There is little question that practice guidelines are a reality in psychiatric practice and that their influence will grow.

But the real question may be, which guidelines? For as APA prepares to publish two new practice guidelines on psychiatric disorders, other institutions are formulating guidelines on the same disorders.

The medical faculties of Duke, Columbia, and Cornell universities issued guidelines on bipolar disorder and schizophrenia, published as a supplement to the November issue of the Journal of Clinical Psychiatry.

APA past president John McIntyre, M.D., chairs APA's Steering Committee on Practice Guidelines and has also reviewed the alternative guidelines developed by the university group.

"The guidelines are very, very nice consensus statements from some outstanding experts in the country," said McIntyre. "But the difference [from the APA guidelines] is that they are consensus statements. They are [the product of] a group of individuals who got together and said what makes sense for the care of these patients."

Guidelines Not Evidence Based

Thus, these guidelines are based on opinion, albeit expert opinion, said McIntyre. "I think [the guidelines] are a marvelous contribution, and I will use them and encourage my students, residents, and others to use them also. But they are not, like the APA guidelines, evidence based."

In the APA guidelines, the evidence to support various recommendations is "clearly laid out," said McIntyre. "So the psychiatrist is able to identify what is the evidence to support the various recommendations."

Another difference between the APA guidelines and the consensus guidelines involves the review process, according to McIntyre. The process of approving the APA guidelines "involves hundreds of members of our Association and review by people in dozens of other organizations, including major medical organizations; nonpsychiatric mental health organizations; and patient, consumer, and advocate organizations," said McIntyre. After going through this "extensive process, [the guidelines] are actually voted on by the governance of this Association so the message to the field is that these are the guidelines that APA approves."

In contrast, the other guidelines "have a list of whatever individuals" formulated them, said McIntyre. "In terms of the process of development, and the fact ours are evidence based, they are significantly different."

Moreover, APA has an explicit policy prohibiting use of pharmaceutical company money in the development and publication of practice guidelines, while the consensus guidelines are fully funded by pharmaceutical company grants, McIntyre noted.

Allen Frances, M.D., chaired APA's Task Force on DSM-IV and is professor and chair of the department of psychiatry at Duke University Medical Center. He directed the development of the consensus guidelines in conjunction with John Docherty, M.D., professor and vice chair of the department of psychiatry at Cornell University Medical College, and David Kahn, M.D., an associate clinical professor at Columbia University College of Physicians and Surgeons.

Need for Both Guidelines

There is a need for both their guidelines and the APA guidelines, Frances asserted. The APA guidelines have "summarized the clinical literature with elegance and class," he added. "It will always be necessary to build on the scientific literature, and having the APA guidelines constantly updated will always be a benefit to the field," Frances said.

But the inherent limitations of controlled research make it impossible for the APA guidelines to keep pace with clinical practice, according to Frances. "The clinical research literature is usually behind clinical application," he said.

Because clinical research controls variables so rigidly, it differs from the complexity of real clinical practice. The scientific literature may not generalize well because of the nature of inclusion and exclusion criteria, Frances noted. Further, "there are lots of questions that clinicians are faced with every day that have not been researched and probably won't be researched in the near future because there are so many questions to ask," and the economic incentives for formal research are often lacking, he pointed out. In this category fall such questions as how many days a psychiatrist should wait between discharging a patient and scheduling his or her first appointment, Frances said.

There is, for example, almost no literature on bipolar depression, but clinicians see people with that disorder and must do something, he noted. "So there are clinical practice habits, and the best way of systematizing those is to get a wide array of expert opinion."

In clinical practice comorbidity, such as substance abuse, is a real issue. But in the artificial constraints of the laboratory, comorbidity may cause researchers to exclude prospective subjects. This is especially significant since about one-half to two-thirds of people with severe and persistent mental illness abuse alcohol or drugs, Frances added.

"I think the APA guidelines have the problem of not being nearly specific enough for the questions clinicians need to have answered," Frances continued. "Our guidelines fill in those holes with expert opinion but at the expense of not having scientific proof for all these opinions."

In addition to giving clinicians a valuable tool, the consensus guidelines can help "manage managed care," according to Frances. "Practice is being standardized as we speak. At this point there is no way we will be moving into the future without guidelines. But the question is, Who will be writing them and with what values? One of the dangers would be to have guidelines developed by the insurance industry with the major goals of limiting costs and denying care. The guidelines we've developed help to provide a voice to the field about what's necessary to provide adequate care."

The university group's practice of accepting drug company money to develop the guidelines could present a possible conflict of interest, Frances conceded. But he said that he and his colleagues used "a systematic method" in analyzing and reporting all results and gave themselves "almost no editorial freedom" even if he, Docherty, and Kahn disagreed with the consensus opinion.

"In few cases is anything stated without reference to expert opinion, and where that is done we say so," Frances said. "I think we've built in enough protections. But the bigger problem it raises is that the field itself may be unduly influenced. It is possible that our clinical researchers themselves have been influenced."

In an effort to neutralize such bias, he and colleagues sought opinions from "several hundred" nonacademic clinical experts in practice who do not publish papers or have a need to seek research funding that could involve drug company money, said Frances.

The guidelines were developed over a two-year period and were funded entirely by unrestricted educational grants from Abbott Laboratories and Janssen Pharmaceutica.

(Psychiatric News, January 17, 1997)