Heterogeneity in the associations between common mental disorders and labour outcomes – a population study from southern Sweden

Table 4 Sensitivity analysis: Logistic regression (odds ratios) – The association between CMD and subsequent disability pension

	Disability pension DP Yr − 1 excluded [1]
No CMD	Ref.
CMD	11.45
CMD	[2.44–53.81]
Man*CMD	1.44
Man*CMD	[0.69–3.00]
Secondary education*CMD	1.67
Secondary education*CMD	[0.73–3.80]
Higher education*CMD	1.94
Higher education*CMD	[0.74–5.10]
30–39 years*CMD	0.13
30–39 years*CMD	[0.02–0.66]
40–49 years*CMD	0.25
40–49 years*CMD	[0.05–1.21]
50–59 years*CMD	0.26
50–59 years*CMD	[0.06–1.16]
Foreign born*CMD	3.96
Foreign born*CMD	[1.35–11.66]
n	68,861
Pseudo-R²	0.14

Note: Disability pension at least once during the three follow-up years. The interaction effect (denoted by *) shows the difference in associations between CMD diagnosis and the labour outcome for the interacted group compared to the reference group, as a ratio of odds ratios. The reference groups are women, mandatory education, 20–29 years of age and born in Sweden. We control for sex, education, age-group, being foreign-born, income and year of diagnosis/inclusion in the study. Income is measured by disposable income in the year before diagnosis and is stratified in quintiles

ISSN: 1471-2458