Leisure-time physical activity (PA) has been established to be related to more years lived without disability. However, less is known about the relationship between occupational PA and disability in old age. The aim of the study was 1) to investigate whether midlife occupational PA is related to late-life disability, and 2) to test the hypothesis that the association differs according to the occupational categories of blue and white collar work.
The study population was derived from the Swedish National Study on Aging and Care, and consisted of a random sample of 1804 subjects aged 72 and above. The association of occupational PA during the longest held occupation with disability in old age was determined using logistic regression.
There was no significant relationship between occupational PA and disability in personal or instrumental activities of daily living (ADL) after controlling for demographic and health-related factors. However, in stratified analyses moderate levels of occupational PA was associated with a lower odds ratio of dependency in personal ADL amongst white collar workers, compared to low level of occupational PA (OR = 0.34 95% C1 0.12–0.98).
Moderate levels of midlife occupational PA were associated with a decreased risk of personal ADL disability in old age among white collar workers, but not among blue collar workers. Our results highlight the importance of encouraging white collar workers to engage in physical activity during or outside work hours.
Citation: Rydwik E, Welmer A-K, Angleman S, Fratiglioni L, Wang H-X (2013) Is Midlife Occupational Physical Activity Related to Disability in Old Age? The SNAC-Kungsholmen Study. PLoS ONE 8(7): e70471. https://doi.org/10.1371/journal.pone.0070471
Editor: Jose Vina, University of Valencia, Spain
Received: April 24, 2013; Accepted: June 19, 2013; Published: July 30, 2013
Copyright: © 2013 Rydwik et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: This work was supported by the Swedish Ministry of Health and Social Affairs; the Swedish Research Council; Swedish Council for Working Life and Social Research; the American Alzheimer foundation; the Alzheimer foundation, Sweden; the Swedish Brain Power program; the Fredrik and Ingrid Thurings foundation; Gun and Bertil Stohnes foundation; Gamla Tjänarinnor foundation; Stockholm County Council; Söderström-Königska foundation; Handlanden Hjalmar Svenssons Forskningsfond; the Konung Gustaf V:s och Drottning Victorias Frimurare foundation, the Board of Research at Karolinska Institutet; the Centre for Health Care Science at Karolinska Institutet. The authors are not employed by the funders. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Accumulated evidence has shown that leisure-time physical activity (PA) has a beneficial effect on various health outcomes. Specifically, non-occupational PA has been related to more years lived without disability, and higher levels of non-occupational PA has been shown to further increase the number of disability-free years and mortality , . However, studies concerning occupational PA are limited and have shown contradictory results. Both low levels of leisure-time and occupational PA in middle-aged adults have been associated with excessive risk of ischemic heart disease . Moderate to high levels of occupational PA have been suggested to protect against hip fracture , heart failure , myocardial infarction , and mortality –, although the protective effect of occupational PA was weaker compared to leisure-time PA –, . Yet in another study occupational PA was not associated with future disability among middle-aged adults with self-reported arthritis over a 2-year follow-up . On the other hand, high occupational PA has been related to an increased risk of cardiovascular mortality in middle-aged men with low and moderate aerobic fitness, but not among men with high level of aerobic fitness .
To date, population-based evidence on the relationship between midlife occupational PA and disability in old age is scarce. Strenuous occupational PA in competitive or recreational middle-aged runners has been associated with a greater likelihood of disability in their sixties . In addition, it is known that blue collar workers are more likely to have a higher total PA level  and are at higher risk of developing arthritis or receiving a disability pension – and of developing disability and mobility limitation in old age compared to white collar workers –. Therefore, the association between occupational PA and late-life disability may differ among blue collar workers and white collar workers. To our knowledge, no studies have examined occupational PA in white collar workers in relation to disability in old age.
Given this evidence, we hypothesize that the association between occupational PA and disability in old age differs for white and blue collar workers. This study aimed: 1) to investigate whether midlife occupational PA is related to disability in old age, and 2) to test the hypothesis that the association between occupational PA and late-life disability differ according to the occupational categories of blue and white collar work.
The Swedish National Study on Aging and Care in Kungsholmen (SNAC-K) is an ongoing longitudinal study in  Kungsholmen, a geographically-defined urban neighborhood of Stockholm, Sweden. The SNAC-K study population consists of a random sample of people within specific age cohorts, as described below, living either at home or in institutions. The baseline data collection was conducted during March 2001–August 2004. Eleven age cohorts were chosen with different intervals, with six years between groups in the younger cohorts (60, 66, 72 and 78 years old), and three years in the older cohorts (81, 84, 87, 90, 93, 96, 99 and above). The study population in SNAC-K has been enlarged by doubling the younger cohorts (aged 60–78 years), and an over-sampling of the 90+ population, due to the specific interest in longitudinal evaluation of the health and needs of care in the oldest-old. A total of 5111 persons were selected to be invited for participation. Of these 4590 were alive and eligible to participate (200 dead, 262 not able to be contacted, 4 deaf, 23 did not speak Swedish, 32 had moved), and of these 3363 (73.3%) participated at the baseline examination (1227 refused). The age groups 60 and 66 (n = 1304) were excluded from this analysis because they were either currently working or had only recently stopped working. Of the remaining 2059 participants, there were missing data for 248 participants regarding occupational PA and for two participants on instrumental activities of daily living (ADL), leaving 1809 participants in the analyses. There were no significant differences between the participants that had missing data and those included in the analyses in terms of personal and instrumental ADL, age, leisure-time PA, education, and socioeconomic position. However, participants with missing data were significantly more likely to be female, and to have multimorbidity (defined as two or more chronic diseases) and a lower cognitive level.
The study was approved by the Regional Ethical Review Board in Stockholm, Sweden. Informed consent in writing was obtained from all participants or their next of kin.
The baseline data were collected through interviews, clinical examinations, testing and self-administered questionnaires performed by nurses, medical doctors and psychologists. For those who agreed to participate but were unable to come to the research center, home visits were conducted.
The subjects were asked if they independently managed both personal activities of daily living (PADL; dressing, hygiene, bathing/showering) and instrumental activities of daily living (IADL; cooking, cleaning, and running errands). If they answered no to any of these questions, they were coded as dependent in PADL or IADL, respectively. All subjects that lived in institutions were coded as dependent in IADL. If the subjects couldn’t provide information, a proxy, usually a next-of-kin was asked.
Occupational physical activity.
The respondents were asked about the five longest previously held occupations. The subjects rated the PA level for each occupation according to seven levels from very light to very strenuous/full time. The occupational PA level was categorized into three groups in the later analyses: light (very light and light); moderate; and strenuous (strenuous and very strenuous some hours a day or full time). The occupational PA level during the longest held occupation was used in the analyses. If the subjects could not give information about time periods but gave information on type of work and PA level, the most recent work before retirement was used in the analyses.
Data regarding age, gender and highest level of formal educational level were recorded during the nurse interview. Socioeconomic position was derived from the longest held occupation and were categorized into three groups (according to classifications determined by Statistics, Sweden) : 1) blue collar workers: no trained skill, goods-producing; no trained skill, service-producing; trained skill, goods-producing; trained skill, service-producing; 2) white collar workers: junior office worker, less than 2 years education after elementary school; junior office worker, ≥2 - <3 years after elementary school; office worker, ≥3 - <6 years after elementary school; senior office worker, ≥6 years after elementary school; 3) entrepreneurs: entrepreneurs; academic professions; farm owners.
Chronic diseases were recorded by a study physician. To avoid misreporting and recall biases, different sources of medical diagnoses were used to limit potential errors. The classification was based on clinical examination, medical history, laboratory data and current drug use. A disease/condition was classified as chronic if it was prolonged in duration, and if one or more of the following characteristics were present: 1) leaving residual disability or worsening of quality of life; and/or 2) requiring a long period of care or treatment/rehabilitation. Number of co-occurring chronic diseases/conditions was recorded. Multimorbidity was defined as having ≥2 chronic diseases . Cognitive level was measured with Mini Mental State Examination (MMSE) .
Leisure-time physical activity.
Information on leisure-time PA was obtained through a self-administered questionnaire concerning both intensity and frequency. Intensity was divided into two levels: light exercises (walks on the sidewalk or paved surfaces, in parks, in forests, short bike rides, light aerobic or gym classes, golf) and moderate to intense exercises (brisk walking, jogging, heavy gardening, long bike rides, intense aerobic or gym classes, skating, skiing, swimming, ball games or similar activities). Information on frequency included: every day; several times a week; two-to-three times per month; less than two-to-three times per month; never. Following the recommendations by the World Health Organization and the American College of Sports Medicine –, the participants were categorized into three different groups according to the intensity and frequency of the activities (from low to high): 1) inadequate: never, <2–3 times per month, 2–3 times per month in light and/or moderate/intense exercise; 2) health-enhancing: light exercise several times per week or every day; and 3) fitness-enhancing: moderate/intense exercise several times per week or every day .
Descriptive analysis was performed by presenting the number and the proportion of participants with light, moderate and strenuous occupational PA as well as dependency in PADL and IADL according to characteristics of the participants, and statistical differences were calculated using chi square tests.
Multinomial logistic regression analyses were conducted using dependency in personal and instrumental ADL, respectively, as the dependent variable and occupational PA level as the independent variable with having a light level of occupational PA level as the reference group.
Missing data for light (n = 707) and moderate to intense (n = 882) leisure-time PA were imputed in two steps. First, a manual imputation was performed. Subjects who were wheelchair bound (n = 37), could not move indoors or outdoors (n = 101), and who had severe physical problems or were incapable of walking 50 meters (n = 60) were coded as never taking part in any exercise. Secondly, we used fully conditional multiple regression imputation, which is a multivariate technique for multiply imputing missing values using a sequence of regression models . The following variables were included in the imputation according to the literature: age; gender; light exercise; moderate to intense exercise; living in own home or institution; PADL; IADL; walking speed; number of chronic diseases; presence of dementia; chronic stroke; Parkinson’s disease; pain while moving; presence of emotional support; social anchorage; and locus of control. The imputation was repeated five times and the results were saved in five different data sets. The data sets were pooled into one dataset with five observations for each person and the mean of the five observations for each person were used in the analyses.
In the first set of models (model A) demographic factors were controlled for. The factors included were age, gender, education (below/above high school level), socioeconomic position, and number of years in the longest held occupation. In the second set of models (model B) the analyses were additionally adjusted for current health-related factors: multimorbidity (below/above ≥2 diseases), cognitive function, leisure-time PA as well as instrumental ADL disability in the personal ADL-analyses. To test the hypothesis that the associations between occupational PA and late-life disability are different according to the category of occupation, stratified analyses by blue collar workers and white collar workers (including entrepreneurs) were performed.
Because the youngest and the oldest age groups were oversampled, a weighted variable was created and used in all proportion (percentage) and aggregated analyses. The analyses were conducted in SPSS (Statistics 18.0) and Stata/SE 11.2.
Participants that were dependent in personal and instrumental ADL were more likely to be older, female, have lower education, multimorbidity, and lower cognitive function (Table 1). Participants who reported light occupational PA were more likely to be female, have higher education, white collar occupation and higher cognitive function (Table 2). Subjects with moderate or heavy occupational PA were more likely to have lower education and a blue collar occupation.
Occupational PA was not significantly associated with either PADL disability (moderate occupational PA OR = 1.37; 95% CI 0.92–2.06 and strenuous occupational PA OR = 1.49; 95% CI 0.77–2.92), or IADL disability (moderate occupational PA OR = 1.18 95%; CI 0.91–1.54 and strenuous occupational PA OR = 0.88; 95% CI 0.56–1.39), as compared to light occupational PA in the unadjusted analyses. Although moderate occupational PA was related to a decreased risk and heavy PA was related to an increased risk of PADL, these associations were not statistically significant after controlling for demographic and health-related factors (Table 3).
When stratifying for blue and white collar workers, moderate occupational PA was associated with a significantly decreased risk of dependency in PADL (OR = 0.29 95% C1 0.09–0.92) in white collar workers (Figure 1) but not in blue collar workers after controlling for demographic and health-related factors. Strenuous occupational PA was associated with a non-significant elevated risk of dependency in PADL risk for both blue and white collar workers. Occupational PA was not significantly related to IADL in either group (data not shown).
To our knowledge, this is the first study that has investigated the association between occupational PA during the longest held occupation and disability in old age and at the same time examined the associations among blue and white collar workers, respectively. The study showed that occupational PA in midlife was not associated with PADL or IADL disability in old age. However, midlife moderate occupational PA decreased the risk of late-life PADL disability among white collar workers. The most common occupations among the white collar workers with moderate occupational PA were in the areas of health care, sales, teaching and engineering.
The non-significant relationship between occupational PA and disability is in line with a previous study in middle-aged people with arthritis , and another study showing a significantly increased risk of disability with strenuous occupational PA , although we observed only a non-significant risk effect of strenuous occupational PA on PADL disability. The non-significant result is reasonable because it is known that levels of occupational PA are related to occupation itself. Levels of occupational PA are usually described in terms of physical load or physical risk factors such as bending, twisting, pushing, pulling and lifting in a repetitive manner . The majority of the blue collar workers that reported strenuous occupational PA in this study were mechanics, construction workers and plumbers. High levels of occupational PA that are mostly found among types of work such as construction, agriculture, mining and production  have been shown to increase the risk of musculoskeletal disorders and of absence from work due to long-term sickness , , . Similar to our study, low levels of occupational PA are common among types of work such as administration, management and programming , and they are related to increased time in sitting, which has been shown to increase the risk of all-cause mortality and obesity , . Therefore, the beneficial effect of the occupational PA and the harmful effects of the occupation on disability might be cancelled out. Further, disability is a complex construct influenced by a variety of factors such as diseases, environment, demographic factors, psychosocial factors and physical fitness .
This is the first study showing that midlife moderate occupational PA in white collar workers was associated with a decreased risk of late-life disability. The result is in line with an earlier study showing that moderate to high occupational PA protected against hip fracture in women . Similar with another study, we found that strenuous occupational PA was related to an elevated risk of disability, although not statistically significant, in both blue and white collar workers . These results highlight the importance of encourage moderate but not strenuous occupational PA. The beneficial effect of moderate occupational PA in white collar workers should be emphasized because white collar workers might be more prone to long work hours which in turn have been shown to reduce leisure PA . If white collar workers can be more physically active at work, the beneficial effect of occupational PA may be retained because both leisure-time and occupational PA have been associated with a reduced risk of diseases and mortality, although the effect of occupational PA was weaker than leisure-time PA –, . In addition, worksite exercise has been shown to improve physical fitness, work ability and sleep quality –. Our finding that moderate occupational PA reduced the risk of PADL disability even when health-related factors including leisure-time physical exercise was controlled for highlights the importance of encouraging and facilitating the possibility to perform PA during or in conjunction to work hours. Although, it has been shown that leisure-time exercise is of importance, occupational PA and household activities are also of importance in order to reach recommended levels .
In contrary to previous studies showing an increased risk of disability in relation to occupational PA ,  in blue collar workers, we found that occupational PA did not relate to increased risk of late-life disability. Even though blue collar workers have a higher total PA  this may not always be beneficial because they are more subject to hazardous and/or repetitive working situations. Further, people in lower socioeconomic classes, measured with occupational category and/or education, have a higher risk of disability ,  up to early old age . The beneficial effect of occupational PA may not have enough power to counteract the negative effect of the occupational hazards. Therefore, it seems as though the quality of the PA is important.
A limitation of the study is that the study sample has a higher educational level compared to the Swedish average level, which hampers the generalization of the results. Further, data on leisure-time PA were collected only at late-life but not at midlife, which limits the possibility to examine the impact of midlife leisure activities on the studied association. However, we did stratified analysis for late-life leisure-time PA, similar results were observed among those who were doing inadequate exercise and those who were doing health-enhancing or fitness-enhancing exercise. This suggests that leisure PA might not moderate the studied association because those who are active in late-life were more likely to be active in midlife as well . Another limitation is the retrospective design, since data on occupation and occupational PA was retrospectively collected, which may introduce recall bias. However, the recall bias could be of less importance because face-to-face interviews were conducted in all participants, and may lead to an underestimation of the studied associations. In addition, a representative random population-based sample of participants has been included and health measures such as multimorbidity and cognitive level have been objectively measured, which increases the validity of the study.
In conclusion, this study showed that midlife occupational PA was not associated with either PADL or IADL disability in old age. However, midlife moderate occupational PA was related to a decreased risk of late-life PADL disability among white collar workers, findings that need to be confirmed in further studies. The findings highlight the importance of socioeconomic factors related to PA-level and health related outcomes, such as disability. It also highlight the importance of encouraging and facilitating people performing PA during or outside work hours, especially for white collar workers who have less occupational related activities and taking into account both level and quality of PA.
We thank all the SNAC-K participants for their invaluable contributions and our colleagues in the SNAC-K Group for their collaboration in data collection and management.
Conceived and designed the experiments: ER LF H-XW. Analyzed the data: ER. Had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis: ER. Interpretation of data and preparation of manuscript: ER A-KW SA LF H-XW. Read and approved the final version of the manuscript: ER A-KW SA LF H-XW.
- 1. Nusselder WJ, Looman CWN, Franco OH, Peeters A, Slingerland AS, et al. (2008) The relation between non-occupational physical activity and years lived with and without disability. J Epidemiol Community Health 62: 823–828.
- 2. Holtermann A, Marott JL, Gyntelberg F, Søgaard K, Suadicani P, et al. (2013) Does the benefit on survival from leisure time physical activity depend on physical activity at work? A prospective cohort study. PLoS One 8: e54548.
- 3. Salonen JT, Slater JS, Tuomilehto J, Rauramaa R (1988) Leisure-time and occupational physical-activity - Risk of death from ischemic heart-disease. Am J Epidemiol 127: 87–94.
- 4. Jaglal SB, Kreiger N, Darlington GA (1995) Lifetime occupational physical activity and risk of hip fracture in women. Ann Epidemiol 5: 321–324.
- 5. Wang YJ, Tuomilehto J, Jousilahti P, Antikainen R, Mahonen M, et al. (2010) Occupational, Commuting, and Leisure-Time Physical Activity in Relation to Heart Failure Among Finnish Men and Women. J Am Coll Cardiol 56: 1140–1148.
- 6. Wennberg P, Lindahl B, Hallmans G, Messner T, Weinehall L, et al. (2006) The effects of commuting activity and occupational and leisure time physical activity on risk of myocardial infarction. Eur J Cardiovasc Prev Rehabil 13: 924–930.
- 7. Graff-Iversen S, Selmer R, Sorensen M, Skurtveit S (2007) Occupational physical activity, overweight, and mortality: A follow-up study of 47,405 Norwegian women and men. Res Quart Exerc Sport 78: 151–161.
- 8. Menotti A, Lanti M, Maiani G, Kromhout D (2005) Forty-year mortality from cardiovascular diseases and their risk factors in men of the Italian rural areas of the Seven Countries Study. Acta Cardiol 60: 521–531.
- 9. Menotti A, Lanti M, Maiani G, Kromhout D (2006) Determinants of longevity and all-cause mortality among middle-aged men. Role of 48 personal characteristics in a 40-year follow-up of Italian Rural Areas in the Seven Countries Study. Aging Clin Exp Res 18: 394–406.
- 10. Andersen LB, Schnohr P, Schroll M, Hein HO (2000) All-cause mortality associated with physical activity during leisure time, work, sports, and cycling to work. Arch Int Med 160: 1621–1628.
- 11. Feinglass J, Thompson JA, He XXZ, Witt W, Chang RW, et al. (2005) Effect of physical activity on functional status among older middle-age adults with arthritis. Arthritis Rheum -Arthritis Care Res 53: 879–885.
- 12. Holtermann A, Mortensen OS, Burr H, Sogaard K, Gyntelberg F, et al. (2010) Physical demands at work, physical fitness, and 30-year ischaemic heart disease and all-cause mortality in the Copenhagen Male Study. Scand J Work Environ Health 36: 357–365.
- 13. Ward MM, Hubert HB, Shi H, Bloch DA (1995) Physical-disability in older runners - Prevalence, risk-factors and progression with age. J Gerontol A Biol Sci Med Sci 50: M70–M77.
- 14. Kirk MA, Rhodes RE (2011) Occupation Correlates of Adults’ Participation in Leisure-Time Physical Activity A Systematic Review. Am J Prev Med 40: 476–485.
- 15. Arndt V, Rothenbacher D, Daniel U, Zschenderlein B, Schuberth S, et al. (2005) Construction work and risk of occupational disability: a ten year follow up of 14,474 male workers. Occup Environ Med 62: 559–566.
- 16. Caban-Martinez AJ, Lee DJ, Fleming LE, Tancredi DJ, Arheart KL, et al. (2011) Arthritis, occupational class, and the aging US workforce. Am J Public Health 101: 1729–1734.
- 17. Rossignol M, Leclerc A, Hilliquin P, Allaert FA, Rozenberg S, et al. (2003) Primary osteoarthritis and occupations: a national cross sectional survey of 10 412 symptomatic patients. Occup Environ Med 60: 882–886.
- 18. Li CY, Wu SC, Wen SW (2000) Longest held occupation in a lifetime and risk of disability in activities of daily living. Occup Environ Med 57: 550–554.
- 19. Russo A, Onder G, Cesari M, Zamboni V, Barillaro C, et al. (2006) Lifetime occupation and physical function: a prospective cohort study on persons aged 80 years and older living in a community. Occup Environ Med 63: 438–442.
- 20. von Bonsdorff MB, Seitsamo J, Ilmarinen J, Nygard CH, von Bonsdorff ME, et al. (2011) Work ability in midlife as a predictor of mortality and disability in later life: a 28-year prospective follow-up study. CMAJ 183: E235–242.
- 21. Lagergren M, Fratiglioni L, Hallberg IR, Berglund J, Elmstahl S, et al. (2004) A longitudinal study integrating population, care and social services data. The Swedish National study on Aging and Care (SNAC). Aging Clin Exp Res 16: 158–168.
- 22. Statistics Sweden (SCB) (1982) Swedish Socioeconomic Classification.
- 23. van den Akker M, Buntinx F, Metsemakers JF, Roos S, Knottnerus JA (1998) Multimorbidity in general practice: prevalence, incidence, and determinants of co-occurring chronic and recurrent diseases. J Clin Epidemiol 51: 367–375.
- 24. Folstein MF, Folstein SE, McHugh PR (1975) Mini-mental state - Practical method for grading cognitive state of patients for clinician. J Psychiatr Res 12: 189–198.
- 25. Global recommendations on physical activity for health. WHO (2010) Available: http://whqlibdoc.who.int/publications/2010/9789241599979_eng.pdf.Accessed 2012 25 September.
- 26. Nelson ME, Rejeski WJ, Blair SN, Duncan PW, Judge JO, et al. (2007) Physical activity and public health in older adults: recommendation from the American College of Sports Medicine and the American Heart Association. Med Sci Sports Exerc 39: 1435–1445.
- 27. Rydwik E, Welmer AK, Kåreholt I, Angleman S, Fratiglioni L, et al. Adherence to physical exercise recommendations in people over 65–The SNAC-Kungsholmen study. Eur J Public Health. 2012 Oct 31. [Epub ahead of print]
- 28. Raghunathan TE, Lepkowski JM, Van Hoewyk, J, Solenberger P (2001) A multivariate technique for multiply imputing missing values using a sequence of regression models. Surv Methodol 27: 86–95.
- 29. Lund T, Labriola M, Christensen KB, Bultmann U, Villadsen E (2006) Physical work environment risk factors for long term sickness absence: prospective findings among a cohort of 5357 employees in Denmark. Br Med J 332: 449–451.
- 30. Torgen M, Kilbom A (2000) Physical work load between 1970 and 1993 - did it change? Scand J Work Environ Health 26: 161–168.
- 31. Holtermann A, Hansen JV, Burr H, Sogaard K, Sjogaard G (2012) The health paradox of occupational and leisure-time physical activity. Br J Sports Med 46: 291–295.
- 32. Mummery WK, Schofield GM, Steele R, Eakin EG, Brown WJ (2005) Occupational sitting time and overweight and obesity in Australian workers. Am J Prev Med 29: 91–97.
- 33. van der Ploeg HP, Chey T, Korda RJ, Banks E, Bauman A (2012) Sitting Time and All-Cause Mortality Risk in 222 497 Australian Adults. Arch Int Med 172: 494–500.
- 34. Fiatarone Singh MA (2002) Exercise to prevent and treat functional disability. Clin Geriatr Med 18: 431–462.
- 35. Leino-Arjas P, Solovieva S, Riihimaki H, Kirjonen J, Telama R (2004) Leisure time physical activity and strenuousness of work as predictors of physical functioning: a 28 year follow up of a cohort of industrial employees. Occup Environ Med 61: 1032–1038.
- 36. Atlantis E, Chow CM, Kirby A, Singh MA (2006) Worksite intervention effects on sleep quality: a randomized controlled trial. J Occup Health Psychol 11: 291–304.
- 37. Pohjonen T, Ranta R (2001) Effects of worksite physical exercise intervention on physical fitness, perceived health status, and work ability among home care workers: five-year follow-up. Prev Med 32: 465–475.
- 38. Vuillemin A, Rostami C, Maes L, Van Cauwenberghe E, Van Lenthe FJ, et al. (2011) Worksite physical activity interventions and obesity: a review of European studies (the HOPE project). Obes Facts 4: 479–488.
- 39. Ratzlaff CR (2012) Good news, bad news: sports matter but occupational and household activity really matter - sport and recreation unlikely to be a panacea for public health. Br J Sports Med 46: 699–701.
- 40. Reinhardt JD, Wahrendorf M, Siegrist J (2013) Socioeconomic position, psychosocial work, environment and disability in an ageing workforce: a longitudinal analysis of SHARE data from 11 European countries. Occup Environ Med 70: 156–163.
- 41. Ramsay SE, Whincup PH, Morris RW, Lennon LT, Wannamethee SG (2008) Extent of social inequalities in disability in the elderly: results from a population-based study of British men. Ann Epidemiol 18: 896–903.
- 42. Frändin K, Mellström D, Sundh V, Svanborg A (1995) A life span perspective on patterns of physical activity and functional performance at the age of 76. Gerontology 41: 109–120.