New model using genetic, psychosocial factors predicts incident CHD

A polysocial score (PSS) for coronary heart disease (CHD) can predict incident CHD, suggests a study. Combining PSSCHD with a polygenic risk score (PRS) results in improvements in the performance of clinical risk calculators.

This cohort study involved 388,224 UK Biobank participants (mean age 55.5 years, 42.5 percent men, 94.9 percent White) recruited between 2006 and 2010. Incident CHD was characterized by myocardial infarction and/or coronary revascularization.

Measurements included the 10-year clinical risk based on pooled cohort equations (PCE), Predicting Risk of cardiovascular disease EVENTS (PREVENT), and QRISK3; PRSCHD; and PSSCHD from 100 related covariates. The investigators also used machine-learning and time-to-event analyses and model performance indices.

Each 1-SD increase in PSSCHD resulted in a 43-percent higher risk of CHD (hazard ratio [HR], 1.43, 95 percent confidence interval [CI], 1.38–1.49; p<0.001), and every 1-SD increase in PRSCHD pointed to a 59-percent increased CHD risk (HR, 1.59, 95 percent CI, 1.53–1.66; p<0.001). PSSCHD was higher among non-White than White persons.

“The effects of PSSCHD and PRS CHD on CHD were independent and additive,” the investigators said.

At a 10-year CHD risk threshold of 7.5 percent, the addition of PSSCHD and PRSCHD to PCE led to the reclassification of about one in 10 (12 percent) participants. The risk of CHD was 1.86 times higher in the up- versus down-reclassified individuals.

In addition, the combined model was superior to PCE alone as shown by the improved net benefit while maintaining good calibration relative to the clinical risk calculators. Notably, there were similar results noted when PSSCHD and PRSCHD were included in PREVENT and QRISK3.