Assisted Outpatient Treatment Enters the Mainstream

Apr 29, 2017

Here: common misconceptions about assisted outpatient treatment--and how this tool can help selected mentally ill patients who are most difficult to treat.

After lounging on the doorstep of respectability for the past decade, assisted outpatient treatment (AOT) has finally entered the mainstream.

In September, the Substance Abuse and Mental Health Services Administration (SAMHSA) announced more than $13 million in federal funding for AOT demonstration grants (Table 1). Three months later, as part of the 21st Century Cures Act, Congress extended the AOT grant program until 2022, increased the funding, and made AOT programs eligible for Department of Justice funding. This was part of a larger recognition by Congress that the US mental health system was in dire need of reform, led by Congressman Tim Murphy (R-PA) and Senator John Cornyn (R-TX). Thus, AOT has now been officially endorsed by SAMHSA and Congress, joining such groups as the National Sheriffs’ Association, the International Association of Chiefs of Police, the National Alliance on Mental Illness, and the American Psychiatric Association.

Understanding AOT

AOT is court-supervised treatment of severe mental illness within the community. It is a 2-way commitment that requires mental health systems to serve participants at the same time it commits participants to adhere to their treatment plans. It is similar in intent to conditional release, mental health courts, and conservatorship-all of which seek to improve the person’s adherence to a treatment plan and quality of life.

The specific procedures vary by state, but generally, to be a candidate for AOT, a person must meet clinical and legal criteria, such as a history of repeated hospitalizations or arrest resulting from nonadherence to treatment. For example, in New York the person must, among other factors, have a history of treatment nonadherence that has led to 2 or more psychiatric hospitalizations or incarcerations within the past 3 years or 1 or more acts of serious violence in the past 4 years; be unlikely to voluntarily participate in outpatient treatment; and be in need of treatment to prevent a relapse that would likely result in serious harm to self or others. Court petitions for AOT are usually filed by the county mental health department, which also arranges for development of a treatment plan and delivery of services (Figure).

Joy Torres illustrates the life-changing capabilities of AOT. Plagued by delusions and auditory hallucinations for years, she refused care, was hospitalized multiple times, jailed once, and lost her job and children. She recalls being “handcuffed and shackled on the cold, wet cement of a tiny (jail) cell with only a tin can for a bathroom,” “running around naked,” “sleeping in dumpsters to avoid being raped,” and “being treated like an animal because no one is willing to help you when you are too ill to help yourself.” Finally, she was court-ordered to adhere to a treatment plan under Laura’s Law-California’s AOT program. She says this gave her the right to “sleep on a bed, go to school, make friends, love my family, and to get and stay well.” She added, “Should I ever become sick again, please give me Laura’s Law.”1

AOT was reportedly first used in 1972 at St. Elizabeth’s Hospital in Washington, DC, by an agreement between Dr. Roger Peele, a psychiatrist, and Harry Fulton, the chief public defender. A patient was ready for discharge but still needed medication supervision, so they decided to commit her to outpatient treatment in place of additional inpatient treatment. This was consistent with a previous court ruling that mandated patient placement in the “least restrictive alternative.” By 1984, AOT had been used at the hospital on 293 patients with very favorable results and had spread to a few other states. However, it was not widely known until 1999 when New York passed an AOT law called Kendra’s Law, named after Kendra Webdale, a young woman pushed under a subway train by a man with grossly undertreated schizophrenia.

AOT has spread in recent years but is still underutilized. New York, and more recently New Jersey, are using it most extensively. Ohio has excellent programs in Butler and Summit Counties and is extending it to other counties. In California, Laura’s Law was pioneered in Nevada County and has now been implemented in counties covering two-thirds of the state’s 38 million population. Seminole County in Florida and Oakland County in Michigan also have excellent programs. All states have laws permitting AOT except for Massachusetts, Connecticut, Maryland, and Tennessee, but in at least 15 states it is virtually never used.


Clinical effectiveness

The clinical effectiveness of AOT has been clearly established. Multiple studies have shown that it dramatically decreases the number of psychiatric admissions, hospital days, arrests, incarcerations, days of homelessness, and alcohol and drug abuse. People on AOT are more adherent to medication and, probably associated with this, are victimized less often.2 Four studies have also reported a decrease in violent behavior associated with the use of AOT (Table 2).3-6

Those who deny the effectiveness of AOT have offered 2 arguments. First, they claim that AOT’s effectiveness is based mostly on before-and-after studies of patients placed on AOT, rather than by studies that randomly assign patients to AOT or a control. But it can be argued that random assignment would be both unethical and legally impossible, as individuals eligible for AOT have already met criteria evidencing either dangerousness or need for treatment and a history of being unable to participate in such care voluntarily. Second, they claim that studies in other countries have not shown community treatment orders to be effective. However, community treatment orders are dissimilar to AOT because they fail to utilize a court order and thus cannot benefit from a judge’s authority-often deemed the “black robe effect.” The design of community treatment order studies themselves has also been subject to criticism.

These academic arguments diminish as data on AOT’s effectiveness accumulate. For example, in San Francisco where AOT was implemented in 2015 after 4 years of debate, the first-year report noted that almost everyone who met AOT criteria accepted voluntary treatment, leading to fewer emergency department visits and “a small but helpful step toward addressing” the problem of homelessness.


As impressive as the clinical effectiveness of AOT-and likely more salient to policymakers-is its demonstrated cost-effectiveness. In North Carolina, according to the researchers, “outpatient commitment [AOT] of six months or more, combined with provision of outpatient services, appeared to result in cost savings of 40%.”7 In New York, in the first year after AOT was implemented, costs declined 43% in New York City and 49% in 5 counties, and even more in the second year.8

Related content:

4 Myths About Assisted Outpatient Treatment

In a study in Summit County, Ohio, “annualized aggregate costs per participant in AOT were found by researchers to have declined 50% in the period before and after participation.”9 In a small study in Nevada County, California, the cost reduction for each AOT participant the first year was 80%-$1.81 saved for every $1.00 invested in the program.10

Much of these cost savings come from improved medication adherence and decreased hospitalizations. In a separate study, researchers estimated that improved medication adherence for individuals with schizophrenia could save Medicaid $1580 per patient, or $3.3 billion per year.11

Mainstream acceptance

Given the clearly established benefits of AOT, why has it taken so long for the program to gain mainstream acceptance? In some states, AOT statutes are poorly written or contain illogical criteria. In Pennsylvania, for example, eligibility criteria for AOT mirror the inpatient standard, requiring a judge to determine a person to be both a “clear and present danger” and eligible for community treatment.

In other communities, a lack of understanding may hinder its use. Judges are often surprised that AOT statutes do not allow for incarceration or direct placement into inpatient care as a noncompliance sanction. In many communities, the delay in adoption reflects the realities of a struggling mental health system, in which providers are reluctant or unable to address the upfront costs of program implementation, despite evidence that AOT saves money over time.

One source of AOT opposition is concerns about coercion or a restriction of individual liberties. AOT serves as a lightning rod for opposition to all forms of involuntary psychiatric treatment, despite AOT being no more coercive than conditional release or mental health courts-both of which are widely accepted.

Some common misconceptions include:

1) AOT is forced treatment infringing on civil liberties. This argument has been soundly rejected in the courts. In the 2004 case Matter of KL, the plaintiff challenged Kendra’s Law, arguing it authorized “forced treatment” without adequate due process. The New York Court of Appeals, widely regarded as a champion of civil liberties, unanimously held that AOT is not forced treatment. The court noted Kendra’s Law prohibits forcible administration of medicine and only allows re-hospitalization if the usual commitment criteria are met. Thus, the court found the law relies on “the compulsion generally felt by law abiding citizens to obey court directives,” not force.12

2) People will voluntarily seek help if we make psychiatric services more widely available. This ignores the key factor addressed by AOT-anosognosia. Approximately half of individuals with serious mental illness experience anosognosia, damage to the parts of the brain used to think about ourselves, confirmed by at least 20 studies.13 People who think nothing is wrong with them-the signature symptom of anosognosia-have no reason to seek treatment. It is simply illogical to assume system improvements will draw in these patients, and the experience of many communities unfortunately bears this out.

3) AOT drives patients away from seeking treatment. There is simply no defensible evidence that this is true. In fact, when individuals receiving AOT under Kendra’s Law in New York were asked, 81% said AOT had helped them get and stay well.14

4) Why spend resources on patients who don’t want to be treated when so many who do want to be treated are not getting services? Research shows that the population served by AOT disproportionately uses public resources. Consequently, research has found that mental health services, including voluntary services, can actually be expanded when AOT is used because of savings associated with decreased resource utilization by a small population.15 More broadly, addressing the outsized costs associated with high utilization is a growing concern for public health and policymakers.


AOT is here to stay, not as a cure-all but rather as another tool to help treat selected patients who are otherwise the most difficult to treat. Dr. Darold Treffert summarized the rationale for treatment approaches such as AOT nicely:

It is not “freedom” to be wandering the streets, severely mentally ill, deteriorating and getting warmth from a steam grate or food from a garbage can; that’s abandonment. And it is not “liberty” to be in a padded jail cell instead of a hospital, hallucinating and delusional, without treatment because that is all the law will allow.16


Dr. Torrey is a research psychiatrist who specializes in schizophrenia and bipolar disorder. He is Founder of the Treatment Advocacy Center and Associate Director of the Stanley Medical Research Institute, and he is Professor of Psychiatry at the Uniformed Services University of the Health Sciences in Bethesda, MD. Mr. Snook is Executive Director of the Treatment Advocacy Center.

The authors report no conflicts of interest concerning the subject matter of this article.


1. Treatment Advocacy Center. If I get sick again, please give me Laura’s Law. April 2014. Accessed March 4, 2017.

2. Torrey EF. The Insanity Offense. New York: WW Norton; 2008:190-192.

3. Swanson JW, Swartz MS, Borum R, et al. Involuntary out-patient commitment and reduction of violent behavior in persons with severe mental illness. Br J Psychiatry. 2000;176:224-231.

4. New York State Office of Mental Health. Kendra’s Law: Final Report on the Status of Assisted Outpatient Treatment. March 2005. Accessed March 4, 2017.

5. Phelan JC, Sinkewicz M, Castille DM, et al. Effectiveness and outcomes of assisted outpatient treatment in New York State. Psychiatr Serv. 2010;61: 137-143.

6. Link BG, Epperson MW, Perron BE, et al. Arrest outcome associated with outpatient commitment in New York State. Psychiatr Serv. 2011;62:504-508.

7. Swanson JW, Swartz MS. Can states implement involuntary outpatient commitment within existing state budgets? Psychiatr Serv. 2013;64:7-9.

8. Swanson JW, Van Dorn RA, Swartz MS, et al. The cost of assisted outpatient treatment: can it save states money? Am J Psychiatry. 2013;170: 1423-1431.

9. Health Management Associates. State and Community Considerations for Demonstrating the Cost Effectiveness of AOT Services: Final Report. February 2015. Accessed March 4, 2017.

10. Tsai G. Assisted outpatient treatment: preventive, recovery-based care for the most seriously mentally ill. Am J Psychiatry Resid J. 2012;7:16-18.

11. Predmore ZS, Mattke S, Horvitz-Lennon M. Improving antipsychotic adherence among patients with schizophrenia: savings for states. Psychiatr Serv. 2015. Accessed March 4, 2017.

12. Matter of KL, 1 NY3d 362; 2004.

13. Treatment Advocacy Center. Serious Mental Illness and Anosognosia. Accessed March 4, 2017.

14. Kendra’s Law: Final Report on the Status of Assisted Outpatient Treatment. Albany, NY: New York State Office of Mental Health; March 2005:20-21.

15. Swanson JW, Van Dorn RA, Swartz MS, et al. Robbing Peter to pay Paul: did New York State’s outpatient commitment program crowd out voluntary service recipients? Psychiatr Serv. 2010;61:988-995.

16. Treffert DA. 1995 Wisconsin Act 292: finally, the fifth standard. Wisconsin Med J. 1996;95:537-540. ❒