Why do these 'medical experts' keep changing their minds?
Let’s say that for the past 10 years you’ve been paying close attention to health news as reported on TV and in newspapers including this column. Perhaps you learned about antioxidants (notably vitamins E and C and beta carotene), which you can get both from food and supplements. These antioxidants may help lower the risk of heart disease, cancer, cataracts, and other ills. So, you started taking the pills. Next thing, you read that a study done in Finland showed that not only was beta carotene not protective against lung cancer, it actually seemed to increase the risk of getting it. Feeling deceived, you stopped taking your supplements.
You may have seen something similar happening with margarine (you switched to this supposed health food a few years ago, and now it has been tagged as an artery-clogger), alcohol and breast cancer (first, moderate drinking for women is good for the heart, then only four months ago, NBC News reported that even one drink a day is a risk factor for breast cancer), hot dogs and childhood leukemia (a headline-maker that soon pooped out, since even the researchers had a hard time explaining their findings), and cell phones (cancer again — but by then you had gotten bored!). Do these folks really know what they are taking about?
In fact, the experts don’t change their minds as often as it may seem. A few researchers and journalists eagerly spread the idea that your power line, electric toaster, microwave oven — and more recently, cell phones — could give you cancer. Most experts thought all along that the evidence was pretty thin. It’s the headline writers who seem to change their minds more often than scientists.
Science is a process, not a product, a work in progress rather than a book of rules. Scientific evidence accumulates bit by bit. This doesn’t mean that scientists are bumblers (though perhaps a few are), but that they are trying to accumulate enough data to get to the truth, which is always a difficult job. Within the circle of qualified, well-informed scientists, there is bound to be disagreement, too. The same data may look different to different people. A good scientist is often his/her own severest critic.
The match for truth in a democracy is also complicated by:
• Intense public interest in health
• Hunger for quick solutions
• Journalists trying to make a routine story sound exciting
• Publishers and TV producers looking for audiences
• Scientists looking for fame and grants
• Medical journals thirsting for prestige
• Entrepreneurs thirsting for profits
It pays to keep your wits about you as you listen, watch, and read.
The search for evidence
In general, there are three ways to look for evidence about health:
• Basic research is conducted in a laboratory, involving “test tube” or “in vitro” (within glass) experiments, or experiments with animals, such as mice. Such work is vital for many reasons. For one, it can confirm observations or hunches and provide what scientists call plausible mechanisms for a theory. If, for example, a link between heart disease and smoking is suspected, laboratory experiments might show how nicotine affects blood vessels. The beauty of lab research is that it can be tightly controlled. Its limitation is that what happens in a test tube or a laboratory rate may not happen in a free-living human being.
• Clinical or interventional trials are founded on observation and treatment of human beings. As with basic research, the “gold standard” clinical trial can and must be rigorously controlled. There’ll be an experimental group or groups (receiving bona fide drug or treatment) and a control group (receiving a placebo or dummy, treatment). A valid experiment must also be “blinded,” meaning that no subject knows whether he or she is in the experimental or control group. In a double-blind trial, the researchers don’t know either. But even clinical trials have their limitations, too, such as those imposed by ethical considerations. The researchers must not knowingly endanger human life and health — there are ethics committees to make sure of this.
• Epidemiological studies generate the most news because so many of them have potential public appeal. An indispensable arm of the research epidemiology looks at the distribution of disease and risk factors for disease in a human population in an attempt to find disease determinants. Compared with clinical trials, epidemiology is beset with pitfalls. That’s because it deals with people in the real world and with situations that are hard to control.
Pitfalls and dead ends
Epidemiology studies cannot usually prove cause and effect, but can identify associations and risk factors — smoking for lung cancer, for example. It is less good at risk assessment when associations are weak.
No matter how well done, any epidemiologic study may be open to criticism. Here are just a few examples:
• People may not reliably report their eating and exercise habits. People aware of the benefits of eating vegetables may unconsciously exaggerate their vegetable consumption on a questionnaire. That’s known as “recall bias.”
• Hidden variables or “confounders” may cloud results. A study might indicate that eating broccoli reduces the risk of heart disease. But broccoli eaters may be health-conscious and get a lot of exercise. Was it the broccoli or the exercise?
• Those included in a study may seem to be a randomly selected, unbiased sample and then turn out not to be.
• Health effects, especially where cancer is concerned, may take 20 years or more to show up. It’s not always financially or humanly possible to keep a study running that long.
However, the bottom line is pretty good
None of this means epidemiology doesn’t work. One study may not prove anything, but a body of research, in which evidence accumulates bit by bit, can uncover the truth. Research into human health has made enormous strides and is still making them. There may be no such thing as a perfect study, but here is only the briefest list of discoveries that came out of epidemiologic research:
• Smoking is the leading cause of premature death in developed countries.
• High blood cholesterol is a major cause of coronary artery disease and heart attack.
• Exercise is important for good health.
• Good nutrition offers protection against cancer or, conversely, poor nutrition is a factor in the development of cancer.
• Obesity is a risk factor for heart disease, cancer, and diabetes.
Word for the wise
• “May”: does not mean “will.”
• “Contributes to,” “is linked to,” or “associated with”: does not mean “causes.”
• “Proves”: Scientific studies gather evidence in a systematic way, but one study, taken alone, seldom proves anything.
• “Breakthrough”: This happens only now and then — for example, the discovery of penicillin or the polio vaccine. But today, the word is so overworked as to be meaningless.
• “Doubles the risk” or “triples the risk”: may or may not be meaningful. Do you know what the risk was in the first place? If the risk was one in a million, and you double it, that’s still only one in 500,000. If the risk was one in 100 and doubles, that’s a big increase.
• “Significant”: A result is “statistically significant” when the association between two factors has been found to be greater than might occur at random (this is worked out by a mathematical formula). But people often take “significant” to mean “major” or “important.”
Some commonsense pointers
• Don’t jump to conclusions. A single study is no reasons for changing your health habits. Distinguish between an interesting finding and broad-based public health recommendations.
• Always look for context. A good reporter — and a responsible scientist — will always place findings in the context of other research. Yet, the typical news report seldom alludes to other scientific work.
• If it is an animal study or some other kind of laboratory study, be cautious about generalizing. Years ago, lab studies suggested that saccharin caused cancer in rats, but epidemiologic studies later showed it didn’t cause cancer in humans.
• Note the number of study participants and the study’s length. The smaller the number of subjects and the shorter the time, the greater the possibility that the findings are erroneous.
• Beware of press conferences and other hypes. Some researchers, not to mention the editors of medical journals, love to make the front page of major newspapers. The fact that the study may have been flawed, or inconclusive, or old news, may not be mentioned. This doesn’t mean though that you shouldn’t believe anything. Truth, too, may be accompanied by hype.
Of course, there is also a great deal of good reporting, and it’s an interesting challenge to follow health news. You don’t believe everything you read or see on TV about politics, business, or foreign relations, so it’s no surprise that you shouldn’t believe some health news, as well. Luckily, there are many sources for health news — none infallible, but some a lot better than others.
