Psychiatric research guidelines developed
Consensus can be difficult, but worth it
Patients with psychiatric illness pose a particular problem to researchers and to the IRBs that review studies involving them.
They can have widely varying capacity to give informed consent — and that capacity can wax and wane depending upon the progression of the illness and the patient’s current treatment. Issues such as the use of placebo and the role of surrogates can be thorny.
After dealing with several such studies on a case-by-case basis, the Human Studies Committee at Washington University School of Medicine in St. Louis created a task force to tackle a set of guidelines to better delineate what is required of researchers who deal with this vulnerable population.
The resulting "Guidelines for the Evaluation of Studies in Persons with Psychiatric Illnesses" was one of two winners of the Health Improvement Institute’s Awards for Excellence in Human Research Protection for Innovation.
John Csernansky, MD, a professor of psychiatry and chairman of one of the school’s IRBs, says the need for the guidelines grew out of the high number of studies at the university into such illnesses as dementia, schizophrenia and depression.
"There’s a lot of activity at our university around neuropsychiatric disorders, which would include subjects who might have compromised capacity to give informed consent," says Csernansky, who was one of the task force members. "Questions kept coming up at the committee level as to whether certain consent procedures were appropriate, whether certain aspects of research design were appropriate for the populations.
"After dealing with these kinds of issues on a case-by-case basis for some time, we realized that we really could benefit from some guidelines that committees could use, and for that matter that investigators could use as they prepared their proposals, he adds."
A task force was formed from about 10 members of various IRBs at the university — members included psychiatrists, clergy, and lay members such as a relative of a mentally ill patient.
The task force used schizophrenia as a prototype illness in creating the guidelines, because it had so many of the elements that make protecting subjects challenging, Csernansky says.
"Patients with this illness could have delusions, they could have hallucinations that interfere with their capacity," he says. "They could also have fundamental cognitive deficits — they don’t remember. They’re not invested in protecting their own self-interest because of their apathy. They may not be adequate advocates for themselves."
But he contends that the guidelines fit a variety of other psychiatric illnesses as well.
"We developed them thinking of all those aspects of the clinical situation that could come up in the context of research," Csernansky says. "Because schizophrenia is such a multifaceted syndrome, it made a good prototype to develop the guidelines around. If we could make it work for schizophrenia, in other words, it almost works for anything."
Breadth of opinion
The membership of the task force was varied in its viewpoints of the issues surrounding psychiatric research, a point that Csernansky says he found troublesome at first, but that he now appreciates for the breadth of opinion that was brought to the table. He credits the decision to Philip Ludbrook, MD, associate dean and chair of Washington University’s Human Studies Committee.
"It turned out to be a very wise thing to do, because by coming to a consensus that we could all live with, those guidelines had broad appeal," he says. "It wasn’t just one faction or one point of view that was represented, but we really did have to struggle to come to some sort of consensus or compromise, something we could all live with."
Reaching that consensus took about a year, with debate focusing on two main issues:
- Placebo controlled experiments — The panel decided that placebo controlled trials were justified only in very rare situations because of the danger of placing stable patients at risk for relapse.
The guidelines state that placebo controlled trials are generally acceptable only in cases where subjects have failed to respond satisfactorily to available standard therapies, or who have experienced unacceptable side effects on those therapies.
They also allow for giving placebos during a "washout" period, designed to allow a prior treatment to clear from a subject’s system.
"Placebo in that sense would be used during a brief washout period as a safety measure, but not to make the study easier to do and not for any purely scientific reason," Csernansky says. "It was justifiable as an additional safety measure but not as a purely scientific measure."
He admits that the decision was a little bit painful for him as an investigator who has done placebo controlled treatment studies in the past.
"There’s no question that some studies are not being done now that were being done five years ago," Csernansky says. "By and large, we’re talking about early Phase II studies, where the basic efficacy of the treatment is being evaluated. Some of the studies that would have been judged ethical five years ago or so would not be judged ethical now, because of the placebo issue."
- Informed consent — The guidelines treat informed consent as an ongoing process that requires monitoring of the patients to ensure they still have capacity to make decisions in their own interest.
Investigators are asked to make special efforts to procure consent, including a waiting period between consent and enrollment and asking subjects to paraphrase the major points of a trial. A qualified person independent of the research project should evaluate the subject’s ability to understand what he or she has been told.
Subjects are given the opportunity to identify someone who can act as a surrogate in case they lose decision-making capacity during the trial. When a subject gives permission, the researchers can give their contact information to family or significant others, so that someone can be notified if a subject’s condition changes.
"Sometimes a person is recruited into a study at one phase of their illness, when they may be relatively stable," Csernansky says. "But something could occur in the course of the study where that stability is interrupted or deteriorates and then of course the capacity has to be rejudged again."
Guidelines hold up over time
Csernansky says the debate at times was contentious, as the group worked to balance the competing concerns of scientific and lay members. But he contends that the difficult process of working out their differences actually made the guidelines more useful and less subject to being challenged later as inadequate.
"It would have been easier and quicker to have avoided that [debate]," Csernansky says. "We probably would have ended up with something that would have been wanting.
"I think we succeeded at finding a balance between protecting research subjects and yet creating guidelines that are usable by people who want to do this kind of research," Csernansky says. "We’ve used them now for a couple of years, and nobody has made any big demands to make revisions to them, nor has anybody strongly objected to them, so I think we must have had some sort of a balance."