A physician assistant in our hospital's primary care clinic asked me to provide a consultation for a young woman who had been discharged from the community mental health center's crisis unit. Her discharge form listed a diagnosis of schizoaffective disorder, admission and discharge dates spanning two weeks, and the treating psychiatrist's directive to follow up with her primary care physician for medication management. Attached to the form was a mind-stunning list of discharge medications—nothing short of a psychoactive granola, composed of three antipsychotics, three mood stabilizers, two antidepressants, and an antianxiety agent.

The physician assistant had been so overwhelmed by this pharmacologic complexity that he did not provide refill prescriptions, except for one antipsychotic. By the time I met with the patient it was unclear what were the symptoms of her mental illness, what were the lingering iatrogenic effects of polypharmacy, and what was the consequence of abrupt discontinuation. Her suffering was heartbreaking. It was also avoidable.

As community psychiatrists we live by ethical commitments, anchored in our personal and professional identities, to do the right thing by patients who come under our care. Unfortunately, organizational budgetary priorities sometimes shape our allocation decisions. As personally confounding as such decisions may be, I trust that most of us have been guided by the ethical commitment to benefit our patients above all else. In the rare situations in which we cannot lay the groundwork for beneficial outcomes, we have an even greater duty to try to minimize or prevent potential harm.

Community psychiatry has struggled for years to treat the most seriously ill individuals within the constraints of severely limited resources. However, the practice of referring the most vulnerable and seriously ill patients in our community mental health systems to agencies and providers whom we know are ill-trained to treat them can hardly be justified morally. Clearly, the fundamental ethical brake to this practice ought to be based on the realization that we cannot do something that we know—or that we should know—could harm those we have a covenant to serve. The choice not to participate in the cost-shifting practice of "turfing" uninsured—but clinically complex and seriously ill—individuals onto primary care providers should be an ethical no-brainer. We ought to know better and act better. And our patients have a right to expect nothing less of us.