An individual’s estimated risk of having a heart attack or stroke in the next 10 years is widely used to guide preventative medication prescriptions with statins or antihypertensive drugs in those who have not yet had such an event.
To estimate that risk, doctors use equations that include different risk factors, such as age, cholesterol levels, and blood pressure. The current equations, known as the pooled cohort equations, are considered to be outdated as they were developed in 2013 based on population data from the 1960s and 70s. A new set of risk equations — known as the PREVENT equations — were developed by the American Heart Association in 2023, and are based on a more contemporary population. It is anticipated that AHA will recommend these new risk equations be used in clinical practice in the next primary prevention guidelines.
But could these new risk equations do more harm than good?
Two recent studies found that applying the PREVENT risk equations to the US population results in a much lower overall level of risk compared with the pooled cohort equations. And, if the current threshold for starting statin treatment — which is an estimated 7.5% risk of having a heart attack or stroke in the next 10 years — is kept the same, this would result in many fewer patients being eligible for statin treatment.
As cardiovascular risk is also used to guide antihypertensive treatment, the new risk equations would also result in fewer people with borderline high blood pressure being eligible for those medications.
This has raised concerns in the medical community, where there is a widespread view that many more people would benefit from primary prevention treatment, and that anything that may cause fewer people to receive these medications would be harmful.
“I believe the new equations more accurately predict the risk of the current US population, but we need to be aware of what effect that may have on use of statins,” said Tim Anderson, MD, who studies healthcare delivery at the University of Pittsburgh and is lead author of one of the studies evaluating the equations.
Anderson told Medscape Medical News that the pooled cohort equations have long been viewed as problematic.
“Because these equations were based on cohorts from the 1960s and 70s, it is believed they overestimate the current population’s risk of MI and stroke as the burden of disease has shifted in the intervening 50-60 years,” he said.
Current Equations Overestimate Risk
The new equations are based on more recent, representative, and diverse cohorts that capture a wider spectrum of the population in terms of race, ethnicity, and socioeconomic status. They also include factors that are now known to be relevant to cardiovascular risk, such as chronic kidney disease.
Anderson compared how the two sets of equations estimated risk of cardiovascular disease in the next 10 years in the US population using the NHANES survey — a large nationally representative survey conducted between 2017 and 2020.
He found that the pooled cohort equations estimated the population average 10-year risk of cardiovascular disease to be about 8%, but the PREVENT equations estimated it at just over 4%.
“The new equations estimate that the middle-aged US population have almost half the level of risk of MI and stroke over next 10 years compared with the equations used currently. So, we will substantially change risk estimates if the new equations are introduced into practice,” Anderson said.
The study found that, if the PREVENT equations are adopted in the next set of primary prevention guidelines and the current threshold of a 7.5% risk of having an MI or stroke in the next 10 years is maintained as the starting point for statin treatment, then 17.3 million adults who were previously recommended primary prevention statin therapy would no longer be eligible.
A second, similar study, conducted by a different team of US researchers, estimated that using PREVENT would decrease the number of US adults receiving or recommended for statin therapy by 14.3 million and antihypertensive therapy by 2.62 million.
The researchers, led by James A. Diao, MD, from Harvard Medical School, Boston, also suggested that over 10 years, reductions in treatment eligibility could result in an estimated 107,000 additional MI or stroke events.
Anderson points out that using the new equations would not affect the highest-risk patients. “They are still going to be high risk whichever equations are used. If you smoke a pack of cigarettes a day, have very high blood pressure or cholesterol and are older, then you are high risk. That part hasn’t changed. These people will qualify for statin treatment many times over with both sets of guidelines,” he said.
Rather it will be the large population at moderate risk of cardiovascular disease that will be affected, with far fewer of these individuals likely to get statins.
“If you are on the fence about whether to take a statin or not and you’re currently just on the threshold where they might be recommended then these new equations could mean that you’ll be less likely to be offered them,” he said. “Using the new equations may result in a delay of a couple of years to have that conversation.”
A Red Flag
Steve Nissen, MD, a cardiologist the Cleveland Clinic, is not a fan of cardiovascular risk equations in general. He points out that less than half of those currently eligible for statins are actually treated. And he believes the studies suggesting fewer people will be eligible with the new risk equations raise a red flag on whether they should be used.
“Anything that may result in fewer people being treated is a huge problem,” he told Medscape Medical News. “We have abundant evidence that we should be treating more people, not fewer people. Every study we have done has shown benefit with statins.”
The risk calculators were initially developed to limit use of statins and other medications to high-risk patients, he said, but now that we know more about safety of these drugs, it’s clear that the risks are almost non-existent.
“We really need something else to guide the prescription of statins,” said Nissen.
Nissen suggests the risk calculators and guidelines have resulted in undertreatment of the population because they lack nuance and put too much emphasis on age. We should be more interested in reducing the lifetime risk of cardiovascular events, he said. “Calculators don’t do a good job of that. Their time horizons are too short. Young people with a family history of cardiovascular disease may have a low 10-year risk on a risk calculator but their lifetime risk is elevated, and as such, they should be considered for statin treatment. We need to find a more nuanced approach to understanding the lifetime risk of individuals,” he said.
Nissen says risk calculators can be useful in high-risk patients to help demonstrate their need for treatment. “I can show them the calculator and that they have a 20% chance of an event — that can help convince them to take a statin.”
But at the lower end of the risk scale, “all it does is keep patients who should be getting treatment from having that treatment.”
Nissen said changing the risk calculator won’t affect how he treats patients. “I use judgment to decide who to treat based on scientific literature and the patient in front of me. We will engage in a discussion and make a shared decision on what is the best course of action. Calculators will never be a substitute for medical judgment,” he said.
Equations Don’t Decide
Sadiya Khan, MD, a cardiologist at Northwestern University, Evanston, Illinois, and lead author of the PREVENT equations, told Medscape Medical News that it is important to put this discussion into context.
“The two recent papers do a good job of describing differences in predictive risk between the two sets of equations but that’s where they stop,” she said. “The translation from that to the decision on who should or should not be on statins or other medications is a step too far.”
Clinical guidelines will need to be updated to take the PREVENT equations into account, as Khan argued in a JAMA editorial. So it is not clear whether the current 7.5% 10-year risk figure will remain the threshold to start treatment. Khan anticipates the guidelines committee will have to re-evaluate that threshold.
“The 7.5% risk threshold was advised in the 2013 guidelines, based on what we knew then about the balance between benefit and harm and with the knowledge that the risk equations overestimated risk,” she said. “We now have a lot more data on the safety of statin therapy. We see this frequently in preventive care. Treatments often becomes more widespread in time and use expands into lower-risk patients.”
She also pointed out that the current primary prevention guidelines encourage consideration of other factors, not just predictive risk scores, when thinking about starting statins, including very high LDL cholesterol, family history, and apo B and Lp(a) levels.
“The recommendation on who would qualify for statin therapy is not based on one number,” she said. “It is based on many considerations, including both qualitative and quantitative factors, and discussions between the patient and the doctor. It is not a straightforward yes or no based on a 7.5% risk threshold.”
The equations, she said, should only be viewed as the first step in the process, and she said she agrees with Nissen that when applying the equations, doctors need to use additional data from each individual patient to make a judgement. “Equations do not decide who gets treated. Clinical practice guidelines do that.”
Khan also agreed with Nissen that more effort is needed to identify longer term cardiovascular risk in younger people, and so the PREVENT equations include 30-year risk estimates.
“I totally agree that we need to start earlier in having these prevention conversations. The PREVENT model starts at age 30 which is 10 years earlier than the pooled cohort equations and they add a 30-year time horizon as well as the 10-year period for these discussions on predicted risk estimates,” she said. “We need to make sure we are not missing risk in young adults just because we are waiting for them to get into some arbitrary age category.”
Khan says she believes that, used correctly, the new equations will not limit access to statins or other cardiovascular treatments. “Because they are a more accurate reflection of risk in the contemporary population, the new PREVENT equations should identify the correct patients to be treated, within the confines of knowing that no risk prediction equation is perfect,” she said. “And if everything else is considered as well, not just the numbers in the risk equations, it shouldn’t result in fewer patients being treated.”
Anderson reported receiving grants from the American Heart Association, the American College of Cardiology, and the US Deprescribing Research Network. Nissen is leading a development program for a nonprescription low dose of rosuvastatin. He is also involved in trials of a new cholesterol lowering drug, obicetrapib, and on trials on drugs that lower Lp(a). Khan reported receiving grants from the American Heart Association and National Heart, Lung, and Blood Institute.