Moving from Principle to Moving from Principle to Execution - - PowerPoint PPT Presentation

SLIDE 1

Moving from Principle to Moving from Principle to Execution Execution

Applications of the Risk-Dosage Relationship

Kimberly Gentry Sperber, Ph.D.

SLIDE 2

Today Today’ ’s Session s Session

• Review of the risk principle
• Discussion of ways to operationalize the

risk principle

• What does the research say about
• perationalizing the risk principle into

practice?

• What counts as dosage?
• Findings from first Talbert House

dosage study

• Building a dosage research agenda
• Questions and practitioner input

SLIDE 3

The Risk Principle The Risk Principle

• Identify those offenders with

higher probability of re-offending

• Target those offenders with higher

probability of re-offending

• Targeting lower risk offenders can

increase recidivism

• Referred to as the WHO principle

SLIDE 4

2 0 0 2 UC Study of Halfw ay 2 0 0 2 UC Study of Halfw ay Houses and CBCFs Houses and CBCFs

Findings on the Risk Principle

SLIDE 5

Increased Recidivism

Reduced Recidivism

SLIDE 6

Treatment Effects For High Risk Offenders

• 34
• 18
• 15 -14
• 6 -5
• 2 -2

2 3 3 3 5 6 7 8 8 9 10 10 12 12 12 13 13 13 15 21 22 24 25 27 30 32 34 R i v e r C i t y F r e s h S t a r t A l t e r n a t i v e A g e n c y T a l b e r t H

• u

s e C

• r

n e r s t

• n

e C

• m

m u n i t y A s s e s s m e n t P r

• g

r a m ( M e n ’ s ) M

• n

d a y W O R T H C i n c i n n a t i V O A M c M a h

• n

H a l l T a l b e r t H

• u

s e S p r i n g G r

• v

e N E O C A P O r i a n a H

• u

s e R I P A l v i s H

• u

s e D u n n i n g H a l l L

• r

a i n / M e d i n a A l l C B C F F a c i l i t i e s C a n t

• n

C

• m

m u n i t y T r e a t m e n t C e n t e r L u c a s C

• u

n t y S R C C C A l l F a c i l i t i e s L i c k i n g / M u s k i n g u m S u m m i t C

• u

n t y B u t l e r S E P T A C

• m

m u n i t y T r a n s i t i

• n

s F r a n k l i n C

• u

n t y S m a l l P r

• g

r a m s O r i a n a H

• u

s e T M R C C i n c i n n a t i V O A C h e m i c a l D e p e n d e n c y P r

• g

r a m A l v i s H

• u

s e A l u m C r e e k T a l b e r t H

• u

s e B e e k m a n C

• m

p D r u g H a r b

• r

L i g h t S a l v a t i

• n

A r m y C

• m

m u n i t y C

• r

r e c t i

• n

s A s s

• c

i a t i

• n

T

• l

e d

• V

O A M a h

• n

i n g C

• u

n t y E O C C 10 20 30 40

• 10
• 20
• 30
• 40

Probability of Reincarceration

SLIDE 7

2 0 1 0 UC Study of Halfw ay 2 0 1 0 UC Study of Halfw ay Houses and CBCFs Houses and CBCFs

Findings on the Risk Principle

SLIDE 8

Treatm ent Effects for Low Risk Treatm ent Effects for Low Risk

SLIDE 9

Treatm ent Effects for High Risk Treatm ent Effects for High Risk

SLIDE 10

Generalizability Generalizability

• To whom does the risk principle

apply?

– Adults – Juveniles – Males – Females – Sex Offenders – Violent Offenders

SLIDE 11

Operationalizing the Risk Operationalizing the Risk Principle Principle

• Separate living quarters
• Separate groups
• Risk-specific caseloads
• Varying dosage by risk
• Varying length of stay by risk

SLIDE 12

W hat Does the Prior Research Say? W hat Does the Prior Research Say?

• Lipsey (1999)

– Meta-analysis of 200 studies on serious juvenile delinquents – Minimum length of stay of 6 months – Approximately 100 hours

SLIDE 13

W hat Does the Prior Research Say? W hat Does the Prior Research Say?

• Bourgon and Armstrong (2005)

– 620 incarcerated males – 12 months recidivism rates – 100 hours to reduce recidivism for moderate risk OR few needs

• 100 hours not enough for high risk

– 200 hours for high risk OR multiple needs – 300+ hours for high risk AND multiple needs

SLIDE 14

First Talbert House Dosage Study First Talbert House Dosage Study

Sperber, Latessa, & Makarios (forthcoming):

• Conceptual understanding of the risk

principle versus operationalization of the risk principle in real world setting to achieve maximum outcome

• “Can we quantify how much more service

to provide high risk offenders?”

SLIDE 15

The Program The Program

• 100-bed CBCF for adult male

felons

• Prison diversion program
• Average length of stay = 4 months
• Serves 3 rural counties
• Cognitive-behavioral treatment

modality

SLIDE 16

Community Correctional Center Risk Level Structure Guide

Medium High High Medium Low/ Moderate Low LSI Score Range 34+ 31-33 24-30 19-23 0-18 Length of Stay Target (days) 147 133 119 105 60 Corrective Thinking 200 180 132 92 52 AOD 62 54 46 38 28 Individualized Relapse Prevention 21 Anger Management 24 24 24 24 if needed Domestic Violence 24 15 15 15 if needed Vocational* 15 15 15 15 8 Life Skills* 16 16 16 16 8 Personal Development* 10 10 10 10 if needed *not counted in dosage total Total hours available: 351 314 258 210 117

SLIDE 17

Methodology Methodology

• Sample size = 689 clients
• Clients successfully discharged between

8/ 30/ 06 and 8/ 30/ 09

– 300 clients pre-implementation – 123 clients during implementation – 266 clients post-implementation

• Excluded sex offenders
• Dosage defined as number of group hours per

client

• Multiple measures of recidivism – arrest,

conviction, reincarceration

– All offenders out of program minimum of 12 months

SLIDE 18

Sam ple Characteristics Sam ple Characteristics

• 88.8% White
• Average age 33
• 59.7% single, never married
• 43.2% less than high school education
• 95.5% Felony 3 or higher

– Almost half Felony 5

• 80% moderate risk or higher
• 88% have probability of substance

abuse per SASSI

SLIDE 19

low moderate high

• verall

0‐99 Tx hours 39 52 46 100‐199 Tx hours 26 45 81 43 200+ Tx hours 43 57 48

Recidivism Rates by Treatment Intensity and Risk Levels

Average low=78, Moderate= 155 High =241

SLIDE 20

Unansw ered Questions Unansw ered Questions

• Is there a saturation effect?

– What is maximum return on investment?

• What counts as dosage?
• Does dosage requirement vary by

setting?

– Halfway house versus CBCF

• What is the impact of sequencing
• f dosage?

SLIDE 21

Building a Dosage Research Agenda Building a Dosage Research Agenda

• Treatment Dosage and Risk: An Extension and

Refinement of the Appropriate Levels of Dosage by Risk Level

– Makarios, Sperber, & Latessa (in process)

• Treatment Dosage and Personality: Examining the

Impact of Personality on the Dosage Recidivism Relationship

– Latessa, Makarios, & Sperber (in process)

• Examining the Dosage Recidivism Relationship in

Female Offenders

– Spiegel & Sperber (in process)

• Frontloading Treatment Dosage and the Impact on

Recidivism

– Sperber and Smith (in process)

SLIDE 22

Treatment Dosage and Risk: An Extension and Refinement Treatment Dosage and Risk: An Extension and Refinement

• f the Appropriate Levels of Dosage by Risk Level
• f the Appropriate Levels of Dosage by Risk Level
• The limited research available on dosage indicates

a broad range of treatment hours from a minimum

• f 200 hours to more than 300 hours of

programming for high risk offenders.

• Consequently, this study seeks to refine the

existing extant knowledge about tailoring dosage to offender risk by testing the impact of more narrowly defined categories of dosage on recidivism for moderate and high risk offenders.

• No findings yet; just recently secured recidivism

data.

SLIDE 23

Treatment Dosage and Personality: Examining the Impact Treatment Dosage and Personality: Examining the Impact

• f Personality on the Dosage Recidivism Relationship
• f Personality on the Dosage Recidivism Relationship
• Research on the risk principle confirms that correctional

practitioners should differentiate services by offender risk.

• Research also confirms that these services should be

based on a cognitive-behavioral modality.

• At the same time, there is some research to suggest

that offenders with certain personality types (e.g. neurotics) are higher risk for re-offending and may not fare as well as other personality types within cognitive behavioral programs.

• If this is true, increasing cognitive behavioral dosage

for high risk neurotic offenders may have a detrimental impact on recidivism for those offenders.

• Consequently, this study examines personality type as

a moderator of the risk dosage relationship to determine the impact on recidivism.

SLIDE 24

Personality Types Personality Types

• Jesness Inventory
• 9 Types collapsed into 4:

– Aggressives – Neurotics – Dependents – Situationals

SLIDE 25

Prelim inary Findings Prelim inary Findings

• Bivariate Results:

– Risk positively correlated to dosage – Risk positively correlated to recidivism – Risk positively correlated to neurotic personality – Risk negatively correlated to dependent personality – Age negatively correlated to risk and to recidivism – Age positively correlated to neurotic personality and negatively correlated to aggressive personality

SLIDE 26

W hat Does This Mean? W hat Does This Mean?

• Higher risk offenders received more dosage
• Higher risk offenders more likely to re-offend
• Younger offenders higher risk and more

likely to re-offend

• Neurotics more likely to be higher risk and to

be older

• Aggressives more likely to be younger
• Dependents more likely to be low risk

SLIDE 27

Percent Returned to Prison by Percent Returned to Prison by Risk and Dosage Risk and Dosage

1 0 2 0 3 0 4 0 5 0 6 0 7 0 8 0 9 0 Low Mod High Overall 0 - 9 9 tx hours 1 0 0 - 1 9 9 tx hours 2 0 0 + tx hours

SLIDE 28

Next Steps Next Steps

• Address limitations

– Only have 190 cases total – Results is that many of the categories have very few people (e.g., high risk/ high dosage/ recidivated = 19) – Add a new cohort of clients to increase number of neurotic clients – Re-run analyses with larger sample and compare findings

SLIDE 29

Examining the Dosage Recidivism Examining the Dosage Recidivism Relationship in Female Offenders Relationship in Female Offenders

• Studies on the number of treatment hours

necessary to reduce recidivism for high risk

• ffenders are few in number.
• Studies to date have relied on male samples.
• Cannot assume that a standard number of

treatment hours necessary to reduce recidivism exists for both men and women.

• Consequently, present study examines the

impact of varying levels of treatment dosage by risk for female offenders in a halfway house setting.

SLIDE 30

Methodology Methodology

• Sample size = 314 clients
• Clients successfully discharged between

10/ 1/ 07 and 2/ 28/ 10

• Dosage defined as number of group

hours per client

• Recidivism defined as re-arrest in

– Checked Hamilton County and referral/ home county websites – All offenders out of program minimum of 12 months

SLIDE 31

Sam ple Characteristics Sam ple Characteristics

• 71% White
• 82.8% single
• Average age is 36
• 52.9% previous mental health history
• 95.5% moderate risk or below

– 53.8% low or low-moderate risk

• 23.2% re-offended
• Average time at risk was 3 years; max

was 4.5 years

SLIDE 32

Dosage and Recidivism by Risk Dosage and Recidivism by Risk

5 1 0 1 5 2 0 2 5 Low / Low Mod Moderate + Overall 0 - 5 0 Hours 5 1 - 1 0 0 Hours 1 0 1 + Hours

SLIDE 33

Prelim inary Findings Prelim inary Findings

• Bivariate Relationships:

– LSI scores positively correlated to recidivism – LSI scores negatively correlated to dosage

• Lower risk = higher dosage
• Influence of IOP drug court clients residing in the

halfway house

– Significant positive relationship between mental health history and recidivism – Bivariate results suggest that increasing dosage

• ver 100 hours for lower risk population may

lead to higher recidivism rates

• Still need to conduct multiple regression, however

SLIDE 34

Next Steps Next Steps

• Investigate adding additional clients to the

sample to increase variation in both risk and recidivism

• Eliminate dosage outliers to determine

impact

• Eliminate drug court clients from the

sample to use pure halfway house sample

• Re-run recidivism checks in 9/ 12 to

increase variation in recidivism

• Attempt multivariate analyses at that time

SLIDE 35

Frontloading Treatment Dosage and the Frontloading Treatment Dosage and the Impact on Recidivism Impact on Recidivism

• The emerging literature suggests that high risk
• ffenders require at least 200 hours of service to reduce

recidivism.

• Dosage studies conducted to date have focused on
• ffenders residing in secure residential environments.
• Little attention paid to the infrastructural barriers to

providing adequate dosage in halfway houses and other non-secure settings.

• In environments such as these, achieving 200 hours of

dosage often may not be feasible. Consequently, some community programs load treatment hours into the beginning of programming before allowing offenders to begin employment in the community.

• This study seeks to examine the impact of this practice
• n post-release recidivism in a sample of adult male

halfway house participants.

SLIDE 36

Methodology Methodology

• Sample size = 345 clients
• Clients successfully discharged between

7/ 1/ 08 – 6/ 30/ 10

• Dosage defined as number of group

hours per client

• Recidivism defined as re-arrest in

– Checked Hamilton County and referral/ home county websites – All offenders out of program minimum of 12 months

SLIDE 37

Defining Frontloading Defining Frontloading

• Percentage of treatment hours received

prior to first employment

• Difference between hours received

during first half of treatment and second half of treatment

• Whether 60% or more of hours were

received prior to employment

• Whether 70% or more of hours were

received prior to employment

SLIDE 38

Sam ple Characteristics Sam ple Characteristics

• 80.8% White
• 63.5% High School/ GED or above
• 80.6% Single
• 61.7% moderate risk or higher
• 28.4% re-offended

SLIDE 39

Prelim inary Results Prelim inary Results

• Bivariate Results:

– Risk positively correlated to recidivism. – Risk positively correlated to overall dosage. – Risk significantly correlated to frontloaded dosage. – Age negatively correlated to recidivism.

• Summary:

– Higher risk more likely to re-offend, receive more treatment, and to receive more hours of treatment prior to employment – Younger offenders more likely to re-offend

SLIDE 40

I m pact of Frontloading I m pact of Frontloading

Lower Risk Higher Risk 70% Frontloaded 70% Frontloaded Yes No Yes No Rearrest 17.6% 31.25% 40% 38.5% No Rearrest 82.4% 68.75% 60% 61.5%

SLIDE 41

Does Frontloading W ork? Does Frontloading W ork?

• No clear support provided
• However

– Limited risk distribution – Limited overall dosage

• Average was 35.5 hours
• Majority received between 20 and 50 hours

– Frontloading did not significantly change the way in which service delivery occurred

• More studies needed

SLIDE 42

Questions and Answ ers Questions and Answ ers

Practitioner Input

Kimberly.Sperber@talberthouse.org