Exercise Patterns and Peak Oxygen Uptake in a Healthy Population

The HUNT Study

Abstract

Purpose 

The objective of this study is to examine how different approaches of the current exercise recommendations for adults associate with V˙O_2peak in a large healthy population. We further examined how a lower duration than recommended, if performed at very vigorous intensity, was related to V˙O_2peak.

Methods 

A total of 4631 healthy adults age 19–89 yr (2263 men and 2368 women) were tested for V˙O_2peak (mean = 44.3 and 35.9 mL·kg⁻¹·min⁻¹ for men and women, respectively). Information on exercise habits was collected through a questionnaire, including questions on frequency, duration, and relative intensity (Borg scale 6–20). A general linear model was applied to assess the associations between physical activity and V˙O_2peak.

Results 

V˙O_2peak did not differ considerably between people who reported to exercise ≥150 min·wk⁻¹ (average = 216 min·wk⁻¹, V˙O_2peak = 45.2 and 36.5 mL·kg⁻¹·min⁻¹ for men and women, respectively) with moderate intensity and people who reported 75–149 min·wk⁻¹ (average = 112.5 min·wk⁻¹, V˙O_2peak = 47.5 and 37.3 mL·kg⁻¹·min⁻¹ for men and women) with vigorous intensity, but it was higher than that in people who reported inactivity (V˙O_2peak = 40.1 and 32.3 mL·kg⁻¹·min⁻¹ for men and women) or low-intensity exercise (V˙O_2peak = 41.2 and 40.1 mL·kg⁻¹·min⁻¹ for men and women). Reporting exercise at very vigorous intensity but with a duration of less than 75 min·wk⁻¹ (average = 49 min·wk⁻¹) was associated with a V˙O_2peak that was similarly high (47.6 and 36.7 mL·kg⁻¹·min⁻¹ for men and women).

Conclusion 

Our findings support current recommendations by showing that exercise of both “moderate intensity–long duration” and “vigorous intensity–short duration” was associated with similarly high V˙O_2peak. Our results also suggest that exercising at very vigorous intensity may be beneficial for V˙O_2peak even with considerably lower total exercise time than expressed in today’s recommendations.

Peak oxygen uptake (V˙O_2peak) and physical activity (PA) are both strongly and inversely associated with cardiovascular morbidity and mortality (^3,14,24). Habitual PA and V˙O_2peak are modestly correlated at the population level, and the results of some studies suggest that PA and V˙O_2peak may influence metabolic and cardiovascular risk through different pathways (^16,17,26). It has been proposed that estimated or directly measured V˙O_2peak is a stronger determinant of cardiovascular disease risk and longevity than measures of PA (^3,17,26). However, PA and V˙O_2peak cannot be easily distinguished because an increase in PA is the main method to increase fitness. Therefore, health-promoting effects of a physically active lifestyle may partly be mediated through increased V˙O_2peak.

Leisure time PA that improves V˙O_2peak is commonly denoted as exercise training (⁹) and typically assessed by its frequency, intensity, and duration, of which intensity may be the most important determinant of V˙O_2peak (^6,27). Current recommendations from the American College of Sports Medicine/American Heart Association suggest that adults at all ages should perform regular moderate-intensity exercise most days of the week to achieve at least 150 min of exercise per week or vigorous-intensity exercise of at least 75 min·wk⁻¹ (¹²). Thus, the recommendations suggest that the aim may be reached either by a “long duration–moderate intensity” approach or by a “shorter duration–vigorous intensity” approach. However, in most population-based studies, it has not been taken into account that vigorous-intensity exercise will generate higher energy expenditure, and there is a lack of studies that have compared effects of vigorous and moderate-intensity exercise with the same amount of net energy expenditure (⁹). However, the results of some randomized controlled trials have shown greater benefits of vigorous-intensity exercise compared with low- and moderate-intensity exercise given the same total exercise volume (^5,10,21). Those studies were targeted interventions and included small or selected samples of participants (^1,5,10,19). No population-based study has examined how different unsupervised, freely selected exercise patterns are associated with directly measured V˙O_2peak.

In this large cross-sectional population study of healthy individuals, we compared self-reported exercise at a “vigorous intensity–short duration” level and at a “moderate intensity–long duration” level in relation to measurements of V˙O_2peak. We also examined how a low duration of very vigorous-intensity exercise was associated with V˙O_2peak.

MATERIALS AND METHODS

Study population

The HUNT study is a community-based health survey to which the total adult population in Nord-Trøndelag County in Norway has been invited to participate at three occasions (¹³). The third wave of the HUNT study was carried out between October 2006 and June 2008. In total, 94,194 persons were invited and 50,821 attended. In a substudy (the HUNT Fitness Study), 12,609 persons from four selected municipalities within the county were offered testing of V˙O_2peak. At study closure, 4631 persons had appeared and completed the test. Assessment of PA was based on self-report, and those who were invited to the HUNT Fitness Study had to be free from self-reported cardiovascular disease, lung diseases, cancer, and hypertension. The HUNT Fitness Study population was similar to the total healthy HUNT population regarding level of conventional cardiovascular disease risk factors such as age, blood pressure, waist circumference, body mass index (BMI), and blood glucose (²). However, a smaller proportion of the HUNT Fitness Study participants reported to be inactive compared with the overall healthy HUNT study participants (2.5% vs. 4.4%, respectively). Further descriptive data of the population have been published elsewhere (²).

The study was approved by the regional committee for medical research ethics, the Norwegian Data Inspectorate, and the National Directorate of Health. All participants signed a document of consent, and the study was conducted in conformity with the Declaration of Helsinki.

Exercise testing procedures

The participants were introduced to treadmill walking through a 10-min warm-up at a moderate relative intensity. Immediately after the warm-up period, the participants were equipped with an HR monitor (Polar S610 or RS400; Polar, Kempele, Finland) and face mask (Hans Rudolph V, Shawnee, KS) before entering the test treadmill (DK7830; DK City, Taichung, Taiwan). An individualized ramp protocol previously published was used to measure V˙O_2max (²⁰). Oxygen uptake was measured continuously using a portable mixing chamber gas analyzer (MetaMax II; Cortex Biophysik Gmbh, Leipzig, Germany). Speed and/or inclination were increased regularly when the participants reached an oxygen uptake that was stable over 30 s, until they reached exhaustion. Before maximal effort, and as a prolongation of the warm-up period, most participants had their steady-state V˙O₂ estimated during one (n = 2773) or two (n = 2543) submaximal levels. For each level, speed, inclination, and V˙O₂ were registered in addition to their subjective level of exertion at the Borg scale 6–20. A test was considered maximal when the expiratory exchange ratio (RER) was above 1.05 combined with less than 2 mL·kg⁻¹·min⁻¹ increases in oxygen uptake despite increased workload. Because 17.7% of the participants did not attain the requirements of a maximal test, the term V˙O_2peak was used. A person’s V˙O_2peak was measured as the mean of the three successively highest 10-s V˙O₂ registrations. Trained personnel performed all the tests, and calibration of equipment was executed regularly with volume ventilation calibrated every third test and gas calibrated every fifth test.

PA questionnaire

Information on leisure time PA was collected from a detailed questionnaire. Questions were related to frequency, duration, and intensity of leisure time PA. The question on frequency was “How often do you exercise?”, with the following response options: “never,” “less than once a week,” “once a week,” “two to three times a week,” and “almost every day.” A corresponding question on duration of exercise was reported as average minutes per session and included four options: “less than 15 min,” “between 15 and 30 min,” “between 30 and 60 min,” and “more than 60 min.” Intensity of exercise was assessed by the Borg Rating of Perceived Exertion Scale, where participants were asked to assess their usual intensity level during exercise on a 6- to 20-point scale (⁴). Questions about frequency and duration were similar to the questions found in the HUNT questionnaire, which is previously validated and proved to be an appropriate tool for use in epidemiological studies (¹⁵). For intensity, the well-known and validated Borg scale of perceived exertion was chosen to give more detailed information on relative exercise intensity (⁴).

Categorization of PA

Participants who responded “none” or “less than once per week” to the frequency question were defined as “inactive.” Total exercise time in minutes per week was calculated by multiplying the average duration of exercise with the average frequency of exercise sessions. For example, reporting “between 30 and 60 min” and “two to three times a week” was interpreted as 45 min × 2.5 = 112.5 min·wk⁻¹. Total exercise time was then subdivided according to suggested cutoff values in the updated recommendations from the American College of Sports Medicine and the American Heart Association (^9,12), which correspond to <75, 75–149, and ≥150 min·wk⁻¹. Intensity was classified according to the Borg scale, with a range from 6 (very, very light/no exertion at all) to 20 (extremely hard/maximal exertion). Ratings from 6 to 11 were classified as low intensity, 12 to 13 as moderate intensity, and 14 to 20 as vigorous intensity, according to the recommendations (²³).

In a subsequent analysis, those who reported a relative intensity of 14 or above on the Borg scale were subdivided into “vigorous” (Borg scale 14–15) and “very vigorous” (Borg scale 16–20) to examine high intensities at different durations. For comparison, the moderate-intensity groups reporting ≥150 min·wk⁻¹ were retained. Those reporting moderate intensity (Borg scale 12–13) in less than 150 min·wk⁻¹ or low intensity (Borg scale <12) were categorized as “less than recommended activity,” which is in accordance with current recommendations.

During the V˙O_2peak test, participants were asked for their Borg scale rating at one or two submaximal levels plus at the termination of the test. From the submaximal steady-state V˙O₂ measures and the V˙O_2peak test, the test subject’s individual Borg–V˙O₂ association was calculated (Y = aX + b, where Y is V˙O₂ and X is Borg). For a given Borg scale rating, a corresponding V˙O₂ value as a percentage of V˙O_2peak could be established on the basis of the individual treadmill test. The individual energy expenditure during self-reported exercise was calculated by multiplying the V˙O₂ (L·min⁻¹) corresponding to the self-reported Borg scale rating by 5 kcal·min⁻¹ (equivalent to 1-L oxygen consumption for 1 min). Furthermore, average energy expenditure was multiplied by the self-reported weekly duration of exercise to estimate total energy expenditure. For example, in one individual reporting a relative exercise intensity of Borg scale 13 during regular exercise, this corresponded to 70% of the individual V˙O_2peak. Because the individual’s V˙O_2peak was 3 L·min⁻¹, the relative intensity corresponds to an average oxygen consumption of 2.1 L·min⁻¹ or 10.5 kcal·min⁻¹ during exercise (2.1 L·min⁻¹ × 5 kcal = 10.5 kcal·min⁻¹). If the individual reported an average exercise time of 150 min·wk⁻¹, the estimated weekly energy expenditure would be 1575 kcal·wk⁻¹ (10.5 kcal·min⁻¹ × 150 min·wk⁻¹). The estimated resting energy expenditure for each individual was subtracted from the total energy expenditure to obtain an estimate of net energy expenditure caused by exercise. Because the oxygen consumption at rest is proposed to be approximately 3.5 mL·kg⁻¹·min⁻¹, we used individual body weights to recalculate this to liters per minute (L·min⁻¹) and multiplied the result by 5 kcal·min⁻¹ (⁹). For a 70-kg person reporting 150 min of exercise, the subtracted resting energy expenditure would be approximately 0.25 L·min⁻¹ × 5 kcal·min⁻¹ × 150 min, corresponding to approximately 188 kcal. To categorize participants on the basis of total volume of exercise during a normal week, approximate tertiles of estimated net energy expenditure were created and compared with the inactive group (0 kcal·wk⁻¹ spent on exercise). The total volume categories were <1000, 1000–2000, and >2000 kcal·wk⁻¹ for men and <750, 750–1250, and >1250 kcal·wk⁻¹ for women.

Statistical analysis

Descriptive statistics of the total population tested for V˙O_2peak are given as means and SDs for continuous variables and frequency count and percentage for categorical variables. The data set was examined for missing values and erroneous outliers. In total, 4427 people were included in the analysis for the present study (2158 men and 2269 women) after exclusion due to missing values on body weight (n = 6) and frequency, duration, or intensity of self-reported exercise habits (n = 187). Continuous variables were tested for normality and heteroscedasticity of the residuals. The independent variables were checked for colinearity by examination of tolerance and variation inflation factor.

The independent associations of frequency, duration, and intensity with V˙O_2peak were examined using a general linear model with age as a covariate. Tertiles of total exercise time (<75, 75–149, and >150 min·wk⁻¹) were combined with the three intensity categories (“low,” “moderate,” and “vigorous”), yielding nine separate groups for each sex. Associations with V˙O_2peak were estimated with V˙O_2peak as the dependent variable and total exercise time by intensity groups as a categorical independent variable. All analyses were adjusted for age. In a secondary model, further adjustment was done for BMI, smoking status (“never,” “former,” and “regular”), and occupational PA (“sedentary,” “regular walking,” “regular walking and lifting,” and “heavy manual labor”). Precision of the estimated means and mean differences were assessed by 95% confidence intervals (CIs). Tests for linear trend across total exercise time groups were performed separately for each intensity category. Including an interaction term together with weekly exercise time and relative intensity groups assessed the combined effect of exercise time and intensity in the full age-adjusted model. In a further analysis, exercise categories in accordance with current recommendations, but divided by “vigorous” and “very vigorous” intensity, were entered as a categorical independent variable, and V˙O_2peak or estimated net energy expenditure was entered as a dependent variable. Age was retained as a covariate. The inactive group was set as a reference category. To assess the independent contribution of relative intensity within groups of different total net energy expenditure, we entered V˙O_2peak as a dependent variable, intensity groups as an independent variable with estimated weekly net energy expenditure (kcal·wk⁻¹) and age as continuous variables and covariates. The analyses were then done separately for tertiles of weekly net energy expenditure.

All statistical analyses were performed with PASW Statistics version 18.0 (SPSS Inc., Chicago, IL).

RESULTS

Descriptive characteristics of the participants in the HUNT Fitness Study are presented in Table 1. The mean V˙O_2peak was 44.3 ± 9.3 and 35.9 ± 7.7 mL·kg⁻¹·min⁻¹ for men and women, respectively. In total, 12.5% of the participants were classified as inactive, i.e., reporting exercise less than once a week, and a larger proportion of men than women reported to be inactive (16.7% of men and 8.4% of women).

TABLE 1: Descriptive data for the total, male, and female participants.

A self-reported moderate intensity (12–13 on the Borg scale) corresponded to approximately 70% of V˙O_2peak during the exercise test, whereas vigorous intensity (14–15 on the Borg scale) corresponded to approximately 80%–90% of V˙O_2peak and very vigorous intensity (≥16 on the Borg scale) >90% of V˙O_2peak (Table 2).

TABLE 2: Average percentage of V˙O_2peak by reported relative intensity^a of exercise.

When examining the association between V˙O_2peak and self-reported frequency per week, duration per session, and usual intensity of exercise, adjusted for age, only frequency (P < 0.001) and intensity (P < 0.001) were significantly associated with V˙O_2peak, whereas duration (P = 0.192 and P = 0.783 for women and men, respectively) did not contribute to the explained variance (full model R², 0.41 and 0.45 for women and men, respectively). To comply with the current recommendations, we collapsed duration and frequency and calculated total exercise time in minutes per week. Every 30 min of higher exercise time per week was associated with 0.5 and 0.7 mL·kg⁻¹·min⁻¹ of higher V˙O_2peak for women and men, respectively. For every unit of higher relative intensity during exercise (Borg scale 6–20), V˙O_2peak was approximately 1 mL·kg⁻¹·min⁻¹ higher (B = 0.83 and B = 1.08 for women and men, respectively).

Table 3 and Figure 1 show the mean V˙O_2peak associated with total exercise time in men and women combined with their self-reported relative intensity of exercise. Among men and women who reported low intensity (6–11 on the Borg scale), higher exercise time was not associated with higher V˙O_2peak compared with the inactive groups. Women and men who reported moderate exercise intensity (12–13 on the Borg scale) had a higher V˙O_2peak than the inactive groups (age-adjusted average, 35.2 vs. 32.3 mL·kg⁻¹·min⁻¹ for women and 43.2 vs. 40.1 mL·kg⁻¹·min⁻¹ for men). Also, among men and women who reported moderate intensity, higher exercise time was associated with higher V˙O_2peak (P trend <0.001 for both sexes). Reporting a vigorous relative intensity (14–20 on the Borg scale) was associated with a higher age-adjusted and multiadjusted V˙O_2peak (age-adjusted average, 37.3 and 47.1 mL·kg⁻¹·min⁻¹ for women and men, respectively). With higher exercise time within the vigorous-intensity category, the association with V˙O_2peak was even stronger (P trend <0.001 for both sexes). We also assessed interactions by sex for the association of total exercise time and intensity and found evidence for a significant interaction in women (P < 0.001), but not in men (P = 0.582), as is also visually apparent in Figure 1.

TABLE 3: Age-adjusted and multiadjusted V˙O_2peak in women and men according to self-reported relative intensity^a and total exercise time.

FIGURE 1: Peak oxygen uptake (95% CI) adjusted for age in women (A) and men (B) according to relative intensity of exercise and total exercise duration in minute per week. Low, moderate, and vigorous intensity corresponds to 6–11, 12–13, and 14–20 on the Borg scale 6–20. V˙O_2peak, peak oxygen uptake.

Table 4 shows age-adjusted V˙O_2peak and estimated weekly energy expenditure in men and women who reported “vigorous” and “very vigorous” intensity at different total durations compared with the “moderate intensity–long duration” groups and those who reported a lower exercise level than recommended. Men who reported less than 75 min·wk⁻¹ (mean, 46 min) and “vigorous” intensity (14–15 on the Borg scale) had considerably higher V˙O_2peak than the inactive and “less than recommended” groups. A comparable V˙O_2peak was observed among men who reported more than 150 min of exercise per week (mean, 219 min) at moderate intensity (12–13 on the Borg scale). The estimated net weekly energy expenditure, however, was more than four times higher in the latter group compared with the group with <75 min at vigorous intensity. A higher V˙O_2peak was observed among men who reported very vigorous intensity (≥16 on the Borg scale) and less than 75 min of total exercise time (mean, 47 min) compared with both the moderate-intensity group and the “less than recommended” group.

TABLE 4: Age-adjusted V˙O_2peak in women and men according to relative intensity^a and total exercise time.

For women who reported <75 min (mean, 51 min) at “very vigorous” intensity, V˙O_2peak was comparable with the moderate-intensity group and higher than the “less than recommended” group. Women reporting “vigorous” intensity and <75 min (mean, 49 min) had V˙O_2peak comparable with the “less than recommended” group and slightly lower than the group with moderate intensity above 150 min. Also for women, the estimated weekly energy expenditure in “moderate intensity–high duration” group was approximately three to four times higher compared with the vigorous-intensity groups reporting less than 75 min of total duration. For both sexes reporting “vigorous” intensity within the recommended weekly duration (75–149 min, mean = 112.5 min), V˙O_2peak was comparable with the moderate-intensity groups with approximately 30%–40% lower net energy expenditure.

Men and women reporting to exercise at vigorous intensities (14–20 on the Borg scale) had higher V˙O_2peak than that observed in those reporting to exercise at moderate intensity (12–13 on the Borg scale), even after adjustment for age and estimated weekly net energy expenditure (43.5 vs. 46.1 in men, P < 0.001, and 36.1 vs. 37.0 in women, P = 0.016; Fig. 2). When categorizing participants into quartiles of estimated weekly net energy expenditure, reporting vigorous intensity was associated with slightly higher V˙O_2peak than reporting moderate intensity over all categories among men. The percentage differences were 5.1% (P = 0.008), 2.4% (P = 0.205), and 6.0% (P = 0.004) between moderate and vigorous intensity within the low-, medium-, and high-volume groups, respectively. Among women, a significant difference was observed only in those reporting an estimated weekly net energy expenditure of at least 1250 kcal·wk⁻¹ (4.4% difference between vigorous and moderate intensity, P = 0.023).

FIGURE 2: Peak oxygen uptake (95% CI) adjusted for age in women (A) and men (B) reporting moderate and vigorous intensity at different levels of weekly net energy expenditure. Moderate and vigorous intensity refers to self-reported relative exercise intensity of 12–13 and 14–20 on the Borg scale 6–20. kcal·wk⁻¹, estimated weekly net energy expenditure in kilocalories per week; V˙O_2peak, peak oxygen uptake.

DISCUSSION

To our knowledge, this is the first large population-based study that validates today’s recommendation of PA related to the level of directly measured V˙O_2peak in healthy men and women. We observed that V˙O_2peak among men and women reporting to perform exercise with either “long duration–moderate intensity” or “short duration–vigorous intensity” was similar, both being considerably higher than that observed among individuals reporting to be inactive or performing low-intensity exercise. Therefore, our findings support the notion that the two exercise approaches can be used interchangeably to obtain improved V˙O_2peak. In a previous study from the HUNT Fitness Study population, 35.1 mL·kg⁻¹·min⁻¹ for women and 44.2 mL·kg⁻¹·min⁻¹ for men represented thresholds below which an unfavorable cardiovascular risk profile was apparent (²). Here, we demonstrate that men and women following today’s recommendations of either “long duration–moderate intensity” or “short duration–vigorous intensity” had age-adjusted V˙O_2peak values above these sex-specific thresholds, whereas the inactive group and those reporting less than recommended exercise at low/moderate intensity had average V˙O_2peak below the suggested cutoff values. However, increasing the exercise intensity level to a Borg scale of 16 and above was associated with similar V˙O_2peak in women and slightly higher V˙O_2peak in men with an even lower total duration than recommended. Hence, individuals that reported to perform exercise at very vigorous intensity, regardless of the total time spent, obtained an average V˙O_2peak above the age-adjusted V˙O_2peak thresholds.

We showed that the individuals that followed the recommended exercise paradigm of “long duration–moderate intensity” had a considerably higher net energy expenditure during exercise compared with that observed in those following the “short duration–vigorous intensity” paradigm. Despite this, V˙O_2peak was comparable or higher in the latter group. Hence, we could not support the notion that multiples of intensity and duration are essentially equal. Interestingly, a recent large prospective cohort study showed that approximately 90 min of vigorous-intensity exercise (corresponding to approximately 347 kcal·wk⁻¹) and approximately 250 min of moderate-intensity exercise (corresponding to approximately 770 kcal·wk⁻¹) were associated with 27% and 18% reduced risk of all-cause mortality (²⁴). That study was novel in showing the advantages of vigorous exercise over moderate exercise for a considerably smaller total volume performed. Nevertheless, they conclude that even a smaller than recommended amount of moderate-intensity exercise was sufficient to reduce all-cause and cardiovascular mortality.

To estimate the isocaloric associations of vigorous- and moderate-intensity exercise to V˙O_2peak, we adjusted for weekly exercise energy expenditure. Among men, those reporting to regularly perform vigorous-intensity exercise had higher V˙O_2peak than those reporting moderate-intensity exercises. A similar pattern was observed in women with an estimated energy expenditure of at least 1250 kcal·wk⁻¹. The observed difference between men reporting moderate and vigorous intensity, respectively, was in the range of 2.5%–6%, which is in accordance with the difference observed in randomized controlled trials evaluating the effects of isocaloric training interventions at moderate and vigorous intensity for 8 to 16 wk (^6,8,10). The reason why we found no difference in V˙O_2peak between vigorous and moderate relative intensity among women spending less than 1250 kcal·wk⁻¹ is uncertain. However, we might speculate that less active women tend to overestimate their relative intensity level, which may potentially mask a true difference (⁷).

Our results showed that low-intensity exercise, regardless of the number of hours of exercise, was not associated with higher levels of V˙O_2peak, which is in accordance with several studies (^22,25). Self-reported low intensity (Borg scale 6–11) in the present study corresponded to 40%–60% of V˙O_2peak during the graded exercise test, which is a training stimuli proposed to be too low to elicit changes in V˙O_2peak in healthy persons (^9,22). Furthermore, we observed a linear dose–response pattern with increasing exercise time only among those reporting at least moderate intensity (i.e., 12–13 on the Borg scale). Hence, there seems to be an intensity threshold above which increased total exercise time positively alters V˙O_2peak levels. However, it has been suggested that a threshold of exercise intensity may vary depending on fitness level (^21,25). Our population must be considered relatively fit, and a benefit from low-intensity exercise has been detected in more sedentary populations (²²).

Strengths and limitations

A major strength of the present study is the population-based approach including approximately 4300 participants over a wide age span with objectively measured V˙O_2peak. Furthermore, we believe that the relative intensity measure applied (i.e., the Borg scale rating) is a strength compared with population-based studies considering absolute intensity levels. Most studies showing that low- or moderate-intensity exercise is sufficient to increase or maintain cardiorespiratory fitness as well as reduce risk of disease and mortality have been conducted in older, sedentary, or obese men or women (^1,11,18). Because these studies in general have used a total MET value (i.e., 1 MET ≈ 3.5 mL·kg⁻¹·min⁻¹) as a measure of intensity, we may speculate that the relative intensity of exercise was higher than suggested. Total MET may be a surrogate for fitness rather than unveiling the true exertion of exercise in the individual. The widespread use of absolute intensity may have contributed to a spurious interpretation that low- to moderate-intensity exercise is sufficient to increase fitness and reduce risks in some individuals but not others. In addition, prescribing PA on a relative scale may be more suitable because older and unfit individuals may have problems adhering to a given MET level, which may be close to their maximal aerobic capacity (i.e., 6 MET ≈ 21 mL·kg⁻¹·min⁻¹). Relative intensity by personal perception of exertion relative to the subject’s aerobic fitness may therefore be a more valuable tool for interpretation of intensity of PA in heterogeneous populations.

The nature of self-reported questionnaires implies a risk of misclassification, and several studies have proposed that vigorous activity is more accurately recalled than moderate- and low-intensity activity, which may contribute to the stronger association observed (⁷). Moreover, low-intensity activity may yield health benefits independent of increased V˙O_2peak, which may not be a complete surrogate for health and does not mimic all positive effects of PA.

Another limitation of the present study is the cross-sectional design, which does not allow for conclusions about causality in the associations between parameters of exercise and V˙O_2peak. It is possible that more fit people (i.e., higher V˙O_2peak) are more likely to participate in vigorous intensity and/or more frequent activity. Nevertheless, the present study identifies self-selected exercise patterns associated with high and low fitness in a large population, which is in conformity with current recommendations. Moreover, our findings indicate that the self-reported questionnaire applied was able to assess the associations between exercise and V˙O_2peak in a biologically plausible dose–response manner.

CONCLUSIONS

The present study adds support to the current recommendations by indicating that increased V˙O_2peak can be achieved both through a “long duration–moderate intensity” and a “short duration–vigorous intensity” approach. Our results also suggest that exercising at very vigorous intensity may be beneficial for V˙O_2peak even with a considerably lower total exercise time than expressed in today’s recommendations. This information is important and should be incorporated in future guidelines.

The study was supported by grants from the K.G. Jebsen Foundation, Norwegian Council on Cardiovascular Disease (B. N.), and the Norwegian Research Council Funding for Outstanding Young Investigators (U. W.). Dr. Janszky is supported by the liaison committee between the Central Norway Regional Health Authority and the Norwegian University of Science and Technology, by the Swedish Research Council, and by the Swedish Council of Working Life and Social Research. There are no further disclosures to report and no conflicts of interest. The Nord-Trøndelag Health Study (the HUNT study) is a collaboration between the HUNT Research Centre (Faculty of Medicine, Norwegian University of Science and Technology), the Nord-Trøndelag County Council, and the Norwegian Institute of Public Health.The results of the present study do not constitute endorsement by the American College of Sports Medicine.