Provincial digital development was positively associated with internet use and physical exercise, and exercise with health outcomes, but 'digital expansion did not automatically correspond to comparable growth in physical exercise or multidimensional health,' with the complete serial indirect association being level-dependent rather than uniformly supported.
Key Findings
Results
Provincial digital development increased substantially over the 10-year period, as did internet use and physical exercise scores.
Provincial digital development (21-indicator composite index) increased from 0.117 in 2013 to 0.176 in 2018, then was 0.162 in 2023.
Internet use increased from 2.132 in 2013 to 3.593 in 2023.
Physical exercise increased from 2.091 in 2013 to 2.537 in 2023.
Data were drawn from harmonized 2013, 2018, and 2023 Chinese General Social Survey waves, including 29,295 adults across 31 provincial-level units.
Results
In population-average models, digital development was positively associated with internet use, internet use with physical exercise, and physical exercise with all three health outcomes.
The complete serial indirect associations (digital development → internet use → exercise → health) were statistically supported in population-average path models.
However, these complete serial indirect associations were described as 'small.'
The study included three health outcomes (specific outcomes not named in the abstract) assessed via survey-weighted population-average path models.
Models were supplemented by constrained-model comparisons and formal between-group heterogeneity tests.
Results
Multilevel decomposition revealed that the serial indirect association was level-dependent, with within-cluster associations supported but the complete between-cluster serial chain not supported.
Within-cluster (province-year) indirect associations of internet use → exercise → health were described as 'robust.'
At the between-cluster level, digital development → internet use and internet use → exercise were supported.
However, at the between-cluster level, the exercise → health link and the complete digital development → internet use → exercise → health indirect association were not supported.
Multilevel structural equation models used 87 observed province-year clusters and additionally adjusted for province-year sports-facility resources.
Results
Formal heterogeneity tests indicated selective, stage-specific differences in the internet use–exercise–health pathway across sociodemographic subgroups.
Statistically supported heterogeneity was found by age, hukou (urban/rural registration), marital status, and education.
There was weaker evidence for differences by household economic position.
No clear evidence of heterogeneity by sex was found.
The authors characterized these differences as 'selective' and 'stage-specific,' meaning they appeared at particular steps of the serial pathway rather than uniformly across the full chain.
Results
The study found support for a 'use-to-benefit' perspective, whereby individual internet use—rather than provincial digital infrastructure alone—was consistently linked to exercise and health within local contexts.
The findings showed that digital expansion at the provincial level 'did not automatically correspond to comparable growth in physical exercise or multidimensional health.'
Individual-level internet use within province-year clusters was consistently associated with exercise and health outcomes.
The authors conclude that the proposed sequence is 'level-dependent: individual internet use was consistently linked to exercise and health within province-year contexts, but the complete between-cluster serial association was not supported.'
The study used repeated cross-sectional design (not longitudinal), limiting causal inference.
Background
The study was designed to assess digital exclusion beyond access, examining whether digital participation links to health-promoting behaviors and health outcomes across a 10-year period in China.
Three survey waves (2013, 2018, 2023) were 'deliberately selected at approximately 5-year intervals to provide repeated cross-sectional snapshots across a 10-year period.'
Provincial digital development was operationalized as a 21-indicator province-year composite index.
Sample included 29,295 adults across 31 provincial-level units in China.
The study assessed whether the internet use → physical exercise → health sequence differed across sociodemographic groups.
What This Means
This research suggests that in China, living in a province with greater digital development is associated with higher rates of internet use, and individual internet use is associated with more physical exercise, which in turn is associated with better health. However, these links do not form a simple, automatic chain from provincial digital growth all the way through to better health outcomes. When the researchers separated effects occurring at the individual level (within a given province and year) from effects at the provincial level (comparing different provinces and time points), they found that the individual-level links between internet use, exercise, and health were consistently supported, but the full chain from provincial digital development to health was not.
The study also found that the strength of the internet-exercise-health connection varied across population groups—differing by age, urban versus rural residency status (hukou), marital status, and education level—but did not clearly differ by sex. This suggests that certain groups may benefit more or less from digital participation when it comes to physical activity and health, pointing to the need for targeted support rather than assuming digital access alone will improve health equally for everyone.
This research suggests that simply expanding digital infrastructure at a regional level is not sufficient to produce population-wide health benefits. Instead, public health strategies may need to actively connect digital participation with specific behavioral support—such as encouraging physical activity—and ensure that offline opportunities for exercise remain accessible, particularly for groups who may be less likely to translate internet use into health-promoting behaviors.
Check Your Own Numbers
Upload your bloodwork. We'll cross-reference your results against this study and 4,700 others.
Yang X, Wei M, Fang C, Zhang F. (2026). Digital development and health in China: a serial pathway model through internet use and physical exercise in repeated cross-sectional surveys.. Frontiers in public health. https://doi.org/10.3389/fpubh.2026.1976170