The protective effect of physical activity against type 2 diabetes is modified by obesity and age, being most pronounced in normal-weight and middle-aged individuals, and absent in those with obesity or older adults.
Key Findings
Results
Higher physical activity levels were independently associated with reduced type 2 diabetes mellitus risk after comprehensive adjustment for confounders.
Cross-sectional data from 5042 adult participants were analyzed
Data sourced from the 2017-2020 National Health and Nutrition Examination Survey (NHANES)
Weighted multivariable logistic regression models were applied to evaluate the PA-T2DM association
The association remained significant after comprehensive adjustment for confounders
Results
The protective effect of physical activity against T2DM demonstrated a clear 'BMI gradient,' varying systematically by weight category.
The protective effect was most pronounced in normal-weight individuals
The effect was somewhat attenuated in the overweight group
The protective association was absent in participants with obesity
Restricted cubic splines (RCS) were used to delineate this dose-response relationship
BMI was identified as a significant effect modifier of the PA-T2DM association
Results
An 'age stratification' pattern emerged in the physical activity and T2DM association, with the relationship varying significantly across age groups.
The protective association of physical activity was strongest in middle-aged adults
Among younger adults, the association exhibited a borderline significant trend
No significant association between physical activity and T2DM was found in older adults
Age was identified as a significant effect modifier of the PA-T2DM association
RCS dose-response analysis was used to reveal this age-stratified pattern
Background
Individuals with severe mental disorders were identified as a population with markedly elevated susceptibility to diabetes, for whom physical inactivity is a prominent modifiable risk factor.
Physical inactivity and prolonged sedentary behavior were described as 'prominent, modifiable risk factors contributing to type 2 diabetes mellitus'
The findings were noted as 'particularly relevant for integrated care models targeting high-risk populations, such as individuals with mental disorders'
This framing was used to motivate the investigation of PA as a modifiable protective factor
Conclusions
The authors concluded that future preventive strategies should shift from universal physical activity recommendations toward more precise, stratified guidance incorporating individual BMI and age profiles.
The study found that the PA-T2DM protective association is heterogeneous rather than uniform across subgroups
The authors recommend that guidance 'incorporates individual BMI and age profiles'
This stratified approach was described as evolving from 'universal PA recommendations toward more precise, stratified guidance'
The recommendation was specifically extended to integrated care models for high-risk populations including those with mental disorders
What This Means
This research suggests that while being more physically active is generally linked to a lower risk of developing type 2 diabetes, this benefit does not apply equally to everyone. Using data from over 5,000 American adults surveyed between 2017 and 2020, researchers found that body weight and age both change how much protection physical activity provides against diabetes. Specifically, people at a normal weight appeared to benefit the most from being active, overweight individuals benefited somewhat less, and people with obesity showed no statistically significant benefit from physical activity in terms of diabetes risk reduction. Similarly, middle-aged adults showed the strongest protective association, younger adults showed only a borderline trend, and older adults showed no significant association.
These findings matter because they challenge the common assumption that a single physical activity recommendation works equally well for everyone. The research suggests that the relationship between exercise and diabetes prevention is more nuanced than previously understood, and that simply telling everyone to 'move more' may not be the most effective public health strategy. For people with obesity, for instance, physical activity alone may be insufficient to reduce diabetes risk without addressing weight or other factors.
The researchers also highlighted that people with severe mental health conditions face a particularly high risk of developing diabetes, partly because mental illness and its treatments can contribute to physical inactivity and weight gain. This research suggests that diabetes prevention programs for these and other high-risk groups could be made more effective by tailoring physical activity guidance to a person's specific age and body weight, rather than applying a one-size-fits-all approach.
Check Your Own Numbers
Upload your bloodwork. We'll cross-reference your results against this study and 4,700 others.
Liu J, Zeng Q. (2026). Physical activity and type 2 diabetes risk: Heterogeneity in the dose-response by obesity and age.. Medicine. https://doi.org/10.1097/MD.0000000000050570