Cross‐country skiing is associated with lower all‐cause mortality: A population‐based follow‐up study

Abstract

Objectives:
The prospective relationship between leisure time cross-country skiing and any fatal events is uncertain. We aimed to assess the associations of leisure time cross-country skiing habits with the risk of all-cause mortality in a general population.

Methods:
A 12-month physical activity questionnaire was used at baseline to assess the frequency, average duration, and intensity of cross-country skiing in a prospective population-based cohort of 2,087 middle-aged men from Eastern Finland. Hazard ratios (HRs) (95% confidence intervals) were calculated for all-cause mortality.

Results:
During a median (interquartile range) follow-up of 26.1 (18.7-28.0) years, 1,028 all-cause mortality outcomes were recorded. In analyses adjusted for several established risk factors and other potential confounders, when compared to men who did not do any cross-country skiing, the HRs (95% CIs) of all-cause mortality were 0.84 (0.73 to 0.97) and 0.80 (0.67 to 0.96) for men who did 1-200 and > 200 MET hours per year of cross-country skiing, respectively. Similarly, compared to men who did not do any cross-country skiing, the corresponding adjusted HRs (95% CIs) for all-cause mortality were 0.84 (0.72 to 0.97) and 0.82 (0.69 to 0.97) for men who did 1-60 mins per week and > 60 mins week of cross-country skiing respectively. The associations were similar across several subgroups, except for evidence of effect modification by body mass index and history of diabetes.

Conclusions:
Total volume as well as duration of leisure time cross-country skiing are each inversely and independently associated with all-cause mortality in a middle-aged Caucasian male population. This article is protected by copyright. All rights reserved.

Full text

Scand J Med Sci Sports. 2017;1–9. wileyonlinelibrary.com/journal/sms
|
1
© 2017 John Wiley & Sons A/S.
Published by John Wiley & Sons Ltd
Accepted: 12 September 2017
DOI: 10.1111/sms.12980
ORIGINAL ARTICLE
Cross- country skiing is associated with lower all- cause mortality:
A population- based follow- up study
J. A. Laukkanen1,2
|
T. Laukkanen1
|
S. K. Kunutsor3
1Institute of Public Health and Clinical
Nutrition,University of Eastern Finland,
Kuopio, Finland
2Central Finland Hospital
District,Department of Medicine,
Jyväskylä, Finland
3Translational Health Sciences, Bristol
Medical School,University of Bristol,
Bristol, UK
Correspondence
Jari A. Laukkanen, Institute of Public
Health and Clinical Nutrition, University of
Eastern Finland, Kuopio, Finland.
Email: jariantero.laukkanen@uef.fi
Funding information
The Finnish Foundation for Cardiovascular
Research, Helsinki, Finland, supported
the Kuopio Ischemic Heart Disease Study.
These sources had no role in design
and conduct of the study; collection,
management, analysis, and interpretation
of the data; and preparation, review, or
approval of the manuscript.
The prospective relationship between leisure- time cross- country skiing and any fatal
events is uncertain. We aimed to assess the associations of leisure- time cross- country
skiing habits with the risk of all- cause mortality in a general population. A 12- month
physical activity questionnaire was used at baseline to assess the frequency, average
duration, and intensity of cross- country skiing in a prospective population- based co-
hort of 2087 middle- aged men from eastern Finland. Hazard ratios (HRs; 95% confi-
dence intervals) were calculated for all- cause mortality. During a median (interquartile
range) follow- up of 26.1 (18.7- 28.0) years, 1028 all- cause mortality outcomes were
recorded. In analyses adjusted for several established risk factors and other potential
confounders, when compared to men who did not do any cross- country skiing, the
HRs (95% CIs) of all- cause mortality were 0.84 (0.73- 0.97) and 0.80 (0.67- 0.96) for
men who did 1- 200 and >200 metabolic equivalent- hours per year of cross- country
skiing, respectively. Similarly, compared to men who did not do any cross- country
skiing, the corresponding adjusted HRs (95% CIs) for all- cause mortality were 0.84
(0.72- 0.97) and 0.82 (0.69- 0.97) for men who did 1- 60 min/wk and >60 min/wk of
cross- country skiing, respectively. The associations were similar across several sub-
groups, except for evidence of effect modification by body mass index and history of
diabetes. Total volume as well as duration of leisure- time cross- country skiing is
each inversely and independently associated with all- cause mortality in a middle-
aged Caucasian male population.
KEYWORDS
all-cause mortality, cross-country skiing, physical activity
1
|
INTRODUCTION
Many studies have indicated that higher- intensity exercise
(with shorter duration) may yield more favorable effects on
mortality and cardiovascular disease (CVD) risk than lower-
intensity exercise.1-3 These previous studies consistently found
a greater reduction in the risk of CVD with vigorous (typi-
cally ≥6 metabolic equivalent, METs) than with moderate-
intensity physical activity (PA) and reported more favorable
risk profiles for individuals engaged in vigorous, as opposed
to moderate- to light- intensity PA.4 Brisk walking as a PA has
many advantages as it is easy to perform. The most important
advantage of walking is its high safety factor that can be sus-
tained for months or years, although it may not be so effective
in reducing CVD risk.5 The aerobic exercise training section
of recent PA guidelines2 recommends 30 min/d of moderate-
intensity exercise or 75 min/wk of vigorous- intensity exer-
cise, with emphasis on low- to moderate- intensity PA due
to better adherence and safety issues at the population level
compared to higher- intensity PA. Although high- intensity
PA with shorter duration could be potentially protective, data
Disclaimers: The views expressed in the submitted article are author’s own
and not an official position of the institution or funder.

2
|
LAUKKANEN EtAL.
are very limited on the associations between high- intensity
physical activities such as cross- country skiing and all- cause
mortality outcome events.
A previous study found that running, even less than the
recommended amount of time or at slower speeds, was as-
sociated with significant benefits.6 Individuals who were
inactive had an increased risk of mortality compared with
individuals in the low- volume PA group; even 15 minute
a day or 90 minutes a week of moderate- intensity exercise
might be of benefit, even for individuals at risk of CVD.7
The benefits of running were superior to walking not only
at the same volume but also at lower volumes than walk-
ing. Compared with moderate- intensity activity, vigorous-
intensity activity, such as running, may be a better option for
time efficiency, producing similar, if not greater mortality
benefits for 5- 10 min/d in many healthy but sedentary in-
dividuals who may find 15- 20 min/d of moderate- intensity
activity time-consuming. There is convincing observational
evidence on the benefits of higher- intensity physical exercise
in preventing chronic diseases, including coronary heart dis-
ease (CHD), stroke, hypertension, diabetes, and hypercho-
lesterolemia8,9 and reducing the levels of cardiovascular risk
factors.10,11 Exceeding intensity levels of PA suggested in the
guidelines may produce even more important CHD and mor-
tality risk reductions.9,12 Indeed, long- term endurance cross-
country skiing has been shown to be associated with reduced
risk of all- cause mortality.13,14
Leisure- time cross- country skiing, which is a high-
intensity type conditioning PA and undertaken during the
winter season, may have potential health benefits; however,
the prospective associations of cross- country skiing hab-
its with all- cause mortality events in general populations
have not been previously investigated. We therefore aimed
to assess the nature and magnitude of the associations of
the total volume (MET- hours) and duration of leisure- time
cross- country skiing with the risk of all- cause mortality
events using the established Kuopio Ischemic Heart Disease
(KIHD) population- based cohort comprising apparently
healthy middle- aged men from eastern Finland.
2
|
METHODS
2.1
|
Participants and study setting
The study population comprised a representative sample of
men living in the city of Kuopio and its surrounding rural
communities in eastern Finland, who were recruited into the
KIHD Risk Factor Study. The KIHD study is a population-
based prospective cohort study designed to investigate risk
factors for CVD and related outcomes.15 Participants age
ranged from 42 to 61 years of age during study recruitment.
Of 3235 potentially randomly selected eligible men, 2682
(82.9%) volunteered to participate in this study. The baseline
examinations were performed between 20 March 1984 and 5
December 1989. The present analysis is based on 2087 men
participating in the follow- up study with no missing data on
the assessment of cross- country skiing activity, total PA,
relevant covariates, and all- cause mortality outcomes. The
study was approved by the Research Ethics Committee of
the University of Eastern Finland, and each participant gave
written informed consent. This study was performed fol-
lowing the STrengthening the Reporting of OBservational
studies in Epidemiology (STROBE) guidelines for reporting
observational studies in epidemiology (Table S1).16
2.2
|
Assessment of cross- country
skiing activity
The assessment of cross- country skiing activity as well
as total PA (kcal/d) was based on a 12- month PA ques-
tionnaire modified from the Minnesota Leisure- Time PA
Questionnaire.17 Cross- country skiing, which is a high-
intensity conditioning PA and the main exposure of this
study, was assigned an intensity of 9.6 MET, which is the
mean intensity of cross- country skiing activities. The total
volume of cross- country skiing (MET- hours per year) was
estimated using the intensity in METs multiplied by the dura-
tion. As skiing is only performed during the winter months,
assessment of the frequency, average duration, and intensity
of cross- country skiing was based on the winter season. Total
PA was classified as a conditioning PA or non- conditioning
PA. This detailed quantitative questionnaire deals with the
most common leisure- time physical activities of middle- aged
Finnish men (conditioning PA, eg, walking, skiing, bicy-
cling, swimming, rowing, ball games, and non- conditioning
PA, eg, crafts, repairs, building, gardening, hunting, fishing)
and enables the assessment of all components of PA.18,19 For
each type of PA performed, participants were asked to record
the frequency (number of sessions per month), average dura-
tion (hours and minutes per session), and intensity (scored as
0 for recreational activity, 1 for conditioning activity, 2 for
brisk conditioning activity, and 3 for competitive, strenuous
exercise). A trained interviewer collected missing data. The
intensity of PA was expressed in metabolic units (MET or
metabolic equivalents of oxygen consumption). The four cat-
egories of intensity of activity (range of possible scores, 0 to
3) were assigned their own metabolic- unit values, revised on
the basis of a synthesis of available empirical data.19
2.3
|
Ascertainment of outcomes
All deaths that occurred from study entry through to 2014
were included. No losses to follow- up were recorded as
participants are under continuous annual surveillance for
the development of new fatal events. Deaths were as-
certained from hospital records, wards of health centers,

|
3
LAUKKANEN EtAL.
questionnaires administered to health workers, death cer-
tificates, autopsy reports, and medico- legal reports.15,20,21
The Independent Events Committee of the KIHD study,
blinded to clinical data, performed classification of all
deaths.
2.4
|
Assessment of baseline risk factors and
confounders
Risk markers and all other characteristics were assessed
during the same visit at study entry. A subject was defined
a smoker if he had ever smoked cigarettes, cigars, or a pipe
on a regular basis within the past 30 days. Resting blood
pressure was measured between 8:00 and 10:00 am with
a random- zero sphygmomanometer. Alcohol consump-
tion was assessed using the Nordic Alcohol Consumption
Inventory.18 Body mass index (BMI) was computed as the
ratio of weight in kilograms to the square of height in me-
ters, using measured body weight and height. Diabetes was
defined as a fasting blood glucose level ≥7.0 mmol/L or
clinical diagnosis of diabetes with dietary, oral, or insu-
lin treatment. The collection of blood specimens and the
measurement of serum lipids, lipoproteins, creatinine, and
glucose have been described previously.22,23 The use of
medications, baseline diseases, and the level of PA were
assessed by self- administered questionnaires. Prevalent
CHD was defined as either a previous myocardial infarc-
tion, angina pectoris, or the use of nitroglycerin for chest
pain once a week or more frequently.
2.5
|
Statistical analyses
Baseline characteristics of the participants were summarized
using descriptive analyses. To assess the cross- sectional as-
sociations of total volume of cross- country skiing with du-
ration of cross- country skiing and several risk markers, we
calculated partial correlation coefficients. Hazard ratios
(HRs) with 95% confidence intervals (CIs) for all- cause mor-
tality were estimated using Cox proportional hazard models,
after confirmation of the assumptions of the proportionality
of hazards using Schoenfeld residuals.24 Subjects were clas-
sified into groups on the basis of the participants’ total vol-
ume of cross- country skiing (0, 1- 200 and >200 MET- hours
per 1 year) and the weekly average duration of cross- country
skiing (0, 1- 60, >60 min/wk). Hazard ratios were adjusted
for (i) age and (ii) established risk factors (age, BMI, systolic
blood pressure [SBP], high- density lipoprotein cholesterol
[HDL- C], smoking status, alcohol consumption, prevalent
CHD, history of diabetes mellitus, resting heart rate, and total
PA). We evaluated effect modification by age, BMI, SBP,
HDL- C, total PA, history of diabetes, smoking status, and
prevalent CHD, using tests of interaction. We conducted sen-
sitivity analysis which involved excluding the first 5 years of
follow- up. All statistical analyses were conducted using Stata
version 14 (Stata Corp, College Station, TX, USA).
3
|
RESULTS
3.1
|
Baseline characteristics and correlates
of total volume of cross- country skiing
The mean age of participants at study entry was 53 (standard
deviation [SD], 5) years. The baseline median (interquar-
tile, [IQR]) total volume of cross- country skiing was 42.5
(0- 195.5) MET- hours per year, and the corresponding value
for duration of cross- country skiing was 60 (60- 90) minutes
per week (Table 1). There were weak inverse correlations of
total intensity of cross- country skiing with alcohol consump-
tion, physical measures (BMI, diastolic blood pressure, and
resting heart rate), triglycerides, and eGFR. Strong positive
correlations were observed for total volume of cross- country
skiing with average duration of cross- skiing (r = .43) and
total PA (r = .40).
3.2
|
Cross- country skiing and
all- cause mortality
During a median (IQR) follow- up of 26.1 (18.7- 28.0)
years, a total of 1028 all- cause mortality events occurred.
All- cause mortality rates per 1000 person- years of follow-
up across the three groups of total volume of cross- country
skiing (0, 1- 200, and >200 MET- hours per year) were
28.1 (95% CI: 25.7- 30.7), 18.9 (95% CI: 16.9- 21.0), and
17.4 (95% CI: 15.2- 20.0), respectively. Cumulative hazard
curves demonstrated a greater risk of all- cause mortality
among participants who did not do any cross- country skiing
compared to the other groups (P < .0001 for log- rank test;
Figure 1A). In analyses adjusted for age at baseline, when
compared to participants who did not do cross- country ski-
ing, the HRs (95% CIs) of all- cause mortality were 0.67
(0.58- 0.77) and 0.60 (0.51- 0.70) for participants who did
1- 200 and >200 MET- hours per year of cross- country ski-
ing, respectively (Table 2). After adjustment for several
established risk factors and potential confounders (age,
BMI, SBP, HDL- C, smoking status, alcohol consump-
tion, prevalent CHD, history of diabetes mellitus, resting
heart rate, and total PA), the respective HRs (95% CIs) of
all- cause mortality were attenuated to 0.84 (0.73- 0.97) for
1- 200 MET- hours per year of cross- country skiing and 0.80
(0.67- 0.96) for >200 MET- hours per year of cross- country
skiing.
All- cause mortality rates per 1000 person- years of follow- up
across the three groups of average duration of cross- country ski-
ing (0, 1- 60, >60 min/wk) were 28.1 (95% CI: 25.7- 30.7), 18.4
(95% CI: 16.5- 20.5), and 18.1 (95% CI: 15.9- 0.7), respectively.
Similarly, cumulative hazard curves demonstrated a greater risk

4
|
LAUKKANEN EtAL.
of all- cause mortality among participants who did not do any
cross- country skiing compared with the other groups (P < .0001
for log- rank test; Figure 1B). Compared with participants who
did not do any cross- skiing, there was a 35% lower risk of
all- cause mortality among participants who did cross- country
skiing for 1- 60 min/wk and a 38% lower risk of all- cause
TABLE 1 Baseline participant
characteristics and correlates of total volume
of cross- country skiing
Mean (SD), median
(IQR), or %
Pearson correlation
r (95% CI)a
Total volume of skiing (MET- h/y)b42.5 (0- 195.5) -
Duration of skiing (min/wk)b60 (60- 90) .43 (0.38, 0.47)***
Questionnaire/prevalent conditions
Age at survey (years) 52.6 (5.2) −.02 (−0.06, 0.02)
Alcohol consumption (g/wk) 76.8 (140.5) −.04 (−0.09, −0.00)*
History of diabetes
No 96.3 -
Yes 3.7 -
Smoking status
Other 67.9 -
Current 32.1 -
History of hypertension
No 69.3 -
Yes 30.7 -
History of CHD
No 75.4 -
Yes 24.6 -
Use of antihypertensives
No 78.5 -
Yes 21.5 -
Medication for dyslipidemia
No 99.4 -
Yes 0.6 -
Physical measurements
BMI (kg/m2) 26.9 (3.6) −.07 (−0.11, −0.03)**
SBP (mm Hg) 134.2 (16.8) −.03 (−0.07, 0.01)
DBP (mm Hg) 89.1 (10.5) −.05 (−0.09, −0.01)*
Resting heart rate (bpm) 62.5 (10.8) −.08 (−0.13, −0.04)***
Total physical activity (kcal/d) 368.1 (359.7) .40 (0.36, 0.44)***
Lipid markers
Total cholesterol (mmol/L) 5.93 (1.08) −.01 (−0.05, 0.04)
HDL- C (mmol/L) 1.28 (0.30) .08 (0.04, 0.12)***
Triglycerides (mmol/L) 1.10 (0.80- 1.56) −.09 (−0.13, −0.04)***
Metabolic and renal markers
Fasting plasma glucose (mmol/L) 5.35 (1.28) −0.03 (−0.07, 0.01)
Serum creatinine (μmol/L) 89.5 (22.0) .02 (−0.03, 0.06)
Estimated GFR (mL/min/1.73 m2) 87.5 (17.5) −.05 (−0.09, −0.00)*
Asterisks indicate the level of statistical significance: *P < .05; **P < .01; ***P < .001.
BMI, body mass index; CHD, coronary heart disease; DBP, diastolic blood pressure; GFR, glomerular filtration
rate; HDL- C, high- density lipoprotein cholesterol; IQR, interquartile range; SBP, systolic blood pressure.
aPearson correlation coefficients between duration of skiing and the row variables.
bVolume and duration of cross- country skiing are estimated based on the frequency, average duration, and inten-
sity of cross- country skiing during the winter months.

|
5
LAUKKANEN EtAL.
mortality among participants who did cross- country skiing for
>60 min/wk, in analysis adjusted for age (Table 3). The risk was
16% and 18% lower, respectively, after adjustment for several
established risk factors and potential confounders. The associa-
tions of both total volume and duration of cross- country skiing
with all- cause mortality risk remained consistent in analyses that
excluded the first 5 years of follow- up (Tables S2 and S3).
3.3
|
Associations of cross- country skiing
with all- cause mortality in subgroups
Figure 2 shows the associations of total volume and average
duration of cross- country skiing with all- cause mortality in
clinically relevant subgroups. Except for evidence of statisti-
cally significant interactions by BMI (P for interaction for all
FIGURE 1 Cumulative Kaplan- Meier curves for all- cause mortality during follow- up according to total volume (A) and duration of cross-
country skiing (B)
(A) (B)
TABLE 2 Associations of total volume of cross- country skiing with all- cause mortality
Total volume of cross- country
skiing (MET- h/y)aEvents/total
Model 1 Model 2
HR (95% CI) P-value HR (95% CI) P-value
0 480/821 ref ref
1- 200 337/759 0.67 (0.58 to 0.77) <.001 0.84 (0.73 to 0.97) .020
>200 211/507 0.60 (0.51 to 0.70) <.001 0.80 (0.67 to 0.96) .015
P- value for trend <.001 .007
Model 1: adjusted for age.
Model 2: model 1 plus body mass index, systolic blood pressure, high- density lipoprotein cholesterol, smoking status, alcohol consumption, prevalent coronary heart
disease, history of diabetes mellitus, resting heart rate, and total physical activity.
aTotal volume of cross- country skiing in MET- h/y is based on cross- country skiing during the winter months.
TABLE 3 Associations of duration of cross- country skiing with all- cause mortality
Duration of cross- country
skiing (min/wk)aEvents/total
Model 1 Model 2
HR (95% CI) P-value HR (95% CI) P-value
0 480/821 ref ref
1- 60 330/761 0.65 (0.56 to 0.75) <.001 0.84 (0.72 to 0.97) .017
>60 218/505 0.62 (0.53 to 0.73) <.001 0.82 (0.69 to 0.97) .019
P- value for trend <.001 .010
Model 1: adjusted for age.
Model 2: model 1 plus body mass index, systolic blood pressure, high- density lipoprotein cholesterol, smoking status, alcohol consumption, prevalent coronary heart
disease, history of diabetes mellitus, resting heart rate, and total physical activity.
aDuration of cross- country skiing is based on cross- country skiing during the winter months.

6
|
LAUKKANEN EtAL.
<.05) and history of diabetes (P for interaction = .025) for the
associations of total volume and duration of cross- country ski-
ing with all- cause mortality, there was no modification of the
associations by other clinically relevant characteristics. For the
associations of both total volume and duration of cross- country
skiing with risk of all- cause mortality, statistically significant
inverse associations were observed in men with lower BMI
(<26.5 kg/m2) compared with a non- significant association in
men with a higher BMI (≥26.5 kg/m2). A statistically signifi-
cant inverse association was found in men with no history of
diabetes, while a trend toward a positive association was ob-
served in men with a history of diabetes for the respective as-
sociation between the total volume of cross- country skiing and
all- cause mortality risk.
4
|
DISCUSSION
4.1
|
Key findings
In the first prospective analysis of the associations of leisure-
time cross- country skiing habits and the risk of all- cause
mortality in a middle- aged Caucasian population, we found
inverse associations of both the total volume and average
duration of cross- country skiing with the risk of all- cause
mortality outcomes. The associations were independent of
several established risk factors as well as a number of other
potential confounders. There were weak inverse correlations
of the total volume of cross- country skiing with several risk
factors and physiological measures. As expected, a strong
correlation was observed with duration of cross- country ski-
ing and the total amount of PA. Except for evidence of effect
modification by BMI and history of diabetes, the associa-
tions of total volume and duration of cross- country skiing
with all- cause mortality were not modified by several clini-
cally relevant characteristics. Both total volume and duration
of cross- country skiing were strongly and inversely associ-
ated with all- cause mortality risk in men with lower BMI
compared with less robust associations in men with higher
BMI. This observation might reflect evidence that people
with low BMI are usually more likely to be active individu-
als participating in high- intensity physical exercise, have
better metabolic profiles, and a lower mortality risk. The
same explanation might also be applicable to the stronger
protective effect of total volume of cross- country skiing on
the risk of all- cause mortality in men without a history of
diabetes, which is an opposite finding to the trend toward an
increased mortality in men with prevalent diabetes. These
findings from subgroup analyses however need replication in
further studies given the multiple statistical tests for interac-
tion conducted.
FIGURE 2 Associations of total volume and duration of cross- country skiing with all- cause mortality in clinically relevant subgroups.
Hazard ratios (HRs) were adjusted for age, body mass index (BMI), systolic blood pressure (SBP), high- density lipoprotein cholesterol (HDL- C),
smoking status, alcohol consumption, prevalent coronary heart disease, history of diabetes mellitus, resting heart rate, and physical activity. CI,
confidence interval; *P- value for interaction; cut- offs used for age, BMI, SBP, HDL- C, and total physical activity are median values. For total
volume of cross- country skiing, HRs are reported comparing >200 metabolic equivalent- h/y of cross- country skiing with no skiing activity; for
duration of cross- country skiing, HRs are reported comparing >60 min of cross- country skiing/wk with no skiing activity

|
7
LAUKKANEN EtAL.
4.2
|
Comparison with previous work and
implications of findings
Substantial evidence shows that high- or vigorous- intensity
PA is associated with reduced risk of all- cause mortality.1-3
A recently published cohort study of 80 306 British adults
found robust associations between participation in certain
types of sport activities and mortality; substantial reduc-
tions in both cardiovascular and all- cause mortality were
observed for aerobics, swimming, and racquet sports.25
The specific associations of cross- country skiing habits
with overall mortality have however been less widely stud-
ied in general population. Our overall results show that
cross- country skiing, a high- intensity conditioning PA, is
associated with a reduced risk of mortality from all causes.
However, there has been some evidence that vigorous PA
might cause some harms as well as exercise- related deaths.
Indeed, biochemical as well as echocardiographic evidence
of cardiac injury has been observed in both high endurance
recreational runners and athletes.26-28 There is increased
short- term risk of sudden coronary events during high-
intensity exercise. It has been reported that participants
in long- distance ski races have an increased risk of acute
cardiovascular events.29,30 Among male participants of a
90 km Swedish cross- country skiing event, a faster finish-
ing time and a high number of completed races were as-
sociated with higher risk of arrhythmias.29 However, it can
be argued that since physically active people spend only a
small amount of their time in these competitive races, the
short- term adverse effects or excess mortality during these
periods is by far outweighed by the long- term beneficial
effects of regular high- intensity exercise such as cross-
country skiing.
4.3
|
Possible explanations for findings
Multiple mechanisms have been postulated to underpin the
protective effects of PA on adverse outcomes such as CVD,
diabetes, and cancer, which eventually lead to death, and
these involve both physiological and metabolic processes.
As cross- country skiing is a high- intensity PA, it is hy-
pothesized that these same processes may underlie the as-
sociations demonstrated. PA exerts beneficial effects on
cardiovascular function through (i) improvements in en-
dothelial function and levels of cardiovascular risk factors
such as blood pressure, natriuretic peptides, lipoproteins,
glucose tolerance, hemostatic factors, and cardiac troponin
T;10,31–33 (ii) its anti- inflammatory effects;34 and (iii) en-
hancement of cardiac function.35,36 PA may also reduce
cancer risk via its effect on obesity, which modulates circu-
lating levels of adipokines and cytokines.37 Direct effects of
PA on target organs and tissues have also been reported.37
Furthermore, it is possible that increasing cardiorespiratory
fitness (CRF) through aerobic PA such as cross- country ski-
ing could be the key mechanism in reducing mortality risk.
Indeed, CRF has been consistently shown to be strongly and
robustly associated with reduced risk of all- cause mortal-
ity.38 Although CRF is largely determined by PA, it has
been reported that genetic factors account for 25%- 40% of
the variation in CRF.39 Emerging evidence from both animal
and human studies indicates that the same genetic factors
influence PA levels, CRF, and the risk of mortality, lead-
ing to the suggestion that genetic pleiotropy might partly
explain the observed associations between high baseline PA
and reduced mortality in long- term observational follow- up
studies such as ours.40
4.4
|
Strengths and limitations
Several strengths of our study deserve mention. To our
knowledge, this is the first comprehensive investigation of
the observational epidemiological prospective association
between cross- country skiing habits and all- cause mortal-
ity risk in the general male population. Other strengths
include the prospective cohort design; a homogenous
community- based cohort which was a nationally representa-
tive sample of middle- aged Finnish men; large number of
well- characterized participants and outcome events; the high
participation rate and the absence of loss to follow- up; the
long- term follow- up of over 25 years; confirmation of out-
comes by independent physician review; and measurements
on a comprehensive panel of lifestyle and biological markers
which permitted adequate adjustment for several potential
confounders. Several limitations should be considered when
interpreting our results. First, cross- country skiing habits as
well as other physical activities were self- reported, leading
to potential misclassification. Second, there is a possibility
that the higher mortality rates in men who did not ski could
be partly due to underlying chronic diseases (diagnosed or
undiagnosed) which are associated with an increased risk
of mortality (reverse causation). However, the associations
remained robust in subgroup analyses and in a sensitivity
analysis which involved exclusion of the first 5 years of
follow- up. Third, residual confounding remains a potential
alternative explanation because of the observational nature
of the data. Fourth, participants in our study were middle-
aged Caucasian men, so generalizability to other popula-
tions and age groups might be limited. Fifth, skiing as a PA
is usually only possible in the winter months and therefore
may reflect individuals’ seasonal PA habits and the ability to
perform high- intensity PA. Sixth, we could not correct for
regression dilution bias which may have underestimated the
associations demonstrated, as only baseline assessments of
cross- country skiing habits were available; there is a pos-
sibility that skiing habits may have changed during follow-
up. Seventh, our findings may only be limited to the effects

8
|
LAUKKANEN EtAL.
of aerobic exercise conducted during the winter months, as
cross- country skiing cannot be performed all year- round in
eastern Finland. The amount of cross- country skiing may be
related to the total amount of physical endurance exercise
performed by study participants, which suggests that those
who are active skiers during the winter months are the same
people commonly undertaking other types of aerobic endur-
ance exercise such as walking, cycling, and running during
the year. Thus, the protective effect of cross- country skiing
may rather be due to total endurance PA. However, while
this is possible, we adjusted for the total amount of PA in
our analyses, thus emphasizing an independent relation-
ship between skiing amount and mortality. Finally, although
cross- country skiing is mainly a high- intensity conditioning
PA with an assigned mean intensity of 9.6 MET, participants
may have skied at various intensities ranging from low-
intensity to very high- intensity. Nevertheless, in this study,
assessment of the intensity of cross- country skiing was
based on a well- known previously validated leisure- time PA
questionnaire.18,19
In conclusion, new data show that total volume as well
as duration of leisure- time cross- country skiing is each in-
versely and independently associated with all- cause mortal-
ity in a middle- aged Caucasian male population. The overall
results add to accumulating evidence that conditioning high-
intensity PA is associated with reduced risk of mortality.
5
|
PERSPECTIVE
There is some evidence on the inverse association between
participation in certain types of sports activities and the risk
of mortality. Data on the associations between high- intensity
physical activities such as cross- country skiing and all-
cause mortality are limited. Although, long- term endurance
cross- country skiing has been shown to be associated with
reduced risk of all- cause mortality, the prospective associa-
tion between leisure- time cross- country skiing and the risk of
all- cause mortality in general populations has not been pre-
viously investigated. Using a population- based prospective
study, we found robust evidence that total volume as well as
duration of leisure- time cross- country skiing is each associ-
ated with reduced risk of all- cause mortality in a middle- aged
male population. This study adds to accumulating evidence
that high- intensity PA such as cross- country skiing is an
effective form of health- enhancing PA.
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 data collection in the study.
COMPETING INTERESTS
None declared.
AUTHORS CONTRIBUTORS
JL and TL conceived the idea for the study. JL, TL, and SKK
made a study plan. JL, TL, and SKK contributed to the de-
velopment of the data analysis design. SKK processed and
analyzed the data. JL and SKK drafted the manuscript. TL
contributed to the writing of the manuscript. All authors con-
tributed to drafting the rebuttal and revising the final draft of
the manuscript.
ORCID
J. A. Laukkanen http://orcid.org/0000-0002-3738-1586
REFERENCES
1. Lee IM, Paffenbarger RS Jr. Associations of light, moderate, and
vigorous intensity physical activity with longevity. The harvard
alumni health study. Am J Epidemiol. 2000;151:293‐299.
2. Physical activity guidelines advisory committee. Physical Activity
Guidelines Advisory Committee Report, 2008. Washington, DC:
US Department of Health and Human Services; 2008;683.
3. Tanasescu M, Leitzmann MF, Rimm EB, Willett WC, Stampfer
MJ, Hu FB. Exercise type and intensity in relation to coronary
heart disease in men. JAMA. 2002;288:1994‐2000.
4. Swain DP, Franklin BA. Comparison of cardioprotective bene-
fits of vigorous versus moderate intensity aerobic exercise. Am J
Cardiol. 2006;97:141‐147.
5. Wen CP, Wai JP, Tsai MK, Chen CH. Minimal amount of exer-
cise to prolong life: to walk, to run, or just mix it up? J Am Coll
Cardiol. 2014;64:482‐484.
6. Lee DC, Pate RR, Lavie CJ, Sui X, Church TS, Blair SN. Leisure-
time running reduces all- cause and cardiovascular mortality risk.
J Am Coll Cardiol. 2014;64:472‐481.
7. Wen CP, Wai JP, Tsai MK, etal. Minimum amount of physical
activity for reduced mortality and extended life expectancy: a pro-
spective cohort study. Lancet. 2011;378:1244‐1253.
8. Chomistek AK, Cook NR, Flint AJ, Rimm EB. Vigorous- intensity
leisure- time physical activity and risk of major chronic disease in
men. Med Sci Sports Exerc. 2012;44:1898‐1905.
9. Williams PT. Reductions in incident coronary heart disease risk
above guideline physical activity levels in men. Atherosclerosis.
2010;209:524‐527.
10. Kraus WE, Houmard JA, Duscha BD, etal. Effects of the amount
and intensity of exercise on plasma lipoproteins. N Engl J Med.
2002;347:1483‐1492.
11. Williams PT, Thompson PD. Walking versus running for hy-
pertension, cholesterol, and diabetes mellitus risk reduction.
Arterioscler Thromb Vasc Biol. 2013;33:1085‐1091.
12. Williams PT. Reduced total and cause- specific mortality from walk-
ing and running in diabetes. Med Sci Sports Exerc. 2014;46:933‐939.
13. Grimsmo J, Maehlum S, Moelstad P, Arnesen H. Mortality and
cardiovascular morbidity among long- term endurance male cross

|
9
LAUKKANEN EtAL.
country skiers followed for 28- 30years. Scand J Med Sci Sports.
2011;21:e351‐e358.
14. Sarna S, Sahi T, Koskenvuo M, Kaprio J. Increased life ex-
pectancy of world class male athletes. Med Sci Sports Exerc.
1993;25:237‐244.
15. Laukkanen JA, Kurl S, Salonen R, Rauramaa R, Salonen JT. The
predictive value of cardiorespiratory fitness for cardiovascular
events in men with various risk profiles: a prospective population-
based cohort study. Eur Heart J. 2004;25:1428‐1437.
16. von Elm E, Altman DG, Egger M, Pocock SJ, Gotzsche PC,
Vandenbroucke JP. The strengthening the reporting of obser-
vational studies in epidemiology (strobe) statement: guide-
lines for reporting observational studies. J Clin Epidemiol.
2008;61:344‐349.
17. Taylor HL, Jacobs DR Jr, Schucker B, Knudsen J, Leon AS,
Debacker G. A questionnaire for the assessment of leisure time
physical activities. J Chronic Dis. 1978;31:741‐755.
18. Lakka TA, Venalainen JM, Rauramaa R, Salonen R, Tuomilehto
J, Salonen JT. Relation of leisure- time physical activity and car-
diorespiratory fitness to the risk of acute myocardial infarction. N
Engl J Med. 1994;330:1549‐1554.
19. Lakka TA, Salonen JT. Intra- person variability of various physi-
cal activity assessments in the kuopio ischaemic heart disease risk
factor study. Int J Epidemiol. 1992;21:467‐472.
20. Kunutsor SK, Kurl S, Khan H, Zaccardi F, Laukkanen JA.
Associations of cardiovascular and all- cause mortality events
with oxygen uptake at ventilatory threshold. Int J Cardiol.
2017;236:444‐450.
21. Kunutsor SK, Kurl S, Khan H, Zaccardi F, Rauramaa R,
Laukkanen JA. Oxygen uptake at aerobic threshold is inversely
associated with fatal cardiovascular and all- cause mortality
events. Ann Med. 2017;49:698‐709.
22. Kunutsor SK, Khan H, Nyyssonen K, Laukkanen JA.
Lipoprotein(a) and risk of sudden cardiac death in middle- aged
finnish men: a new prospective cohort study. Int J Cardiol.
2016;220:718‐725.
23. Salonen JT, Nyyssonen K, Korpela H, Tuomilehto J, Seppanen
R, Salonen R. High stored iron levels are associated with excess
risk of myocardial infarction in eastern finnish men. Circulation.
1992;86:803‐811.
24. Therneau TM, Grambsch PM. Modeling Survival Data: Extending
the cox Model. New York: Springer; 2000.
25. Oja P, Kelly P, Pedisic Z, et al. Associations of specific types
of sports and exercise with all- cause and cardiovascular- disease
mortality: a cohort study of 80 306 British adults. Br J Sports
Med. 2016;51:812‐817.
26. Neilan TG, Januzzi JL, Lee-Lewandrowski E, etal. Myocardial
injury and ventricular dysfunction related to training levels
among nonelite participants in the boston marathon. Circulation.
2006;114:2325‐2333.
27. Jaworski CA. Medical concerns of marathons. Curr Sports Med
Rep. 2005;4:137‐143.
28. Fortescue EB, Shin AY, Greenes DS, etal. Cardiac troponin in-
creases among runners in the boston marathon. Ann Emerg Med.
2007;49:137‐143. 143 e131.
29. Andersen K, Farahmand B, Ahlbom A, etal. Risk of arrhythmias
in 52 755 long- distance cross- country skiers: a cohort study. Eur
Heart J. 2013;34:3624‐3631.
30. Farahmand B, Hållmarker U, Brobert GP, Ahlbom A. Acute mor-
tality during long- distance ski races (vasaloppet). Scand J Med
Sci Sports. 2007;17:356‐361.
31. Tran ZV, Weltman A, Glass GV, Mood DP. The effects of exercise
on blood lipids and lipoproteins: a meta- analysis of studies. Med
Sci Sports Exerc. 1983;15:393‐402.
32. deFilippi CR, de Lemos JA, Tkaczuk AT, etal. Physical activ-
ity, change in biomarkers of myocardial stress and injury, and
subsequent heart failure risk in older adults. J Am Coll Cardiol.
2012;60:2539‐2547.
33. Mora S, Lee IM, Buring JE, Ridker PM. Association of physical
activity and body mass index with novel and traditional cardio-
vascular biomarkers in women. JAMA. 2006;295:1412‐1419.
34. Lavie CJ, Church TS, Milani RV, Earnest CP. Impact of phys-
ical activity, cardiorespiratory fitness, and exercise training
on markers of inflammation. J Cardiopulm Rehabil Prev.
2011;31:137‐145.
35. Hambrecht R, Wolf A, Gielen S, etal. Effect of exercise on coro-
nary endothelial function in patients with coronary artery disease.
N Engl J Med. 2000;342:454‐460.
36. Myers J, Prakash M, Froelicher V, Do D, Partington S, Atwood
JE. Exercise capacity and mortality among men referred for exer-
cise testing. N Engl J Med. 2002;346:793‐801.
37. McTiernan A. Mechanisms linking physical activity with cancer.
Nat Rev Cancer. 2008;8:205‐211.
38. Kodama S, Saito K, Tanaka S, etal. Cardiorespiratory fitness as
a quantitative predictor of all- cause mortality and cardiovascu-
lar events in healthy men and women: a meta- analysis. JAMA.
2009;301:2024‐2035.
39. Bouchard C, Shephard RJ, Stephens T, eds. Physical Activity,
Fitness, and Health: International Proceedings and Consensus
Statement. Champaign, III: Human Kinetics Publisher Inc;
1994.
40. Karvinen S, Waller K, Silvennoinen M, etal. Physical activity in
adulthood: genes and mortality. Sci Rep. 2015;5:18259.
SUPPORTING INFORMATION
Additional Supporting Information may be found online in
the supporting information tab for this article.
How to cite this article: Laukkanen JA, Laukkanen
T, Kunutsor SK. Cross- country skiing is associated
with lower all- cause mortality: A population- based
follow- up study. Scand J Med Sci Sports. 2017;00:
1–9. https://doi.org/10.1111/sms.12980