If millions of Americans no longer qualify for a statin or a blood pressure medication based on a new calculator updated to better predict their risk, that could lead to 107,000 more heart attacks and strokes over 10 years, a new study estimates.
The research paper, published Monday in JAMA, is the second in two months drawing attention to widely used medicines designed to prevent the leading cause of death in the United States.
The research is creating a buzz in cardiology circles while two medical societies formulate new guidelines to inform practice, weighing the new risk models and existing thresholds that trigger prescriptions.
“This is concerning that we could reverse eligibility for many millions of Americans,” Raj Manrai, assistant professor of biomedical informatics in the Blavatnik Institute at Harvard Medical School and senior author of the new study, said in an interview. “We really need to reexamine the other side of the equation here, which is how those risk estimates are going to be used by patients and physicians to decide who and when individuals receive preventative care, particularly statins and antihypertensive blood pressure medications.”
Just who should be taking a statin is at the heart of both research papers analyzing risk equations released in November 2023, called Predicting Risk of cardiovascular disease EVENTs (PREVENT). That model, developed by the American Heart Association, aimed to improve on a version created in 2013, known as the Pooled Cohort Equations (PCE) and widely criticized for overestimating risk.
PREVENT drew on billing and electronic health record data from a more diverse real-world population than the older one, incorporating current statin use as well as metabolic and kidney diseases. Current guidelines used by primary care doctors are based on PCE, while the American Heart Association and American College of Cardiology review the newer PREVENT model and complete new guidelines.
Under guidelines based on the 2013 calculator, most people with a 10-year risk of 7.5% or more for developing cardiovascular disease are advised to take a statin, while at a 5% risk, they’re told only that they and their doctors should consider doing so. The threshold for hypertension treatment is blood pressure at or above 140/90 mmHg or blood pressure at or above 130/80 mmHg combined with a 10% risk for other reasons.
Rather than debate which risk calculator is better, some experts said resetting the thresholds they use is more urgent.
“It is a well-known fact that PCE overestimates risk by about two-fold,” Sadiya Khan, a cardiologist and associate professor of medicine and preventive medicine at Northwestern University Feinberg School of Medicine, told STAT in an email. She was not involved in either study but chaired the group that developed the PREVENT equations. “The PREVENT estimates more closely match contemporary event rates and therefore are approximately half of the PCE estimate. So the question isn’t more deaths or events if the threshold is the same, but what should the guidelines consider as the new threshold.”
Harvard’s Manrai concurred. “To me, the data really says that’s a very, very important and urgent conversation to be having, whether that 7.5% is that magic number at which we should initiate statins and which rests on some risk-benefit calculus which I think needs to be reexamined.”
Statins, available for about $40 a year in the U.S., work very well to lower LDL, the “bad” cholesterol that clogs blood vessels. If that atherosclerotic condition gets bad enough, it can starve the heart or the brain of oxygen-rich blood, causing a heart attack or a stroke. A variety of blood pressure medications, also at low cost, relax and open blood vessels that when blocked could lead to heart attack or stroke.
Statins’ drawbacks include potentially higher risk of type 2 diabetes. But the drugs have dramatically improved cardiovascular health, although about 40% of Americans who could benefit from statins under current guidelines don’t take them.
“Even if we leave aside the question of what is the best and most accurate model for our patients in 2024 (which I think is the first fundamental question), we still know we have a huge gap in patients being on the right treatment even when eligible,” Khan said.
Last month’s JAMA Internal Medicine study used the newer PREVENT risk equations to determine that overall, 4% of the 3,785 adults who were 40 to 75 years old in a 2017-2020 the National Health and Nutrition Examination Survey survey had a 10-year risk of developing cardiovascular disease, compared to the 8% previously predicted by the PCE. If applied in doctors’ offices around the country, the study estimated, the number of adults recommended for statins could drop by 17.3 million.
Monday’s JAMA study similarly applied the PREVENT risk calculator to 7,765 adults age 30 to 79 who answered the same national survey but from 2011-2020 (also ending as the pandemic began in March) before concluding 14.3 million fewer people would no longer meet eligibility for statins and 2.6 million fewer people would be candidates for blood pressure medications.
Then the authors went on to predict what that could mean for heart attacks and strokes if new risk tools reduced the number of people advised to take statins or blood pressure drugs. Their answer: 107,000 more heart attacks and strokes over 10 years if people are reclassified to lower risk and thus no longer benefit from preventive medications. More men than women would potentially be affected, but similar proportions of Black and white adults. People determined to be ineligible for treatment had fewer risk factors, including lower rates of obesity, hypertension, and chronic kidney disease.
“This estimate is based on the assumption that fewer people would be treated, which assumes the same threshold would be applied,” Khan said about the JAMA paper. “I think we have to expect that the threshold should and needs to change because we now have a more accurate model.”
Manrai said he was struck by the potential magnitude of changing the risk equation. “I think anytime you see a number like 107,000 heart attacks and strokes — we checked this a hundred times. We did, like, 15 sensitivity analyses to it,” he said. “I really do think it’s a pretty stunning projection for the potential impact of changing these equations across the U.S.”
Timothy Anderson, a primary care physician and an assistant professor of medicine at University of Pittsburgh Medical Center, was an author of the June analysis. He thinks there is going to be a conversation in the both primary care and cardiology communities about what is the right risk threshold.
“I’ve never seen some sort of study that tells us 5% is the perfect number versus 7% versus 2% versus 20%. Ultimately, statins are safe drugs and they do work very well,” he said in an interview. “For a lot of people, they’re willing to tolerate some amount of risk and they’re also willing to perhaps think about if I decline that statin, I will work on exercise and weight loss and get my blood pressure under control.”
Manrai and his co-authors noted there are downsides to both statins and blood pressure medications. There is an increased risk of type 2 diabetes in people treated with statins, possibly because the drugs interfere with insulin. So under the PREVENT model, which suggests less statin use, the researchers project there would be 57,800 fewer new cases of type 2 diabetes over 10 years. People on blood pressure medications can suffer sudden drops in blood pressure, lower blood supply, or sexual dysfunction, side effects that enter into decisions for patient and doctor about costs and benefits.
“There is no perfect risk estimation equation,” Jelani Grant, Chiadi Ndumele, and Seth Martin of Johns Hopkins wrote in a companion editorial. “These equations provide a starting point for risk discussions and shared decision-making in the primary prevention setting. Further risk assessment via risk-enhancing factors and arterial imaging can help tailor individual treatment plans.”
Manrai has heard that cardiologists are already plugging values from PREVENT into their risk estimates.
Not Steven Nissen, a cardiologist at the Cleveland Clinic. Nissen believes problems with risk calculators arise when they’re misconstrued as mandates.
“In my humble opinion, there is no substitute for a well-informed physician who knows what the current medical literature looks like, and he looks at the patient in front of him, and factors in everything that we know,” he said in an interview.
Asserting statins’ value across the lifespan, he has been leading an effort in collaboration with AstraZeneca to make the 5-milligram dose of its drug Crestor (rosuvastatin) available without a prescription.
“What they’re trying to say here is that if we make a mistake with a risk calculator, it can be very costly,” Nissen said about the new JAMA paper, in which he did not play a role. “I think they’re right that if we miss the mark, we are going to produce societal harm. We need to have a more thoughtful approach to deciding who to treat and not put all of our emphasis on somebody’s mathematical risk calculator.”
Khan said new thresholds for treatment would ideally be based on randomized clinical trials following people to see how well the risk model predicted later heart attacks or strokes. That would help establish at what level of risk the most benefit and least harm could be achieved from prescribing statins and blood pressure medications.
“Guidelines do not recommend a one-size-fits-all threshold,” she said. Instead, doctors and patients should discuss individual risk-enhancing factors, such as a family member who’d had a heart attack at a young age or developed the dangerous condition preeclampsia in pregnancy, as well as imaging of calcium levels in the coronary arteries.
“One of the most important parts of this discussion is to make sure that patients do not stop their statins without speaking with their doctors based on a new model until there are new guidelines available to indicate when therapy should be started or stopped,” Khan said.
STAT’s coverage of chronic health issues is supported by a grant from Bloomberg Philanthropies. Our financial supporters are not involved in any decisions about our journalism.