It’s all in the numbers; and one model comes out the winner.
In order to prevent heart disease, you have to know who is at risk for developing it. The answer depends on whether the Framingham, Reynolds, or Adult Treatment Panel III risk scoring system is used. The three models have many similarities, but also differences in the risk factors that are included in the equation. Until recently, how the systems compared was a matter of debate.
A team led by Dr. Nancy Cook, a biostatistician at Brigham and Women’s Hospital and associate professor of medicine at Harvard Medical School, took on the task. They examined how well the three systems performed in a representative sample of the nearly 72,000 racially and ethnically diverse healthy postmenopausal women in the Womens’ Health Study. The most accurate results were predicted by the Reynolds risk scoring system. The study was published in Circulation on April 10, 2012.
Why the finding is important
A good risk-prediction model allows doctors to identify who is at risk for heart disease, then weigh the risks and benefits of measures to prevent it—statins, in particular. These powerful drugs do an excellent job of lowering levels of low-density, or LDL, cholesterol, and with it, risk of heart attack. But they have side effects that can include muscle pain and diabetes.
“If your heart risk is high enough, any risk associated with statin use is outweighed by its benefits. On the other hand, if you overestimate benefit, you are not taking risk into account,” she explains.
Dr. Cook’s findings are particularly valuable for people who are at very high risk. “We can’t change a risk factor like age, but we can say, ‘Compared to someone else your age, your risk is three times higher,'” she notes. “That’s easy to understand and can be motivation for change.”