It happened slowly over several decades and now millions of healthy Americans are taking drugs for the rest of their lives because they might have a heart attack, stroke, or hip fracture some time in the future. As recently as 25 years ago, for example, only people who already had a heart attack were told to lower their cholesterol with drugs; now high cholesterol (the definition keeps changing) must be reduced in everyone, even kids.
Drug treatment of risk factors like bone loss is firmly entrenched in the American psyche as prevention. More accurately, though, this type of drug treatment is just risk reduction, or lowering the chance of some dire occurrence like a hip fracture. Once told you’re at high risk for something, you are expected to take a drug to lower that risk and to comply with the followup doctor visits. You have, in effect, become a patient, albeit one without any symptoms. Rarely, does the prescribing physician explain the math to you. Simply put, what is the “patient’s” risk of having a heart attack now and how much of that risk is lowered by taking the drug. One thing that is almost always left out is an estimate of the drug-related adverse effects.
I thought of all this while trawling the Web for risk calculators that might fill in the missing information for people with uninformative doctors. The websites aren’t very good, but here’s a standout: www.thennt.com The website’s name refers to the number needed to treat (NNT), a statistical term researchers use to describe how many people must be given a drug or treatment to save one person from potentially fatal condition like a stroke. Another way of putting it: how many people will risk the side effects of treatment and derive no benefit. The calculations are based are based on results of clinical trials.
TheNNT.com was started in October by several New York doctors who want other doctors and the public to understand the size of the benefits and risks for drugs or other common medical treatments. Click here for the section about cholesterol-lowering statin drugs for people without heart disease. It’s blunt and to the point: “98% of the people who take statins saw no benefit” and “0.4% were helped by preventing a stroke”. Harms are also calculated: “0.6% were harmed by developing diabetes”. Click here if you want to see how marginally effective statins are even for people with heart disease.
The sources for each treatment’s calculation are described under “Where we get the numbers”. Usually it is a systematic review, most often a Cochrane review, and there are caveats, when appropriate. A caveat example: “Virtually all of the major statin studies were paid for and conducted by their respective pharmaceutical company. A long history of misrepresentation of data and occasionally fraudulent reporting of data suggests that these results are often much more optimistic than subsequent data produced by researchers and parties that do not have a financial stake in the results.”
You are not likely to find this level of honesty on other websites, especially those funded by the pharmaceutical industry. Typical is this widely used “risk assessment tool” from the National Cholesterol Education program. (This government agency made news in 2004 when its guidelines for who should be on cholesterol-lowering drug therapy were expanded. Eight of the nine experts who wrote the guidelines had financial ties to companies that make statin drugs.) Click here for what I would call a “Should I go on drugs” quiz with seven personal questions such as age, gender, and systolic blood pressure. This calculator estimates your risk of having a heart attack—and that’s about it. If, after answering the quiz, your risk comes up as 3%, for example, which is “3 of a 100 people with this level of risk will have a heart attack in the next ten years.” If you’re inclined to look on the bright side, just flip that statistic around to “97 out of 100 people like me” are not going to have a heart attack in ten years.
Then there’s the American Heart Association “blood pressure health risk calculator”. On the opening page Schering-Plough, maker of heart drugs, and Omron, maker of home blood pressure monitoring machines, are featured as the “proud sponsors” of this website. (Note to AHA: Thanks for being so upfront but try finding sponsors without such obvious conflicts of interest.) Could this proud sponsorship explain the fearmongering I found on this website? At the start of the quiz, it says that “women’s risk of heart disease starts to rise after menopause.” That’s a tad alarmist, considering that 78% of female deaths from heart disease in the U.S. occur after the age of 75. (I got this stat from the CDC last year but don’t expect to see it in any women and heart disease awareness campaigns.)
The AHA explains its risk calculator this way: “Your risk estimates are shown in comparison to a person your same age with normal blood pressure below 120/80.” It’s easy to see how this newly lowered threshold for normal will lead doctors to prescribe inappropriate drug treatment, especially if they are unaware of this Cochrane review that found no benefit to reducing blood pressure below 140/90. Yes, the new definition of “normal” has recently become “below 120/80″. I guess we can’t expect a pharmaceutical industry-funded website to warn about the futility of taking drugs to bring blood pressure below 140/90 just because studies show this will not prolong life or reduce the chances of having a heart attack, stroke, or kidney failure.
The AHA and the National Cholesterol Education Program provide the calculations for hypertension and cholesterol that are most likely to influence physicians. Let’s hope that more of them make their way to www.thennt.com
Maryann Napoli, Center for Medical Consumers©