Physical activity and metabolic syndrome: A population base study (forest and tree model algorithms)

Hossein-Ali Nikbakht; Abbas Rezaianzadeh; Mozhgan Seif; Layla Shojaie; Masoumeh Ghoddusi Johari; Haleh Ghaem

doi:10.1016/j.clnesp.2023.05.014

Physical activity and metabolic syndrome: A population base study (forest and tree model algorithms)

Clin Nutr ESPEN. 2023 Aug:56:173-179. doi: 10.1016/j.clnesp.2023.05.014. Epub 2023 May 25.

Authors

Hossein-Ali Nikbakht¹, Abbas Rezaianzadeh², Mozhgan Seif³, Layla Shojaie⁴, Masoumeh Ghoddusi Johari⁵, Haleh Ghaem⁶

Affiliations

¹ Social Determinants of Health Research Center, Health Research Institute, Babol University of Medical Sciences, Babol, Iran.
² Colorectal Research Center, Shiraz University of Medical Sciences, Shiraz, Iran.
³ Department of Epidemiology, School of Health, Shiraz University of Medical Sciences, Shiraz, Iran.
⁴ Division of GI/Liver, Department of Medicine, Keck School of Medicine, University of Southern California, Los Angeles, CA, USA.
⁵ Breast Diseases Research Center, Shiraz University of Medical Sciences, Shiraz, Iran.
⁶ Non-Communicable Diseases Research Center, Department of Epidemiology, School of Health, Shiraz University of Medical Sciences, Shiraz, Iran. Electronic address: ghaemh@sums.ac.ir.

PMID: 37344069
DOI: 10.1016/j.clnesp.2023.05.014

Abstract

Background & aim: Lifestyle changes, prominently low mobility in recent years, have increased the prevalence of metabolic syndrome (MetS), and cardiovascular disease risk. This study aimed to determine the relationship between physical activity and MetS using modern statistical methods in a population-based study.

Methods: The target population included 10,663 people aged 40-70 years in phase 1 of the Persian Kharameh cohort study conducted in 2017. The data used in this study had questions about physical activity, demographic, anthropometrics, blood pressure, and biochemical data.

Results: Participants who their activity was within the fourth quarter were 36% less likely to develop MetS than the participants in the first quarter. In the decision-Tree algorithm with all variables, physical activity was significant after gender and comorbidity. With a lack of comorbidities and physical activity less than 2338 Metabolic Equivalent of Task (MET) and age greater than 53 years, the probability was 26.7% for the male population. For the female population, if associated with comorbidities, a history of diabetes in first-degree relatives, or both, the chance of developing MetS was estimated to be 70.4%. In the decision-tree algorithm, 56.0% of the predictions for MetS were due to gender. After gender, the presence of comorbidities, age, occupation, family history of diabetes, place of residence, and physical activity was discovered as the essential variables in predicting and identifying factors associated with MetS, respectively.

Conclusion: Modern statistical methods can be used in similar research due to better presentation of results in applied clinical laws. An essential approach for treating the syndrome and preventing its complications is a lifestyle change, including educating about physical activity and promoting it.

Keywords: Cohort; Lifestyle; Machine learning methods; Metabolic syndrome; Physical activity.

MeSH terms

Cohort Studies
Exercise
Forests
Humans
Metabolic Syndrome*
Risk Factors