Abstract

Aim: To investigate potential differences in sickness absence among public sector employees in Sweden and Denmark, and to what extent a difference was associated with age, gender, physical and psychosocial work environment exposures, lifestyle factors, self-rated health or work ability. Methods: In 2000, two cross-sectional samples of 8562 public sector employees in Sweden and Denmark were surveyed. The study outcome, self-reported number of sick-leave days the year preceding interview, was dichotomized into 7 days or less, and more than 7 days. Chi square test was used to analyse distribution of dependent and independent variables in the two sub-cohorts. Stratified logistic regression analysis was performed to identify causes for absence within the two sub-cohorts, and logistic regression analysis was performed to study differences in sickness absence levels between the two sub-cohorts. Results: More subjects from the Swedish study population reported more than 7 days of sickness absence. Factors associated with sickness absence were largely similar in the two countries. The difference in absence level between Sweden and Denmark was not associated with differences in age, gender, skill level, lifestyle, psychosocial or physical work environment, musculoskeletal symptoms or self-rated health, whereas work ability score decreased the difference in sickness absence level. Conclusion: The results could indicate an increased retention of employees with health problems in the Swedish labour market compared with the Danish labour market. A possible explanation for the differences in sickness absence ascertained in this study could be due to differences in the sickness insurance legislation.

Introduction

Sickness absence and the related factors vary over time, among different population sub-groups and between countries. In recent decades, the number of studies of sickness absence has increased in the Nordic countries as well as in many other industrial countries due to extensive consequences for the individual, employers and society. However, international comparisons of sickness absence and its underlying causes are urgently needed to increase the scientific knowledge in this research area.1,2

The existing, surprisingly few, comparisons of sickness absence between countries show higher proportions of sickness absence in Sweden, Norway and The Netherlands, compared to other European countries.3,4 Also, higher levels of and risk for sickness absence have been reported among employees in the public sector compared with employees in the private sector.4–6 Especially long-term sick-leave spells have been in focus since longer sickness absence spells could have negative effects on the life situation7 and increase the risk of disability pension.8–10

National data on sick leave in Sweden show that persons employed by the municipalities are overrepresented among those on sick leave for ≥60 days per annum and that sickness absence varies between occupational groups.11–14 Compared to Sweden, the overall absence rate in Denmark is considered to be relatively low.15 However, to date, international comparisons have suffered from incompatibility of sickness absence data sources. Nevertheless, there seems to be a similar trend in Denmark as in Sweden, with regards to an excess risk of sickness absence among employees in the municipal sector5 and among job groups overrepresented in the municipal sector.6

There is a vast body of literature explaining causes of sickness absence within the countries of Sweden16–20 and Denmark.5,6,21–30 However, no studies have been performed to explain the causes of the excess sickness absence in Sweden vs. Denmark. The aim of this study was to investigate potential differences in sickness absence levels assessed with identical sickness absence measures, and to investigate to which degree such a difference was associated with differences in age, gender, skill level, work environment, lifestyle, health or work ability.

Methods

Data

The cross-sectional study is based upon the HAKuL study17 and the Danish Work Environment Study, DWECS.23 Both samples of public sector employees from Sweden and Denmark were selected in 2000 as described in the following.

Sweden

HAKuL started in 1999–2000 in four county councils and six municipalities in the northern, middle and southern parts of Sweden.17 The selection of municipalities was not random but based on interest from the municipalities to participate in the project. Information on occupation and age was derived from employers’ registers. The study population represents almost all types of employees as well as the age and sex distribution of county councils and municipalities in Sweden. A comprehensive questionnaire was constructed using validated instruments as far as possible. The self-administered questionnaire included questions about individual factors, social situation, health, lifestyle, work ability and physical and psychosocial work environment. In year 2000, the questionnaire was distributed to 9003 employees with permanent employment, who at the time of inclusion were not on long-term sick leave for 3 months or longer. A total of 7533 employees (84%) completed and returned the baseline questionnaire. Of these, 511 employees were excluded due to missing data on either sickness absence or type of occupation. Accordingly, 7022 employees were included in the present study.

Denmark

DWECS23 features a random sample of 11 437 people living in Denmark, of whom 8583 (75%) participated in telephone interviews. Of these, 5357 were employees aged 18–69 who had not been sickness absent for at least 8 weeks prior to interview in 2000. A total of 1584 of the interviewed employees were employed in the municipal sector. Of these, 44 were excluded due to missing values on important variables, leaving 1540 for basis of analysis.

The HAKNAK study population thus consists of those 7022 Swedish and 1540 Danish public sector employees supplying information on sickness absence and skill level, totalling 8562 employees.

Sickness absence compensation systems in Sweden and Denmark

As described below, sickness absence systems were compatible during the time of study, even though the two systems are based on different models: Denmark practices the ‘Flexicurity model’, which conserves the employers’ right from the so-called September Settlement off 1899 to freely hire and fire their employees. Sweden has introduced an active labour market policy that, among other things, protects employees against dismissal during sickness. All subjects in the Swedish cohort were covered by the Swedish national health insurance, which provides sickness benefits to those who fulfil the requirement of being unable to work due to disease or injury. At the time of this study, compensation for the first 14 days of a sick-leave spell was paid by the employer, with the exception of the first qualifying (‘waiting’) day, for which no benefit was paid. The first 7 days of absence in a sick-leave spell could be self-certified, thereafter a medical certificate was required. Almost similar conditions applied for Denmark at the time of the study: The initial 14 days of sickness absence was paid by the employer, after which the employer would receive a refund for sickness absence compensation from the State. There was no first qualifying (‘waiting’) day in Denmark at the time—it had been abolished in 1987. The rules with regards to certification were compatible.

The two sub-studies, HAKuL and DWECS, feature an array of common measurements of outcome and determinants of sickness absence. For this study, a working group consisting of researchers from the Swedish and Danish underlying sub-studies was formed, especially focusing on selecting these common features. These are described in the following:

Study outcome

The outcome of this study, ‘Sickness absence in year preceding follow-up’, was based on self-reported days of sickness absence the year preceding the date of follow-up using one question: ‘How many days, approximately, have you been off work for health reasons the last 12 months?’ Days of sickness absence was dichotomized as to identify those with 0–7 days of accumulated days of sickness absence per annum vs. those with more than 7 days.

Determinants

Age and gender

The study includes data on gender and age of the individual employee.

Occupational skills

Occupational skill level was classified according to the International Standard Classification of Occupations, ISCO-88.31 Fifteen levels were identified (table 1).

Table 1

Characteristics of the study population

 Country 
 Sweden Denmark 
N 7022 1540 
Age, mean years 45.7 41.7 
Gender,% females 82.9 75.3 
Occupational skills,% in 15 ISCO-88 job groups   
    Legislators, senior officials and managers 4.1 3.4 
    Clerks 5.1 8.3 
    Personal and protective service workers 39.6 27.7 
    Skilled agricultural and fishery workers 0.8 0.3 
    Craft and related trade workers 2.3 1.0 
    Plant and machine operators and assemblers 0.7 1.3 
    Elementary occupations 5.0 10.5 
    Physical, mathematical and engineering science professionals 0.2 0.7 
    Life science and health professionals 7.0 2.7 
    Teaching professionals 10.0 13.4 
    Other professionals 3.7 2.9 
    Physical and engineering science associate professionals 1.0 3.0 
    Life science and health associate professionals 12.5 8.5 
    Teaching associate professionals 7.0 11.8 
    Other associate professionals 1.0 4.5 
Sickness absence,% with >7 days/year 29.5 24.5 
 Country 
 Sweden Denmark 
N 7022 1540 
Age, mean years 45.7 41.7 
Gender,% females 82.9 75.3 
Occupational skills,% in 15 ISCO-88 job groups   
    Legislators, senior officials and managers 4.1 3.4 
    Clerks 5.1 8.3 
    Personal and protective service workers 39.6 27.7 
    Skilled agricultural and fishery workers 0.8 0.3 
    Craft and related trade workers 2.3 1.0 
    Plant and machine operators and assemblers 0.7 1.3 
    Elementary occupations 5.0 10.5 
    Physical, mathematical and engineering science professionals 0.2 0.7 
    Life science and health professionals 7.0 2.7 
    Teaching professionals 10.0 13.4 
    Other professionals 3.7 2.9 
    Physical and engineering science associate professionals 1.0 3.0 
    Life science and health associate professionals 12.5 8.5 
    Teaching associate professionals 7.0 11.8 
    Other associate professionals 1.0 4.5 
Sickness absence,% with >7 days/year 29.5 24.5 

Lifestyle

Lifestyle factors were measured in term of smoking status (current smokers, previous smokers and never), body mass index (underweight was BMI < 18.5 kg/m2, normal weight was 18.5 ≤ BMI < 25 kg/m2, overweight was 25 ≤ BMI < 30 and obesity was BMI ≥ 30 kg/m2) and physical activity (active or passive).

Physical work environment

Extreme bending of back was divided into two categories according to work time spent in a position with extreme bending of the back: ‘No’ (never) and ‘Yes’ (others).

Working with arms above shoulders was divided into two categories according to work time spent in a position with arms lifted above shoulder height: ‘No’ (never) and ‘Yes’ (others).

Lifting of heavy loads was divided into two categories according to frequency of lifting: ‘No’ (never) and ‘Yes’ (others).

Psychosocial work environment

Quantitative demands was measured with one item: ‘Do you have to work very fast?’ The response options were dichotomized into the categories ‘High’ (‘always’ or ‘often’) and ‘Low’ (‘sometimes’, ‘rarely’ or ‘never’).

Decision authority was measured with one item: ‘Have you got influence on what you do in your job?’ Response options were categorized in ‘High’ (‘always’, ‘often’ or ‘sometimes’) and ‘Low’ (‘rarely’ or ‘never’).

Management quality was assessed with one item ‘Is your work recognized and appreciated by the management?’ and dichotomized: ‘High’ (response options ‘in a high degree’ and ‘partly’) and ‘Low’ (response options ‘in a low degree’ and ‘in a very low degree’).

Health status

The health status of employees was considered in terms of musculoskeletal symptoms and self-rated health. Three types of musculoskeletal symptoms were identified: Neck symptoms previous 12 months, low back symptoms during the last 12 months and knee symptoms during the last 12 months. Response options were ‘yes’/‘no’.

Global Self-rated Health (SRH) was measures in DWECS using a single question: ‘How do you rate your health in general?’ with five response options (‘very good’, ‘good’, ‘fair’, ‘poor’, ‘very poor’). In the analyses the categories poor and very poor were collapsed into one category: ‘poor’. In HAKuL the response options where ‘excellent’, ‘very good’, ‘good’, ‘fair’, ‘poor’ and in the analyses ‘excellent’ and ‘very good’ were collapsed into one category ‘very good’.

Work ability

Work ability was measured with one question on self-rated work ability score on a scale ranging from 0 to 10, a measure previously used in cross country comparisons of sickness absence.32 The respondents were asked to rate their work ability between 0 (‘cannot work at all’) and 10 (‘has never been better’) according to the question: ‘Consider that your work ability was rated 10 when it was best. How would you rate your current work ability?’

Analysis

Initial χ2 test was performed to identify potential differences in distribution of dependent and independent variables in the two sub-cohorts. To establish which factors were significantly associated with the outcome in the two sub-cohorts, country stratified logistic regression analysis was performed, yielding mutually adjusted odds ratios (ORs) with 95% confidence intervals (95% CIs) for each independent variable within Sweden and Denmark. Finally, logistic regression analysis was performed on the total data set consisting of the Swedish and Danish sub-samples to study differences in sickness absence levels between the two sub-cohorts. This part of the analysis was done in eight steps, adjusting the 95% ORs with CI 95% as follows: (i) Gender and age, (ii) Gender, age and occupational skill level, (iii) Gender, age, occupational skill level and lifestyle factors, (iv) Gender, age, occupational skill level and psychosocial work environment factors, (v) Gender, age, occupational skill level and physical work environment factors, (vi) Gender, age, occupational skill level and musculoskeletal symptoms, (vii) Gender, age, occupational skill level and self-rated health and (viii) Gender, age, occupational skill level and work ability.

Results

The Swedish sub-cohort consisted of relatively more women, and the mean age was higher, compared to the study participants from Denmark. There were certain variations with regards to composition of occupations, but the overall picture was similar in the two countries, where personal and protective service workers (including health care personnel) and teaching professionals totalled roughly 50% of both sub-cohorts. More subjects from the Swedish study population than from the Danish reported more than 7 days of sickness absence (table 1).

There were significant differences on all variables related to lifestyle, health and work environment in the two countries: more Danish study participants had a normal BMI, whereas more Swedish study participants than Danish study participants tended to be overweight or obese. More Danish study participants tended to be underweight. Smoking was more pronounced among Danish study participants, who on the other hand tended to be more physically active. More Swedish study participants complained about musculoskeletal symptoms in neck, low back or knees. More Danish study participants than Swedish study participants reported high-quantitative demands and low-decision authority, whereas low-management quality was reported profoundly more often in Sweden. More Swedish study participants reported working positions straining the back and having heavy lifting, whereas work postures implying having arms lifted above shoulder height was reported more often among the Danish study participants. Danish study participants scored higher on self-rated health and work ability scores (table 2).

Table 2

Frequency of exposure levels and differences between Sweden and Denmark

 Sweden n (%) Denmark n (%) P-value 
Body mass index    
    Underweight 6.21 11.44 <0.0001 
    Obesity 10.10 6.44  
    Overweight 32.14 26.63  
    Normal 51.55 55.49  
Smoking    
    Smokers 26.90 35.41 <0.0001 
    Ex-smokers 26.54 23.72  
    Non-smokers 46.56 40.87  
Physical activity    
    Passive 15.51 11.83 0.0002 
    Active 84.49 88.17  
Work environment    
    Quantitative demands (High) 19.47 32.92 <0.0001 
    Decision authority (Low) 66.4 78.72 <0.0001 
    Management quality (Low) 29.99 5.63 <0.0001 
    Extreme bending of back (Yes) 68.63 48.5 <0.0001 
    Work w. arms above shoulder (Yes) 32.89 37.48 0.0006 
    Lifting heavy loads (Yes) 49.98 34.98 <0.0001 
Health    
    Neck symptoms (Yes) 56.43 45.06 <0.0001 
    Low back symptoms (Yes) 56.09 51.63 0.0014 
    Knee symptoms (Yes) 30.42 26.27 0.0013 
    Self-rated health 
        Poor 2.50 2.14 <0.0001 
        Fair 18.85 12.60  
        Good 39.78 43.64  
        Very good 38.87 41.62  
Work ability    
    Low score(0–7) 28.59 11.1 <0.0001 
    Medium score(8–9) 47.06 46.36  
    High score (10) 24.36 42.54  
    Mean (SD) 8.0 (2.0) 8.8 (1.5)  
 Sweden n (%) Denmark n (%) P-value 
Body mass index    
    Underweight 6.21 11.44 <0.0001 
    Obesity 10.10 6.44  
    Overweight 32.14 26.63  
    Normal 51.55 55.49  
Smoking    
    Smokers 26.90 35.41 <0.0001 
    Ex-smokers 26.54 23.72  
    Non-smokers 46.56 40.87  
Physical activity    
    Passive 15.51 11.83 0.0002 
    Active 84.49 88.17  
Work environment    
    Quantitative demands (High) 19.47 32.92 <0.0001 
    Decision authority (Low) 66.4 78.72 <0.0001 
    Management quality (Low) 29.99 5.63 <0.0001 
    Extreme bending of back (Yes) 68.63 48.5 <0.0001 
    Work w. arms above shoulder (Yes) 32.89 37.48 0.0006 
    Lifting heavy loads (Yes) 49.98 34.98 <0.0001 
Health    
    Neck symptoms (Yes) 56.43 45.06 <0.0001 
    Low back symptoms (Yes) 56.09 51.63 0.0014 
    Knee symptoms (Yes) 30.42 26.27 0.0013 
    Self-rated health 
        Poor 2.50 2.14 <0.0001 
        Fair 18.85 12.60  
        Good 39.78 43.64  
        Very good 38.87 41.62  
Work ability    
    Low score(0–7) 28.59 11.1 <0.0001 
    Medium score(8–9) 47.06 46.36  
    High score (10) 24.36 42.54  
    Mean (SD) 8.0 (2.0) 8.8 (1.5)  

Similar patterns for causes for sickness absence were observed in the two countries. There was an increased risk associated with BMI above normal. The risk associated with overweight was not significant in Denmark, and the association with underweight was not significant in either country. Both former and current smoking was associated with absence in both countries, seemingly more pronounced among the Danish study participants. Leisure time physical inactivity was associated with increased risk of sickness absence only among the Swedish study participants, whereas no association was found among the Danish study participants. This could possibly be due to small sample size. Musculoskeletal symptoms in the neck, low back or knees were associated with increased absence in both Sweden and Denmark. As was extreme bending of the back, working with arms lifted above the shoulders and lifting heavy loads. However, the two latter exposures were not significantly associated with increased absence in the Danish sub-cohort, possibly due to sample size issues. There was a strong association between self-rated health score and sickness absence, and workability score and sickness absence, both most pronounced among the Swedish study participants (table 3).

Table 3

Sickness absence >7 days in relation to lifestyle, physical and psychosocial work environment and health status in Sweden (n = 7022) and Denmark (n = 1540), mutually adjusted OR with 95% CI

 Sweden
 
Denmark
 
 n Cases OR (95% Ci) n Cases OR (95% Ci) 
Body mass index       
    Underweight 427 125 1.18 (0.94–1.47) 180 48 1.17 (0.79–1.71) 
    Obesity 696 285 1.84 (1.55–2.19) 98 34 1.99 (1.25–3.17) 
    Overweight 2211 701 1.32 (1.17–1.48) 415 107 1.32 (0.99–1.76) 
    Normal 3545 919 1.00 859 187 1.00 
Smoking status 
    Current 1872 643 1.34 (1.18–1.52) 558 158 1.61 (1.22–2.13) 
    Former 1845 562 1.20 (1.06–1.37) 372 97 1.54 (1.12–2.11) 
    Never 3237 845 1.00 641 125 1.00 
Physical activity 
    Passive 1083 354 1.18 (1.02–1.36) 186 46 0.95 (0.66–1.37) 
    Active 5897 1710 1.00 1384 334 1.00 
Neck symptoms 
    Yes 3915 1343 1.62 (1.45–1.81) 708 198 1.39 (1.09–1.76) 
    No 3026 709 1.00 862 182 1.00 
Low back symptoms 
    Yes 3896 1323 1.58 (1.41–1.76) 806 235 1.69 (1.33–2.16) 
    No 3049 726 1.00 764 145 1.00 
Knee symptoms 
    Yes 2094 739 1.44 (1.29–1.61) 411 120 1.49 (1.14-1.93) 
    No 4801 1292 1.00 1159 260 1.00 
Quantitative demands 
    High 1361 454 1.28 (1.12–1.46) 512 128 1.16 (0.90–1.50) 
    Low 5637 1616 1.00 1057 251 1.00 
Decision authority 
    Low 4647 1331 0.92 (0.82–1.03) 1232 289 0.97 (0.72–1.30) 
    High 2343 734 1.00 337 89 1.00 
Management quality 
    Low 2059 653 1.18 (1.05–1.32) 88 32 1.68 (1.05–2.69) 
    High 4803 1367 1.00 1469 343 1.00 
Extreme bending of back 
    Yes 4785 1502 1.26 (1.10–1.43) 758 210 1.34 (1.04–1.73) 
    No 2185 558 1.00 810 169 1.00 
Working with arms above shoulders 
    Yes 2292 813 1.43 (1.27–1.60) 585 167 1.26 (0.98–1.63) 
    No 4675 1244 1.00 984 213 1.00 
Lifting heavy loads 
    Yes 3480 1127 1.13 (1.00–1.28) 543 150 1.21 (0.93–1.58) 
    No 3494 931 1.00 1003 2241 1.00 
Self-rated health 
    Poor 175 127 12.94 (9.11–18.38) 33 22 9.25 (4.25–20.14) 
    Fair 1325 658 4.75 (4.08–5.52) 198 77 2.99 (2.07–4.30) 
    Good 2790 841 2.16 (1.89–2.46) 688 155 1.26 (0.96–1.65) 
    Very good 2728 445 1.00 653 126 1.00 
Work ability 
    0 (lowest) 100 93 69.45 (31.75–151.91) * (*) 
    1 23 20 42.43 (12.39–145.32) * (*) 
    2 43 37 37.00 (15.34–89.24) * (*) 
    3 89 55 9.24 (5.86–14.58) 3.24 (0.69–15.15) 
    4 101 58 7.53 (4.93–11.49) 11 2.47 (0.66–9.23) 
    5 394 212 6.19 (4.86–7.89) 34 14 2.96 (1.41–6.21) 
    6 352 189 6.55 (5.09–8.44) 21 10 3.80 (1.48–9.73) 
    7 872 329 3.46 (2.85–4.20) 85 25 1.94 (1.14–3.30) 
    8 1705 483 2.24 (1.88–2.66) 333 93 1.81 (1.30–2.51) 
    9 1543 311 1.45 (1.21–1.75) 369 86 1.36 (0.99–1.87) 
    10 (highest) 1683 257 1.00 644 118 1.00 
 Sweden
 
Denmark
 
 n Cases OR (95% Ci) n Cases OR (95% Ci) 
Body mass index       
    Underweight 427 125 1.18 (0.94–1.47) 180 48 1.17 (0.79–1.71) 
    Obesity 696 285 1.84 (1.55–2.19) 98 34 1.99 (1.25–3.17) 
    Overweight 2211 701 1.32 (1.17–1.48) 415 107 1.32 (0.99–1.76) 
    Normal 3545 919 1.00 859 187 1.00 
Smoking status 
    Current 1872 643 1.34 (1.18–1.52) 558 158 1.61 (1.22–2.13) 
    Former 1845 562 1.20 (1.06–1.37) 372 97 1.54 (1.12–2.11) 
    Never 3237 845 1.00 641 125 1.00 
Physical activity 
    Passive 1083 354 1.18 (1.02–1.36) 186 46 0.95 (0.66–1.37) 
    Active 5897 1710 1.00 1384 334 1.00 
Neck symptoms 
    Yes 3915 1343 1.62 (1.45–1.81) 708 198 1.39 (1.09–1.76) 
    No 3026 709 1.00 862 182 1.00 
Low back symptoms 
    Yes 3896 1323 1.58 (1.41–1.76) 806 235 1.69 (1.33–2.16) 
    No 3049 726 1.00 764 145 1.00 
Knee symptoms 
    Yes 2094 739 1.44 (1.29–1.61) 411 120 1.49 (1.14-1.93) 
    No 4801 1292 1.00 1159 260 1.00 
Quantitative demands 
    High 1361 454 1.28 (1.12–1.46) 512 128 1.16 (0.90–1.50) 
    Low 5637 1616 1.00 1057 251 1.00 
Decision authority 
    Low 4647 1331 0.92 (0.82–1.03) 1232 289 0.97 (0.72–1.30) 
    High 2343 734 1.00 337 89 1.00 
Management quality 
    Low 2059 653 1.18 (1.05–1.32) 88 32 1.68 (1.05–2.69) 
    High 4803 1367 1.00 1469 343 1.00 
Extreme bending of back 
    Yes 4785 1502 1.26 (1.10–1.43) 758 210 1.34 (1.04–1.73) 
    No 2185 558 1.00 810 169 1.00 
Working with arms above shoulders 
    Yes 2292 813 1.43 (1.27–1.60) 585 167 1.26 (0.98–1.63) 
    No 4675 1244 1.00 984 213 1.00 
Lifting heavy loads 
    Yes 3480 1127 1.13 (1.00–1.28) 543 150 1.21 (0.93–1.58) 
    No 3494 931 1.00 1003 2241 1.00 
Self-rated health 
    Poor 175 127 12.94 (9.11–18.38) 33 22 9.25 (4.25–20.14) 
    Fair 1325 658 4.75 (4.08–5.52) 198 77 2.99 (2.07–4.30) 
    Good 2790 841 2.16 (1.89–2.46) 688 155 1.26 (0.96–1.65) 
    Very good 2728 445 1.00 653 126 1.00 
Work ability 
    0 (lowest) 100 93 69.45 (31.75–151.91) * (*) 
    1 23 20 42.43 (12.39–145.32) * (*) 
    2 43 37 37.00 (15.34–89.24) * (*) 
    3 89 55 9.24 (5.86–14.58) 3.24 (0.69–15.15) 
    4 101 58 7.53 (4.93–11.49) 11 2.47 (0.66–9.23) 
    5 394 212 6.19 (4.86–7.89) 34 14 2.96 (1.41–6.21) 
    6 352 189 6.55 (5.09–8.44) 21 10 3.80 (1.48–9.73) 
    7 872 329 3.46 (2.85–4.20) 85 25 1.94 (1.14–3.30) 
    8 1705 483 2.24 (1.88–2.66) 333 93 1.81 (1.30–2.51) 
    9 1543 311 1.45 (1.21–1.75) 369 86 1.36 (0.99–1.87) 
    10 (highest) 1683 257 1.00 644 118 1.00 

According to table 4, the differences in sickness absence could not be attributed to differences in occupational skill level composition in the two sub-cohorts. Inclusion of lifestyle factors also had no association with the excess absence in Sweden. In models including occupational skill level and psychosocial or physical work environment respectively, the sickness absence estimate was still significantly higher in Sweden. Musculoskeletal symptoms or self-rated health score did not affect the difference either. However, the introduction of work ability score in a model also containing the effects of age, gender and occupational skill level, caused the differences between to two countries to become statistically insignificant (table 4).

Table 4

Risk (OR and 95% CI) for sickness absence in Sweden vs. Denmark adjusted for other risk factors

OR (95%CI) Adjusted for 
1.26 (1.11–1.43) Age, gender 
1.22 (1.07–1.39) Age, gender, occupational skill level 
1.25 (1.10–1.43) Age, gender, occupational skill level, lifestyle factors 
1.19 (1.04–1.37) Age, gender, occupational skill level, psychosocial work environment 
1.18 (1.03–1.35) Age, gender, occupational skill level, physical work environment 
1.15 (1.01–1.31) Age, gender, occupational skill level, musculoskeletal symptoms 
1.18 (1.03–1.34) Age, gender, occupational skill level, self-rated health 
0.92 (0.80–1.06) Age, gender, occupational skill level, work ability score 
OR (95%CI) Adjusted for 
1.26 (1.11–1.43) Age, gender 
1.22 (1.07–1.39) Age, gender, occupational skill level 
1.25 (1.10–1.43) Age, gender, occupational skill level, lifestyle factors 
1.19 (1.04–1.37) Age, gender, occupational skill level, psychosocial work environment 
1.18 (1.03–1.35) Age, gender, occupational skill level, physical work environment 
1.15 (1.01–1.31) Age, gender, occupational skill level, musculoskeletal symptoms 
1.18 (1.03–1.34) Age, gender, occupational skill level, self-rated health 
0.92 (0.80–1.06) Age, gender, occupational skill level, work ability score 

Discussion

More Swedish study participants than Danish study participants reported more than 7 days of sickness absence. Causes for sickness absence were largely similar in the two countries: BMI above normal (Sweden only), former and current smoking, leisure time physical inactivity (Sweden only), musculoskeletal symptoms in the neck, low back or knees, extreme bending of the back, working with arms lifted above the shoulders (Sweden only) and lifting heavy loads (Sweden only) and low self-rated health score all increased the risk of sickness absence. The difference in absence level between Sweden and Denmark was not associated with differences in composition gender, age, skill level, lifestyle, psychosocial or physical work environment, musculoskeletal symptoms or self-rated health, whereas taking into account differences in work ability score caused the differences between to two countries to become statistically insignificant.

Design

This is the first cross national study addressing causes of sickness absence among public sector employees in Sweden and Denmark, and furthermore the only study addressing the underlying causes for this difference. One should take the following study limitations into account, when interpreting the results: Firstly, methods of selection into the two national cohorts differ: In Denmark, study participants were randomly drawn from the Central Population Register, and are representative for all employees working in the municipal sector in Denmark. In Sweden, the selection of participants was not random but was based on interest from the employing municipalities to participate in the project. This could theoretically select municipalities with focus on, and resources for, employee absence. However, we do not think that this has resulted in a selection of workplaces that either functioned very well or had a lot of problems and the authors do not think it likely that this affects the conclusions of the present study.

Although the questions from both Denmark and Sweden originated from the same validated instruments used in several other investigations we are aware that the wording in some of the questions somewhat differed slightly. However, we do not think that this should affect the results to a great extent but it should be taken into account when interpreting our results.

The study utilizes a cross-sectional design which provides information about a possible association between the exposures and sickness absence but not about the direction of a possible cause–effect relation. If the reporting of lifestyle, work environment and health related factors are affected by the level of sickness absence this could lead either to an over or an underestimation of the associations, whereas the impact on the analysis of the international sickness absence difference is assumed to be minimal. This is also the case with the bias potentially caused by common method variance due to gathering exposure and outcome data with the same data source. We recognize that this study only covers a small array of factors known to affect sickness absence, leaving considerable room for residual confounding.

The categorization of sickness absence should be commented upon: The dichotomization around 7 days does not specifically capture the aspects associated with the longest sickness absence cases. The dichotomization was chosen because of the small sample size for the Danish sub-population. It was not possible to establish useful (sufficiently large) categories of subjects with long-term sickness absence in the Danish sample. We used the dichotomization around 7 days, as this has been used in other previous studies,40 as this identifies the quartile (roughly) with most sickness absence.

With regard to the use of self-reported data on sickness absence to establish outcome only a few studies feature comparisons of self-reported data on sickness absence and data from employer records.33–38 Some of these studies are based on small populations within specific occupations (coal miners, office and production work33–36) and two studies focus on sickness absence due to back pain.33,34 Nearly all report a high specificity of a single question for detecting workers sickness absence but one author suggests not using a recall period of more than 2 months.35 Two more recent studies based on larger cohorts have compared the total number of retrospectively self-reported sickness absence days for 12 months with register data for the same period from the employers’ registers on sickness absence irrespective of diagnoses. Both studies concluded that there were a relatively good agreement between self-reported and register data on sickness absence and that self-reported information can be useful when no information from employers’ registers are available.37,38 In relation to the present study there is no reason to believe, that any possible systematic over or underestimation of sickness absence should differ between to two sub-cohorts.

Finally, we cannot rule out that the different data collection methods (self-administered questionnaire in Sweden, telephone interview in Denmark) can affect the reporting of items. A recent study by Feveile and colleagues has found this to be the case for certain measures of health.39

Other studies

Previous studies have found larger differences in sickness absence between the two countries.3,4 These differences might be partly due to differences in data collection methods, partly the fact that this study only focuses on employees in the municipal sector, and, lastly, the selection criteria for this study: We only included those in work. This implies that people with an ongoing case of long-term sickness would not be included in the present study.

To our knowledge no other studies have examined possible differences of sickness absence between Denmark and Sweden and to what degree these differences could be attributed to by gender composition, work environment, lifestyle, global self-rated health or work ability. A recent cross-sectional study of associations between ill health, unemployment and retirement has shown a related trend: Ill health seemed to be unrelated to unemployment in Sweden, whereas there was a significant association in Denmark.40

Conclusion

The difference in absence level between Sweden and Denmark was not associated with differences in job level composition, lifestyle factors, psychosocial or physical work environment, musculoskeletal symptoms or self-rated health, whereas taking into account differences in work ability score caused the differences between to two countries to become statistically insignificant. This could indicate an increased retention of employees with health problems in the Swedish labour market compared to the Danish labour market. As suggested by Alavinia and Burdorf,40 a possible explanation for the differences in sickness absence ascertained in this study could be explained by differences in legislation: Sweden practises an active labour policy that increases the opportunities for people with chronic illness to remain in work, implying that Swedish employees are protected against dismissal during sickness absence, whereas this is not the case in Denmark.

Key points
  • Sickness absence is a major public health problem in both Sweden and Denmark.

  • Especially employees in the municipal sector contribute disproportionally to sickness absence in both countries.

  • No studies have investigated the underlying causes for cross country differences in sickness absence in Sweden and Denmark.

  • The present study found that study participants from Sweden had more sickness absence than those from Denmark, and that causes for sickness absence were largely similar in the two countries. Differences in sickness absence were not due to differences in gender, age, skill level, lifestyle, work environment or health, but were associated with differences in work ability score.

  • The study indicates that the higher sickness absence level in Sweden could be due to increased retention of employees with health problems.

References

1
Gimeno
D
Benavides
FG
Benach
J
Amick
BC
3rd
Distribution of sickness absence in the European Union countries
Occup Environ Med
 , 
2004
, vol. 
61
 (pg. 
867
-
9
)
2
Alexanderson
K
Norlund
A
Sickness absence - causes, consequences, and physicians' sickness certification practice. A systematic literature review by the Swedish Council on Technology Assessment in Health Care
Scand J Public Health
 , 
2004
, vol. 
32:
 
Suppl 63
3
Bergendorff
S
Sickness absence in Europe: a comparative study.
International social security association
 , 
2003
 
4
Nyman
K
Palmer
E
Bergendorff
S
Den svenska sjukan − sjukfrånvaron i åtta länder (The Swedish malady − sickness absence in eight countries) [In Swedish].
 
Stockholm
ESO Ministry of Finance
 
Ds, 2002, 49
5
Labriola
M
Lund
T
Burr
H
Prospective study of physical and psychosocial risk factors for sickness absence
Occup Med
 , 
2006
, vol. 
56
 (pg. 
469
-
74
)
6
Lund
T
Labriola
M
Villadsen
E
Who is at risk for long-term sickness absence? A prospective cohort study of Danish employees
Work
 , 
2007
, vol. 
28
 (pg. 
225
-
30
)
7
Floderus
B
Göransson
S
Alexanderson
K
Aronsson
G
Self-estimated life situation in patients on long-term sick leave
J Rehabil Med
 , 
2005
, vol. 
37
 (pg. 
291
-
9
)
8
Borg
K
Hensing
G
Alexanderson
K
Predictive factors for disability pension − an 11-year follow up of young persons on sick leave due to neck, shoulder, or back diagnoses
Scand J Public Health
 , 
2001
, vol. 
29
 (pg. 
104
-
12
)
9
Labriola
M
Lund
T
Self-reported sickness absence as a risk marker of future disability pension. Prospective findings from the DWECS/DREAM study 1990-2004
Int J Med Sci
 , 
2007
, vol. 
4
 (pg. 
153
-
8
)
10
Lund
T
Kivimäki
M
Labriola
M
, et al.  . 
Using administrative sickness absence data as a marker of future disability pension: the prospective DREAM study of Danish private sector employees
Occup Environ Med
 , 
2008
, vol. 
65
 (pg. 
28
-
31
)
11
Långtidssjukskrivna – demografi, arbete, yrke, diagnos, sjukpenningrätt och återgång i arbete 2003, 2005 och 2006.
(Long-term sick-listed - demography, work, occupation, diagnose, eligible for sickness benefit and return to work 2003, 2005 and 2006.) [In Swedish].
 , 
2007
, vol. 
6
 
Swedish Social Insurance Agency
 
2007
12
Knutsson
A
Goine
H
Occupation and unemployment rates as predictors of long term sickness absence in two Swedish counties
Soc Sci Med
 , 
1998
, vol. 
47
 (pg. 
25
-
31
)
13
Brulin
C
Goine
H
Edlund
C
Knutsson
A
Prevalence of long-term sick leave among female home care personnel in northern Sweden
J Occup Rehabil
 , 
1998
, vol. 
8
 (pg. 
103
-
11
)
14
Alexanderson
K
Leijon
M
Åkerlind
I
, et al.  . 
Epidemiology of sickness absence in a Swedish county in 1985, 1986 and 1987. A three year longitudinal study with focus on gender, age and occupation
Scand J Soc Med
 , 
1994
, vol. 
22
 (pg. 
27
-
34
)
15
Nordic Council 2007
Nordic Statistical Yearbook 2007
 
16
Lidwall
U
Skogman Thoursie
P
Gustafsson
R
Lundberg
I
Sickness absence in Sweden
Worklife and Health in Sweden 2004.
 , 
2005
Stockholm
National Institute for Working Life & National Board of Occupational Safety
17
Vingård
E
Lindberg
P
Josephson
M
, et al.  . 
Long-term sick-listing among women in the public sector and its associations with age, social situation, lifestyle and work factors: a three-year follow-up study
Scand J Public Health
 , 
2005
, vol. 
33
 (pg. 
370
-
5
)
18
Voss
M
Floderus
B
Diderichsen
F
How do job characteristics, family situation, domestic work, and life style factors relate to sickness absence? A study based on Sweden Post
J Occup Environ Med
 , 
2004
, vol. 
46
 (pg. 
1134
-
43
)
19
Krantz
G
Lundberg
U
Workload, work stress, and sickness absence in Swedish male and female white-collar employees
Scand J Public Health
 , 
2006
, vol. 
34
 (pg. 
238
-
46
)
20
Voss
M
Josephson
M
Stark
S
, et al.  . 
The influence of household work and of having children on sickness absence among employed women in Sweden
Scand J Public Health
  
(in press)
21
Christensen
KB
Labriola
M
Lund
T
Kivimäki
M
Explaining the social gradient in long term sickness absence. Prospective results from DWECS/DREAM
J Epidemiol Community Health
 , 
2008
, vol. 
62
 (pg. 
181
-
3
)
22
Lund
T
Labriola
M
Christensen
KB
, et al.  . 
Psychosocial work environment exposures as risk factors for long-term sickness absence among Danish employees: results from DWECS/DREAM
J Occup Environ Med
 , 
2005
, vol. 
47
 (pg. 
1141
-
7
)
23
Lund
T
Labriola
M
Christensen
KB
, et al.  . 
Physical work environment risk factors for long term sickness absence: prospective findings among a cohort of 5357 employees in Denmark
Br Med J
 , 
2006
, vol. 
332
 (pg. 
449
-
52
)
24
Christensen
KB
Lund
T
Labriola
M
, et al.  . 
The fraction of long term sickness absence attributable to work environment factors: results from DWECS/DREAM
Occup Environ Med
 , 
2007
, vol. 
64
 (pg. 
487
-
9
)
25
Labriola
M
Christensen
KB
Lund
T
, et al.  . 
Multilevel analysis of workplace and individual risk factors for long-term sickness absence
J Occup Environ Med
 , 
2006
, vol. 
48
 (pg. 
923
-
9
)
26
Bultmann
U
Rugulies
R
Lund
T
, et al.  . 
Depressive symptoms and the risk of long-term sickness absence: a prospective study among 4747 employees in Denmark
Soc Psychiatry Psychiatr Epidemiol
 , 
2006
, vol. 
41
 (pg. 
875
-
80
)
27
Christensen
KB
Lund
T
Labriola
M
, et al.  . 
The impact of health behaviour on long term sickness absence: results from DWECS/DREAM
Ind Health
 , 
2007
, vol. 
45
 (pg. 
348
-
51
)
28
Nielsen
ML
Rugulies
R
Christensen
KB
, et al.  . 
Impact of the psychosocial work environment on registered absence from work: a two-year longitudinal study using the IPAW cohort
Work Stress
 , 
2004
, vol. 
18
 (pg. 
323
-
35
)
29
Nielsen
ML
Rugulies
R
Smith-Hansen
L
, et al.  . 
Psychosocial work environment and registered absence from work: estimating the etiologic fraction
Am J Ind Med
 , 
2006
, vol. 
49
 (pg. 
187
-
96
)
30
Christensen
KB
Nielsen
ML
Rugulies
R
, et al.  . 
Workplace levels of psychosocial factors as prospective predictors of registered sickness absence
J Occup Environ Med
 , 
2005
, vol. 
47
 (pg. 
933
-
40
)
32
Bloch
FS
Prins
R
Who returns to work and why? A six-country study on work incapacity and reintegration.
 , 
2001
New Brunswick, New Jersey
Transaction Publishers
33
Agius
RM
Lloyd
MH
Campbell
S
, et al.  . 
Questionnaire for the identification of back pain for epidemiological purposes
Occup Environ Med
 , 
1994
, vol. 
51
 (pg. 
756
-
60
)
34
Burdorf
A
Post
W
Bruggeling
T
Reliability of a questionnaire on sickness absence with specific attention to absence due to back pain and respiratory complaints
Occup Environ Med
 , 
1996
, vol. 
53
 (pg. 
58
-
62
)
35
Severens
JL
Mulder
J
Laheij
RJ
Verbeek
AL
Precision and accuracy in measuring absence from work as a basis for calculating productivity costs in The Netherlands
Soc Sci Med
 , 
2000
, vol. 
51
 (pg. 
243
-
349
)
36
van Poppel
MN
De Vet
HC
Koes
BW
, et al.  . 
Measuring sick leave: a comparison of self-reported data on sick leave and data from employer records
Occup Med
 , 
2002
, vol. 
52
 (pg. 
485
-
90
)
37
Ferrie
JE
Kivimäki
M
Head
J
, et al.  . 
A comparison of self-reported sickness absence with absences recorded in employers' registers: evidence from the Whitehall II study
Occup Environ Med
 , 
2005
, vol. 
62
 (pg. 
74
-
9
)
38
Voss
M
Stark
S
Alfredsson
L
, et al.  . 
Comparisons of self-reported and register data on sickness absence among public employees in Sweden
Occup Environ Med
 , 
2008
, vol. 
65
 (pg. 
61
-
7
)
39
Feveile
H
Olsen
O
Hogh
A
A randomized trial of mailed questionnaires versus telephone interviews: response patterns in a survey
BMC Med Res Methodol
 , 
2007
, vol. 
7
 pg. 
27
 
40
Alavinia
SM
Burdorf
A
Unemployment and retirement and ill-health: a cross-sectional analysis across European countries
Int Arch Occup Environ Health
 , 
2008
, vol. 
82
 (pg. 
39
-
45
)

Comments

0 Comments