Fallout from controversial OHRP decision regarding IC may confuse PIs, IRBs
Media attention highlights unclear guidance
Whatever IRB members might think happened in the recent regulatory controversy over studying the use of infection control checklists in Michigan hospitals, the truth might be stranger.
It was at least a complicated situation when the Office of Human Research Protection (OHRP) of Rockville, MD, clashed with researchers at Johns Hopkins University (JHU) School of Medicine in Baltimore, MD.
What started as a straightforward research assessment of using proven infection control practices to reduce infections among patients on catheters in the intensive care unit of hospitals, turned into a national controversy that involved hospitals across the state of Michigan and a 45-mile radius in Maryland that includes Johns Hopkins researchers, IRBs, and OHRP regulatory officials.
OHRP's investigation into the infection control research was prompted by a telephone caller who complained that investigators had not obtained IRB approval. Only, this complaint was inaccurate, a Johns Hopkins investigator says.
Before conducting the study, JHU investigators had contacted the IRB and were told the study was exempt from IRB review,1 says Peter J. Pronovost, MD, PhD, professor of the departments of anesthesiology and critical care, surgery, and health policy and management at JHU School of Medicine. Pronovost also is the medical director of the Center for Innovations in Quality Patient Care and director of the Quality and Safety Research Group, also in Baltimore.
"The Hopkins IRB is quite sophisticated," Pronovost says. "We originally had intended to submit the study for expedited review, and they said that it was quality improvement and it was exempt, so we submitted it as exempt."
After some exchange back forth, OHRP Director Kristina C. Borror, PhD, sent JHU officials a letter on Feb. 14, 2008, stating that the research likely would have been eligible for review by the IRB in an expedited manner.
"This highlights how there is lack of clarity in the regulations if a sophisticated IRB could interpret it differently [than OHRP]," Pronovost says.
A bigger problem was OHRP's initial stand that JHU needed to obtain informed consent from all patients and IRB approval from each of the 67 hospital sites, he notes.
Since the infection control measures were used for all patients to compare outcomes to previous experience, obtaining informed consent would have been a major obstacle. Plus, it could be argued that the health care team, whose IC practices were being evaluated, could be the study's participants.
Even if the patients are seen as the likely subjects, their risk was greater if the infection control practices were not implemented, Pronovost says.
The infection control practices were based on previous research evidence that such practices as using full barrier precautions when inserting central venous catheters were optimal for preventing infections.2,3
Waiver of informed consent is allowed when the below four conditions are met, according to Ivor Pritchard, PhD, acting director of OHRP, who answered IRB's questions about the controversy in an e-mail:
- the risk to the subject is minimal;
- subjects' rights and welfare will not be adversely affected by the waiver;
- conducting the research without a waiver is not feasible or practical; and
- if appropriate, subjects are provided subsequently with additional pertinent information.
OHRP's original stance was that since the research had not sought IRB approval, the informed consent could not be waived. But in the Feb. 14, 2008, letter by Borror, OHRP noted that "the human subjects research activities described in the initial grant and IRB applications, as well as similar activities, would likely have been eligible for waiver of informed consent."
Unintended consequences
The IRB review requirement also could have unintended consequences.
If IRB review is required for all sites involved in research that reviews quality improvement data about the use of evidence-based best practices, then there also is a social justice issue.
"I think requiring IRB approval at each site creates some challenges with social justice," Pronovost says. "Forty percent of Michigan hospitals are rural, and this [requirement] can constrain hospitals that don't have an IRB."
What might happen is that smaller hospitals that don't have an IRB would choose not to participate in the project, and then their staff would not learn to follow infection control best practices, and their patients would suffer from greater infection rates, as a result, Pronovost explains.
Pronovost suggests the entire controversy might have been avoided if investigators had checked a different box on the IRB submission forms.
"Even if we had submitted it as expedited instead of exempt, it would have been a trivial difference in work," Pronovost says.
When researchers responded to OHRP's request and letters, they did resubmit the protocol to the IRB, asking for an expedited review, he notes.
"We checked a different box, and it was quickly approved," he says.
The issue also highlights the problem of regulatory agencies overreacting to complaints, Pronovost says.
"I believe the complaint about our study was signed by a concerned citizen, and that same person has filed several other letters to OHRP," he says. "If you think about the enormous work and burden that one letter caused, then we may need a more judicious response to letters."
The Michigan hospitals can continue to use the checklist and collect data for their quality improvement programs without considering the U.S. Department of Health and Human Services regulations regarding the protection of human research subjects, Pritchard says.
"The regulations do not apply when institutions are only implementing practices to improve the quality of care," Pritchard says.
"At the same time, if institutions are planning research activities examining the effectiveness of interventions to improve the quality of care, then the regulatory protections are important to protect the rights and welfare of human research subjects," Pritchard adds.
OHRP sends the wrong message to health care organizations when the agency suggests that medical institutions should engage in quality improvement projects without analyzing their effectiveness, Pronovost says.
"If there are any quality inferences made about quality of care, then we need to make sure data are accurate," he says.
In the JHU case, the follow-up research that would show whether the infection control measures and checklist have resulted in long-term reductions in infections is on hold until the Michigan hospitals receive local IRB approval, Pronovost says.
Although OHRP's latest response to Michigan hospitals is that they do not need to obtain IRB review because they are not using federal funding for the QI efforts, there remains a chilling effect on the data collection out of fear of further OHRP scrutiny, he suggests.
And this chilling impact can be seen with other institutions and IRBs around the country.
"Colleagues around the country say IRBs are hardening up because they don't want to be at risk, and no institution or investigator will want to accept that risk," Pronovost says. "So if there's that fear then what we need is greater clarity in the regulations."
OHRP has indicated that the issue should be laid to rest at the Michigan hospitals because the data collecting doesn't involve human subjects research.
"At this point the interventions that were originally part of the research study are now being implemented by the Michigan hospitals solely for clinical care purposes, and the only data being released by hospitals in Michigan to JHU for research purposes are de-identified data that are collected for clinical purposes," Pritchard says. "We believe that the project has now evolved to a stage where JHU is no longer engaged in human subjects research as defined by the regulations related to this project."
IOM may provide clarity
Pritchard also notes that OHRP will continue to work to develop guidance to assist institutions in determining when the regulatory protections are required and appropriate and when they're not.
"This area has always been murky for IRBs, and it's been murky for OHRP," Pronovost says. "OHRP says 'We're interpreting the regulations quite narrowly,' and they're certainly much tighter than some IRBs."
There will soon be an opportunity for a discussion about achieving greater clarity when the Institute of Medicine (IOM) of Washington, DC, convenes an invitation-only session on this controversy on May 1, Pronovost says.
"I think the IOM dialogue is going to get us there," Pronovost says.
References
- Lynn J, Baily MA, Bottrell M, et al. The ethics of using quality improvement methods in health care. Ann Intern Med 2007;146:666-673. Epub 2007 April 16.
- Pronovost P, Needham D, Berenholtz S, et al. An intervention to decrease catheter-related bloodstream infections in the ICU. N Engl J Med 2006;355:2725-2732.
- Miller FG, Emanuel EJ. Quality-improvement research and informed consent. N Engl J Med 2008;358:765-767.
You have reached your article limit for the month. Subscribe now to access this article plus other member-only content.
- Award-winning Medical Content
- Latest Advances & Development in Medicine
- Unbiased Content