| Study | Design & population | Finding & comparison |
|---|---|---|
| 2018 · Laukkanen T ↗ | prospective cohort Middle-aged/older population Participants: 1688. habitual sauna | After the fuller reported adjustment, cardiovascular-mortality HRs versus once-weekly sauna were 0.75 (95% CI 0.52–1.08) for 2–3 sessions and 0.23 (0.08–0.65) for 4–7. Adding frequency improved prediction, but neither prediction nor association establishes causal benefit. Comparison: Once-weekly sauna Prediction is not intervention efficacy |
| 2015 · Laukkanen T ↗ | prospective cohort Population-based sample Participants: 2315. habitual sauna | The adjusted sudden-cardiac-death HR for 4–7 versus 1 weekly session was 0.37 (95% CI 0.18–0.75). Higher frequency was associated with several mortality outcomes. In the duration analysis, longer sessions were associated with fatal cardiovascular endpoints but not all-cause mortality; the abstract does not provide an exact all-cause HR to reuse. Comparison: Once-weekly sauna; duration below 11 minutes for duration analyses Not a comparison with never-users |
| 2018 · Kunutsor SK ↗ | prospective cohort Population-based male cohort Participants: 2277. habitual sauna | High fitness combined with frequent sauna use was associated with lower cardiovascular and overall mortality than the low-fitness/low-frequency category. Comparison: Joint fitness/frequency categories Observational joint analysis |
Original publication: Sauna bathing is associated with reduced cardiovascular mortality and improves risk prediction in men and women: a prospective cohort study.
What questions did the investigators separate?
The paper asked both whether sauna habits were associated with later fatal cardiovascular events and whether information about those habits improved risk prediction. Its original title joins those questions, but they require different interpretations. An association describes how outcomes differed across observed groups. Prediction asks whether knowing an additional characteristic helps distinguish who later experiences an outcome.
That distinction is the main reason to read this report separately from the original male mortality analysis. A characteristic may be useful in a forecast even when deliberately changing it has no established benefit. Conversely, identifying a plausible intervention does not automatically show that it improves an existing forecasting model. The study contributes evidence about observed habits and model performance, with intervention efficacy remaining a further question.
Who was represented in this cohort?
The cohort included middle-aged and older Finnish adults, with women making up approximately half of the sample. This extends the sex representation beyond the earlier male-only mortality report. It does not, however, mean that every subgroup had identical estimates or that the paper established an equally precise association for each sex considered separately.
A useful applicability check asks whether the reader resembles the recruited population, uses a comparable heat exposure and shares the same clinical circumstances. Representation is broader here, while the setting remains a particular sauna culture. The study did not enroll a worldwide population of new users or randomly allocate different sauna technologies. Its mixed-sex design is a meaningful addition without making the findings universal.
What did the exposure comparison actually mean?
Sauna habits were assessed at baseline. The principal frequency estimates compared more frequent bathing with a once-weekly group. The reference therefore consisted of sauna users, rather than people who had never used a sauna. This changes the question from whether starting sauna helps to how different established habits relate to subsequent outcomes.
Frequency and weekly time in sauna were examined as related descriptions of exposure. Their associations cannot be read as independent instructions to increase both. A person can accumulate weekly time in different ways, and observational categories do not select the safest or most effective combination. The abstract does not supply a validated temperature-duration prescription, so this summary does not construct one from the mortality curves.
What did the fuller adjustment show?
In the more extensively adjusted model, the cardiovascular-mortality hazard ratio was 0.75, with a 95% confidence interval of 0.52–1.08, for two to three weekly sessions compared with once-weekly use. For four to seven sessions, the corresponding estimate was 0.23, with an interval of 0.08–0.65. These values belong to that model and reference group.
A hazard ratio compares estimated event rates through follow-up; it is not the percentage of participants who died. The intermediate-frequency interval includes the no-association value. The highest-frequency estimate is lower, with substantial uncertainty around its size. Selecting the smallest estimate from a simpler model or dropping the reference category would give the reader a different and less faithful account.
Why is improved prediction a separate result?
The investigators reported that adding sauna frequency improved their cardiovascular-mortality prediction model. That means the additional information changed measures of model performance in the analyzed data. It does not show that the model should be adopted in every clinic, that it is equally calibrated elsewhere or that changing sauna frequency will alter the predicted outcome.
Imagine a model that uses an existing habit as a clue to a wider pattern of behavior. The clue can help a forecast without being the mechanism that produces the outcome. This is a methodological possibility, rather than a demonstrated explanation for this study. External validation and an intervention study would answer different follow-up questions from the association and prediction analyses reported here.
| Question | What this report assessed | What remains separate |
|---|---|---|
| Association | Habitual sauna and later fatal cardiovascular events | Effect of assigning a new routine |
| Prediction | Adding sauna frequency to a risk model | External clinical validation |
| Personal benefit | Not directly estimated for a new user | An individual absolute intervention benefit |
How does this fit the broader KIHD literature?
The original mortality report and later fitness analyses provide useful context for the questions being asked. Their presence in the bibliography does not establish independent replication for every paper. KIHD publications can share participants while examining different outcomes, exposure definitions or combinations of characteristics.
This report adds mixed-sex representation and a prediction component. The fitness analysis instead investigates joint observed categories. Reading them together helps distinguish what each contributes, but counting them as unrelated populations would overstate the breadth of confirmation. A source map should identify the cohort framework before tallying papers, especially when several impressive-looking estimates appear to reinforce the same mortality narrative.
What can adjustment leave unresolved?
The fuller model considered established cardiovascular risk factors and other measured characteristics, including activity and socioeconomic information. This makes the analysis more informative than a comparison that ignores those differences. Adjustment still depends on how well variables were measured, which variables were available and how relationships were modeled.
Baseline habits also need not describe exposure throughout follow-up. A participant may change routine as health, access or daily life changes. The abstract cannot establish how completely such changes were captured. These limitations define the additional evidence needed before moving from a population association to advice about starting, intensifying or maintaining a personal routine.
What can a reader take into a clinical discussion?
The useful takeaway is a bounded research signal: habitual sauna use carried prognostic information in this cohort. It can support a discussion about evidence, preferences and unanswered questions. It cannot supply an individual absolute benefit, establish clearance for cardiovascular disease or replace other clinical decisions.
The safety review belongs beside the mortality literature because benefits and suitability are different questions. A long-term association among existing users is not a screening test for a person considering heat exposure. This is particularly relevant when a reader is drawn to the large relative estimate but has not yet considered whether the observed population, comparator or exposure resembles their own situation.
What has this editorial review checked?
The indexed abstract and the existing verified source abstraction were read for this summary. The corpus links a free full-text version, but this particular appraisal does not claim to have assessed the complete methods, supplementary analyses or formal risk of bias. Access to a paper and completion of a full-text appraisal are separate states.
Funding and conflict-of-interest details are not recorded in the abstract-level review used here. That means they remain unreported in this summary, rather than assumed absent. A deeper appraisal should inspect the full disclosure statements, model specification, missing-data handling and external validation evidence. Until then, the prediction result and mortality association should retain the limits stated above.
Explore the evidence with a tool.
Questions, considered
Questions about this study
Did the later cardiovascular-mortality sauna cohort include a substantial number of women?
Yes. Women made up approximately half of the participants, unlike several earlier male-only analyses. That improves the population relevance of this particular association, but does not turn it into a randomized intervention. Sex representation, age range and the Finnish sauna context still need to accompany its mortality findings.
Follow the sourceDoes improved cardiovascular risk prediction prove sauna causes lower risk?
No. Adding a variable can improve a model’s ability to distinguish later outcomes even when changing that variable would not cause improvement. This paper examined prediction as well as association. A modest model improvement should not be presented as proof that prescribing more sauna will reduce a person’s cardiovascular mortality.
Follow the sourceWere weekly duration and session frequency both examined in the cardiovascular-mortality paper?
Yes. The investigators analyzed how often people bathed and their weekly time in sauna. These related exposure descriptions were associated with mortality, but were not assigned independently. The paper therefore cannot establish that increasing either number produces a guaranteed benefit or identify a universally optimal combination for individual users.
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 is associated with reduced cardiovascular mortality and improves risk prediction in men and women: a prospective cohort study.
Laukkanen T, Kunutsor SK, Khan H, Willeit P, Zaccardi F, Laukkanen JA
Volume 16, issue 1, pages 219 · PMID 30486813
Open-access flag: yes.Abstract-level editorial review - Association between sauna bathing and fatal cardiovascular and all-cause mortality events.
Laukkanen T, Khan H, Zaccardi F, Laukkanen JA
Volume 175, issue 4, pages 542-548 · PMID 25705824
Open-access flag: not recorded.Abstract-level editorial review - Joint associations of sauna bathing and cardiorespiratory fitness on cardiovascular and all-cause mortality risk: a long-term prospective cohort study.
Kunutsor SK, Khan H, Laukkanen T, Laukkanen JA
Volume 50, issue 2, pages 139-146 · PMID 28972808
Open-access flag: not recorded.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