Unraveling the prevalence and key influencers of metabolic syndrome in Drug-Naïve, first-episode major depressive disorder patients with psychotic symptoms: A decision tree modeling approach

BACKGROUND: Investigating the relationship between metabolic syndrome (MetS) and first-episode, drug-naïve psychotic major depressive disorders (FEDN-PMD), this study employed decision tree modeling to identify key factors. METHODS: A cross-sectional study was conducted with 1718 FEDN-PMD patients. Sociodemographic, clinical, and blood biochemical parameters were collected, and the prevalence of MetS was assessed. RESULTS: MetS was more prevalent in the PMD group (16.96 %) compared to non-PMD (5.95 %) and whole MDD groups (7.04 %). Utilizing decision tree modeling, four key variables were identified: TSH, duration of disease, A-TPO, and age. Specifically, when TSH was less than 6.9 IU/mL, MetS incidence was 2.67 %. Higher TSH levels, in conjunction with other factors, substantially influenced MetS incidence, especially when age was 51.5 years or older, and A-TPO was greater or equal to 763.025 IU/mL. CONCLUSIONS: MetS is prevalent in FEDN-PMD patients. The decision tree model highlighted the importance of TSH, duration of disease, A-TPO, and age in predicting MetS risk. These insights could lead to personalized interventions, potentially mitigating the risk of MetS and enhancing patient outcomes.