Work environment and somatic hospital admissions in Denmark 1994-1999

Objective: The aim was to elucidate the usefulness of the excess hospitalisation fraction (EHF) approach-as a practical alternative to the calculation of etiological fractions-for prioritising national preventive measures in the total environment. In this study we used the inequality in somatic hospitalisation across industrial sectors as an example. The presented EHFs may provide an estimate of the order of magnitude of the prevention potential, which may also be useful in priority setting in countries comparable to Denmark. Methods: All economically active adults in Denmark were followed-up for 6 years after hospital discharge, and industrial specific EHFs for 58 industrial sectors were calculated for 11 main diagnostic groups. Assuming 'education and research' to be the industrial sector which is least exposed to occupational hazards, we calculated the EHFs as a practical approximation to the etiologic fraction for all other industries, using hospitalisation rates adjusted for age. In addition, we also controlled for social class since many risk factors may also be related to life style or living conditions. Results: Compared to 'education and research' the EHF in all other industries was (additional control for social class in brackets): Women 11% (7%), men 15% (8%). The EHFs for some of the main diagnostic groups were as follows: Circulatory diseases: Women 18% (12%), men 16% (10%). Neoplasms: Women 3% (4%), men 8% (6%). Musculoskeletal diseases: Women 19% (12%). men 21% (10%). Diseases in the respiratory system: Women 12% (8%), men 16% (9%). Diseases in the nervous system: Women 12% (7%), men 17% (12%). Violent events (injuries and trauma): Women 6% (5%), men 17% (6%). Conclusions: The advantage of this method is that the EHF can be calculated for all diseases and for the whole population without detailed knowledge of causal risk factors and their distribution in the population. Two main methodological problems are that we cannot control for selection bias related to occupation, and that the control for social class may lead to conservative estimates. Our estimates are, however, close to estimates of etiologic fractions published in the literature. Large unexpected EHFs for diseases of the blood and blood-forming organs should be seen as challenges for future research. (C) 2004 Elsevier B.V. All rights reserved.