Predictors of changes in physical activity in a prospective cohort study of the Danish adult population.

Aim: To investigate predictors of changes in physical activity, in a prospective population-based study.

Methods: Data were from the Danish Health Interview Surveys in 1994 and 2000, and included persons between 16 and 64 years of age who answered the questions on physical activity and various covariates in 1994, and who were re-interviewed in 2000. In total 2,957 subjects participated (62% of the original sample). Data were analysed using logistic regression.

Results: Predictors of physical inactivity were, for men and women respectively, heavy smoking (odds ratio (OR) 2.07, 95% confidence interval (CI) 1.24-3.48, and OR 2.23, 95% CI 1.34-3.71), poor self-rated health (OR 2.11, 95% CI 1.25-3.58, and OR 1.75, 95% CI 1.10-2.80), and believing that one's own effort has no effect on health (OR 1.86, 95% CI 1.20-2.88, and OR 1.63, 95% CI 1.00-2.65). For men, further predictors for physical inactivity were obesity as compared to normal weight (OR 2.10, 95% CI 1.11-3.98), and being unmarried as compared to being married (OR 2.22, 95% CI 1.40-3.51). The only predictor for becoming physically active among initially sedentary respondents was meeting often with family (p=0.011).

Conclusions: Our results show a number of strong predictors for physical inactivity. Variables of an unhealthy lifestyle tend to cluster in some individuals, and physical inactivity interventions may therefore be more successful if they also address smoking, self-belief, and obesity. Interventions tailored to raise inactive people's level of activity should focus on social support from the family.