Changes in work factors and concurrent changes in leisure time physical activity: a 12-year longitudinal analysis

OBJECTIVES: There is a paucity of longitudinal population-based studies examining whether changes in work factors are associated with concurrent changes in leisure time physical activity (LTPA). This study examines this issue using 12 years of longitudinal survey data. METHODS: Data were drawn from the Canadian National Population Health Survey. The initial analytical sample in 1994 of 6407 working individuals was followed every 2 years from 2000 to 2010. Seven work factors were measured as independent variables: skill discretion, decision authority, psychological demands, physical exertion, number of jobs, hours at work and shift schedule. LTPA was categorised as inactive, moderately active or active based on metabolic equivalent task values. Fixed-effects multinomial logistic models were used to examine associations between work factors and LTPA controlling for time-invariant effects and adjusted for covariates. RESULTS: Workers with lower skill discretion (OR=0.96; 95% CI 0.92 to 0.99), higher psychological demands (OR=0.95; 95% CI 0.92 to 0.99), higher physical exertion (OR=0.93; 95% CI 0.88 to 0.99) and longer work hours (OR=0.97; 95% CI 0.95 to 098) were associated with a lower odds of transitioning from inactive to active and moderately active. There was no evidence of effect modification by age or sex. CONCLUSION: Results suggest that as participants' skill discretion decreased, and their physical and psychological demands, and work hours increased, their likelihood of becoming more active and moderately active also decreased, supporting the value of targeting improvements in these work factors for physical activity interventions