, and stress reduction that accompany Finnish sauna practice are all independently associated with lower dementia risk, and no statistical adjustment can fully disentangle them from the thermal stimulus itself. As Yuko Hara, then director of ageing and Alzheimer’s prevention at the Alzheimer’s Drug Discovery Foundation, has noted, observational studies of this kind cannot establish causation, and the cultural specificity of the cohorts makes generalisation uncertain.
Nobody knows whether a person who begins sauna use at age fifty, in a commercial facility, will receive the same protection as a Finn who has been bathing since childhood. That question has not been studied. What we can say is that the epidemiological signal is large, consistent in direction, replicated across two independent cohorts totalling more than 16,000 people, and biologically plausible. Which brings us to the mechanisms.
The psychosis finding
Before the molecular case, one more finding from the Laukkanen research programme deserves attention.
In 2018, the same team published a study in Medical Principles and Practice examining the association between sauna frequency and psychosis risk. Using the same Kuopio cohort, they found that men who used the sauna four to seven times per week had a roughly 77% lower risk of developing psychotic disorders compared with those bathing once per week.
The finding is not limited to cognitive decline, but it considerably broadens the neuroprotective picture. Whatever sauna use is doing to the brain, it does not appear limited to a single disease pathway. The associations span dementia, Alzheimer’s specifically, and psychosis.
How heat protects the brain
So the epidemiology establishes a signal. The question is what is happening at the molecular level, and on that, the evidence is more specific than you might expect.
At the centre of the mechanistic case is a family of proteins called heat shock proteins, particularly HSP70. When the body is exposed to sustained heat, cells upregulate the production of these proteins as part of a stress-response pathway that Rhonda Patrick, a biomedical scientist whose review of sauna and healthspan appears in Experimental Gerontology, describes as hormesis: a beneficial adaptation triggered by manageable stress.
HSP70 functions as a molecular chaperone. It helps other proteins fold correctly and flags misfolded proteins for disposal. HSP70’s function is directly relevant to Alzheimer’s disease, which is characterised by the accumulation of two types of misfolded protein: beta-amyloid plaques outside neurons and hyperphosphorylated tau tangles inside them.
HSP70’s connection to Alzheimer’s pathology is not speculative. Campanella and colleagues, writing in the International Journal of Molecular Sciences in 2018, reviewed evidence showing that HSP70 has direct neuroprotective activity against intracellular amyloid aggregation. A 2011 study by Hoshino and colleagues demonstrated that overexpressing HSP70 in transgenic Alzheimer’s mice suppressed disease phenotypes, reducing both pathological markers and cognitive decline.
But the most revealing piece of evidence may be the most recent. In 2022, Guisle and colleagues published a study in Neurobiology of Aging showing that sauna-like heat conditions — applied to both mice and cell cultures — reduced tau phosphorylation. Rather than a generic anti-inflammatory effect, this reduction targets tau hyperphosphorylation, the specific molecular event that drives tangle formation inside neurons. Tangle burden correlates more closely with cognitive decline than amyloid plaque load does in most Alzheimer’s patients. What Guisle’s team found provides a direct, mechanism-level bridge between sitting in a hot room and reduced Alzheimer’s-specific pathology.
Two other mechanisms warrant mention, though neither carries the same specificity. Sauna exposure increases heart rate and cardiac output, driving blood flow to the brain. Chronic cerebral hypoperfusion — reduced blood flow over time — is increasingly recognised as a contributor to neurodegeneration, and Laukkanen’s broader review of sauna health benefits, published in Mayo Clinic Proceedings in 2018, details the cardiovascular pathways involved. Heat stress has also been shown to increase BDNF (brain-derived neurotrophic factor), a protein supporting neuron survival and synaptic plasticity, though evidence linking sauna-induced BDNF increases to long-term cognitive outcomes remains thin.
The HSP70/tau pathway is the most Alzheimer’s-specific mechanism the evidence currently offers.
The contrast therapy hypothesis
There is one more piece to this picture, and it comes from the cold side.
In 2015, Giovanna Mallucci’s laboratory at the University of Cambridge published a study in Nature identifying a cold-shock protein called RBM3 as a mediator of synapse preservation in mouse models of neurodegeneration. When mice were cooled, they upregulated RBM3 production, and this protein protected synapses — the connections between neurons — from being destroyed as disease progressed. Overexpressing RBM3 prevented neuronal loss entirely and prolonged survival in prion-diseased mice, even without the cooling stimulus.
RBM3 is activated by cold in much the same way that HSP70 is activated by heat. They are mirror-image stress responses. And the neuroprotective actions they perform are complementary, not redundant. HSP70 addresses protein misfolding and tau pathology — the accumulation side of neurodegeneration. RBM3’s role in brain health addresses synapse loss — the connectivity side.
In theory, an individual who regularly alternates between sauna and cold water immersion may be activating both pathways. Heat upregulates the molecular chaperone that clears misfolded proteins. Cold upregulates the protein that preserves the synaptic connections those misfolded proteins would otherwise destroy.
This hypothesis has not been tested in humans. The RBM3 findings are preclinical, from mouse models, not human brains. The degree of cold exposure required to upregulate RBM3 in humans is not yet established. These are real limitations.
But the molecular logic is coherent. Two distinct temperature stresses, two distinct protective proteins, two distinct mechanisms of action against two distinct features of neurodegenerative pathology. The idea that deliberate alternation between heat and cold might offer broader neuroprotection than either stimulus alone is not wishful thinking. It is an untested inference from published, peer-reviewed science.
What to do with all of this
For someone reading this with a genuine concern about cognitive decline, the practical question deserves a straight answer.
Regular sauna use at moderate temperatures — 80–99°C, roughly two to four times per week, sustained over years — is associated with reduced dementia risk in two large Finnish cohorts. That frequency and temperature range represents the intersection where both studies agree.
Going hotter does not appear to help and may cause harm. Going more frequently, to daily or near-daily, does not appear to add benefit beyond the moderate range in the Knekt data. Consistency, not intensity, produces the strongest protective signal.
Session duration in the Finnish studies typically ranged from 15 to 20 minutes per visit — long enough to raise core temperature and trigger heat shock protein production. This aligns with the unhurried sauna culture from which the data emerged. A rushed ten minutes in a poorly heated gym sauna is a different stimulus from a relaxed session at 85°C, and it would be a mistake to assume equivalence.
Whether infrared saunas confer the same benefit is unknown. The Finnish studies measured traditional sauna use at 80°C and above; infrared saunas typically operate at lower air temperatures (45–60°C). The core temperature increase may be comparable, but no one has studied the long-term cognitive outcomes specifically.
And the question of starting later in life — whether a 55-year-old who begins sauna use today can expect the same protection as someone who has bathed since childhood — is unanswered. The Finnish cohorts measured lifetime users. What we know about the underlying mechanisms, HSP70 production and cerebral blood flow among them, suggests they respond to present-tense stimulation, not only to historical habit. That is encouraging but it is not proof.
The full shape of the evidence
Two large cohort studies agree on direction and diverge on dose. A set of molecular mechanisms connects heat exposure to Alzheimer’s-specific pathology with a precision that surprises even some researchers. A temperature sweet spot runs counter to the assumption that hotter is better. And a forward-looking molecular argument for combining heat and cold remains genuinely interesting and genuinely unproven.
What this body of evidence suggests is that the brain’s defences against neurodegeneration can be activated by something as ordinary as deliberate heat exposure, repeated consistently, at moderate intensity, over time. How fully, for whom, and in what combination with cold, are questions the science is still answering. The outline of the picture is already large enough to act on. The details are still being filled in.
