From July 1995 to June 1998 I had the privilege of meeting weekly with eight residents in the Harvard Longwood psychiatry residency training program as coleader of a three-year seminar sequence— "Patient and Psychiatrist." The seminar focused on developing professional identity as a psychiatrist and addressed ethics, the patient-psychiatrist relationship, the vicissitudes of residency, and career planning. Although the seminar included some readings, the sessions generally built on the residents' experiences.

The seminar was not about managed care, but managed care for clinicians today is like water for fish—for better or worse, it is the medium we swim in. As a result, the seminar provided an extraordinarily rich view of managed care as experienced by eight fine residents as they progressed from the beginning of their second year of residency in July 1995 to their graduation three years later.

In the still-unresolved debate about whether caring for patients within resource limits is an ethical imperative or an ethical abomination (1), I have consistently argued that ethical managed care is possible, necessary, and preferable to an "unmanaged" health care system (2,3). Pronouncements about ethics, however, do not deserve respect until they are subjected to the test of experience. The two best experiential tests of a position about clinical ethics are whether we can look patients and families in the eye and honestly discuss our treatment recommendations and our reasons for making them (4) and whether our students' challenges about why we do what we do elicit pride or shame. The seminar provided the opportunity to respond to our students' challenges.

From review of the notes I made after each seminar and discussion of our three years together, I have concluded that these students taught their teachers four key lessons about managing care ethically. The lessons centered on threats to coordination of care and empathy, confusing messages about the role of the psychiatrist, the importance of a new form of "differential diagnosis," and a changing psychological contract for young professionals.

As hard as my own PG2-PG4 residency cohort worked 30 years earlier, from 1965 to 1968, the intensity of practice experienced by today's residents is vastly more demanding. The residents in the seminar were deeply distressed during their hospital rotations when patients were discharged with what appeared to be substandard treatment plans, such as referrals to an outpatient clinic for follow-up without the residents' having spoken with a designated outpatient clinician before discharge.

It is not technically difficult to shorten the length of inpatient hospital care. However, as was the case with deinstitutionalization, many patients will suffer if appropriate follow-up care is not provided. The residents were legitimately confused about whether the national movement to short hospital stays represents an inherently bad idea or a potentially good idea sometimes executed poorly. As a clinician who has practiced in responsible managed care settings for 28 years—five years in community mental health and 23 in a prepaid group practice—I could tell them that the incidents of inadequate referral represented "mismanaged care," not managed care per se. However, their experience made clear that even in a clinical environment that is excellent as measured by national standards, the managed care revolution must provide much better continuity of care. A relay team does not succeed simply by running fast—it must also pass the baton smoothly. The residents help us see that managed care has certainly learned how to run fast, but it has not managed the art of baton passing with comparable zeal.

Efficiency can entail other costs. One resident was practicing in a partial hospital unit that allowed more extended care than inpatient units generally do. She noted that only when she had time to allow a severely ill patient to explain to her how a pot roast should be cooked was she able to identify previously hidden strengths, which proved important for long-term treatment planning. Although it might be hard to argue that discussing pot roast was medically necessary, in fact, it was!

The residents received very confusing signals about the future role of psychiatrists from their discussions with attending physicians, reading, and travels to investigate postresidency positions. Their worst fear was that psychologists would develop stronger psychotherapeutic skills and that other medical specialists would develop stronger medical and neurological skills, leaving psychiatrists in a kind of limbo. The confused signals were especially prominent in the experience of a fourth-year resident who was interested in cognitive-behavioral as well as pharmacotherapeutic treatment of anxiety disorders. At one program he was asked, "How do you expect to find such a position when managed care won't pay psychiatrists to do that kind of treatment?" Two months later, another program, valuing his cognitive-behavioral skills and interests, hired him to run its anxiety disorders program.

During the course of their residency, all members of the seminar appeared to consolidate strong identities as clinicians who would take comprehensive care of patients and not restrict themselves by default to a narrow "medical" slice of the care process. However, the national scene did not make it easy for them to achieve this robust clinical identity. We can hope that a recent report suggesting that care of depressed patients is less costly when psychiatrists provide both psychotherapy and pharmacotherapy than when those roles are split between medical and nonmedical clinicians (5) reflects the beginning of a correction in what will probably appear in retrospect to be five to ten years of short-sighted constriction of the psychiatric role.

Fifteen years ago the late Donald Schön (6) developed a powerful model of professionalism, which he conceptualized as the capacity to reflect within the scene of action in a way that allows for innovation and creativity as well as application of specialized knowledge. On reviewing the seminar sequence, it was clear that throughout the three years, we worked recurrently with the "differential diagnosis" of problematic clinical situations in managed care in the way that Schön described. We continually asked to what extent a patient's problem required a better diagnosis and treatment plan (clinical quality improvement), enhanced skills on the part of the resident (personal learning), activism and advocacy about suboptimal system functioning (system-level quality improvement), or some other approach.

One resident commented on how helpful the group dynamic was in addressing this differential diagnosis in circumstances of ambiguity and pressure. Her comment was, "I get too upset when I think about these things on my own." Over three years, the residents appeared to develop considerable skill in this domain of professional action. Clinicians who were established in practice before the managed care revolution have rarely had an analogous opportunity to reflect on the turbulent environment and arrive at a thoughtful differential diagnosis and action plan in a context providing consultation and group support for learning. The absence of such an opportunity has had serious consequences for professional morale and the quality of professional efforts to influence the course of change (1).

The residents' readiness to advocate for change when they encountered system-level problems, their recognition of the national need to find a way of improving system performance and controlling costs, and the striking absence among them of the bitter sense of betrayal so pervasive among established professionals suggest something about their expectations and "psychological contract" with the profession. When I asked about this issue, a resident commented, "We all started medical school knowing how much uncertainty and change was ahead of us." Unlike the disillusioned graduates described so poignantly by Gabbard (7) six years ago, this group of 1998 residency graduates is not experiencing a sense of betrayal of their implicit contract with society and the profession.

The opportunity to explore managed care through the reactions and reflections of a talented group of residents as they moved from anxious PG2 "beginners" to impressively confident and competent graduates three years later gives me hope about the concept of ethical managed care, for which I have sometimes been a rather lonely advocate. These students saw the need for a society even as wealthy as ours to set some limits on health care expenditures, especially given our grotesque national failure to provide health insurance for all. They were especially critical of managed care formats that failed to provide the "seamless" continuity of care that mission statements so regularly invoke. They challenged overly stringent definitions of medical necessity that impeded their ability to align with their patients' strengths by getting to know them in enough depth.

Although the residents' ability to develop the skills of reflective practitioners was encouraging for their future as clinicians and the future of the field, acquisition of these skills required recurrent study and practice in the supportive learning environment of a residency program carefully designed to prepare residents for the future (8,9). It is worrisome that these ingredients are almost wholly lacking for the typical mid-career clinician struggling to deal with profound system change. It is reasonable to expect that students finishing their professional training in 1998 and in the future will find ways of developing and applying skills that will let them pursue our traditional commitment to patient welfare and be advocates and activists when needed. However, we will almost certainly continue to see substantial career casualties among mid-career clinicians who have been asked to adapt to profound change with little help in making this adaptation. From the perspective of the 21st century, the decade of the 1990s is likely to look like an extended annus horribilis for mid-career professionals in their encounters with managed care.

The author thanks Hudie B. Siegel, M.D., coleader of the seminar, and Stella Bae, M.D., Jean F. Cochrane, M.D., Liza Donlon, M.D., James A. Evans, M.D., Scott A. Freeman, M.D., John H. Halpern, M.D., Rebecca Lundquist, M.D., and Eric D. Teitel, M.D., for their devoted participation in our three-year venture in shared learning.

Dr. Sabin, who is editor of this column, is associate clinical professor of psychiatry at Harvard Medical School and codirector of the Center for Ethics in Managed Care at Harvard Pilgrim Health Care and Harvard Medical School. Send correspondence to him at Harvard Pilgrim Health Care, Teaching Programs, 126 Brookline Avenue, Suite 200, Boston Massachusetts 02215 (e-mail, jim_sabin@hphc.org).