IRBs tie up study with informed consent changes
Changes have little to do with protecting subjects
Gaining approvals from 36 IRBs for a multisite, multilingual epidemiologic study of tuberculosis gave Dolly Katz, who serves on the Centers for Disease Control and Prevention's IRB, insight as to what it's like on the other side of the protocol.
Her assessment: "I thought it was a nightmare," in large part because of the many requests for changes to the study's informed consent documents. Most, Katz says, had little to do with subjects' safety and seemed to focus more on protecting the institution.
Her message to other IRBs members? "Do not fuss with consent language, beyond making sure it meets the federal requirements. And do not seek to protect yourself because that's not the purpose of the consent."
As an IRB member, Katz says she's already learned that lesson as a result of her experience.
"I fuss much less with consent form language than I used to," she says. "I don't suggest changes in language unless I think those changes are central to what the study is doing."
The study, conducted at 22 sites in the United States and Canada, sought to question foreign-born residents with tuberculosis about their illness and treatment experiences. It required consent documents in 10 languages and dealt with a population that was likely to include illegal immigrants and children.
Katz, PhD, MPH, an epidemiologist with the CDC in Atlanta, says that her research team designated the CDC IRB as a central IRB for the effort, and hoped to get all 36 IRBs involved to defer to it. In the end, a third of the IRBs agreed, an accomplishment she describes as a moderate success.
But she estimates she spent hundreds of hours dealing with the IRBs, primarily answering requests to change the informed consent.
"We had a questionnaire that was an hour long to administer," Katz says. "No one requested the slightest changes to the questionnaire or the protocol. But clearly everybody is focused on this poor little consent form."
Subtle distinctions
For example, Katz says, one IRB asked that the phrase "We are asking you to be in a study" be changed to "We are inviting you to be in a study."
"You invite somebody to a party, you don't invite somebody to a study. We want to use value-neutral language."
She also notes that the consent was being translated into 10 languages and it would be nearly impossible to make such subtle distinctions in every one.
Katz says other IRBs wanted to insert language that would release the institutions from responsibility in case of unexpected injuries.
"The federal government has repeatedly said in a minimal risk study you do not need that kind of injury language, because you can presume that injury is not going to happen," she says.
And Katz suspects that some of the responses were not from IRB members but from clerical staff who didn't understand the study, since they asked for language to be inserted that applied to a clinical trial. "The person who read our consent form has not even realized that this is an epidemiologic study, not a clinical trial."
Katz says that for the most part, she declined to make the requested changes, and negotiated with the IRBs to explain her reasoning. As a result, she says, she was able to get one set of consents approved by all of the IRBs, which she says is unusual in multisite studies.
"I pointed out to the IRBs that we could not practically continue the study if we had to create site-specific consent forms in 10 different languages," Katz says. "If we had to do that for each of the three different forms – adolescent assent, pediatric permission and adult consent – that came to over 700 documents."
Engaging in dialogue
Katz says she thinks part of the problem keeping consent forms consistent across multisite studies is that many investigators aren't as persistent as she was in asking IRBs to reconsider their requested changes.
"I think that what commonly happens is that principal investigators delegate the job of getting a document through an IRB to the lowest person on the totem pole," she says. "This person simply wants to get the job done. When the IRB responds, the person is thinking, it's my job to get this fixed. So whatever the IRB wants, I'm going to change it."
Katz argues that it's better for IRBs and study researchers to engage in dialogue and listen to each other's concerns.
"I think IRBs are absolutely essential not only to the protection of human subjects, but to the health of a study," she says. "They are good partners, but in order for them to be good partners, both study researchers and IRBs need to do their jobs and act like it's a dialogue and a partnership. Study researchers should not slavishly obey IRBs and IRBs should pay attention to concerns of study researchers."
She also recommends that IRBs be more willing to defer in the cases of multisite studies, using their time to focus more fully on single-site protocols.
"An IRB needs to ride herd on a study that is only at its site, because nobody else is going to," she says. "But if I'm doing a study that's at 14 sites, it makes more sense to centralize all that – to have one IRB that says we're going to look at this and make sure it's correct and focused. If you've got 14 IRBs all responsible, then nobody is responsible."
Reference
- Davidow AL, Katz D, Reves R, et al. The challenge of multisite epidemiologic studies in diverse populations: design and implementation of a 22-site study of tuberculosis in foreign-born people. Public Health Rep 2009 May-Jun;124(3):391-9.
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