Physical activity levels and determinants of change in young adults: a longitudinal panel study

BACKGROUND: There is growing concern about physical inactivity in adolescents and young adults. Identifying determinants that are associated with low levels of physical activity and with changes in physical activity levels will help to develop specific prevention strategies. The present study describes the prevalence and potential determinants of physical activity behavior and behavior changes of young adults. The study is based on the Swiss Household Panel (SHP), a longitudinal study assessing social changes in a representative sample of Swiss households since 1999. METHODS: Data is collected yearly using computer-assisted telephone interviews. Information is obtained from each household member over 14 years of age. Participants between 14 and 24 years entering the SHP between 1999 and 2006 were included (N = 3,068). "Inactive" was defined as less than 1 day/week of at least 30 minutes of moderate physical activity, "no sport" as exercising less than once a week. Age, gender, nationality, linguistic region, household income, education, membership in a sport club, reading, and Internet use were included as potential determinants of physical activity behavior and behavior change. RESULTS: In both young men and young women, the prevalence of inactivity, "no sport", and non-membership in a sport club was increasing with age. Women were less active than men of the same age. From one wave to the following, 11.1% of young men and 12.1% of young women became active, and 11.9% of men and 13.7% of women became inactive, respectively (pooled data over all eight waves). Non-membership in a sport club was the strongest predictor for "no sport" (OR(men )6.7 [4.9-8.9]; OR(women )8.1 [5.7-11.4]), but also for being inactive (OR 4.6 [3.5-6.0]; 4.6 [3.3-6.4]). Leaving a sport club (OR 7.8 [4.4-14.0]; 11.9 [5.9-24.1]) and remaining non-member (OR 7.8 [4.7-12.9]; 12.4 [6.4-24.1]) were the strongest predictors of becoming "no sport". Effects for becoming inactive were similar, though smaller (OR 5.9 [3.4-10.5] and 5.1 [2.7-9.6] for leaving a club, OR 5.1 [3.1-8.4] and 6.9 [4.0-11.8] for remaining non-member). CONCLUSIONS: The most important findings were the strong effects of sport club membership on general physical activity. The correlation between sport club membership and exercise was not surprising in its nature, but in its strength.