Cuyahoga wasn't the only place with this problem, and the study is upfront about that. As of 2024, SAKI's national LODNA initiative had identified roughly 220,000 people across the country who owe a DNA sample. Texas identified more than 43,000 between 1995 and 2020 and, after four years of active collection, had closed only about 12 percent of the gap. Alaska identified 22,000 and closed 425 cases in its first year. A survey of Idaho law enforcement agencies found such widespread confusion about which agency was even responsible for collecting owed DNA that the report's authors described the state's approach as fragmented and unguided, with no agreed-upon standard for who does it or when.

Ohio's own statute has evolved in stages. A 1996 law first required DNA collection from people convicted of felonies. A 2010 expansion extended that requirement to felony arrestees, and since July 2011, the law has allowed collection at booking, during jail or prison intake, at sentencing, at post-conviction hearings, or while someone remains under community supervision such as probation or parole. The catch, and it's the one that shapes almost everything else in this study, is that once someone is no longer under the criminal justice system's supervision, Ohio law doesn't allow the sample to be collected at all. Miss the window, and the person becomes untouchable unless, and until, they reenter the system on a new charge.

Cuyahoga County's turnaround didn't come from a single fix. It came from three mechanisms working together over time, and the study is specific about what each one actually contributed.

The most visible was the 2017 internal policy that turned DNA status into a checkpoint at every stage a case touches the court. At arraignment, prosecutors are required to confirm whether a defendant already has a DNA sample in CODIS, checking a line in the Ohio Law Enforcement Gateway (OHLEG) called the DNA statement, which reads either “A DNA profile is in the Ohio DNA Offender Databank (CODIS)” or “A DNA profile is not.” If it's missing, they request a court-ordered collection. Before accepting a plea, prosecutors have to check again. At sentencing, they ask the judge to determine DNA status and order collection if needed. And at post-conviction, probation violation, or any later hearing, the same check happens once more. Intake staff and prosecutors are also required to log DNA status directly into the office's case management system before a case can close.

None of that works, though, without a reliable way to know who still owes a sample in the first place, and that's the second mechanism, the one that did the most quiet work. Starting in mid-2018, the Ohio Attorney General's Office, which oversees the state's CODIS lab, began sending the Task Force quarterly updates on every census member's DNA status. The Task Force sent over a list of everyone tied to a felony case, pulled from its own case management system, along with identifying information: name, date of birth, Social Security number, state ID number. The Attorney General's Office matched that list against CODIS and reported back who was confirmed and who still wasn't. The Prosecutor's Office then built the ability to flag a defendant's DNA status directly inside its own case management system, so the assigned prosecutor saw an alert when a sample was still owed, rather than having to remember to look. The study credits this quarterly matching process with surfacing the majority of the county's confirmed CODIS entries, more than any single courtroom checkpoint, simply because it was checking on everyone continuously rather than waiting for a specific court date to trigger a look.

The third mechanism came first, chronologically. Early in the initiative, the Task Force requested “sweeps” of the county jail and probation department, collecting DNA on the spot from anyone still under supervision who owed a sample. That netted 623 samples: 77 from the jail, 546 from probationers already in the system. It was a head start, not a long-term fix. It only reached people already in custody or under supervision at that specific moment, which is exactly the population the ongoing policy and quarterly matching were built to keep catching afterward.

Of the 3,802 census members confirmed in CODIS by August 2024, two specific pushes account for 1,264 of them directly: 623 from the early sweeps and 641 more from OHLEG checks run in the day or two before a new felony arraignment. The study is candid about the rest. Some had actually been in CODIS for a while, but a data-matching error in their name, date of birth, Social Security number, or state ID number had hidden that fact until a later quarterly update caught and corrected it. Others were confirmed for reasons the researchers couldn't fully pin down, such as a new charge in a different Ohio county triggering routine collection there. The quarterly matching process is how the county found out about all of it. It's the reason anyone could say, with confidence, at any point over eight years, exactly how big the remaining gap still was.

Two probation audits, conducted two years apart, are where the county's progress becomes hardest to argue with, because they measure the policy's effect across the whole system rather than any one mechanism in isolation. In September 2021, researchers pulled a list of everyone on felony probation and checked how many still lacked a CODIS sample: 2.2 percent, or about 1 in 45. By November 2023, that number had fallen to 0.2 percent, about 1 in 455. The study's own language for that shift doesn't hedge: “a near-elimination of missed DNA sample collections among felony probationers.” Before the initiative began, the county had estimated that 20 to 50 percent of felony arrestees were falling through the cracks entirely.

In July 2023, the study's lead researcher, Rachel Lovell, and the prosecutor supervising the initiative sat in on roughly 15 arraignments to see the policy work in real time. Each case took two to three minutes. When a defendant lacked a CODIS sample, the court ordered one on the spot, and deputies walked the person into a small holding room connected to the courtroom, informally known as the DNA sample collection room, to collect it. Lovell asked deputies what happens if someone refuses. They said they're permitted to compel a felony arrestee to submit, though none of them had ever personally had to. She didn't witness a single refusal. But she did find two real, unglamorous problems: paperwork completed at the collection point sometimes had insufficient or improperly inked fingerprints, and jail intake staff weren't always sure whether an incoming charge was a felony or a misdemeanor, a distinction that determines whether DNA should be collected at booking at all.

Not every strategy the county tried worked equally well. The Task Force also flagged roughly 1,000 census members connected to the most serious offenses, including first-degree felonies and sexually based crimes. Of those, 801 also still owed a DNA sample as of 2022, and the Task Force reasoned that actively pursuing this narrower group made sense given the stakes. Only 17 of those 801 turned out to have both an owed sample and a pending case in the county, and by the end of the grant period, just 59 people, 7.4 percent of the cohort, had actually gained a CODIS entry as a result. The lesson the researchers draw from that isn't that the effort was wasted. It's that chasing the worst offenders individually is far less efficient than a policy and a data pipeline that check everyone, automatically, every time a case moves through the system.

Even the failures were rare and specific. Over an 18-month window, only 33 DNA samples collected at arraignment or jail intake had to be resubmitted because they failed to meet CODIS entry standards, out of nearly 6,000 total collection requests during the same period. The most common reasons were mundane: not enough DNA on the swab, a missing or poorly inked fingerprint, no name on the sample. And despite early concern about overlap with Ohio's sex offender registry, only 1.8 percent of the census turned out to be currently registered, since Ohio, unlike Texas or Washington, doesn't require DNA collection as a condition of registration in the first place.

The study doesn't pretend the remaining gap is easy to close. An estimated 21.5 percent of census members without a confirmed sample are likely deceased or living out of state, both effectively unreachable under a statute that only permits collection while someone remains under supervision. Another 10.4 percent are “unverified,” meaning their name, date of birth, Social Security number, and state ID number couldn't be reliably matched across two separate databases, a data infrastructure problem the authors flag as its own unresolved issue.

There's a warning built into the trend line itself. The number of new census members entering CODIS declined every year of the initiative, dropping 70.4 percent from the first year of tracking to the last. Fewer people were reentering the system on new charges over time, which meant fewer new chances to catch a missed sample after the fact. The paper's recommendation follows directly from that curve: collect it right the first time, because the odds of a second chance get worse with every year that passes.

The researchers are careful to note that Cuyahoga County's exact playbook won't transfer everywhere without adjustment. Every state's DNA collection statute is different, and how much real-time access law enforcement and court personnel have to a person's DNA status varies just as much. But several of the specific tools behind this county's numbers are worth any jurisdiction's attention.

The first is a real-time, statewide DNA status lookup, the kind OHLEG already gave Ohio before this initiative even started. Texas built something similar into its own law enforcement database: a color-coded flag that marks a person's record the moment DNA is owed, letting any officer check status as a matter of routine rather than guesswork. That kind of system solves a second, less visible problem at the same time. Some crime laboratories, the study notes, report receiving thousands of duplicate samples because there was no way to confirm a sample had already been collected somewhere else in the state. A shared, searchable status field fixes both problems with one piece of infrastructure.

The second is the quarterly data-sharing arrangement itself, which Cuyahoga County and the Ohio Attorney General's Office formalized through a memorandum of understanding. Without that written agreement, the quarterly matching process that surfaced most of this county's confirmed CODIS entries wouldn't exist. Any jurisdiction attempting something similar needs that same kind of formal commitment between its prosecutor's office and whichever agency administers its state DNA lab, not an informal understanding that depends on which two people happen to hold those jobs at a given time.

The study also passes along two smaller lessons other jurisdictions might not think to ask about. Colleagues at other LODNA sites recommended working with credit bureaus, rather than state motor vehicle or Social Security records, to track down whether a census member had died or moved out of state, since Cuyahoga's own attempts to automate that search through government databases didn't work. And in states where the law allows it, sheriff's offices already have a built-in reason to make repeated contact with registered sex offenders through mandatory address verification, a touchpoint that could double as a DNA status check, even though Ohio doesn't currently tie DNA collection to registry status. Emerging rapid DNA technology, capable of generating a usable DNA sample in as little as 90 minutes, may also cut down on the kind of collection and duplication problems this study documented.