The relationship between physical activity and metabolic syndrome risk: differential effects by exercise intensity and decision tree-based prediction model.

Authors

Woo Hyeon Son, Min Seong Ha, Yu Mi Jang

Published in

Physical activity and nutrition. Volume 30. Issue 2. Pages 23-29. Epub Jun 30, 2026.

Abstract

Metabolic syndrome (MetS) is a major risk factor for cardiovascular disease and diabetes. However, ecological evidence regarding the differential effects of exercise intensity remains limited. In this study, we examined the associations between three types of exercise (high-intensity, moderate-intensity, and strength training) and five MetS components at the population level and developed a decision tree-based model to identify high-risk groups.
Age group-level aggregate data (N = 150) from the National Health Screening Statistical Yearbook (2014-2023) were analyzed. Pearson's and age-adjusted partial correlations were used to assess associations between exercise frequency and the prevalence of metabolic conditions. A CART-based decision tree model was developed and evaluated using five-fold stratified cross-validation.
In simple correlation analyses, high-intensity exercise was negatively correlated with hypertension and hyperglycemia. After adjusting for age, moderate-intensity exercise was positively correlated with abdominal obesity, hypertension, and hyperglycemia, whereas strength training was positively correlated with hypertriglyceridemia and low HDL cholesterol (all p < 0.001). The decision tree model achieved an accuracy of 78.7% using high- and moderate-intensity exercise variables, which improved to 82.2% when all three exercise types were included. Under the MetS criterion (≥ 3 risk factors), accuracy reached 85.6% with a precision of 0.94.
Partial correlations differed in direction from findings at the individual level, indicating a potential ecological fallacy. Nevertheless, the decision tree model demonstrated acceptable predictive performance, suggesting that exercise frequency may be a useful population-level screening indicator.

PMID:
42438841
Bibliographic data and abstract were imported from PubMed on 13 Jul 2026.

Read full publication at:
Please sign in to see all details.

Stats

1-terrible, 9-excellent. How would you rate this publication? Sign in in to submit your rating.

Post a comment

You need to be signed in to post comments. You can sign in here.

Comments

There are no comments yet.