On Making Science and Life Indistinguishable By Herbert S. Sacks, M.D. APA President

A lively debate focused upon the uneasy relationship between evidence-based science and clinical consensus has been precipitated by the sixth draft of the APA Practice Guideline for the Treatment of Patients With Panic Disorder. The arguments mirror the discourse in all of medicine underscoring further the dissolution of boundaries during the past decade between psychiatry and the other medical specialties.

Evidence-based medicine involves integrating current best evidence with clinical expertise, pathophysiological knowledge, and patient preferences in making decisions about their care. Culturally sensitive and competent care, together with shared decision making, is part of the calculus. The relevant skills include defining a patient problem, proficiently searching and critically appraising relevant information from the literature, and deciding whether and how to use this information in practice. Dr. David Sackett, a founder of this approach and director of the Center for Evidence-Based Medicine at Oxford, has elaborated the concept further in an illuminating article in the British Medical Journal (1996; 312: 71-72). The debate has been popularly distilled into trying whenever possible to base medical decisions on sound research data derived from randomized controlled trials.

Sackett points to a million randomized controlled trials that have been reported in the past 30 years. A doctor with a computer and modem can find relevant data within a few minutes. Proponents believe this strategy will eliminate dangerous and obsolete practices that many of us have been teaching for many years.

Objections to evidence-based medicine have questioned the statistical and ethical validity of making patient decisions on the basis of what happens to large selected groups.

Outcomes analysis, like evidence-based medicine, also uses the experiences of large patient groups to guide patient care, but it approaches the groups from a different direction—looking backward through the medical records of the treatment of thousands of patients. Hospital and insurance provider databases find groups of similar patients; judgments are then made about best outcomes, and the best treatment recommendations are made for similar patients.

Randomized control trials almost always exclude patients who are too old, too sick, or too confused to participate, but outcomes research includes everyone. This point has been made by geriatric psychiatrists and geriatric medicine specialists.

The practice of evidence-based medicine is dependent upon probabilistic thinking, which some observers claim ignores the complexity of a single patient. Dr. James Goodwin, a geriatrician, wrote in a JAMA editorial in November that "everything that makes an individual an individual, everything that importantly defines an individual’s life, is outside the realm of science. The practice of medicine involves only individuals." The complexity of a single patient’s problems invokes the art of medicine, but Dr. Sackett’s fiery rebuttal is "Art kills. Art gave us purging, puking, leeching, and gastric freezes. There’s a science to the art of medicine."

Evidence-Based Clinical Guidelines

The writing of evidence-based clinical practice guidelines reduces a large body of knowledge into a clinically usable and readily convenient format. This burdensome task combs and assesses the evidence, makes the best judgments about the relative importance of health and economic outcomes and then makes recommendations. The latter leaves space for creative approaches by skilled, experienced clinicians and does not advocate for cookbook treatment. Compliance with guidelines is somewhat problematic, since the clinician may be only marginally aware of the guidelines’ significance and the latter may not fit the subtlety of the patient’s needs. In looking at internists’ experiences, Tunis in 1994 in the Annals of Internal Medicine emphasized that clinicians may not be confident of the recommendations because of controversies extant and differing views held by respected scholars.

APA Practice Guideline for Patients With Panic Disorder

By mid-December, the sixth draft of the ninth guideline was presented to the APA Board of Trustees after review by the Assembly in November. Revisions have benefited from comments by hundreds of reviewers. Of special importance, extensive and detailed responses were sought from senior professors and outstanding clinicians who have contributed widely to the literature and to the teaching of psychodynamic psychotherapy to medical students and psychiatric residents. Among them were Board members Drew Clemens, Michele Riba, and Marcia Goin, in addition to Sheila Hafter Gray, Barbara Rosenfeld, and Sherwin Woods. The APA Commission on Psychotherapy is chaired by Dr. Clemens. The guideline section on psychiatric management was recast, providing richer descriptions of psychotherapy’s role. Cognitive-behavioral therapy, developed by psychiatrists, is now more appropriately balanced by the discussion of psychodynamic psychotherapy.

The most troublesome criticism claimed that only evidence-based research findings were cited in the guideline and clinical consensus (or clinical wisdom) has been ignored. From the outset of the guidelines project, APA’s position has been that evidence utilized will be of two forms: clinical research and clinical consensus based upon decades of experience.

An organized group of critics raised serious questions about the scientific perspective of the guidelines project, seeking a more inclusive biopsychosocial view of mental disorders. One commentator feared that not adhering to the guidelines might expose him to litigation risk. While the challenge of the critics merited a considered response, the details of the criticism reflected factual inaccuracies and a lack of understanding of distinguished colleagues about the nature of guideline reviews, which allow for substantive revision after reasoned discussion. Most importantly, the multilayered APA process is designed to encourage dissonant voices to be raised. Guidelines must maintain their integrity, reflecting available science as well as rich clinical experience. The meeting with the critics, in a climate touched by comity, led to my earnest invitation to them to contribute actively to our work at every level in our component structure. Unless all of our concerned constituencies are players at the table of psychiatry, misreadings, misprisions, and misapprehensions are inevitable.

Science, Art, and Life

The stirring debate occasioned by the panic disorder guideline has served us well for the present and for the millennium. Freud, in "Analysis Terminable and Interminable," in looking at treatment expectations, cited government and education as impossible professions. He also might have mentioned parenting as an impossible life task. He inferred that while we seek the highest level of performance, our striving cannot yield to perfection, given the inherent nature of the tasks and the human condition. But strive we must. We would add another impossible task: attempting to bring together, indistinguishably, emergent science, the art of medicine, and an imperfect grasp of the complexity of a single patient’s life. Despite what may be seen as impossible, we are determined to push beyond the present limits of our understanding.

APA recognizes with pride the splendid leadership of Dr. Jack McIntyre in gathering together a cohort of expert colleagues who often reflect disparate views in order to advance the frontiers of our knowledge. Backed by our superb Office of Research, the work products to date have lent luster to this massive effort.