Judges in Baltimore County are testing a new scoring system to help determine whether to release defendants awaiting trial — and if so, how closely to monitor them.

The change is part of a pilot program that was launched in early December. Officials say the aim is to reduce the number of people held in jail before trial without failing to ensure public safety.

“The purpose of all this is to really provide the court with a very good and serious picture of the defendants in custody,” said Randy Mentzell, management assistant with the county’s Department of Corrections, which oversees pretrial services. “Our goal is to give the court the best information, the most information so they can make a decision — a good decision.”

A risk assessment applied to defendants takes into account factors including the seriousness of the offense, the severity of prior convictions and previous failures to appear in court.

Some factors can help a person’s score — for instance, if they’ve had no arrests within the past year or are employed.

Defendants’ scores are given to judges for bail review hearings.

While judges have long considered many of the same elements that are now on the score sheet, the new method is meant to keep decisions consistent across courtrooms, said Chief Administrative Judge Kathleen Cox.

“Before, you looked intuitively at certain things,” Cox said. “[This] scores that, and it evens it out across different judges.”

In developing the pilot program, Baltimore County looked closely at similar programs in Montgomery and St. Mary’s counties, Cox said. In addition, judges met over months with representatives of the county detention center in Towson and the state’s attorney’s and public defender’s offices to work out details.

The scoring method recommends release for those who score a 5 or below, and detention for those scoring 14 and above.

For those in the middle, there can be release with three levels of supervision, conducted by the corrections department.

The lowest level requires phone check-ins with a corrections employee. The second level requires in-person check-ins and random drug testing, and may include referrals to services such as counseling. The third, most stringent level is home detention. People in home detention pay a $75 weekly fee, corrections officials said. The county can waive the fee if someone can’t afford it.

Judges do not have to follow the recommendations.

“At the end of the day, it’s the judge’s discretion,” Cox said.

County State’s Attorney Scott D. Shellenberger said someone could have a low score, but still be detained for safety reasons.

For instance, a man charged in a domestic-violence incident could have a job and no prior arrests, giving him a score that would seemingly indicate he should be released without supervision. But if the defendant has threatened to kill his wife when he gets out, the judge can “override” the score.

“Public safety is always an override,” Shellenberger said. “That is one of the positives from a prosecutor’s point of view.”

Nearly 57 percent of the county jail’s current population is awaiting trial, Mentzell said. The jail had an average population of 1,171 this year, which is down more than 15 percent since 2012.

Maryland’spretrial services and bail system have been in the spotlight over the past year.

In 2016, Attorney General Brian E. Frosh issued a letter to lawmakers concluding that holding defendants in jail because they can’t afford to pay cash bail likely would be found unconstitutional.

Then, the Maryland Court of Appeals adopted new rules that said before using cash bail, judges should look to other ways to ensure someone appears for trial. The rule change took effect in July.

In the first three weeks of Baltimore County’s pilot program, judges have placed a total of 149 people in supervised release, according to corrections officials, with more than half placed in the second level of supervision.

Donald Zaremba, district public defender for Baltimore County, said effective supervision before trial can address public safety concerns in many cases.

The new scores provide “some objective criteria” that can aid the decision-making process.

“People are still presumed to be innocent, and taking somebody’s liberty away before there has been a trial is a very weighty decision,” Zaremba said. “There should be options that are consistent with the norm, which in this country is liberty.”

Only a few weeks in, those involved with the program say it is too early to draw conclusions. They plan to monitor data over the coming months to determine rates of recidivism before trial and failure to appear among those in the program, as well as the effect on the size of the jail population.

“The question for us is, if they’re under supervision, are they doing things we ordered?” Cox said. “Do they show up to court when they are supposed to?”

alisonk@baltsun.com

twitter.com/aliknez