PSYMETRIC 2.0: DEVELOPMENT, REFINEMENT AND EXTERNAL VALIDATION OF CLINIC-READY CARDIOMETABOLIC PREDICTION MODELS FOR YOUNG PEOPLE WITH PSYCHOSIS SPECTRUM DISORDERS

Background

Young people with psychosis spectrum disorders are at high risk of cardiometabolic morbidity and subsequent premature mortality, but accurate clinic-ready prediction models for this group are lacking.

Methods

We refined the Psychosis Metabolic Risk Calculator (PsyMetRiC) prediction models (comprising age, sex, ethnicity, body mass index, smoking, antipsychotics, highdensity lipoprotein, triglycerides), originally developed to predict incident metabolic syndrome within six years of a first recorded psychosis spectrum disorder in people aged 1635 years. We used primary care (CPRD, n=9,181; QResearch, n=7,457) and secondary care (South London and Maudsley NHS Foundation Trust, n=846) datasets. We revised existing predictors for finer detail (hereafter PsyMetRiC1 models) and added new ones: cardiometabolic disorder family history, antidepressants, systolic blood pressure, glycated haemoglobin (hereafter PsyMetRiC2 models). Different models now predict clinicallysignificant weight gain within one year (PsyMetRiC2-WG; logistic regression); metabolic syndrome within six years (PsyMetRiC1-MetS; PsyMetRiC2-MetS, logistic regression), and type 2 diabetes within ten years (PsyMetRiC2-T2D; Weibull regression). “Partial” model versions without biochemical results were also developed for weight gain and metabolic syndrome models. We conducted discrimination, calibration, and decision curve analyses (whole sample and by demographic subgroup); involved stakeholders; and implemented the models in a web application compliant with regulatory standards in Great Britain.

Results

Models performed well at internal (PsyMetRiC2-WG full-model: C=0·77, 95% C.I., 0·73-0·82; partial-model: C=0·76, 95% C.I., 0·72-0·80; PsyMetRiC2-T2D full-model: C=0·86, 95% C.I., 0·78-0·94) and external validation (PsyMetRiC2-MetS full-model: C=0·81, 95% C.I., 0·77-0·84; partial-model: C=0·79, 95% C.I., 0·76-0·83; PsyMetRiC2-T2D full-model: C=0·81, 95% C.I., 0·75-0·87). Calibration plots were acceptable. All models displayed evidence of clinical usefulness at all plausible thresholds. Subgroup analysis revealed some accuracy differences that did not impair clinical usefulness. The web application (https://psymetric.app/) is available for clinical use in Great Britain.

Importance: We developed prediction models for incident cardiometabolic disorders in young people with psychosis, and a clinically-available web application. PsyMetRiC encourages shared decision-making toward reduction of cardiometabolic morbidity in this high-risk group.

References

Plana-Ripoll O, Pedersen CB, Agerbo E, et al. A comprehensive analysis of mortality-related health metrics associated with mental disorders: a nationwide, register-based cohort study. Lancet 2019; 394(10211): 1827-35. Perry BI, Osimo EF, Upthegrove R, et al. Development and external validation of the Psychosis Metabolic Risk Calculator (PsyMetRiC): a cardiometabolic risk prediction algorithm for young people with psychosis. Lancet Psychiatry 2021; 8(7): 589-98.