We've recently been looking at the impact of depression on employment status in Canadians using longitudinal data from the National Population Health Survey. We felt that this was important since, although it is well known that unemployment and depression are associated, it is possible both that unemployment could cause depression and that depression could cause unemployment. We were particularly interested in looking at the latter question. We selected people who were all working at baseline, divided these into groups that experienced (or did not) an episode of major depression. The effect of depression was greatest in young people - perhaps reflecting the greater difficulties faced by younger people in coping with depression, or their greater vulnerability to its effects (for example, because they may be less well-established in their career). However, it is also possible that older people who were working at baseline represented a subset of those who were more effective at coping with the symptoms - ie. those whose careers had been interrupted by depression prior to the baseline interview (1994) would not have been eligible, a selective effect that may have occurred less often in the younger respondents.
The results are published in the Canadian Journal of Psychiatry, here.