The federal agency charged with protecting patients in medical trials is being challenged over its effectiveness and independence, in new fallout from an experiment suspected of endangering hundreds of premature infants.
Government email records obtained by a private watchdog group depict the agency, the Office for Human Research Protections, or OHRP, as yielding to pressure from the National Institutes of Health—including from the NIH’s director, Francis S. Collins—to halt criticism of the NIH and university researchers involved in the infant study.
Beyond that one high-profile case, the OHRP has also acknowledged a sharp drop in the number of formal investigations it has been conducting, from 91 in 2001 to just one last year, leaving even its allies unsure whether it has been slacking off, grown vastly more efficient, or perhaps become paralyzed by the political intrigue.
Better practices by universities and their institutional review boards are the most likely explanation for the declining enforcement rates, said Edward Greg Koski, an associate professor of anesthesia at Harvard University who led the OHRP for its first two years, starting in 2000. But Dr. Koski admitted that, right now, “none of us know truly what is going on” inside the agency.
Both the OHRP and its organizational parent, the Department of Health and Human Services, issued statements defending their actions. They also declined to answer questions on some key points of contention, including why the OHRP still lets the NIH guide its actions nearly 15 years after the department separated them to avoid conflicts of interest between the NIH’s mission of promoting research and the OHRP’s duty to police it.
“OHRP and NIH, being separate, doesn’t mean they cannot work together,” said a department spokesman, Tait Sye, who declined to elaborate.
The problems are coming at a time when many experts argue that the rules for protecting human subjects in scientific research need an overhaul to reflect modern processes. Yet instead of bringing the research community together to make those ethically and scientifically complicated decisions, the OHRP and the turmoil surrounding it may be driving that community apart, potentially slowing medical progress, driving up costs, and even endangering patients.
“There’s a need for a lot of conversation about these things, and we’re getting a lot of argument,” said Alan R. Fleischman, a professor of clinical pediatrics at the Albert Einstein College of Medicine of Yeshiva University. “We’re not getting a lot of thoughtful, collegial, collaborative thinking to move the field.”
Lines of Authority
The problems with the infant study, known as the “Support” trial, are regarded by experts as exemplifying the change needed in both rules and practices. The trial involved some 1,300 premature infants at two dozen hospitals who, from 2004 to 2009, were assigned to one of two levels of oxygen supply to try to find which was the better amount. Previous experience had suggested that too little oxygen could increase the risk of death, and too much could lead to blindness. The experiment at some locations appeared not to have been fully explained to parents.
The OHRP last year sent a letter to the University of Alabama at Birmingham, the study’s lead institution, citing a series of problems with the work, including a failure to properly notify parents of the risks. But two months later the office retreated, postponing any move toward imposing penalties while telling the university that the OHRP’s guidance may not have been clear enough.
The email records obtained by the watchdog group, Public Citizen, show the NIH as deeply involved in persuading the OHRP to reverse course. The emails include a May 2013 message from Dr. Collins to his staff summarizing the NIH’s work with the office “to develop a consensus set of statements that OHRP could put forward to clarify the situation with the Support study.”
Officials in both agencies referred questions about the emails to the Department of Health and Human Services, which issued a statement saying the OHRP “regularly works with entities such as NIH” and universities’ institutional review boards “to ensure the protection of human subjects in research.” The emails themselves depict department officials and Jerry A. Menikoff, director of the OHRP, as clearly willing to accept the NIH’s advice on how to handle the Support matter.
At least 11 families have filed suit against the University of Alabama and its institutional review board, blaming the study for serious injuries to their babies. Of the infants in the high-oxygen group, 107 of 662 died, compared with 130 of 654 in the low-oxygen group, a statistically significant difference, the OHRP said. The data showed that if the lower oxygen level were chosen over the higher, doctors could expect one additional death for every two cases of eye disease prevented, it said.
Fewer Investigations
Even before the emails were released, questions had arisen about whether the OHRP was just losing effectiveness.
An article published this month in “Report on Research Compliance,” a specialty newsletter, noted that the OHRP opened only one formal investigation in 2013, continuing an almost continuous slide downward from the 91 cases when it began operations, in 2000. Subsequent numbers were 43 in 2005, 15 in 2006, 16 in 2007, five in 2008, and 10 in 2012, the article said. The OHRP confirmed the numbers as accurate.
Some experts, such as Arthur L. Caplan, a professor of bioethics at New York University, called the decline troubling. “It is completely unclear why that would be, and suspicious, because there’s no reason to think anything else changed from last year or years past,” Mr. Caplan said.
Others, such as Dr. Koski, are willing to credit the OHRP and the universities with progress. In part, Dr. Koski said, the decline reflects a backlog of cases that faced the OHRP when it began operating. And around the same time, he said, the office helped introduce the accreditation of human-research-protection programs, making university-based IRBs much more able to handle problems on their own.
The OHRP could help assure the public by issuing better data on its activities, Dr. Koski said, conceding that as a possible agency shortcoming during his own tenure. “I can believe that things are better,” Dr. Koski said. “It would be awfully nice, though, to have some objective statistics to support that, other than simply looking at the numbers of investigations.”
The research-protections office, in response, did provide some figures, showing it received 142 complaints last year, leading to the one investigation and to agency action in nine other instances, such as clarifying misunderstandings about a trial or ensuring a participant got paid. Dozens of other complaints either were outside the OHRP’s jurisdiction or involved clearly nongermane topics, the agency said in a response provided by the Health and Human Services Department.
Over all, the decline in formal case numbers reflects various trends, the department said in its written response, including the use of more informal ways of dealing with less-serious allegations, “which can frequently be handled by a phone call rather than opening a formal case and issuing a determination letter.”
Dr. Koski cautioned against that approach, saying formal processes should be maintained in a regulatory body. “It keeps everything aboveboard,” he said. “It preserves the integrity of the system.”
