Recently I've been reading an award-winning website (http://www.healthnewsreview.org/) run by Gary Switzer, a health journalist and founding editor-in-chief of MayoClinic.com. His insights have made me aware of some of the mistakes I have made in the past in writing articles, and I'm working diligently not to repeat them in the future.
I thought I'd share a few of his tips with you – the readers of articles about health care – to help you decide whether that article you find on the Internet or in the press that promises a new cure for what ails you is as accurate as it sounds.
Does red wine actually prevent falls in elderly people?
As an example of one of the pitfalls of believing everything you read, I present an article I stumbled across today. Its headline was Red Wine May Prevent Senior Falls, Study Finds. It reported on a study that found that subjects fed a diet high in reservatrol, a substance found in red wine, gave the subjects increased mobility, and allowed them to experience fewer falls. The article went on to say that this research suggests that the elderly – for whom falls are a major cause of death and injury – would benefit from drinking red wine.
Bzzzzzt. The "subjects" in the study were mice. There have been no corresponding studies on humans. For a human to get the same amount of reservatrol as the mice, they would have to drink four glasses of red wine a day.
In this case, the discerning reader can figure this out just from the article itself. But in many popular articles about health issues, you can't. So to help you out, I'm passing along some of Mr. Switzer's invaluable tips, along with my own.
How to read articles about health care
• Track down the original research. Read what it has to say, as opposed to what the writer of the article said about it. Often you'll find that the study authors were very careful to state that their findings were preliminary, and that the findings only hinted at or suggested benefits that the article writer presented as facts.
• Understand the difference between association and causation. An association between A and B is not the same thing as A causing B. It only means that the researchers found a statistically meaningful relationship between A and B.
• Learn the 7 words to instantly disbelieve. There are several words commonly used in articles about health issues that, when you see them, should raise a red flag for you. Those words are cure, breakthrough, miracle, promising, dramatic, hope, and victim. All of these words are intended to provoke an emotional reaction in the readers, especially if those readers or members of their families have the disease or condition that the article says that a new drug may provide a promising and dramatic miracle breakthrough cure that finally offers victims of the disease hope. Get the point?
• What is meant by "lowers your risk?" A common phrase in health care articles is that the drug or therapy or whatever "lowers your risk of contracting disease X." But what exactly is your risk of contracting disease X? Recent articles touted the benefit of a drug that reduced the risk of hip fractures by 50%. This is what is called relative risk. When you look into the real numbers, the drug reduced the risk of hip fracture from 2 per 100 women to 1 per 100 women. So the absolute risk in this case has been reduced by 1%, not 50%.
• In research that cites survey results, who was surveyed? I was recently asked to review two studies that claimed that among several hundred people surveyed, having bad teeth was their "number one dating turnoff." What I found when I looked into these two studies was fascinating. Both did, in fact, survey a number of people, who reported exactly what was claimed. But both studies 1) were paid for by providers of cosmetic dentistry services, and 2) the people surveyed were their customers, each of whom had just shelled out between five and twenty thousand dollars each to fix their own teeth. Does this constitute what in science is called "selection bias?" I think it does.
• What is the difference between Phase I and Phase II in drug trials? Often journalists report on a drug by saying that it's in "early phase drug trials." Phase I of drug trials seek only to prove that the drug is safe. Nothing is implied about the drug's actual effectiveness until Phase II of the research has been completed.
• Was the study on humans, or animals? The example mentioned above about red-wine drinking mice should make any further explanation unnecessary. Animal studies are important, but their findings do not necessarily imply similar findings in humans.
• Most important, has the study been replicated? The scientific method depends on the repeatability of the science. One study never "proves" anything. To become fact, other researchers have to replicate the first group of scientists' research, using the same methodology, and find the same or similar results.