Hastings Center Report

As soon as I read the headline, I thought, here we go again. As a former director of the Office for Human Research Protections at the Department of Health and Human Services, I was hardly surprised when several colleagues sent me the op-ed by Atul Gawande in the New York Times last December called "A Lifesaving Checklist."¹

Dr. Gawande, a surgeon at Brigham and Women's Hospital in Boston and staff writer at the New Yorker, suggested that the federal agency responsible for protection of human subjects in research had gone too far—over the bureaucratic edge. Perhaps he was right, but his representation of the facts was not, as OHRP was quick to point out in a statement of its own.² OHRP's statement did not evoke the energetic public response that Gawande's did, but then facts have a way of being complicated and nuanced, making drama a bit harder.

Gawande's piece and the outcry it provoked gave the impression that OHRP had truly gone over the top—that it had, against all common sense and good judgment, prohibited hospitals and physicians from implementing a simple checklist for inserting a central venous catheter. Getting doctors to follow such a checklist could be a safe and effective quality improvement program that might reduce the incidence of bloodstream infections, which are believed to be caused by poor attention to technique by doctors but have become the eighth leading cause of death in the United States, according to the Centers for Disease Control and Prevention. Gawande's article might also have left the impression that OHRP viewed all efforts to improve quality of care in hospitals as clinical research [End Page 27] and had required that they be reviewed and approved by an institutional review board. At least, that seems to be the impression that prompted the chief executive officer of the American Hospital Association, Rich Umberdston, to write an irate letter to Secretary of Health and Human Services Michael Leavitt demanding that OHRP be put in its place and its determinations and statements retracted. The simple truth was that OHRP took no such draconian action.

The lead investigator in the study that sparked this controversy was Peter Pronovost, an anesthesiologist who has dedicated his career to improving safety and quality in medicine. Pronovost is a true believer in quality improvement, but he holds that quality improvement initiatives should be rigorously evaluated to determine whether a specific quality improvement initiative really works. He had proposed a research study to evaluate the effectiveness of the "lifesaving checklist," and he had submitted the proposal to the Johns Hopkins institutional review board for review and approval, just as any other properly trained, responsible investigator would. The IRB held that the study was exempt from review because it was a quality improvement program, not human subjects research. This decision prompted a complaint to OHRP that research was being done without prior review and approval and without informed consent of the human subjects who participated in it. OHRP agreed that the project was not exempt from review, forcing the investigators and the institution to take "corrective action." The trial was suspended.

Gawande—and his readers—were incensed. But when I spoke to Dr. Pronovost, even he agreed that his study was designed to be "research," as OHRP understands that term, and that OHRP's determination that IRB review was necessary was consistent with the applicable regulations. While there is room for discussion about risk levels, involvement of human subjects, and requirements for informed consent, the bottom line is that some QI initiatives are research involving human subjects, and that those initiatives are subject to federal regulations and institutional policies for research oversight, whether they turn out to be exempt from IRB review and approval, approvable by expedited review (because they pose no greater than minimal risk), or subject to full IRB review. This is where clarification and guidance are badly needed.

Ironically, hospitals can and do track catheter-related infection rates, and they can and do use procedure-related checklists to improve and ensure quality of care. Further, Pronovost...