Our work suggests three specific recommendations for researchers and care providers who are discussing with patients the possibility of enrolling in a phase I clinical trial.

First, we should always communicate the likelihood of benefit in terms of the number of participants expected to derive benefit from among a total population—eg, fewer than 3 out of every 100 people. Whether this is done in words or graphics, the important point is that a frequency-type concept should be used, because it is the most sincere statement about what we know. We don’t know the chance that this particular patient will benefit, but we can calculate a historical rate of benefit from prior studies.

Second, if we are interested in checking the potential participant’s understanding of the chance of benefit, we should query the person using the same frequency-type concept used to communicate the chance of benefit. For example, we might ask, “On average, how many patients enrolled in phase I clinical trials have had their cancer controlled?” with responses ranging from “0 out of 100” to “100 out of 100.”

Third, if we hear a patient expressing high expectations of direct personal benefit from participating in a phase I trial, we should not immediately conclude that the patient does not understand the nature of the phase I trial. In our research, the overwhelming number of patients who said such things were trying to voice a positive attitude for various reasons, not to describe their understanding. It is possible for a person to express hope for the best while still planning for the worst; thus, clinicians and researchers might attend more to the behaviors of the person than to their stated expectations to determine whether the person seems to appreciate his or her situation. Instead of trying to dissuade patients from expressing hope or positive attitude, we might work with them to make sure they are planning for all possible outcomes. ■

Disclosure: Dr. Weinfurt reported no potential conflicts of interest.