Interruption in physical activity behavior among Chinese adults in 2015 and 2020: trends and multilevel correlates based on two national surveys

Background This study provides empirical evidence for the formulation of public health strategies to promote sustained physical activity (PA) by analyzing the changing trends and main influencing factors of interruption in PA behavior among Chinese adults in 2015 and 2020. Methods Data were drawn from the 2015 and 2020 National Fitness Surveys that included 31 provinces (autonomous regions and municipalities) in China, including 120,067 adults aged 20–59 years (54,854 in 2015 and 65,213 in 2020). χ² tests were used to compare the rates of interruption in PA behavior. Generalized linear models were used to examine the individual-, family-, and community-level factors associated with interruption in PA behavior. The covariates included age, sex, residence, education, occupation, and marital status. Results The prevalence of interruption in PA behavior was higher in 2020 than in 2015 (16.4% vs. 6.8%, χ²=2596.723, p < 0.001). At the individual level, lack of interest in PA (2015 odds ratio [OR] = 1.64, 95% confidence interval [CI]: 1.37–1.93; 2020 OR = 1.51, 95% CI: 1.32–1.73), feeling lazy about PA engagement (2015 OR = 1.16, 95% CI: 1.03–1.29; 2020 OR = 1.17, 95% CI: 1.05–1.31), and perceiving a good physical condition (2015 OR = 1.24, 95% CI: 1.04–1.49; 2020 OR = 1.11, 95% CI: 1.00–1.24) were positively associated with interruption in PA behavior. Receiving professional guidance was negatively associated with interruption (2015 OR = 0.87, 95% CI: 0.79–0.96; 2020 OR = 0.73, 95% CI: 0.55–0.93). At the family level, household responsibility in 2020 (OR = 1.13, 95% CI: 1.01–1.34) and work demands (2015 OR = 1.08, 95% CI: 1.02–1.15; 2020 OR = 1.19, 95% CI: 1.05–1.42) were associated with a higher likelihood of interruption in PA behavior. Childcare responsibility showed a modest positive association in 2020 (OR = 1.11, 95% CI: 1.02–1.22). At the community level, lack of access to PA was negatively associated with interruption in PA behavior (2015 OR = 0.80, 95% CI: 0.64–0.95; 2020 OR = 0.85, 95% CI: 0.79–0.92). Lack of suitable facilities in 2020 (OR = 1.22, 95% CI: 1.10–1.33), lack of organized groups or teams in 2020 (OR = 1.33, 95% CI: 1.21–1.47), and financial constraints in 2015 (OR = 1.09, 95% CI: 1.03–1.16, p = 0.047) were positively associated with interruption. Conclusions Between 2015 and 2020, interruption in PA behavior among Chinese adults increased substantially. Multiple-level factors (individual, family, and community) were associated with the interruption patterns. Understanding these characteristics may help inform the development of public health strategies that support sustained PA engagement.