Outpatient commitment remains a controversial approach to the treatment of people with serious mental disorders who decline to participate in ordinary care (1). Forty-two states now have some form of statutory authorization for involuntary outpatient treatment (2), although surveys suggest that only a minority actively implement such laws (3). Indeed, some statutes appear to be written in such a way as to preclude outpatient commitment from being widely adopted (4). In recent years, New York State initiated a high-profile effort to develop an outpatient commitment system, and recently new data have become available that provide a look at where New York's efforts stand.

New York's statute, often referred to as Kendra's Law, was passed by the state legislature in 1999 in the wake of the notorious murder of Kendra Webdale, a 32-year-old woman who was pushed in front of an oncoming subway train by a man with untreated schizophrenia. Although the man had sought treatment for his condition and been turned away for lack of available slots, and hence was not the prototypical target of outpatient commitment statutes, the attack galvanized the public and lawmakers in support of the proposed legislation (5). Like many of the newer statutes, Kendra's Law focuses on people with mental illness whose nonadherence has resulted in repeated hospitalization or incarceration, or has led to threats or acts of serious violence, and who could be predicted to deteriorate to a similar state without what the statute terms "assisted outpatient treatment" (6).

Although New York's law largely resembles several other recent statutes, its implementation had some unusual aspects. Many states have adopted outpatient commitment laws without providing for the increased costs of developing more intensive services for a difficult-to-treat population. In contrast, New York appropriated funds to expand case management and other services and to pay for the care of patients who have been committed. Incremental funding support has continued to this day. Kendra's Law also had a sunset clause—which is uncommon in mental health legislation—that required reauthorization of the program by June 30, 2005, if it were to continue. Hence, at the time of this writing, New Yorkers are debating whether to extend their experiment with outpatient commitment. Finally, New York required periodic evaluations of the law's implementation so that its effects could be gauged, a salutary approach to public policy that is all too infrequently observed at either the state or the local level.

As might have been anticipated given the contentious debate over outpatient commitment, the new statute was the target of a number of legal challenges. The most important of these was In the Matter of K.L., a test of the constitutionality of the statute that was ultimately decided by New York's highest court (7). K.L. was a patient with a diagnosis of schizoaffective disorder, a history of noncompliance, and a record of aggressiveness toward family members when he was experiencing decompensation. When a petition for outpatient commitment was filed, rather than merely claiming that he did not qualify for commitment, K.L. argued that the statute itself was constitutionally defective. His primary contention was that the statute violated due process because an order for assisted outpatient treatment was not predicated on a finding that the patient was incapable of making his own decisions about treatment.

In his claim, K.L. relied on an earlier New York decision, Rivers v. Katz (8), which held that even involuntarily committed patients could not be medicated against their will unless they were determined by a judge to be decisionally incapable. But the New York Court of Appeals held that Rivers was not relevant here because, like most outpatient commitment statutes, Kendra's Law did not authorize involuntary medication. Even hospitalization of the committed outpatient required that the usual standards for involuntary admission be met. Indeed, the law's only "teeth" should patients fail to comply with a judicial order for outpatient treatment was to permit them to be picked up and detained for up to 72 hours to determine whether they met inpatient commitment criteria. A challenge to this provision on due process grounds was also turned aside. Reasoning that because outpatient commitment had involved a finding that the patient would be at risk of causing or suffering serious harm if noncompliant, the court held that the state had a legitimate interest in removing noncompliant patients from the streets to determine whether the prediction had been accurate.

The irony in the court's decision should not go unnoted. Although heralded by advocates of outpatient commitment as an endorsement of the constitutionality of such laws, In the Matter of K.L. upheld the statute only because it lacked any direct mechanisms to enforce compliance with treatment. Outpatient commitment laws are often attacked by civil libertarians as being destructive of individual liberty; however, beyond the hortatory effect of a judge's admonition, and perhaps the prospect of a short-term hospital stay, they generally are without any coercive power. And although it is possible that courts in states that, unlike New York, do not require a finding of incapacity before medication can be administered involuntarily would uphold provisions for forcible administration of medication, legislative drafters have shied away from confronting this option.

That Kendra's Law is constitutional, however, does not necessarily mean that it represents good policy. The latter determination depends on an assessment of the law's effectiveness. In this regard, the publication of the required final report on implementation of the statute by New York's Office of Mental Health should be of considerable interest (9). A previous study of a pilot project on outpatient commitment in New York City, the seed from which Kendra's Law grew, was disappointing in that it showed no significant differences in outcome between committed and control subjects (10). But the results of the study were compromised by problems in implementing the law, essentially leaving the question of effectiveness unanswered (11). Unfortunately, the new evaluation lacks a control group, which means that some of its conclusions will be open to varying interpretations. Nevertheless, it offers an interesting look at how the law is actually being implemented.

Over the roughly five years covered by the study, more than 10,000 persons were referred to county-level assisted outpatient treatment coordinators for assessment of their eligibility (9). Of those, petitions were filed for 4,041 and granted in 93 percent of cases. Almost two-thirds of the patients who were subject to a court order had their commitments renewed after the first six months, for an average commitment period of 16 months. Committed patients had a mean age of 38 years and tended to be male, to have never married, to have a psychotic disorder, and to have substance use problems. Non-Hispanic blacks were disproportionately represented in the committed group, a finding that has already aroused concern (12). Over the three years before commitment, 97 percent of patients had been hospitalized, 30 percent arrested, 23 percent incarcerated, and 19 percent homeless.

Outcomes were assessed at six months by clinicians who were involved in the patients' treatment; data were available for 2,745 people. Compared with the situation before commitment, rates of involvement in services had increased markedly—for example, for case management, 100 percent compared with 53 percent; for medication management, 88 percent compared with 66 percent; and for substance abuse services, 40 percent compared with 24 percent. Problems in self-care and community living dropped by 23 percent on average, and comparable improvements were seen in other functional domains. Harmful behaviors dropped by 44 percent, including a 47 percent decrease in physical harm to others. Arrest, incarceration, psychiatric hospitalization, and homelessness all dropped by between 74 percent and 87 percent. It seems clear that, as a group, patients did better with assisted outpatient treatment than they had previously.

However, as with any program evaluation, there are several caveats to keep in mind, even though the findings have been trumpeted in some quarters as indicating that the law is a "remarkable success" (13). Ratings of patients' service use and functioning were made by case managers, who may have been unaware of some aspects of patients' functioning or predisposed to view the program as producing improvement. Long-term outcome data are not yet in, leaving open the question of whether patients sustain their gains, especially once they are no longer subject to the program. It seems clear that the current data will not satisfy many critics who contend that making a similar set of services available on a voluntary basis would have led to similar results, without the need for a judicial order. Only a randomized controlled comparison of voluntary and involuntary patients is likely to answer this question, and the data fall short of that standard.

But that is not to say that we cannot learn something from the experience with Kendra's Law over the past five years. Statutes like this one, which reflect a new generation of legislation on outpatient commitment, are likely to be found constitutional by the courts. In implementing the law, New York State has been able to focus on a very ill and impaired target population and appears to have generated substantial improvements in these individuals' levels of functioning. The availability of funding for additional services probably made a big difference in the outcomes that are now being reported. And, perhaps the most certain conclusion of all, the controversy about the use of outpatient commitment is far from over.

Dr. Appelbaum, who is editor of this column, 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 01655 (e-mail, appelbap@ummhc.org).