Lost somewhere in the frantic growth of topics and competencies expected of psychiatry residents has been an appreciation for the challenges of the most elementary of clinical skills: the prescription of psychotropic medication. The great misconception here is that this area is straight-forward, accessible, evidence-bound, and simple to master. A plethora of algorithms, with their reduction of clinical decisions to a few either-or boxes, implies the obvious: that with a small set of clear-cut rules one can master the field of psychopharmacology.

Teachers of pharmacotherapy know better than anyone that this is not so. Faced with the challenge of equipping the novice resident with the imprimatur of competency, the breadth of knowledge, depth of understanding, and range of skills required suddenly seem less trivial. In this setting, the crucial questions facing the whole of the profession come to the fore. What we teach is what we are—and certainly what we will become—as a discipline. Consequently, the issues that we debate as a field must be faced head-on when we teach.

Here we find a series of questions that force us to examine what we know, how we know it, what we do, and why we do it. These did not begin as questions in the classroom but in the clinic and the answers are not always clear. Our students may be surprised to discover that when we ask these questions we are not so much engaging them in the Socratic method as inviting them to join us at the round table.

This issue of Academic Psychiatry includes a series of articles that address some of these questions. They are as critical to the field as they are to the classroom. Their presentation in the context of education provides a unique and informative venue for their discussion, and they highlight the dilemmas facing the teacher of psychopharmacology.

A host of questions lie at the heart of the debate over evidence-based medicine. How much of what we do is informed by data? How much should be? What information constitutes evidence? How good are the published data that we have? How should published data be applied in the clinic? For the teacher, they pose parallel questions in the classroom. How much information from papers and textbooks must be mastered to proceed? How much of that information is hard evidence, and how much is expert opinion? How can published data be evaluated by a clinician? How much weight should be given to clinical experience?

Surely, if evidence has any role in medicine, it is in education. Here, at the outset of clinical experience, the trainee is open to consider the evidence as it stands, rather than through the lens of inevitably skewed clinical experience. Even here, however, the tools the trainee needs are considerable. The body of evidence to be mastered is large and growing rapidly. The quality of information is highly variable and subject to a variety of biases and influences. Trainees must learn to assess and filter this information. Habits of good scholarship and continuing education are as important to develop as is the mastery of the data currently available.

For the teacher, who struggles not only to know what to teach but how to teach it, the question of evidence in education is doubly important. One useful role for treatment algorithms is the instruction of trainees. The collective guidance of data and expert consensus provides an excellent template for the novice to begin clinical decision making. With time and experience, clinical judgment develops, and the trainee begins to venture with increasing confidence from the prefabricated treatment course. This would seem to be a safe and effective way to provide care in an educational setting.

But if intuitive skill is also valued, how can it be developed? If evidence is only derived from large, controlled databases, should the trainee be expected to acquire an extensive body of clinical experience? If so, to what purpose? An alternative view of training is to set aside algorithms altogether in favor of expert mentorship. By making clinical decisions independent of rigid rules, the novice will develop the essential skills of clinical care.

David Osser et al. (1) present a case for evidence-supported guidelines and algorithms as the core material for education. Residents with this curriculum have the opportunity to compare clinical practice with algorithmic recommendations and evaluate the validity of the guidelines themselves. With faculty guidance, they master the information in the algorithm and develop the skills necessary to assess the clinical data that inform it. Herein lies a balance between practice informed by evidence and an awareness of the limitations of that evidence.

The counterargument is offered by Carl Salzman (2). Here, the weaknesses of algorithms and the limitations of the evidence on which they are based are identified as the appropriate starting points for the development of clinical expertise. Research studies average the experience of many patients; clinical practice invariably focuses on the individual. Clinical trials seek uniform diagnoses and minimize complicating factors; real patients fit categories imperfectly and inevitably carry comorbidities. Blinded, randomized, multisite studies are impersonal by design; clinical pharmacology involves critical interpersonal elements. Competence lies in the practitioner’s capacity to work beyond the guidelines, which cover only the most self-evident of decisions.

Has the trust of science been corrupted by the influence of commercial research and pharmaceutical marketing? This question is debated throughout medicine with vigor and passion. Education is not exempt and, in some cases, lies at the center of the controversy. Marketing is a legitimate and essential function of a pharmaceutical industry based on capitalist principles and financial incentives, but when marketing masquerades as education, it undermines the objectivity and judgment of academia. Residents need to be taught how to interact with industry, even at a time when we struggle as a profession to determine what the appropriate relationship should be. In this regard, there is a place for mentorship, formal curricula, active discussion, and familiarity with professional guidelines.

Educators also need to be taught how to interact with industry, to preserve the integrity of medical education, while benefiting from the information base and financial largesse it controls. Whether this is possible and, if so, how it can be done remain open questions. Involvement of faculty in formal policy development, creation and dissemination of specialty guidelines, and ongoing discussion of the pertinent issues are essential.

Two perspectives on this topic stand in striking contrast. Amy Brodkey (3) argues that the pervasive influence of commercial marketing undermines medical education and harms trainees and the patients they serve. The appropriate response to this situation is rigorous control of education by academia and a firewall between the medical profession and the pharmaceutical industry. This is not where things stand now, and to get from here to there is a daunting but not impossible task.

A more conciliatory and practical approach is offered by Paul Mohl (4). A clear understanding of the functions and mores of the medical profession and the pharmaceutical industry permits constructive engagement without ethical compromise; problematic interactions can be identified and avoided. This is the approach currently used by most training programs, although few do it as systematically and thoughtfully as presented in this article.

A related and no less controversial topic is who should be teaching psychopharmacology. Is this the job of the researcher with the greatest insight into the workings of the drugs, the clinician with the greatest experience in their use, or the educator with the greatest skill in conveying knowledge? What is the role of the paid consultant to industry? Is the balance between access to information and potential bias adequately maintained? Steven Dubovsky (5) reviews this topic and the implications of the involvement of each participant.

What is the appropriate relationship of pharmacotherapy and psychotherapy? Are they compatible? Are they separable? The robust placebo response evident in most clinical trials reminds us that drugs involve more than pharmacology; complex psychological factors are also in play. Drugs and the prescription of drugs have meaning for patients, physicians, and the therapeutic relationship they share. Whether they are viewed as a total solution, a useful but limited tool, a corrupting crutch, or a personal gift depends in large measure on the quality of the relationship between the prescriber and patient and the physician’s skill at understanding and managing that relationship. Much of the current trend toward evidence-based medicine detracts from this more subjective aspect of pharmacotherapy. The challenge for medical educators is to return this topic to a curriculum dominated by large controlled studies designed to minimize these very factors.

Equally complex is the relationship between psychotherapy and medication managed by different practitioners. Whether this treatment is collaborative, competitive, or collinear depends in large measure on the expectations of the professions. Competitive treatment arises from narrowly defined perspectives of therapy; one must be better than another and the selection of one devalues the other. Collinear treatment suggests that psychotherapy and pharmacology are fundamentally separate, work parallel to one another, and need not influence each other. Collaborative treatment seeks to support both processes. Therapeutic insight is enhanced by the relief of overwhelming symptomatic distress; medication compliance is improved by insight into the meaning and value of treatment.

In this context, David Mintz (6) addresses the challenges residents face in learning the prescriber role, particularly as they confront the reality of its complex psychological implications at a time when they are struggling with their own professional identity. James Ellison (7) complements this with a focus on practical suggestions for teaching treatment collaboration and assessment of residents’ skill in the area. Peter Weiden and Nyapati Rao (8) contribute a related paper on medication compliance, touching on relationship issues and expectations patients and physicians have of medications.

Are psychiatrists primary physicians or secondary care specialists? Psychotropic drugs are widely prescribed in primary care settings, a trend that will likely continue as awareness of the public health implications of psychiatric disorders grows. But if family physicians appropriately prescribe fluoxetine and risperidone, what will be the role of the psychiatrist? Instead of acting as the primary treater for a small group of fortunate patients, the psychiatrist will more often be the specialist to whom treatment refractory and side effect addled patients are sent after the primary care physician tries several first-line treatments or exhausts the dominant algorithm. Current trainees see few of the older medications; monoamine oxidase inhibitors, tricyclic antidepressants, and conventional neuroleptics are textbook curiosities for most. Yet, these may well be medications they are called upon to dispense with experience and expertise. James Jefferson (9) addresses this issue with specific suggestions for both what and how to teach, noting important lessons for the profession from available evidence on the appropriate role of these older medications.

Neuroscience is the foundation of psychopharmacology, yet much of what we know about the pathophysiology of mental illness was discovered from study of the action of medications; rarely have we developed new drugs from our knowledge of neural dysfunction. The flow of information would seem to be in the wrong direction. This raises the important question of how much neuroscience should be taught for clinical practice. Knowledge of the mechanism of action of psychotropic drugs has numerous applications in clinical practice, including prediction of clinical response, avoidance of drug-drug interactions, rational polypharmacy, and others. We fall short, however, of the real goal of understanding the pathophysiology of psychiatric disorders. How far into the black box of brain function do clinicians really need to peer?

The answer would seem to be that residents need enough neuroscience to understand the mechanism of action of the drugs they use, to appreciate the biological bases for the illnesses they treat, and to understand the emerging literature in the field. In an educational climate dominated by competencies and their assessment, they must demonstrate a substantial fund of neuroscience knowledge and skill in the practical application of that knowledge. Significant portions of the American Board of Psychiatry and Neurology certification examinations are directed to these topics, as discussed by Dorthea Juul et al. (10) To this end, Ira Glick and Sidney Zisook (11) present their experiences in the development and promotion of the American Society of Clinical Psychopharmacology model curriculum. Their observations on the rationale for the curriculum and the obstacles to its implementation are both thoughtful and provocative, as are their recommendations to educators and the field as a whole.

Is psychopharmacology competency a knowledge or a skill? How effectively can it be taught in the classroom? How much should be done in the clinic? Are other teaching modalities better suited to the topic? Sidney Zisook et al. (11) consider this issue through an evaluation of a range of venues for teaching, some conventional and some unique.

Finally, lest we lose sight of the object of this exercise, Anna Georgiopoulos and Jeff Huffman (13) share a resident perspective on the process. Their advice to faculty is heartening: make it accessible, make it practical, and stay close by. That is not a bad place to start.