The initial decision to detain or release an accused person, either with or without financial and other conditions, is made by a judicial officer—usually pretty quickly, and often with a minimum of information. But this quick decision can have serious long-term consequences, and not just for individual detainees.

So why not take a second look?

Eight Safety and Justice Challenge sites have instituted procedures for revisiting initial detention and bail decisions on a routine basis. From a system point of view, it’s about “saving bed days”—finding ways to minimize unnecessary pretrial holding. For individual defendants, though, second looks like these can save a lot more—jobs, families, even lives. Some of the second-look approaches being taken by Safety and Justice Challenge jurisdictions are simple; some are highly complex and collaborative. Lucas County, Ohio’s is both.

It’s called the Population Review Team: about ten people as a rule, including representatives of the Toledo Legal Aid Society (which represents indigent defendants) and the City of Toledo Law Department (which prosecutes them), along with the Lucas County Sheriff’s Office, Pretrial Services, the Criminal Justice Coordinating Council, and other local agencies and providers. They meet every week for a couple of hours, combing page by page through a fat printout showing charges, criminal histories, risk assessment scores and other details on Lucas County’s entire pretrial jail population.

They’re looking for deals: good candidates for a modified bond, release on appropriate conditions, or expedited case resolution. They concentrate on defendants who were recommended for pretrial release by the county’s risk instrument, the PSA Court, but who for one reason or other were held anyway, as well as defendants charged with low-level offenses but not recommended for release because of past criminal activity or failures to appear. They don’t just rely on what’s in the printout: each of the participants has access to sources of additional information about defendants—affidavits and incident reports on file, needs assessment results, community service histories, active cases in neighboring court systems, etc.

Sean McNulty, the Toledo Legal Aid Society’s Chief Public Defender and a frequent participant in Population Review Team meetings, says that “if we can predict the likely outcome of a case at the next court event, then this review provides us with an opportunity to reach that outcome in two days, not two weeks. It helps the system become more efficient, but it also affords our clients a chance to be released from custody earlier-—thereby decreasing the likelihood that their lives will further destabilize while they remain incarcerated.”

When the meeting is over and the team’s release recommendations have been recorded, the defender representative typically heads straight across the street to meet with clients in the Lucas County Jail, explain the Population Review project, and present the proposed offers. If clients agree, their cases come before a Toledo Municipal Court judge at an expedited hearing a day or two later, and they are released.

Regular deep dives into the county’s pretrial population have given Population Review Team members a chance to spot trends, detect bottlenecks and snags, and come up with simple policy fixes for chronic problems. The project has engendered useful habits of interagency collaboration and information sharing, and created a sense that there are certain basic goals and values—efficiency, proportionality, common sense—that everyone in the justice system ought to own.

Lucas County has plans to expand its Population Review Team approach in 2018, hoping to get similar benefits with a new target population: people being held for felonies who have an identified need for mental health or substance abuse treatment. The hope for the expansion—led by the Toledo Legal Aid Society and dubbed the Opportunity Project—is that a connection to services at the pretrial stage can improve court and other outcomes for a behavioral health population.