- Split View
-
Views
-
Cite
Cite
Annika Åhs, Ragnar Westerling, Self-rated health in relation to employment status during periods of high and of low levels of unemployment, European Journal of Public Health, Volume 16, Issue 3, June 2006, Pages 294–304, https://doi.org/10.1093/eurpub/cki165
- Share Icon Share
Abstract
Background: There is a need for more research on the health impact of changes in the national unemployment rate. Therefore, the present study was carried out to compare levels of self-rated health during periods of high and low levels of unemployment. Methods: Data included cross-sectional interviews from the Swedish Survey of Living Conditions, which were based on random samples of inhabitants between 16 and 64 years of age living in Sweden. Data were collected for the period 1983–89, when unemployment levels were low (n = 35 562; 2.5%) and for the period 1992–97 when unemployment was high (n = 24 019; 7.1%). Results: After adjusting for sociodemographic variables as well as long-term disease or handicap, the differences in self-rated health between the unemployed and employed were larger when unemployment levels were high in the 1990s, than when they were low in the 1980s. More groups of the unemployed were afflicted with poor health when unemployment was high, compared with when it was low. In 1992–97, being married, living in larger cities, or not having a long-term disease or handicap no longer buffered the negative effects on health among the unemployed. Conclusions: Poorer self-rated health among the unemployed seems to be an increasing public health problem during high levels of unemployment.
There is a link between unemployment, poor health and mortality.1–3 Unemployment may affect both psychological and physiological risk factors for ill health. This may be mediated by the social contexts in which unemployment is experienced, by changes in social support,4 economy5 or altered risk behaviours.6,7 Illness or socioeconomic factors existing before the onset of unemployment may also make the unemployed more susceptible to illness or mortality.8–11 This selection into unemployment may be more marked in times of low levels of unemployment, since more severely disadvantaged groups in society may suffer from unemployment during such periods.1,9,12 When it is difficult to get a job, unemployment may affect different groups in the population, including more fit and healthy individuals. The health consequences of unemployment may therefore be reduced or unchanged during periods of high unemployment.
During the 1990s Sweden was transformed from a country of low employment to one of high unemployment. The economic crisis characterizing the labour force and the higher number of psychologically demanding jobs may possibly have lead to an intensification of the negative effects on health among different groups in the labour force.13 The reductions in welfare may have affected the unemployed to a larger extent than other groups in the population.13–16 However, there is evidence indicating that the health of young long-term unemployed did not change significantly during the 1980s and mid-1990s,17 although young persons who were not unemployed reported poorer health during the recession.18 Studies of changes in socioeconomic health inequalities over time in the Nordic countries, show that there was no immediate deterioration in public health in terms of long-standing illness and global self-rated health until the mid-1990s, and that the health inequalities in relation to employment status groups remained constant.19–22 The consequences of the high rates of unemployment have to be further analysed for the period of the mid-1990s and on, since many of the budget deficits were implemented during the late 1990s.20 Research should also address that the group of unemployed was more heterogeneous during the 1990s.
The aim of this study was to estimate the differences in self-rated health among the employed and unemployed during one period of low unemployment and one period of high unemployment, as well as to analyse whether self-perceptions of health were determined by long-standing illness and sociodemographic factors other than that of being unemployed.
Methods
Data and general procedure
Data included cross-sectional interviews from the Swedish Survey of Living Conditions, collected during periods of low and of high unemployment. The analyses were based on a random sample of inhabitants living in Sweden and included information about indicators of health and socio-demographic data.
Global self-rated health was measured using the following question: ‘How would you rate your own health? As: good, poor, or something in between?’ In 1996–97, the measure included a 5-point scale ranging from very good to very poor, where good and very good were combined, as were poor and very poor.23 The participants were also asked to rate their ‘age-related health’—their present health compared with persons their own age—as worse, about the same as most or better than most.24,25 The occurrence of a long-term illness or handicap was operationalized as follows: ‘Do you have any long-standing illness, problem due to an accident, handicap or any other weakness? (yes/no)’.
The level of unemployment rose in 1992 and was relatively stable until 1997.26 When studying the global self-rated health, interviews were used from 1983–89 (unemployment level 2.5%) and 1992–97 (unemployment level 7.1%). The results are presented for a total of 35 562 subjects for the period 1983–89 and 24 019 subjects for 1992–97, all between 16 and 64 years of age. Persons who participated during the period 1983–89 and who were interviewed again in 1992–97 were excluded from the analyses at the time of the second interview. The non-response rate was 18.9% for the period 1983–89 and 20.4% for 1992–97.
The question of health in relation to one's peers was only asked in the interviews in 1988–89 (unemployed n = 189, 1.9%; employed n = 8062, 82.9%; outside the labour force n = 1470, 15.1%) and in 1996–97 (unemployed n = 337, 7.5%; employed n = 3276, 72.7%; outside the labour force n = 894, 19.8%).
The employed group included those who responded that during the week preceding the interview they had worked full time or part time, were self-employed, or worked on a farm. The unemployed included those who said that they had been unemployed and/or sought employment during the whole or part of the previous week, and who were not employed. Students or retired persons who were looking for jobs were classified as unemployed.27 The relationship between unemployment and self-rated health did not change significantly when these persons were excluded from the group of unemployed (results not presented). Persons who were classified as neither employed nor unemployed were categorized as ‘persons outside the labour force’. This group included students, housewives, and persons who were on disability pension or who had retired for other reasons, who were performing compulsory military service or who reported ‘doing something else’.
Educational level was based on the number of years of education and included completed primary school (9 years), secondary school (12 years) and post-secondary school (>12 years). The categories for region of residence were defined according to the level of urbanization. Three age groups were formed, while marital status was dichotomized.
Treatment of data
χ2-tests were carried out to compare ratings of global- and age-related health among the employed and unemployed. The levels of significance were determined at the 95.0% level. Multinomial logistic regression analyses were used to adjust for sociodemographic indicators. The results are only presented for poor global self-rated health, compared with good self-rated health (the reference group) and for self-rated health perceived as worse than one's peers, compared with reporting that one's health was ‘about the same’ as others in the same age (the reference group). Separate logistic regression analyses were performed to see whether the results would differ when using another categorization of global self-rated health, i.e. ‘less than good self-rated health’ compared with ‘good self-rated health’.
If the variables were significantly related to global or age-related health in unadjusted regression analyses, they were entered into the models. The independent variables that were highly correlated were not included in the same multivariate model. Post-secondary level education and occupational status were correlated (Spearman's rho 0.42), as were country of birth and citizenship (Spearman's rho 0.62), and cohabiting versus marital status (Spearman's rho 0.71). Since there were no data on the occupational status of the long-term unemployed, educational level was used as a measure of socioeconomic status. Some individuals who were born outside Sweden were now Swedish citizens, which is why country of birth was thought to be a better measure of foreign ethnicity. Marital status instead of cohabiting status was chosen, since marital status was more strongly associated with self-rated health in the unadjusted analyses.
Separate analyses for periods for low and high levels of unemployment were carried out to study differences in self-rated health among the unemployed and employed. Then, the effects of length of unemployment were tested. Analyses were conducted using cut-off points of being unemployed for 1–25 weeks, 26–51 weeks or for ≥52 weeks. To estimate the difference between the odds ratios (ORs) for self-rated health during the 1980s and the 1990s, the interaction terms between employment status and period were estimated. Finally, stratified analyses were carried out for the chosen sociodemographic variables and reports of long-standing illness. These analyses were not performed for the measure of age-related health, owing to the small sample size.
The analytic models
Employment status, sex, age, and country of birth were entered into the first model (Model 1). These variables were assumed not to change due to employment status and adjusting for these variables were thought to control for confounders (i.e. factors that may correlate with both unemployment and self-rated health), which occurred before the start of unemployment. If the result showed an excess risk for poor health among the unemployed despite adjusting for these factors, this would indicate that poor health among the unemployed was not related to such preexisting factors. In Model 2, educational level, region of residence and marital status were added, and long-standing illness was adjusted for in Model 3. These factors might have existed before the onset of unemployment, but may also have been affected by employment status. They could both be confounders or mediating factor (i.e. factors that lie on the pathway through which unemployment exerts its effect.). The purpose of adjusting for the factors in Models 2 and 3 was not to answer the question of whether ill health among the unemployed is explained by a health selection. The analyses were designed to determine whether ratings of self-rated health were stable among the unemployed and independent of these variables.
Results
Descriptive data
Unemployment was significantly higher during the 1990s, than during the 1980s (table 1 and figure 1). Long-term unemployment (≥52 weeks) was more frequent in 1992–97 (n = 487; 28.7%) than in 1983–89 (n = 146; 16.3%), and more frequent during 1996–97 (n = 125; 37.2%) compared with 1988–89 (n = 35; 18.7%).
Employment status . | 1983–89 (n = 35 562) . | . | 1992–97 (n = 24 019) . | . | P for 1983–89/1992–97 . | |||||
---|---|---|---|---|---|---|---|---|---|---|
. | n . | % . | n . | % . | . | |||||
Unemployed | 906 | 2.5 | 1707 | 7.1 | ** | |||||
Employed | 28 725 | 80.8 | 17 851 | 74.3 | ** | |||||
Persons outside the labour force | 5931 | 16.7 | 4461 | 18.6 | ** | |||||
Of these: | ||||||||||
Taking care of household | 3340 | 9.4 | 2585 | 10.8 | ** | |||||
Students | 2999 | 8.4 | 2632 | 11.0 | ** | |||||
Retired | 1773 | 5.0 | 1230 | 5.1 | NS | |||||
Military service | 139 | 0.4 | 105 | 0.4 | NS | |||||
‘Doing something else’ | 88 | 0.2 | 68 | 0.3 | NS |
Employment status . | 1983–89 (n = 35 562) . | . | 1992–97 (n = 24 019) . | . | P for 1983–89/1992–97 . | |||||
---|---|---|---|---|---|---|---|---|---|---|
. | n . | % . | n . | % . | . | |||||
Unemployed | 906 | 2.5 | 1707 | 7.1 | ** | |||||
Employed | 28 725 | 80.8 | 17 851 | 74.3 | ** | |||||
Persons outside the labour force | 5931 | 16.7 | 4461 | 18.6 | ** | |||||
Of these: | ||||||||||
Taking care of household | 3340 | 9.4 | 2585 | 10.8 | ** | |||||
Students | 2999 | 8.4 | 2632 | 11.0 | ** | |||||
Retired | 1773 | 5.0 | 1230 | 5.1 | NS | |||||
Military service | 139 | 0.4 | 105 | 0.4 | NS | |||||
‘Doing something else’ | 88 | 0.2 | 68 | 0.3 | NS |
The percentage figures refer to the proportion of persons within the whole sample. The types of activity among ‘persons outside the labour force’ are not mutually exclusive
NS = not significant
*P-value ≤ 0.05
**P-value ≤ 0.01
Employment status . | 1983–89 (n = 35 562) . | . | 1992–97 (n = 24 019) . | . | P for 1983–89/1992–97 . | |||||
---|---|---|---|---|---|---|---|---|---|---|
. | n . | % . | n . | % . | . | |||||
Unemployed | 906 | 2.5 | 1707 | 7.1 | ** | |||||
Employed | 28 725 | 80.8 | 17 851 | 74.3 | ** | |||||
Persons outside the labour force | 5931 | 16.7 | 4461 | 18.6 | ** | |||||
Of these: | ||||||||||
Taking care of household | 3340 | 9.4 | 2585 | 10.8 | ** | |||||
Students | 2999 | 8.4 | 2632 | 11.0 | ** | |||||
Retired | 1773 | 5.0 | 1230 | 5.1 | NS | |||||
Military service | 139 | 0.4 | 105 | 0.4 | NS | |||||
‘Doing something else’ | 88 | 0.2 | 68 | 0.3 | NS |
Employment status . | 1983–89 (n = 35 562) . | . | 1992–97 (n = 24 019) . | . | P for 1983–89/1992–97 . | |||||
---|---|---|---|---|---|---|---|---|---|---|
. | n . | % . | n . | % . | . | |||||
Unemployed | 906 | 2.5 | 1707 | 7.1 | ** | |||||
Employed | 28 725 | 80.8 | 17 851 | 74.3 | ** | |||||
Persons outside the labour force | 5931 | 16.7 | 4461 | 18.6 | ** | |||||
Of these: | ||||||||||
Taking care of household | 3340 | 9.4 | 2585 | 10.8 | ** | |||||
Students | 2999 | 8.4 | 2632 | 11.0 | ** | |||||
Retired | 1773 | 5.0 | 1230 | 5.1 | NS | |||||
Military service | 139 | 0.4 | 105 | 0.4 | NS | |||||
‘Doing something else’ | 88 | 0.2 | 68 | 0.3 | NS |
The percentage figures refer to the proportion of persons within the whole sample. The types of activity among ‘persons outside the labour force’ are not mutually exclusive
NS = not significant
*P-value ≤ 0.05
**P-value ≤ 0.01
A higher proportion of men were unemployed in 1992–97 (56.4%) than in 1983–89 (46.6%). During the period of high unemployment, more unemployed were born outside Sweden (19.5% versus 15.5%), had a post-secondary educational level (14.5% versus 11.8%), were living in one of the three largest cities in Sweden (29.3% versus 23%) or were studying (12.5% versus 6.7%). During 1992–97, a lower proportion of the unemployed were between the ages of 16 and 25 years (29.3% versus 38.4%), as was the case for those 26–45 years of age (39.1% versus 47.7%).
The risk of poor global self-rated health was significantly higher among the unemployed than among the employed during both periods (table 2). The unemployed, but not the other employment categories, had significantly poorer health during the period of the mid-1990s and on, than during the beginning of the 1990s (unemployed: 1992–94, 3.7%; 1995–97, 6.5%). Viewing one's own health as worse than that of one's peers was significantly more likely among the unemployed during 1996–97, but not during 1988–89. In 1996–97, 13.8% of the unemployed reported having worse health than others of the same age, while 7.1% of the employed did. In 1988–89, 7.7% of the unemployed and 5.9% of the employed reported that their health was worse than others in the same age bracket.
Employment status . | 1983–89 . | 1992–97 . | 1983–89/1992–97 . | 1983–89 . | 1992–97 . | 1983–89/1992–97 . | 1983–89 . | 1992–97 . | 1983–89/1992–97 . | |||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
. | Poor (%) . | Poor (%) . | P . | Fair (%) . | Fair (%) . | P . | Good (%) . | Good (%) . | P . | |||||||||
Employed (n = 46531) | 2.5 | 2.3 | NS | 13.1 | 14.4 | ** | 84.4 | 83.3 | ** | |||||||||
Unemployed (n = 2608) | 4.2 | 4.8 | NS | 19.0 | 21.0 | NS | 76.8 | 74.2 | NS | |||||||||
P unemployed/employed | ** | ** | ** | ** | ** | ** | ||||||||||||
Persons outside the labour force (n = 10 363) | 9.3 | 8.9 | NS | 19.8 | 22.4 | ** | 71 | 68.7 | * | |||||||||
Of these: | ||||||||||||||||||
Retired | 25.4 | 24.3 | NS | 42.0 | 45.9 | * | 32.5 | 29.8 | NS | |||||||||
Students | 1.4 | 1.7 | NS | 6.9 | 12.8 | ** | 91.7 | 85.5 | ** | |||||||||
Taking care of household | 9.6 | 10.2 | NS | 24.6 | 26.1 | NS | 65.8 | 63.8 | NS | |||||||||
‘Doing something else’ | 9.3 | 8.9 | NS | 19.8 | 22.4 | ** | 71 | 68.7 | * |
Employment status . | 1983–89 . | 1992–97 . | 1983–89/1992–97 . | 1983–89 . | 1992–97 . | 1983–89/1992–97 . | 1983–89 . | 1992–97 . | 1983–89/1992–97 . | |||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
. | Poor (%) . | Poor (%) . | P . | Fair (%) . | Fair (%) . | P . | Good (%) . | Good (%) . | P . | |||||||||
Employed (n = 46531) | 2.5 | 2.3 | NS | 13.1 | 14.4 | ** | 84.4 | 83.3 | ** | |||||||||
Unemployed (n = 2608) | 4.2 | 4.8 | NS | 19.0 | 21.0 | NS | 76.8 | 74.2 | NS | |||||||||
P unemployed/employed | ** | ** | ** | ** | ** | ** | ||||||||||||
Persons outside the labour force (n = 10 363) | 9.3 | 8.9 | NS | 19.8 | 22.4 | ** | 71 | 68.7 | * | |||||||||
Of these: | ||||||||||||||||||
Retired | 25.4 | 24.3 | NS | 42.0 | 45.9 | * | 32.5 | 29.8 | NS | |||||||||
Students | 1.4 | 1.7 | NS | 6.9 | 12.8 | ** | 91.7 | 85.5 | ** | |||||||||
Taking care of household | 9.6 | 10.2 | NS | 24.6 | 26.1 | NS | 65.8 | 63.8 | NS | |||||||||
‘Doing something else’ | 9.3 | 8.9 | NS | 19.8 | 22.4 | ** | 71 | 68.7 | * |
Types of activity among ‘persons outside the labour force’ are not mutually exclusive
NS = not significant; *P-value ≤ 0.05; **P-value ≤ 0.01
Employment status . | 1983–89 . | 1992–97 . | 1983–89/1992–97 . | 1983–89 . | 1992–97 . | 1983–89/1992–97 . | 1983–89 . | 1992–97 . | 1983–89/1992–97 . | |||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
. | Poor (%) . | Poor (%) . | P . | Fair (%) . | Fair (%) . | P . | Good (%) . | Good (%) . | P . | |||||||||
Employed (n = 46531) | 2.5 | 2.3 | NS | 13.1 | 14.4 | ** | 84.4 | 83.3 | ** | |||||||||
Unemployed (n = 2608) | 4.2 | 4.8 | NS | 19.0 | 21.0 | NS | 76.8 | 74.2 | NS | |||||||||
P unemployed/employed | ** | ** | ** | ** | ** | ** | ||||||||||||
Persons outside the labour force (n = 10 363) | 9.3 | 8.9 | NS | 19.8 | 22.4 | ** | 71 | 68.7 | * | |||||||||
Of these: | ||||||||||||||||||
Retired | 25.4 | 24.3 | NS | 42.0 | 45.9 | * | 32.5 | 29.8 | NS | |||||||||
Students | 1.4 | 1.7 | NS | 6.9 | 12.8 | ** | 91.7 | 85.5 | ** | |||||||||
Taking care of household | 9.6 | 10.2 | NS | 24.6 | 26.1 | NS | 65.8 | 63.8 | NS | |||||||||
‘Doing something else’ | 9.3 | 8.9 | NS | 19.8 | 22.4 | ** | 71 | 68.7 | * |
Employment status . | 1983–89 . | 1992–97 . | 1983–89/1992–97 . | 1983–89 . | 1992–97 . | 1983–89/1992–97 . | 1983–89 . | 1992–97 . | 1983–89/1992–97 . | |||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
. | Poor (%) . | Poor (%) . | P . | Fair (%) . | Fair (%) . | P . | Good (%) . | Good (%) . | P . | |||||||||
Employed (n = 46531) | 2.5 | 2.3 | NS | 13.1 | 14.4 | ** | 84.4 | 83.3 | ** | |||||||||
Unemployed (n = 2608) | 4.2 | 4.8 | NS | 19.0 | 21.0 | NS | 76.8 | 74.2 | NS | |||||||||
P unemployed/employed | ** | ** | ** | ** | ** | ** | ||||||||||||
Persons outside the labour force (n = 10 363) | 9.3 | 8.9 | NS | 19.8 | 22.4 | ** | 71 | 68.7 | * | |||||||||
Of these: | ||||||||||||||||||
Retired | 25.4 | 24.3 | NS | 42.0 | 45.9 | * | 32.5 | 29.8 | NS | |||||||||
Students | 1.4 | 1.7 | NS | 6.9 | 12.8 | ** | 91.7 | 85.5 | ** | |||||||||
Taking care of household | 9.6 | 10.2 | NS | 24.6 | 26.1 | NS | 65.8 | 63.8 | NS | |||||||||
‘Doing something else’ | 9.3 | 8.9 | NS | 19.8 | 22.4 | ** | 71 | 68.7 | * |
Types of activity among ‘persons outside the labour force’ are not mutually exclusive
NS = not significant; *P-value ≤ 0.05; **P-value ≤ 0.01
A significantly larger proportion of persons who were outside the labour force had a long-term illness during 1983–89 (n = 2591; 50.7%) than in 1992–97 (n = 2524: 49.3%), but in this group there was no difference between the 1980s and 1990s, in poor or ‘worse’ self-rated health.
Adjusting for sociodemographic variables and long-standing illness
Unemployment was a significant independent risk factor for poor global self-rated health, during both high and low periods of unemployment (table 3). In Model 1, the OR for poor self-rated health among the unemployed was 2.20 in 1983–89 and 2.70 in 1992–97. In Model 2, the OR was 2.03 versus 2.55. After allowing for long-standing illness, the unemployed showed a 1.87 times higher risk of poor health, compared with the employed during low levels of unemployment, while they had a 2.67 times higher risk during high levels of unemployment. These results persisted among both the short-term (1–25 weeks) and the long-term unemployed (>52 weeks) (table 4).
Employment status . | 1983–89 . | . | . | . | 1992–97 . | . | . | . | ||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
. | Multivariate logistic regression [OR (95% CI)] . | . | . | . | Multivariate logistic regression [OR (95% CI)] . | . | . | . | ||||||||
. | Unadjusted . | Model 1 . | Model 2 . | Model 3 . | Unadjusted . | Model 1 . | Model 2 . | Model 3 . | ||||||||
Employed | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | ||||||||
Unemployed | 1.84** (1.32–2.58) | 2.20** (1.56–3.09) | 2.03** (1.44–2.87) | 1.87** (1.30–2.69) | 2.37** (1.86–3.03) | 2.70** (2.10–3.48) | 2.55** (1.98–3.29) | 2.67** (2.05–3.48) | ||||||||
Sex | ||||||||||||||||
Men | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | ||||||||
Women | 1.17** (1.04–1.30) | 1.21** (1.05–1.40) | 1.23** (1.06–1.42) | 1.21* (1.04–1.41) | 1.24** (1.09–1.42) | 1.12 NS (0.93–1.34) | 1.14 NS (0.95–1.37) | 1.12 NS (0.93–1.35) | ||||||||
Age | ||||||||||||||||
16–25 years | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | ||||||||
26–45 years | 2.84** (2.22–3.64) | 2.29** (1.67–3.14) | 2.86** (2.07–3.94) | 2.42** (1.74–3.36) | 2.73** (2.04–3.66) | 2.01** (1.35–2.99) | 2.36** (1.57–3.53) | 2.22** (1.47–3.35) | ||||||||
46–64 years | 10.62** (8.37–13.48) | 6.26** (4.59–8.52) | 7.40** (5.34–10.25) | 4.10** (2.93–5.72) | 7.71** (5.81–10.23) | 4.87** (3.29–7.20) | 5.66** (3.76–8.51) | 3.72** (2.45–5.64) | ||||||||
Country of birth | ||||||||||||||||
Born in Sweden | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | ||||||||
Born outside Sweden | 3.17** (2.77–3.64) | 2.93** (2.45–3.52) | 2.94** (2.45–3.54) | 3.34** (2.74–4.08) | 3.45** (2.94–4.04) | 2.81** (2.25–3.52) | 2.93** (2.33–3.67) | 3.13** (2.47–3.97) | ||||||||
Educational level | ||||||||||||||||
Post-secondary education | 1 | 1 | 1 | 1 | 1 | 1 | ||||||||||
Secondary education | 1.69** (1.41–2.03) | 1.77** (1.42–2.20) | 1.64** (1.31–2.06) | 2.48** (2.08–2.97) | 1.40** (1.13–1.74) | 1.32* (1.06–1.64) | ||||||||||
Primary education | 3.25** (2.72–3.88) | 2.40** (1.92–3.01) | 2.24** (1.78–2.84) | 1.36** (1.15–1.60) | 1.97** (1.52–2.55) | 1.84** (1.41–2.41) | ||||||||||
Marital status | ||||||||||||||||
Married | 1 | 1 | 1 | 1 | 1 | 1 | ||||||||||
Not married | 0.74** (0.66–0.83) | 1.34** (1.14–1.56) | 1.33** (1.13–1.57) | 0.73** (0.64–0.84) | 1.28* (1.05–1.55) | 1.22 NS (1.0–1.49) | ||||||||||
Region of residence | ||||||||||||||||
Stockholm, Gothenburg, or Malmö | 1 | 1 | 1 | 1 | 1 | 1 | ||||||||||
Other larger cities | 0.94 NS (0.82–1.08) | 0.93 NS (0.78–1.12) | 0.91 NS (0.76–1.10) | 1.04 NS (0.88–1.22) | 1.09 NS (0.88–1.36) | 0.98 NS (0.78–1.23) | ||||||||||
Other parts of Sweden | 0.95 NS (0.83–1.09) | 0.89 NS (0.74–1.06) | 0.83* (0.68–0.99) | 0.93 NS (0.78–1.10) | 0.97 NS (0.77–1.23) | 0.88 NS (0.69–1.13) | ||||||||||
Limiting longstanding illness or handicap | ||||||||||||||||
No illness or handicap | 1 | 1 | 1 | 1 | ||||||||||||
Illness or handicap | 72.05** (55.75–93.13) | 46.42** (34.73–62.05) | 46.64** (34.99–62.15) | 30.45** (21.86–42.41) |
Employment status . | 1983–89 . | . | . | . | 1992–97 . | . | . | . | ||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
. | Multivariate logistic regression [OR (95% CI)] . | . | . | . | Multivariate logistic regression [OR (95% CI)] . | . | . | . | ||||||||
. | Unadjusted . | Model 1 . | Model 2 . | Model 3 . | Unadjusted . | Model 1 . | Model 2 . | Model 3 . | ||||||||
Employed | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | ||||||||
Unemployed | 1.84** (1.32–2.58) | 2.20** (1.56–3.09) | 2.03** (1.44–2.87) | 1.87** (1.30–2.69) | 2.37** (1.86–3.03) | 2.70** (2.10–3.48) | 2.55** (1.98–3.29) | 2.67** (2.05–3.48) | ||||||||
Sex | ||||||||||||||||
Men | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | ||||||||
Women | 1.17** (1.04–1.30) | 1.21** (1.05–1.40) | 1.23** (1.06–1.42) | 1.21* (1.04–1.41) | 1.24** (1.09–1.42) | 1.12 NS (0.93–1.34) | 1.14 NS (0.95–1.37) | 1.12 NS (0.93–1.35) | ||||||||
Age | ||||||||||||||||
16–25 years | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | ||||||||
26–45 years | 2.84** (2.22–3.64) | 2.29** (1.67–3.14) | 2.86** (2.07–3.94) | 2.42** (1.74–3.36) | 2.73** (2.04–3.66) | 2.01** (1.35–2.99) | 2.36** (1.57–3.53) | 2.22** (1.47–3.35) | ||||||||
46–64 years | 10.62** (8.37–13.48) | 6.26** (4.59–8.52) | 7.40** (5.34–10.25) | 4.10** (2.93–5.72) | 7.71** (5.81–10.23) | 4.87** (3.29–7.20) | 5.66** (3.76–8.51) | 3.72** (2.45–5.64) | ||||||||
Country of birth | ||||||||||||||||
Born in Sweden | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | ||||||||
Born outside Sweden | 3.17** (2.77–3.64) | 2.93** (2.45–3.52) | 2.94** (2.45–3.54) | 3.34** (2.74–4.08) | 3.45** (2.94–4.04) | 2.81** (2.25–3.52) | 2.93** (2.33–3.67) | 3.13** (2.47–3.97) | ||||||||
Educational level | ||||||||||||||||
Post-secondary education | 1 | 1 | 1 | 1 | 1 | 1 | ||||||||||
Secondary education | 1.69** (1.41–2.03) | 1.77** (1.42–2.20) | 1.64** (1.31–2.06) | 2.48** (2.08–2.97) | 1.40** (1.13–1.74) | 1.32* (1.06–1.64) | ||||||||||
Primary education | 3.25** (2.72–3.88) | 2.40** (1.92–3.01) | 2.24** (1.78–2.84) | 1.36** (1.15–1.60) | 1.97** (1.52–2.55) | 1.84** (1.41–2.41) | ||||||||||
Marital status | ||||||||||||||||
Married | 1 | 1 | 1 | 1 | 1 | 1 | ||||||||||
Not married | 0.74** (0.66–0.83) | 1.34** (1.14–1.56) | 1.33** (1.13–1.57) | 0.73** (0.64–0.84) | 1.28* (1.05–1.55) | 1.22 NS (1.0–1.49) | ||||||||||
Region of residence | ||||||||||||||||
Stockholm, Gothenburg, or Malmö | 1 | 1 | 1 | 1 | 1 | 1 | ||||||||||
Other larger cities | 0.94 NS (0.82–1.08) | 0.93 NS (0.78–1.12) | 0.91 NS (0.76–1.10) | 1.04 NS (0.88–1.22) | 1.09 NS (0.88–1.36) | 0.98 NS (0.78–1.23) | ||||||||||
Other parts of Sweden | 0.95 NS (0.83–1.09) | 0.89 NS (0.74–1.06) | 0.83* (0.68–0.99) | 0.93 NS (0.78–1.10) | 0.97 NS (0.77–1.23) | 0.88 NS (0.69–1.13) | ||||||||||
Limiting longstanding illness or handicap | ||||||||||||||||
No illness or handicap | 1 | 1 | 1 | 1 | ||||||||||||
Illness or handicap | 72.05** (55.75–93.13) | 46.42** (34.73–62.05) | 46.64** (34.99–62.15) | 30.45** (21.86–42.41) |
The following variables were controlled for in the different models: Model 1 = sex, age, country of birth; Model 2 = sex, age, country of birth, educational level, marital status, region of residence; Model 3 = sex, age, country of birth, educational level, marital status, region of residence, longstanding illness or handicap
NS = not significant; *P-value ≤ 0.05; **P-value ≤ 0.01
Employment status . | 1983–89 . | . | . | . | 1992–97 . | . | . | . | ||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
. | Multivariate logistic regression [OR (95% CI)] . | . | . | . | Multivariate logistic regression [OR (95% CI)] . | . | . | . | ||||||||
. | Unadjusted . | Model 1 . | Model 2 . | Model 3 . | Unadjusted . | Model 1 . | Model 2 . | Model 3 . | ||||||||
Employed | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | ||||||||
Unemployed | 1.84** (1.32–2.58) | 2.20** (1.56–3.09) | 2.03** (1.44–2.87) | 1.87** (1.30–2.69) | 2.37** (1.86–3.03) | 2.70** (2.10–3.48) | 2.55** (1.98–3.29) | 2.67** (2.05–3.48) | ||||||||
Sex | ||||||||||||||||
Men | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | ||||||||
Women | 1.17** (1.04–1.30) | 1.21** (1.05–1.40) | 1.23** (1.06–1.42) | 1.21* (1.04–1.41) | 1.24** (1.09–1.42) | 1.12 NS (0.93–1.34) | 1.14 NS (0.95–1.37) | 1.12 NS (0.93–1.35) | ||||||||
Age | ||||||||||||||||
16–25 years | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | ||||||||
26–45 years | 2.84** (2.22–3.64) | 2.29** (1.67–3.14) | 2.86** (2.07–3.94) | 2.42** (1.74–3.36) | 2.73** (2.04–3.66) | 2.01** (1.35–2.99) | 2.36** (1.57–3.53) | 2.22** (1.47–3.35) | ||||||||
46–64 years | 10.62** (8.37–13.48) | 6.26** (4.59–8.52) | 7.40** (5.34–10.25) | 4.10** (2.93–5.72) | 7.71** (5.81–10.23) | 4.87** (3.29–7.20) | 5.66** (3.76–8.51) | 3.72** (2.45–5.64) | ||||||||
Country of birth | ||||||||||||||||
Born in Sweden | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | ||||||||
Born outside Sweden | 3.17** (2.77–3.64) | 2.93** (2.45–3.52) | 2.94** (2.45–3.54) | 3.34** (2.74–4.08) | 3.45** (2.94–4.04) | 2.81** (2.25–3.52) | 2.93** (2.33–3.67) | 3.13** (2.47–3.97) | ||||||||
Educational level | ||||||||||||||||
Post-secondary education | 1 | 1 | 1 | 1 | 1 | 1 | ||||||||||
Secondary education | 1.69** (1.41–2.03) | 1.77** (1.42–2.20) | 1.64** (1.31–2.06) | 2.48** (2.08–2.97) | 1.40** (1.13–1.74) | 1.32* (1.06–1.64) | ||||||||||
Primary education | 3.25** (2.72–3.88) | 2.40** (1.92–3.01) | 2.24** (1.78–2.84) | 1.36** (1.15–1.60) | 1.97** (1.52–2.55) | 1.84** (1.41–2.41) | ||||||||||
Marital status | ||||||||||||||||
Married | 1 | 1 | 1 | 1 | 1 | 1 | ||||||||||
Not married | 0.74** (0.66–0.83) | 1.34** (1.14–1.56) | 1.33** (1.13–1.57) | 0.73** (0.64–0.84) | 1.28* (1.05–1.55) | 1.22 NS (1.0–1.49) | ||||||||||
Region of residence | ||||||||||||||||
Stockholm, Gothenburg, or Malmö | 1 | 1 | 1 | 1 | 1 | 1 | ||||||||||
Other larger cities | 0.94 NS (0.82–1.08) | 0.93 NS (0.78–1.12) | 0.91 NS (0.76–1.10) | 1.04 NS (0.88–1.22) | 1.09 NS (0.88–1.36) | 0.98 NS (0.78–1.23) | ||||||||||
Other parts of Sweden | 0.95 NS (0.83–1.09) | 0.89 NS (0.74–1.06) | 0.83* (0.68–0.99) | 0.93 NS (0.78–1.10) | 0.97 NS (0.77–1.23) | 0.88 NS (0.69–1.13) | ||||||||||
Limiting longstanding illness or handicap | ||||||||||||||||
No illness or handicap | 1 | 1 | 1 | 1 | ||||||||||||
Illness or handicap | 72.05** (55.75–93.13) | 46.42** (34.73–62.05) | 46.64** (34.99–62.15) | 30.45** (21.86–42.41) |
Employment status . | 1983–89 . | . | . | . | 1992–97 . | . | . | . | ||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
. | Multivariate logistic regression [OR (95% CI)] . | . | . | . | Multivariate logistic regression [OR (95% CI)] . | . | . | . | ||||||||
. | Unadjusted . | Model 1 . | Model 2 . | Model 3 . | Unadjusted . | Model 1 . | Model 2 . | Model 3 . | ||||||||
Employed | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | ||||||||
Unemployed | 1.84** (1.32–2.58) | 2.20** (1.56–3.09) | 2.03** (1.44–2.87) | 1.87** (1.30–2.69) | 2.37** (1.86–3.03) | 2.70** (2.10–3.48) | 2.55** (1.98–3.29) | 2.67** (2.05–3.48) | ||||||||
Sex | ||||||||||||||||
Men | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | ||||||||
Women | 1.17** (1.04–1.30) | 1.21** (1.05–1.40) | 1.23** (1.06–1.42) | 1.21* (1.04–1.41) | 1.24** (1.09–1.42) | 1.12 NS (0.93–1.34) | 1.14 NS (0.95–1.37) | 1.12 NS (0.93–1.35) | ||||||||
Age | ||||||||||||||||
16–25 years | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | ||||||||
26–45 years | 2.84** (2.22–3.64) | 2.29** (1.67–3.14) | 2.86** (2.07–3.94) | 2.42** (1.74–3.36) | 2.73** (2.04–3.66) | 2.01** (1.35–2.99) | 2.36** (1.57–3.53) | 2.22** (1.47–3.35) | ||||||||
46–64 years | 10.62** (8.37–13.48) | 6.26** (4.59–8.52) | 7.40** (5.34–10.25) | 4.10** (2.93–5.72) | 7.71** (5.81–10.23) | 4.87** (3.29–7.20) | 5.66** (3.76–8.51) | 3.72** (2.45–5.64) | ||||||||
Country of birth | ||||||||||||||||
Born in Sweden | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | ||||||||
Born outside Sweden | 3.17** (2.77–3.64) | 2.93** (2.45–3.52) | 2.94** (2.45–3.54) | 3.34** (2.74–4.08) | 3.45** (2.94–4.04) | 2.81** (2.25–3.52) | 2.93** (2.33–3.67) | 3.13** (2.47–3.97) | ||||||||
Educational level | ||||||||||||||||
Post-secondary education | 1 | 1 | 1 | 1 | 1 | 1 | ||||||||||
Secondary education | 1.69** (1.41–2.03) | 1.77** (1.42–2.20) | 1.64** (1.31–2.06) | 2.48** (2.08–2.97) | 1.40** (1.13–1.74) | 1.32* (1.06–1.64) | ||||||||||
Primary education | 3.25** (2.72–3.88) | 2.40** (1.92–3.01) | 2.24** (1.78–2.84) | 1.36** (1.15–1.60) | 1.97** (1.52–2.55) | 1.84** (1.41–2.41) | ||||||||||
Marital status | ||||||||||||||||
Married | 1 | 1 | 1 | 1 | 1 | 1 | ||||||||||
Not married | 0.74** (0.66–0.83) | 1.34** (1.14–1.56) | 1.33** (1.13–1.57) | 0.73** (0.64–0.84) | 1.28* (1.05–1.55) | 1.22 NS (1.0–1.49) | ||||||||||
Region of residence | ||||||||||||||||
Stockholm, Gothenburg, or Malmö | 1 | 1 | 1 | 1 | 1 | 1 | ||||||||||
Other larger cities | 0.94 NS (0.82–1.08) | 0.93 NS (0.78–1.12) | 0.91 NS (0.76–1.10) | 1.04 NS (0.88–1.22) | 1.09 NS (0.88–1.36) | 0.98 NS (0.78–1.23) | ||||||||||
Other parts of Sweden | 0.95 NS (0.83–1.09) | 0.89 NS (0.74–1.06) | 0.83* (0.68–0.99) | 0.93 NS (0.78–1.10) | 0.97 NS (0.77–1.23) | 0.88 NS (0.69–1.13) | ||||||||||
Limiting longstanding illness or handicap | ||||||||||||||||
No illness or handicap | 1 | 1 | 1 | 1 | ||||||||||||
Illness or handicap | 72.05** (55.75–93.13) | 46.42** (34.73–62.05) | 46.64** (34.99–62.15) | 30.45** (21.86–42.41) |
The following variables were controlled for in the different models: Model 1 = sex, age, country of birth; Model 2 = sex, age, country of birth, educational level, marital status, region of residence; Model 3 = sex, age, country of birth, educational level, marital status, region of residence, longstanding illness or handicap
NS = not significant; *P-value ≤ 0.05; **P-value ≤ 0.01
Employment status . | Poor health 1983–89 . | . | . | . | Poor health 1992–97 . | . | . | . | ||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
. | Multivariate logistic regression [OR (95% CI)] . | . | . | . | Multivariate logistic regression [OR (95% CI)] . | . | . | . | ||||||
. | Unadjusted . | Model 1 . | Model 2 . | Model 3 . | Unadjusted . | Model 1 . | Model 2 . | Model 3 . | ||||||
Employed | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | ||||||
Unemployed 1–25 weeks (1983–89, n = 494; 1992–97, n = 727) | 1.63* (1.08–2.48) | 2.21** (1.44–3.39) | 2.07** (1.35–3.18) | 2.12** (1.34–3.33) | 1.36 NS (0.91–2.02) | 1.74** (1.16–2.62) | 1.66** (1.10–2.50) | 1.84** (1.21–2.81) | ||||||
Employed | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | ||||||
Unemployed 26–51 weeks (1983–89 n = 96; 1992–97 n = 197) | 0.70 NS (0.17–2.85) | 0.68 NS (0.17–2.80) | 0.64 NS (0.16–2.64) | 0.46 NS (0.11–1.95) | 1.30 NS (0.61–2.78) | 1.32 NS (0.61–2.84) | 1.25 NS (0.58–2.71) | 1.08 NS (0.49–2.38) | ||||||
Employed | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | ||||||
Unemployed ≥52 weeks (1983–89 n = 99; 1992–97 n = 333) | 4.08** (2.23–7.46) | 3.62** (1.95–6.70) | 3.23** (1.74–5.98) | 2.63* (1.33–5.16) | 5.16** (3.75–7.12) | 4.95** (3.56–6.89) | 4.62** (3.32–6.43) | 5.01** (3.49–7.20) |
Employment status . | Poor health 1983–89 . | . | . | . | Poor health 1992–97 . | . | . | . | ||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
. | Multivariate logistic regression [OR (95% CI)] . | . | . | . | Multivariate logistic regression [OR (95% CI)] . | . | . | . | ||||||
. | Unadjusted . | Model 1 . | Model 2 . | Model 3 . | Unadjusted . | Model 1 . | Model 2 . | Model 3 . | ||||||
Employed | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | ||||||
Unemployed 1–25 weeks (1983–89, n = 494; 1992–97, n = 727) | 1.63* (1.08–2.48) | 2.21** (1.44–3.39) | 2.07** (1.35–3.18) | 2.12** (1.34–3.33) | 1.36 NS (0.91–2.02) | 1.74** (1.16–2.62) | 1.66** (1.10–2.50) | 1.84** (1.21–2.81) | ||||||
Employed | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | ||||||
Unemployed 26–51 weeks (1983–89 n = 96; 1992–97 n = 197) | 0.70 NS (0.17–2.85) | 0.68 NS (0.17–2.80) | 0.64 NS (0.16–2.64) | 0.46 NS (0.11–1.95) | 1.30 NS (0.61–2.78) | 1.32 NS (0.61–2.84) | 1.25 NS (0.58–2.71) | 1.08 NS (0.49–2.38) | ||||||
Employed | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | ||||||
Unemployed ≥52 weeks (1983–89 n = 99; 1992–97 n = 333) | 4.08** (2.23–7.46) | 3.62** (1.95–6.70) | 3.23** (1.74–5.98) | 2.63* (1.33–5.16) | 5.16** (3.75–7.12) | 4.95** (3.56–6.89) | 4.62** (3.32–6.43) | 5.01** (3.49–7.20) |
The following variables were controlled for in the different models: Model 1 = sex, age, country of birth; Model 2 = sex, age, country of birth, educational level, marital status, region of residence; Model 3 = sex, age, country of birth, educational level, marital status, region of residence, longstanding illness or handicap
NS = not significant; *P-value ≤ 0.05; **P-value ≤ 0.01
Employment status . | Poor health 1983–89 . | . | . | . | Poor health 1992–97 . | . | . | . | ||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
. | Multivariate logistic regression [OR (95% CI)] . | . | . | . | Multivariate logistic regression [OR (95% CI)] . | . | . | . | ||||||
. | Unadjusted . | Model 1 . | Model 2 . | Model 3 . | Unadjusted . | Model 1 . | Model 2 . | Model 3 . | ||||||
Employed | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | ||||||
Unemployed 1–25 weeks (1983–89, n = 494; 1992–97, n = 727) | 1.63* (1.08–2.48) | 2.21** (1.44–3.39) | 2.07** (1.35–3.18) | 2.12** (1.34–3.33) | 1.36 NS (0.91–2.02) | 1.74** (1.16–2.62) | 1.66** (1.10–2.50) | 1.84** (1.21–2.81) | ||||||
Employed | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | ||||||
Unemployed 26–51 weeks (1983–89 n = 96; 1992–97 n = 197) | 0.70 NS (0.17–2.85) | 0.68 NS (0.17–2.80) | 0.64 NS (0.16–2.64) | 0.46 NS (0.11–1.95) | 1.30 NS (0.61–2.78) | 1.32 NS (0.61–2.84) | 1.25 NS (0.58–2.71) | 1.08 NS (0.49–2.38) | ||||||
Employed | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | ||||||
Unemployed ≥52 weeks (1983–89 n = 99; 1992–97 n = 333) | 4.08** (2.23–7.46) | 3.62** (1.95–6.70) | 3.23** (1.74–5.98) | 2.63* (1.33–5.16) | 5.16** (3.75–7.12) | 4.95** (3.56–6.89) | 4.62** (3.32–6.43) | 5.01** (3.49–7.20) |
Employment status . | Poor health 1983–89 . | . | . | . | Poor health 1992–97 . | . | . | . | ||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
. | Multivariate logistic regression [OR (95% CI)] . | . | . | . | Multivariate logistic regression [OR (95% CI)] . | . | . | . | ||||||
. | Unadjusted . | Model 1 . | Model 2 . | Model 3 . | Unadjusted . | Model 1 . | Model 2 . | Model 3 . | ||||||
Employed | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | ||||||
Unemployed 1–25 weeks (1983–89, n = 494; 1992–97, n = 727) | 1.63* (1.08–2.48) | 2.21** (1.44–3.39) | 2.07** (1.35–3.18) | 2.12** (1.34–3.33) | 1.36 NS (0.91–2.02) | 1.74** (1.16–2.62) | 1.66** (1.10–2.50) | 1.84** (1.21–2.81) | ||||||
Employed | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | ||||||
Unemployed 26–51 weeks (1983–89 n = 96; 1992–97 n = 197) | 0.70 NS (0.17–2.85) | 0.68 NS (0.17–2.80) | 0.64 NS (0.16–2.64) | 0.46 NS (0.11–1.95) | 1.30 NS (0.61–2.78) | 1.32 NS (0.61–2.84) | 1.25 NS (0.58–2.71) | 1.08 NS (0.49–2.38) | ||||||
Employed | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | ||||||
Unemployed ≥52 weeks (1983–89 n = 99; 1992–97 n = 333) | 4.08** (2.23–7.46) | 3.62** (1.95–6.70) | 3.23** (1.74–5.98) | 2.63* (1.33–5.16) | 5.16** (3.75–7.12) | 4.95** (3.56–6.89) | 4.62** (3.32–6.43) | 5.01** (3.49–7.20) |
The following variables were controlled for in the different models: Model 1 = sex, age, country of birth; Model 2 = sex, age, country of birth, educational level, marital status, region of residence; Model 3 = sex, age, country of birth, educational level, marital status, region of residence, longstanding illness or handicap
NS = not significant; *P-value ≤ 0.05; **P-value ≤ 0.01
The separate analyses for ‘less than good health’ compared with good health showed a pattern similar to the previous results, although the ORs were slightly reduced during both time-periods [in Model 3: 1983–89: OR 1.62, 95% confidence interval (CI) 1.34–1.96; 1992–97: OR 1.88, 95% CI 1.67–2.13].
Unemployment was independently associated with viewing one's own health as worse than that of one's peers in 1996–97, but not in 1988–89. This excess risk was found when sex, age and country of birth were included in the regression (OR 1.90; 95% CI 1.33–2.70 in Model 1). In Model 2, the OR was still significant but slightly reduced (OR 1.81; 95% CI 1.26–2.58). The risk was about 65% (OR 1.65; 95% CI 1.13–2.42) higher than among the unemployed, than among employed when long-standing illness was adjusted for (Model 3). The results persisted among both the short-term (OR 1.90; 95% CI 1.09–3.31 in Model 3) and the long-term unemployed (OR 1.87; 95% CI 1.11–3.15 in Model 3).
The interaction between employment status and the level of unemployment was significant for poor self-rated health, as well as for worse self-rated health.
Self-rated health among different groups of the unemployed
The risk of poor self-rated health was significantly higher irrespectively of sex and the level of unemployment (table 5). Among the unemployed, being older, being born outside Sweden or being married was not related to a higher risk of poor health in 1983–89, after adjustment for long-standing disease or handicap, while the same groups had a high risk for poor health in 1992–97. A high OR for poor health was found among the unemployed living in all studied regions during the 1990s, and unemployment was significantly associated with a higher risk of poor health both among the unemployed, with and without a long-standing illness or handicap.
Employment status . | Poor health 1983–89 . | . | . | . | Poor health 1992–97 . | . | . | . | ||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
. | Multivariate logistic regression [OR (95% CI)] . | . | . | . | Multivariate logistic regression [OR (95% CI)] . | . | . | . | ||||||||
. | Unadjusted . | Model 1 . | Model 2 . | Model 3 . | Unadjusted . | Model 1 . | Model 2 . | Model 3 . | ||||||||
Sex | ||||||||||||||||
Men | ||||||||||||||||
Employed | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | ||||||||
Unemployed | 1.60 NS (0.93–2.77) | 2.00* (1.15–3.50) | 1.83* (1.04–3.20) | 1.89* (1.05–3.40) | 2.03** (1.42–2.89) | 2.31** (1.60–3.34) | 2.18** (1.50–3.15) | 2.42** (1.65–3.57) | ||||||||
Women | ||||||||||||||||
Employed | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | ||||||||
Unemployed | 1.98** (1.29–3.03) | 2.33** (1.51–3.60) | 2.18** (1.41–3.38) | 1.78* (1.11–2.83) | 2.80** (1.99–3.92) | 3.15** (2.22–4–65) | 2.98** (2.10–4.22) | 2.92** (2.02–4.21) | ||||||||
Age | ||||||||||||||||
16–25 year | ||||||||||||||||
Employed | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | ||||||||
Unemployed | 4.79** (2.41–9.53) | 4.38** (2.19–8.77) | 4.35** (2.15–8.82) | 3.84** (1.84–8.01) | 2.54* (1.17–5.51) | 2.56* (1.17–5.57) | 2.62* (1.19–5.76) | 2.65* (1.19–5.89) | ||||||||
26–45 years | ||||||||||||||||
Employed | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | ||||||||
Unemployed | 2.35** (1.35–4.07) | 1.87* (1.07–3.27) | 1.69 NS (0.97–2.95) | 1.62 NS (0.90–2.94) | 3.80** (2.67–5.41) | 3.12** (2.17–4.49) | 3.03** (2.10–4.37) | 3.37** (2.30–4.92) | ||||||||
46–64 years | ||||||||||||||||
Employed | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | ||||||||
Unemployed | 1.73 NS (0.97–3.08) | 1.70 NS (0.95–3.04) | 1.57 NS (0.88–2.82) | 1.32 NS (0.71–2.44) | 2.45** (1.65–3.66) | 2.29** (1.53–3.43) | 2.07** (1.38–3.10) | 1.91** (1.24–2.94) | ||||||||
Country of birth | ||||||||||||||||
Born in Sweden | ||||||||||||||||
Employed | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | ||||||||
Unemployed | 1.92** (1.31–2.82) | 2.47** (1.68–3.65) | 2.22** (1.50–3.28) | 2.04** (1.35–3.07) | 2.40** (1.80–3.20) | 3.10** (2.31–4.16) | 2.82** (2.09–3.79) | 2.76** (2.03–3.76) | ||||||||
Not born in Sweden | ||||||||||||||||
Employed | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | ||||||||
Unemployed | 1.13 NS (0.56–2.28) | 1.45 NS (0.71–2.97) | 1.54 NS (0.75–3.16) | 1.36 NS (0.61–3.01) | 1.44 NS (0.90–2.32) | 1.76* (1.08–2.87) | 1.76* (1.08–2.88) | 2.10** (1.24–3.56) | ||||||||
Educational level | ||||||||||||||||
Primary | ||||||||||||||||
Employed | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | ||||||||
Unemployed | 1.94** (1.21–3.12) | 2.41** (1.48–3.91) | 2.32** (1.43–3.78) | 2.20** (1.30–3.74) | 1.21 NS (0.64–2.30) | 1.23 NS (0.64–2.37) | 1.22 NS (0.63–2.35) | 1.07 NS (0.54–2.11) | ||||||||
Secondary | ||||||||||||||||
Employed | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | ||||||||
Unemployed | 1.67* (1.02–2.75) | 2.04** (1–23–3.40) | 2.04** (1.23–3.39) | 1.89* (1.10–3.24) | 2.31** (1.67–3.21) | 2.82** (2.01–3.97) | 2.77** (1.97–3.90) | 3.09** (2.16–2.42) | ||||||||
Post-secondary | ||||||||||||||||
Employed | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | ||||||||
Unemployed | 0.62 NS (0−4.49) | 0.66 NS (0−4.87) | 0.69 NS (0−5.12) | 0.58 NS (0−4.37) | 2.94** (1.49–5.79) | 3.21** (1.59–6.48) | 2.87** (1.41–5.85) | 2.59** (1.24–5.39) | ||||||||
Marital status | ||||||||||||||||
Married | ||||||||||||||||
Employed | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | ||||||||
Unemployed | 2.43** (1.47–4.02) | 2.02** (1.21–3.37) | 1.96* (1.17–3.27) | 1.44 NS (0.83–2.48) | 3.14** (2.18–4.53) | 2.68** (1.83–3.92) | 2.53** (1.73–3.70) | 2.74** (1.82–4.13) | ||||||||
Unmarried | ||||||||||||||||
Employed | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | ||||||||
Unemployed | 1.67* (1.06–2.62) | 2.22** (1.40–3.52) | 2.06** (1.29–3.27) | 2.14** (1.31–3.50) | 2.06** (1.48–2.87) | 2.57** (1.83–3.62) | 2.53** (1.80–3.56) | 2.55** (1.79–3.63) | ||||||||
Region of residence | ||||||||||||||||
Stockholm, Gothenburg or Malmö | ||||||||||||||||
Employed | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | ||||||||
Unemployed | 1.78 NS (0.90–3.53) | 1.92 NS (0.96–3.86) | 1.79 NS (0.89–3.60) | 1.34 NS (0.64–2.81) | 2.70** (1.76–4.13) | 2.66** (1.71–4.15) | 2.46** (1.57–3.84) | 3.02** (1.90–4.81) | ||||||||
Other larger cities | ||||||||||||||||
Employed | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | ||||||||
Unemployed | 2.27** (1.35–3.83) | 3.01** (1.75–5.17) | 2.74** (1.59–4.71) | 2.68** (1.49–4.82) | 2.97** (2.05–4.31) | 3.54** (2.40–5.20) | 3.36** (2.28–4.95) | 3.18** (2.11–4.81) | ||||||||
Other parts of Sweden | ||||||||||||||||
Employed | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | ||||||||
Unemployed | 1.57 NS (0.89–2.78) | 1.83* (1.02–3.27) | 1.88 NS (1.02–3.45) | 1.80 NS (0.95–3.42) | 1.49 NS (0.90–2.47) | 1.88* (1.11–3.17) | 2.08* (1.20–3.62) | 2.07* (1.17–3.67) | ||||||||
Limiting longstanding illness or handicap | ||||||||||||||||
No long-standing illness | ||||||||||||||||
Employed | 1 | 1 | 1 | NA | 1 | 1 | 1 | NA | ||||||||
Unemployed | 0.72 NS (— −5.24) | 0.75 NS (0.10–5.49) | 0.77 NS (0.10–5.66) | 6.56** (3.40–12.66) | 6.00** (3.01–11.99) | 5.52** (2.75–11.10) | ||||||||||
Long-standing illness | ||||||||||||||||
Employed | 1 | 1 | 1 | NA | 1 | 1 | 1 | NA | ||||||||
Unemployed | 1.83** (1.27–2.64) | 2.07** (1.41–3.02) | 1.94** (1.33–2.84) | 2.12** (1.60–2.81) | 2.38** (1.78–3.18) | 2.27** (1.60–3.04) |
Employment status . | Poor health 1983–89 . | . | . | . | Poor health 1992–97 . | . | . | . | ||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
. | Multivariate logistic regression [OR (95% CI)] . | . | . | . | Multivariate logistic regression [OR (95% CI)] . | . | . | . | ||||||||
. | Unadjusted . | Model 1 . | Model 2 . | Model 3 . | Unadjusted . | Model 1 . | Model 2 . | Model 3 . | ||||||||
Sex | ||||||||||||||||
Men | ||||||||||||||||
Employed | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | ||||||||
Unemployed | 1.60 NS (0.93–2.77) | 2.00* (1.15–3.50) | 1.83* (1.04–3.20) | 1.89* (1.05–3.40) | 2.03** (1.42–2.89) | 2.31** (1.60–3.34) | 2.18** (1.50–3.15) | 2.42** (1.65–3.57) | ||||||||
Women | ||||||||||||||||
Employed | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | ||||||||
Unemployed | 1.98** (1.29–3.03) | 2.33** (1.51–3.60) | 2.18** (1.41–3.38) | 1.78* (1.11–2.83) | 2.80** (1.99–3.92) | 3.15** (2.22–4–65) | 2.98** (2.10–4.22) | 2.92** (2.02–4.21) | ||||||||
Age | ||||||||||||||||
16–25 year | ||||||||||||||||
Employed | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | ||||||||
Unemployed | 4.79** (2.41–9.53) | 4.38** (2.19–8.77) | 4.35** (2.15–8.82) | 3.84** (1.84–8.01) | 2.54* (1.17–5.51) | 2.56* (1.17–5.57) | 2.62* (1.19–5.76) | 2.65* (1.19–5.89) | ||||||||
26–45 years | ||||||||||||||||
Employed | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | ||||||||
Unemployed | 2.35** (1.35–4.07) | 1.87* (1.07–3.27) | 1.69 NS (0.97–2.95) | 1.62 NS (0.90–2.94) | 3.80** (2.67–5.41) | 3.12** (2.17–4.49) | 3.03** (2.10–4.37) | 3.37** (2.30–4.92) | ||||||||
46–64 years | ||||||||||||||||
Employed | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | ||||||||
Unemployed | 1.73 NS (0.97–3.08) | 1.70 NS (0.95–3.04) | 1.57 NS (0.88–2.82) | 1.32 NS (0.71–2.44) | 2.45** (1.65–3.66) | 2.29** (1.53–3.43) | 2.07** (1.38–3.10) | 1.91** (1.24–2.94) | ||||||||
Country of birth | ||||||||||||||||
Born in Sweden | ||||||||||||||||
Employed | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | ||||||||
Unemployed | 1.92** (1.31–2.82) | 2.47** (1.68–3.65) | 2.22** (1.50–3.28) | 2.04** (1.35–3.07) | 2.40** (1.80–3.20) | 3.10** (2.31–4.16) | 2.82** (2.09–3.79) | 2.76** (2.03–3.76) | ||||||||
Not born in Sweden | ||||||||||||||||
Employed | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | ||||||||
Unemployed | 1.13 NS (0.56–2.28) | 1.45 NS (0.71–2.97) | 1.54 NS (0.75–3.16) | 1.36 NS (0.61–3.01) | 1.44 NS (0.90–2.32) | 1.76* (1.08–2.87) | 1.76* (1.08–2.88) | 2.10** (1.24–3.56) | ||||||||
Educational level | ||||||||||||||||
Primary | ||||||||||||||||
Employed | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | ||||||||
Unemployed | 1.94** (1.21–3.12) | 2.41** (1.48–3.91) | 2.32** (1.43–3.78) | 2.20** (1.30–3.74) | 1.21 NS (0.64–2.30) | 1.23 NS (0.64–2.37) | 1.22 NS (0.63–2.35) | 1.07 NS (0.54–2.11) | ||||||||
Secondary | ||||||||||||||||
Employed | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | ||||||||
Unemployed | 1.67* (1.02–2.75) | 2.04** (1–23–3.40) | 2.04** (1.23–3.39) | 1.89* (1.10–3.24) | 2.31** (1.67–3.21) | 2.82** (2.01–3.97) | 2.77** (1.97–3.90) | 3.09** (2.16–2.42) | ||||||||
Post-secondary | ||||||||||||||||
Employed | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | ||||||||
Unemployed | 0.62 NS (0−4.49) | 0.66 NS (0−4.87) | 0.69 NS (0−5.12) | 0.58 NS (0−4.37) | 2.94** (1.49–5.79) | 3.21** (1.59–6.48) | 2.87** (1.41–5.85) | 2.59** (1.24–5.39) | ||||||||
Marital status | ||||||||||||||||
Married | ||||||||||||||||
Employed | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | ||||||||
Unemployed | 2.43** (1.47–4.02) | 2.02** (1.21–3.37) | 1.96* (1.17–3.27) | 1.44 NS (0.83–2.48) | 3.14** (2.18–4.53) | 2.68** (1.83–3.92) | 2.53** (1.73–3.70) | 2.74** (1.82–4.13) | ||||||||
Unmarried | ||||||||||||||||
Employed | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | ||||||||
Unemployed | 1.67* (1.06–2.62) | 2.22** (1.40–3.52) | 2.06** (1.29–3.27) | 2.14** (1.31–3.50) | 2.06** (1.48–2.87) | 2.57** (1.83–3.62) | 2.53** (1.80–3.56) | 2.55** (1.79–3.63) | ||||||||
Region of residence | ||||||||||||||||
Stockholm, Gothenburg or Malmö | ||||||||||||||||
Employed | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | ||||||||
Unemployed | 1.78 NS (0.90–3.53) | 1.92 NS (0.96–3.86) | 1.79 NS (0.89–3.60) | 1.34 NS (0.64–2.81) | 2.70** (1.76–4.13) | 2.66** (1.71–4.15) | 2.46** (1.57–3.84) | 3.02** (1.90–4.81) | ||||||||
Other larger cities | ||||||||||||||||
Employed | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | ||||||||
Unemployed | 2.27** (1.35–3.83) | 3.01** (1.75–5.17) | 2.74** (1.59–4.71) | 2.68** (1.49–4.82) | 2.97** (2.05–4.31) | 3.54** (2.40–5.20) | 3.36** (2.28–4.95) | 3.18** (2.11–4.81) | ||||||||
Other parts of Sweden | ||||||||||||||||
Employed | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | ||||||||
Unemployed | 1.57 NS (0.89–2.78) | 1.83* (1.02–3.27) | 1.88 NS (1.02–3.45) | 1.80 NS (0.95–3.42) | 1.49 NS (0.90–2.47) | 1.88* (1.11–3.17) | 2.08* (1.20–3.62) | 2.07* (1.17–3.67) | ||||||||
Limiting longstanding illness or handicap | ||||||||||||||||
No long-standing illness | ||||||||||||||||
Employed | 1 | 1 | 1 | NA | 1 | 1 | 1 | NA | ||||||||
Unemployed | 0.72 NS (— −5.24) | 0.75 NS (0.10–5.49) | 0.77 NS (0.10–5.66) | 6.56** (3.40–12.66) | 6.00** (3.01–11.99) | 5.52** (2.75–11.10) | ||||||||||
Long-standing illness | ||||||||||||||||
Employed | 1 | 1 | 1 | NA | 1 | 1 | 1 | NA | ||||||||
Unemployed | 1.83** (1.27–2.64) | 2.07** (1.41–3.02) | 1.94** (1.33–2.84) | 2.12** (1.60–2.81) | 2.38** (1.78–3.18) | 2.27** (1.60–3.04) |
All models were stratified for sociodemographic variables. In addition, in Model 1 the following variables were used for stratification: sex, age, country of birth; in Model 2: sex, age, country of birth, educational level, marital status, region of residence; and in Model 3: sex, age, country of birth, educational level, marital status, region of residence, longstanding illness or handicap, were included into the analyses
NA = not applicable for analysis; NS = not significant *P-value ≤ 0.05; **P-value ≤ 0.01
Employment status . | Poor health 1983–89 . | . | . | . | Poor health 1992–97 . | . | . | . | ||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
. | Multivariate logistic regression [OR (95% CI)] . | . | . | . | Multivariate logistic regression [OR (95% CI)] . | . | . | . | ||||||||
. | Unadjusted . | Model 1 . | Model 2 . | Model 3 . | Unadjusted . | Model 1 . | Model 2 . | Model 3 . | ||||||||
Sex | ||||||||||||||||
Men | ||||||||||||||||
Employed | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | ||||||||
Unemployed | 1.60 NS (0.93–2.77) | 2.00* (1.15–3.50) | 1.83* (1.04–3.20) | 1.89* (1.05–3.40) | 2.03** (1.42–2.89) | 2.31** (1.60–3.34) | 2.18** (1.50–3.15) | 2.42** (1.65–3.57) | ||||||||
Women | ||||||||||||||||
Employed | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | ||||||||
Unemployed | 1.98** (1.29–3.03) | 2.33** (1.51–3.60) | 2.18** (1.41–3.38) | 1.78* (1.11–2.83) | 2.80** (1.99–3.92) | 3.15** (2.22–4–65) | 2.98** (2.10–4.22) | 2.92** (2.02–4.21) | ||||||||
Age | ||||||||||||||||
16–25 year | ||||||||||||||||
Employed | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | ||||||||
Unemployed | 4.79** (2.41–9.53) | 4.38** (2.19–8.77) | 4.35** (2.15–8.82) | 3.84** (1.84–8.01) | 2.54* (1.17–5.51) | 2.56* (1.17–5.57) | 2.62* (1.19–5.76) | 2.65* (1.19–5.89) | ||||||||
26–45 years | ||||||||||||||||
Employed | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | ||||||||
Unemployed | 2.35** (1.35–4.07) | 1.87* (1.07–3.27) | 1.69 NS (0.97–2.95) | 1.62 NS (0.90–2.94) | 3.80** (2.67–5.41) | 3.12** (2.17–4.49) | 3.03** (2.10–4.37) | 3.37** (2.30–4.92) | ||||||||
46–64 years | ||||||||||||||||
Employed | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | ||||||||
Unemployed | 1.73 NS (0.97–3.08) | 1.70 NS (0.95–3.04) | 1.57 NS (0.88–2.82) | 1.32 NS (0.71–2.44) | 2.45** (1.65–3.66) | 2.29** (1.53–3.43) | 2.07** (1.38–3.10) | 1.91** (1.24–2.94) | ||||||||
Country of birth | ||||||||||||||||
Born in Sweden | ||||||||||||||||
Employed | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | ||||||||
Unemployed | 1.92** (1.31–2.82) | 2.47** (1.68–3.65) | 2.22** (1.50–3.28) | 2.04** (1.35–3.07) | 2.40** (1.80–3.20) | 3.10** (2.31–4.16) | 2.82** (2.09–3.79) | 2.76** (2.03–3.76) | ||||||||
Not born in Sweden | ||||||||||||||||
Employed | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | ||||||||
Unemployed | 1.13 NS (0.56–2.28) | 1.45 NS (0.71–2.97) | 1.54 NS (0.75–3.16) | 1.36 NS (0.61–3.01) | 1.44 NS (0.90–2.32) | 1.76* (1.08–2.87) | 1.76* (1.08–2.88) | 2.10** (1.24–3.56) | ||||||||
Educational level | ||||||||||||||||
Primary | ||||||||||||||||
Employed | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | ||||||||
Unemployed | 1.94** (1.21–3.12) | 2.41** (1.48–3.91) | 2.32** (1.43–3.78) | 2.20** (1.30–3.74) | 1.21 NS (0.64–2.30) | 1.23 NS (0.64–2.37) | 1.22 NS (0.63–2.35) | 1.07 NS (0.54–2.11) | ||||||||
Secondary | ||||||||||||||||
Employed | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | ||||||||
Unemployed | 1.67* (1.02–2.75) | 2.04** (1–23–3.40) | 2.04** (1.23–3.39) | 1.89* (1.10–3.24) | 2.31** (1.67–3.21) | 2.82** (2.01–3.97) | 2.77** (1.97–3.90) | 3.09** (2.16–2.42) | ||||||||
Post-secondary | ||||||||||||||||
Employed | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | ||||||||
Unemployed | 0.62 NS (0−4.49) | 0.66 NS (0−4.87) | 0.69 NS (0−5.12) | 0.58 NS (0−4.37) | 2.94** (1.49–5.79) | 3.21** (1.59–6.48) | 2.87** (1.41–5.85) | 2.59** (1.24–5.39) | ||||||||
Marital status | ||||||||||||||||
Married | ||||||||||||||||
Employed | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | ||||||||
Unemployed | 2.43** (1.47–4.02) | 2.02** (1.21–3.37) | 1.96* (1.17–3.27) | 1.44 NS (0.83–2.48) | 3.14** (2.18–4.53) | 2.68** (1.83–3.92) | 2.53** (1.73–3.70) | 2.74** (1.82–4.13) | ||||||||
Unmarried | ||||||||||||||||
Employed | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | ||||||||
Unemployed | 1.67* (1.06–2.62) | 2.22** (1.40–3.52) | 2.06** (1.29–3.27) | 2.14** (1.31–3.50) | 2.06** (1.48–2.87) | 2.57** (1.83–3.62) | 2.53** (1.80–3.56) | 2.55** (1.79–3.63) | ||||||||
Region of residence | ||||||||||||||||
Stockholm, Gothenburg or Malmö | ||||||||||||||||
Employed | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | ||||||||
Unemployed | 1.78 NS (0.90–3.53) | 1.92 NS (0.96–3.86) | 1.79 NS (0.89–3.60) | 1.34 NS (0.64–2.81) | 2.70** (1.76–4.13) | 2.66** (1.71–4.15) | 2.46** (1.57–3.84) | 3.02** (1.90–4.81) | ||||||||
Other larger cities | ||||||||||||||||
Employed | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | ||||||||
Unemployed | 2.27** (1.35–3.83) | 3.01** (1.75–5.17) | 2.74** (1.59–4.71) | 2.68** (1.49–4.82) | 2.97** (2.05–4.31) | 3.54** (2.40–5.20) | 3.36** (2.28–4.95) | 3.18** (2.11–4.81) | ||||||||
Other parts of Sweden | ||||||||||||||||
Employed | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | ||||||||
Unemployed | 1.57 NS (0.89–2.78) | 1.83* (1.02–3.27) | 1.88 NS (1.02–3.45) | 1.80 NS (0.95–3.42) | 1.49 NS (0.90–2.47) | 1.88* (1.11–3.17) | 2.08* (1.20–3.62) | 2.07* (1.17–3.67) | ||||||||
Limiting longstanding illness or handicap | ||||||||||||||||
No long-standing illness | ||||||||||||||||
Employed | 1 | 1 | 1 | NA | 1 | 1 | 1 | NA | ||||||||
Unemployed | 0.72 NS (— −5.24) | 0.75 NS (0.10–5.49) | 0.77 NS (0.10–5.66) | 6.56** (3.40–12.66) | 6.00** (3.01–11.99) | 5.52** (2.75–11.10) | ||||||||||
Long-standing illness | ||||||||||||||||
Employed | 1 | 1 | 1 | NA | 1 | 1 | 1 | NA | ||||||||
Unemployed | 1.83** (1.27–2.64) | 2.07** (1.41–3.02) | 1.94** (1.33–2.84) | 2.12** (1.60–2.81) | 2.38** (1.78–3.18) | 2.27** (1.60–3.04) |
Employment status . | Poor health 1983–89 . | . | . | . | Poor health 1992–97 . | . | . | . | ||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
. | Multivariate logistic regression [OR (95% CI)] . | . | . | . | Multivariate logistic regression [OR (95% CI)] . | . | . | . | ||||||||
. | Unadjusted . | Model 1 . | Model 2 . | Model 3 . | Unadjusted . | Model 1 . | Model 2 . | Model 3 . | ||||||||
Sex | ||||||||||||||||
Men | ||||||||||||||||
Employed | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | ||||||||
Unemployed | 1.60 NS (0.93–2.77) | 2.00* (1.15–3.50) | 1.83* (1.04–3.20) | 1.89* (1.05–3.40) | 2.03** (1.42–2.89) | 2.31** (1.60–3.34) | 2.18** (1.50–3.15) | 2.42** (1.65–3.57) | ||||||||
Women | ||||||||||||||||
Employed | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | ||||||||
Unemployed | 1.98** (1.29–3.03) | 2.33** (1.51–3.60) | 2.18** (1.41–3.38) | 1.78* (1.11–2.83) | 2.80** (1.99–3.92) | 3.15** (2.22–4–65) | 2.98** (2.10–4.22) | 2.92** (2.02–4.21) | ||||||||
Age | ||||||||||||||||
16–25 year | ||||||||||||||||
Employed | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | ||||||||
Unemployed | 4.79** (2.41–9.53) | 4.38** (2.19–8.77) | 4.35** (2.15–8.82) | 3.84** (1.84–8.01) | 2.54* (1.17–5.51) | 2.56* (1.17–5.57) | 2.62* (1.19–5.76) | 2.65* (1.19–5.89) | ||||||||
26–45 years | ||||||||||||||||
Employed | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | ||||||||
Unemployed | 2.35** (1.35–4.07) | 1.87* (1.07–3.27) | 1.69 NS (0.97–2.95) | 1.62 NS (0.90–2.94) | 3.80** (2.67–5.41) | 3.12** (2.17–4.49) | 3.03** (2.10–4.37) | 3.37** (2.30–4.92) | ||||||||
46–64 years | ||||||||||||||||
Employed | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | ||||||||
Unemployed | 1.73 NS (0.97–3.08) | 1.70 NS (0.95–3.04) | 1.57 NS (0.88–2.82) | 1.32 NS (0.71–2.44) | 2.45** (1.65–3.66) | 2.29** (1.53–3.43) | 2.07** (1.38–3.10) | 1.91** (1.24–2.94) | ||||||||
Country of birth | ||||||||||||||||
Born in Sweden | ||||||||||||||||
Employed | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | ||||||||
Unemployed | 1.92** (1.31–2.82) | 2.47** (1.68–3.65) | 2.22** (1.50–3.28) | 2.04** (1.35–3.07) | 2.40** (1.80–3.20) | 3.10** (2.31–4.16) | 2.82** (2.09–3.79) | 2.76** (2.03–3.76) | ||||||||
Not born in Sweden | ||||||||||||||||
Employed | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | ||||||||
Unemployed | 1.13 NS (0.56–2.28) | 1.45 NS (0.71–2.97) | 1.54 NS (0.75–3.16) | 1.36 NS (0.61–3.01) | 1.44 NS (0.90–2.32) | 1.76* (1.08–2.87) | 1.76* (1.08–2.88) | 2.10** (1.24–3.56) | ||||||||
Educational level | ||||||||||||||||
Primary | ||||||||||||||||
Employed | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | ||||||||
Unemployed | 1.94** (1.21–3.12) | 2.41** (1.48–3.91) | 2.32** (1.43–3.78) | 2.20** (1.30–3.74) | 1.21 NS (0.64–2.30) | 1.23 NS (0.64–2.37) | 1.22 NS (0.63–2.35) | 1.07 NS (0.54–2.11) | ||||||||
Secondary | ||||||||||||||||
Employed | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | ||||||||
Unemployed | 1.67* (1.02–2.75) | 2.04** (1–23–3.40) | 2.04** (1.23–3.39) | 1.89* (1.10–3.24) | 2.31** (1.67–3.21) | 2.82** (2.01–3.97) | 2.77** (1.97–3.90) | 3.09** (2.16–2.42) | ||||||||
Post-secondary | ||||||||||||||||
Employed | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | ||||||||
Unemployed | 0.62 NS (0−4.49) | 0.66 NS (0−4.87) | 0.69 NS (0−5.12) | 0.58 NS (0−4.37) | 2.94** (1.49–5.79) | 3.21** (1.59–6.48) | 2.87** (1.41–5.85) | 2.59** (1.24–5.39) | ||||||||
Marital status | ||||||||||||||||
Married | ||||||||||||||||
Employed | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | ||||||||
Unemployed | 2.43** (1.47–4.02) | 2.02** (1.21–3.37) | 1.96* (1.17–3.27) | 1.44 NS (0.83–2.48) | 3.14** (2.18–4.53) | 2.68** (1.83–3.92) | 2.53** (1.73–3.70) | 2.74** (1.82–4.13) | ||||||||
Unmarried | ||||||||||||||||
Employed | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | ||||||||
Unemployed | 1.67* (1.06–2.62) | 2.22** (1.40–3.52) | 2.06** (1.29–3.27) | 2.14** (1.31–3.50) | 2.06** (1.48–2.87) | 2.57** (1.83–3.62) | 2.53** (1.80–3.56) | 2.55** (1.79–3.63) | ||||||||
Region of residence | ||||||||||||||||
Stockholm, Gothenburg or Malmö | ||||||||||||||||
Employed | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | ||||||||
Unemployed | 1.78 NS (0.90–3.53) | 1.92 NS (0.96–3.86) | 1.79 NS (0.89–3.60) | 1.34 NS (0.64–2.81) | 2.70** (1.76–4.13) | 2.66** (1.71–4.15) | 2.46** (1.57–3.84) | 3.02** (1.90–4.81) | ||||||||
Other larger cities | ||||||||||||||||
Employed | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | ||||||||
Unemployed | 2.27** (1.35–3.83) | 3.01** (1.75–5.17) | 2.74** (1.59–4.71) | 2.68** (1.49–4.82) | 2.97** (2.05–4.31) | 3.54** (2.40–5.20) | 3.36** (2.28–4.95) | 3.18** (2.11–4.81) | ||||||||
Other parts of Sweden | ||||||||||||||||
Employed | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | ||||||||
Unemployed | 1.57 NS (0.89–2.78) | 1.83* (1.02–3.27) | 1.88 NS (1.02–3.45) | 1.80 NS (0.95–3.42) | 1.49 NS (0.90–2.47) | 1.88* (1.11–3.17) | 2.08* (1.20–3.62) | 2.07* (1.17–3.67) | ||||||||
Limiting longstanding illness or handicap | ||||||||||||||||
No long-standing illness | ||||||||||||||||
Employed | 1 | 1 | 1 | NA | 1 | 1 | 1 | NA | ||||||||
Unemployed | 0.72 NS (— −5.24) | 0.75 NS (0.10–5.49) | 0.77 NS (0.10–5.66) | 6.56** (3.40–12.66) | 6.00** (3.01–11.99) | 5.52** (2.75–11.10) | ||||||||||
Long-standing illness | ||||||||||||||||
Employed | 1 | 1 | 1 | NA | 1 | 1 | 1 | NA | ||||||||
Unemployed | 1.83** (1.27–2.64) | 2.07** (1.41–3.02) | 1.94** (1.33–2.84) | 2.12** (1.60–2.81) | 2.38** (1.78–3.18) | 2.27** (1.60–3.04) |
All models were stratified for sociodemographic variables. In addition, in Model 1 the following variables were used for stratification: sex, age, country of birth; in Model 2: sex, age, country of birth, educational level, marital status, region of residence; and in Model 3: sex, age, country of birth, educational level, marital status, region of residence, longstanding illness or handicap, were included into the analyses
NA = not applicable for analysis; NS = not significant *P-value ≤ 0.05; **P-value ≤ 0.01
In 1992–97, being unemployed and having a primary school education was not associated with a higher risk of poor health after adjusting for all variables, although this was the case during the period 1983–89. The unemployed with a higher education showed a higher risk of poor health when unemployment was high, but not when it was low.
Discussion
Poor self-rated health was significantly higher among the unemployed than among the employed during both high and low levels of unemployment, although the difference between the unemployed and employed was slightly higher when job opportunities were few. Unemployment was independently related to a higher risk of rating health as worse than one's peers in 1996–97, but not in 1988–89. This pattern persisted among both the long-term and the short-term unemployed.
Self-rated health has been used as a reliable predictor for total mortality, and may to some extent be related to anticipation of deaths due to specific diseases.28,29 It has been related to psychological and medical symptoms and may play a role in predicting adjustment after illness, by being a predictor of gainful employment and involvement in community organizations.30 The global non-comparative measure of health and the age-comparative measure, where the assessment of health is anchored to one's peers,24,31 may have different thresholds for detecting changes in health, since a larger number of the unemployed viewed their health as worse than that of their peers compared with poor. The age-related measures of health may capture more of those unemployed who have noted a change in their own health, while the global question may be more specific in the way it estimates health. Comparing one's health with that of persons of the same age may also offer a possibility to assess self-rated health in relation to an implicit norm (i.e. ‘health is about the same as most’) instead of in relation to a continuum.24
A heterogeneous group of the unemployed
The unemployment during the 1990s hit larger sections of the population than during previous decades. There was a rise in unemployment among the well educated, the middle-aged and in urban areas.13,32 The changes in the constellation of unemployed during the 1990s may have created new reactions to unemployment. In this study, younger, middle-aged and older unemployed were at higher risk of poor health in the 1990s. The unemployed showed a higher risk of poor health both in densely and sparsely populated areas. The effects of social support received from being married may not be great enough to buffer the negative impact of stress experienced among the unemployed in times of high levels of unemployment. During the 1990s, the risk of poor health was high both among those unemployed who were born in Sweden and among those who were born outside Sweden, while in the 1980s the risk of poor health was only found among persons born in Sweden. Finally, during high levels of unemployment, having attained a post-secondary educational level was associated with a higher risk of poor health, while having a primary education was not. The pattern was the opposite during low levels of unemployment.
The higher risk of poor health among unemployed immigrants during the 1990s may be related to refugees being the dominant immigrant group during this time period. In the 1980s there was a higher number of labour immigrants in the population. A selection bias among the immigrants owing to a ‘healthy immigrant effect’ may have been more salient during the 1980s.33 Possibly, the large number of educational training programs for persons seeking a job during the 1990s34 may have improved health among those who had obtained a primary educational level during the 1990s.
Changes in the labour market from the 1990s
The social structural changes occurring during the 1990s in Sweden may have been detrimental to self-rated health especially among the unemployed. The Swedish labour market was characterized by an economic crisis, followed by mass unemployment. This situation came to have a powerful impact on government finances in terms of both reduced revenue and greater public expenditure, which influenced people's economy.13,20,35 The reductions in economic position and health-related problems were reduced especially among long-term unemployed and persons who were outside the labour force.13–16 This was explained only to a small extent by sex, age, cohabiting status, educational level, citizenship or region of residence.15 Poor mental health increased from 5% to 13% among the long-term unemployed during the 1990s.13
According to a project set up to examine changes of socioeconomic health inequalities over time in the Nordic countries, only small changes in long-standing illness and perceived global self-rated health took place in different employment groups from the 1980s to the mid-1990s.19–22,36 Lundberg and colleagues20 analysed Swedish data on health inequalities for the periods 1986–87 and 1994–95, based on data from the Swedish Survey of Living Conditions. The overall health and differences in health between different employment status groups remained constant during the two periods. However, the situation might have changed during the late 1990s, which may also be supported by data showing that absenteeism due to illness has increased dramatically during this period.37,38 In this study the unemployed had significantly poorer health during the period of the mid-1990s and on, compared with during the beginning of the 1990s.
Novo and colleagues17 surveyed a group of long-term unemployed men and women during 1986 and 1994. The economic fluctuations did not influence health among the young long-term unemployed, during the recession, although complaints about health rose during the recession in the reference group, comprised of short-term unemployed, employed and others.
The contrasting results found in the present study may have to do with the fact that the reference group included only employed persons. Furthermore, the effects of unemployment were analysed against the background that the unemployed became a more heterogeneous group during the 1990s. Although, the association between long-term unemployment and psychological health seem to be stronger in young people than in adults,39 being older and unemployed may put the individual in a more socially and economically deprived position.40 Older unemployed individuals show higher risks of remaining unemployed than do their younger counterparts.40 During times of few job opportunities, this vulnerable position might be even more salient than when jobs are more available.
What can we say about selection?
Since this is a cross-sectional study it is not possible to make any definite statements about the potential influence of health selection. Some of the analyses were based on a small sample of individuals and the results should be interpreted with caution. However, the excess risk for ‘poor’ or ‘worse’ self-rated health among the unemployed remained, despite adjusting for sex, age and country of birth, which were assumed not to change due to employment status. Thus, poor health among the unemployed was not due to the fact that these sociodemographic variables existed before the start of unemployment. The relative ill health among the unemployed was also independent of the occurrence of long-standing illness.
When unemployment is high, the group of unemployed may be more heterogeneous and healthier, consisting of well-educated white-collar workers, for many of whom unemployment may be a new experience. Possibly, one should not expect higher rates of ill health among the unemployed in times when unemployment levels are high, if health selection could explain the poor health among the unemployed.1,9,12 The social structure experienced during unemployment itself, rather than selection, may then affect health status among the unemployed.
Data from Norway41 and Britain42 have shown that the socioeconomic inequalities in health might have widened between the employed and the non-employed, while the inequalities among the employed might have narrowed. This would have to do with a selective exit from and entry to paid employment due to individual health.41 Bartley and Owens42 found that being a manual worker and having a limiting illness was strongly associated with being out of the work force altogether, and that this tendency increased as the rate of unemployment rose. This would result in a ‘healthy worker effect’, where both the unemployed and the employed are being increasingly selected. In this study, self-rated health among persons who were outside the labour force was not generally poorer, nor was long-standing illness more common when unemployment levels were high. Possibly, the Swedish labour market policies provide good opportunities for persons with chronic illnesses to remain in employment.43 However, the future question to be answered concerns the relative weight of the selection causation mechanisms.8,10
In conclusion, the results indicate that a more heterogeneous group of the unemployed may experience poor health when unemployment levels are high, compared with when they are low. Poorer self-rated health among the unemployed seems to be an increasing public health problem during high levels of unemployment.
Acknowledgement is made to Patrik Öhagen, statistician from the Swedish University of Agricultural Sciences and Lars Berglund, statistician from Uppsala Clinical Research Center, for commenting on the statistics used in the study. The study was supported by grants from the Swedish council for working life and social research.
This study examined self-rated health in the employed and unemployed, during periods of high and of low levels of unemployment.
Differences in health between the unemployed and employed were larger when unemployment levels were high, than when they were low.
The results persisted after adjusting for sociodemographic variables and long-term disease or handicap.
A more heterogeneous group of the unemployed reported poor health when unemployment was high compared with when it was low.
Poorer self-rated health among the unemployed seems to be an increasing public health problem during high levels of unemployment.
References
Bartley M. Unemployment and ill health: Understanding the relationship.
Jin RL, Shah CP, Svoboda TJ. The impact of unemployment on health: A review of the evidence [see comments].
Roberts H, Pearson JC, Madeley RJ, et al. Unemployment and health: the quality of social support among residents in the Trent region of England.
Whelan CT. The role of income, life-style deprivation and financial strain in mediating the impact of unemployment on psychological distress: Evidence from the Republic of Ireland.
Hammarström A, Janlert U. Unemployment and change of tobacco habits: a study of young people from 16 to 21 years of age.
Janlert U, Hammarström A. Alcohol consumption among unemployed youths: Results from a prospective study.
Valkonen TMP. The association between unemployment and mortality: causation or selection? In: A. Lopez, G. Caselli and T. Valkonen, editors. Adult mortality in developed countries: From description to explanation. Oxford, England: Clarendon Press,
Martikainen PT, Valkonen T. Excess mortality of unemployed men and women during a period of rapidly increasing unemployment [see comments].
Janlert U. Unemployment as a disease and diseases of the unemployed.
Montgomery SM, Bartley MJ, Cook DG, Wadsworth ME. Health and social precursors of unemployment in young men in Great Britain.
Moser KA, Goldblatt PO, Fox AJ, Jones DR. Unemployment and Mortality. Longitudinal Study: Mortality and Social Organization. London: HMSO,
Ministry Publication Series. 2002:32. Welfare in Sweden: The Balance Sheet for the 1990s/Palme J. Stockholm: Ministry of Health and Social Affairs. p. 216.
SOU, Statens Offentliga Utredningar. Välfärd och arbete i arbetslöshetens årtionde/Palme J. Ministry of Health and Social Affair,
Lindh T. 1990-talets ekonomiska utveckling- vilken roll har åldersfördelningen spelat? (Economic development in the 1990s: The impact of age distribution). In Fritzell J, editor. Välfärdens förutsättningar. Arbetsmarknad, demografi och segregation (Conditions of Welfare: Employment, Demography and Segregation). Stockholm: Fritzes,
SOU, Statens Offentliga Utredningar. Two of a kind? economic crisis, policy responses and well-being during the 1990's in Sweden and Finland (a report from the government commission). A Balance Sheet for Welfare of the 1990s/Mikko Kautto.
Novo M, Hammarström A, Janlert U. Health hazards of unemployment—only a boom phenomenon? A study of young men and women during times of prosperity and times of recession.
Novo M, Hammarström A, Janlert U. Do high levels of unemployment influence the health of those who are not unemployed? A gendered comparison of young men and women during boom and recession.
Dahl E, Elstad JI. Recent changes in social structure and health inequalities in Norway.
Lundberg O, Diderichsen F, Yngwe MA. Changing health inequalities in a changing society? Sweden in the mid-1980s and mid-1990s.
Manderbacka K, Lahelma E, Rahkonen O. Structural changes and social inequalities in health in Finland, 1986–1994.
Lahelma E, Kivela K, Roos E, et al. Analyzing changes of health inequalities in the Nordic welfare states.
Lundberg O, Manderbacka K. Assessing reliability of a measure of self-rated health.
Smith AMA, Shelley JM, Dennerstein L. Self-rated health: Biological continuum or social discontinuity?
Undén AL, Elofsson S. Självupplevd hälsa: Faktorer som påverkar människors egen bedömning. Forskningsrådsnämnden (FRN) (Swedish Council for Planning and Coordination of Research)
The Swedish National Labour Market Board (AMS). Employment and Unemployment Statistics. Department of Analyses and Evaluation. Stockholm,
International Labour Organization. Employment and Unemployment Statistics. Geneva: International Labour Office,
Idler EL. Self-rated health and mortality: A review of twenty-seven community studies.
Bjorner J, Sondergard Kristensen T, Orth-Gomér K et al. Self-rated health: A useful concept in research, prevention and clinical medicine. Forskningsrådsnämnden (FRN)
Garrity TF. Vocational adjustment after first myocardial infarction; comparative assessment of several variables suggested in the literature.
Undén AL, Elofsson S. Comparison between different measures of self-rated health, and an analysis of predictors. Uppsala: Forskningsrådsnämnden (FRN),
Andersson R. Etnisk och socioekonomisk segregation i Sverige 1990 1998 (Ethnic and socioeconomic segregation in Sweden, 1990–1998). Välfärdens förutsättningar. Arbetsmarknad, demografi och segregation (Conditions of Welfare: Employment, Demography and Segregation). Reports of the Government Commissions (SOU) 2000: 37. Stockholm: Fritzes,
Westerling R, Rosén M. Avoidable mortality among immigrants in Sweden.
Prora/Arbetsmarknadsstyrelsen. Uppföljning av deltagare som slutat yrkesinriktad arbetsmarknadsutbildning andra kvartalet 1999. Programenheten, Arbetsmarknadsstyrelsen (The Swedish National Labour Market Board). Stockholm 2000, 1–47.
Bergmark A. Solidarity in Swedish welfare—standing the test of time?
Lissau I, Rasmussen NK, Moss Hesse N, Hesse U. Social differences in illness and health-related exclusion from the labour market in Denmark from 1987 to 1994.
Lahelma E, Lundberg O, Manderbacka K, Roos E. Changing health inequalities in the Nordic Countries?
SOU Statens Offentliga Utredningar. Sjukförsäkringen basfakta och utvecklingsmöjligheter (Sickness insurances—basic facts and possibilities for development). Delbetänkande från sjukförsäkringsutredningen. Stockholm: Fritzes,
Reine I, Novo M, Hammarström A. Does the association between ill health and unemployment differ between young people and adults? Results from a 14-year follow-up study with a focus on psychological health and smoking.
Dahl E, Birkelund GE. Sysselsettning, klasse och helse 1980–1995. En ananlyse av fem norske levekårsunderskelser (Employment, class and health 1980–95).
Bartley M, Owen C. Relation between socioeconomic status, employment, and health during economic change, 1973–93 [see comments].
Comments