Statistics and the Twenty Year Rule: Why the Old Standbys May Be Safer Than Newly Approved Drugs

For much of my adult life, twenty-five years, I had the same doctor. She was a fine internist who took wonderful care of me whenever there was a crisis but we had our differences. In 2000, when my cholesterol was high, she wanted me to take a statin. Instead, I opted for changes in diet and exercise. A couple of years later when a DEXA showed osteopenia, low bone density, she urged me to take a bisphosphonate. I said “No, thanks.” to that too and increased my intake of calcium and vitamin D.

As I told my doctor, my resistance to taking new medications has partly to do with my medical history. In the mid 1940’s when she was pregnant with me my mother took diethylstilbestrol (DES), which was thought at the time to prevent miscarriages. By the 1970’s DES was found to be associated with an increased risk of cancer in girls prenatally exposed to it. In 1978 after I had a complete hysterectomy my doctor at the time said, “Take estrogen or you will get osteoporosis.” So I took estrogen. In 2002 a large Women’s Health Initiative (WHI) study associated hormone replacement therapy with an increased risk of heart disease, stroke, and breast cancer. If I started taking statins in 2000 at the age of 54, I reasoned, new research in the 2020’s would show why that had been a bad idea. By then I would be in my 70’s and I would not be amused.

How can a person weigh and evaluate the benefits and risks associated with various options? An excellent  article in the September 24-25 Wall Street Journal addresses that very topic. In “Designing a Smarter Patient,” Drs. Jerome Groopman and Pamela Hartzband discuss ways to help a patient make wise decisions about whether or not to accept particular treatments. A patient is told that a particular statin will reduce her risk of a heart attack by 30%, which sounds impressive. Then she goes to a government-sponsored web site and calculates her risk of getting a heart attack, which turns out to be 1%. So taking the statin will lower her risk from 1% to .67%, not much of a benefit considering the monetary cost and possibility of side effects.

In my case, the Cleveland Clinic’s Stroke Risk Calculator indicates that my 10-year risk of a stroke is 3%. Based on the WHI study, my increased risk of a stroke because of taking estrogen is .00012 (12 in 10,000 additional strokes), increasing my overall risk to 3.0036. I’m going to keep taking the estrogen to protect my bones.

New medications come on the market constantly. A recent study showed that many Americans believe that drugs approved by the FDA are “extremely effective” and have no serious side effects.  In fact, FDA approval only means that the benefits outweigh the risks, based on the research that has been done up to that point. There may be products already on the market that are more effective and safer. The risks (and benefits) of the new drug may not show up for years until a larger population of patients has tried them.