MetS-Z: A gender- and age-specific scoring system for predicting type 2 diabetes

Abstract

Aims: Current metabolic syndrome (mets) criteria often lack consideration for age and gender differences. This study introduces the mets-Z score, a novel tool designed to enhance mets assessment and improve long-term outcome predictions.

Materials and methods: The mets-Z score was developed using principal component analysis (PCA) to weight five mets indicators-waist circumference, blood glucose, blood pressure, high-density lipoprotein (HDL) cholesterol, and triglycerides-by gender and age. Data from 188,739 Taiwan Biobank participants, stratified by gender and age groups (20-39, 40-54, 55-64, 65+ years), were analyzed. Predictive performance for type 2 diabetes mellitus onset was assessed over a 4- to 5-year follow-up.

Results: The mets-Z score achieved superior accuracy in predicting type 2 diabetes mellitus onset, with an AUC of 0.76 in men and 0.80 in women, significantly outperforming conventional indices (P < 0.0001).

Conclusions: By integrating age- and gender-specific variations, the mets-Z score provides a more personalized and precise tool for assessing metabolic and diabetes risk, surpassing existing methods. The tool is available for public use at http://bioinfolab.nhri.edu.tw/metsz/, supporting broader applications in precision medicine.

Keywords: Metabolic syndrome; Prediction; Type 2 diabetes mellitus.