Implementation research raises unique issues
Implementation research raises unique issues
Studies pose issues of who is the research subject
IRBs ordinarily are concerned with studies that test a specific drug or intervention on patients, students, or other end-users of health care services.
But sometimes the subjects of a study aren't those end-users at all, but rather providers such as physicians, nurses, or teachers. And the study doesn't test whether a drug or intervention works, but rather how best to implement it.
Such studies, known as implementation research, can raise confusing questions for IRB review, says Catarina Kiefe, PhD, MD, a professor of medicine and biostatistics at the University of Alabama at Birmingham.
"IRBs hardly ever understand it," Kiefe says. "The IRB regulatory process, by and large, was designed to protect patients, as it should be. Patients in traditional clinical research are frequently subjected to things that could be dangerous — invasive procedures or medications with potentially harmful side effects.
"When you're talking about implementation research, that's not the arena that you're in," she says. "You know that the intervention works, you're not experimenting on the patients, really. You're experimenting on the providers."
She describes implementation research as one step in the larger process of translational research — using findings generated in one setting and translating them for use in different settings. Along that continuum, an ordinary clinical trial would be the first step.
"Even once you have proven that a certain medication works in the ideal setting of a clinical trial, A) it doesn't necessarily get taken up in practice by clinicians, and B) sometimes it works very differently in the real world compared to the ivory-tower atmosphere of the traditional clinical trial," Kiefe says. "The work that goes from knowing that those interventions — medications or procedures or hardware — can work to actually getting them used and used appropriately in the real world is what's called implementation research."
An example of an implementation study might be one looking at the use of aspirin with patients suffering the first signs of a heart attack. Since previous studies already have shown that aspirin works to prevent heart attack, the implementation researcher would be testing ways to ensure that health care providers administer it promptly when needed.
Kiefe says one of the challenges of implementation research is that it often is conducted across multiple sites, such as individual physicians' practices, and requires review by many different IRBs. Some might give the study expedited review, while others require the full board to review it.
"It's one of the difficulties and challenging aspects of doing this kind of research," she says. "Each IRB will have different requirements."
Because the process and subjects being studied are different from more traditional studies, Kiefe says IRBs can get hung up on issues that confuse implementation research and other types of research.
Patients who aren't subjects
In many implementation studies, patients aren't the ones being studied — the investigator is seeing how a clinician uses a drug or intervention that already has been proven to be safe and effective.
Patients often don't have to sign informed consent documents, since there is no experimental intervention. Kiefe says IRBs often have a hard time understanding this.
"It's an area of disagreement between investigators and IRBs," she says. "It is really very individual IRB-dependent. Some IRBs can be educated to understand the difference and some simply will refuse to say there's a difference and they will insist on treating everything as if we were talking about an invasive procedure for a patient."
Complicating the issue is the fact that sometimes, patients are included as subjects in implementation studies; the investigator may look at their outcomes or ask them to provide feedback, says Sandra Naoom, MSPH, associate director and founding member of the National Implementation Research Network at the University of South Florida in Tampa.
The network provides technical and consulting services for institutions and investigators who conduct implementation studies.
Naoom cites an example of a study about teen dating violence currently being funded by the Centers for Disease Control and Prevention (CDC). She says the CDC wanted to look at the effectiveness of educational programs addressing the issue. One of their measures was how faithfully teachers conducting the program adhered to the lesson plan that had been developed for it.
"So really the target of the study is the teachers who are delivering that intervention," she says. "But as part of our measure of fidelity, we've also asked students to tell us how well the teacher is delivering the intervention."
She says investigators also will look at the outcomes for teens exposed to the program. Because some of those questions involve delicate issues of teenage sex, their parents had to give consent for them to be questioned, Naoom says.
"This CDC study is unique in that lots of studies haven't really looked at asking the person receiving the services whether the person who is delivering those services has delivered them properly," she says.
Differing risks
For providers who are the usual subjects of implementation research, the chief risk is loss of privacy, Kiefe says.
She says clinicians may be concerned about the potential for lawsuits, so investigators can obtain certificates of confidentiality, which protect the data in the study from the threat of subpoena or court order.
"The risk [to providers] is a real risk and it needs to be addressed, but it's a very different risk from putting something in your body that could kill you," Kiefe says. "That difference in risk is not appreciated usually by the IRB."
Because the interventions being studied already have been proven safe and effective, Kiefe says, the risk to patients is minimal. For that reason, she believes implementation research should be subject to a different set of rules than more traditional clinical research.
"It's not just implementation research, but other kinds of epidemiological observational research that should have a different set of rules governing them," she says. "In that kind of research — what one might call minimal risk research — the type of regulations observed should be different than the type of regulations of a chemotherapy trial or an invasive procedure trial."
She doesn't argue that such research should be exempt from review, saying investigators should have to make the case that the study is minimal risk.
While Naoom agrees that many implementation studies bear little risk to patients involved, she does offer a caveat. Implementation studies are based on existing research that shows a medication or intervention is successful in a particular population or setting. Moving the medication or intervention to a different population or setting may entail additional risk to the patients — a point that IRBs should consider in their review.
"In lots of cases, I interviewed developers of evidence-based programs and practices, and they talk about people using their programs and practices and making slight adaptations to them," Naoom says.
While it may be difficult for an IRB to know if the core components of the original research have been changed substantially, Naoom says members can ask whether the investigator is working with the original developer of the evidence-based program. The program developer who did the original research would know best what parts of the intervention should or should not be changed, she says.
"If someone implementing this intervention is going to use it with a different target population than it was tested for, I would want to know, as part of an IRB, whether they're working with the program developer, who knows the program well enough to know [if the changes create additional risk]," Naoom says.
IRBs ordinarily are concerned with studies that test a specific drug or intervention on patients, students, or other end-users of health care services.Subscribe Now for Access
You have reached your article limit for the month. We hope you found our articles both enjoyable and insightful. For information on new subscriptions, product trials, alternative billing arrangements or group and site discounts please call 800-688-2421. We look forward to having you as a long-term member of the Relias Media community.