Watching a loved one fall into the grip of severe mental illness can be painful and terrifying in equal measure, as the person’s behavior becomes increasingly more bizarre and self-destructive. Long-treasured bonds to family and friends often fall by the wayside. Personal hygiene is often neglected, along with the person’s concern for his own basic welfare. Life savings can be rapidly depleted in manic spending sprees. And yet sometimes, to the chagrin and astonishment of those who want to help the ill person find a way back to the life he once had, pleas to seek treatment are met with emphatic insistence that everything is fine. Pressing the matter often only leads to further alienation and hostility. At some point, when things become unbearable for the concerned observer, a call will be made to police or a local mental health facility: “Something is very wrong. He’s not himself. Can you help?”
It is at this moment that the situation becomes not merely a health emergency, but also a legal matter. The caller is asking authorities to override the mentally ill person’s constitutional liberty interests by detaining him against his will for evaluation and/or treatment. Of course, no constitutional right is absolute; as in all things, we rely on law to strike the right balance between individual rights and societal imperatives.
This leads to a critical question: What exactly are the societal imperatives activated by a psychiatric crisis? One obvious answer is the need to eliminate a substantial risk of imminent death or physical injury. But what if the person is neither threatening violence against anyone nor at any apparent imminent risk of injuring himself? What if the concern spurring the family member to seek help is simply that the person is suffering in isolation, tormented by terrifying delusions, yet somehow unaware that he is ill? Do we as a society have reason to intervene? To answer “yes,” we must believe there is a compelling societal imperative beyond preventing imminent injury or death — an imperative to liberate a person from a hellish existence he would never in his “right mind” choose.
This is a major question that states have had to grapple with in crafting their laws on involuntary treatment. And the answer each state reaches has implications far beyond the initial need to detain a person for an emergency evaluation. Laws must also address the circumstances that typically follow a diagnosis of severe mental illness, such as how long and under what criteria the person should be held by a court of law for continued inpatient treatment against his wishes (“civil commitment”); whether the state, upon releasing the person from hospital care, should require him to adhere to a prescribed treatment plan (“assisted outpatient treatment”); and how far down the spiral of relapse a person must fall before he is involuntarily re-hospitalized.
During the past 50 years, this has been a highly turbulent area of the law. Prior to that, obtaining involuntary treatment was straightforward. Typically, state laws hinged on a simple determination that the person required care and allowed commitments to be continued indefinitely without ongoing judicial oversight.1 The de-institutionalization movement of the 1960s brought a national trend to reform these laws, shifting the focus to the person’s “dangerousness to self or others” as the basis for civil commitment.1 The trend accelerated in (over) reaction to the 1975 U.S. Supreme Court ruling in O’Connor v. Donaldson, which held that “a State cannot constitutionally confine without more a nondangerous individual who is capable of surviving safely in freedom by himself or with the help of willing and responsible family members or friends.”2
Kenneth Donaldson, the individual seeking release in the O’Connor case, had been confined for 15 years in a Florida state hospital with no meaningful attempt to offer him treatment for his purported mental illness. This absence of treatment was critical to the court’s analysis of the case, and the ruling was carefully limited to address the constitutionality of confinement “without more” — meaning without treatment. Justice Stewart, writing for the unanimous court, even went out of his way to point out that “there is no reason now to decide … whether the State may compulsorily confine a nondangerous, mentally ill individual for the purpose of treatment.”4 The clarification went largely unnoticed as O’Connor quickly came to be understood by some as a landmark repudiation of all commitments of nondangerous individuals.
Compounding this tragedy in the wake of O’Connor, there began to take hold within American mental health systems an exceedingly narrow understanding of what it means to be “dangerous to self.” Despite a clear statement to the contrary in the O’Connor opinion,4 “dangerous to self” came to be understood to mean that a person is at risk of imminent suicide or intentional self-injury — and nothing else.
In the late 1970s, many psychiatrists, policymakers, and academics began looking around at the results of de-institutionalization and wondering if perhaps the pendulum had swung too far. Although community placement had undoubtedly improved the lives of some, another consequence had become evident: a large number of desperately ill people had been abandoned to the streets and the penal system.1 And so began a counter-movement to re-think state laws, which continues to this day. The goals of this re-examination have been twofold: first, to affirm that there are circumstances other than the imminent risk of violence or suicide that warrant hospital commitment; and second, to minimize the need for such involuntary hospitalizations through the lesser liberty intrusion of court-ordered outpatient treatment, where appropriate.
At a minimum, a state’s statutory scheme for involuntary treatment must address both the criteria for commitment (the legal standard under which the judge decides whether commitment is necessary), and the process of commitment (the nuts and bolts of getting the matter before a judge for consideration).
The widespread misunderstanding of the role of “dangerousness” in the civil commitment equation is rooted in two intertwined misconceptions.
The first is the notion that a person must pose a risk of imminent harm to be deemed dangerous. As noted in the previous discussion of O’Connor, the U.S. Supreme Court has never held that dangerousness in any form is constitutionally required in commitments for purposes of treatment. State high courts have consistently upheld commitment criteria hinging on a risk of foreseeable future harm, rather than risk that appears immediate or imminent.5,6
The second misconception is that “dangerousness” means only one thing (ie, a likelihood to intentionally cause serious physical harm to oneself [suicide or self-mutilation] or another [violence]). Common sense should tell us that there are ways to be dangerous to self or others without intent to harm anyone. Justice Stewart confirmed this in a footnote to his O’Connor opinion: “Of course, even if there is no foreseeable risk of self-injury or suicide, a person is literally ‘dangerous to himself’ if for physical or other reasons he is helpless to avoid the hazards of freedom[.]”7
To varying degrees, most states have by now moved beyond these common fallacies in their statutory civil commitment criteria. The more progressive commitment standards come in two basic varieties, known generally as “grave disability” standards and “need-for-treatment” standards (although specific terminology varies by state.)
Grave disability standards are rooted in the premise that a person poses a physical threat to himself when mental illness renders him unable to provide for the basic necessities of human survival, just as surely as if the illness was causing him to actively attempt self-harm. A grave disability standard encourages the hospital commitment of a person whose untreated mental illness has led him to living under a bridge and foraging in dumpsters for food.
Need-for-treatment standards open the hospital gates wider still. The underlying notion here is that deterioration of general health, psychiatric damage, and loss of ability to function independently, all of which typically follow when severe mental illness goes untreated, are unacceptable harms per se. The crux of eligibility for treatment is a finding that the person’s mental illness prevents him from seeking help on a voluntary basis and, if not treated, will cause him severe suffering and harm his health. Need-for-treatment laws make commitment available to the person who suffers profoundly, even if he manages to meet his basic survival needs and exhibits no violent or suicidal tendencies.
Not all grave disability or need-for-treatment standards are created equal. Utility depends largely upon whether intervention is permitted on the basis of future foreseeable harm. Laws that prohibit action until harm occurs (or appears “imminent”) force a would-be petitioner to bide his time until the ill person deteriorates further. Needless to say, it is not always easy to precisely time a petition for commitment to reach a judge just before “imminent” disaster occurs.
It is also important to note that even in the few states that do not expressly articulate standards beyond a general notion of danger, a mental health evaluator or judge could reasonably interpret “danger” to encompass grave disability or need for treatment. However, experience has shown that where state law does not include an explicit grave disability and/or a need-for-treatment standard, mental health systems too often insist upon likelihood of violence or intentional self-harm as the only basis for commitment. Judges rarely get opportunities to impose a broader view of dangerousness, because the narrow view tends to prevail in the psychiatric evaluations that determine whether commitment cases ever reach the court. This filters down to police officers and crisis outreach workers, who quickly learn that it is a waste of time to detain a mentally ill person for evaluation unless he exhibits a risk of imminent violence or suicide. Families are routinely told to call again when the individual hurts someone, threatens to, or tries to.
Knowing as we do that imminent risk of violence or suicide is demanded even in some jurisdictions that have explicitly broader laws (due to “triaging” that inevitably results when mental health systems do not make an adequate number of inpatient beds available), we harbor no illusions that mental health system dysfunction may be cured legislatively. But there can be little doubt that explicitly broader criteria are a necessary (if not sufficient) condition for sound commitment policies. There is also evidence that more-inclusive commitment laws exert positive pressure on mental health systems and have important ripple effects. As one example, a 2011 study published in Social Psychiatry and Psychiatric Epidemiology found a significant association between broader state commitment standards and lower rates of homicide.8
Although there are many important procedural aspects to a state’s hospital commitment law, we focus here on two we regard as critical.
First, it is important for a state to grant private persons the right to petition the court for commitment. The more broadly this right is extended, the better. We are not advocating here for allowing anyone to be committed on the basis of lay opinion. In all cases, hospital commitment must rest in part on the expert testimony of a professional who has recently examined the person, which is why local mental health authorities or providers are usually the natural parties to petition for commitment. The problem is that in some cases, a brief mental health evaluation conducted by an overextended, unfamiliar public psychiatrist does not in itself offer a compelling case for the need to commit. Those who know and care about the individual are sometimes in a better position to demonstrate this need — not to discount the findings of the professional evaluation, but to place them in a more meaningful context for the judge.
Another critical aspect of a state’s inpatient commitment law is the maximum duration for which it allows the commitment to be imposed. Court-ordered hospitalization is not meant to be punitive, but rather a means to restore mental health. Accordingly, the time period attached to a commitment order is not a “sentence.” It is in fact unconstitutional to detain a person in a hospital under a civil commitment if his treating physician has determined that he no longer meets the state’s inpatient commitment standard. But in practice, it usually takes more than a few days for a person to achieve stability after a mental health crisis.
In most public psychiatric hospitals, there is inherent tension between the need to treat and the need to clear beds to meet incoming demand. The length of commitment orders plays directly into this tension. Although it is always possible to renew an expiring commitment if the patient is thought to continue to need hospital care, in practice, the expiration of an order has the effect of pushing the person out the door, fully stabilized or not. A state law that limits an initial court order of hospital commitment after emergency detention to fewer than 30 days is inadequate, and a limitation of such order to 14 days (as in West Virginia and Washington) is unacceptable.
Assisted Outpatient Treatment
The notion of civil commitment is generally associated with a hospital setting. But there is another longstanding and well-established form of court-ordered commitment that is nearly as essential to the optimal functioning of a mental health system. It is outpatient commitment — or alternatively, “assisted outpatient treatment” (AOT) — which is the practice of court-ordering a person with mental illness who meets certain legal criteria to adhere to a specific program of outpatient treatment as a condition of remaining in the community.
To grasp the importance of AOT, it must be understood that non-adherence to prescribed treatment is the single largest reason that people get caught in the mental health system’s “revolving door,” shuttling endlessly between hospitals, correctional facilities, and the streets. Improved treatment adherence is the key to avoiding this, but it is not easily accomplished, particularly for those with anosognosia, a symptom of brain dysfunction that prevents the sufferer from recognizing his own illness.
AOT is not a panacea to this complex conundrum, but it is a proven “best practice” to mitigate the damage.9 Multiple studies have conclusively established its potential to significantly reduce a number of negative outcomes, such as hospitalization, incarceration, suicide, violence, and crime, among the hardest-to-treat people with severe mental illness,10 and also save money in the process.11
Looking only at the research, one might expect AOT to be practiced universally by local mental health systems overwhelmed by the disproportionate needs of patients who lack insight. But AOT remains controversial in the mental health field by virtue of its involuntariness. Just as there is resistance in some corners to the involuntary hospitalization of anyone not posing an imminent risk of violence or suicide, some are offended by the notion of “coercing” an individual to follow a treatment plan, however compelling the need.
To some extent, the controversy around AOT is rooted in misunderstanding of its aims. In just about any other legal context, the point of a court order is to disincentivize undesirable acts or omissions through the court’s power to punish (with jail, fines, or both) those who show it “contempt.” This association is understandably troubling when considered in the context of a person whose errant behavior is driven by illness. But in fact, the threat of punishment plays no role in AOT. Violation of an AOT court order typically leads to nothing more than a re-evaluation of the person’s need to be committed to hospital care. And such commitment can only occur if the person is found to meet the ordinary criteria for hospitalization, just as it would in the absence of an AOT order.
This might reasonably cause some to wonder what, then, is the point of AOT. Experience suggests that when practiced correctly, AOT works for three fundamental reasons. First, AOT motivates patients by impressing upon them, through the symbolic power of the judge as an authority figure, the seriousness of their need to comply with treatment. This is sometimes called the “black robe effect.” Second, AOT alerts treatment providers that the court identifies a patient as high risk and expects a commensurate level of care. Third, AOT typically provides close monitoring of patients so that non-adherence is detected early and addressed before deterioration makes it harder to intervene effectively.
Because AOT is not merely an approach to outpatient treatment but also a type of court procedure, it requires state law to specifically authorize it. By 2013, all but five states (Connecticut, Maryland, Massachusetts, New Mexico, and Tennessee) had enacted such laws.
However, it must be acknowledged that the existing state AOT laws on the books in 45 states and Washington D.C. vary greatly in quality. Some are carefully conceived and used to great effect in pockets, if not throughout their states. Others give scant indication of how AOT might function in practice, or they include insurmountable barriers to practical use.
States’ approaches to defining the eligibility criteria for AOT take two basic forms.
One approach is to treat AOT and inpatient commitment as entirely separate entities, with distinct criteria. This is helpful in states where inpatient standards emphasize current “dangerousness” and thus do not apply to a person who is presently under treatment and not dangerous but has a history of treatment non-adherence with bad outcomes. In such states, having a need-for-treatment standard specifically for AOT facilitates the use of AOT in hospital discharge planning. This is a point at which the person is entitled to release because he no longer meets inpatient criteria yet may not be fully stable and in any case remains at great risk of recidivating.
The second, more common approach is to treat commitment as a unitary process with a single set of criteria and allow the court to choose the type of commitment — inpatient or outpatient — that is the least restrictive alternative meeting the person’s particular needs at the time. This makes it easy for a court to change the nature of the person’s commitment as circumstances evolve.
Either of these approaches to AOT criteria can work well if structured properly. We caution, however, that under the unitary approach, the key to effectiveness is to ensure that at least one of the alternative statutory commitment standards is a need-for-treatment standard that can potentially apply to a presently stable yet insight-deficient hospital dischargee. The more narrow a conception of dangerousness an AOT eligibility standard imposes, the more it becomes a barrier to any use of AOT at all.
Aside from the eligibility criteria, there are various features that define an effective AOT law. In surveying the states’ various approaches to AOT, one is struck by the contrast between states that specify precisely how the process of treating a person in the community under court order is supposed to work and those that leave the process out. We generally favor the more detailed approach as it provides a “how-to manual” for mental health officials looking to implement AOT.
One helpful feature is a provision to explain how the treatment adherence of the AOT patient will be monitored and specifying what should happen if such monitoring reveals that the patient is not succeeding under the order. The consequences of non-adherence should be oriented toward re-evaluating whether outpatient placement is still appropriate to meet the person’s needs.
We particularly appreciate AOT laws that do not merely order the patient to comply with prescribed treatment, but also in some way press the mental health system to actually provide the treatment ordered. The state that goes furthest in this direction is New York, where judges directly order mental health officials to provide essential services. But this works mainly because New York law requires every county to establish an AOT program. In other states, it is a trickier proposition because local mental health officials are free to forgo AOT altogether, as they are more likely to do if they perceive that opening the process imposes new obligations upon them. A sensible and more common approach to ensuring quality treatment is to require the local mental health system to develop a detailed treatment plan and identify providers prior to the AOT hearing. The plan is then explained at the hearing and incorporated into the court’s order.
For the very reason explained in the preceding discussion of inpatient commitment, we favor AOT laws that allow families and friends of people in need — not just mental health officials — to petition the court.
Finally, laws that empower courts to order AOT of longer duration are generally more effective. The maximum length of AOT orders varies by state, ranging from 2 months to 1 year. Although in most cases these orders are renewable if the need persists, forcing mental health officials to return frequently to court discourages renewal and leads to shorter periods of AOT. Research indicates that AOT of 6 months or shorter in duration is not as successful as AOT of longer periods in leading to gains that are sustained after AOT is terminated.12 An optimal AOT law allows for commitment of up to 1 year. Limiting the maximum duration to 6 months is inadequate, and a 3-month limit is significantly worse.
We have presented herein our conception of an ideal statutory scheme for mental health civil commitments. We are keenly aware that the realization of this vision would get us only halfway to an ideal practice of involuntary treatment. For timely and adequate treatment to become the norm, state mental health systems must be equipped to make full use of the authority that such laws would bestow upon them. For one thing, that will require state legislatures to provide ample funding for hospital beds, appropriate medications, community-based services, and intensive case management. For another, it will require a shift in thinking among leadership and staff of state and local mental health systems toward unapologetic embrace of the strategic and judicious use of involuntary treatment. On both fronts — fiscal and philosophical — our systems have long been under siege. But there can be little hope of reversing these trends so long as state treatment laws continue to give deference to disastrous “personal choices” driven by psychosis.
- Testa M, West SG. Civil commitment in the United States. Psychiatry (Edgmont). 2010;7(10):30–40.
- O’Connor v Donaldson, 42 US 563, 576 (1975).
- O’Connor v Donaldson, 42 US 573 (1975).
- O’Connor v Donaldson, 42 US 572, fn.9 (1975).
- State v. Dennis H, 647 NW2d 851 (Wisc Super Ct. 2002).
- State v. Labelle D, 728 P2d 138 (Wash Super Ct. 1986).
- O’Connor v Donaldson, 42 U.S. 563, 572, fn.9
- Segal SP. Civil commitment law, mental health services, and US homicide rates. Soc Psychiatry Psychiatr Epidemiol. 2012;47(9):1449–1458. doi:10.1007/s00127-011-0450-0 [CrossRef]
- National Institute of Justice. Assisted out-patient treatment. Available at: http://www.crimesolutions.gov/ProgramDetails.aspx?ID=228. Accessed on February 5, 2014.
- Swartz MS. Introduction to the special section on assisted outpatient treatment in New York State. Psychiatr Serv. 2010;61(10):967–969. doi:10.1176/appi.ps.61.10.967 [CrossRef]
- Swanson J, Van Dorn RA, Swartz MS, et al. (2013). The cost of assisted outpatient treatment: can it save states money?Am J Psychiatry. 2013;170(12):1423–1432. doi:10.1176/appi.ajp.2013.12091152 [CrossRef]
- Van Dorn R, Swanson JW, Swartz MS, et al. Continuing medication and hospitalization outcomes after assisted outpatient treatment in New York. Psychiatr Serv. 2010;61(10):982–987. doi:10.1176/appi.ps.61.10.982 [CrossRef]