Revised Pooled Cohort Equations for CVD Risk Improve Accuracy of Estimates

A revision of the 2013 pooled cohort equations that includes up-to-date data and statistical techniques can improve estimates of cardiovascular disease risk, according to a study recently published in Annals of Internal Medicine. 

This population-based study included 6 cohorts with a total of 26,689 adults between the ages of 40 and 79, none of whom had prior cardiovascular disease. The study researchers measured the incidents of stroke, nonfatal myocardial infarction, or fatal coronary heart disease. Using the revised pooled cohort equations resulted in increased accuracy across all groups, including across race and sex. Approximately 11.8 million adults who were previously classified as having a 10-year risk of at least 7.5% would be assigned a lower risk with the revised equations.

The 2013 pooled cohort equations’ overall estimates for 10-year risk for atherosclerotic cardiovascular disease were 20% too high. Black individuals faced the highest rates of misestimation, with 33% of eligible black adults showing <70% or >250% the risk estimates of white individuals with the same risk factors. The 2013 equations were compared with 2 new options: an updated version with a derivation technique that accounted for statistical problems with the older equations and another that incorporated new cohort data. 

Limitations to these results include the fact that reducing the number of subjects classified as being at high risk also reduces the likelihood of those persons being recommended aspirin or medication for blood pressure or cholesterol. These newly updated equations will also need revisions over time.

The study researchers conclude that “by revising the [pooled cohort equations] with newer data and statistical methods, we could substantially improve the accuracy of [cardiovascular disease] risk estimates.”

Reference

Yadlowsky S, Hayward RA, Sussman JB, McClelland RL, Min YI, Basu S. Clinical implications of revised pooled cohort equations for estimating atherosclerotic cardiovascular disease risk [published online June 5, 2018]. Ann Intern Med. doi: 10.7326/M17-3011