Smoking as a predictor of long-term work disability in physically active and inactive people.

Data from a Norwegian 4-year prospective questionnaire study were used to test our hypothesis that smoking is a weaker predictor of long-term work disability in physically active than inactive persons. Of 1,788 respondents who were working and not older than 62 years in 1990, 1,426 (80%) also returned a questionnaire in 1994. For respondents who were taking physical exercise less than once a week, smoking predicted long-term work disability four years later [odds ratio (OR) = 2.24; 95% confidence interval (CI) = 1.30-3.87; p < 0.01], when adjusting for age, sex, work-hours per week, heavy lifting at work, emotional symptoms and musculoskeletal pain at time 0. In respondents who were exercising at least once a week, there was no association between smoking and long-term work disability.