Exercise & Training

Internet use and physical activity among older adults in China: a multi-dimensional analysis with endogeneity mitigation.

TL;DR

Internet use is significantly and positively associated with exercise participation among older adults in China (OR = 1.36), primarily through direct mechanisms not captured by learning-oriented empowerment or entertainment-oriented instant gratification pathways.

Key Findings

Internet use was significantly and positively associated with exercise participation among older adults in China.

  • Odds ratio of 1.36 (p < 0.01) for exercise participation among internet users versus non-users
  • Sample consisted of 6,860 adults aged 60 and above from the 2022 China Family Panel Studies (CFPS)
  • Data were nationally representative microdata
  • Baseline logistic regression models were used for the participation outcome

The association between internet use and exercise frequency and intensity was heterogeneous rather than uniformly positive.

  • While participation showed a clear positive association, frequency and intensity did not show consistent patterns
  • OLS regression models were used for frequency and intensity outcomes
  • Results were described as 'heterogeneous' across these two dimensions
  • The study examined three distinct dimensions of exercise behavior: participation, frequency, and intensity

Neither learning-oriented health empowerment nor entertainment-oriented instant gratification served as a statistically significant mediating pathway between internet use and exercise participation.

  • KHB (Karlson-Holm-Breen) decomposition was used to assess mediation
  • The indirect effect of learning-oriented health empowerment accounted for only 1.14% of the total effect
  • The indirect effect of entertainment-oriented instant gratification accounted for only 3.49% of the total effect
  • Neither indirect effect was statistically significant
  • These findings suggest the association operates primarily through direct mechanisms not captured by these two hypothesized pathways

Doubly robust treatment-effect estimators confirmed a meaningful positive association between internet use and exercise participation after addressing endogeneity.

  • IPWRA (Inverse Probability Weighted Regression Adjustment) estimated an Average Treatment Effect on the Treated (ATET) of +14.9 percentage points
  • AIPW (Augmented Inverse Probability Weighted) estimated an Average Treatment Effect (ATE) of +11.0 percentage points
  • These methods were used to address endogeneity arising from health selection and unobserved confounding
  • Results were described as 'stable across alternative variable definitions, estimation methods, and subsample analyses'

Results were robust across multiple sensitivity analyses including alternative variable definitions, estimation methods, and subsample analyses.

  • Robustness was confirmed across alternative variable definitions
  • Robustness was confirmed across alternative estimation methods
  • Subsample analyses also yielded consistent findings
  • Both doubly robust estimators (IPWRA and AIPW) produced substantively similar conclusions

The authors concluded that digital inclusion policies should be integrated into broader public health strategies for active aging, with emphasis on purposeful online engagement.

  • The paper calls for emphasis on 'enabling purposeful online engagement rather than passive digital consumption'
  • The context is accelerating population aging globally and the challenge of promoting physical activity among older adults
  • The study was conducted in China, described as a 'large developing economy'
  • The findings were framed as relevant to digital inclusion policy design

What This Means

This research suggests that older adults in China who use the internet are more likely to exercise than those who do not. Using data from nearly 7,000 Chinese adults aged 60 and older, researchers found that internet users were about 36% more likely to participate in exercise. Even after using advanced statistical methods to account for the possibility that healthier people are simply more likely to use the internet in the first place, the positive association held up — internet users were between 11 and 15 percentage points more likely to exercise than non-users. Interestingly, the study tested two common explanations for why the internet might encourage exercise: that people learn health information online (health empowerment) and that online entertainment might substitute for or motivate physical activity (instant gratification). Neither of these pathways explained the association in any meaningful statistical way — together they accounted for less than 5% of the total effect, and neither was statistically significant. This means the internet's connection to exercise participation appears to work through other, less obvious mechanisms that were not directly measured in this study. The findings have practical relevance for public health policy in aging societies. This research suggests that simply getting more older adults online — digital inclusion — may have benefits for physical activity levels, but that the type of online engagement matters. Policies encouraging purposeful, active use of the internet (rather than passive scrolling or entertainment consumption) may be more effective at promoting healthy aging. The authors suggest that digital inclusion efforts should be woven into broader strategies for keeping older populations physically active.

Check Your Own Numbers

Upload your bloodwork. We'll cross-reference your results against this study and 4,700 others.

Upload Your Labs

Have a question about this study?

Citation

Li X, Zhan B. (2026). Internet use and physical activity among older adults in China: a multi-dimensional analysis with endogeneity mitigation.. Frontiers in public health. https://doi.org/10.3389/fpubh.2026.1862565