BERTHEUSSEN, GRO F.1; ROMUNDSTAD, PÅL R.2; LANDMARK, TORMOD3; KAASA, STEIN4; DALE, OLA5; HELBOSTAD, JORUNN L.6
Health-related quality of life (HRQoL) is an umbrella term for multiple outcomes related to self-reported physical, emotional, and social functioning (25) and is a goal for most health interventions (26). A person's evaluation of HRQoL involves perceived and appraised function (25). More knowledge is required on modifiable lifestyle factors, such as physical activity (PA), that might change health or functioning.
PA is defined as any bodily movement produced by contraction of skeletal muscle that increases energy expenditure (24). The level of PA is typically assessed in terms of frequency, duration, and intensity (2). PA has been shown to be associated with reduced risk of cardiovascular diseases, thromboembolic stroke, hypertension, type 2 diabetes mellitus, obesity, osteoporosis, colon cancer, breast cancer, depression, and anxiety by improving physiological and psychological processes (15,33).
The exercise dimension of PA was assessed as self-reported frequency, duration, and intensity. Exercise is a subcategory of PA that is "planned, structured, and repetitive and purposive in the sense that the improvement or maintenance of one or more components of physical fitness is the objective" (11). It refers to PA performed during leisure time with the primary purpose of improving or maintaining physical fitness, physical performance, or health (http://www.health.gov/paguidelines/guidelines/default.aspx).
A systematic review from the United States has found leisure time PA being relatively stable or slightly increasing during the last 50 yr, although there has been decline in work-related PA, active transportation, and home-based activity and increasing sedentary activity, thus indicating an overall decrease in total PA (9). The prevalence of leisure time physical inactivity in the United States from 1994 to 2004 showed a decline from 29.8% to 23.7%. The largest decline of inactivity was among males age 50-59 yr and females age 60-69 yr. In 2004, the highest prevalence of inactivity was in those age ≥70 yr in both males (30%) and females (40%) (16).
Current international recommendations on PA for health benefits in adults are at least 150 min·wk−1 of moderate-intensity PA, corresponding to 3-6 METs, or a total of 75 min·wk−1 of vigorous-intensity aerobic PA, corresponding to >6 METs, or combinations constituting an equivalent MET dose. A subsequent updated recommendation appends vigorous-intensity aerobic PA for a minimum of 20 min for 3 d·wk−1 or at least 30 min moderate-intensity PA 5 d·wk−1 or equivalent combinations. Aerobic activity should be performed in episodes of at least 10 min, and preferably, it should be spread throughout the week (12). The proportion of the population being active at a recommended level varies from 20% to 50% in different studies, depending on assessment methods and study populations (23).
Cross-sectional studies in the general adult population 20-65 yr old (8,27,34) and people > 65 yr old (1,8) have concluded that there is better HRQoL in more active people compared with the more inactive group. General population studies have demonstrated a positive association between PA of at least moderate intensity and physical health for adults < 65 yr old (4,27,34,37). Among older adults, PA performed daily at low-to-moderate intensity may have shown stronger associations with QoL than vigorous exercise 3 d·wk−1 (26). The amount of PA was reported to have positive effects on HRQoL in both females and males, but one study showed that a maximum intensity of PA had positive effects on most of the HRQoL scales in females but not in males (21). However, comparison of studies is complicated since the cutoff value between active and inactive people varied between studies, as did assessment methods for HRQoL and PA. The dimensions of PA such as type, frequency, duration, or intensity, which are closest related to HRQoL are still uncertain (3,31,37).
The aim of this study was to investigate the association between HRQoL and self-reported exercise of a general adult population ≥20 yr. More specifically, we wanted to assess how different dimensions of PA (frequency, duration, or intensity) were associated with physical and mental health aspects of HRQoL in younger (<65 yr) and older (≥65 yr) females and males, and how socioeconomic status, education, and disease influence the relationship. We hypothesized that 1) the physically active part of the population reports better physical and mental health independent of gender and age; 2) frequency, duration, and intensity of PA are associated with physical and mental health in a curvilinear manner; and 3) exercise dimensions of PA have a stronger association with physical health than mental health.
The PainHUNT study was a part of the third Nord-Trøndelag Health Study, HUNT 3 (http://www.ntnu.no/hunt/english) (22). In HUNT 3, a group of 50,839 of 94,194 people ≥ 20 yr old participated (54% response rate). The demographics of the HUNT population are similar to the average for the Norwegian population, except from slightly lower average income and education. The PainHUNT population is similar to the HUNT 3 population with respect to gender and age distribution, body mass index (BMI), type of work, smoking habits, disability pension, and annual income. Education is somewhat higher in the PainHUNT population, that is, 34% versus 26% with college/university education, whereas fewer are living alone or as single parent, that is, 26% versus 34%, respectively, for the PainHUNT and the HUNT 3 population. The HUNT population is stable, with a net out-migration of 0.3%, and is homogeneous (97% Caucasian), making it suitable for epidemiological studies (12). A random sample of 6419 HUNT 3 participants from Levanger and Verdal (with a total of 23,358 inhabitants and a response rate of 51%) were invited to participate in the PainHUNT. These two municipalities were considered to be representative for the population in Nord-Trøndelag County. Informed consent was obtained from all participants, and the study was approved by the Regional Committee for Medical and Health Research Ethics and by the Norwegian Social Data Inspectorate.
HRQoL was measured with the SF-8 Health Survey (1-wk recall version), a short version of the SF-36 version 2.0. SF-8 is recommended for use in large population surveys and has been translated into Norwegian (35,36). SF-8 consists of single items/scales that describe each of the eight dimensions of health in the SF-36: general health, physical function, role physical, bodily pain, vitality, role emotional, mental health, and social function. The Physical (PCS-8) and Mental (MCS-8) Summary Scales were computed according to the SF-8 manual (35). The scales are constructed so that, in the general US population, they have a mean of 50 and SD of 10 (35). A higher score indicates better functioning and less symptoms.
In this study, the frequency of PA was defined as the number of times the person exercised last week: never, one to three times, four to six times, and daily. Duration was described as the average exercise (min) per session last week: <15 min, 15-29 min, 30-60 min, and >60 min. Intensity was assessed by the Borg Ratings of Perceived Exertion scale (the Borg RPE scale) (7) where we asked for average intensity of exercise last week. The RPE scale ranges from 6 (very, very light/no exertion at all) to 20 (extremely hard/maximal exertion). Scores from 6 to 11 were classified as light intensity, scores from 12 to 13 were classified moderate intensity, and scores from 14 to 20 were classified as high intensity according to recommendations (7,32). The questions of frequency, duration, and intensity addressed exercise as follows: "for example going for walks, skiing, swimming or training/sport." The HUNT 1 PA questionnaire including items on frequency, duration, and intensity has earlier proved to be an appropriate tool for use in epidemiological studies, having acceptable test-retest reliability in adult males, with κ values ranging from 0.52 to 0.77 and a significant correlation with V˙O2max (r = 0.31 and above) (18). However, this study has a somewhat different response scale regarding the categories of answers for the frequency of PA, where the validated version used five categories (never, less than once a week, once a week, two to three times per week, almost everyday), whereas the PainHUNT study used the four categories described above. In addition, the PainHUNT study used the well-validated Borg scale for assessing intensity of PA instead of the HUNT 1 question, "How hard do you push yourself?", which had three categories of answers: without breaking into a sweat or losing my breath, I lose my breath and break into a sweat, to near exhaustion. The reason for using the Borg scale was to have a more detailed assessment of intensity.
PA is presented separately for frequency, duration, and intensity. Participants who reported "never exercise" for frequency, independent of whether or what they had reported for duration and intensity, and participants reported exercise ≥1 d·wk−1 of <15 min each session were classified as "no exercise." In addition, we classified PA based on the combination of certain frequency, duration, and intensity of PA corresponding to the best fit of the PA recommendations from the American College of Sports Medicine (12). Participants who reported exercise ≥1 d·wk−1 of at least 15 min and at moderate or high intensity (≥12 on the Borg scale) were classified as having "PA level as recommended." Participants were classified as having "PA level less than recommended" if they replied one of the following: never exercise, exercise <15 min or low-intensity exercise (≤11 on the Borg scale).
Information on diseases was based on self-report. Participants were asked if they have had any diseases during the past 12 months. The reported diseases were classified as organ system (cardiovascular, lung, gastrointestinal, kidney, diabetes, neurological, and cancer diseases), musculoskeletal (arthritis, musculoskeletal disorders, osteoporosis, fractures, nerve injury, and other injuries), or psychiatric.
Variables from the HUNT 3 were collected from self-administered questionnaires and included gender, age, smoking habits, and cohabitation (http://www.ntnu.no/hunt/english) (22). Data on education (dated 2008) were obtained from the National Educational Database, and the classification of education level was based on compulsory school, high school, and college/university. Data on annual household income (dated 2007) were obtained from Statistics Norway, and data on disability pension (dated 2008 or earlier) were obtained from the National Insurance Register in Norway.
Two months after participation in the HUNT 3 postal survey, the PainHUNT questionnaire was sent out. The PainHUNT data used in this article were collected from September 2007 to February 2008 through a postal questionnaire. A reminder was sent after approximately 1 month to participants who did not return the questionnaire.
Statistical analyses were conducted using STATA version 10 (StataCorp, College Station, TX) and SPSS version 16.0 for Windows (Chicago, IL). Participants with missing values on frequency, duration, or intensity of PA and/or any of the SF-8 items were excluded from the analysis (n = 282).
Analyses of associations between PA and PCS-8 and MCS-8 were carried out separate for females and males and for people <65 and ≥65 yr old. PCS-8 and MCS-8 scores were presented as mean scores adjusted for mean age within each age group because different age groups have different norm-based means of PCS-8 and MCS-8 (35). In addition, PCS-8 and MCS-8 scores adjusted for the combination of mean age, education, and annual household income with 95% confidence interval (CI) are presented.
Associations between PA and physical and mental health were estimated by using univariate general linear modeling, with PCS-8 or MCS-8 as the dependent variables and frequency, duration, and intensity as independent variables. Adjustments for possible covariables, such as the presence of diseases, BMI, smoking habits, cohabitation, and disablement were performed. Only age-adjusted values and values adjusted for the combination of age, education, and annual household income were presented because these variables have been related to HRQoL and to PA in previous studies (27,34,37). When adjusting for possible covariables, the same categorizations as those presented in Table 1 were used. To test for potential interactions with gender and age in the association between PA and HRQoL, likelihood ratio tests were performed in which models with and without the interaction terms were compared. We used P < 0.05 as significant in the analysis of interactions.
Of the 4782 available PainHUNT respondents (75% of the invited), a total of 4500 participants were included (70% of the invited) after exclusion owing to missing values on frequency (n = 116), duration or intensity (n = 38) of exercise, or any of the eight SF-8 items (n = 180, some missed data on more than one of these variables).
There are only small differences in frequency, duration, and intensity of PA between the HUNT 3 and PainHUNT populations. In both populations, 76% reported having at least 30 min of daily work-related or leisure time PA.
Females comprise 56% of the PainHUNT population sample (Table 1). Participants were age 19-91 yr, with a mean age of 53.1 yr: 51.9 yr for females and 54.6 yr for males. Nearly a quarter of the PainHUNT population is >65 yr.
More males than females reported no exercise (24% and 19%, respectively). Males reported a small predominance regarding the highest category of answer for frequency (8% in males and 7% in females), duration (20% in males and 15% in females), and intensity (31% in males and 29% in females). On the other hand, females reported a predominance in the middle categories of answers for exercise one to three times last week (55% in females and 51% in males) and of 30-60 min per session (49% in females and 41% in males). Regarding intensity of exercise, more females than males reported light intensity (21% and 16%, respectively) and moderate intensity (31% and 29%, respectively).
Table 1 shows that approximately 40% of both females and males are less physically active than recommended. These reported <150 min last wk of moderate-intensity exercise or <75 min last week of high-intensity exercise or less than an equivalent combination of moderate- and high-intensity exercises. The highest proportion of those being less physically active than recommended tended to be ≥65 yr, have organ system diseases, a BMI ≥ 30 kg·m−2 (obese), be current smokers, have only finished compulsory schooling, be on disability pension, and have an annual household income of ≤32,000 euros. Mean PCS-8 and MCS-8 is 1.4 scale points higher in males than in females, with the highest PCS-8 in people from 20 to 44 yr and highest MCS-8 in people ≥65 yr. The no-diseases group reported a substantially higher PCS-8 (about 9 points) and somewhat higher MCS-8 than the groups with diseases. MCS-8 and PCS-8 are almost equal in the no-diseases group, whereas MCS-8 is higher than PCS-8 in the organ system and musculoskeletal group and lower in the psychiatric disease group.
The age-adjusted mean of PCS-8 and MCS-8 by recommended PA level is given in Table 2 according to gender and age group. The PCS-8 and MCS-8 are higher in people reporting exercise at or above the recommended level compared with the exercise below this level irrespective of gender and age group. We found no substantial interactions between recommended PA and age groups for PCS-8 (P = 0.4 in females and P = 0.09 in males) or for MCS-8 (P = 0.7 in females and P = 0.8 in males). Similarly, no indications of interactions are found between recommended PA level and gender for PCS-8 or with respect to MCS-8 (all P values for interaction > 0.2).
For PCS-8, we found evidence for interactions between frequency, duration, and intensity of PA and age groups with respective P values for interactions of 0.02, 0.05, and 0.003 for females and respective P values for interactions of 0.1, 0.02, and 0.006 for males. For MCS-8, we found no clear evidence for interactions between frequency, duration, and intensity of PA and age groups (P values for interaction between 0.1 and 0.8). Similarly, no significant interactions are observed between the levels of exercise and gender (P values for interaction between 0.07 and 0.6) except for PCS-8 for duration of PA (P value for interaction of 0.01) in the youngest age group. Although no interactions were found for gender and age in the association between recommended PA and PCS and MCS, we did find interactions in the associations between frequency, duration, and intensity of PA and PCS and MCS. To enhance comparability and because there are obvious biological differences between genders and age groups, we decided to present all further results stratified for gender and age.
In general, the participants reported a higher MCS-8 than PCS-8, and males reported slightly higher PCS-8 and MCS-8 than females, with the most pronounced gender differences for people reporting no exercise (Tables 3-6). Thus, differences of PCS-8 between genders in the analyses adjusted for age, education, and annual household income in the no-exercise group are 3.5 points for people <65 yr (Table 3) and 2.2 points for people ≥65 yr (Table 4). For MCS-8, the corresponding gender differences in scores are 1.9 points for both age groups (Tables 5 and 6). Mean PCS-8 is slightly lower and mean MCS-8 is slightly higher for older compared with younger people. On the whole, mean PCS-8 and mean MCS-8 increase with increasing exercise level for both females and males in both age groups. Exceptions were PCS-8 for females and males in the oldest age group who reported daily exercise and high-intensity exercise (Table 4) and MCS-8 for just males in the oldest age group reporting daily exercise, where there is a slightly lower score (Table 6). The largest difference in PCS-8 between exercise levels was seen among those reporting no exercise compared with any level of exercise for frequency, duration, and intensity. The differences are more pronounced in females than in males and in the oldest compared with the youngest age group. MCS-8 shows the same tendency but is less pronounced than for PCS-8. There are only small changes in scores when adjusted for age, education, and household income compared with adjustments for age only (at most 0.5 points). Adjusting for the presence of disease, BMI, smoking habits, cohabitation, and disablement only changes the results by a maximum of 1.5 points (results not tabulated).
In this population-based cross-sectional study, PA was consistently associated with better physical health and mental health in males and females and in young and old adults for all aspects of PA (frequency, duration, and intensity). We found only small gender differences in the associations between PA and both physical health and mental health. The associations between the level of exercise and physical health were stronger in those who are ≥65 yr old compared with younger people, although no such differences were found for mental health. The stronger positive associations between PA and physical health are in line with findings from other cross-sectional studies, whereas longitudinal studies have found stronger positive associations with mental health (4,27,34,37).
Mean PCS-8 and MCS-8 rose with increasing level of frequency, duration, and intensity of PA in a curvilinear manner. In general, the associations tended to level off for the highest level of frequency. This may indicate that duration and intensity of PA were more important for physical and mental health at high exercise levels than the frequency of PA. One exception was the high intensity of PA in the oldest age group where the PCS-8 curve leveled off. This may indicate that a high intensity of PA was less important for physical health in the oldest age group. The associations were most pronounced in the oldest age group, demonstrated by a significant interaction between PA and age group with physical health (except for frequency of PA in men) but not with mental health. In comparison with these findings, a cross-sectional U.S. population study of 175,850 adults found a bell-shaped association curve between PA and HRQoL. Compared with our findings, they found a more pronounced reduction in HRQoL for moderate to high daily PA of ≥90 min (8). The lack of substantial decrease in HRQoL for the highest exercise levels in our study may possibly be explained by the fact that we have looked at the three dimensions of PA, namely, frequency, duration, and intensity, separately, whereas the other study used a composite measure of PA. In our opinion, which of the aspects of PA that is most closely related to physical and mental health in a general population of both younger and older females and males have received little attention in the literature until now. This was why we chose to look at the three dimensions of PA separately. Our results may suggest that duration and intensity were more important for physical health than frequency in the youngest age group, whereas duration was more important than frequency and intensity in the oldest age group. For mental health, duration and intensity were more important than frequency, particularly for older people.
In the present study, the differences in PCS-8 and MCS-8 between people who are active at the recommended level and those who are less active were somewhat smaller than the differences in PCS-8 and MCS-8 between those reporting no exercise and those who are reporting any level of exercise. The reporting of better health in people who are active at or above the recommended level is in accordance with previous studies showing positive associations between PA and morbidity and mortality (5,6). In addition, for these outcomes, the intensity of PA seems to be of greatest importance (5,38).
For quality-of-life measures, half an SD difference has been interpreted to be a clinically meaningful difference. According to the SF-8 manual, a 0.3 SD difference is suggested as clinically meaningful difference at group level, and 0.5 SD at an individual level (29,30). This suggests that the observed differences between inactive and active people in physical health are clinically meaningful in the present study independent of the level of exercise. Similarly, the differences in physical health between people who are less active than recommended and those who are active as recommended are also clinically important. For mental health, the estimated differences in the present study are less prominent. However, using SD when interpreting clinically meaningful differences should be interpreted with care.
Our results, that males scored higher than females on both PCS-8 and MCS-8, are in accordance with the norms for the general U.S. population (35) and normative SF-36 data from the general Norwegian population (20). However, there was no clear evidence of interactions between gender and PA for PCS-8 or MCS-8, suggesting that the associations between PA and HRQoL were similar in magnitude for females and males.
Among those who reported "no exercise," 4% more females than males reported having diseases. Females also reported a higher prevalence of diseases in every exercise category. This is in line with previous studies showing higher rates of morbidity, depression, and psychological distress in females compared with males (10,19,28). This suggests that the gender difference in self-reported health may be explained by the higher prevalence of diseases among females. However, when adjusting for diseases in combination with age (dichotomized as no diseases or one or more diseases), the gender differences do not diminish.
In this study, males have a minor domination in the lowest and the highest exercise levels and females in the middle exercise levels. Regarding frequency of exercise, fewer females (17%) than males (22%) reported "never exercise" last week. In HUNT 1 (1984-1986), the first population survey on the same population, more females than males reported "never PA last week" (14.8% vs 13.7% in males), whereas in HUNT 2 (1995-1997), there was no distinction between the genders (8.3%) (17). The tendency toward a decrease in the proportion of females reporting "never PA" and an increase in the portion of males reporting "never PA" from HUNT 1 to HUNT 2 may suggest a change in the activity patterns in females and males over time.
We found that 60% of our subjects were as active as recommended by international guidelines. This is somewhat higher than comparable figures from most other studies (23). A Norwegian population study on PA in adults, using the International Physical Activity Questionnaire, found 43% of the males and 32% of the females to be active at or above the recommended level. When objectively assessing PA on the same study population by an activity monitor (ActiGraph GT1M), it was found that 22% of females and 18% of males were active at or above the recommended level (3). Various methods for assessing PA and differences in the criteria to classify "as recommended" may explain some of the differences in the results from these studies. The high portion within the category "PA level as recommended" in this study may be explained by the conservative cutoff point based on the best fit of the combination of the categories of answers in the questionnaire and the PA recommendations from the American College of Sports Medicine (12). The consequence of the conservative cutoff point is a possible "misclassification" and may, in accordance to previous studies, be an underestimation of PCS-8 and MCS-8 scores in the group being "active as recommended." It may thus have given an underestimation of the effect size of PCS-8 and MCS-8 between those being active and not being active as recommended.
PA was registered by assessing frequency, duration, and intensity of exercise. Measuring exercise as a proxy for PA has some limitations. PA not performed during leisure time and without the purpose of improving or maintaining physical fitness or performance has not been registered, for example, transportation and household activities.
The PainHUNT population is a random sample from Nord-Trøndelag County, which is a region with demographics similar to the Norwegian population. The Norwegian population is characterized as quite uniform, with fewer social inequalities than many other countries. Norway has a well-established social security system with family income supplement and a well-organized public health care system. The population as a whole is generally in good health and condition. This may explain the small deflections in physical and mental health in relation to PA. The PainHUNT sample size and the fact that the PainHUNT population is representative for the HUNT 3 population strengthen the external validity of the study. A study of nonparticipants in the HUNT 1 study did not find evidence of selection in health measures in the young age groups, whereas old nonparticipants had more health problems than participants of same age (14). In the region of recruitment to the PainHUNT sample, only 31% who participated in the HUNT 3 study were in the age group 20-29 yr, with an increasing participation up to 60-69 yr, where 71% participated. Accordingly, the most selected age group has the highest participation and may bring along an underestimation of health problems in our sample. Still, adjustment for the presence of disease did not change our results. A strength of the present study is that the answers of exercise and HRQoL were all corresponding to the same single week and that recall bias may be lower when mapping 1 wk compared with a longer period (3). However, the short time frame may be more sensitive to seasonal or individual basis (climatic factors, infections, travel, vacations, etc.). The inclusion period lasted from September 2007 to February 2008 and constituted five mailings. However, there were no differences regarding the answers of exercise, which may attribute to the point in time of mailings. Because of the observational nature of this cross-sectional study, we cannot rule out potential effect from reverse causality or unmeasured confounding. Thus, we were only able to estimate the associations between PA and physical and mental health and not the causal effects.
The present study suggests that PA is associated with better physical and mental health in a curvilinear manner. In particular, PA is associated with physical well being in the age group ≥65 yr. The largest age-adjusted differences in physical and mental health scores were between those reporting "no exercise" and any degree of exercise. Accordingly, the present study supports the view that people with any degree of exercise may have better physical health and slightly better mental health than those with "no exercise," also after adjusting for factors related to PA and health. On the whole, we found the same relation between HRQoL and frequency, duration, and intensity of PA, suggesting that future studies on PA and HRQoL should use several dimensions of PA. To better assess the direction of these associations and whether changes in PA can explain changes in HRQoL over time, longitudinal studies are needed.
The Nord-Trøndelag Health Studies (The HUNT Study, including the HUNT 3 study) is a collaboration between HUNT Research Centre (Faculty of Medicine, Norwegian University of Science and Technology, NTNU), Nord-Trøndelag County Council, and the Norwegian Institute of Public Health.
This work has been funded by the Research Council of Norway.
The authors declare no conflict of interest.
The results of the present study do not constitute endorsement by the American College of Sports Medicine.
1. Acree LS, Longfors J, Fjeldstad AS, et al. Physical activity is related to quality of life in older adults. Health Qual Life Outcomes
2. Ainsworth BE, Macera CA, Jones DA, et al. Comparison of the 2001 BRFSS and the IPAQ Physical Activity Questionnaires. Med Sci Sports Exerc
3. Anderssen SA, Hansen BH, Kolle E, et al. [Norwegian Directorate of Health. Physical Activity among Adults and Elders in Norway: A Report of the Public Health Department]
. Oslo (Norway): Norwegian Directorate of Health; 2009. p. 105.
4. Bize R, Johnson JA, Plotnikoff RC. Physical activity level and health-related quality of life in the general adult population: a systematic review. Prev Med
5. Blair SN, Connelly JC. How much physical activity should we do? The case for moderate amounts and intensities of physical activity. Res Q Exerc Sport
6. Blair SN, Lamonte MJ, Nichaman MZ. The evolution of physical activity recommendations: how much is enough? Am J Clin Nutr
7. Borg G. Borg's Perceived Exertion and Pain Scales
. Champaign (IL): Human Kinetics; 1998. p. 104.
8. Brown DW, Brown DR, Heath GW, et al. Associations between physical activity dose and health-related quality of life. Med Sci Sports Exerc
9. Brownson RC, Boehmer TK, Luke DA. Declining rates of physical activity in the United States: what are the contributors? Ann Rev Public Health
10. Carmona L, Ballina J, Gabriel R, Laffon A. The burden of musculoskeletal diseases in the general population of Spain: results from a national survey. Ann Rheum Dis
11. Caspersen CJ, Powell KE, Christenson GM. Physical activity, exercise, and physical fitness: definitions and distinctions for health-related research. Public Health Rep
12. Haskell WL, Lee IM, Pate RR, et al. Physical activity and public health: updated recommendation for adults from the American College of Sports Medicine and the American Heart Association. Med Sci Sports Exerc
13. Holmen J, Midthjell K, Krüger Ø, et al. The Nord-Trøndelag Health Study 1995-97 (HUNT 2): objectives, contents, methods and participation. Norsk Epidemiol. 2003;13(1):19-32.
14. Holmen J, Midthjell K, Forsen L, Skjerve K, Gorseth M, Oseland A. A health survey in Nord-Trondelag 1984-86: participation and comparison of attendants and non-attendants [in Norwegian]. Tidsskr Nor Laegeforen
15. Kesaniemi YK, Danforth E Jr, Jensen MD, Kopelman PG, Lefebvre P, Reeder BA. Dose-response issues concerning physical activity and health: an evidence-based symposium. Med Sci Sports Exerc
. 2001;33(6 suppl):S351-8.
16. Kruger J, Ham SA, Kohl MS. Trends in leisure-time physical inactivity by age, sex, and race/ethnicity-United States, 1994-2004. MMWR Morb Mortal Wkly Rep
17. Kurtze N, Gundersen KT, Holmen J. Self-reported physical activity in 1984-86 and 1995-97-the Nord-Trøndelag Health Study (HUNT) [in Norwegian]. Norsk Epidemiol
18. Kurtze N, Rangul V, Hustvedt BE, Flanders WD. Reliability and validity of self-reported physical activity in the Nord-Trondelag Health Study: HUNT 1. Scand J Public Health
19. Leach LS, Christensen H, Mackinnon AJ, Windsor TD, Butterworth P. Gender differences in depression and anxiety across the adult lifespan: the role of psychosocial mediators. Soc Psychiatry Psychiatr Epidemiol
20. Loge JH, Kaasa S. Short Form 36 (SF-36) health survey: normative data from the general Norwegian population. Scand J Soc Med
21. Morimoto T, Oguma Y, Yamazaki S, Sokejima S, Nakamayma T, Fukuhara S. Gender differences in effects of physical activity on quality of life and resource utilization. Qual Life Res
23. Oja P. Descriptive epidemiology of health-related physical activity and fitness. Res Q Exerc Sport
24. Pate RR, Pratt M, Blair SN, et al. Physical activity and public health. A recommendation from the Centers for Disease Control and Prevention and the American College of Sports Medicine. JAMA
25. Rejeski WJ, Brawley LR, Shumaker SA. Physical activity and health-related quality of life. Exerc Sport Sci Rev
26. Rejeski WJ, Mihalko SL. Physical activity and quality of life in older adults. J Gerontol A Biol Sci Med Sci
27. Riise T, Moen BE, Nortvedt MW. Occupation, lifestyle factors and health-related quality of life: the Hordaland Health Study. J Occup Environ Med
28. Seedat S, Scott KM, Angermeyer MC, et al. Cross-national associations between gender and mental disorders in the World Health Organization World Mental Health Surveys. Arch Gen Psychiatry
29. Sloan JA, Cella D, Hays RD. Clinical significance of patient-reported questionnaire data: another step toward consensus. J Clin Epidemiol
30. Sloan JA, Vargas-Chanes D. Detecting worms, ducks, and elephants: a simple approach for defining clinically relevant effects in quality-of-life measures. J Cancer Integr Med
31. Tessier S, Vuillemin A, Bertrais S, et al. Association between leisure-time physical activity and health-related quality of life changes over time. Prev Med
32. US Department of Health and Human Services. Physical Activity and Health: A Report of the Surgeon General
. Atlanta (GA): US Department of Health and Human Services; 1996. p. 278.
33. von Huth SL, Borch-Johnsen K, Jorgensen T. Commuting physical activity is favourably associated with biological risk factors for cardiovascular disease. Eur J Epidemiol
34. Vuillemin A, Boini S, Bertrais S, et al. Leisure time physical activity and health-related quality of life. Prev Med
35. Ware JE, Kosinski M, Dewey JE, Gandek B. How to Score and Interpret Single-Item Health Status Measures: A Manual for Users of the SF-8 Health Survey
. 2nd ed. Boston (MA): Quality Metric, Inc., Lincoln, RI, Health Assessment Lab; 2001. p. 238.
36. Ware JE Jr. SF-36 health survey update. Spine
37. Wendel-Vos GC, Schuit AJ, Tijhuis MA, Kromhout D. Leisure time physical activity and health-related quality of life: cross-sectional and longitudinal associations. Qual Life Res
38. Wisloff U, Nilsen TI, Droyvold WB, Morkved S, Slordahl SA, Vatten LJ. A single weekly bout of exercise may reduce cardiovascular mortality: how little pain for cardiac gain? The HUNT study, Norway. Eur J Cardiovasc Prev Rehabil