OBGYN.net: The Universe of Women's Health

A Perspective on Medical News Reporting

by
R. Daniel Braun, MD FACOG, OBGYN.net Editorial Advisor

Medical reporting on television news programs seems to have become a standard procedure. This means that every Wednesday night, I can turn on the local news at six and get a report on the content of the "New England Journal of Medicine" that will not arrive in its subscribers' offices until tomorrow morning. If the subscribers are lucky, it will arrive prior to the multiple phone calls from their patients who heard about it on the news last night. If the subscribers have any patients in their office that morning, they certainly will not have the time (30 minutes or more per article) necessary to read, digest, interpret, and evaluate the significance of this article for the practice of medicine.

The problem with all of this is that many people don't realize what goes into medical research because they don't have the training to do so. The basic problem is that many people think "If A then B, and as a result of this, given A therefore B." This is generally true. The problem is that frequently with biological systems, we are given, "sometimes A" and think that we have been given "A". This leads to the conflicting studies that get reported on the news and have the public wondering: "What are these idiots in medical research saying now? "

Not all people are alike in their response to medications, exercise, dietary modifications, poisons, or other interventions. Perhaps the most important thing to look at in a medical study is: 'Who were the subjects of the study?'

An example is that men and women have a different incidence of heart attacks at different ages. Premenopausally, women have a much lower risk of heart attack than do men of the same age. Once women have gone through menopause the incidence increases and approaches that of men. We think that we can lower this postmenopausal incidence by giving estrogen to these postmenopausal women. In order to study this we need to study postmenopausal women and not men. In fact one reason that this was not studied previously is the finding that older men treated with estrogen for prostate cancer had a higher mortality from heart attack than they would have had from the cancer if they weren't treated. This finding in men postponed the study of this hormone in postmenopausal women for a significant period of time. Another problem with the interpretation of medical data is that not everything works such that if I give medicine A to all patients with disease C, they will all get well. Some will get well without any treatment and very few treatments will lead to cure in 100% of patients. For this reason, we have to statistically analyze our results. Mark Twain said in his autobiography: "There are three kinds of lies; lies, damn lies, and statistics."

Two hours a week for one semester of my freshman year in medical school was devoted to "Biostatistics." Most postgraduate medical training programs devote at least one hour a month to a "Journal Club." Journal clubs usually consist of a faculty member who assigns 1-3 articles to be presented and has one resident responsible for presenting each article. This resident then presents the article and discusses it. Then a faculty member usually discusses it as well and then discussion from the floor occurs. The purpose of this is to train physicians in how to read and interpret the medical literature and to sift out what is important.

The above information should explain why one week there is a study that shows that estrogen replacement protects against breast cancer and the next week an article shows that it doesn't. Review of the medical literature shows many (more than 100) articles describing no difference in the incidence of breast cancer as related to hormone replacement. There are 2 articles that describe a slight increase in the incidence of breast cancer in those women who took estrogens. These two articles received a large amount of publicity and the others did not. Why? These two articles are sensational and all the others aren't. The network or station will not increase ratings if they report the majority of these articles. Reporting one article that shows: "Physicians are giving medicine to women which causes them to have breast cancer." does wonders for the ratings.

Have I written this because I have a solution to this problem? No. I wish I did. I have written this because I think people need to be made aware of the facts. Then they can perhaps understand some of these incongruities. They might even realize that not all medical researchers are crazy or idiots.