Try 1 month for 99¢
041318 PSA

A employee with the 7th Judicial District Department of Correctional Services inputs data to update the Scott County Sheriffs web-site Wednesday, April 11, 2018. Scott is one of four counties in Iowa to start using the Public Safety Assessment, which takes into account a number of factors (such as age, prior criminal history, prior failure to appear in court) to determine whether a defendant should be released, placed on supervision, or detained while their case is pending.

Scott County is using a new tool to help judges decide whether a criminal defendant should be in jail or out on release, with or without supervision, while their case is pending.

The Public Safety Assessment, implemented March 12, uses evidence-based risk factors to determine the likelihood that a defendant who is released before trial will fail to return for future court hearings or commit another crime or violent crime, once released.

Ultimately, the goal is to prevent defendants who otherwise would be eligible for bond to be held because they don’t have the money to post it, Assistant Scott County Attorney Steve Berger said.

“I think that’s what mostly attracted our (Iowa) Supreme Court to this process,” he said. “They view that it is inherently unfair that people can’t get out of jail simply because they don’t have money,” he said.

“We have a problem with disproportionate minority confinement,” Wayln McCulloh, district director for the 7th Judicial District Department of Correctional Services, said. “We know that persons of a lower socio-economic stratum in society are more likely to be incarcerated.”

Scott County Sheriff Tim Lane said that the jail population has dropped by about 10 percent, but there is not enough data to determine if it is because of the new PSA tool.  

Although it’s still new, the PSA tool is causing concern for some in the court system and in law enforcement.

Some say the tool doesn’t put enough emphasis on the current charge, especially if it’s violent, and that prior juvenile adjudications are not considered.

How does it work?

The PSA was developed by the non-profit, Houston-based Laura and John Arnold Foundation after researchers analyzed more than 1.5 million cases from more than 300 jurisdictions across the country.

Polk County became the first in Iowa to implement the PSA on Jan. 16, Steve Davis, communications director for the Iowa Judicial Branch, said.

Scott and Woodbury counties implemented it last month. Linn County will be the next to implement it, Davis said.

Berger said he anticipates that it will eventually be implemented throughout the entire Seventh Judicial District, which includes Cedar, Clinton, Jackson and Muscatine counties.

Polk and Linn counties will participate in a Harvard University study, where PSA will be given only to judges in half of all new initial appearances for the next 18 months.

Researchers then will compare the PSA and non-PSA groups to see if the assessment was successful at making better informed decisions on bail.

Implementation of the PSA is a statewide collaboration between the Iowa Judicial Branch, the Iowa Department of Corrections, the State Public Defender and the County Attorney’s Association.

The assessment tool looks at nine factors: age at the time of the arrest; whether the current offense is violent; whether there is a pending charge at the time of the offense; prior misdemeanor convictions; prior felony convictions; prior violent convictions; prior failures to appear in the past two years; prior failures to appear older than two years; and prior sentences of incarceration.

It does not take into account a defendant’s race, gender, income, education, home address, drug use history, family status, marital status, national origin, employment or religion.

The PSA then produces a score from 1 (low risk) to 6 (high risk) that represents the likelihood that a defendant who is released before trial will commit a new crime or fail to appear in court.

It also flags defendants who pose an elevated risk of committing a crime of violence if they are released before trial.

Judges then can use the PSA to determine whether a defendant should be released on a recognizance bond, meaning no money down; released on supervision with or without posting bond; or remain in jail.

The judge is not bound to follow that recommendation.

“Public safety is always foremost in every judge's mind, but we also know that pretrial incarceration can cause real harm, such as loss of employment, economic hardship, interruption of education or training, inability to care for children, loss of place of residence, and impairment of health or injury because of neglected medical issues,” Marlita Greve, chief judge of the Seventh Judicial District, said.

“Judges will continue to exercise their independent judicial discretion to make these difficult decisions with the additional help of the PSA tool while keeping in mind the defendant's interest in release, the defendant's likelihood of appearing for court, and the community's public safety.”

To release or not release

Berger said all Class A felonies, as well as charges of second-degree murder, first-degree robbery, and second-degree sexual abuse are offenses that automatically get a “release not recommended” indication on the PSA.

“People have been extradited from another state; we’re not too fond of releasing them again,” Berger said. “If their current charge is escape, we’re not too fond of releasing them.

“We’re also not too fond of releasing people with federal detainers, interstate detainers, intrastate detainers, because we have an interest in getting our cases done.”

If the defendant doesn’t face any of those offenses, then the PSA applies the failure to appear and new criminal activity scores to determine a preliminary recommendation release type and corresponding conditions level.

Offenses such as domestic abuse stalking, no contact order violation, any person-to-person sex crime, arson, operating while intoxicated third, vehicular homicide, or offenses involving the use of a dangerous weapon or failure to appear on any of these charges will bump up the recommended supervised release type by one level.

“What we’ve seen so far is that age is a very significant factor in determining this violence flag,” Berger said.

Over the last several years, Scott County has seen an uptick in juveniles committing vehicle thefts and other violent crimes.

There is no bond system in the juvenile court system. However, juveniles who are charged as adults are subject to bond.

One such defendant is 17-year-old Darion Lemont Thomas of Davenport, who was arrested March 20 following a drive-by-shooting and high speed chase (he was not the driver). Police also charged him in connection with two earlier other drive-by-shootings.

The PSA filed in his cases did not find that a violence flag existed and scored him at a “1” on the failure to appear scale and a “2” on the new criminal activity scale.

It also recommended that he be released on his own recognizance and placed on the lowest level of supervision.

Court records show that Associate Court Judge Christine Dalton instead set a $10,000 cash or surety bond in all three of his cases and. If he posts bond, he will be on the maximum level of supervision.

That includes court reminders, weekly face-to-face contact, curfew, home visits, supervision contract and electronic monitoring if appropriate.

Berger said there is a combination of factors in the PSA algorithm that triggers the violence flag.

“What I’ve seen so far is that if they are 20 and under and have a present violent offense and have any prior adult conviction, it seems to trigger that flag, particularly if they have a violent offense,” he said. “But, if they have no prior convictions, it’s not going to be triggered by current violent offenses.”

That’s where the judge’s discretion comes in, he said.

“Sometimes, you have judges that are seeing juveniles that they’ve seen in juvenile court, so they know exactly who they are dealing with and they are well equipped to deal with them accordingly,” he said.

Still, it’s a concern for Assistant Scott County Attorney Amy DeVine.

“What mortifies me the most is this does not seem to address the current offense that much,” she said.

“It’s about the protection of the community and I feel like that, quite frankly, is where that’s falling short.”

DeVine said that because the PSA tool is so new, the county attorney’s office has not been able to keep track yet of how many defendants have had their pretrial release revoked because they didn’t show up for court or have committed a new offense.

And the process of filing those petitions and then serving them is going to create more work for the court system, she said. If it’s a case where the person has to be extradited from another jurisdiction, the county has to shoulder the cost.

“It’s so new that it’s hard to see how yet how much that’s increasing, but, I mean, it’s going to increase,” she said.


McCulloh said a defendants who is detained before they are tried are “more likely to be found guilty of a crime and, if found guilty of a crime, is more likely to go to prison because the person is not able to maintain employment.”

That’s what this tool is hoping to better address, he said.

Lisa Chapman, a probation and parole supervisor with the Seventh Judicial District, said the new tool also allows for defendants to be under supervision while their case is pending.

“That allows for more public safety as opposed to them just being able to go through a bondsman and bond out,” she said.

Prior to the implementation of the PSA, the Department of Corrections was monitoring about a dozen people who were on pretrial release, McCulloh said.

That number has climbed to more than 100, on top of the people they are monitoring who are on probation and parole, he said.

McCulloh said the department has had to reallocate resources, such as moving one probation officer to Scott County, to keep up with increased work load.

“That’s something that causes sleepless nights, trying to balance the expectations with the resources,” he said.

Two probation officers are doing the bulk of the work to supervise those on pretrial release, Chapman said. Two other probation officers also are in the rotation, she said.

Chapman said there are a lot of defendants on the lowest level, PM1, that doesn’t use as many staff resources. However, there also a lot of defendants on the PM2 and PM3 levels that requires “enough bodies to assign them to.”

Still, McCulloh and Chapman see the value of the PSA.

“I think what’s important to point out … our 1960s previous process for pretrial (release) was not necessarily evidence-based practices,” Chapman said.

“I had many years of interviewing people in the jail in the morning before court and a lot of times they are under the influence … a lot of their behaviors in the morning when they’re still maybe drunk or high wouldn’t necessarily determine how they are normally,” she said. “So, to be utilizing that information to decide whether to hold them would not be evidence-based practices.”

Beginning at 5 a.m. each day, Department of Corrections staff look up each defendant’s background through crime databases, such as the Iowa Corrections Offender Network, and inputs the data into the computer.

Chapman said there is a section on the report that allows the probation officer to include comments, such as whether the defendant is already on probation so that the judge has all the information he or she needs to make an informed decision.

DeVine said county attorneys also may weigh in if they have more information, such as the possibility that more charges could be coming.

“I think we’ve been doing that a little bit more than what we used to in the past,” she said.

McCulloh said the ICON system is helpful to look up a defendant’s criminal history. But for a defendant who lives in another jurisdiction, like Chicago, “we’re at the mercy of what the officials in Cook County have entered into the system.”

“And many times, those records are incomplete,” he said.

Chapman said that probation officers completing a pre-sentence report prior to a defendant’s sentencing have contacts in Cook County to help solve that problem.

“We’ve shared information with the pretrial unit who are conduct the PSA, so we’re able to get more information than we have previously been able to get,” she said. “But, on weekends it’s going to be hard because our contacts and people that provide us more thorough records aren’t available a lot of times on the weekend,” she said.

Law enforcement weighs in

Lane said that one of the things that worries him about the PSA is the lack of face-to-face encounters with an inmate when they are first brought in.

“There is no direct interview, so things like mental illness, and intoxication, and addiction are not even remotely being considered,” he said.

Lane said another concern is where the offender is from, especially if they live outside of Iowa.

If that person decides not to come back for trial, Lane said, it falls to his department to go get that person, and it is money that has to be spent out of the Sheriff’s Department’s general fund. The more money Lane has to spend out of the general fund, the less money he has to buy equipment for his deputies.

Also, Lane said, an offender can have an extensive juvenile record, but once they turn 18, they start with a clean slate in terms of the Public Safety Assessment, which means even a violent juvenile record is not being considered.

Already, he said, there have been people who have been arrested in connection with some of the stolen cars and shots fired calls in Davenport, and they’ve been released and gone out and offended again.

“The Public Safety Assessment is supposed to produce a score that tells the likelihood of a new crime or failure to appear,” Lane said. “You can already see that that isn’t working.”

One case that popped out is that of Timothy Brian O’Haver, 42, of Davenport, who was arrested March 19 by Bettendorf police in connection with a methamphetamine trafficking investigation.

O’Haver also has an extensive record of drug convictions dating to 2003 and has been in and out of prison since at least 2006. He was last released from the Iowa Department of Corrections on June 20, 2016.

A PSA completed in his case noted that he did not have any violent convictions on his record and that he had three instances of failing to appear in court. The preliminary recommendation, according to the PSA, was to place him on the lowest level of supervision – court date reminders and one phone contact a month.

Even though he was charged with a felony that could land him in prison for up to 25 years, O’Haver was granted pre-trial release with supervision after his initial bond had been set at $100,000.

Magistrate Christine Frederick, however, bumped up his level of pretrial supervision to include monthly face-to-face contact and ordered him to obtain a substance abuse evaluation.

That one case alone has Iowa Quad-City law enforcement raising their eyebrows.

Davenport Police Chief Paul Sikorski even is wondering when the distribution of narcotics not a public safety risk.

As for Davenport's own cases, Sikorski said he and his command staff are educating themselves on the new pre-trial release process, “and we’ve been in communication with the County Attorney and we’ll continue to provide our input into the new process.”

“Anything that we’re concerned with we will communicate with the County Attorney and they can communicate with the judges as they see fit,” he said.

Lane said he already has seen a reduction in the number of inmates at the jail.

“As a Sheriff, and as the Sheriff’s Office, we do what the court system tells us,” Lane said. “Whoever is remanded to the jail, we hold, and whoever is ordered released, we release. We don’t petition the court to do otherwise.”

But this new system of the Department of Corrections monitoring people who have been placed on pre-trial release is unfunded and no new employees have been hired to contend with the extra work.

“The idea is to get as many people out of jail as possible, but where do you get the personnel to monitor all of these people?” Lane said.