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

Table 2 Long-term effect of unemployment at 28–30 years of age on self-rated health (n = 620)

		Model
Method	Estimate	Full	Reduced	Crude
Logistic regression	Odds ratio	1.73	1.74	1.78
Logistic regression	Confidence interval	1.07–2.8	1.08–2.6	1.15–2.8
G-computation	Risk difference	0.111	0.113	-
	Mean square error	0.0029	0.0027	-
	Confidence interval	0.025–0.199	0.029–0.200	-
Propensity scores, inverse probability weighting	Risk difference	0.114	0.103	-
	Mean square error	0.0038	0.0012	-
	Confidence interval	0.018–0.220	0.016–0.188	-

The p-value is less than 0.05 for all effect estimates in the table. Analyses in the reduced models controlled for education level, marital status, self-rated health 1995, and occupation. The full model also controlled for gender, social network, cash margin, smoking, alcohol intake, and body mass index. Crude refers to the estimate with unemployment as the only predictor. Estimates represent the health effect on unemployed compared to employed individuals.

ISSN: 1471-2458