Is there an ethical obligation to disclose controversial risk? A question from the ACCORD Trial.

Researchers designing a clinical trial may be aware of disputed evidence of serious risks from previous studies. These researchers must decide whether and how to describe these risks in their model informed consent document. They have an ethical obligation to provide fully informed consent, but does this obligation include notice of controversial evidence? With ACCORD as an example, we describe a framework and criteria that make clear the conditions requiring inclusion of important controversial risks. The ACCORD model consent document did not include notice of prior trials with excess death. We develop and explain a new standard labeled risk in equipoise. We argue that our approach provides an optimal level of integrity to protect the informational needs of the reasonable volunteers who agree to participate in clinical trials. We suggest language to be used in a model consent document and the informed consent discussion when such controversial evidence exists.