Millions of people with normal cholesterol levels in their blood could be started on cholesterol-lowering statin drugs based on a new research study, but if patients understood the numbers behind the study, they might not move so fast to put statins in their medicine cabinet. Every patient can benefit from a closer understanding of how statistics work in medicine to push people toward treatments that they may or may not really benefit from.
The latest study involves people who were put on cholesterol-lowering statins because they had a high result on a blood test called C-Reactive Protein, even though the same people did not have high cholesterol.
As reported by Tara Parker-Pope in the New York Times’ “Well” blog, here are the key numbers:
* The researchers reported an impressive sounding 50 percent reduction in heart attacks in the group treated with statins, as compared to patients in the same study who got a sugar pill (placebo) instead.
* But the real numbers of actual patients helped by the statins were only around nine in every 1,000 people treated — less than one percent.
How do those numbers fit together? In the placebo group, 18 of every 1,000 patients suffered a heart attack or some other serious heart event during the study. In the group taking the statin drug, nine of every 1,000 patients had a serious heart event. That’s how the researchers could report that the risk had been cut in half — from eighteen to nine — although the actual numbers of patients were few. Comparing eighteen to nine is called a relative risk ratio. Comparing 18/1,000 to 9/1,000 is called comparing the absolute risk. The absolute risk number is usually more meaningful.
Another important number for patients to understand in figuring out if a new medicine is for them is called the “number needed to treat.” How many patients need to be treated with the new drug for one patient to benefit?
According to a New England Journal of Medicine editorial which analyzed the new study, 120 patients would need to be treated with statins over two years for just one of those patients to benefit.
That number might be enough to persuade some patients to take the drug. But it’s a lot different than fifty percent. Bottom line: to make intelligent choices about treatments, patients need to understand how many patients like them are really expected to benefit from the treatment.