527526
research-article2014
SJP0010.1177/1403494814527526A.M. Hansen et al.Mobility limitations
Scandinavian Journal of Public Health, 2014; 42: 417–424
Original Article
Psychosocial factors at work and the development of mobility limitations among adults in Denmark
ÅSE MARIE HANSEN1,2, LILIE DARSØ1, MINNA MANTY1,3, CHARLOTTE NILSSON1,3, ULLA CHRISTENSEN1, RIKKE LUND1,3, ANDREAS HOLTERMANN2 & KIRSTEN AVLUND1,3,4 1Department of Public Health, University of Copenhagen, Denmark, 2National Research Centre for Working Environment, Copenhagen, Denmark, 3Center for Healthy Ageing, University of Copenhagen, Denmark, and 4Danish Aging Research Center, Universities of Aarhus, Southern Denmark (Odense) and Copenhagen, Denmark
Abstract Aim: Psychosocial factors in the working environment have been shown to be associated with mobility limitations, but this has not yet been confirmed in a Danish population. We aimed to examine how psychosocial factors at work are related to developing mobility limitations in Denmark. Methods: This study is based on data from 2952 middle-aged men and women without mobility limitations in 2000. Results: We found increased risk of incident mobility limitations during 6-year follow up among men who often perceived high work pace (OR 5.45, 95% CI 1.21–24.52) vs. never, who only sometimes or/ never perceived the work to be meaningful (OR 6.54, 95% CI 1.55–27.55) vs. always, and who sometimes perceived high emotional demands at work (OR 7.85, 95% CI 1.78–34.65) vs. never. Among women, lower risk of incident mobility limitations was observed among those who in 2000 perceived high work pace sometimes (OR 0.46, 95% CI 0.24–0.87) or often (OR 0.43, 95% CI 0.22–0.85) vs. never in 2000. Also, women who always or often experienced high emotional demands had an increased risk. Conclusions: The most important finding was that high work pace was strongly associated with increased risk of mobility limitations among men, but associated with lower risk of mobility limitations among women. This knowledge may be used to better target interventions among men and women in midlife from physical deterioration later in life. Key Words: Emotional demands, high work pace, mobility limitations, psychosocial factors
Introduction Mobility limitations are predictive of disability [1]. Mobility limitations are observed frequent among elderly, but are already observed in midlife [2,3]. In the English Longitudinal Study of Ageing, 8% of women and 9% of men aged 50–64 years have difficulties in walking a quarter mile increasing to 20% of women and 17% of men aged 65–79 years of age [2]. A slightly different result was found in the American Health and Retirement Survey, where the onset of mobility limitations among adults aged 51–61 was found to be 6% after 2-year follow up [4]. It is
important to identify risk factors for and prevent mobility limitations among the middle-aged as attention paid to risk factors and early signs of disability may lead to disability reductions in later life [3]. High demands and effort–reward imbalance are independent powerful predictors of poor functioning [5]. In the Whitehall II study, a ratio of high efforts to low rewards predicted poor physical functioning during 11-year follow up among 6895 male and 3413 female civil servants aged 35–55 [6]. Another study showed that low job control, high job demands, and
Correspondence: Åse Marie Hansen, Department of Public Health, University of Copenhagen, Øster Farimagsgade 5 opg. B, Postboks 2099, 1014 Copenhagen K, Denmark. E-mail: [email protected] (Accepted 18 February 2014) © 2014 the Nordic Societies of Public Health DOI: 10.1177/1403494814527526
Downloaded from sjp.sagepub.com at The University of Iowa Libraries on March 14, 2015
418 A.M. Hansen et al. low work-related social support were associated with functional decline years later among 21,290 female nurses [7]. There has been little attention on gender differences in the early stages of functional limitations, but gender differences in the development of mobility limitations [8] has been shown. The perception of the psychosocial work environment differ between men and women: e.g. women experience worse conditions in the work environment [9] and are more vulnerable to the negative effects than men [10]. A few studies showed that higher job strain was related to poor physical health among men [9] and women [7]. Mobility limitations may have progressive impact on the workers’ capacity to do their job [11], and to our knowledge, no studies have focused on the impact of adverse psychosocial factors at work and on mobility limitations among Danish men and women. The objective of this study was to examine if five psychosocial factors at work (quantitative demands, work pace, emotional demands, meaningfulness, and social support from coworkers and supervisors) were associated with onset of mobility limitations, that is, limitations in running and climbing stairs, during a 6-year prospective study among middle-aged Danish men and women. Materials and methods Study population The present study was based on data from the Danish Longitudinal Study on Work, Unemployment and Health, including a random sample of Danish adults aged 40 and 50 years in 1999 (n=11,082). Baseline data were collected in 2000 by postal questionnaire (n=7588, response rate 69%). Follow-up data on surviving participants and participants who had not emigrated or refused to participate in research (n=651) were collected in 2006 by postal questionnaire (n=4893, response rate 71%). Of these, 715 responders were excluded due to unemployment in 2000. Nonresponders of the questionnaire differed from responders (i.e. a larger proportion of these were men with low socioeconomic status). Responders and nonresponders did not differ on number of contacts with general practitioner [12]. Participants with mobility limitations in 2000 (n=174) were excluded. Participants still active in working life, who returned the questionnaires in 2000 and 2006, and who had full records on all included variables and who did not have mobility limitations at baseline (n=2952) were included. Five scales from the Copenhagen Psychosocial Questionnaire (COPSOQ) [13] were used. Quantitative demands consisted of three items: (1)
“Is your workload unevenly distributed so it piles up?”, (2) “How often do you not have time to complete all your work tasks?” (3) “Do you have to do overtime?”. Work pace consisted of one item: (1) “Do you have enough time for your work tasks?” All items were responded to on 4–point scales using a frequency (0=never to 3=always). Meaning of work consisted of three items: (1) “Is your work meaningful?”, (2) “Do you feel that the work you do is important?”, and (3)”Do you feel motivated and involved in your work?”. Social support was measured by a modified version of the original scale, using only two items: (1) “How often do you get help and support from your colleagues?” and (2) “How often do you get help and support from your immediate superior?”. All items were responded to on 4–point scales using a frequency (0=always to 3=never). The scales were established as the mean score of the items mentioned in each scale and categorized as always (scores from 0 to 0.75), often (scores from 0.76 to 1.50), sometimes (scores from 1.51 to 2.25), and never (scores from 2.26 to 3.0). Cronbach’s α for each scale were: quantitative demands, α=0.66; for meaningfulness, α=0.80; and social support from coworkers and supervisor, α=0.72. Emotional demands were measured by a modified version of the original scale, using only one item: “Does your work put you in emotionally disturbing situations?”. The item was responded to and a four-point scale ranging from 0 (always) to 3 (never). Mobility limitations in 2000 and 2006 was measured by a modified version of the Rosow–Breslau functional status measure [14] assessing whether the study participant feels limited because of health problems, when running 100 m and climbing stairs to the second floor. The item was: “Are you due to your health limited in the following two activities: (a) running 100 m, (b) climbing stairs to the second floor?” (yes very, yes a little, no not at all). Mobility limitations were dichotomized as no (no not at all and yes a little) and yes (yes very). We excluded participants that responded “very limited” in running 100 m and/or climbing stairs to the second floor at baseline. The excluded participants had more cases of mobility limitations at follow up, were older, more likely to be women, had more chronic diseases, and belonged to the lower social classes compared to the study participants. Covariates Covariates measured at baseline (2000) were: age (40 or 50 years), social class, and number of chronic diseases diagnosed by a physician, body mass index (BMI), back pain, and life-style factors (smoking
Downloaded from sjp.sagepub.com at The University of Iowa Libraries on March 14, 2015
Mobility limitations 419 status, alcohol consumption, leisure time physical activity, and work-related physical activity). Covariates were chosen in relation to their theoretical association to the psychosocial factors at work [9,13] and to mobility limitations [15,16]. Socioeconomic position was measured by the Danish Occupational Social Class measurement, including social class I–V, which represents all individuals in the workforce. Number of chronic diseases was evaluated by one item: “have you ever been diagnosed with… myocardial infarction, angina pectoris, or other heart disease; asthma or bronchitis; stroke or cerebral haemorrhage; depression or other mental illness/eczema or other skin disease; diabetes or other regulatory disease; arthritis or other diseases in back, muscles, or joints; other chronic disease?” (response categories: no, yes, don’t know). In the present study, chronic diseases were classified in four groups (none; one; two or more chronic diseases; don’t know). BMI was divided in four groups (≤18.5, 18.5–24.9, 25– 29.9, >30 kg/m2). We included one item on pain: “Have you during the past four weeks had pain in limbs?”, with responses on a four-point scale ranging from not at all to all the time. Smoking status was divided in four groups: (heavy smoker (>15 cigarettes per day), light smoker (6 hours/week). Work-related physical activity was measured by three items: “How would you describe the physical activity in your work?”, with four possibilities: (1) heavy/fast work; (2) standing with lifting; and (3) standing without lifting. All items were responded to using a 4-point scale using frequency (0, always; 3, never). A scale was established as the mean score of the items and finally dichotomized between 0 and >0. Statistics We used descriptive statistics (χ2 tests) to test for gender differences. Multiple logistic regression analyses were used to test for risk of onset mobility limitation at follow up from psychosocial factors at work
at baseline. Scales on quantitative demands, work pace, meaningfulness, and social support had four levels (always, often, sometimes, never). The item on emotional demands also had four response categories from never to always. Groups consisting of n≤25 were joined with the group next to it (i.e. never, n=21; sometimes meaningfulness at work was collapsed; and always, n=20, and often emotional demands was collapsed). The analyses of risk for onset of mobility limitations (dichotomized) were performed in four models. Model 1 included the psychosocial factors at work (quantitative demands, work pace, meaningfulness, social support, and emotional demands) and adjusted for age. Model 2 was model 1 + social class (5 levels); model 3 was model 2 + chronic conditions; and model 4 was model 3 + smoking + BMI + alcohol consumption + leisure time physical activity (two levels) + physical activity at work (four levels). All analysis analyses were conducted in SPSS/PASW Statistics 18.0. Results Table I shows demographic characteristics, chronic illness, and mobility limitations among men and women in the cohort at baseline. At follow up in 2006, significantly more women (10.5%) than men (7.6%) had experienced onset of mobility limitations. The men were significantly older than the women and had fewer cases of comorbidity at baseline. Table II presents the gender-specific distributions of the psychosocial work-related factors at baseline. Compared to men, women experienced always and/or often higher degree of work pace (45.7 vs. 40.0%), emotional demands (20.0 vs. 12.8%), and social support (54.5 vs. 43.5), and men perceived more quantitative demands at work (28.6%) than women (23.2%). The majority of both male and female participants (95%) reported that they always or often regarded their work as meaningful. Table III (men) and Table IV (women) show odds ratios for onset of mobility limitations at 6-year follow up by psychosocial factors at work. Men who often vs. never experienced high work pace at baseline (OR 5.45, 95% CI 1.21–24.52), who sometimes/ never vs. always experienced meaningfulness of work (OR 6.54, 95% CI 1.55–27.55), and who sometimes vs. never felt emotional demands (OR 7.85, 95% CI 1.78–34.65) at work had higher risk of mobility limitations (Table III). Women who often (OR 0.43, 95% CI 0.22–0.85) or sometimes (OR 0.46 (0.24–0.87) experienced high work pace had significantly lower risks of developing mobility limitations compared to women who never experienced high work pace. Women who often/always perceived high emotional
Downloaded from sjp.sagepub.com at The University of Iowa Libraries on March 14, 2015
420 A.M. Hansen et al. Table I. Demographical characteristics in 2000 of the cohort stratified by gender (n=2952).
Aged 40 in 1999 Aged 50 in 1999 Social group I II III IV V Chronic illness None 1 ≥2 Mobility limitations in 2006 Not limited Limited
Total (n)
Total (%)
Men (%)
Women (%)
p-value
1468 1484
49.7 50.3
46.4 53.6
53.0 47.0
312 644 865 709 422
10.6 21.8 29.3 24.0 14.3
14.9 22.6 24.5 21.8 16.2
6.3 21.0 34.1 26.2 12.4
research-article2014
SJP0010.1177/1403494814527526A.M. Hansen et al.Mobility limitations
Scandinavian Journal of Public Health, 2014; 42: 417–424
Original Article
Psychosocial factors at work and the development of mobility limitations among adults in Denmark
ÅSE MARIE HANSEN1,2, LILIE DARSØ1, MINNA MANTY1,3, CHARLOTTE NILSSON1,3, ULLA CHRISTENSEN1, RIKKE LUND1,3, ANDREAS HOLTERMANN2 & KIRSTEN AVLUND1,3,4 1Department of Public Health, University of Copenhagen, Denmark, 2National Research Centre for Working Environment, Copenhagen, Denmark, 3Center for Healthy Ageing, University of Copenhagen, Denmark, and 4Danish Aging Research Center, Universities of Aarhus, Southern Denmark (Odense) and Copenhagen, Denmark
Abstract Aim: Psychosocial factors in the working environment have been shown to be associated with mobility limitations, but this has not yet been confirmed in a Danish population. We aimed to examine how psychosocial factors at work are related to developing mobility limitations in Denmark. Methods: This study is based on data from 2952 middle-aged men and women without mobility limitations in 2000. Results: We found increased risk of incident mobility limitations during 6-year follow up among men who often perceived high work pace (OR 5.45, 95% CI 1.21–24.52) vs. never, who only sometimes or/ never perceived the work to be meaningful (OR 6.54, 95% CI 1.55–27.55) vs. always, and who sometimes perceived high emotional demands at work (OR 7.85, 95% CI 1.78–34.65) vs. never. Among women, lower risk of incident mobility limitations was observed among those who in 2000 perceived high work pace sometimes (OR 0.46, 95% CI 0.24–0.87) or often (OR 0.43, 95% CI 0.22–0.85) vs. never in 2000. Also, women who always or often experienced high emotional demands had an increased risk. Conclusions: The most important finding was that high work pace was strongly associated with increased risk of mobility limitations among men, but associated with lower risk of mobility limitations among women. This knowledge may be used to better target interventions among men and women in midlife from physical deterioration later in life. Key Words: Emotional demands, high work pace, mobility limitations, psychosocial factors
Introduction Mobility limitations are predictive of disability [1]. Mobility limitations are observed frequent among elderly, but are already observed in midlife [2,3]. In the English Longitudinal Study of Ageing, 8% of women and 9% of men aged 50–64 years have difficulties in walking a quarter mile increasing to 20% of women and 17% of men aged 65–79 years of age [2]. A slightly different result was found in the American Health and Retirement Survey, where the onset of mobility limitations among adults aged 51–61 was found to be 6% after 2-year follow up [4]. It is
important to identify risk factors for and prevent mobility limitations among the middle-aged as attention paid to risk factors and early signs of disability may lead to disability reductions in later life [3]. High demands and effort–reward imbalance are independent powerful predictors of poor functioning [5]. In the Whitehall II study, a ratio of high efforts to low rewards predicted poor physical functioning during 11-year follow up among 6895 male and 3413 female civil servants aged 35–55 [6]. Another study showed that low job control, high job demands, and
Correspondence: Åse Marie Hansen, Department of Public Health, University of Copenhagen, Øster Farimagsgade 5 opg. B, Postboks 2099, 1014 Copenhagen K, Denmark. E-mail: [email protected] (Accepted 18 February 2014) © 2014 the Nordic Societies of Public Health DOI: 10.1177/1403494814527526
Downloaded from sjp.sagepub.com at The University of Iowa Libraries on March 14, 2015
418 A.M. Hansen et al. low work-related social support were associated with functional decline years later among 21,290 female nurses [7]. There has been little attention on gender differences in the early stages of functional limitations, but gender differences in the development of mobility limitations [8] has been shown. The perception of the psychosocial work environment differ between men and women: e.g. women experience worse conditions in the work environment [9] and are more vulnerable to the negative effects than men [10]. A few studies showed that higher job strain was related to poor physical health among men [9] and women [7]. Mobility limitations may have progressive impact on the workers’ capacity to do their job [11], and to our knowledge, no studies have focused on the impact of adverse psychosocial factors at work and on mobility limitations among Danish men and women. The objective of this study was to examine if five psychosocial factors at work (quantitative demands, work pace, emotional demands, meaningfulness, and social support from coworkers and supervisors) were associated with onset of mobility limitations, that is, limitations in running and climbing stairs, during a 6-year prospective study among middle-aged Danish men and women. Materials and methods Study population The present study was based on data from the Danish Longitudinal Study on Work, Unemployment and Health, including a random sample of Danish adults aged 40 and 50 years in 1999 (n=11,082). Baseline data were collected in 2000 by postal questionnaire (n=7588, response rate 69%). Follow-up data on surviving participants and participants who had not emigrated or refused to participate in research (n=651) were collected in 2006 by postal questionnaire (n=4893, response rate 71%). Of these, 715 responders were excluded due to unemployment in 2000. Nonresponders of the questionnaire differed from responders (i.e. a larger proportion of these were men with low socioeconomic status). Responders and nonresponders did not differ on number of contacts with general practitioner [12]. Participants with mobility limitations in 2000 (n=174) were excluded. Participants still active in working life, who returned the questionnaires in 2000 and 2006, and who had full records on all included variables and who did not have mobility limitations at baseline (n=2952) were included. Five scales from the Copenhagen Psychosocial Questionnaire (COPSOQ) [13] were used. Quantitative demands consisted of three items: (1)
“Is your workload unevenly distributed so it piles up?”, (2) “How often do you not have time to complete all your work tasks?” (3) “Do you have to do overtime?”. Work pace consisted of one item: (1) “Do you have enough time for your work tasks?” All items were responded to on 4–point scales using a frequency (0=never to 3=always). Meaning of work consisted of three items: (1) “Is your work meaningful?”, (2) “Do you feel that the work you do is important?”, and (3)”Do you feel motivated and involved in your work?”. Social support was measured by a modified version of the original scale, using only two items: (1) “How often do you get help and support from your colleagues?” and (2) “How often do you get help and support from your immediate superior?”. All items were responded to on 4–point scales using a frequency (0=always to 3=never). The scales were established as the mean score of the items mentioned in each scale and categorized as always (scores from 0 to 0.75), often (scores from 0.76 to 1.50), sometimes (scores from 1.51 to 2.25), and never (scores from 2.26 to 3.0). Cronbach’s α for each scale were: quantitative demands, α=0.66; for meaningfulness, α=0.80; and social support from coworkers and supervisor, α=0.72. Emotional demands were measured by a modified version of the original scale, using only one item: “Does your work put you in emotionally disturbing situations?”. The item was responded to and a four-point scale ranging from 0 (always) to 3 (never). Mobility limitations in 2000 and 2006 was measured by a modified version of the Rosow–Breslau functional status measure [14] assessing whether the study participant feels limited because of health problems, when running 100 m and climbing stairs to the second floor. The item was: “Are you due to your health limited in the following two activities: (a) running 100 m, (b) climbing stairs to the second floor?” (yes very, yes a little, no not at all). Mobility limitations were dichotomized as no (no not at all and yes a little) and yes (yes very). We excluded participants that responded “very limited” in running 100 m and/or climbing stairs to the second floor at baseline. The excluded participants had more cases of mobility limitations at follow up, were older, more likely to be women, had more chronic diseases, and belonged to the lower social classes compared to the study participants. Covariates Covariates measured at baseline (2000) were: age (40 or 50 years), social class, and number of chronic diseases diagnosed by a physician, body mass index (BMI), back pain, and life-style factors (smoking
Downloaded from sjp.sagepub.com at The University of Iowa Libraries on March 14, 2015
Mobility limitations 419 status, alcohol consumption, leisure time physical activity, and work-related physical activity). Covariates were chosen in relation to their theoretical association to the psychosocial factors at work [9,13] and to mobility limitations [15,16]. Socioeconomic position was measured by the Danish Occupational Social Class measurement, including social class I–V, which represents all individuals in the workforce. Number of chronic diseases was evaluated by one item: “have you ever been diagnosed with… myocardial infarction, angina pectoris, or other heart disease; asthma or bronchitis; stroke or cerebral haemorrhage; depression or other mental illness/eczema or other skin disease; diabetes or other regulatory disease; arthritis or other diseases in back, muscles, or joints; other chronic disease?” (response categories: no, yes, don’t know). In the present study, chronic diseases were classified in four groups (none; one; two or more chronic diseases; don’t know). BMI was divided in four groups (≤18.5, 18.5–24.9, 25– 29.9, >30 kg/m2). We included one item on pain: “Have you during the past four weeks had pain in limbs?”, with responses on a four-point scale ranging from not at all to all the time. Smoking status was divided in four groups: (heavy smoker (>15 cigarettes per day), light smoker (6 hours/week). Work-related physical activity was measured by three items: “How would you describe the physical activity in your work?”, with four possibilities: (1) heavy/fast work; (2) standing with lifting; and (3) standing without lifting. All items were responded to using a 4-point scale using frequency (0, always; 3, never). A scale was established as the mean score of the items and finally dichotomized between 0 and >0. Statistics We used descriptive statistics (χ2 tests) to test for gender differences. Multiple logistic regression analyses were used to test for risk of onset mobility limitation at follow up from psychosocial factors at work
at baseline. Scales on quantitative demands, work pace, meaningfulness, and social support had four levels (always, often, sometimes, never). The item on emotional demands also had four response categories from never to always. Groups consisting of n≤25 were joined with the group next to it (i.e. never, n=21; sometimes meaningfulness at work was collapsed; and always, n=20, and often emotional demands was collapsed). The analyses of risk for onset of mobility limitations (dichotomized) were performed in four models. Model 1 included the psychosocial factors at work (quantitative demands, work pace, meaningfulness, social support, and emotional demands) and adjusted for age. Model 2 was model 1 + social class (5 levels); model 3 was model 2 + chronic conditions; and model 4 was model 3 + smoking + BMI + alcohol consumption + leisure time physical activity (two levels) + physical activity at work (four levels). All analysis analyses were conducted in SPSS/PASW Statistics 18.0. Results Table I shows demographic characteristics, chronic illness, and mobility limitations among men and women in the cohort at baseline. At follow up in 2006, significantly more women (10.5%) than men (7.6%) had experienced onset of mobility limitations. The men were significantly older than the women and had fewer cases of comorbidity at baseline. Table II presents the gender-specific distributions of the psychosocial work-related factors at baseline. Compared to men, women experienced always and/or often higher degree of work pace (45.7 vs. 40.0%), emotional demands (20.0 vs. 12.8%), and social support (54.5 vs. 43.5), and men perceived more quantitative demands at work (28.6%) than women (23.2%). The majority of both male and female participants (95%) reported that they always or often regarded their work as meaningful. Table III (men) and Table IV (women) show odds ratios for onset of mobility limitations at 6-year follow up by psychosocial factors at work. Men who often vs. never experienced high work pace at baseline (OR 5.45, 95% CI 1.21–24.52), who sometimes/ never vs. always experienced meaningfulness of work (OR 6.54, 95% CI 1.55–27.55), and who sometimes vs. never felt emotional demands (OR 7.85, 95% CI 1.78–34.65) at work had higher risk of mobility limitations (Table III). Women who often (OR 0.43, 95% CI 0.22–0.85) or sometimes (OR 0.46 (0.24–0.87) experienced high work pace had significantly lower risks of developing mobility limitations compared to women who never experienced high work pace. Women who often/always perceived high emotional
Downloaded from sjp.sagepub.com at The University of Iowa Libraries on March 14, 2015
420 A.M. Hansen et al. Table I. Demographical characteristics in 2000 of the cohort stratified by gender (n=2952).
Aged 40 in 1999 Aged 50 in 1999 Social group I II III IV V Chronic illness None 1 ≥2 Mobility limitations in 2006 Not limited Limited
Total (n)
Total (%)
Men (%)
Women (%)
p-value
1468 1484
49.7 50.3
46.4 53.6
53.0 47.0
312 644 865 709 422
10.6 21.8 29.3 24.0 14.3
14.9 22.6 24.5 21.8 16.2
6.3 21.0 34.1 26.2 12.4
