Unemployment, depressiveness and disability retirement: a follow-up study of the Finnish HeSSup population sample.

Background: Disability retirement because of depression is increasingly common in Finland. The rise of such retirement coincided with the rise of unemployment in the second half of the 1990s. In this study we sought potential connections between these two epidemics. We assumed that depressiveness incurs a higher risk of disability retirement among the unemployed than among the employed population.

Methods: Data were derived from HeSSup Study, which follows a large population sample. A total of 14,487 participants responded in 1998 and in 2003 in postal surveys about their employment status. Their depressiveness was measured in 1998 with Beck's depression inventory. Labour market status in 1998 was classified as employed, short-term unemployed and long-term unemployed. In 2003 the status was dichotomised into disability retired or not.

Results: A total of 329 participants (2.3%) retired during the 5-year follow-up. The retirement percentage was 1.9 among the employed, 4.8 among short-term unemployed and 6.7 among long-term unemployed. The percentage for the employed ranged from 1.4 (no depressiveness) to 7.1 (severe depressiveness), for the short-term unemployed, the range was from 3.3 to 17.9% and for the long-term unemployed the range was from 2.6 to 14.2%. Retirement was more probable among unemployed participants than among employed (OR 2.24 [95% CI 1.50-3.36] for short-term and OR 2.82 [95% CI 1.93-4.13] for long-term unemployed), when employment status was adjusted by sociodemographic background and somatic comorbidity and interaction term employment status with depressiveness was taken into account. Moreover, the interaction of employment status and depressiveness with the probability of disability retirement was significant. Our final analysis also revealed that disability retirement was particularly common among the short-term unemployed with severe depressiveness. Among the long-term unemployed, in contrast, retirement was more common in the mildly depressed than in the severely depressed.

Conclusions: Our results suggest that high level of unemployment may mean high level of premature exit from the work-force due to disability retirement. The epidemic of depression-related disability contributes significantly to this general trend. In future, it will be important to find ways to support depression patients to hold on to their jobs and to develop services, combined with health policy and employment policy, can interrupt the vortex of deteriorating mental well-being, prolonged unemployment and marginalisation from health care.