While the Pooled Cohorts Equation for cardiovascular risk assessment has been criticized for overestimating risk, part of the problem may just be overlooked events, an analysis of the Women's Health Study suggested.
Looking just at self-reported atherosclerotic cardiovascular disease (ASCVD) events adjudicated by medical record review, the risk calculator overestimated 10-year risk:
- In the less than 5% baseline score group, observed rate of events 1.7% versus 2.8% predicted
- In the 5% to less than 7.5% baseline score group, observed 4.4% versus 6.2% predicted
- In the 7.5% to less than 10% group, observed 5.3% versus 8.7% predicted
- In the 10% or greater baseline risk score group, observed 12.4% versus 18.2% predicted
But adding Medicare claim data for additional event surveillance among the 65-and-older women in the trial brought the observed and predicted risks closer, Samia Mora, MD, MHS of Brigham and Women's Hospital in Boston, and colleagues reported in
The more complete surveillance analysis observed versus predicted risk across the four risk groups was: 3.8% versus 4.3%, 7.1% versus 6.4%, 8.3% versus 8.7%, and 18.9% versus 18.7%, respectively.
And while the Pooled Cohort Equations have been particularly problematic in minorities, risk discrimination across ethnic and racial groups was similar in the more complete surveillance data. The overall C statistic was 0.726.
The concern with overestimation of risk leading to calls for recalibration with more current cohorts is overtreatment with statins, antihypertensive drugs, and aspirin, according to an accompanying editor's note by Gregory Curfman, MD, also of Brigham and Women's.
"The central message of the study by Mora et al is that accurate measurement of ASCVD risk in populations depends on comprehensive surveillance of events, but whether this provides a complete explanation of the overestimation of risk by the [Pooled Cohort Equations] remains uncertain," he concluded.
A key limitation was that the additional Medicare data-identified events were not self-reported and were not confirmed by medical record review, Curfman noted.
Disclosures
The WHI Program is funded by the National Heart, Lung, and Blood Institute (NHLBI), National Institutes of Health, Department of Health and Human Services.
Mora reported research grant support from the NHLBI and Atherotech Diagnostics for research outside the current work and personal fees for scientific advisory board participation for Amgen, Lilly, Pfizer, and Quest Diagnostics.
Curfman disclosed no relevant relationships with industry.
Primary Source
JAMA Internal Medicine
Mora S, et al "Evaluation of the pooled cohort risk equations for cardiovascular risk prediction in a multiethnic cohort from the women's health initiative" JAMA Intern Med 2018; DOI:10.1001/jamainternmed.2018.2875.
Secondary Source
JAMA Internal Medicine
Curfman GD "Prediction of cardiovascular risk to guide primary prevention" JAMA Intern Med 2018; DOI:10.1001/jamainternmed.2018.3131.