Medical news often seems to follow an all-too-familiar pattern: New drugs or therapies are introduced with glowing reports, followed a few years later by headlines blaring their dangers. Consider just a few recent examples:

Vitamin E and supplemental estrogen, portrayed in the media for years as cure-alls, end up being anything but when studies show that estrogen increases the risk of breast cancer, heart attack and stroke, and vitamin E may do the same for heart failure.

Initially heralded as safer than existing drugs, the pain relievers rofecoxib (Vioxx) and valdecoxib (Bextra) make the front pages again when they get pulled from the market because of increased risk of heart attack and other problems.

After riding a long wave of good press, the reputation of the antidepressant paroxetine (Paxil) crashes amid reports that the drug may make some teenagers suicidal.

While some of the confusion stems from the natural unfolding of scientific knowledge, some comes from shortcomings in the way medical research is published and the way the mass media present medical news. Here are the questions you should keep in mind when reading medical news with a critical eye.

• Has the study been published? News reports often trumpet tantalizing results of preliminary research presented at medical meetings, relying only on the investigator's description of ongoing studies. A 2002 review of such reports found that only half ever got published in respected journals.
• Who funded--and promoted--the study? Because pharmaceutical companies help fund much medical research, it's unreasonable to dismiss all industry-related studies. It is vital, however, to know about the potential conflict of interest of the researchers involved. Be leery of any news account that omits that crucial information.
• How good is the study? The gold standard in medical research is the double-blind, controlled clinical trial, in which subjects are randomly assigned to a control (placebo) or an experimental (the real thing) group. Neither the subjects nor the researchers know who is in which group until the study is over and the code is broken. Observational studies, on the other hand, compare people who independently chose a particular health intervention against others who didn't. These studies can suggest a probable link but can't prove a causal effect. Other factors, such as lifestyle habits, can influence results.
• What's the supporting evidence? A single study seldom constitutes strong evidence of anything. Look for descriptions of previous research that pointed in the same direction or at least provided a plausible biological explanation for the finding.
• What do others have to say? Don't rely on a single news report. Check whether other sources give additional details or perspectives that provide a fuller picture. Also, look for responses of governmental agencies and reputable organizations, which can often help you gauge how seriously to take the news.
• How new is the therapy? News reports are far more likely to describe the possible benefits of a new therapy than its potential risks. And research into new drugs or treatments is likely to underestimate the risks they pose and overestimate the potential benefits, for several reasons.

First, positive studies are far more likely than negative ones to make it to print, and thus far more likely to get media attention. (Consumers Union believes drugmakers should be mandated to publish results of all clinical trials.) Second, studies testing new treatments are often conducted by only the best doctors and in the best hospitals--and only on patients most likely to benefit from the treatment and least likely to be harmed. As a result such studies might not reflect the real risks and benefits of that therapy in the general population. Finally, even the best research of new therapies can't provide evidence of their long-term safety and efficacy.