Evaluation of pre-Games effects of the Tokyo 2020 Olympic Games on Japanese population-level physical activity: a time-series analysis

Background The Olympic Games represent an opportunity to create a 'physical activity legacy' that promotes physical activity at the population level in the host nations and cities. However, previous studies showed little increase in population-level physical activity following the Olympics. The upsurge of public interest in sports and physical activity participation before the Olympics may diminish rapidly following the Games. We examined the pre-Games effects of the Olympics on Japanese population-level physical activity after the announcement of Tokyo's successful bid for the 2020 Olympic and Paralympic Games in September 2013. Methods We used publicly available data from serial cross-sectional surveys conducted with nationally or regionally representative samples in Japan seven years before and after the announcement (from 2006-2020). The outcomes were 1) daily step counts and 2) exercise habit prevalence (>= 30 min/day, >= 2 days/week, and over a year) from the National Health and Nutrition Surveys Japan (NHNS-J; 14 time points; aggregated data); and 3) sports participation (at least once a week) from the National Sports-Life Survey conducted every two years (NSLS; eight time points; individual-level data of 18,867 adults) and from the Public Opinion Survey on Sports Participation of Tokyo Residents (POSSP; eight time points; aggregated data). Age- and gender-adjusted regression models were used to estimate changes in the outcomes before and after the announcement. Results There were no significant pre-Games effects of the Olympics on national-level physical activity participation among Japanese adults. Sports participation (56.4% and 57.5%, respectively; P = 0.518), daily steps (6,535 and 6,686 steps/day; P = 0.353), and exercise habit (30.7% and 29.1%, P = 0.309) did not change significantly before and after the announcement. Although an increase in sports participation among Tokyo residents was not found in the NSLS (61.5% and 59.3%, P = 0.227), it was observed in the POSSP (49.1% and 57.7%, P = 0.019). Nonetheless, this increase might not be related to the pre-Games effects since the trend diminished following the announcement. Conclusions Population-level physical activity did not show significant changes until 2020. Realising the physical activity legacy of an Olympics may require strategic promotion and cross-agency partnership implementation in the pre- and post-event period.