County Associations and State Governments: Working Together Toward Smart Justice

County Associations and State Governments: Working Together Toward Smart Justice By Michael Thompson October 24, 2013

National non-profit, non-partisan membership association of state government officials Represents all three branches of state government Provides practical, nonpartisan advice informed by the best available evidence Council of State Governments Justice Center 2

90 staff in 4 offices Seattle, WA New York, NY Austin, TX Bethesda, MD Council of State Governments Justice Center 3

Major Initiatives Underway Regarding Youth, Mental Health, and Reentry School Discipline Consensus Project Criminal Justice / Mental Health Consensus Project National Reentry Resource Center Council of State Governments Justice Center 4

Justice Reinvestment in 18 States NH NH VT NV ID KS WI IN MI OH WV PA RI CT AZ OK NC TX HI Council of State Governments Justice Center 5

County data and stakeholder input enriches analysis and justice reinvestment policies Council of State Governments 6 STATE FINDING POLICY 42% of the regional jail population is pretrial Require the use of a pretrial risk screen within 3 days of booking to identify those with greatest risk of flight 25% of prison admissions are misdemeanor offenders with a 3- month average length of stay Misdemeanor offenders may be housed in a jail if: The sheriff voluntarily accepts Bed space exists Reimbursed by new state funds Probation violators spend lengthy periods in jail awaiting hearings Create a 30-day cap in statute for probationers awaiting violation hearings

Overview Understand the latest science and research Improve outcomes for people with mental illnesses in contact with justice system Reduce re-offense rates for people released from jail Council of State Governments Justice Center 7

Recidivism across the states Council of State Governments Justice Center 8

Knowledge on Improving Criminal Justice Outcomes Has Increased Dramatically Over the Last 20 Years Academics and practitioners have contributed to this growing body of research Council of State Governments Justice Center 9

What are 3-4 things I need to know about what works to reduce recidivism? 1. Focus on individuals most likely to reoffend 2. Base programs on science and ensure quality 3. Implement effective community supervision policies and practices 4. Apply place-based strategies Just as helpful as pointing out commonly made mistakes are the cutting-edge practices identified in the report. The report, in other words, should be required reading. -The Washington Post, February 27, 2011 Council of State Governments Justice Center 10

1. Focus on individuals most likely to reoffend Who? Identify and Focus on Higher-Risk Offenders Without Risk Assessment With Risk Assessment Risk of Re-offending LOW 10% re-arrested MODERATE 35% re-arrested HIGH 70% re-arrested Council of State Governments Justice Center 11

1. Focus on individuals most likely to reoffend TITLE (Ohio) Low Risk + 3 % Average Difference in Recidivism by Risk for Halfway House Offenders Moderate Risk - 6 % High Risk - 14 % Adopted a common set of risk assessment instruments across the state s criminal justice system. Ensured that program placement is driven by risk assessment score. *Presentation by Latessa, What Works and What Doesn t in Reducing Recidivism: Applying the Principles of Effective Intervention to Offender Reentry Council of State Governments Justice Center 12

1. Focus on individuals most likely to reoffend What? Housing Target the Factors that Evidence Shows Are Most Central to Criminal Behavior Employment/ Education Thinking Antisocial The Big Four (impacting these are the major drivers to reducing criminal behavior) Past Criminality* Criminal Behavior Peers Family Higher-risk offenders are likely to have more of the Big Four. Substance Use Personality Programs targeting these needs can significantly lower recidivism rates Leisure * Past criminality cannot be changed. Council of State Governments Justice Center 13

2. Base programs on science and ensure quality After Who and What Are Answered, Supervision and Programming Should Be Well Targeted Risk of Re-offending LOW 10% re-arrested MODERATE 35% re-arrested HIGH 70% re-arrested Low Supervision/ Program Intensity Moderate Supervision/ Program Intensity High Supervision/ Program Intensity Council of State Governments Justice Center 14

2. Base programs on science and ensure quality Ensure Funded Programs Are Reducing Recidivism Council of State Governments Justice Center 15

2. Base programs on science and ensure quality How Well? Ensure Programs Are High Quality and Properly Implemented Is the program based on principles demonstrated to be effective? Program Effectiveness (reduced recidivism) Is program matched with appropriate client population? Is program implemented as designed? Are program staff properly trained? Is performance tracked and measured against expectations? Council of State Governments Justice Center 16

2. Base programs on science and ensure quality Where and How Treatment Is Delivered Impacts the Degree of Recidivism Reduction Impact of Treatment Intervention on Recidivism Rates Drug Treatment in Prison Drug Treatment in the Community Supervision with Risk Need + Responsivity -17% -24% -30% Supervision, with effective RNR principles, yields the biggest recidivism reduction Source: Lee, S., Aos, S., Drake, E., Pennucci, A., Miller, M., & Anderson, L. (2012). Return on investment: Evidence-based options to improve statewide outcomes, April 2012 (Document No. 12-04-1201). Olympia: Washington State Institute for Public Policy. Source: Latessa, Lovins, and Smith, Follow-up Evaluation of Ohio s Community Based Correctional Facilities, Outcome Study, February 2010 Council of State Governments Justice Center 17

3. Implement effective community supervision policies and practices Elements of Effective Supervision Dosage/Intensity Focus supervision officer time and program resources on the highest-risk offenders. Consistency Use a graduated range of sanctions and incentives to guide specific type of response to violations and compliance. Swiftness Enable officers to respond meaningfully to violations without delay or time-consuming processes. Cost-effectiveness Prioritize the most expensive, restrictive sanctions for offenders committing the most serious violations. Council of State Governments Justice Center 18

4. Apply placebased strategies Prison Admissions Hotspots Arizona, 2004 60% of the State s prison population comes from and returns to the Phoenix-Mesa metropolitan area. Council of State Governments Justice Center 19

4. Apply placebased strategies Prison Admissions Hotspots Maricopa County, 2004 A single neighborhood in Phoenix is home to 1% of the state s total population but 6.5% of the state s prison population South Mountain Zip Code 85041 Prison Admissions = 31.8 per 1000 adults Jail Bookings = 96.5 per 1000 adults Probation = 25.1 per 1000 adults

4. Apply placebased strategies Prison Expenditures in Dollars Maricopa County, 2004 North Mountain Alhambra Glendale Encanto Camelback East Maryvale Estrella Central City Laveen South Mountain $1.1 Million $1.8 Million $1.6 Million Within high expenditure neighborhoods there are numerous, smaller area, million dollar block groups

4. Apply placebased strategies High Density of Probationers in South Phoenix Council of State Governments Justice Center 22

Reducing Criminal Behavior Requires Focusing on Risk, Need, and Responsivity Traditional Approach Evidence-Based Practices Supervise everyone the same way Risk Assess risk of recidivism and focus supervision on the highest-risk offenders Assign programs that feel or seem effective Need Prioritize programs addressing the needs most associated with recidivism Deliver programs the same way to every offender Responsivity Deliver programs based on offender learning style, motivation, and/or circumstances Council of State Governments Justice Center 23

Alcohol and Drug Use Disorders: Significant Factor in Jail and Prisons Percent of Population 60 50 40 30 20 10 8 % 47 % 54 % 44 % 53 % Alcohol use disorder (Includes alcohol abuse and dependence) Drug use disorder (Includes drug abuse and dependence) 0 2 % Household Jail State Prison Source: Abrams & Teplin (2010) Council of State Governments Justice Center 25

Prevalence of Serious Mental Illness and Co-Occurring Disorders in Jail Populations General Population Jail Population 95% 5% 83% 17% 72% 28% Serious Mental Illness No Serious Mental Illness Serious Mental Illness No Serious Mental Illness Co-Occurring Substance Use Disorder No Co-Occurring Substance Use Disorder Council of State Governments Justice Center 26

County Officials: Jails are the wrong place to treat mental illnesses Our jails are increasingly a place of last resort for offenders who are mentally ill. Even as the department s total inmate population continues to fall, this group is unable to get out or stay out. New York City jails, like jails across the country, have disproportionately high numbers of inmates whose mental health needs present unique challenges. -- Commissioner Dora Schriro, Department of Corrections, New York, NY -- Deputy Mayor Gibbs, Health and Human Services, New York, NY I would welcome the chance to take all of our mentally ill and medically challenged inmates and put them somewhere they could get programming, but I haven t heard anyone stepping up to do that. -- Sheriff David Mahoney, Dane County, WI In every city and state I have visited, the jails have become the de facto mental institutions there are not enough resources out there to care for them [mentally ill]. -- Sheriff Tom Dart, Cook County, IL Council of State Governments Justice Center 27

Case Study: Florida Counties Ocala Star-Banner, November 2007 Lakeland Ledger, February 2001 St. Petersburg Times, November Council 2007 of State Governments Justice Center 28

Flowchart of select events in the Orange County Criminal Justice System http://static.nicic.gov/library/022134.pdf Council of State Governments Justice Center 29

Identification of mental health pretrial release program participants http://static.nicic.gov/library/022134.pdf

Entry into mental health pretrial release program 53,091 7,260 http://static.nicic.gov/library/022134.pdf 36 percent of the 1,101 detainees referred to the program did not participate because they had already been released from jail. Others did not participate because they did not meet the clinical criteria, declined to participate, had a history of violence, were not an Orange County resident, or had been referred to another program. Detainees who participated in the program were released from jail and put under the MHPTR program s supervision.

Completion of mental health pretrial release program 53,091 7,260 http://static.nicic.gov/library/022134.pdf Offenders are returned to jail on an outstanding warrant, another arrest/charge, or a revocation due to noncompliance with the conditions of the MHPTR program prior to the disposition of current charges Offenders have satisfied their charges with the court system without further arrests. Time in program for successful participants varies depending on when participants satisfy their charges without further arrests

Not all Substance Use Disorders are Alike Abstinence Dependence The Substance Use Disorder Continuum Council of State Governments Justice Center 33

Not All Mental Illnesses Are Alike Mental Illnesses In the General Population Diagnosable mental disorders 16% Serious mental disorders 5% Severe mental disorders 2.5% Council of State Governments Justice Center 34

Criminal Justice Risk on a Continuum Risk/Needs Assessment 101: Science Reveals New Tools to Manage Offenders, http://www.pewstates.org/uploadedfiles/pcs_assets/2011/pew_risk_assessment_brief.pdf Council of State Governments Justice Center 35

Why a Framework Was Needed Develop a shared language around the risk of criminal activity and public health needs Integrate the best practices in mental health treatment, substance abuse treatment, and recidivism reduction Help system administrators allocate scarce resources more wisely Maximize the impact of interventions on public safety and public health Council of State Governments Justice Center 36

Council of State Governments Justice Center 37

The Framework Low Criminogenic Risk (low) Medium to High Criminogenic Risk (med/high) Low Severity of Substance Abuse Substance Dependence Low Severity of Substance Abuse Substance Dependence (low) (med/high) (low) (med/high) Low Severity of Mental Illness (low) Serious Mental Illness (med/high) Low Severity of Mental Illness (low) Serious Mental Illness (med/high) Low Severity of Mental Illness (low) Serious Mental Illness (med/high) Low Severity of Mental Illness (low) Serious Mental Illness (med/high) Group 1 I L CR: low SA: low MI: low Group 2 II L CR: low SA: low MI: med/high Group 3 III L CR: low SA: med/high MI: low Group 4 IV L CR: low SA: med/high MI: med/high Group 5 I H CR: med/high SA: low MI: low Group 6 II H CR: med/high SA: low MI: med/high Group 7 III H CR: med/high SA: med/high MI: low Group 8 IV H CR: med/high SA: med/high MI: med/high Council of State Governments Justice Center 38

NYC: Why is the number of people with mental illnesses increasing in our jail? NYC Jail Population (2005-2012) Average Daily Jail Population (ADP) and ADP with Mental Health Diagnosis 13,576 Total 11,948 Total Analyses Revealed Greatest Potential Impact by Addressing Pretrial Detainee Population ALOS for Pretrial Admissions (Days) 40 10,257 7,557 76% 63% 3,319 4,391 24% 37% 2005 2012 M Group Non-M Group 79 Pretrial 62% Sentence d 24% State Prison Sentence 15% ALOS for Sentenced Admissions (Days) 60 81 Council of State Governments Justice Center 39

NYC Analyses Reveal Greatest Potential Impact by Addressing Pretrial Detainee Population Similar minimum bail amounts set but M group much less likely to make bail % Making Bail Post-Arraignment 25.0% 20.0% 15.0% 10.0% 5.0% 20.8% 11.8% 0.0% Felony Misd. $4,784 $4,769 $1,055 $1,001 Lowest amount needed for release (grouped median) Council of State Governments Justice Center 40

NYC Adopts Changes & Begins Implementation Justice Center Final Report Mayor announces the allocation of nearly $10 million to create Court-based Intervention and Resource Teams (CIRTs) to serve over 3,000 clients with mental health needs annually City selects CIRT providers for each borough and negotiates performance based contracts Borough-specific implementation planning begins with goal of serving first clients by December 2013 December Spring Spring Summer Fall 2012 2013 2013 2011 2013 City releases RFP for CIRT Provider Applications City starts development of IT infrastructure to share appropriate information Council of State Governments Justice Center 41

Significant Challenge #1: No county effectively screening and assessing, and recording in an info system Johnson County, KS Hillsborough, NH New York City, NY Bexar County, TX Mental Health Assessment Substance Abuse Assessment Risk Assessment Council of State Governments Justice Center 42

Significant Challenge #2: Results of assessment not driving decision making Concerns that too much information undermines defense or prosecution strategies Confusion about distinctions between risk and need, dangerousness, and failure to appear Few sites using information to inform conditions of release, supervision intensity, or service package Council of State Governments Justice Center 43

Significant Challenge #3: Workforce not ready to apply these concepts/strategies Static Risk Factors Criminal history number of arrests number of convictions type of offenses Current charges Age at first arrest Current age Gender Dynamic Risk Factors Anti-social attitudes Anti-social friends and peers Anti-social personality pattern Substance abuse Family and/or marital factors Lack of education Poor employment history Lack of pro-social leisure activities Council of State Governments Justice Center 44

Significant Challenge #3: Workforce not ready to apply these concepts/strategies Behavioral Health Reluctance to focus on higher risk individuals Insufficient skillsets involving cognitive behavioral interventions targeting needs Serious shortages in treatment capacity Criminal Justice Judges are skeptical (NY training example) Community corrections, jail, and prison staff need improved understanding of mental health issues and how to serve this population effectively Council of State Governments Justice Center 45

Texas Counties: How do we compare rates of recidivism across counties? Comparable Recidivism Measure County Recidivism Rate Harris 40% Dallas 30% Bexar 28% Tarrant 30% Travis 37% Average 33% Step 1: Set baseline for counties to get comparable recidivism measure Measure recidivism of jail releases, probationers and parolees to the county Later steps: Track changes in recidivism over time Review why recidivism rates may vary by county Review relation of recidivism rates to county practices and programs Council of State Governments Justice Center 47

Recidivism is a Return to Criminal Activity after Previous Criminal Involvement Recidivism Rate The number of offenders in a county under community supervision (probation or parole) or released from custody during time period (X) with outcome (Y) within (Z) number of months Number of offenders under community supervision or released from custody during time period (X) X. Released during time period Y. Outcome tracked Z. Uniform follow up period Fiscal year Calendar year Rearrest Conviction Return to Incarceration Follow up matters a one year rate will track offenders for 12 months Fabelo, Tony and Nancy Arrigona. March 1991. Recommended Methodologies for State Criminal Justice Agencies. State of Texas. Accessed March 4, 2013 http://www.lbb.state.tx.us/public_safety_criminal_justice/reports/unirecid.pdf Council of State Governments Justice Center 48

Need Standard Definition of Recidivism Definition Follow-up Re-arrest Uniform one, two and three year follow-up One Year Uniform Tracking Period Percent Re-arrested Council of State Governments Justice Center 49

Re-arrest Rate for Probationers in Harris County Felons 1 year 2 year 3 year 18% 27% Percent Rearrested after Three Years 36% Misdemeanors Overall Recidivism Rate 30% 1 year 2 year 3 year 13% 21% Percent Rearrested after Three Years 27% Harris Probation Learning Site Report to Judicial Officials, April 18, 2012 Council of State Governments Justice Center 50

Need to Access Data in Multiple Storage Systems Recidivism Data Collection Elements Case Data to Be Requested Calendar Year 2011 Data CY 2011-12 County Jail Releases CSTS Probation Admissions TDCJ Releases to Parole DPS Criminal History SID Admission Date Release Date Release Type Offense Type Offense Level SID Admission Date Completion Date County Offense Type Offense Level Risk Score Risk Cutoff SID Parole Start Date Parole End Date County of Release Offense Type Offense Level Risk Score Risk Cutoff Arrests for each SID Date of Arrest Date of Birth Gender Race/Ethnicity Council of State Governments Justice Center 51

Need to Make Rates Comparable Across Counties by Controlling for Risk Ideal Measure If unavailable Working Measure Use risk assessment from county jail data for offenders Build a proxy risk score using static factors (age, gender, criminal history) from the county jail and criminal history file Risk assessment in criminal justice is a systematic methodology to determine the risk of recidivism of an offender for a variety of purposes using: Static Factors (cannot change) Prior conviction history Age at first conviction Dynamic Factors (can change) Employment status Treatment completion Council of State Governments Justice Center 52

Recidivism Probationers by Risk in Harris County Three Year Recidivism Rate for Felons by Wisconsin and Combined LSI-R Risk Categories Minimum Medium Maximum 47.9% 36.8% 41.3% 34.1% 23.6% 25.2% LSI-R Wisconsin LSI-R Wisconsin LSI-R Wisconsin Wisconsin = 11 Risk Factors LSI-R = 44 Risk Factors Harris Probation Learning Site Report to Judicial Officials, April 18, 2012 Council of State Governments Justice Center 53

Summary Key Elements Outcome Measurement Uniform Follow-Up Period Control for Risk County Jail Release: Rearrest Probation and Parole: Rearrest 365 days from release for each offender Risk Offense level Demographics Council of State Governments Justice Center 54

Determine Logistics for Extracting Data Who Houses the data? What Data is available for extraction? When Can the data be extracted? Where How Do we need to get permission for the extraction? Do we want to structure the file extract? Council of State Governments Justice Center 55

Counties Are Where It s At Leverage state county association networks to offer trainings, disseminate latest information Coordinate closely with state county associations when we work intensively in a state/county Partner with state county associations to pilot concepts in particular sites and then replicate lessons in other counties Imbed capacity in state county associations to provide technical assistance to county officials Council of State Governments Justice Center 56

Thank You Report at: www.justicecenter.csg.org mthompson@csg.org The presentation was developed by members of the Council of State Governments Justice Center staff. The statements made reflect the views of the authors, and should not be considered the official position of the Justice Center, the members of the Council of State Governments, or the funding agency supporting the work. Citations available for statistics presented in preceding slides available on CSG Justice Center web site. Council of State Governments Justice Center 57