Related Experiment Video
Updated: Jul 13, 2026
Implementation of a Real-Time Psychosis Risk Detection and Alerting System Based on Electronic Health Records using CogStack
Published on: May 15, 2020
Ten-year risk prediction models for atherosclerotic cardiovascular disease in Chinese patients with
Jennifer Yee-Man Tang1, Gloria Hoi-Yan Wong2, Huiquan Zhou3
1Department of Educational Psychology, The Chinese University of Hong Kong, Hong Kong, China.
Background:
Existing atherosclerotic cardiovascular disease (ASCVD) prediction models developed for the general population may inaccurately estimate risk in Chinese patients with schizophrenia-spectrum disorders. This underscores the need to validate an existing risk score or to develop a population-specific risk score.
Objective:
To develop a Hong Kong-Chinese (HK-Chinese) model for 10-year ASCVD risk estimation, incorporating both psychiatric and non-psychiatric risk factors, and to compare it against an existing comparison model.
Methods:
The sample consisted of 39,250 Chinese adults with schizophrenia-spectrum disorders between 2005 and 2015. The index date was set one year following the date of the diagnosis (range 2006-2016). The HK-Chinese model examined psychiatric comorbidities, psychiatric prescriptions and other non-psychiatric factors. The comparison model was developed by the Prediction and Management of Cardiovascular Risk in People with Severe Mental Illnesses (PRIMROSE) study, with a reduced set of predictors. The primary outcome was incident ASCVD over 10 years. Model performance was evaluated.
Results:
Over 10 years, 4.4% of patients developed an ASCVD. Significant predictors included duration of FGA, sex, age, hypertension, diabetes mellitus, and calendar year at index date. The HK-Chinese model showed satisfactory performance, comparable to the reduced PRIMROSE model but with greater parsimony. Both models tend to overestimate risk among patients in the highest predicted risk category (>90th percentile).
Conclusions:
The HK-Chinese model performs reasonably well in estimating ASCVD risk among Chinese patients with schizophrenia-spectrum disorders. With further external validation, it could support cardiovascular monitoring and promote integrated care for ASCVD in this population.
