We are witnessing an unprecedented wave of interest in outpatient commitment. In 1999 outpatient commitment statutes were introduced in six state legislatures and adopted in two—Wyoming and New York—bringing to 41 the number of states permitting some form of involuntary outpatient treatment (1). Several states have established task forces to consider the issue, including West Virginia, Nevada, Minnesota, Maryland, and California. Professional and advocacy organizations are lining up on one side or another of the question or updating their long-held positions. Why has outpatient commitment become the issue of the hour?
In part, as with so many aspects of mental health law, enthusiasm for outpatient commitment stems from concerns about highly publicized acts of violence by persons with mental disorders. Thus New York's recently adopted outpatient commitment statute has been dubbed "Kendra's law," after the woman pushed to her death in a New York City subway station by a man with untreated schizophrenia (2). It is no small irony that the murderer, far from rejecting treatment like the prototypical target of outpatient commitment laws, had actually sought assistance on several occasions, only to be turned away because of a lack of available slots in the treatment programs (3).
New York's experience suggests that outpatient commitment, in some quarters, is currently perceived as a virtual panacea for the problems caused by severe mental illness. Even when the unfortunate outcome appears to have been related to the absence of sufficient services, the prescription, rather than increasing service availability, is to institute a system of involuntary outpatient care.
To be sure, other recent tragedies might constitute stronger examples of the potential prophylactic effect of outpatient commitment. These include the 1998 murders at the U.S. Capitol by Russell Weston, who had been lost to follow-up after discharge from a Montana state hospital (4). However, fear of violence by persons with mental illness, although probably the strongest motivating force behind the current push for outpatient commitment statutes, is one of the weaker justifications for new laws. Most people with severe mental illness are not violent (5); only a tiny fraction of the nation's violence can be attributed to mentally ill persons. The best available estimate is about 3 percent (6). Also, it is unclear whether many of these acts are perpetrated by persons who would be eligible for or deterred by outpatient commitment. Here, as in most circumstances, a handful of highly visible cases constitutes a dubious basis for social policy.
Is the current focus on outpatient commitment, then, completely misguided? Perhaps not. Provision of involuntary outpatient treatment may be an important component of a system of care for persons with schizophrenia, bipolar disorder, and other serious mental illnesses for reasons entirely unrelated to the prevention of headline-grabbing acts of violence. Put somewhat differently, it may be the right answer to another, no less important question than how to prevent violence: how do we care for patients subject to recurrent psychotic episodes that lead to severe deterioration and may require repeated hospitalization, but who are resistant to participating in treatment that might maintain them in a nonpsychotic state?
The "revolving-door" patient, as the most problematic members of this group are called, is a familiar figure to most mental health professionals who work with the severely and chronically ill population. Admitted to a psychiatric inpatient facility, this patient often recompensates quickly in response to psychotropic medications, removal of access to intoxicating substances, and the structure provided by ward routine. Once discharged, however, he or she stops taking medications, drifts away from outpatient care, frequently resumes substance abuse, and soon appears at the hospital door, requiring admission once again. The cycle can be repeated three, four, ten, or 20 times in a year. Although good data on the frequency of such cases are lacking, we cannot ignore the observations of involved clinicians that such cases are not rare and that they consume a vastly disproportionate amount of limited mental health system resources (7,8).
Can we say that outpatient commitment is likely to be effective with this group? Here we are on uncertain ground, even though data from studies of outpatient commitment have been accumulating for almost two decades. There is no shortage of reports demonstrating positive effects of involuntary outpatient treatment, including decreased rates of hospital readmission and shortened lengths of stay (9,10,11,12), increased utilization of aftercare (13,14), and continued compliance with outpatient treatment even after the termination of the outpatient commitment order (15,16). Not all studies have shown positive effects (17,18), but negative findings have tended to be correlated with erratic enforcement, lack of clinician support, and unavailability of the resources needed to provide appropriate services.
However, all of the studies of outpatient commitment have been limited by the ways in which their subjects were selected. Investigators who reported changes in rates of service use in their patient sample before and after commitment were unable to control for the effects of concomitant changes in patients' mental states or environments, such as the stabilizing effects of hospitalization or the availability of a concerned and aggressive treatment team. Studies comparing committed and uncommitted groups suffered from nonrandom assignment, which created the possibility that patients with a better prognosis might have been selected for involuntary outpatient treatment, thus biasing the findings in favor of the intervention.
Hence a certain amount of excitement greeted word that the two studies summarized in this special section were under way (19,20). They represent the first randomized trials of outpatient commitment ever attempted—extraordinary achievements in themselves. Yet, even randomization does not ensure that data are free from contention. The apparently negative findings of the New York study have been challenged by several of those parties most directly involved with the Bellevue program. Telson and colleagues (21) have contended that during the 11-month study period, there was considerable confusion among both patients and treaters as to who was in the experimental group and who was in the control group, and that mechanisms for enforcing court orders were lacking. As this situation demonstrates, even the best-designed evaluation studies are hostage to the manner in which the intervention being examined is implemented.
Questions of a different sort can be raised about the North Carolina study. On its face, the study offers evidence of the efficacy of outpatient commitment in reducing hospital stays and the incidence of violence, as long as the intervention is sustained and frequent services are provided. These findings were obtained even though North Carolina law limits recourse when patients fail to comply with the commitment order; the law excludes both involuntary administration of medication and automatic rehospitalization. The research team speculated that the positive effects they observed stem from a combination of greater motivation by patients to comply with treatment and the effects of court orders on the diligence with which treatment teams pursued patients and made resources available (22). But subjects were not selected randomly for extension of commitment beyond 90 days, and differences in outcomes between the subjects with shorter and longer durations of commitment began almost immediately. Thus the possibility cannot be ignored that the two groups were different in composition or were treated differently from the inception of the study.
Undoubtedly, considerable discussion of these issues will ensue as more data from the study are published. However, enough questions exist at this point to ensure that persons who doubt the efficacy of outpatient commitment will not have their qualms stilled by the North Carolina data alone.
On what basis, then, can we determine whether to support adoption of outpatient commitment statutes or their application in the many jurisdictions where they already exist but are infrequently employed? It is unfortunately common for us to have to answer policy questions of this sort in the absence of conclusive data on the efficacy of the proposed intervention. Indeed, the existence of such data constitutes the rare exception rather than the rule. Consider, for example, that well-designed studies comparing the efficacy of involuntary inpatient commitment to possible alternatives have never been performed, yet every state in the country has an inpatient commitment statute, and few would question the need for some power of this sort.
The existing data, all imperfect, tend to favor the efficacy of outpatient commitment as a means of stabilizing patients in the community, and many clinicians who have been involved in the process share this view. In my judgment, these conclusions should not be ignored, even as we await more definitive data on the effects of different models of involuntary outpatient care. Moreover, it is clear that legislators are looking to outpatient commitment—sometimes unrealistically—to solve a variety of problems in the treatment of mental illnesses. This alone makes it worthwhile to consider how statutes governing the process might be fashioned. Among the decisions legislators must make are what criteria to use to define the target group for outpatient commitment and what remedies ought to be available to respond to patients who fail to comply with the terms of their commitment orders.
A majority of the states that have statutory provisions for outpatient commitment use the same criteria as for inpatient commitment—almost always danger to self or others or inability to meet one's basic needs. Experience in these jurisdictions, however, has suggested that homologous standards discourage the use of outpatient commitment. Clinicians and courts alike have a difficult time determining which patients are sufficiently impaired to meet dangerousness criteria for inpatient commitment and yet might be appropriate candidates for enforced outpatient care. Concerns about liability play no small role in this situation. Having declared a patient dangerous for purposes of commitment, many clinicians understandably shy away from recommending outpatient treatment, fearing that they will be held responsible for any harm that occurs to the patient or to other people.
Moreover, it is not at all clear that a standard based on a patient's current level of dangerousness will succeed in identifying the population most likely to benefit from outpatient commitment: those who have made several passes through the revolving door. At the point of hospital discharge, when outpatient commitment might be invoked, most of these patients are quite stable, although their subsequent deterioration—given their past history—seems all but preordained. To encompass this group, eligibility criteria in outpatient commitment statutes should focus on:
• a history of repeated deteriorations that required involuntary hospitalization
• the likelihood that, without intervention, the patient will again deteriorate to a state in which inpatient commitment criteria are met
• a treatment plan that holds out the prospect of stabilizing the patient in the community
Involvement by the community treatment team in fashioning the terms of the outpatient order and the team's concurrence in the final plan is another important dimension of an effective outpatient commitment process.
The other conundrum facing drafters of outpatient commitment statutes is how to deal with a patient's noncompliance with the outpatient commitment order. Jurisdictions with identical inpatient and outpatient commitment criteria may be able to rehospitalize noncompliant patients, although it is likely that a hearing would have to take place soon after hospitalization to confirm the patient's committability. For jurisdictions whose outpatient criteria are broader than their inpatient standards, however, this is not an option, unless patients can be shown independently to meet the requirements for inpatient commitment.
A number of states permit noncompliant patients to be apprehended by law enforcement personnel and brought to the mental health center where their care is being supervised. Such apprehension can occur at the initiative of the patient's clinician or, as in the North Carolina statute, only at the order of a clerk of court acting on a petition filed by the clinician.
A subject of considerable debate is what happens when the uncooperative patient arrives at the facility. Should treatment staff be allowed to administer medication over the patient's objections? Earlier data from North Carolina, where staff can only exhort the patient to comply with treatment, suggest that even the power to have patients brought to the clinic forcibly was rarely used (14). Many clinicians there and elsewhere believe that patients' compliance can generally be obtained without the threat of involuntary administration of medication. However, some observers think that high rates of compliance are due to the assumption by committed outpatients faced with a judicial order that they will be treated involuntarily if they demur. If this assumption is true, it suggests that noncompliance may become increasingly problematic as patients gain awareness of the limited remedies available to caregivers. For now, however, the number of patients who require or would benefit from compelled administration of medication appears to be small.
Perhaps the most important issue that legislators must consider in adopting or revising outpatient commitment statutes has to do with the availability of resources to provide care to committed patients. Most community-based services of the sort that would be called upon to supervise outpatient commitment orders are already overworked and underfunded. Adding to their workload the care and oversight of patients who, by definition, have been uncooperative with treatment makes little sense. This factor may account for the low rates of use of outpatient commitment in many jurisdictions. The patients targeted by these statutes will require the kind of active outreach and intensive service coordination exemplified by assertive community treatment models (23). Legislatures that fail to provide new resources to support such programs in conjunction with the passage of outpatient commitment laws are engaging in a sham effort unlikely to have any real impact on the care of the mentally ill.
What objections might be offered to a framework for outpatient commitment such as the one outlined here? Many opponents of this approach maintain that we cannot legitimately employ a compulsory option until all less coercive possibilities have been demonstrated to be ineffective. Specifically, they argue for the development of a full range of voluntary outpatient programs, with input from the persons who will be using them and with adequate funding to serve every person who seeks care. Only then will we know which patients are genuinely treatment resistant and thus reasonably considered candidates for outpatient commitment. Advocates of this perspective believe that the number will be sufficiently small to vitiate the utility of an outpatient commitment law.
Without question, the availability of adequately funded services for persons with mental illness would make a big difference in the effectiveness of care. It is a goal toward which the efforts of all those who care about the treatment of persons with serious mental illness should strive. But given the nature of the many major mental disorders in which denial of the illness and the need for treatment plays a prominent role, it seems unlikely that even an optimal system of care will put a stop to the revolving door. In addition, although it would be wonderful to see a fully funded system of psychiatric care, nearly two centuries of experience with public mental health systems in this country casts doubt on how realistic a goal that is. Policy must be made for the real world, and in that world too many severely ill patients simply drift away from care.
To say that the weight of the evidence and of clinical experience now favors efforts to implement reasonable statutory schemes of outpatient commitment is not to say that the issue has been settled conclusively. Research must continue in order to confirm the effectiveness of the approaches being used, to identify the populations most likely to be helped, and to explore alternatives to compulsory treatment. The lessons learned from the New York and North Carolina studies will help us along that road.
Dr. Appelbaum is A. F. Zeleznik professor and chair in the department of psychiatry at the University of Massachusetts Medical School, 55 Lake Avenue North, Worcester, Massachusetts 01605 (email, firstname.lastname@example.org). This paper is part of a special section on outpatient commitment.