Objective: The aim of this longitudinal study was to examine (a) the causal effect of unemployment 1990-1997 on health 1998-2001 (in-patient periods) and (b) the selection effect of health 1996-1999 on unemployment 2000-2002. We examined the effects of different diagnoses, namely, all causes, circulatory diseases, diseases of the digestive system, musculoskeletal diseases, and mental disorders.
Methods: The data from the Central Register of Health Care Professionals of persons born 1945-1970 were linked to data from employment statistics and Finnish Hospital Discharge Register including 78,195 women and 12,455 men aged 31 to 56 in 2001. The associations were analyzed with logistic regression analyses and expressed as odds ratios (OR) and their 95% confidence intervals (CI).
Results: After adjustments, existence of unemployment periods was associated with lower odds for in-patient periods due to musculoskeletal diseases for both women (OR=0.82, 95% CI=0.76-0.89) and men (OR=0.74, 95% CI=0.60-0.89). Unemployment periods were more likely among women (OR=1.65, 95% CI=1.33-2.04) and men (OR=2.54, 95% CI=1.44-4.48) having had in-patient periods due to mental diseases and among women also due to diseases of the digestive system (OR=1.17, 95% CI=1.04-1.31).
Conclusion: The present study found evidence for selection to unemployment according to mental diseases and diseases of the digestive system among health care professionals. In addition, our results show that unemployment periods may protect health care workers from musculoskeletal diseases.