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

Fredrik Norström, Urban Janlert, Anne Hammarström

Many studies have reported that unemployment has a negative effect on health. However, little is known about the long-term effect for those who become unemployed when they are young adults. Our aim was to examine how unemployment is related to long-term self-rated health among 30 year olds, with an emphasis on how health differs in relation to education level, marital status, previous health, occupation, and gender. In the Northern Swedish Cohort, 1083 teenagers (~16 years old) were originally invited in 1981. Of these, 1001 participated in the follow-up surveys in 1995 and 2007. In our study, we included participants with either self-reported unemployment or activity in the labor force during the previous three years in the 1995 follow-up so long as they had no self-reported unemployment between the follow-up surveys. Labor market status was studied in relation to self-reported health in the 2007 follow-up. Information from the 1995 follow-up for education level, marital status, self-reported health, and occupation were part of the statistical analyses. Analyses were stratified for these variables and for gender. Analyses were performed with logistic regression, G-computation, and a method based on propensity scores. Poor self-rated health in 2007 was reported among 43 of the 98 (44%) unemployed and 159 (30%) of the 522 employed subjects. Unemployment had a long-term negative effect on health (odds ratio with logistic regression 1.74 and absolute difference estimates of 0.11 (G-computation) and 0.10 (propensity score method)). At the group level, the most pronounced effects on health were seen in those with upper secondary school as their highest education level, those who were single, low-level white-collar workers, and women. Even among those becoming unemployed during young adulthood, unemployment is related to a negative long-term health effect. However, the effect varies among different groups of individuals. Increased emphasis on understanding the groups for whom unemployment is most strongly related to ill health is important for future research so that efforts can be put towards those with the biggest need. Still, our results can be used as the basis for deciding which groups should be prioritized for labor-market interventions.