Richard Dobyns, MD
University of Iowa
Department of Family Medicine
First Published: 2004
Last Revised: October 2004
Peer Review Status: Internally Peer Reviewed

With so many health stories appearing each day, what should you believe? It might help to know how doctors decide what to believe.

First, medical researchers have a grading system that assesses the quality of medical research when it is published in journal articles. That system is never mentioned in news articles, however.

A look at the grading system reveals no direct correlation between the quality of a journal and quality of the research it publishes. In other words, just because research is published in a well-known journal like the New England Journal of Medicine does not mean it should immediately change how your doctor practices medicine or your own thinking.

Also, a single investigation on one group of people in one place should not change medical practice. Doctors seldom change how they provide care unless a question has been answered by meta-analysis, which looks at many different investigations on a subject before arriving at a conclusion.

Doctors can then confidently adjust their practice to incorporate new information. But this takes time, so doctors do not make practice changes as quickly as the headlines change.

Here are other approaches, also used by physicians, to be a savvy consumer of health news.

• See if the writer puts the current finding in context with similar information and reveals whether the finding changes medical practice.
• If the story is about one study, realize that the results need to be reproduced several times. If the story says, "Recommendations were based on several studies," then you can be more confident in accepting those suggestions. Of course, it is best to check with your own doctor before incorporating any advice into your health care routine.
• Find out what kind of study was used. The best kind of patient care study is a randomized clinical trial, in which clinicians randomly assign patients to receive one of two or more medical treatments, then analyze the results. In most cases, until this kind of study is done and duplicated, practice should not change.
• If a story says a condition is associated with some behavior or outcome, be aware that "associated" does not mean cause. For instance, "too many soft drinks are associated with diabetes" does not necessarily mean that pop causes this condition.
• Likewise, be wary of "scares." A finding may apply only to a small segment of the population or in unusual conditions.
• Studies that are very small or done in one center or geographical area may not apply to the general population. See if the article tells you how many patients were studied and where they lived.
• See if the story mentions who funded the study. Be wary if the only sponsor was a private industry with a special interest in the study outcome.