Say you have high blood pressure. There's a new blockbuster drug on the market, and your doctor lets you know about a new clinical trial you can join that is testing the new treatment against an old tried-and-true one. What's not to like? You're going to be taking, under the care of experts, one of two U.S. Food and Drug Administration–approved medications.

What you might not know—even after you sign up for the trial and have inked the informed-consent form—is that scattered reports are starting to suggest that the new medication might occasionally cause severe side effects. And the real reason the trial is being conducted with these previously released drugs is to test whether the new medication really is a lot riskier to everyone or just to a subset of patients.

If you found that out, would you still sign up for the trial? The problem is that many patients—and often even the institutional review boards that approve the trials—are never informed of these lingering questions.

This is one of the big ethical holes often left open in post-market trials, says Ruth Faden, director of the Johns Hopkins Berman Institute of Bioethics, who co-authored a new essay on this topic in The New England Journal of Medicine, which was published online August 22. She and a team of co-authors released a formal Institute of Medicine (IOM) report earlier this year recommending that the FDA improve this and other ethical aspects of post-market trials—especially those it requires.

This issue hit the headlines two years ago, when the FDA suddenly restricted the use of Avandia, a drug that had been approved to treat diabetes.  It had become clear that patients in post-market trials were at increased risk for heart attack and stroke—risks about which drugmaker GlaxoSmithKline knew but had not informed patients.

When patients sign up for clinical trials to test an experimental drug, procedure or device, they give their informed consent after doctors counsel them about the risks and uncertainty associated with a not-yet-approved intervention. Volunteers are often more than willing to accept those risks if the experimental intervention offers potentially more promising treatment for their condition, compared with what is on the market.

The benefits of signing up for a trial testing a drug that is already on the market are less clear. Patients and their doctors will already have the option to choose that same drug without being part of a study, ideally taking any known risks into consideration. If they sign up for a trial, however, they will be randomized into groups and required to take one of the test therapies—especially if one is suspected to cause occasional severe side effects—"Now you've got this heightened ethical obligation toward these people," Faden says. "They need to understand why this particular trial is being conducted." This step is often lacking in post-market trials, especially those conducted by the pharmaceutical companies and their contractors, she notes.

With so many new therapies being fast-tracked through the FDA approval process, post-release vigilance is becoming increasingly important, Faden and her colleagues noted in the piece. And when suspicion arises that a new medication might be causing more harm than it is worth, a large, formal, randomized controlled clinical trial is necessary. But informing patients fully of the risks they are taking is ethically imperative, the researchers concluded. "That's a hard sell," but potential trial participants would need to knowingly shoulder the risks to hopefully help others down the road.

FDA required post-market trials are also different ethically than comparative-effectiveness research, Faden points out. For the latter, the medical establishment truly does not know what intervention works best and for whom, as opposed to testing something that they suspect might be causing harm against an alternative.

Pulling a therapy from the market without testing is also undesirable, Faden points out, because "sometimes the worries are unfounded," or the intervention proves perfectly safe for a particular set of patients.

The FDA, which requested the IOM begin its analysis in 2010, said in response to the report that it is "currently engaged in developing a systematic process for assessing and communicating new information about a drug after it is marketed," according to a prepared statement from the Center for Drug Evaluation and Research released this spring. The statement (to which FDA representatives referred comment requests) does not make specific reference to whether the agency will take into account the IOM's recommendations.

The FDA also has ethical obligations to use the data from these post-market trials that it requires, Faden and her colleagues argue. She acknowledges that the agency already faces huge resource challenges and that overseeing additional ethical controls could be a strain. But she projects that if it could reduce the chances of future trials ending like the one for Avandia did, it would ultimately be an efficient long-term investment.

The bottom line on which both the FDA and the IOM agree is that there needs to be better, more transparent tracking of drug safety throughout a product's development, release and use. Just because a treatment gains FDA approval does not mean that it is ultimately safe or even safer than older options. "There's no line in the sand that magically happens," Faden says. "We keep learning about drugs over their lifetime."