The Psychiatric Patient’s Right to Effective Treatment: Implications of Osheroff v. Chestnut Lodge

Resolution of both the clinical and scientific issues is made difficult by the divisions within psychiatry in the United States, where psychiatry is divided theoretically and clinically into different schools—biological, psychoanalytic, and behavioral (7). This aspect of the sociology of psychiatry and other mental health professions and its effect on training and practice have been documented for a number of years (8–11). Various terms have been used to describe these divisions and splits—schools, movements, ideologies, and paradigms, for example (10, 12, 13). Whatever term is used, there is agreement that the differences in theory and practice involve controversies over the nature of mental illness, the appropriateness of different forms of treatment, and the nature of the evidence for the safety and efficacy of such treatments.

Chestnut Lodge has played an important role in the modern history of psychiatry in the United States. For more than 40 years, Chestnut Lodge has been one of the major centers of theory and clinical practice in intensive individual psychotherapy based on psychoanalytic and interpersonal paradigms (14). Harry Stack Sullivan (15), who formulated the interpersonal theory of psychiatry, was a consultant to the institution. Many of his lectures and seminars at Chestnut Lodge have been published posthumously. Frieda Fromm-Reichmann was also on the staff at the same time. She had immigrated to the United States from Germany along with a large number of other leading psychoanalysts driven out of Europe by the Nazi regime. Fromm-Reichmann wrote a number of influential papers and books about the psychotherapeutic treatment of schizophrenia and manic-depressive illness (16, 17).

Several prominent U.S. psychiatrists were trained at Chestnut Lodge; many subsequently became leaders in clinical psychiatry. Alfred Stanton, who became psychiatrist-in-chief at McLean Hospital in Massachusetts, and Otto Will, who became medical director of the Austin Riggs Center in Massachusetts, are two notable examples. The writings of Sullivan (15), Fromm-Reichmann (16), Will (18), and others were influential in many psychiatric residency training programs from 1950 through the 1970s.

In the 1950s and 1960s, new psychopharmacological agents and the findings of neuroscientific research began to influence psychiatric teaching, practice, and research. New forms of psychotherapy based on approaches other than psychoanalytic were applied. Professional controversies increased, particularly over the comparison of the therapeutic efficacy of the different forms of psychotherapy (psychoanalytic, behavioral, family, group) and over the relative efficacy and safety of the psychotherapies, used either alone or in combination with psychopharmacological agents (19).

At both Chestnut Lodge and Silver Hill Foundation there was agreement that Dr. Osheroff suffered from a severe depressive condition. There was disagreement, however, as to the diagnosis of narcissistic personality disorder. In a discussion of this case, Dr. Stone (3) described a “dispute” over the appropriate diagnosis: “The patient’s psychiatric experts, in depositions that reflected their biological orientation, diagnosed him as having an obvious case of biological depression, emphasizing his vegetative disturbances. The private psychiatric hospital contended that the patient was properly diagnosed as having a narcissistic personality disorder.”

It is to be noted that Dr. Osheroff’s diagnoses at both Chestnut Lodge and Silver Hill Foundation were made in 1979 in accordance with DSM-II, APA’s official nomenclature at the time of his hospitalization. DSM-III, which is the current diagnostic nomenclature for clinical psychiatric practice in the United States, did not come into use until 1980. DSM-II does not include a diagnostic category of narcissistic personality disorder, although that diagnostic category is included in DSM-I and in DSM-III.

DSM-II includes diagnostic categories of psychotic depressive reaction and manic-depressive illness, depressed type. Both refer to severe forms of depression. There is no evidence of clinical features of hypomania or mania in Dr. Osheroff's history or in the case records from either institution. The patient would not meet DSM-III criteria for bipolar disorder or DSM-II criteria for manic-depressive illness, manic or circular types.

The DSM-II diagnostic category of psychotic depressive reaction was replaced in DSM-III by major depressive episode with melancholia and/or major depressive episode with psychotic features. Melancholia is a term from the past denoting a particularly severe form of depression uniquely responsive to somatic drugs and/or ECT therapies. It is of note that the term “biological depression” does not appear in DSM-II, DSM-III, or ICD-9.

According to Chestnut Lodge records, there were differences in medical opinion as to the relative importance to be given to the patient’s personality conflicts and his depressive diagnosis as they influenced treatment decisions, not over the depressive diagnosis itself. As was the practice at that institution, the patient had two physicians, a psychiatrist-administrator and a psychotherapist (20). The hospital records suggest there may have been disagreement between these two physicians: the psychotherapist emphasized the need to treat the patient's personality problems as the major condition, and the administrator expressed concern over the continued severity of the patient’s depressive symptoms and distressed behavior.

This aspect of the clinical process illustrates the tendency for many psychoanalytically oriented psychotherapists, both in institutional and in community practice, to focus treatment on a patient’s personality conflict and character pathology rather than on symptoms. In DSM-III terms, there tends to be an emphasis on the axis II diagnosis and relatively less attention given to the axis I diagnosis. The axis I diagnosis, a severe depression in the case of Dr. Osheroff, is often missed, or, even if it is formulated, the personality disorder is chosen as the major target for treatment planning.

An important clinical consideration at issue in Osheroff is whether the patient suffered from a personality disorder as well as from depression and whether the presence of the narcissistic personality disorder militated against the use of medication for the depression. Long-term psychoanalytically oriented psychotherapy is often justified by the theory that some states of clinical depression derive from unresolved personality conflicts whose origins lie in developmental problems related to childhood intrafamilial psychopathology (17, 21). This theory of etiology and pathogenesis of depression is the subject of scientific research and professional discussion (22). Expert witnesses testified on this issue at the Osheroff hearings.

It should be noted that the psychiatric experts who testified in this case did not agree on the validity of the diagnosis of narcissistic personality disorder for the patient. One expert, a trained psychoanalyst who is currently responsible for Dr. Osheroff's treatment and who had treated him when the patient was 29 years old and at the time of his divorce (when he was 34 years old), did not accept the diagnosis of narcissistic personality disorder and testified to this effect at the court hearing. He noted the patient’s successful life achievements before the onset of the illness episode that led to hospitalization at Chestnut Lodge, including his professional success as a nephrologist, his ability to sustain a high income, and his loving, empathic, and sensitive relationship with his children.

The admitting psychiatrist at Silver Hill Foundation did not make the diagnosis of any personality disorder. An expert witness called by Chestnut Lodge to testify at the court hearing also did not think that the patient had a narcissistic personality disorder. In contrast to the near unanimity of expert opinion as to the patient's severe depressive condition, disagreement existed as to whether the patient met any criteria for narcissistic personality disorder.

With regard to all kinds of therapeutics—pharmacotherapy, surgery, radiation, psychotherapy—the most scientifically valid evidence as to the safety and efficacy of a treatment comes from randomized controlled trials when these are available. Although there may be other methods of generating evidence, such as naturalistic and follow-up studies, the most convincing evidence comes from randomized controlled trials.

There have been many controlled clinical trials of psychiatric treatments; most have been conducted to evaluate psychopharmacological agents. These trials were initiated in the 1950s and 1960s in response to the controversy that followed the introduction of chlorpromazine, reserpine, and the other “tranquilizers.” The application of controlled trials in psychopharmacology expanded after the passage in 1962 of the Kefauver-Harris Amendments to the Food, Drug, and Cosmetic Act, which mandated evidence of efficacy before a pharmaceutical compound could be approved by the Food and Drug Administration and marketed.

Research on the efficacy of psychotherapy has lagged behind that of psychopharmacology but has, nevertheless, been extensive. Smith et al. (23) analyzed more than 400 reports of psychotherapy research. Specific reviews of the evidence have appeared with regard to psychotherapy of neurosis (24), schizophrenia (25), depression (26), and obsessive-compulsive disorders (27).

In view of these developments, a review of the state of evidence regarding the treatments of the two psychiatric conditions diagnosed for Dr. Osheroff at the time of his hospitalization is in order.

With regard to the treatment of the patient's diagnosis of narcissistic personality disorder, there were no reports of controlled trials of any pharmacological or psychotherapeutic treatment for this condition at the time of his hospitalization (28). The doctors at Chestnut Lodge decided to treat Dr. Osheroff’s personality disorder with intensive individual psychotherapy based on psychodynamic theory.

With regard to the treatment of the patient’s DSM-II diagnosis of psychotic depressive reaction, there was very good evidence at the time of his hospitalization for the efficacy of two biological treatments—ECT and the combination of phenothiazines and tricyclic antidepressants. The combination pharmacotherapy was the treatment later prescribed at Silver Hill Foundation.

There are no reports of controlled trials supporting the claims for efficacy of psychoanalytically oriented intensive individual psychotherapy of the type advocated and practiced at Chestnut Lodge and administered to Dr. Osheroff. The closest approximation to a controlled clinical trial of this form of intensive individual psychotherapy has been reported with hospitalized schizophrenic patients at two institutions in the Boston area (30). Contrary to the expectations of the investigators, one of whom was Dr. Alfred Stanton (who had held a senior position at Chestnut Lodge and was one of the authors of The Mental Hospital [20], which describes the Chestnut Lodge institution), the results indicated that intensive individual psychotherapy offered no advantage over standard treatment (hospitalization, medication, and supportive psychotherapy) for these patients.

McGlashan and Dingman (30, 31) have reported results from follow-up studies of groups of patients treated at Chestnut Lodge. The findings from this naturalistic study do not support the efficacy of long-term psychotherapy and hospitalization for severely depressed patients such as Dr. Osheroff.

It should not be concluded there is no evidence for the value of any psychotherapy in the treatment of depressive states. Depressive states are heterogeneous, and there are many forms of psychotherapy. There is very good evidence from controlled clinical trials for the value of a number of brief psychotherapies for nonpsychotic and nonbipolar forms of depression in ambulatory patients (26). The psychotherapies for which there is evidence include cognitive-behavioral therapy (32), interpersonal psychotherapy (14), and behavioral therapy (33). However, no clinical trials have been reported that support the claims for efficacy of psychoanalysis or intensive individual psychotherapy based on psychoanalytic theory for any form of depression.

An important clinical issue raised by Osheroff has to do with the possible influence of a patient’s diagnosis of personality disorder on the decision to use medication and on the expected response to medication of depressed patients treated either with medication alone or with medication in conjunction with psychotherapy.

Even if we assume that the personality disorder was correctly diagnosed in Dr. Osheroff's case, there is no evidence to support the premise that the presence of a narcissistic personality disorder militates against the use of antidepressant medication. Patients with a personality disorder in addition to depressive illness may be relatively less responsive to medication than those without an associated personality disorder (34). However, the presence of a personality disorder by itself does not contraindicate the prescription of appropriate medication or predict complete failure to respond.

A related therapeutic issue raised by the case has to do with the possible negative interactions between psychotherapy and pharmacotherapy for depression. Many psychoanalytically oriented psychotherapists have argued against the use of medication in patients receiving psychotherapy because of the possible adverse effects of the pharmacotherapy on the conduct of the psychotherapy (35), although there is evidence that the combination of drugs and psychotherapy does not interfere with the psychotherapy of depression (36). Moreover, findings from controlled trials suggest that the combination of drugs and psychotherapy may have beneficial additive effects in the treatment of depression (37).

Given this state of evidence, it is difficult to justify the rationale used by the Chestnut Lodge staff in forming their treatment plan and in making specific decisions. On the one hand, there was a body of scientific evidence from controlled trials attesting to the value of medication and/or ECT for the type of severe depression that the institution diagnosed this patient as having. On the other hand, there was no scientific evidence for the value of psychodynamically oriented intensive individual psychotherapy for either the patient’s depressive condition or his diagnosis of personality disorder. Nevertheless, the patient was treated only with intensive psychotherapy.

It might have been reasonable to have undertaken a period of psychotherapy, particularly in view of the tendency of many depressive states to remit spontaneously. However, several clinical studies (38, 39) have concluded that, in the absence of intervention with somatic treatments, severe health impairment and greater mortality are associated with deep depressions.

The hospital continued its treatment plan for many months in the face of continued worsening of the patient's clinical state. Meanwhile, the prolonged hospitalization was having adverse effects on the patient's medical practice, financial resources, and marital and family relations.

There is a federal agency, the Food and Drug Administration, that has statutory authority to review the evidence for the efficacy and safety of pharmacological treatments. Because of the Kefauver-Harris Amendments, a pharmaceutical firm that makes promotional claims for the efficacy of a drug is expected to present evidence from controlled trials in support of its assertions.

Consider, however, the situation with regard to psychotherapy. There are no statutory constraints on claims made for psychotherapy. No government body is authorized to review the evidence for psychotherapy or comment on its status. In the late 1970s, the Senate considered the creation of a National Commission on Mental Health Treatments, but the proposal was opposed by the mental health professions and was not enacted into law (40).

The National Institutes of Health (NIH) conduct consensus development conferences to review the evidence about specific procedures relevant to health and medicine, including the efficacy of treatments. An NIH consensus development conference was held on longterm drug treatments of affective disorders in 1984 (41), and a conference on electroconvulsive therapies was held in June 1985. However, the efficacy of psychotherapies has not been addressed by NIH.

It might be expected that two other federal government agencies concerned with health financing and disability—the Health Care Financing Administration and the Social Security Administration—would be involved in judgments as to the appropriateness of treatment, inasmuch as they are involved in the disbursement of large amounts of funds. The Health Care Financing Administration provides reimbursement under both Medicare and Medicaid, and the Social Security Administration determines the disability status of individuals with psychiatric illness. However, only limited efforts have been undertaken by these agencies to establish criteria for the safety and efficacy of treatments for which reimbursement will be provided. In this respect it is of note that the legislation establishing Medicaid and Medicare did not include criteria of safety or efficacy but, rather, discussed the criteria of reasonable and medical necessities. These criteria have not been explicated in specific regulations or procedures.

Although the federal government has no direct regulatory role with regard to psychotherapy, as it does with regard to drugs, it has a major role in supporting scientific research on mental illnesses and their treatment. The current imbalance in available evidence for efficacy of psychotherapy in relation to psychopharmacology has many sources; one is the social and economic structure of treatment research. In the case of pharmacological agents, the pharmaceutical industry is organized into large corporate bodies with considerable resources and incentives for research on the efficacy and safety of their products. In contrast, the psychotherapy “industry” is made up of many small firms and practitioners whose resources are less extensive and who are less capable of concerted action. It might be expected that the institutes of the Alcohol, Drug Abuse, and Mental Health Administration, particularly the National Institute of Mental Health (NIMH), would devote leadership and resources to treatment research, but here again, for complex reasons, NIMH’s record on funding psychotherapy research is inadequate in total grants and not reflective of clinical practice or professional judgment. Efforts to correct this imbalance require greater cooperation between officials of the Alcohol, Drug Abuse, and Mental Health Administration and the professional leadership than has been achieved to date.

State governments have an important potential role with respect to these issues because licensure and certification of health professionals are the responsibility of state governments, as is the licensing of hospitals and clinics. Almost all state governments have established standards for professional licensing of physicians. An increasing number of state governments have established criteria for licensing and/or certification of psychotherapists, particularly psychologists and social workers. Similarly, almost all hospitals, including private psychiatric hospitals such as Chestnut Lodge and Silver Hill Foundation, require licensing in their respective states. However, no state has attempted to establish guidelines for the selection of treatments based on efficacy as part of licensing or certification requirements.

In the absence of a government body similar to the Food and Drug Administration, patients and the public might expect that professional associations such as APA, the American Psychological Association, or the National Association of Social Workers would undertake to provide this service to the public. No guidelines for treatment have emerged, however, although peer review criteria have been established. APA issued a report on the status of ECT in 1978 (42). The Royal Australian and New Zealand College of Physicians has contracted with the Australian Ministry of Social Security to undertake a quality assurance program, which has issued a series of reports reviewing the state of scientific evidence for selected diagnoses, including depression (43).

As of the late 1970s, when Dr. Osheroff was hospitalized, APA had published a manual for peer review of hospital utilization (44). With regard to the DSM-II diagnosis of psychotic depressive reaction, this manual recommended the use of drugs or ECT. It did not recommend individual psychotherapy. Furthermore, this manual recommended that if hospitalization has continued beyond 1 or 2 months, the case should be reviewed and the use of ECT or drug treatments considered. Therefore, although there were no government bodies offering legal guidelines, APA had established peer review criteria for the hospital treatment of psychotic depressive reaction (44).

APA is currently completing a project on psychiatric treatments under the leadership of T. Byram Karasu (45). Preliminary reports from this project have been published (46).

Given that there are no government bodies judging the efficacy of claims for psychotherapy, and given the limited efforts undertaken by professional associations, it is understandable that individual patients use the courts to seek redress for their grievances.

Governmental and professional bodies have been urged to issue judgments recommending treatments so that these criteria could be used by reimbursement agencies. In response, the Senate considered possible legislation to establish a National Commission on Mental Health Treatments in the late 1970s and, more recently, APA established the Commission on Psychiatric Therapies, led by Dr. Karasu. Some have advocated that the profession not make such recommendations in regard to treatment, assuming that if the profession did not take such actions the courts would ignore the issue or not take a position. The opposite seems to be the case. In the absence of professional criteria for standards of care, the courts are increasingly becoming the arena in which these disputes are adjudicated. Thus, case law and individual precedents may become the criteria for adequacy of diagnosis and treatment.

Dr. Stone (3) raised the possibility that patients who have not improved after prolonged psychotherapeutic treatment may have found a way around their frustrations—a way provided by “biological psychiatrists.” Dr. Stone noted that biological psychiatry appears to be on the scientific ascendancy over psychodynamic psychiatry due to the prestige of the neurosciences and the evidence for efficacy of biological treatments.

My conclusion, however, is that the issue is not psychotherapy versus biological therapy but, rather, opinion versus evidence. The efficacy of drugs and other biological treatments is supported by a large body of controlled clinical trials. This body of evidence is all the more relevant to public policy in view of the paucity of studies indicating efficacy for individual psychotherapy.

It is regrettable that psychoanalysts and psychodynamic psychotherapists have not developed evidence in support of their claims for therapeutic efficacy. Twenty years ago, psychodynamic psychotherapy was the dominant paradigm of psychiatry in the United States, particularly in academic centers. A number of European psychiatrists, mostly psychoanalysts, contributed intellectual leadership and imaginative ideas to psychiatry here. Currently, however, psychoanalysis is on the scientific and professional defensive. This situation is, in part, a consequence of the failure of psychoanalysis to provide evidence for the efficacy of psychoanalysis and psychodynamic treatments for psychiatric disorders (47, 48).

In the period between World War I and World War II, biological psychiatry was in poor repute. Numerous treatments, often of a heroic nature, were advocated: colonic resection, adrenalectomy, excision of teeth, lobotomy. These interventions were based on biological laboratory research of dubious quality and without any systematic studies of safety and efficacy. The situation changed after World War II, with evidence for the value of ECT for depression and insulin coma therapy for schizophrenia and, later, with the introduction of chlorpromazine and other drugs.

The case of Osheroff v. Chestnut Lodge prompts a reevaluation of the doctrine of the respectable minority. Until recently, this doctrine held that if a minority of respected and qualified practitioners maintained a standard of care, this was an adequate defense against malpractice. I propose that this doctrine no longer holds if there is a body of evidence supporting the efficacy of a particular treatment and if there is agreement within the profession that this is the proper treatment of a given condition. Moreover, the respectable minority have a duty to inform the patient of the alternative treatments. In an unpublished 1985 paper discussing Osheroff v. Chestnut Lodge, K. Livingston wrote,

Under this view, the respectable minority view would still constitute a defense to a malpractice action where even 10% of practitioners would adhere to the treatment in question. However, the shield of the respectable minority rule would not be available unless the patient had been given informed consent after a disclosure of risk/benefits and alternatives to the therapy.

When there is consensus in the profession as to the appropriate treatment for a given condition (in the case of Osheroff, the essential nature of biological treatment for severe depression), then a standard of care can be agreed on and can provide the basis for malpractice action.

However, how are we to evaluate claims for the efficacy of treatments for clinical conditions about which there is no consensus? What are the standards to be applied in diagnostic and clinical situations where there is no consensus within the field with regard to the treatment of the particular disorder? This is a serious policy question that, in the future, may become a legal question. In my opinion, there are three aspects to this issue: 1) What constitutes evidence for efficacy? 2) Who is responsible for generating the evidence? and 3) Who is to make the appropriate evaluation of treatments?

What constitutes evidence of treatment? In my view, the best available evidence as to efficacy comes from controlled trials. I am not taking the position that the only source of evidence for efficacy comes from such trials. Clinical experience, naturalistic studies, and follow-up studies are also sources of relevant evidence. However, when results from controlled clinical trials are available, they should be given priority in any discussion of scientific evidence.

Who should be responsible for generating the evidence? What should be society’s policy in regard to treatments for which there is no positive or negative evidence? This issue has not reached resolution, and I feel it merits further discussion within the profession.

My opinion is that the responsibility for generating evidence for efficacy rests with the individual, group, or organization that makes the claim for the safety and efficacy of a particular treatment. In the case of drugs, this responsibility is established by statute. If a pharmaceutical firm makes a claim for the efficacy of one of its products, it must generate enough evidence to satisfy the Food and Drug Administration before it can market the drug for prescription use.

No such mandate of responsibility exists for psychotherapy. Anyone can make a claim for the value of a form of psychotherapy—psychoanalysis, Gestalt, est, primal scream, etc.—with no evidence as to its efficacy.

What should be our position toward the claims of the efficacy in certain conditions of multiple treatments for which the evidence varies in quality and quantity? In my view, those treatments which make claims but have not generated evidence are in a weak position.

The efficacy of psychoanalysis and psychoanalytic treatments is in question for conditions for which there is evidence of efficacy with other treatments. For example, how many psychiatrists would justify longterm psychoanalytic treatment of panic disorder and/or agoraphobia when there is no evidence that this treatment works for these disorders but reasonably good evidence for the efficacy of certain drugs and/or forms of behavioral psychotherapy?

Who is to evaluate the evidence? A major problem arises as to the process by which the evidence regarding psychiatric treatments is to be evaluated. I believe there are serious deficiencies in our current professional and governmental arrangements for evaluating psychiatric treatments. In the case of drugs, we have the Food and Drug Administration, which makes such judgments according to established legal statutes and regulatory processes. There is no comparable statutory mandate for assessing the efficacy and safety of nonpharmacological treatments such as radiation, surgery, and psychotherapy.

In this situation, I believe the public has the right to expect that the medical profession will provide appropriate judgments as to the state of the evidence for treatments and establish criteria for standards of care. I maintain that the psychiatric profession has been lax in this responsibility and that the absence of professional consensus statements in our field leaves it open for the courts to be used by individuals, such as Dr. Osheroff, who feel they have been poorly treated and who believe they are entitled to redress of their grievances.

The fact that evidence changes is to my mind irrelevant to any policy or clinical discussion. The judgment on treatment of individual patients should be made according to the state of knowledge and professional practice at the time the individual patient is treated. In the case of Osheroff, this was 1979.

My strong preference would be for the profession to be more vigorous and more responsible in accepting this responsibility. I have stated these views on a number of occasions.