Sauna bathing and mortality risk: unraveling the interaction with systolic blood pressure in a cohort of Finnish men

Abstract

Objectives: This cohort study aimed to investigate the potential interplay between systolic blood pressure (SBP), frequency of sauna bathing (FSB), and all-cause mortality risk among Caucasian men. Design: A prospective study was conducted, involving 2575 men aged 42 to 61 years. Baseline assessments included resting blood pressure measurements and self-reported sauna bathing habits. SBP levels were categorized as normal (<140 mmHg) or high (≥140 mmHg), while FSB was classified as low (≤2 sessions/week) or high (3–7 sessions/week). Hazard ratios (HRs) with 95% confidence intervals (CIs) were estimated using Cox regression analysis, while adjusting for lifestyle factors, lipids, inflammation, and comorbidities. Results: Over a median follow-up of 27.8 years, 1,618 deaths were recorded. In the adjusted analysis, individuals with high SBP versus low SBP showed a 29% increased all-cause mortality risk (HR 1.29, 95% CI 1.16–1.43). Similarly, those with low FSB versus high FSB exhibited a 16% elevated mortality risk (HR 1.16, 95% CI 1.02–1.31). When considering combined effects, participants with high SBP-low FSB had a 47% higher mortality risk (HR 1.47, 95% CI 1.24–1.74) compared to those with normal SBP-high FSB. However, no significant association was observed between individuals with high SBP-high FSB and mortality risk (HR 1.24, 95% CI 0.98–1.57). There were potential additive and multiplicative interactions between SBP and sauna bathing concerning mortality risk. Conclusions: This study reveals a potential interplay between SBP, sauna bathing, and mortality risk in Finnish men. Frequent sauna bathing may mitigate the increased mortality risk associated with elevated SBP.

Keywords:

• Sauna
• systolic blood pressure
• mortality
• interaction
• cohort study

Introduction

Among the top 20 risk factors identified in the Global Burden of Disease Study 2019, the five leading contributors to global deaths were high systolic blood pressure (SBP), tobacco use, dietary risks, high fasting plasma glucose, and air pollution [1]. These modifiable factors explain a large proportion of the risk of cardiovascular disease (CVD) [2], which remains the foremost cause of death worldwide [3]. Extensive research has examined the individual associations of these risk factors with adverse vascular outcomes, including CVD mortality [1, 4, 5]. However, studies investigating the combined impact of these risk factors are limited.

Hypertension, commonly defined as SBP ≥140 mmHg and/or diastolic blood pressure (DBP) ≥90 mmHg [6], stands out as the primary preventable risk factor for CVD and premature death [7]. While both elevated systolic and diastolic blood pressure independently predict these adverse outcomes, elevated SBP has consistently emerged as the more critical predictor of cardiovascular events [8, 9]. Consequently, SBP is commonly used in determining cardiovascular risk, with tools such as the American College of Cardiology (ACC)–American Heart Association (AHA) risk estimation tool focusing solely on SBP. The relationship between SBP and mortality from CVD and all causes is robust, independent, and log-linear, showing no evidence of a threshold for SBP down to at least 115 mmHg [10]. Moreover, reducing SBP below recommended targets significantly reduces the risk of CVD and mortality [11].

Sauna bathing, a traditional Finnish practice involving short-term passive exposure to heat for leisure and relaxation, has been associated with various health benefits [12]. Sauna bathing is emerging as a protective vascular risk factor and may extend lifespan [13]. Recent epidemiological studies have shown strong and independent associations between frequent sauna bathing and decreased risk of adverse vascular outcomes [12]. Frequent sauna bathing serves as a robust protective lifestyle factor that may complement the effects of other risk-reducing factors like physical activity and cardiorespiratory fitness (CRF) [14, 15], or even counteract the adverse impact of other risk factors [16, 17]. Both sauna bathing and SBP are modifiable factors with significant public health implications, underscoring the need for a better understanding of their interplay to inform targeted public health interventions. In our recently published population-based cohort study on the interplay between SBP, sauna bathing, and cardiovascular mortality risk in middle-aged and older Finnish men, we showed the existence of an interplay and frequent sauna baths was observed to offset the increased risk of CVD mortality in men with high-normal SBP but not elevated SBP [18]. Given that overall mortality is the oldest, simplest, and most widely used summary measure of population health, encompasses a broader spectrum of health outcomes and provides a more comprehensive understanding of the public health impact of diseases, we decided to extend our previous analysis to this important outcome. We hypothesized that frequent sauna bathing might attenuate the increased mortality risk in adults with elevated SBP. In this context, we aimed to (i) evaluate the joint contributions of SBP and frequency of sauna bathing (FSB) and their interactions in relation to all-cause mortality risk and (ii) assess the independent prospective associations of SBP and FSB with the risk of all-cause mortality.

Materials and methods

Study design and participants

This study adheres to the broad EQUATOR guidelines [19] and and follows the STROBE (STrengthening the Reporting of OBservational studies in Epidemiology) guidelines for reporting observational studies in epidemiology (Supplementary Material 1). The analysis utilizes data from the Kuopio Ischemic Heart Disease (KIHD) prospective cohort study, an ongoing investigation involving a representative sample of men aged 42 to 61 years recruited from the region of Kuopio in eastern Finland. The primary objective of the KIHD study is to assess risk factors for atherosclerotic CVD and related outcomes. Between March 1984 and December 1989, a total of 3433 men were invited to participate in the screening process for potential inclusion in the study. Among them, 3235 men were deemed eligible for participation, and 2682 of these eligible men voluntarily agreed to take part in the study and underwent baseline assessments. For the present analysis, individuals with missing data on the exposures or potential confounding variables (n = 107) were excluded (Supplementary Material 2), resulting in a final cohort of 2575 men with complete information on SBP, FSB, relevant covariates, and all-cause mortality events. There were no losses to follow-up during the course of the study (Supplementary Material 3). The Research Ethics Committee of the University of Eastern Finland granted approval for the study (reference #:143/97), which was conducted according to the Declaration of Helsinki. Additionally, written informed consent was obtained from all participants included in the study.

Assessment of exposures, covariates, and outcomes

The study’s recruitment methods and assessment of risk markers have been previously described [20]. Prior to assessments including blood sample collection between 08:00 and 10:00 AM, participants were instructed to fast overnight, abstain from drinking alcohol for at least 3 days and from smoking. Resting blood pressure measurements were taken using a random-zero sphygmomanometer between 8:00 and 10:00 AM. Participants were asked to rest in a supine position for 5 min before blood pressure was measured three times: once in a standing position and twice in a sitting position in that order, with 5-min intervals between each measurement. The mean of all available measurements was calculated [21]. To evaluate the weekly frequency of sauna sessions, a standardized self-reported questionnaire was utilized. An experienced nurse verified the responses during the baseline examination [22]. Socioeconomic status (SES) was assessed using a self-reported questionnaire that encompassed various indicators, including income, education, occupational prestige, material standard of living, and housing conditions. Each indicator was scored and then summed to create a composite SES index, ranging from 0 to 25, with higher values indicating lower SES [17]. For the purposes of this analysis, all deaths that occurred from the baseline period through to 2018 were included. Mortality events were identified through data linkage to the Finnish national hospital discharge registry and a comprehensive review of death certificate registers.

Statistical analysis

Stata version MP 17 (Stata Corp, College Station, Texas) was utilized for the statistical analyses.

Cox proportional hazards models were employed to estimate multivariable-adjusted hazard ratios (HRs) with 95% confidence intervals (CIs) for all-cause mortality, after confirming no significant departure from the proportionality of hazards assumptions using Schoenfeld residuals [23]. Systolic blood pressure was categorized into two groups: normal (<140 mmHg) and high (≥140 mmHg), consistent with blood pressure clinical guidelines [6]. Additionally, given the well-established linear association between SBP and adverse vascular outcomes, we also modeled SBP as a continuous variable per 10 mmHg increase. Based on our previous studies of the associations of sauna bathing with various adverse outcomes [22, 24], sauna bathing sessions of 3 or more per week have been shown to be consistently and substantially protective of these outcomes; hence, FSB was categorized as low and high (designated as ≤ 2 and 3–7 sauna sessions per week, respectively) as in previous reports [18, 25]. To evaluate the associations of the combinations of SBP and FSB with mortality risk, study participants were classified into the following four groups based on the above-defined cutoffs of SBP and FSB: normal SBP-high FSB (reference comparison); normal SBP-low FSB; high SBP-high FSB; and high SBP-low FSB. Cox proportional hazards models were progressively adjusted for age, body mass index (BMI), smoking status, total cholesterol, high-density lipoprotein cholesterol (HDL-C), prevalent type 2 diabetes (T2D) and coronary heart disease (CHD), use of antihypertensive medication, alcohol consumption, physical activity, SES, high sensitivity C-reactive protein (hsCRP), and mutual adjustment for each exposure. The selection of covariates was based on their established roles as risk factors for mortality, evidence from previous research, previously published associations with mortality in the KIHD study [17, 26], or their potential as confounders based on known associations with mortality outcomes and observed associations with the exposures using the available data [27]. To assess statistical evidence of effect modification on the separate associations of SBP and FSB with mortality risk by clinically relevant characteristics, we performed subgroup analyses using tests of interaction.

Additive and multiplicative two-way interactions were evaluated between SBP and FSB in relation to all-cause mortality risk. Additive interactions were evaluated using the “relative excess risk due to interaction” (RERI), which is estimated for binary variables as RERI_HR =HR₁₁−HR₁₀−HR₀₁+1 [28], where HR₁₁ is the HR of the outcome (i.e. all-cause mortality) if both risk factors (high SBP and low FSB) are present, HR₁₀ is the HR of the outcome if one risk factor is present and the other is absent, with HR₀₁ being vice versa. Though RERI has been demonstrated as the best choice among measures of additivity using a proportional hazards model [28], we also estimated other measures of additive interaction: attributable proportion (AP)=(RERI/HR₁₁) and synergy index (S)= [(HR₁₁ − 1)/(HR₁₀ − 1)+ (HR₀₁ − 1)]. The Stata code -ic- which implements the procedure described in Hosmer and Lemeshow [29], was used to generate the measures of additive interaction, their corresponding 95% CIs and two-tailed tests for no interaction. Multiplicative interactions were evaluated using the ratio of HRs = HR₁₁/(HR₁₀ × HR₀₁) [28]. In the absence of interaction, RERI and AP = 0, S = 1 and the ratio of HRs = 1. As reported in the article by Knol et al. [30] measures of interaction were developed for risk factors rather than preventive factors. Hence, preventive factors should be recoded to risk factors before estimating these measures. Furthermore, recoding of preventive factors should be done such that the stratum with the lowest risk becomes the reference category when two factors are considered jointly (rather than one at a time) [30]. Hence, to evaluate the joint association of SBP and FSB with mortality risk, normal SBP-high FSB was used as the reference comparison. Given a sample of 2575 individuals including 1618 mortality events, we had 100% power allowing for a two-sided significance level of 5%, to detect a clinically important HR of 1.40 for the association between elevated SBP and mortality risk; the power was 96% to detect a clinically important HR of 1.20 for the association between low FSB and mortality risk.

Results

The mean (standard deviation, SD) age and SBP of men at the start of the study was 53 (5) years and 134 (17) mmHg, respectively. The frequency of sauna bathing ranged from 0 to 7 sessions per week, with a median (interquartile range, IQR) of 2 (1–2) sessions per week. The median (IQR) duration of a single sauna session was 10 (8–15) min, and the mean (SD) temperature in the sauna room was 78.9 (9.6)°C (Table 1).

Table 1. Baseline characteristics of study participants (N = 2575).

Characteristics	Overall (N = 2575)	Survived (N = 957)	Died (N = 1618)
	Mean (SD) or median (IQR)	Mean (SD) or median (IQR)	Mean (SD) or median (IQR)
Exposures and related variables
Systolic blood pressure, mmHg	134 (17)	131 (14)	136 (18)
Frequency of sauna bathing (sessions/week)	2 (1–2)	2 (1–3)	2 (1–2)
Duration of sauna bathing (minutes/session)	10 (8–15)	10 (10–15)	10 (7–15)
Temperature of sauna room (°C)	78.9 (9.6)	79.0 (9.2)	78.8 (9.9)
Questionnaire/Prevalent conditions
Age, yr	53 (5)	51 (6)	55 (4)
Alcohol consumption, g/wk	31.8 (6.3–91.5)	27.0 (6.6–76.4)	34.1 (6.0–103.3)
Socioeconomic status	8.49 (4.23)	7.01 (4.02)	9.37 (4.11)
Current smoking, n (%)	814 (31.6)	189 (19.8)	625 (38.6)
History of hypertension, n (%)	811 (30.2)	212 (22.2)	566 (35.0)
History of type 2 diabetes, n (%)	104 (4.0)	15 (1.6)	89 (5.5)
History of coronary heart disease, n (%)	649 (25.2)	146 (15.3)	503 (31.1)
Use of antihypertensive medication, n (%)	600 (22.4)	127 (13.3)	449 (27.8)
Physical measurements
Body mass index, kg/m^2	26.9 (3.6)	26.4 (3.2)	27.2 (3.7)
Diastolic blood pressure, mmHg	89 (11)	88 (10)	89 (11)
Physical activity (KJ/day)	1204 (628–2000)	1217 (679–2000)	1188 (598–2001)
Blood biomarkers
Total cholesterol, mmol/l	5.91 (1.08)	5.77 (1.02)	5.99 (1.11)
HDL-C, mmol/l	1.29 (0.30)	1.30 (0.30)	1.29 (0.31)
Fasting plasma glucose, mmol/l	5.36 (1.27)	5.14 (0.69)	5.48 (1.49)
High sensitivity C-reactive protein, mg/l	1.29 (0.71–2.48)	1.00 (0.59–1.87)	1.54 (0.80–2.90)

HDL-C, high-density lipoprotein cholesterol; IQR, interquartile range; SD, standard deviation.

Men who frequently used the sauna and had normal SBP were characterized by being younger and having higher SES. They spent more time in the sauna during each session, smoked less, had a lower prevalence of comorbidities (hypertension, T2D, and CHD), had higher levels of HDL-C, and lower levels of fasting plasma glucose and hsCRP (Supplementary Material 4). Among the 2,575 men included in this cohort, a total of 1618 deaths occurred during the median (IQR) follow-up period of 27.8 (18.4–31.1) years. The annual mortality rate was 26.0 per 1000 person-years at risk, with a 95% CI of 24.8–27.3.

Compared with men with normal SBP, those with high SBP had an increased risk of all-cause mortality in analysis adjusted for age 1.40 (95% CI 1.27–1.55) (Figure 1 Model 1), which was attenuated to 1.29 (95% CI 1.16–1.43) on further adjustment for BMI, smoking status, total cholesterol, HDL-C, prevalent T2D and CHD, use of antihypertensive medication, alcohol consumption, physical activity, SES, and hsCRP (Figure 1 Model 2). There was evidence of an association when SBP was modeled as a continuous variable. Following adjustment for the covariates in Model 2, men with low FSB had an increased risk of all-cause mortality compared with those with high FSB 1.16 (95% CI 1.02–1.31) (Figure 1 Model 2). The HRs were unchanged when SBP was adjusted for FSB and FSB was adjusted for SBP (Figure 1 Model 3). The associations did not significantly differ according to categories of relevant clinical characteristics (p-values for interaction for all >0.05) (Figure 2).

Figure 1. Separate and combined associations of systolic blood pressure and frequency of sauna bathing with risk of all-cause mortality using a systolic blood pressure cutoff of ≥140 mmHg. CI, confidence interval; FSB, frequency of sauna bathing; HR, hazard ratio; ref, reference; SBP, systolic blood pressure. Model 1: Adjusted for age. Model 2: Model 1 plus body mass index, total cholesterol, high-density lipoprotein cholesterol, smoking status, history of type 2 diabetes, history of coronary heart disease, use of antihypertensive medication, alcohol consumption, physical activity, socioeconomic status, and high sensitivity C-reactive protein. Model 3: Model 2 plus FSB for SBP or SBP for FSB.

Figure 2. Separate associations of systolic blood pressure and frequency of sauna bathing with risk of all-cause mortality stratified by relevant characteristics. (A) SBP and all-cause mortality (B) FSB and all-cause mortality. CI, confidence interval; FSB, frequency of sauna bathing; HR, hazard ratio; ref, reference; SBP, systolic blood pressure. Analyses were adjusted for age, body mass index, total cholesterol, high-density lipoprotein cholesterol, smoking status, history of type 2 diabetes, history of coronary heart disease, use of antihypertensive medication, alcohol consumption, physical activity, socioeconomic status, and high sensitivity C-reactive protein.

In evaluation of joint associations of SBP and FSB with the risk of all-cause mortality, cumulative hazard curves showed an increased risk of all-cause mortality among participants with high SBP-low FSB compared with other groups (p-value for log-rank test < .001; Supplementary Material 5). Compared with men with normal SBP-high FSB, those with high SBP-low FSB had an increased risk of all-cause mortality 1.47 (95% CI 1.24–1.74) in multivariable analysis, with no evidence of an association between men with high SBP-high FSB and all-cause mortality risk 1.24 (95% CI 0.98–1.57) (Figure 1 Model 2; Supplementary Material 6). Interaction analysis showed the RERI = 0.09 (95% CI −0.21, 0.40; p = .56), AP = 0.06 (95% CI −0.15, 0.27; p = .56), S = 1.24 (95% CI 0.54, 2.84; p = .61), and the ratio of HRs = 1.04 (95% CI 0.77, 1.31; p = .26), suggesting there might be positive interactions between SBP and FSB in relation to all-cause mortality risk on both the additive and multiplicative scales. The results were qualitatively similar in a sensitivity analysis that excluded men with prevalent CHD (n = 649) or stroke (n = 58) at baseline (Supplementary Materials 7 and 8). Given that the American College of Cardiology/American Heart Association guidelines defines hypertension using a SBP cutoff of ≥130 mmHg [31], we conducted a subsidiary analysis in which SBP was categorized as low and high (<130 and ≥130 mmHg, respectively) and we re-evaluated the joint associations of SBP and FSB with all-cause mortality risk. The results were consistent with the main findings (Table 2).

Table 2. Separate and joint associations of systolic blood pressure and frequency of sauna bathing with risk of all-cause mortality using a systolic blood pressure cutoff of 130 mmHg.

Exposure	Events/total	Model 1		Model 2		Model 3
		HR (95% CI)	p-value	HR (95% CI)	p-value	HR (95% CI)	p-value
SBP, mmHg
Normal SBP (<130)	622 / 1117	ref		ref		ref
High SBP (≥130)	996 / 1458	1.29 (1.17–1.43)	<.001	1.22 (1.10–1.35)	<.001	1.22 (1.10–1.36)	<.001
FSB, sessions/week
High FSB (3–7)	311 / 560	ref		ref		ref
Low FSB (≤2)	1307 / 2015	1.23 (1.09–1.9)	.001	1.16 (1.02–1.31)	.023	1.15 (1.02–1.31)	.026
SBP-FSB combination
Normal SBP-High FSB	113 / 242	ref		ref	ref	NA
Normal SBP-Low FSB	509 / 875	1.24 (1.01–1.52)	.039	1.08 (0.88–1.33)	.45	NA
High SBP- High FSB	198 / 318	1.30 (1.03–1.64)	.026	1.11 (0.88–1.41)	.38	NA
High SBP-Low FSB	798 / 1140	1.61 (1.32–1.96)	<.001	1.36 (1.11–1.66)	.003	NA

CI, confidence interval; FSB, frequency of sauna bathing; HR, hazard ratio; ref, reference; NA, not applicable; SBP, systolic blood pressure.

Model 1: Adjusted for age.

Model 2: Model 1 plus body mass index, total cholesterol, high-density lipoprotein cholesterol, smoking status, history of type 2 diabetes, history of coronary heart disease, use of antihypertensive medication, alcohol consumption, physical activity, socioeconomic status, and high sensitivity C-reactive protein.

Model 3: Model 2 plus FSB for SBP or SBP for FSB.

Discussion

Our analysis of a cohort of Finnish men revealed independent associations between SBP and FSB with all-cause mortality risk, consistent with previous research findings [9, 20]. These associations were not significantly modified by relevant clinical characteristics including age, BMI, smoking status, use of antihypertensive medication, and comorbidities such as T2D and CHD. New findings indicate the presence of an interplay between SBP, sauna bathing, and all-cause mortality risk. Men with elevated SBP who were not frequent sauna users exhibited an increased risk of all-cause mortality compared to men with normal SBP who were frequent sauna bathers. However, the adverse risk associated with elevated SBP was offset in men who had elevated SBP but were also frequent sauna bathers. This protective effect of sauna bathing persisted when SBP was categorized using a threshold of ≥130 versus <130 mmHg, suggesting that frequent sauna bathing may ameliorate the adverse effects of high-normal SBP. Furthermore, interaction analysis revealed that the combined effect of high SBP and low FSB on all-cause mortality risk may exceed the additive or multiplicative sum of their individual associations. Nonetheless, the estimates for this interaction were relatively modest, possibly due to low statistical power as a result of the several combined categories.

Given that CVDs are the leading cause of death globally, the pathophysiological mechanisms underlying the associations of SBP and sauna bathing with the risk of CVD may also be relevant to overall mortality. The link between elevated blood pressure and increased cardiovascular risk is well-established and considered causal [10]. High blood pressure can lead to various short-term and long-term consequences, including stroke, CHD, heart failure, atrial fibrillation, valvular heart disease, chronic kidney disease, T2D, and dementias, all contributing to the elevated risk of death [32]. Passive heat therapies, such as sauna bathing, have been reported to elicit physiological responses and adaptations similar to those observed with exercise training or physical activity [33]. Frequent exposure to sauna may consequently reduce the risk of CVD and, by extension, the risk of mortality, through various mechanisms. These mechanisms include improvements in levels of cardiovascular risk factors such as blood pressure, lipids, blood glucose, and insulin [12]. Additionally, sauna bathing has been associated with improvements in endothelial function, arterial stiffness, vascular compliance, cutaneous microvascular function, and intima media thickness [12]. Moreover, sauna bathing has been linked to a reduction in reactive oxygen species and systemic inflammation, an enhancement of overall cardiorespiratory function, and a boost to the immune system [12, 34].

The potential ability of frequent sauna baths to enhance the beneficial effects of other protective risk factors [14, 15, 35] and neutralize the adverse effects of cardiovascular risk factors [16, 36], suggests that the benefits of frequent sauna bathing may be potent enough to mitigate the adverse cardiovascular effects of high and high-normal SBP, including the risk of death. These findings hold public health importance and have substantial clinical implications. Clinical trials have consistently shown that lowering blood pressure with common antihypertensive regimens reduces the risk of CVD and all-cause mortality, with larger reductions in blood pressure leading to proportional risk reductions [11, 37]. Lifestyle modifications, such as regular physical activity and exercise training, also play a critical role in managing high blood pressure [31, 38]. Regular physical activity has been found to reduce blood pressure, prevent the development of hypertension in individuals with normal blood pressure and prehypertension, and attenuate the progression of CVD in those with hypertension [39, 40]. Sauna exposure has been shown to produce blood pressure reductions of up to 7 mmHg [41–44], similar to the effects observed following aerobic exercise training [40], and these reductions have significant clinical relevance. However, most of these studies only evaluated the short-term/acute effects of sauna exposure on blood pressure [41–45]. A 2 mmHg reduction in SBP could reduce mortality from stroke and vascular causes by 10 and 7%, respectively [10]. It has also been reported that blood pressure reductions of 5–7 mmHg among individuals with hypertension translate to a 20–30% reduced risk of CVD [39]. It is uncertain how long the beneficial effects of sauna on blood pressure lasts. However, it is documented that some effects of physical activity are short-lived, such as reductions in blood pressure and lipids [46, 47]; every sustained bout of aerobic physical activity has acute physiological effects which may last up to 24 h [48], therefore, these beneficial effects need to be sustained by chronic regular physical activity [47]. This is likely to apply to sauna bathing given their similar mechanistic pathways. Indeed, it has been demonstrated that life long sauna use reduces the risk of hypertension [49]. There is evidence showing that the combination of frequent sauna baths with habitual physical activity may offer substantial benefits in blood pressure reduction, as well as the risk of CVD and all-cause mortality, than the use of each modality alone [50]. However, for individuals unable to engage in exercise due to physical limitations, regular sauna use alone may still offer benefits in blood pressure reduction. Indeed, the use of sauna alone has been shown to elicit similar haemodynamic responses such as blood pressure reduction when compared with a combination of exercise and sauna [44].

It has often been argued anecdotally that because sauna bathing is a tradition and more accessible to people in Nordic countries, its public health impact may be limited in other populations. For instance, in Finland, there are about 3 million saunas in Finland for a population of about 5.5 million [51]. Though there has been a steep increase in research outputs on the health benefits of saunas over the last decade with an accompanying increase in the use of saunas as a lifestyle activity globally [52], there are still challenges in accessing sauna facilities and promoting its uptake. In most parts of the world, facilities are often located in gyms, which may require paid memberships and travel, making them less affordable for some individuals. Policymakers should consider investing in services to increase sauna accessibility and address any inequalities that might hinder its uptake, similar to efforts made to promote physical activity. Encouragingly, sauna bathing is generally well-tolerated, enjoyable, and involves little physical exertion, making it an attractive option for many individuals, and its safety profile is considered favorable [12]. As research on the health benefits of sauna continues to emerge, promoting its use among the wider population may offer valuable contributions to cardiovascular health and overall well-being.

Strengths and limitations

The current study has several strengths. The study employed a population-based prospective cohort design with a relatively large sample of middle-aged and older men. Additionally, the study benefits from an extended follow-up duration, allowing for a comprehensive assessment of long-term mortality outcomes. Robust analytical methods were employed, including the adjustment for a comprehensive panel of potential confounders, subgroup analyses, exploration of additive and multiplicative interactions between SBP and sauna bathing in relation to all-cause mortality, and sensitivity analysis to assess the robustness of the results. However, several limitations should be considered while interpreting the study’s findings. A significant limitation is the use of baseline data for the exposures and confounders during the lengthy follow-up period. This approach may fail to account for potential changes in lifestyle, medication usage, chronic diseases, and aging, which can lead to underestimation of the true strength of associations due to regression dilution bias. Additionally, changes in sauna bathing habits may have occurred during follow-up due to aging and disease, although it has been reported that sauna habits tend to remain fairly stable in the Finnish population. Reproducibility studies yielded encouraging results [53], indicating a moderate stability of sauna bathing habits over several years. Another limitation arises from the potential for type 1 errors due to multiplicity [54], given the testing of multiple variables for associations with exposures and mortality outcomes within the KIHD study. Furthermore, the use of random-zero sphygmomanometers for blood pressure measurements during baseline examinations in the 1980s may have led to underestimations compared to the standard mercury sphygmomanometer [55]. As with all observational studies, this investigation is susceptible to biases such as reverse causation, and it cannot establish causation directly. Although the study adjusted for numerous potential confounders to enhance the validity of the associations, there remains a possibility of residual confounding due to measurement errors and relevant unmeasured confounders such as comorbidities and medication usage. Lastly, the generalizability of the findings is limited to middle-aged and older Finnish men, and may not be applicable to other populations. Given these limitations, the results need to be interpreted with caution.

Conclusions

Systolic blood pressure and FSB are each associated with the risk of all-cause mortality, independently of several established risk factors in Finnish men. There may be interactive effects between SBP and sauna bathing in relation to mortality risk. The most notable finding is that frequent sauna bathing appears to counteract the adverse effects of elevated blood pressure on mortality risk. This observation holds potential clinical implications, as it highlights the potential benefits of sauna bathing as a lifestyle intervention to improve cardiovascular health and longevity, especially for individuals with elevated blood pressure.

Authors’ contributions

Setor K. Kunutsor: Conceptualization, data collection, data analysis, data interpretation, writing the original draft, reviewing, editing, and finalization. Sae Young Jae: Conceptualization, data interpretation, reviewing, editing, and finalization. Sudhir Kurl: Funding acquisition, conceptualization, data interpretation, reviewing, editing, and finalization. Jari A. Laukkanen: Funding acquisition, conceptualization, data interpretation, reviewing, editing, and finalization.

Acknowledgements

We thank the staff of the Kuopio Research Institute of Exercise Medicine and the Research Institute of Public Health and University of Eastern Finland, Kuopio, Finland for the data collection in the study.

Disclosure statement

No potential conflict of interest was reported by the authors.

Additional information

Funding

JAL is supported by grants from the Finnish Foundation for Cardiovascular Research, Helsinki, Finland. SKK is supported by the National Institute for Health and Care Research (NIHR) Applied Research Collaboration East Midlands (ARC EM) and Leicester NIHR Biomedical Research Centre (BRC). The views expressed are those of the author and not necessarily those of the NIHR or the Department of Health and Social Care. The funders had no role in study design, data collection and analysis, decision to publish or preparation of the manuscript.