Is unemployment in young adulthood related to self-rated health later in life? Results from the Northern Swedish cohort

Table 3 Long-term effect of unemployment at 28–30 years of age on self-rated health at age 42 for groups of individuals (n = 620)

	Logistic Regression		G-computation	Propensity scores, inverse probability weighting
	Odds ratio	Confidence interval	Risk difference	Risk difference
Education level^a
Secondary education (n = 255)	1.45	0.60–3.32	0.074*	0.052
Upper secondary education (n = 112)	5.99	2.03–19.3^b	0.345*^b	0.372*^b
University (n = 253)	1.22	0.59–2.49	0.039*	0.013^b
Marital status
Married (n = 468)	1.58	0.90–2.74	0.069	0.072
Single (n = 152)	2.46	0.98–6.34	0.194*	0.211*^b
Self-rated health 1995
Poor (n = 125)	1.75	0.65–5.11	0.109*	0.169^b
Good (n = 495)	1.77	1.02–3.01	0.116*	0.089
Occupation
Blue-collar workers (n = 246)	1.56	0.74–3.25	0.087*	0.099
Low white-collar workers (n = 103)	1.51	0.49–4.54^b	0.083*^b	0.058^b
Medium–high white-collar workers (n = 271)	2.29	1.04–4.96	0.169*	0.119
Gender
Man (n = 338)	1.35	0.68–2.64	0.060*	0.056
Woman (n = 282)	2.29	1.15–4.55	0.168*	0.155*

* p-value below 0.05
^a Secondary education corresponds to at most 2-years of secondary education, and upper-secondary education corresponds to 3–4 years of secondary education
^b Logistic regression was used with fewer than the recommended 10 outcomes per variable for the least-occurring outcomes
Analyses controlled for education level, marital status, previous health status (self-rated health in 1995), and occupation, excluding the factor for which the stratification was done. Estimates represent the effect on unemployed compared to employed individuals

ISSN: 1471-2458