The association of mineral element concentrations and dietary habits with the risk of cardiovascular diseases: A random forest analysis

Document Type : Original Article

Authors
1 Cardiovascular Research Center, Health Policy and Promotion Institute, Kermanshah University of Medical Sciences, Kermanshah, Iran
2 Substance Abuse Prevention Research Center, Research Institute for Health, Kermanshah University of Medical Sciences, Kermanshah, Iran
3 Medical Toxicology and Drug Abuse Research Center, Birjand University of Medical Sciences, Birjand, Iran
4 Research Center for Environmental Determinants of Health (RCEDH), Health Institute, Kermanshah University of Medical Sciences, Kermanshah, Iran
5 Student Research Committee, Kermanshah University of Medical Sciences, Kermanshah, Iran
10.22034/emt.2026.598803.1000
Abstract
Background and aims: Cardiovascular diseases (CVDs) are a major cause of death worldwide, with environmental and dietary factors significantly contributing to the development of these diseases. This study aims to examine how mineral element concentrations and dietary habits together affect CVDs risk. We used a random forest model (RFM) to identify the most predictive factors associated with this risk.
Methods: We conducted a cross-sectional study of 224 participants, including 102 patients with CVDs and 122 healthy controls aged 35 to 65 years from western Iran. Serum concentrations of 11 mineral elements (aluminum (Al), boron (B), calcium (Ca), cobalt (Co), potassium (K), lithium (Li), magnesium (Mg), manganese (Mn), molybdenum (Mo), phosphorus (P), and vanadium (V)) were measured using Inductively Coupled Plasma Mass Spectrometry (ICP-MS). Dietary habits were assessed through validated questionnaires. Random forest analysis was employed to rank variable importance, followed by logistic regression to evaluate significant predictors of CVDs.
Results: The RFM demonstrated high predictive performance, with V, Li, Al, Mn, and Co emerging as the most influential mineral elements. In multivariable logistic regression analysis, higher serum V concentrations were significantly associated with CVDs (Odds ratio (OR) = 8.80, 95% CI: 5.069–15.252), whereas the association between Li and CVDs was not statistically significant based on the reported confidence interval (OR = 3.32, 95% CI: 0.431–25.467). Older age was also associated with higher odds of CVDs (OR = 1.20 per year). Among dietary factors, daily consumption of grilled foods (OR=6.435) and use of solid oil (OR=1.963) were associated with elevated risk. The RFM achieved 97% sensitivity and 100% specificity for mineral elements.
Conclusion: Serum V concentrations and selected dietary practices, particularly frequent consumption of grilled foods and use of solid oils, were associated with CVDs. These findings highlight the importance of monitoring mineral element levels and promoting healthier dietary patterns for CVDs prevention.

Highlights

  • Random forest model shows 97% sensitivity and 100% specificity for CVD prediction.
  • Daily grilled food consumption increases CVD odds 6.4 times versus monthly intake.
  • Using solid oil for cooking doubles CVD risk compared to liquid oil (OR = 1.96).
  • Serum vanadium significantly elevates CVD risk (OR = 8.80).

Keywords
Subjects


Articles in Press, Accepted Manuscript
Available Online from 15 September 2026

  • Receive Date 23 August 2026
  • Revise Date 12 September 2026
  • Accept Date 15 September 2026
  • First Publish Date 15 September 2026
  • Publish Date 15 September 2026