This page was translated automatically from Turkish. Read the original

Bu sayfa, 26. Dünya Psikiyatri Kongresi'nde sunulan çalışmanın yapay zekâ desteğiyle hazırlanmış Türkçe özetidir. Çalışmanın sahibi sunan yazarlardır; atıf için orijinal kaynağa başvurun.

EV0490

Validation of PsyMetRiC for predicting metabolic syndrome in first-episode psychosis: a retrospective cohort study protocol

VALIDATING PSYMETRIC FOR METABOLIC SYNDROME PREDICTION IN FIRST-EPISODE PSYCHOSIS: A RETROSPECTIVE COHORT STUDY PROTOCOL

PsyMetRiC and Framingham identify different patients as being at risk, discrimination appears to favor PsyMetRiC, but risk stratification changes the type of care provided rather than its quantity.

The study aims to compare PsyMetRiC with a Framingham-based metabolic algorithm for predicting new-onset metabolic syndrome (MetS) over 12 months and to map the care cascade comprising recognition, recommendation, and action. Consecutive patients with first-episode psychosis (FEP) were examined in a single-center retrospective cohort in Québec; scores could be calculated for 270 of the 291 consenting patients, and 214 were included in the analysis. MetS was defined using the harmonized IDF/AHA/NHLBI 2009 criteria, and the scores were based on the studies by D'Agostino 2008 and Perry 2021; the analysis used AUC, kappa, and cascade analysis. Baseline MetS prevalence was 17.3%, and the 12-month incidence was 20.5% (CI 13.2–30.4); discrimination was calculated as 0.705 for the full PsyMetRiC model, 0.677 for the partial model, and 0.591 for Framingham, with discordance between scores expressed as kappa = 0.59. There were 26 patients (11.3%) identified as being at risk by PsyMetRiC alone and 5 patients (2.3%) identified by Framingham alone; in other words, Framingham flags five patients for every twenty-six young patients it misses. In the cascade analysis, of the 155 visits for which risk levels were calculated, 34 were classified as moderate risk and 121 as low risk, with none classified as high risk; at least one action was taken in 23/34 (68%) moderate-risk visits, and action was taken for 12 of 16 indications (75%), whereas this proportion remained at 14% in low-risk visits. Interventions per 100 visits were 16.5 (n=20) in the low-risk group and 55.9 (n=19) in the moderate-risk group for lifestyle and education; 2.5 (n=3) and 23.5 (n=8) for pharmacological interventions; and 0.0 (n=0) and 14.7 (n=5) for referral to clinical pharmacy services. Among visits resulting in action, intervention intensity was similar across strata (1.35 versus 1.39 per visit with an intervention). In conclusion, the two scores identify different patients, the difference in discrimination is considered indicative rather than definitive because confidence intervals were not calculated, and care remains largely education-focused (50 of 69 interventions involved education and group programs). The authors emphasize that a better score alone will not be sufficient if capacity at subsequent stages of the care cascade remains limited.

Tüm WCP 2026 özetleri