Cardiovascular Risk Scores Overestimate Likelihood of Cardiovascular Events

By Will Boggs MD

NEW YORK (Reuters Health) - Four of five common cardiovascular risk scores overestimate the likelihood of cardiovascular events by as much as 115%, researchers report.

"Atherosclerotic cardiovascular disease is the leading cause of death worldwide," Dr. Andrew P. DeFilippis, from the University of Louisville, Kentucky, told Reuters Health by email. "While multiple therapies are available to prevent this common disease, accurate risk assessment is essential to effectively balance the risks and benefits of therapy in primary prevention."

Dr. DeFilippis and colleagues used data from the Multi-Ethnic Study of Atherosclerosis (MESA) to see how well five cardiovascular risk scores performed in predicting actual cardiovascular events.

These risk scores included the original Framingham risk prediction algorithm to predict coronary heart disease (FRS-CHD), a modified Framingham score that incorporated ATPIII recommendations (ATPIII-FRS-CHD), the Reynolds Risk Score (RRS), a modified Framingham score to predict cardiovascular disease (FRS-CVD), and the most recent score developed by the American Heart Association and American College of Cardiology to predict atherosclerotic cardiovascular disease (AHA-ACC-ASCVD).

FRS-CHD, FRS-CVD, ATPIII-FRS-CHD, and AHA-ACC-ASCVD overpredicted cardiovascular events by 51%, 25%, 115%, and 78%, whereas RRS performed best, underpredicting cardiovascular events by only 3%, according to the Annals of Internal Medicine report online February 16.

The scores that overestimated cardiovascular risk performed slightly better in women than in men, whereas RRS overpredicted cardiovascular events by 9% in men but underpredicted cardiovascular events by 21% in women.

The researchers found discordance between observed and expected risk throughout the risk continuum, including those at moderate risk.

Limiting the evaluation to untreated participants resulted in even greater risk overestimation for all risk scores examined.

"These findings may be because subjects not treated with these modalities have healthy characteristics not captured by the risk scores or the effect of the medication is accounted for by variables in the prediction models (cholesterol levels in subjects taking statins)," Dr. DeFilippis said.

"Objective cardiovascular risk assessment tools remain a major medical advancement; however, continued effort is needed to produce more accurate risk assessment tools for today's patients," Dr. DeFilippis said. "Until new risk scores are more convincingly demonstrated to be accurate, physicians should consider interpreting the absolute risk generated by the new risk AHA-ACC score with caution."

"We recommend using risk prediction scores as a starting point, not a decision maker, for cardiovascular risk prediction therapy," Dr. DeFilippis said. "In difficult treatment decision cases, additional testing, like coronary artery calcium, may help guide therapy."

Dr. DeFilippis added, "Our next step is to explore the impact of multiple individual cardiovascular risk factors on risk score accuracy. Such an analysis will generate important insights about which factors need to be recalibrated and what new variables should be considered to develop new, more accurate risk scores for today's patients."

"There has been considerable debate about the importance of such overestimation of risk. On the one hand, systematic overestimation of risk is a significant concern if it leads to prescription of statin therapy (and its potential adverse effects) among groups in which actual risk scores are lower than predicted," write Dr. Paul M. Ridker and Dr. Nancy R. Cook from Brigham & Women's Hospital, Boston, in a related editorial.

"On the other hand, systematic overestimation of risk may be less concerning for statin use because meta-analyses indicate that statins achieve risk reductions even among patients at low levels of absolute risk and the benefits of vascular event reduction seem to exceed the risk for diabetes," they write.

How should physicians respond to this overestimation? Dr. Ridker and Dr. Cook suggest, "Clinicians might recalibrate the algorithm so that it tracks more closely with contemporary evidence, simultaneously calculate multiple risk algorithms as currently done in some Mayo Clinic prevention programs, or elect to ignore the problem and accept that more persons will be treated with a class of drugs proven to reduce vascular event rates. Physicians might also consider including revascularization procedures as an end point because they are relevant to our patients, are expensive, and are part of the trial end points used for efficacy evaluation."

Dr. Donald M. Lloyd-Jones, from Northwestern University Feinberg School of Medicine, Chicago, co-chaired the Risk Assessment Work Group that developed the ACC/AHA equations.

He told Reuters Health, "There will never be a perfect risk estimator. It is, after all, just a statistical model designed to estimate a probability (not a certainty) of having an event. The ACC/AHA Pooled Cohort Equations represent a significant step forward over prior efforts given that they provide sex- and race-specific estimates and capture risk for stroke as well as coronary heart disease. Thus, they do a substantially better job of representing the risks facing women and African-Americans, especially."

"I believe that, rather than making further incremental changes to risk scores, we would do far better to focus on implementing the highly effective and safe therapies we have for preventing cardiovascular disease, which remains the leading cause of preventable death and will affect 60% of us before we die," Dr. Lloyd-Jones said.

"The ACC/AHA 2013 guidelines provide a user-friendly framework for preventing cardiovascular disease using safe and effective therapies in patient groups shown to benefit in high-quality randomized controlled trials."

The other co-chair of the Work Group, Dr. David Goff, from the Colorado School of Public Health, Aurora, added by email, "The guidelines, while not perfect, represent our best current evidence-based approach to furthering the mission of preventing atherosclerotic cardiovascular disease (ASCVD). The paradigm of matching the intensity of preventive therapy to the level of ASCVD risk is a foundation of this approach."

Dr. Goff concluded, "We welcome efforts to improve risk assessment as well as efforts to examine new paradigms for guiding preventive interventions, such as screening for subclinical disease. While awaiting these new scientific advances, we recall the advice to not let the perfect be the enemy of the good. Implementation of these guidelines will result in additional public health benefit through further reductions in ASCVD. Let's get on with that effort while pursuing new knowledge to improve future guidelines."

SOURCE: https://bit.ly/1CBwNf5 and https://bit.ly/1AfnRxx

Ann Intern Med 2015.

(c) Copyright Thomson Reuters 2015. Click For Restrictions - https://about.reuters.com/fulllegal.asp