State and federal guidelines are increasingly restricting the scope and depth of psychiatric research. Under new guidelines, born out of a New York State lawsuit, however, the tools available to researchers would be expanded.

A primary ethical dilemma for psychiatric researchers in the last 15 years has revolved around the enrollment of patients in psychiatric protocols when the patient’s decision-making capability is impaired. Is a patient with schizophrenia or a patient in the manic phase of bipolar disorder competent to give informed consent to participate in psychiatric research? How about a patient with dementia due to Alzheimer’s disease?

If patients are not capable of giving consent themselves, could a surrogate or medical proxy give consent on their behalf?

Paul S. Appelbaum, M.D., APA vice president and chair of the psychiatry department at the University of Massachusetts Medical School, posed these difficult questions during a workshop at APA’s 2000 annual meeting held in Chicago in May. The workshop, titled "When Can Incompetent Patients Be Research Subjects?," was aimed more at raising awareness of key ethical issues facing psychiatric researchers than trying to answer these controversial questions.

"If we are to make progress on treatment of many severe mental disorders, treatment-refractive schizophrenia, dementia, refractive major depression, bipolar disorder, just to name a few," Appelbaum said, "it seems likely that, at some point in the research process, we are going to need to include people who lack decision-making capacity in the research effort."

He asked whether in light of their mental illness people with dementia, for example, are competent to give true informed consent, understanding all of the implications of their participation in the research?

John M. Oldham, M.D., acting chair of psychiatry at the New York State Psychiatric Institute of Columbia University and chief medical officer of the New York State Office of Mental Health, offered his view on the informed consent issue.

"People assume," Oldham said, "that when one is mentally ill, he or she loses the decision-making capacity." This is not necessarily true, Oldham argued. Mentally ill patients can move in and out of decision-making competence, he said, calling it a "dynamic" process.

"There is also a differentiation," he added, "between a patient being ‘incapable of making decisions’ and ‘incapacitated or incompetent.’" The issue at hand is one of protecting the patient while allowing reasonable, approved, research to go forward.

The issue workshop was chaired by APA Area 2 Trustee Herbert S. Peyser, M.D. Also on the panel were Barry Perlman, M.D., vice president of the New York State Psychiatric Association, and Seth J. Stein, J.D., executive director and general counsel for the New York State Psychiatric Association.

Stein described current national guidelines, based on federal regulations and common law, and details in a report by the National Bioethics Advisory Commission (NBAC) that categorize research into two broad areas: research that carries minimal risk, with or without the potential for direct benefit to the subject, and research that poses greater than minimal risk to the subject.

Federal regulations and common law define "minimal risk" as that level of risk an individual would be subject to in everyday life; the NBAC also included in its definition the level of risk encountered in routine medical care.

The potential for a patient to benefit directly from a research protocol is considered in all of the sets of guidelines. A patient may directly benefit from a trial of a new medication, for example, seeing symptom improvements while on the study medication. But a patient would not directly benefit from a study measuring alterations in brain anatomy or physiology in dementia, for example, even though the brain study may lead to advances in diagnosis and treatment in the future.

All of the guidelines agree on at least two points: patients may be enrolled in studies if they are capable of giving informed consent, and patients objecting to enrollment may not be enrolled, regardless of their mental status.

"The differences of opinion start to appear," said Stein, "when you consider a non-objecting patient who is decisionally incapable.

"Surrogate decision making," Stein continued, "is acceptable, in some form, in each of the guidelines." A surrogate is most often defined as a primary relative, legal guardian, medical proxy, or someone holding power of attorney, or consent expressed on the patient’s behalf through an advance directive.

The difficulty arises when for protocols that carry a greater than minimal risk. This would include protocols involving, for example, MRI and PET scans, repeated blood draws, the use of short-term indwelling catheters, and lumbar punctures.

Under the current guidelines, if a protocol presents greater than minimal risk, the potential for benefit must be addressed. If the protocol has the potential to show direct benefit to the patient, then in most instances, the guidelines say surrogate decision making is acceptable.

What happens if a protocol presents no potential for benefit? "Then these protocols," said Stein, "become very problematic." The problem, indeed, is that most neuropsychiatric research protocols would be deemed to pose greater than minimal risk due to the types of monitoring and follow-up procedures necessary and the type of invasive interviewing or therapy that is often used. And most neuropsychiatric protocols do not show the potential for direct benefit for the patient; therefore, according to Stein, under current guidelines, most neuropsychiatric protocols would be prohibited.

A task force, set up by the state of New York following a legal challenge to its research guidelines, developed a third category of risk to help clarify research protocols.

Stein played a significant role in preparing legal briefs on behalf of the New York State Psychiatric Association and APA that were cited in the lawsuit. The third category, he explained, is "a minor increase over minimal risk." Most of the procedures noted above, such as MRI and PET scans, now fall into this category in New York.

"Minor increase over minimal risk," said Stein, "includes the risks that are of the same quality as those risks associated with the person’s specific illness or treatment that the individual would otherwise [encounter in treatment] for their illness."

The New York guidelines are not in place in other states. They directly conflict, for example, with the NBAC recommendations, which would set up a national advisory committee to review and make policy recommendations on protocols with greater than minimal risk.

Critics of NBAC, including Appelbaum and Oldham, maintained that the so called "super-IRB" envisioned by the NBAC report is not workable. The process, they said, would undoubtedly be cumbersome and time consuming.

To date, the new risk category has had a slow acceptance in New York state. Oldham and Stein strongly advocated its inclusion in federal guidelines.

"Fortunately," said Appelbaum, "the NBAC recommendations have not yet been enshrined into regulations."