In research on standard of care interventions, when and to what extent does randomization itself pose foreseeable risks?

by Elisa A. Hurley, PhD, Education Director

Those who have been following the Surfactant, Positive Pressure, and Oxygenation Randomized Trial (SUPPORT) controversy know that the story has continued to unfold over the past month. Here I review several of the latest developments, some of which may have wide implications for the human research protections community.

Three weeks after the New England Journal of Medicine (NEJM) published a letter to the editor signed by 46 prominent bioethicists, which called for the Office for Human Research Protections (OHRP) to retract its determination letter to the University of Alabama at Birmingham (UAB), the NEJM posted a second letter, signed by a group of 45 high-profile bioethicists, that stated their agreement with OHRP’s original determination that “the informed-consent documents that were used in … SUPPORT were seriously inadequate” because they failed to appropriately describe the purpose of the research, the research procedures, the foreseeable risks, and the relevant alternatives to study participation.

The authors of this second letter further criticize the statement, included in about half the SUPPORT sites’ consent forms, that because all the treatments proposed in the study were standard of care, there was no foreseeable increase in risk:  “The potential risks and benefits of being in the study could not be said to be the same as the potential risks and benefits of receiving care outside the study, in settings in which infants were not randomly assigned and held to an oxygen level at either end of a wider range of oxygen-saturation levels generally considered to be safe.”

That very same day, making good on a promise in OHRP’s June 4 letter to UAB, the Department of Health and Human Services (DHHS) announced  that it will hold a public meeting on August 28 to gather input for shaping guidance about how to apply the federal regulations governing randomized human subjects research to studies on standard of care interventions. The announcement also invites the submission of written public comments by September 9 “regarding how an IRB should assess the risk of research involving randomization to one or more treatments within the standard of care for particular interventions, and what reasonably foreseeable risks of the research should be disclosed to research subjects in the informed consent process.” The notice then outlines five specific questions on which DHHS is seeking public guidance.

The SUPPORT study has generated heated discussion about everything from the definitions of forseeable risk and clinical equipoise, to the proper purview of OHRP, to who gets to count as a bioethicist. It’s easy to become bewildered and overwhelmed by all that has been said by those with strong views on both sides, especially when the disagreements seem to extend to some of the crucial facts around SUPPORT, including, for instance, whether the increased mortality rate at the lower ends of the oxygenation range was a foreseeable risk, and how much tailoring of oxygen levels to individual needs is involved in the standard of care for extremely premature infants.

What has emerged, however, as a core question raised by the SUPPORT study is: When we are talking about research on standard of care, when and to what extent does randomization itself pose foreseeable risks that then must be disclosed in the informed consent process? PRIM&R encourages all those within the research protections community to share their input on this question with DHHS and thereby take advantage of this important opportunity to shape future guidance for the growing domain of comparative effectiveness research.