When new guidelines for breast cancer screening appeared in 2009, the response among health professionals was swift and surprisingly contentious. Some experts praised the recommendations of the U.S. Preventive Services Task Force (an independent panel of experts appointed by the U.S. Department of Health and Human Services). But to others, the advice—that most women should wait until age 50 to start having mammograms and then only get them once every two years instead of annually—was wrongheaded and potentially dangerous. One professional group even demanded that the task force reverse its decision and return to its previous position, that most women should start mammograms at age 40.
The crossfire left millions of women confused about how to handle their health care. And many were downright angry that the experts seemed unable to give them a straight answer.
It’s easy to understand the uproar. When lives are at stake, we want doctors to know what they’re talking about and give advice we can trust, not debate. But the latest controversy underscores an uncomfortable truth: More often than we would like to think, health recommendations are based on educated guesses. Experts have to make critical decisions relying on incomplete and sometimes even contradictory evidence. In following their recommendations, the rest of us have to contend with contradictions and uncertainty.
Recent history is full of examples of how shaky the basis of even widely held assumptions can be. For years doctors recommended hormone therapy at midlife in order to lower heart disease risk after menopause. Then came studies showing that HT provided very little benefit to the heart—and created potentially serious cancer risks. Another example: Many of us remember when researchers were so convinced that beta-carotene and vitamin E prevented heart disease that they were popping the pills by the handful themselves. Enthusiasm suddenly waned when subsequent findings showed that the supplements didn’t help much and were actually dangerous for some people.
Why is certainty so elusive in medicine? One major reason is the limitations of what researchers call a randomized controlled trial, which is the gold standard for scientific evidence. In these trials, one group of people may be randomly chosen to take a new cholesterol-lowering drug while another group takes a sugar pill. When the results are clear-cut, such trials provide the most reliable evidence available for a drug’s effectiveness
and safety. And indeed, dozens of randomized controlled studies have shown that cholesterol-lowering drugs really do lower cholesterol.
But the effect of drugs is not the same in everyone. Some people see a dramatic change. In others, the numbers barely budge. Working out the averages, researchers can see when there’s a good probability the drugs will work. But as ads on television are required to state “Individual results may vary”; you may turn out to be one of those with non-budging numbers.
Drug effects are uncertain in other ways. Even if cholesterol falls, for instance, there’s no guarantee that the pills will prevent a heart attack, since people with normal cholesterol have those too. Studies suggest there’s a good probability that lowering your cholesterol will lower your odds of heart disease. But most of us aren’t comfortable with probabilities. We want to know for sure if a treatment will work, and that’s something even the best research findings can’t guarantee.
Beyond individual differences, there are other reasons uncertainty persists. In some instances, randomized trials are difficult to conduct. Take the purported health benefits of alcohol: Numerous studies show that moderate drinkers have a lower risk of heart disease than nondrinkers. But people who drink moderately may have lifestyle habits that explain their lower heart disease risk. The best way to know for sure that alcohol improves heart health would be through a randomized trial. But it is difficult to divide volunteers into groups that are told to drink or not to drink for a period of years. As a result, scientists may never know beyond a shadow of a doubt if moderate alcohol intake protects against heart disease.