IRB, researchers team up on card study review
Method allows practice-based researchers to conduct quick surveys during appointments
Practice-based researchers have long complained that IRBs don't understand their work, which is based in physicians' practices and often consists of low-risk activities such as chart reviews and surveys.
But the experience of a practice-based research network in Cleveland shows that IRBs and researchers can work collaboratively and creatively to find more appropriate ways to review research methodologies that don't fit the usual mold.
The subject of the collaboration was a type of survey called a "card study," in which physicians use pocket-sized survey cards to collect information from patients during regular appointments.
While the approach has been used for more than 20 years, IRBs tend to be unfamiliar with it, says Michelle Hamilton, BA, who served as a clinical research facilitator with PBRN Shared Resource at Case Western Reserve University in Cleveland.
That unfamiliarity can lead to questions and restrictions that discourage physicians from participating, Hamilton says.
"In general, most of the people in practice-based research networks are practicing clinicians," she says. "They're not really used to going through the IRB process in the first place. So when the IRB starts raising all of these issues, they immediately say, 'I don't know what to do with this.' And just the thought of going through that can be a really big barrier."
Recognizing the problem, lead investigators at the network reached out to Philip Cola, MA, vice president for research and technology with University Hospitals in Cleveland and head of its IRB administrative office.
"They approached me and said, 'Phil, we're worried that as we try to bring research into the community and into these practices, practice-based researchers are going to get scared away by the perception of the regulatory burden that they're not used to,'" Cola says.
One of the researchers' major concerns, Cola says, was that as studies evolved and new questions were added to cards, researchers and physicians would have to go back to the IRB and start the approval process all over again — re-educating the board about how card studies worked each time.
So Cola came up with a solution: Create one "parent protocol," detailing a list of standard procedures for a generic low-risk card study, which included collection of de-identified data. Each time a new card study was created, it would be added to the parent protocol as an amendment. This amendment would still require IRB approval, but wouldn't be treated as a fresh protocol.
"You're starting from close to the end point, versus starting from back at the beginning each time you do this," Cola says.
Cola himself helped introduce the idea to the two IRB committees that would be handling the studies. He says they initially had a lot of questions about the approach. Would the model work if the studies were about two different conditions, such as diabetes and hypertension?
"They wanted to know if the generic framework would hold in both those instances," he says. "And I said, 'You guys tell me — is there something we have to do differently at a fundamental level if we're asking questions about diabetes or hypertension?' And the answer really was no."
He says that the parent protocol has been open now for more than a year and a half, with numerous amendments to it. The IRB actually likes the approach, because it helps members focus on what is important about the individual study, Cola says.
"They've seen the basic protocol over and over again," he says. "Where they need to focus now is on what's different within that framework. From the perspective of a reviewer, if you can tell me coming in to the review where to focus and what's new and different, that makes my life easier."
One key to the success of this project is that it places strict boundaries on what can be accepted as an amendment to the parent protocol, Cola says. For example, he says a card study applies to visits in which a physician and a patient interact face-to-face.
"This is designed for a busy private practice physician to pull a card out of his pocket and say, 'Now that we've done our visit, here are a couple of extra questions I'm going to use as part of a research study, and here is a consent form for you to sign," Cola says.
He said it wouldn't apply, for example, to a physician who wanted to call patients at home and ask these questions. "You've changed the method and now that would have to be reviewed separately as a new protocol for the IRB."
Cola sees the potential for this protocol to continue to be open indefinitely, with periodic reviews to ensure that it continues to satisfy regulations.
"The regulatory environment changes over time, at least the interpretations of those regulations change," he says. "The question is will we view this the same way in 2015 that we did when we first reviewed this protocol in 2010?"
In addition to yearly continuing review, the IRB office looks at protocols that have been open for at least five years to specifically question whether they still are in accord with current regulatory thinking.
While this parent protocol approach may not have wide application to other types of research methodologies, Cola says the lesson from this experience is the value of bringing people from the research and IRB sides together to work out solutions before a protocol is submitted.
"Investigators should be going to their IRBs as they're planning a study, and saying, 'Hey, can you help me think about this differently and meet me in the middle somewhere?'" Cola says. "This idea lends itself to any type of IRB, any discipline, but it brings the IRB into the development of research design much earlier than most people do."
Hamilton MD, Cola PA, Terchek JJ, et al. A Novel Protocol for Streamlined IRB Review of Practice-based Research Network Card Studies. J Am Board Fam Med 2011 Sep-Oct;24(5):605-9.