Turning JDAI’s Focus to the Deep End of the Juvenile Justice System

In the nine years since the city of St. Louis embraced deten­tion reform, the pop­u­la­tion in its juve­nile deten­tion facil­i­ty has plum­met­ed, and the con­tin­u­um of deten­tion alter­na­tive pro­grams has swelled. A new ethos has enveloped all cor­ners of the city’s juve­nile sys­tem ded­i­cat­ed to reduc­ing the deten­tion pop­u­la­tion as much as pos­si­ble con­sis­tent with pub­lic safety.

But how are St. Louis youth far­ing at lat­er stages of the juve­nile court process? And how well is the city doing in min­i­miz­ing the use of cor­rec­tion­al com­mit­ments and oth­er res­i­den­tial place­ments for youth who do not pose a sig­nif­i­cant threat to pub­lic safety?

Begin­ning in late 2012, St. Louis lead­ers set out to dig deep and find out, join­ing the first cohort of sites in the Casey Foundation’s new effort to broad­en the focus of JDAI to the dis­po­si­tion­al (or ​“deep”) end of juve­nile jus­tice. St. Louis and the oth­er five pilot sites — Bernalil­lo Coun­ty (New Mex­i­co), Jef­fer­son Parish (Louisiana), Lucas Coun­ty (Ohio), Mar­i­on Coun­ty (Indi­ana) and Washoe Coun­ty (Neva­da) — spent all of 2013 mobi­liz­ing deep end study groups and con­duct­ing an exten­sive quan­ti­ta­tive and qual­i­ta­tive assess­ment to bet­ter under­stand local dis­po­si­tion­al trends and to iden­ti­fy oppor­tu­ni­ties for safe­ly reduc­ing placements.

The ulti­mate impact of their efforts will not be known for some time — the sites are just now devel­op­ing their action plans. But the assess­ment process revealed a num­ber of impor­tant lessons.

First, all six sites found that many youth who pose min­i­mal risks to pub­lic safe­ty are nonethe­less being removed from their homes and placed into cor­rec­tion­al insti­tu­tions or oth­er res­i­den­tial facil­i­ties. Thanks in part to their work in JDAI, all six sites had already reduced com­mit­ments sig­nif­i­cant­ly before launch­ing their deep end efforts. How­ev­er, the assess­ment process made clear that many oppor­tu­ni­ties remain to safe­ly reduce res­i­den­tial con­fine­ment below cur­rent levels.

Sec­ond, all of the par­tic­i­pat­ing sites have found that unnec­es­sary place­ments are occur­ring both at ini­tial dis­po­si­tion and through vio­la­tions of pro­ba­tion. Every site iden­ti­fied room for improve­ment in the use of risk assess­ments to deter­mine ini­tial dis­po­si­tions, and every site found that a sub­stan­tial share of place­ments — at least 30 per­cent — stemmed from pro­ba­tion vio­la­tions rather than new offenses.

Third, the sites’ efforts to reduce deep end place­ments rely on many of the same tools and val­ues used in JDAI — includ­ing col­lab­o­ra­tion, data analy­sis, objec­tive deci­sion mak­ing and an inten­tion­al focus on racial/​ethnic equi­ty — but it is more com­plex. Com­pared with deten­tion reform, sites are find­ing that reduc­ing dis­po­si­tion­al place­ments requires more elab­o­rate and sophis­ti­cat­ed data analy­ses, a top-to-bot­tom review of risk and needs assess­ment pro­ce­dures, far greater atten­tion to front­line pro­ba­tion prac­tice and more inten­sive efforts to engage and sup­port families.

Despite the chal­lenges, sites are mak­ing progress and build­ing momentum.

“It’s been a lot of work. Your brain is con­stant­ly work­ing,” reports Cathy Hore­jes, Chief Deputy Juve­nile Offi­cer in the St. Louis City pro­ba­tion depart­ment. ​“But it’s just a nat­ur­al flow,” Hore­jes says. ​“Once you’ve improved the front end of your sys­tem [through JDAI], I think you need to crit­i­cal­ly look at the back end. You need to ensure that you’re not just look­ing at the data, but that you’re tak­ing the next step and doing some­thing with the data. That’s what the deep end focus has done for us.”

• • •

By Decem­ber 2013, when teams from the six deep end sites con­vened for an inter-site work­shop, all had com­plet­ed their assess­ments, boiled down the results into Pow­er­Point pre­sen­ta­tions and shared the find­ings with their local collaboratives.

In St. Louis, the data revealed that half of all place­ments stemmed from mis­de­meanors or pro­ba­tion vio­la­tions. Just 13 per­cent of the place­ments were for vio­lent felony offens­es. Indeed, the data showed that St. Louis youth had a greater like­li­hood of out-of-home place­ment for a pro­ba­tion vio­la­tion than a vio­lent felony.

Equal­ly pow­er­ful, says Cathy Hore­jes, have been the insights gleaned from sur­veys and inter­views with youth, fam­i­lies, pub­lic defend­ers, ser­vice providers, pro­ba­tion staff and oth­ers. Many state­ments were pos­i­tive, but oth­ers were crit­i­cal, and they made a pow­er­ful impres­sion on her staff of 35 pro­ba­tion offi­cers. ​“To hear what came out of people’s mouths was like, ​‘wow,’” Hore­jes says. ​“Those were learn­ing moments.”

In Lucas Coun­ty, Ohio, the deep end data analy­sis revealed two eye-open­ing trends. First, while the over­all num­ber of com­mit­ments and res­i­den­tial place­ments has fall­en in recent years, the decline has lagged an even larg­er drop in arrests and refer­rals. In fact, for youth who are arrest­ed, the odds of being placed in a res­i­den­tial facil­i­ty have grown by more than half since 2008.

“When we saw that the com­mit­ments were ris­ing in rela­tion to arrests and refer­rals, it was a big wow,” reports Rachael Gard­ner, the local deep end coor­di­na­tor. ​“It pro­vid­ed a dif­fer­ent frame.”

The Lucas Coun­ty assess­ment also revealed a strik­ing racial dis­par­i­ty in place­ments. Where­as the pro­por­tion of African-Amer­i­can youth holds rough­ly steady (about 60 per­cent) at most deci­sion points — includ­ing arrests, deten­tion admis­sions, peti­tions, adju­di­ca­tions and pro­ba­tion place­ments — 92 per­cent of youth sent to res­i­den­tial treat­ment cen­ters and 94 per­cent of youth sent to state train­ing schools are African American.

“We didn’t expect that our sys­tem wouldn’t have dis­pro­por­tion­al­i­ty and dis­par­i­ty,” says Gard­ner, ​“but see­ing that jump at the point of out-of-home place­ment, that was surprising.”

• • •

In ear­ly 2014, the pilot deep end juris­dic­tions are devis­ing action plans and tak­ing ini­tial steps to address the issues they have iden­ti­fied. Review­ing risk and needs assess­ments is a focus for all sites — both the design of the tools, and the way they are used to inform the deci­sion-mak­ing process.

In St. Louis, local lead­ers are reex­am­in­ing their risk assess­ment tool. Until now, Hore­jes explains, staff have seen the tools ​“as a task to com­plete, not some­thing to help them in real deci­sion making.”

To lim­it place­ments stem­ming from pro­ba­tion vio­la­tions, St. Louis has begun requir­ing pro­ba­tion offi­cers to con­vene a meet­ing with the youth, fam­i­ly, pro­ba­tion lead­ers and oth­er part­ners before fil­ing a pro­ba­tion vio­la­tion. In May 2014, Hore­jes hired a for­mer child pro­tec­tion work­er to train pro­ba­tion staff on how to con­duct these team deci­sion-mak­ing meet­ings effectively.

St. Louis is already show­ing a dra­mat­ic drop in place­ments. Just 54 city youth were com­mit­ted to state cus­tody in 2013, down from 91 in 2012 (and more than 100 each year from 2008 to 2011). Hore­jes attrib­ut­es most of the decline to the arrival of a new pre­sid­ing juve­nile court judge, David Mason, a cham­pi­on of deep end reform. In addi­tion, Hore­jes says, through train­ing ses­sions on fam­i­ly involve­ment and through their involve­ment in the assess­ment process, the deep end efforts are begin­ning to influ­ence how pro­ba­tion staff work with court involved youth.

In Lucas Coun­ty, local lead­ers are work­ing with the Nation­al Coun­cil on Crime and Delin­quen­cy to design a new dis­po­si­tion­al matrix to guide ini­tial place­ment deci­sions. In addi­tion, Lucas Coun­ty has opened a new day treat­ment pro­gram for pro­ba­tion youth exhibit­ing con­tin­u­ing con­duct prob­lems who might oth­er­wise be placed in a res­i­den­tial facility.

“The work is fan­tas­tic,” says Gard­ner. ​“It real­ly lifts up what’s next for your juris­dic­tion, and it’s led to some real­ly impor­tant con­ver­sa­tions.” Asked what advice she would offer to oth­er JDAI juris­dic­tions con­sid­er­ing whether to apply and become a deep end site, Gard­ner responds: ​“Come on. Jump in with us. Jump into the deep end, because it’s total­ly worth it.”