| Study | Design & population | Finding & comparison |
|---|---|---|
| 2017 · Zaccardi F ↗ | prospective cohort Without hypertension at baseline Participants: 1621. habitual sauna | After the fuller reported adjustment, incident-hypertension HRs versus once-weekly sauna were 0.83 (95% CI 0.59–1.18) for 2–3 sessions and 0.53 (0.28–0.98) for 4–7. The middle-frequency estimate was compatible with no association. Comparison: Once-weekly sauna Not a randomized prevention intervention |
| 2018 · Laukkanen T ↗ | Single-session experimental before-after study At least one cardiovascular risk factor Participants: 102. Sauna | Blood pressure and pulse-wave velocity fell after exposure, but the study did not establish durable changes or fewer cardiovascular events. Comparison: Before, immediately after and recovery No unheated comparator described |
| 2022 · Lee E ↗ | randomized controlled trial Low activity with at least one cardiovascular risk factor Participants: 47. postexercise sauna · 8 weeks | Relative to exercise alone, exercise plus sauna produced an additional fitness change of +2.7 mL/kg/min (95% CI +0.2 to +5.3) and a systolic-pressure difference of −8.0 mmHg (−14.6 to −1.4). This small combined-program trial did not test sauna alone or clinical-event prevention. Comparison: Exercise alone; control Small eight-week multi-arm trial |
Original publication: Sauna Bathing and Incident Hypertension: A Prospective Cohort Study.
The question begins before hypertension
The question was whether existing sauna habits predicted later development of hypertension. That starting point is the defining feature of this paper. Participants did not enter as a group receiving a new sauna intervention for elevated pressure. Instead, researchers identified men without hypertension and examined what happened after their habitual exposure had been recorded.
The publication is titled Sauna Bathing and Incident Hypertension: A Prospective Cohort Study. Incident means newly occurring during observation. Keeping that word visible prevents a common switch in meaning: evidence about developing a diagnosis becomes evidence about lowering an already high reading. The latter requires a different population, a different comparison and attention to ongoing clinical care. The useful contribution here is a prospective association with a defined new outcome, rather than a general verdict on every blood-pressure question.
Who was compared with whom
The study followed men in the Finnish KIHD cohort who were free of hypertension at baseline. Frequency came from questionnaires about existing sauna habits. The principal reference was once-weekly sauna, so the comparison concerns people already using sauna at different frequencies. It should not be summarized as sauna versus never using sauna.
This distinction matters when considering a reader who does not currently bathe. Starting a new activity changes more than membership in a statistical category: adherence, tolerability and reasons for starting also matter. The cohort did not allocate that change. Nor can findings in this male Finnish sample directly establish the same relationship for women, other sauna technologies or people already receiving hypertension care. Those are questions of transferability rather than reasons to discard the original association.
Read the fuller model rather than the largest percentage
The fuller reported adjustment model estimated a hypertension hazard ratio of 0.83, with a 95% confidence interval of 0.59–1.18, for 2–3 weekly sessions compared with one. For 4–7 sessions, the corresponding estimate was 0.53, with an interval of 0.28–0.98. These are the model-specific values retained in the site’s fact registry.
A hazard ratio describes the relative rate of a recorded outcome over follow-up in the compared groups. It is not an individual probability. The middle-frequency interval crosses the no-association value, so its estimate does not support a precise claim of protection. The higher-frequency interval also remains broad. Choosing a simpler adjustment model because it yields a more attractive headline would hide information about how the result responds when measured clinical and lifestyle differences are considered.
Adjustment narrows explanations without assigning exposure
The investigators reported several adjustment stages, and the estimates changed between them. That is informative: measured differences between frequency groups can affect the apparent relationship. A careful reader should ask which model a summary is using before interpreting a percentage or comparing it with another paper.
Even a detailed model cannot guarantee that groups were otherwise exchangeable. Sauna habits may accompany other routines, resources or health characteristics that were measured imperfectly or not included. Conversely, including additional variables is not an automatic solution to every bias; which variables belong in a causal model also requires justification. This abstract-level reading does not adjudicate the complete statistical model. It identifies the inferential limit: adjusted observation is stronger than an unqualified anecdote, but it remains different from randomly assigning a prevention strategy.
An immediate pressure response answers another question
An acute experimental report found lower blood pressure and pulse-wave velocity after a sauna session. That provides short-term physiological context, but an immediate measurement is not the same outcome as a new hypertension diagnosis years later. The acute report also lacked an unheated comparator in its abstract, so its before-and-after change should not be treated as an isolated assigned heat effect.
The cohort and the experiment can therefore sit beside each other without being counted as direct replications. One asks about later diagnosis among people with different habits; the other asks how measurements change around a particular exposure. A proposed pathway connecting them would need testing across the intervening steps, including persistence of any change outside the session and whether that persistence affects clinically relevant outcomes.
What the exercise combination trial adds
A separate randomized study compared an exercise-plus-sauna program with exercise alone and a control condition. Its favorable systolic-pressure comparison is relevant because it tested an assigned program rather than observing a pre-existing sauna habit. However, it tested sauna added to exercise, not sauna in isolation, and its outcomes were risk markers rather than incident hypertension.
The comparison does not turn the cohort frequency groups into an evidence-based weekly prescription. A combined intervention can change fitness, routines and exposure together. Its value is to make a more specific causal question testable: under the studied conditions, did adding postexercise sauna alter outcomes beyond exercise alone? Distinguishing that question from prevention in initially nonhypertensive men keeps the evidence useful without borrowing certainty from a different design.
What further work would change confidence
A stronger prevention case would require evidence that follows assigned exposure long enough to distinguish a transient response from a sustained difference in new hypertension. Independent populations would also help determine whether the Finnish male cohort’s association depends on its social setting, established bathing practices or participant selection.
Repeated measures of habits and pressure would make changes during follow-up easier to interpret. Reporting absolute event risks alongside relative estimates would improve communication, but this summary does not calculate personal absolute benefit from a hazard ratio. The present grade reflects a worthwhile cohort signal with important uncertainty. It is not a recommendation to increase exposure until a risk estimate is reached, and it does not support adjusting medication or delaying evaluation of high blood pressure.
Access, safety and review boundaries
This appraisal uses the indexed abstract and the existing verified fact record. The original publication can be followed through its PubMed record and DOI. The complete statistical protocol, handling of missing data, follow-up ascertainment details and funding or conflicts of interest have not been assessed from full text here. An unreported funding field should not be read as evidence that no funding or competing interests existed.
A separate narrative safety review names unstable angina, recent myocardial infarction and severe aortic stenosis as contraindications and warns about alcohol-related harms. That historical source supplies specific cautions, not comprehensive clearance for someone without those labels. For a personal decision, keep the cohort’s prevention question separate from an individual’s current blood-pressure status, medications and clinical advice.
Explore the evidence with a tool.
Questions, considered
Questions about this study
Did the KIHD hypertension study enroll people who already had hypertension?
No. This analysis began with men without hypertension and examined new cases during follow-up. It therefore addresses an association with developing hypertension, rather than whether sauna lowers pressure in people already receiving treatment. It cannot answer medication-adjustment questions or establish a safe routine for uncontrolled high blood pressure.
Follow the sourceWhy do different hypertension risk estimates appear for the same sauna study?
The paper reports models with different sets of adjustments. Estimates changed when additional clinical and lifestyle factors were included, particularly for the middle-frequency category. A summary should name the model used and preserve its uncertainty rather than selecting whichever percentage reduction sounds most impressive to readers.
Follow the sourceDoes the hypertension cohort show that more sauna always means less risk?
It reports an inverse association across frequency categories, but existing users were not randomly assigned to their habits. The middle-frequency estimate was also imprecise after fuller adjustment. Those features prevent the results from establishing an unlimited dose response or a personal prescription for preventing hypertension through sauna use.
Follow the sourceFollow the evidence
Sources & references
Research summaries in this edition are based on indexed abstracts unless otherwise noted. A source link is not a claim of medical review.
- Sauna Bathing and Incident Hypertension: A Prospective Cohort Study.
Zaccardi F, Laukkanen T, Willeit P, Kunutsor SK, Kauhanen J, Laukkanen JA
Volume 30, issue 11, pages 1120-1125 · PMID 28633297
Open-access flag: not recorded.Abstract-level editorial review - Acute effects of sauna bathing on cardiovascular function.
Laukkanen T, Kunutsor SK, Zaccardi F, Lee E, Willeit P, Khan H, Laukkanen JA
Volume 32, issue 2, pages 129-138 · PMID 29269746
Open-access flag: not recorded.Abstract-level editorial review - Effects of regular sauna bathing in conjunction with exercise on cardiovascular function: a multi-arm, randomized controlled trial.
Lee E, Kolunsarka I, Kostensalo J, Ahtiainen JP, Haapala EA, Willeit P, Kunutsor SK, Laukkanen JA
Volume 323, issue 3, pages R289-R299 · PMID 35785965
Open-access flag: yes.Abstract-level editorial review - Benefits and risks of sauna bathing.
Hannuksela ML, Ellahham S
Volume 110, issue 2, pages 118-126 · PMID 11165553
Open-access flag: not recorded.Abstract-level editorial review