Editor’s note: Reporters from the Topeka Capital-Journal and KHI News Service collaborated for a six-month exploration of how the state’s legal system deals with people with mental illness. This is one of the stories in a four-day series.
Who: A research team in Florida.
What: A study that found it costs more to have people with mental illness in the criminal justice system than it does to treat them effectively outside the system.
When: Published in 2013.
Quote: “Medication was really sort of the issue. That’s the real fork in the road for folks.” — John Petrila, chairman of the University of South Florida’s department of health policy and management.
John Petrila already knew that a breakdown of the mental health system was increasing prison populations when he became chairman of the University of South Florida’s department of health policy and management in 2012.
He knew it from his years working as the top attorney for the New York Office of Mental Health and a department chair at the Florida Mental Health Institute.
And he knew that groups with combined expertise in psychiatry and law enforcement had done nationwide studies that came to the same conclusion.
In addition to the obvious humanitarian implications of a system breakdown, Petrila knew there was plenty of anecdote-based speculation that states were spending more to incarcerate people with mental illness than they would to treat them.
But little had been done to scientifically quantify that theory until Petrila and a team of researchers published a study called “Effects of Outpatient Treatment on Risk of Arrest of Adults with Serious Mental Illness and Associated Costs” in a May 2013 edition of the journal Psychiatric Services.
The name of the study was long, but its conclusion was rather simple: It costs more to have people with mental illness in the criminal justice system than it does to treat them effectively outside the system.
“There’s not a lot (of other studies) out there, but the couple that are suggest the same thing,” Petrila said.
His team’s study provides a blueprint for Kansas researchers who want to see if the same holds true here.
Florida officials granted Petrila’s team access to seven years of claims data for about 4,000 Medicaid recipients who had been hospitalized for schizophrenia or bipolar disorder. Petrila’s team then compared that with public arrest data.
The researchers found that adults with severe mental illness who received routine outpatient treatment like medication and counseling after their hospitalizations were less likely to be arrested.
They also found that the state’s costs were lower for those who were not arrested, even though they were using more mental health services. Petrila said that was the key point for policymakers, who often make the bottom line a priority.
“I thought that if you could tell them something about utilization and cost, that those were measures that regardless of which party you were dealing with, folks would be interested in,” Petrila said.
Petrila’s report only provides a snapshot of Florida from 2005 to 2012. The cost of corrections and psychiatric care — especially prescription medications — fluctuates with time and geography.
A similar analysis could be done for Kansas, Petrila said, but researchers would face a few obstacles.
For one thing, the information his team used came from a traditional fee-for-service Medicaid system in which each treatment produces an individual claim that the team could use as a data point.
States are increasingly moving to managed care Medicaid models, like KanCare in Kansas, that provide flat per-member payments to private insurance companies. In those models, individual claims information often is held by the companies and not the state, making it more difficult to access.
“That does create an issue for sure,” Petrila said.
But Cassie Sparks, a spokeswoman for the Kansas Department of Health and Environment, said the state agency still has access to individual claims data that could be used for Kansas-based research.
Then it’s a matter of having the resources to clean up the data and identify the same individuals in separate datasets.
That’s harder than it sounds, Petrila said, and takes people with unique skillsets that command high salaries.
Funding for the research team was always an issue.
“We did a lot of stuff on the cheap,” Petrila said. “I was better at getting access to data than I was at finding funding to support the analyses. I never had enough money to do the analyses with the level of depth or sophistication the data would allow us to.”
The analysis the team was able to do showed one thing emphatically: Medication adherence is key to keeping people with severe mental illness out of the criminal justice system.
“Medication was really sort of the issue,” Petrila said. “That’s the real fork in the road for folks.”
Keeping people medicated
Keeping people with severe and persistent mental illness from going off their medications can be difficult, especially if they’re adults.
In some states, parents or friends have no legal recourse in those situations short of getting a judge to commit the person to a psychiatric facility against his or her will. That generally requires law enforcement and the judiciary to determine that the person is a danger to himself or herself and others. Some families say it’s hard to meet that standard before it’s too late to help a loved one.
The Treatment Advocacy Center, a Virginia-based nonprofit that focuses on “eliminating legal and other barriers to the timely and effective treatment of severe mental illness,” has pressed for another solution: assisted outpatient treatment, or AOT.
Under state AOT laws, courts can order that people with severe mental illness comply with their prescribed medication or therapy regimen. If they fail to do so, they can be committed to an inpatient facility.
Nearly every state has an AOT law, but they vary widely and some are rarely used. The Treatment Advocacy Center graded the Kansas AOT law a C- in part because it requires the same “danger to themselves or others legal standard” as an involuntary hospital commitment.
By contrast, New York’s law sets a standard that is easier to meet: that the person being considered for AOT is deemed “unlikely to survive safely in the community without supervision.”
The Treatment Advocacy Center grades the New York statute, known as Kendra’s Law, an A+. One of the first AOT laws in the country, it was enacted in 1999 and named for Kendra Webdale, who was killed when a man with an untreated mental illness pushed her in front of a New York City subway train.
Luis Marcos, now a professor at New York University, helped spearhead the law change as commissioner of New York City’s Department of Mental Health at the time.
In a phone interview, Marcos said Webdale’s death was just one incident in a long line of tragedies that convinced civic leaders like Mayor Ed Koch that the city had to do more to ensure people with mental illness were getting appropriate treatment.
“Things don’t happen all of a sudden,” Marcos said. “There’s an increasing concern, at least that’s the way I experienced it in the ’80s and ’90s. … It does take leadership on the part of influential people — whether it’s a mayor, state legislators or other people.”
AOT does not require courts to order medication — and some forms of the law exclude court-ordered medication entirely in favor of counseling therapy only. Still, it has proven controversial both in terms of its implications for the rights of people with mental illness and its effectiveness.
The New York Civil Liberties Union opposed Kendra’s Law, and other groups affiliated with the American Civil Liberties Union have successfully challenged similar laws in other states.
But an evaluation of several studies found that AOT “results in no significant difference in service use, social functioning or quality of life compared with standard care.”
As the review also noted, “People receiving compulsory community treatment were, however, less likely to be victims of violent or non-violent crime. It is unclear whether this benefit is due to the intensity of treatment or its compulsory nature.”
The review authors estimated that courts would have to use AOT 238 times to prevent one arrest.