Background The London 2012 Olympic Games aimed to “inspire a generation” by increasing participation in sport through ‘demonstration effects’ – where individuals’ motivations are influenced by observing the actions of others. Here we evaluate whether adolescents’ participation in Olympic-related sports increased post-games, and whether changes were mediated by level of engagement with the Olympic experience.

Methods The Olympic Regeneration in East London (ORiEL) study is a longitudinal controlled quasi-experimental study of a sample of Year 7 adolescents from 25 schools in four East London boroughs. Respondents were surveyed 6 months pre-Games and were followed-up 6 months post-Games (n = 2727). Games engagement was assessed on a 10-point scale ranging from not excited (0) to very excited (10). The Youth Physical Activity Questionnaire (Y-PAQ) assessed participation in ten Olympic sports over the past seven days (cycling, basketball, football, gymnastics, hockey, martial arts, running, swimming, horse riding, racquet sports) transformed to a binary ‘ever/never’ response. Sedentary activities (computer gaming, online browsing, TV viewing) were analysed by duration. Random effects logistic regression estimated the odds of participation in Olympic sports post-Games compared to pre-Games. Changes in the duration of sedentary activities were estimated with random effects linear regression. The effect of engagement was assessed by stepwise inclusion to the model along with confounding variables.

Results Overall there was a statistically significant decline in the odds of participating in eight of the ten Olympic sports post-Games controlling for co-variates. The largest decreases were observed for swimming (OR: 0.51 [0.40, 0.64]), gymnastics (0.61 [0.45, 0.81]) and racquet sports (0.65 [0.51, 0.83]. There were no significant differences in pre-post participation in cycling and running. There was a significant increase in the duration per week spent online (p < 0.01), with no significant differences for time spent gaming/watching television. However individuals who were most excited by the prospect of the Olympics pre-games were more likely than the least excited to participate in nine of the ten sports post-games compared to pre-games, though this was only significant for basketball, football, racquet sports and running. Time spent online decreased significantly in the engaged group (p < 0.01).

Conclusion Adolescents who were positively engaged with the London 2012 Olympic Games reported a higher frequency of physical activity and less sedentary time spent online than those less engaged. Sporting mega-event demonstration effects are unlikely to motivate population-level increases in sporting participation unless engagement strategies can excite the interest of a greater proportion of young people.