Last year 227 boys who committed crimes were incarcerated in the Connecticut Juvenile Training School, the state's maximum-security facility in Middletown for juvenile offenders. They lived there for months before it was determined when they would be released. Their stay averaged nine months.

This uncertainty came to an end in October.

Now, the Department of Children and Families, which runs the training school, gives every inmate sent there either a six- or 12-month sentence from the day they arrive, though they can be released somewhat earlier or later depending on behavior.

The change was made to correct a sentencing system that many felt had become subjective, as well as to respond to research showing that long incarceration times actually increase the chances an inmate will return to jail.

"Before kids and families didn't know when they were going home. Now they know the expectations," said William Rosenbeck, the superintendent of the jail that houses around 130 boys on any given day. "It's providing the opportunity for kids to work very hard here and let them earn a couple of days a month to get out early."

Rosenbeck said the new policy will help get children home or into a less-secure residential facility more quickly.

While child advocates applaud the goal of getting children out of a locked facility and into the community sooner, there are some concerns that the new protocols are too standardized.

"It's going to a purely criminal model. There's nothing personal or individualized about this. It's similar to a zero tolerance model," Abby Anderson, executive director of the Connecticut Juvenile Justice Alliance, said of the standard sentences for offenders.

DCF made the changes in sentencing policy after the number of children incarcerated at the training school reached a 10-year high, a new jail to house girls was opened, and the living conditions of an incarcerated transgender teen drew criticism from child advocates.

It also came after researchers at Georgetown University's Center of Juvenile Justice Reform reported that children were locked up at the training school for too long and too frequently.

"The current pattern is likely counterproductive," researchers wrote in a 72-page analysis of the state's juvenile justice system.

A discharge should be related to a child's risk to the community and behavior improvement, said Anderson of the Bridgeport-based advocacy group. However, the Georgetown researchers pointed out that approach is not an option in Connecticut since DCF has no valid way to measure an offender's risk. These juvenile offenders are committed to DCF when their crimes are not serious enough to land them in the adult system.

Purpose of jail: 'Gain self-control'

Research shows that how long children are confined matters.

"Longer stays in juvenile institutions do not reduce recidivism, and some youth who had the lowest offending levels reported committing more crimes after being incarcerated," Edward Mulvey, a psychologist at the University of Pittsburgh School of Medicine, found in a study of delinquent youths. "Youth who stayed in institutions longer showed little or no decrease in their rates of re-arrest compared with those with shorter stays."

"Shorter stays are always better for my clients. We always want to see our clients out in the community earlier," said James Connolly of the state's public defenders' office, who represents children after they are convicted and sent to DCF for placement.

State officials say another goal of the new standard approach is to eliminate potential disparities in how youths are treated, based on who their parole officer is or which office handles their case. The state Office of Policy and Management reports that of the 227 boys who lived at the training school last year, 200 were black, Hispanic or another minority - well above the 30 percent they represent in the general state population.

This sentencing protocol "treats every man that comes in here fairly," said Rosenbeck.

Under the new policy, every inmate has a meeting to create a treatment plan within his first 30 days at the training school. Inmates are told that participation in clinical and educational programs will get them released up to one month early for those serving six months and two months early for those serving 12 months.

"Ultimately, secure confinement should only be long enough for a youth to stabilize behavior and gain self-control over his/her daily regime. Thus, such a system would support more rapid returns to community," the new sentencing protocols read.

"Discharge dates can only be negatively impacted though major behavioral events or new charges."

Sarah Eagan, the state's child advocate, said while she supports the goal of shortening stays that the new protocols aims to fix, she has mixed feelings.

"What is the purpose of CJTS: Is it correction or treatment?" she said of the locked facility. "It would be very challenging to be correctional and therapeutic."

Many advocates say the only reason to lock up these juveniles is if they pose a risk to the community, pointing out the significance of the Georgetown finding that DCF has no valid risk-assessment instrument.

"Any risk assessment instrument that has not been validated on the specific population for which it is intended is problematic," the researchers concluded in June. Using such results "to make individual case decisions, especially where there is a liberty interest at stake, is akin to medical malpractice or off-label pharmaceutical usage and arguably misses the mark in any reasonable test of fairness."

Officials have since said they are working to set up a system to better identify a child's risk to the community.

"We have a risk assessment. It is outdated. We are working to get it resolved," said DCF's Rosenbeck during an interview.

Half return to jail

When a child leaves Connecticut's juvenile jail, the odds of returning to custody are one in two, the Georgetown researchers found.

"There are a lot of kids going back and forth. Something is not quite right," said Martha Stone, the executive director of the Center for Children's Advocacy and a lawyer who represents some of the boys who have been sent to the facility.

The largest driver of incarceration is violation of the conditions of probation or parole.

For Hakeem A., this meant that he was sent to the training school for smoking marijuana and leaving his house without his parents' permission - both violations of his probation.

When parole officers and their supervisors were questioned about the high rate of incarceration for non-criminal infractions, Georgetown researchers were told that, "CJTS is the best we have."

"The supervisors and managers informed us there were no less-restrictive alternatives available in the current system," the researchers said. "Violations say as much about the quality of parole services as they do about the parolee."

Inmates "on occasion" remained jailed "because of a lack of an appropriate and viable community-based or home placement," the researchers reported.

DCF officials told state legislators last month that they are aware of the issues and are working to improve their services.

This includes an "approval process that goes up to the commissioner" for placement at the training school. They also report that, on any given day, they have 40 vacant beds in residential facilities ready for children who leave the training school for a less-restrictive environment.

But Stone, who won a class-action lawsuit against the state in 1997 over the way the state's juvenile justice system handled adolescents before they are convicted, said these vacancies are proof "there is still a mismatch of services for the needs."

Fewer residential placements

Of the children committed to DCF for breaking the law, the percent that are sent to residential facilities has steadily declined over the last several years, the DCF reports. Meanwhile, the percent living at home or at the training school has somewhat increased.

The shift away from using congregate facilities has angered providers.

"If you don't get referrals, you don't get paid," said Ron Cretaro, the executive director of the Connecticut Association of Nonprofits, said of the vacancies.

He estimates hundreds of workers have lost their jobs since DCF moved to place fewer children in residential facilities.

"With the closing of group care beds/programs for juvenile justice, there has not been corresponding development of new non-residential services in the community," he said.

The Georgetown researchers pointed out that the services available for those sent to live in the community are not always available or adequate, and the agency has no way to know which programs succeed at preventing future bad behavior.

"It appears that Connecticut does not have a well-developed quality assurance system, either for its parole supervision services or for its residential confinement system," the Georgetown researchers concluded.

The issues have captured the attention of Rep. Toni Walker, the chairwoman of the legislature's powerful budget-writing committee.

"What we have to do is find out what's really working and shift those dollars to those areas," she told DCF officials last month. "We can no longer continue with [incarcerating kids]. This has got to change."

Jacqueline Rabe Thomas is a reporter for The Connecticut Mirror (www.ctmirror.org). Copyright 2014 © The Connecticut Mirror.

jrabe@ctmirror.org.