Benefit/Risk Assessment: What Patients Can Know That Scientists Cannot

Abstract

While it is generally recognized that medical scientists may know things about pharmacological agents that lay people cannot know, this paper suggests that there are also things about which lay people may have more knowledge. The concepts of benefit, harm, safety, and risk are analyzed suggesting that they all involve evaluative judgments about which scientists have no specialized knowledge. Insofar as these judgments apply to patients, the patients themselves may have better knowledge of whether a drug's effects will be considered beneficial or harmful and to what degree. The implications for clinical trials are explored, including the notion that randomized trials are ethical only if one is approximately indifferent to the projected net effects of each arm. Since patients may assess projected effects differently, only they can determine whether they are at the indifference point. This can mean that some patients may rationally prefer one arm of a study while the investigators are legitimately indifferent or that some patients may be indifferent when investigators have come to favor one of the arms. The implications for a right of access to experimental agents on a nonrandomized basis and for completing randomized trials are examined.