Social, genetic factors improve prediction of CHD risk

Adding a polysocial score, which included social determinants of health (SDOH) and lifestyle-psychological factors, to a polygenic risk score improved prediction of coronary heart disease (CHD) risk, according to a new study.

Researchers used data from 388,224 UK Biobank participants ages 40 to 70 years (average age, 55.5 years; 42.5% men; 94.9% White) to develop and evaluate a model that used social and polygenic scores to improve CHD risk prediction. Participants who were known to have CHD or who were taking statins were excluded. Data about demographic, genetic, SDOH, lifestyle-psychological, and other health factors were obtained from participant questionnaires, physical assessments, and biochemical assays. The correlation of SDOH and lifestyle-psychological factors, including occupation, sleep, and physical activity, with CHD was evaluated and used to develop a polysocial score. The researchers investigated the performance of the score when added to clinical risk calculators and polygenic risk scores. The study was published Dec. 10 by Annals of Internal Medicine.

The polysocial score was associated with CHD risk. SDOH and lifestyle-psychological factors that had the highest variable of importance in CHD risk prediction were related to physical activity, sleep, education, and psychological health. With every SD increase in the polysocial score, the hazard ratio for incident CHD was 1.43 (95% CI, 1.38 to 1.49; P<0.001); for polygenic risk scores, the hazard ratio was 1.59 (95% CI, 1.53 to 1.66; P<0.001). Non-White persons had higher polysocial scores than White persons, although the authors noted the limitation of a mostly White study cohort.

The effects of the two scores on CHD were independent and additive. At a 10-year CHD risk threshold of 7.5%, adding the scores to pooled cohort equations reclassified 12% of participants, with CHD risk 1.86 times higher in those reclassified up versus down, while maintaining good calibration relative to the clinical risk calculators. Factors such as having a slow walking pace, snoring, having financial difficulties, and having no educational qualification had the highest positive coefficients, while brisk walking pace, high education, and no tiredness or lethargy had the highest negative coefficients.

The findings suggest that the integration of polygenic risk scores, SDOH, and lifestyle-psychological factors in CHD risk equations can improve risk prediction and inform targeted interventions, according to the study authors. “Public health interventions, such as smoking cessation and weight reduction, are crucial but could be more effective by improving lifestyle-psychological factors, health care access, and education,” they wrote.