2009

A

Texas

Annual

Youth

Review

of

Agency

Commission

Report

Cheryln K. Townsend, TYC Executive Director
Submitted December 31, 2009

Submitted in compliance with Texas Human Resources Code, Section 61.0315(a)

REVIEW OF AGENCY TREATMENT EFFECTIVENESS
DECEMBER 31, 2009

TEXAS YOUTH COMMISSION

FY2009 Review of Agency Treatment Effectiveness

TABLE OF CONTENTS

Executive Summary ........................................................................................................ 2

Youth Needs for Specialized Treatment.......................................................................... 6

TYC Specialized Treatment Programs ............................................................................ 7

Study .............................................................................................................................. 9
Hypothesis ............................................................................................................... 9
Methodology............................................................................................................. 9
2009 Sample Selection and Tracking..................................................................... 10
Analysis.................................................................................................................. 13
Results: Sexual Behavior Treatment Program ...................................................... 14
Results: Capital & Serious Violent Offender Treatment Program .......................... 16
Results: Chemical Dependency Treatment Program ............................................ 18
Results: Mental Health Treatment Program .......................................................... 20
Results: Female Youth .......................................................................................... 21

Conclusion .................................................................................................................... 24

Texas Youth Commission

1

FY2009 Review of Agency Treatment Effectiveness

EXECUTIVE SUMMARY

The Texas Human Resources Code, Section 61.0315, requires the Texas Youth Commission
(TYC) to review annually the effectiveness of agency programs for the rehabilitation and
reestablishment in society of youth committed to the commission. The agency reports on the
effectiveness of those programs to the Legislative Budget Board no later than December 31 of
each year. The rehabilitation treatment programs required for inclusion in this report are for sex
offenders, capital and serious violent offenders, chemically dependent youth, youth with mental
health needs, and female youth. This report measures the effectiveness of treatment in these
programs by tracking one-year and three-year recidivism rates among juvenile offenders after
their release from TYC.
Recidivism is the best available measure of treatment effectiveness because arrests and
incarcerations following releases are objective data that can be gathered systematically. The
measure is limited, however, in clearly indicating the effectiveness of any program because
there is a balance of many factors after release that influences youth behavior, including peer
pressure in the community, family circumstances, treatment continuity, and education and job
successes. The report analyzes in one-year and three-year increments youth released from
TYC beginning July 1, 2001 through June 30, 2008. Youth who participated in any TYC program
and were released after June 2008 were not included within the scope of the study because a
full year of re-arrest and incarceration data was not available. The report is organized to
describe youth treatment needs, TYC’s treatment programs, and the study’s hypothesis,
methodology, and analysis by treatment category.
Major findings:
ƒ Overall, the results of the analysis were mixed in determining the effectiveness of
pre-reform programming. Programs for sex offenders and capital and serious
violent offenders were effective, while the results for specialized treatments for
chemically dependent youth, youth with mental health needs, and non-genderspecific treatment services for girls reflected opportunities for improvement.
ƒ

Youth who had participated in TYC sex offender treatment were 45% less likely
than expected to be rearrested for a violent offense and 16% less likely to be
rearrested for any offense within one year.

ƒ

Youth enrolled in the Capital and Serious Violent Offender Program were 72%
less likely than expected to be arrested for a violent offense within one year.

ƒ

Youth who participated in the chemical dependency treatment program were
more likely to be arrested within one year or incarcerated within one or three
years.

ƒ

Youth who participated in specialized programs for mental heath services and
girls who participated in any non-gender-specific specialized treatment were
about as likely to be rearrested or incarcerated within one or three years as male
and female youth who did not participate in those programs.

Texas Youth Commission

2

FY2009 Review of Agency Treatment Effectiveness

The following table contains an overall summary of the study results.
EFFECTS OF TREATMENT ON SELECTED OUTCOMES: LIKELIHOOD OF PARTICIPANTS TO RECIDIVATE
Measure

Sexual
Behavior

Arrest Any Offense: 1 Year

16.2%
less likely

Arrest Violent Offense: 1 Year
Incarceration Any Offense: 1 Year

Capital &
Serious

Treatment Groups
Chemical
Mental
Dependency
Health

Females

ns

8.9%
more likely

ns

ns

45.3%
less likely

71.8%
less likely

ns

ns

ns

ns

ns

13.4%
more likely

ns

ns

12.3%
Incarceration Any Offense: 3 Years
ns
ns
more likely
ns
ns = no significant differences (statistical non-significance can result with small sample sizes)

ns

These study results are important. The agency is using the results and the previous 2008
Report results to assess programs for those elements that are effective and those that must be
changed to improve outcomes. This report and previous reports show that pre-reform
specialized programming for sex offenders and capital and serious violent offenders was
effective; therefore, these services have been expanded to more youth and offered earlier in
treatment regimens. New programming for Aggression Replacement Training® is offered as a
service enhancement for capital and serious violent offenders. A new staff development
initiative is ensuring that the agency will comply with statutory requirements for employing
Licensed Sex Offender Treatment Providers. Five staff received licensure in 2009 and other
clinicians are targeted to attain licensure by October of 2010.
The study results also affirm the need for reforms adopted by the 80th and 81st Legislatures, but
they do not yet reflect the impacts of those reforms due to the time lag for data availability.
Although the focus of the report is on one outcome—recidivism, improving recidivism also
requires improvement in other cross-cutting outcomes, such as educational achievement. A
youth’s ability to read and comprehend leads to positive results on multiple outcome indicators,
including a lower probability for arrest or incarceration.
Results of long-term studies in 2004 of healthy human brain development supported new
understanding of adolescent brain development and the implications for changing and
correcting behavior. With juvenile offenders, experience shows that rehabilitation programs are
often ineffective because many of these youth were never “habilitated” in the first place.
Historically, there was either no data or poor data on which to base decisions for service
improvements. Increasingly, scientific data from other jurisdictions are available for reforming or
refining programs. Typically high relapse rates for alcohol and drug dependencies following
release from juvenile correctional systems support the allocation of more and better targeted
resources for halfway houses and parole to reduce the frequency and severity of relapse. The
81st Legislature supported stronger re-entry programming with an appropriation for specialized
aftercare pilot programs. The first pilot program will target youth with substance abuse treatment
needs. Functional Family Therapy© (FFT©) is an evidence-based model with proven success
that will be initiated in FY 2010. The model is a structured family intervention program for youth
diagnosed with conduct disorders, violent behavior, and substance abuse that has been shown
in multiple studies to reduce felony recidivism.
Two new alcohol and drug treatment programs were implemented in 2009 and services were
expanded to include moderate-need youth. Also, better coordinated and strengthened
psychiatric services in the last two years are expected to improve longer term results for mental

Texas Youth Commission

3

FY2009 Review of Agency Treatment Effectiveness

health needs in youth. Early results indicate a reduced number of mental health discharges due
to an inability to progress in programs. Specialized treatment programming for females was
expanded in FY 2009 and Girls Circle programming is being implemented in FY 2010 to target
better outcomes for female juvenile offenders that will reduce recidivism.
As TYC continues to receive the most serious offenders with more intense, complex, and longer
term needs for specialized services, one of the agency’s overarching goals will continue to be
greater public safety through fewer re-arrests or incarcerations. Treatment challenges will
continue to require great determination, focus, and effective coordination across jurisdictional
entities with shared responsibilities. TYC’s focus for improving recidivism will continue to revolve
around three treatment objectives:
1. Ensure that program models are well-matched to youth needs as reflected by data from
intake assessments and periodic assessments through the youth’s discharge from TYC.
2. Concentrate agency initiatives on reentry programming for youth preparing to be
released from TYC facilities, including early warning signals for public safety risks.
3. Invest in staff development to expand knowledge, skills, and abilities to manage
increasingly complex youth needs in an increasingly complex and fast changing
correctional environment.

Texas Youth Commission

4

Timeline of Reform Events
Period of Reform

Identifying and correcting problems;
developing new policies and programming

2007

March 2007

June 2007

TYC Board
resigns

SB 103
becomes
law

2008

•

FY2009 Review of Agency Treatment Effectiveness

Period of Rebuilding

2009

June 2009

October 2008
August 2008
©

CoNEXTions pilot
program reviewed
and revised

TYC placed in
conservatorship
December 2007
Created CoNEXTions
Treatment Program
prototype

Implemented
Youth
assessment
tool for
individualized
treatment

TYC removed from
conservatorship
Executive
Commissioner
appointed

September 2009
First meeting of new TYC
Board
Board hires Executive
Director

©

December 2008

September 2009

Gender-specific specialized
treatment programs implemented

New comprehensive
specialized treatment policy
goes into effect

Key Event Categories
January 2009

Agency Leadership

Staff development program begun for
sex offender treatment providers
(LSOTP) to comply with licensing
requirements by October 2010

Creating & Implementing the General
CoNEXTions© Treatment Program
Creating & Implementing New
Specialized Treatment Programs

Implementing new policies
and programming

December 2009
All components of
©
CoNEXTions fully
implemented

Specialized treatment
expanded to accommodate
youth with varying needs
Specialized outpatient
aftercare available to high
and medium need youth

New curriculum and programming started
for sexual behavior and chemical
dependency treatment

FY 2009 Treatment Effectiveness Review Report
July 2001-June 2002
First cohort of youth
released and tracked
3 years
Texas Youth Commission

July 2003-June 2004
First cohort of youth
released and tracked
1 year

July 2005-June2006
Last cohort of youth
released and
tracked 3 years

July 2007-June 2008
Last cohort of youth
released and tracked 1 year

5

June 2009
End of tracking for last
cohorts of youth in this report

FY2009 Review of Agency Treatment Effectiveness

YOUTH NEEDS FOR SPECIALIZED TREATMENT
Youth whose offenses were between their 10th and 17th birthdays can be committed to TYC for
felony offenses, including violation of felony probation. Youth entering TYC for the first time or
returning to TYC present increasingly complex and intense needs for a range of specialized
treatment. Recent legislative changes have reshaped the juvenile justice system allowing only
the most serious juvenile offenders to be sent to the agency. These offenders are often
characterized by multiple severe needs for sex offender treatment, alcohol and drug treatment,
mental health services, treatment for violent behavior, and special education services. Local and
county level resources are typically insufficient to address such complex needs.
The characteristics of youth committed to TYC are not typical of the general population. Many
of the characteristics are highly correlated with a probability of future criminal behavior. Most of
these characteristics are static risk factors and cannot be changed. Examples of static risk
factors are prior juvenile justice history, prior placements, IQ scores, and history of abuse and
neglect.
Non-static factors and protective factors can be changed. Examples of these factors are
education level, peer relationships, gang membership, and substance abuse.
Interventions can improve non-static and protective factors and reduce the influence of the
unchangeable static history. However, even when youth committed to TYC make progress on
non-static and protective factors, the risk of future criminal behavior is higher than that of the
general population and community-based juvenile justice populations.
A typical TYC youth is male, 16 or 17 years old, of a minority race or ethnicity, from an urban
setting, and reflects family impacts often associated with conditions of economic poverty. The
table provides a profile of youth entering TYC for the first time with comparisons for fiscal years
2005 through 2009.

Profile of New Commitments to TYC
FY 2005

FY 2007

FY 2009

Committed for Felony Offenses

81%

84%

100%

Committed for a Violent Offense

33%

39%

48%

Self-Reported Gang Member

35%

40%

43%

Chemically Dependent

39%

36%

47%

Sex Offense History

11%

9%

12%

Had Serious Mental Health Diagnosis

36%

38%

37%

Had History of Abuse or Neglect

35%

37%

38%

IQs Less Than 100

81%

83%

85%

Median Educational Achievement

Texas Youth Commission

5 years behind

5 years behind

4-5 years behind

6

FY2009 Review of Agency Treatment Effectiveness

TYC SPECIALIZED TREATMENT PROGRAMS

The rehabilitation treatment programs analyzed for this report are those for sex offenders,
capital and serious violent offenders, chemically dependent youth, youth with mental health
needs, and female youth. Youth who were released from a secure TYC program after June 30,
2008 were not included within the scope of the study because a full year of re-arrest and
incarceration data was not available.
While the focus of the report is on specialized treatment services, the basic treatment program
is the foundation for specialized services. All TYC youth receive basic treatment services;
therefore, both youth enrolled and those not enrolled in specialized treatment received basic
treatment. Youth in this study other than those in the final one-year cohort received treatment
when TYC was offering Resocialization© as its basic treatment intervention. Resocialization©
focused on three major areas of intervention: academic and workforce development, behavior
modification, and correctional therapy. Program completion was determined by progress
through a system of four “phases” that required youth to learn and demonstrate competency to
reduce the probability of offending. Program completion was defined as completing and
maintaining Phase 4 in each area.
TYC replaced Resocialization© with a new rehabilitation strategy called CoNEXTions©, which
incorporates nationally recognized best and promising practices. The new program was first
piloted at TYC as a general treatment program in December 2007 at one facility. As a result of
the pilot, the agency made substantial changes to the program in August 2008, developing
CoNEXTions© as a rehabilitative strategy including education, specialized treatment, and a
positive behavioral change system. This rehabilitation strategy was implemented in its current
form after youth in this study were released.
At admission to TYC, all youth entered the Orientation & Assessment Unit where the
assessment process identified the presence of specialized needs that required additional
emphasis while in TYC. Two specialized need areas were based primarily on the youth’s
classifying offense: capital and serious violent offenses and sexual offenses. The need for
specialized chemical dependency treatment was based on a diagnosis of chemical dependency.
In addition, for all three of these specialized treatment programs, the presence of a high risk for
re-offending in the respective area was a factor for determining specialized need. The fourth
specialized treatment program was for youth with identified mental health needs. Inclusion in
this group was based on the presence of a mental health diagnosis and impaired adaptive
functioning indicating an ongoing need for supportive psychiatric and mental health services not
available in non-specialized TYC programs.
Pre-reform specialized treatment programs were based on the Resocialization© model but with
specific and more intensive emphasis on the specialized treatment need. In addition to factors
relating to general delinquent or criminal behavior, Life Stories included additional focus on the
etiology and development of those specific risk areas associated with the specialized need
being addressed. Offense cycles were expanded to include understanding of how aggressive,
sexual or drug related behavior patterns emerged and were maintained with emphasis on how
they could be modified. Success Planning addressed specific risk management issues in the
community to address these specialized risks. Additionally, those in the specialized programs
received psycho-educational programming to help them better understand “normal”
development and social customs. When indicated, the programs would introduce specialized
modalities to facilitate the treatment process. Caseload sizes were smaller than in the general
programs to facilitate more intensive individual and group work, and the staff assigned to the
programs received additional specialized training. Brief descriptions of the pre-reform programs
in place for youth in the current study are included below.

Texas Youth Commission

7

FY2009 Review of Agency Treatment Effectiveness

Capital & Serious Violent Offender Treatment Program (C&SVOTP) ● This
treatment program was a dormitory-based, structured 24 week program. The
residential component assisted in follow-up processing and exploration of issues
identified in the intensive process group. It provided an opportunity to analyze
the degree to which treatment gains observed in the group would generalize to
daily behavior on the dorm. The residential element allowed for better
coordination of treatment services between the program therapists, case workers
and dormitory staff members. It was designed to facilitate cognitive, emotional
and social developmental processes and facilitate empathic development,
emotional regulation and appropriate expression of feelings to improve
interpersonal functioning.
Sexual Behavior Treatment Program (SBTP) ● This treatment program was a
dormitory-based, structured 12 -18 month program designed to provide services
to youth with high risk to commit a new sexual offense. Gender specific programs
were offered to youth with adjudicated sexual offenses. The residential
advantage allowed for intensive work with youth on a daily basis to ensure that
gains and plans made in group were carried out in the less structured day to day
living situations. Motivational techniques specific to sex offender treatment were
coupled with intensive psychotherapeutic groups to identify issues and facilitate
change.
Chemical Dependency Treatment Program (CDTP) ● This treatment program
was a dormitory-based, 6 month program based on the belief that dependency
was a primary chronic disease which is progressive and influenced by biological,
psychological and social factors. The consequences of continued drug use
include problems in spiritual, moral, physical, emotional, intellectual and social
functioning. The program sought to address not only underlying emotional
dynamics that fueled delinquent and criminal behaviors but also addressed CD
issues that impacted the youth, their families and other victims. Entry to the
program was based on a diagnosed Chemical Dependency and a high score on
the agency’s violent risk assessment instrument.
Mental Health Treatment Program (MHTP) ● While most youth with mental
health needs were able to participate in the agency’s general programs with
psychiatric and psychological support and follow-up, a small percentage had
either more serious diagnoses or mental disorders that did not respond to
standard psychological and psychiatric interventions. These youth were generally
less able to manage the demands of the basic treatment program without
additional support and treatment overlay. The MHTP provided this support and
treatment. It offered enhanced psychiatric and psychological assistance along
with smaller caseloads and specially trained direct care staff. Adaptations of the
Resocialization© were made to address and minimize the obstacles to treatment
participation of the primary symptoms of the mental disorder. Most youth were
able to progress in the treatment program with these additional supports in place.
A very small percentage of TYC youth had mental health symptoms that would
periodically increase in severity and that required a protective environment to
provide stabilization of the most severe symptoms. Youth who were dangerous
to themselves or to others were eligible for admission to the Corsicana
Stabilization Unit (CSU) for short term treatment or could be admitted to a public
psychiatric hospital.

Texas Youth Commission

8

FY2009 Review of Agency Treatment Effectiveness

STUDY
HYPOTHESIS
The current study was designed to determine whether specialized treatment programs reduced
recidivism more than TYC’s basic treatment program.
Hypothesis: Specialized Treatment is more effective than basic
treatment. After statistically controlling for differences among
youth, youth with specialized needs who received specialized
treatment had lower recidivism rates than did youth with
comparable needs who did not receive specialized treatment.
METHODOLOGY
This review examines recidivism of youth in TYC specialized treatment programs by tracking
subsequent involvement in the juvenile and adult criminal justice systems. Recidivism is a
general term describing a ‘return to criminal or delinquent behavior’ and can be defined in
multiple ways. Therefore, caution is advised when comparing rates across different studies or
justice systems. As depicted and defined below, this study used multiple measures of
recidivism to capture different types of recidivating behavior at various points in time after
release from a secure location.

Recidivism Tracking Starting Point

For secure programs,
recidivism tracking
began the day youth
were released from a
secure program to a
non-secure residential
program, parole or
discharge, and
continued for the
designated time period
from that day.

Secure
Programs

Non-Secure
Residential
Programs

Parole

Discharge

Youth progress through TYC

Youth were tracked for up to three years from the date of release from a secure residential
program to a non-secure residential program or parole or discharged from the agency.
The specialized treatment analyses applied the following measures:
Arrest Rate for Violent Offense: The percent of youth in the cohort released
from secure programs within one year who were known to have been arrested for
any new violent offense. Violent offenses are those for which a youth committed
to TYC would be classified as a Violent Offender. These are generally felony
level offenses defined in the Texas Penal Code as being committed against
persons. Some specific examples are aggravated offenses, sexual assault,
murder, assault, robbery, arson, and non-violent offenses committed with intent
to commit a violent offense.

Texas Youth Commission

9

FY2009 Review of Agency Treatment Effectiveness

Arrest Rate for Any Offense: The percent of youth in the cohort who were
released from secure programs that, within one year, was known to have been
arrested for any offense or technical violation of parole.
Incarceration Rate for Any Offense: The percent of youth released from secure
programs that, within one or three years (depending on the cohort), were known
to have been incarcerated in secure juvenile confinement or an adult prison
facility for any offense or technical violation.
The table, 2009 Sample Selection and Tracking, shows that data on youth with an initial release
from secure confinement over two different five year periods were tracked to calculate
measures at one and three years after release. The two cohorts had some overlap as depicted
in the table. Note that the release periods for the groups overlap and, therefore, contain
duplicate cases.
2009 SAMPLE SELECTION AND TRACKING
Cohort
SubGroup

Release Period

Tracked for One Year
After Release Date with
Cohort Tracking Ending…

Tracked for Three Years
After Release Date with
Cohort Tracking Ending…

3 Year

July 1, 2001-June 30, 2002

NA

June 30, 2005

3 Year

July 1, 2002-June 30, 2003

NA

June 30, 2006

1 & 3 Year

July 1, 2003-June 30, 2004

June 30, 2005

June 30, 2007

1 & 3 Year

July 1, 2004-June 30, 2005

June 30, 2006

June 30, 2008

1 & 3 Year

July 1, 2005-June 30, 2006

June 30, 2007

June 30, 2009

1 Year

July 1, 2006-June 30, 2007

June 30, 2008

NA

1 Year

July 1, 2007-June 30, 2008

June 30, 2009

NA

The cohort for one-year recidivism and the cohort for three-year recidivism both include youth
released over five years. The following table, Enrollment Rates by Cohort, shows the total
cohort of youth released from TYC and the smaller group who were enrolled in the identified
specialized treatment. In the one-year cohort, for example, the total cohort was 1,337 released
youth with an identified high need for specialized treatment for sexual behavior, while only 556
(42%) were actually enrolled to receive such treatment prior to their release.
ENROLLMENT RATES BY COHORT 1
Total
Cohort

1

Total
Enrolled

Enrollment
Rate

One Year Treatment Cohorts
Sexual Behavior
Capital & Serious Violent Offender
Chemical Dependency
Mental Health

1,337
464
4,100
4,597

556
98
1,696
1,455

42%
21%
41%
32%

Three Year Treatment Cohorts
Sexual Behavior
Capital & Serious Violent Offender
Chemical Dependency
Mental Health

1,091
522
2,566
4,388

440
104
1,188
1,207

40%
20%
46%
28%

Cohorts include duplicate youth when youth present multiple needs. Cohorts include males and females.

Texas Youth Commission

10

FY2009 Review of Agency Treatment Effectiveness

ƒ

Enrollment rates for youth who had a high need for treatment ranged from 20%
to 46%, depending on the type of need. Reasons for not enrolling a youth were
limited bed space, enrollment in another program that the youth needed, and
disciplinary problems limiting the youth’s ability to participate.

ƒ

Only youth with the most severe impact in adaptive functioning because of their
mental health diagnoses are enrolled in the agency’s specialized mental health
treatment program. Most youth with mental health treatment needs function well
in a general population setting where they have access to licensed therapists,
psychological care, and nursing assistance.

ƒ

Beginning in September 2009, all youth with an identified sexual behavior
treatment need now receive specialized treatment based on a needs
assessment. In addition, improved coordination and treatment-focused aftercare
services are underway to retain treatment gains after reentry into the community.
The impact of these changes is expected to be seen in future youth data.

Characteristics of each youth in the sample included: assessed as high need for specialized
treatment by TYC and initial release from a secure program during the established time frame.
Only those youth with an initial release from secure confinement were included in order to
exclude youth who may have participated in specialized treatment during one stay but not
another.
The analysis compared the percent of youth that recidivated within defined intervals of time
following their release dates. Each measure had a treatment group and a group that did not
receive specialized treatment. The group that did not receive specialized treatment consisted of
youth with an initial release during the specified time period and had been assessed with a high
need for a specialized treatment program, but who were not assigned to such a program.
WH AT IS THE PREDICTED RATE?
The likelihood of re-offending based on certain
characteristics like age at first referral and gang
membership.
Similar to actuarial tables used by the health care industry to identify a
person’s probability of developing heart disease based on
characteristics such as blood pressure, smoking, age, and gender; or
by the insurance industry to identify a driver’s probability of being
involved in an accident based on age, prior accidents, marital status,
and distance from work.

The treatment group consisted
of youth meeting the same
criteria but who were also
enrolled
in
a
specialized
treatment program for at least
one day. Youth did not need to
complete
the
specialized
treatment program in order to
have comparability with the
group not receiving treatment,
who did not have corresponding
completion criteria.

Youth with different probabilities to recidivate and with different characteristics were accounted
for by creating a probability to recidivate variable for each youth based on individual
characteristics and history in TYC. The resulting probability was a control variable in the
analysis.

Texas Youth Commission

11

FY2009 Review of Agency Treatment Effectiveness

The probability of recidivating was calculated for each youth and an overall expected value
established for each groups. The difference between the predicted and actual values was the
means by which the effectiveness of treatment was determined. 2
The reason for controlling these differences is illustrated below in a hypothetical example. In this
example, Program B initially appears to have a lower recidivism rate than Program A. However,
the difference in program effectiveness is actually due to gender differences between programs
rather than treatment received.
Several factors underlie the probability that a youth will offend after release. For example, age
at first referral is highly associated with re-offending. Predicted rate is a scientifically credible
way to determine the likelihood of recidivism using known predictors such as age at first referral,
juvenile justice history, and gang membership.
In TYC, participation in treatment programs is only one factor that can have an impact on
lowering the probability of re-offending. In order to understand how much impact treatment has
on recidivism, other factors that are known predictors of recidivism must be taken into account.

P ROGRAM E FFECTIVENESS C ALCULATION E XAMPLE
Males

Females

Rearrest

Program

Total
Violent
Rate
Released Offense

2

Total

Rearrest
Total
Released

Violent
Rate
Offense

Rearrest
Total
Released

Violent
Offense

Rate

A

100

9

9.0%

100

3

3.0%

200

12

6.0%

B

100

9

9.0%

200

6

3.0%

300

15

5.0%

•

This hypothetical table shows two programs with slightly different rates for rearrest for a
violent offense. As can be seen in the far right column, Program A has a 6.0% rate,
compared to Program B, which has a rate of 5.0%. Not looking at the characteristics of
who is in the program, it could be concluded that the recidivism rate for A is 20% higher
than that of B (just like $6 is 20% more than $5).

•

However, both programs were equally successful with males (9%) and with females (3%).
Both programs had the same number of males (100), but Program B had more females
than did Program A (200 vs. 100). Therefore Program A was handicapped by having a
higher percentage of their releases being high risk youths, namely males.

•

A statistical program would demonstrate that the treatment effect of Program A compared
to Program B was 0%, while the effect of initial differences in youth characteristics between
the programs was 20%.

•

While Programs A and B are hypothetical, the difference in recidivism risk between males
and females is real. This example demonstrates why the comparative risk levels must be
taken into account when evaluating program effectiveness.

Probabilities were based on the actual recidivism rate of the group not receiving treatment and characteristics
empirically found in the cohorts to predict recidivism. Among the characteristics included in the specialized
treatment analysis were: age at commitment, age of first referral, classifying offense, citizenship, commitment
county, escape history, ethnicity, gender, self-reported gang membership, prior placements, prior felony
adjudications, prior felony referrals, prior violent offense referrals, prior probations, type of release program, prior
runaway referrals, specialized treatment needs, incidents during orientation and assessment, grade level at
commitment, criminal involvement by relatives, and total incidents reported prior to release. The predicted rate
statistically assumed that the cohort did not receive specialized treatment.

Texas Youth Commission

12

FY2009 Review of Agency Treatment Effectiveness

ANALYSIS
Youth who received treatment (treatment group) were compared with similar youth who did not
receive treatment. Each youth was tracked from the release date until the end of the tracking
period. For each measure of specialized treatment, the analysis addressed both the magnitude
of differences between groups and the probability of the differences occurring by chance. For
the latter, the smaller the probability, the more likely the difference arose from a real effect and
did not occur by chance. This calculation is illustrated below for the Sexual Behavior Treatment
Program using the outcome measure “One-year Rearrest Rate for Any Offense.”

Difference Calculation Example:
1 Year Rearrest for Any Offense
Sexual Behavior Treatment Program

33%

33%

33%
29%

Step 1: 4%
due to initial
differences
Step 2: 5%
due to
specialized
treatment

24%

100%

80%

60%

Step 3: 16%
reduction
due to
specialized
treatment

84%
remaining

40%

20%
0%

High Need, No Expected Specialized
Treatment
Value
Specialized
Group
Treatment
Treatment
Group

0%

0%

Step 1: Control for differences between groups due to factors other than specialized treatment.
Step 2: Calculate the difference in recidivism due to specialized treatment.
Step 3: Calculate the percentage reduction in recidivism due to specialized treatment.
Formula: % Reduction = Reduction due to specialized treatment (Expected rate minus Actual rate)/
Expected rate: (29.0% - 24.3%)/29.0% = 4.7%/29.0% = 16.2%

Texas Youth Commission

13

FY2009 Review of Agency Treatment Effectiveness

RESULTS: SEXUAL BEHAVIOR TREATMENT PROGRAM
As presented in the methodology section, data on youth with an initial release from secure
confinement over two different five year periods were tracked to calculate measures at one and
three years after release. Characteristics of each youth in the sexual behavior treatment
sample included sex offenders with an initial release from a secure program during the
established time frames.
The total sample for measures at one year post-release included 1,337 youth and for measures
at three years post-release included 1,091 youth.
Sexual Behavior Treatment Enrollment Rates
One Year
Cohort
Total Cohort
Total Enrolled
Enrollment Rate

Three Year
Cohort

1,337
556
42%

1,091
440
40%

The enrolled group made up the study’s treatment group while the not-enrolled group made up
the group with no specialized treatment. Controlling for statistical differences between the
groups, calculations were made regarding the treatment group’s expected outcomes,
probabilities of the outcomes occurring by chance, and the statistical significance of the
differences between the actual and expected values.
The following table summarizes the study results in this treatment category.
Sexual Behavior Treatment Effectiveness Results
High Need,
No
Specialized
Treatment

Arrest for Any Offense: 1 Year
Arrest for Violent Offense: 1 Year

32.9%

Treatment
Group

Expected
Value
Treatment
3
Group

Difference
between
Actual and
4
Expected

24.3%

29.0%

-16.2%

Specialized

Probability

5

Level of
Significance
6

*
**
ns
ns

5.4%

4.9%

2.9%

5.3%

-45.3%

3.7%

Incarceration for Any Offense: 1 Year

15.9%

11.9%

12.1%

-1.7%

90.2%

Incarceration for Any Offense: 3 Year

35.0%

25.5%

28.9%

-11.8%

21.8%

Findings:

3

4

5

6

ƒ

Youth who had participated in TYC sex offender treatment were 45% less likely than
expected to be rearrested for a violent offense and 16% less likely to be rearrested for
any offense within one year.

ƒ

The strongest result is that 2.9% of the Specialized Treatment Group was arrested for
a violent offense within one year, compared to the expectation that 5.3% of this group

The expected outcome when differences between the youth receiving and not receiving specialized treatment were
controlled.
The difference is the percentage difference between the actual and expected rate. Negative values indicate that
the treatment group had a lower number than expected as the outcome.
The percent of time the effect would have occurred by chance with everything else held constant. Lower values
indicate that the difference was due to the intervention while higher values indicate a higher likelihood that the
difference would have occurred without the intervention.
*** p< .01 ** p< .05 * p< .10
ns = not significant at .10

Texas Youth Commission

14

FY2009 Review of Agency Treatment Effectiveness

would be rearrested for a violent offense within one year. The results also showed that
24.3% of youth receiving treatment were arrested for any offense within one year,
compared to the expected 29%. Although 24.3% might still be considered a high
recidivism rate for rearrests, this result clearly demonstrates that the program had a
significant positive impact.
ƒ

Although the treatment group had different rates for the incarceration measures, the
expected values that were calculated while controlling for differences between the
groups were not significantly different from the actual results of the treatment group.
This indicates that the differences in these outcomes could have occurred by chance
at least 10 times in 100 had there been no specialized treatment effect.

ƒ

The sexual behavior treatment program has shown significant reductions in recidivism
in the last several studies.

Discussion
Youth who participated in the sex offender program and were released after June 2008 were not
included within the scope of this study because a full year of arrest and incarceration data was
not available. As a result, this study provides an analysis of pre-reform programming.
Study results continue to guide the agency as it builds and expands programs, and the sex
offender program has shown positive outcomes but has had limited access. During FY 2009,
the agency has focused on expanding access to this program.
Highlights of changes to the sex offender program:
•

Beginning September 2009 100% of newly-committed youth needing all levels of
specialized sexual behavior treatment have access to treatment, either through the
traditional intensive program or a new moderate intensive program. As a result, TYC
anticipates greater reductions in youth recidivism in future reports.

ƒ

In September 2009 an enhanced assessment element was added for this population.
The Juvenile Sex Offender Assessment Protocol – II is a nationally recognized
assessment tool that identifies treatment sensitive issues related to sexual offending.
This dynamic risk information will be used to determine population needs, guide
treatment interventions, and measure treatment effectiveness.

ƒ

In September 2009 moderate intensity programming was added and will be expanded in
FY 2010 for youth that do not require intensive residential treatment. Moderate intensity
programming is also offered to youth with a co-occurring mental health or other
specialized treatment need if they are primarily in need of a different program.

ƒ

By spring 2010 a more integrated model of providing needed treatment services after
release from a high restriction setting will be initiated. This model implements
community reentry plans after release, provides any continuing treatment needs, and
includes close collaboration with local partners.

ƒ

Texas Occupations Code requires that all TYC staff who provide sex offender
counseling to be Licensed Sex Offender Treatment Providers (LSOTP) by October 2010.
In January 2009 implementation of the development program began for licensing internal
staff and attracting external LSOTP candidates for these hard-to-fill positions. Five staff
received licensure in 2009 and the other candidates are targeted to attain licensure by
October 2010.

Texas Youth Commission

15

FY2009 Review of Agency Treatment Effectiveness

RESULTS: CAPITAL & SERIOUS VIOLENT OFFENDER TREATMENT PROGRAM
All youth in the cohorts who were capital offenders or in need of treatment based on other
serious violent offenses were included in this analysis. As presented in the methodology
section, data on youth with an initial release from secure confinement over two different five
year periods were tracked to calculate measures at one and three years after release.
Characteristics of each youth in the capital and serious violent treatment sample included
capital and serious violent offenders with an initial release from a secure program during the
established time frames.
The total sample for measures at one year post-release included 464 youth and for measures at
three years post-release included 522 youth.
C&SVO Enrollment Rates

Total Cohort
Total Enrolled
Enrollment Rate

One Year
Cohort
464
98
21%

Three Year
Cohort
522
104
20%

The enrolled group made up the study’s treatment group while the not enrolled group made up
the group not receiving specialized treatment. Controlling for statistical differences between the
groups, calculations were made regarding the treatment group’s expected outcomes,
probabilities of the outcomes occurring by chance, and the statistical significance of the
differences between the actual and expected values. The following table summarizes the
results in this specialized treatment category.
Capital & Serious Violent Offender Treatment Effectiveness Results
High Need,
No
Specialized
Treatment

Arrest for Any Offense: 1
Year

Specialized

Treatment
Group

Expected
Value
Treatment
7
Group

Difference
between
Actual and
8
Expected

Probability

9

Level of
10
Significance

27.9%

20.4%

20.1%

1.5%

94.4%

ns

Arrest for Violent
Offense: 1 Year

7.7%

2.0%

7.1%

-71.8%

6.0%

*

Incarceration for Any
Offense: 1 Year

4.6%

3.1%

3.9%

-20.5%

71.8%

ns

Incarceration for Any
Offense: 3 Year

23.2%

15.4%

10.2%

51.0%

33.8%

ns

7

The expected outcome when differences between the youth receiving and not receiving specialized treatment were
controlled.
8
The difference is the percentage difference between the actual and expected rate. Negative values indicate that
the treatment group had a lower number than expected as the outcome. Positive values indicate that the treatment
group had a higher number than expected as the outcome.
9
The percent of time the effect would have occurred by chance with everything else held constant. Lower values
indicate that the difference was due to the intervention while higher values indicate a higher likelihood that the
difference would have occurred without the intervention.
10
*** p<.01 ** p<.05 * p<.10 ns = not significant at .10
Texas Youth Commission

16

FY2009 Review of Agency Treatment Effectiveness

Findings:
ƒ

Youth enrolled in the Capital and Serious Violent Offender Program were 72% less likely
than expected to be arrested for a violent offense within one year. The arrest rate for the
treatment group was only 2%, while the expected arrest rate was 7.1%.

ƒ

Although the treatment group had different rates for the other three measures, the
expected values that were calculated while controlling for differences between the
groups were not significantly different from those that occurred without controls. This
indicates that the differences in these outcomes could have occurred by chance at least
10 times in 100 had there been no specialized treatment effect.

Discussion
Youth who participated in the C&SVO program and were released after June 2008 were not
included within the scope of this study because a full year of re-arrest and incarceration data
was not available. As a result, this study provides an analysis of pre-reform programming.
Study results continue to guide the agency as it builds and expands programs, and the C&SVO
program has shown positive outcomes but has had limited program access. During FY 2009,
the agency expanded access to this program.
Highlights of C&SVO program expansions include:
ƒ

In 2009 the program was expanded to provide treatment for girls at the Ron Jackson
State Juvenile Correctional Complex, Unit I. The program is designed to provide girls
the opportunity to develop skills that will aid in the reduction of risk factors, develop
problem-solving skills, address trauma-related issues, improve interpersonal functioning
and affect regulation, facilitate empathic development, and receive psycho-social
education programming on drug and alcohol and sexual and anger issues as needed.

ƒ

In November 2009 a program for medium need violent offenders was introduced.
Aggression Replacement Training® (ART®) is a nationally recognized program and has
been demonstrated as an empirically supported intervention program with violent
adolescents. The ART® treatment program has been implemented on eight TYC high
restriction campuses and will enhance the effectiveness of specialized services to violent
youth and those in need of intervention to address anger related behaviors. The
introduction of ART® will help to reach youth unable to participate in the time-intensive
C&SVO program.

Texas Youth Commission

17

FY2009 Review of Agency Treatment Effectiveness

RESULTS: CHEMICAL DEPENDENCY TREATMENT PROGRAM
As presented in the methodology section, data on youth with an initial release from secure
confinement over two different five year periods were tracked to calculate measures at one and
three years after release. Characteristics of each youth in the chemical dependency treatment
sample included: assessed as being chemically dependent with an initial release from a secure
program during the established time frames.
The total sample for measures at one year post-release included 4,100 youth and for measures
at three years post-release included 2,566 youth.
Chemical Dependency Enrollment Rates

Total Cohort
Total Enrolled
Enrollment Rate

One Year
Cohort
4,100
1,696
41%

Three Year
Cohort
2,566
1,188
46%

The enrolled group made up the study’s treatment group while the not enrolled group made up
the group not receiving specialized treatment. Controlling for statistical differences between the
groups, calculations were made regarding the treatment group’s expected outcomes,
probabilities of the outcomes occurring by chance, and the statistical significance of the
differences between the actual and expected values. The following table summarizes the results
in this specialized treatment category.

Chemical Dependency Treatment Effectiveness

Treatment
Group

Expected
Value
Treatment
11
Group

Difference
between
Actual and
12
Expected

57.3%

67.0%

61.5%

8.9%

1.4%

**

8.7%

11.3%

10.1%

11.9%

25.6%

ns

Incarceration Any Offense: 1
Year

21.1%

27.0%

23.8%

13.4%

1.6%

**

Incarceration Any Offense: 3
Year

42.0%

48.3%

43.0%

12.3%

1.8%

**

High Need,
No
Specialized
Treatment

Arrest for Any Offense: 1 Year
Arrest for Violent Offense: 1
Year

Specialized

Probability
13

Level of
14
Significance

11

The expected outcome when differences between the youth receiving and not receiving specialized treatment were
controlled.
12
The difference is the percentage difference between the actual and expected rate. Positive values indicate that the
treatment group had a higher number than expected as the outcome.
13
The percent of time the effect would have occurred by chance with everything else held constant. Lower values
indicate that the difference was due to the intervention while higher values indicate a higher likelihood that the
difference would have occurred without the intervention.
14
*** p<.01 ** p<.05 * p<.10 ns = not significant at .10
Texas Youth Commission

18

FY2009 Review of Agency Treatment Effectiveness

Findings:
ƒ

Youth who were enrolled in chemical dependency treatment were significantly more
likely to have been arrested within one year or reincarcerated within one or three years
of release from a secure program.

ƒ

Although the treatment group had a higher rate for the Arrest for Violent Offense
measure, the expected value that was calculated while controlling for differences
between the groups was not significantly different from the value that occurred without
controls. This indicates that the difference in the outcome could have occurred by
chance at least 10 times in 100 had there been no specialized treatment.

Discussion
Youth who participated in the chemical dependency program and were released after June
2008 were not included within the scope of this study because a full year of re-arrest and
incarceration data was not available. As a result, this study provides an analysis of pre-reform
programming.
Study results continue to guide the agency as it builds and expands programs, and the prereform chemical dependency program has not shown positive results. Significant alcohol and
drug abuse needs are extremely challenging conditions to treat effectively for long term
success. Current research indicates that inappropriate placement of youth in a level not
matched with their needs can be ineffective and lead to worse outcomes. Based on the 2008
report, the agency focused on modifying this program.
Highlights of chemical dependency program modifications:
ƒ

In January 2009 an evidenced-based treatment curriculum, Pathways to Self- Discovery
and Change: a Guide to Responsible Living by Dr. Harvey Milkman, was adopted and
implemented at TYC high restriction institutions. This program was chosen because it is
designed to work most effectively with chemically dependent incarcerated adolescents.

ƒ

In spring 2009 programming was expanded to halfway houses, providing short-term
counseling for youth with treatment needs. This initiative will individualize the continuum
of services for youth.

ƒ

In January 2010 a comprehensive assessment will be added to identify the level of
treatment need for youth with drug and alcohol use history. The Adolescent SelfAssessment Profile-II provides assessment information to target the specific risk and
protective factors associated with drug and alcohol use and abuse. It will also serve as
a measure of treatment progress.

ƒ

The 81st Legislature supported stronger reentry programming with an appropriation for
substance abuse treatment with Functional Family Therapy© (FFT©). The program is an
evidence-based model with proven success that will be initiated in FY 2010. The
program is a structured family intervention protocol for youth diagnosed with conduct
disorders, violent behavior and substance abuse that has been shown in multiple studies
to reduce felony recidivism. FFT© has three phases: motivate the family toward change;
teach the family how to change a specific critical problem; and help the family generalize
their problem-solving skills.

ƒ

In November 2009 enhancements to community reentry case plans were implemented.
As part of the development of this plan, the Regional Chemical Dependency Specialist
provides the youth’s case manager with information concerning support groups,
including Alcoholics Anonymous and Narcotics Anonymous, and aftercare providers in
the area where the youth will reside. The Regional Specialist also provides support to
the youth as needed when the youth returns to the community.

Texas Youth Commission

19

FY2009 Review of Agency Treatment Effectiveness

RESULTS: MENTAL HEALTH TREATMENT PROGRAM
All youth in the cohorts who had a high need for mental health treatment were included in this
analysis. Selection for this program was based on diagnoses and the Global Assessment of
Functioning (GAF) score.
As presented in the methodology section, data on youth with an initial release from secure
confinement over two different five year periods were tracked to calculate measures at one and
three years after release. Characteristics of each youth in the mental health treatment sample
included: assessed as ‘high need’ by TYC for specialized mental health treatment with an initial
release from a secure program during the established time frames.
The total sample for measures at one year post-release included 4,597 youth and for measures
at three years post-release included 4,388 youth.
Mental Health Treatment Enrollment Rates

Total Cohort
Total Enrolled
Enrollment Rate

One Year
Cohort
4,597
1,455
32%

Three Year
Cohort
4,388
1,207
28%

The enrolled group made up the study’s treatment group while the not enrolled group made up
the group not receiving specialized treatment. Controlling for statistical differences between the
groups, calculations were made regarding the treatment group’s expected outcomes,
probabilities of the outcomes occurring by chance, and the statistical significance of the
differences between the actual and expected values. The summary of results is contained in
the following table.

Mental Health Treatment Effectiveness Results
High Need,
No
Specialized
Treatment

Specialized
Treatment
Group

Expected
Value
Treatment
15
Group

Difference
between
Actual and
16
Expected

57.6%

50.1%

52.5%

-4.6%

12.7%

ns

9.0%

9.0%

8.1%

11.1%

32.8%

ns

Incarceration for Any Offense: 1
Year

24.0%

22.2%

23.5%

-5.5%

39.6%

ns

Incarceration for Any Offense: 3
Year

47.0%

43.7%

40.2%

8.7%

20.6%

ns

Arrest for Any Offense: 1 Year
Arrest for Violent Offense: 1
Year

Probability

17

Level of
18
Significance

15

The expected outcome when differences between the youth receiving and not receiving specialized treatment were
controlled.
16
The difference is the percentage difference between the actual and expected rate. Negative values indicate that
the treatment group had a lower number than expected as the outcome. Positive values indicate that the treatment
group had a higher number than expected as the outcome.
17
The percent of time the effect would have occurred by chance with everything else held constant. Lower values
indicate that the difference was due to the intervention while higher values indicate a higher likelihood that the
difference would have occurred without the intervention.
18
*** p<.01 ** p<.05 * p<.10 ns = not significant at .10
Texas Youth Commission

20

FY2009 Review of Agency Treatment Effectiveness

Findings:
ƒ

Although the treatment group had different rates for the four measures, the expected
values that were calculated while controlling for differences between the groups were not
significantly different from those that occurred without controls. This indicates that the
differences in these outcomes could have occurred by chance at least 10 times in 100
had there been no specialized treatment effect.

ƒ

The FY 2008 Review of Agency Treatment Effectiveness report included two very
significant positive one-year results in this treatment category for arrest and
incarceration for any offense. The treatment group in that study had significantly fewer
arrests and incarcerations following release compared to what was expected statistically.

Discussion
Youth who participated in the mental health program and were released after June 2008 were
not included within the scope of this study because a full year of re-arrest and incarceration data
was not available. As a result, this study provides an analysis of pre-reform programming.
Study results continue to guide the agency as it builds and expands programs, and the mental
health treatment program has not shown results. During 2009, the program was modified to
more fully integrate services with youth needs:
Highlights of the mental health program modifications include:
ƒ

February 2009 clinical staffings were enhanced to increase collaboration between the
psychology and contracted psychiatric staff. Staffings are designed to address the
medical and clinical aspects of the youth’s mental health issues and how best to
minimize their effects on the daily functioning of the youth.

ƒ

February 2009 multi-disciplinary team meetings and individualized case planning were
enhanced to have a greater focus on criminogenic needs (dynamic risk and protective
factors), to increase collaboration between staff disciplines, the youth, and the youth’s
family, and to increase individualization of programming for more effectives services to
youth in the program. The integrated plan is designed to help youth focus on reentry into
the community from the very beginning of their TYC commitment.

ƒ

October 2009 HHSC received a grant to assess the prevalence of brain trauma and
impact on youth within TYC and TJPC and county systems. Approximately 3,000 youth
per year will be assessed during the project period. Program implementation will be the
second phase of the grant.

ƒ

November 2009 services were added to address drug and alcohol risks, aggression
management issues, sex offender risks, and trauma related services in addition to the
mental health support groups for youth.

ƒ

February 2008 coordination of psychiatric services was enhanced and psychiatry was
integrated more effectively with other mental health services. For all youth, psychiatric
visits per youth increased about two-fold through FY 2009 from 0.65 psychiatric visits
per youth in the first quarter to 1.24 visits per youth in the fourth quarter. For youth
residing at Corsicana Residential Treatment Center, where youth with the highest level
of mental health needs are located, the number of psychiatric visits per youth increased
over the same period from 1.48 psychiatric visits per youth in the first quarter of FY 2009
to 2.90 psychiatric visits per youth in the fourth quarter.

Texas Youth Commission

21

FY2009 Review of Agency Treatment Effectiveness

RESULTS: FEMALE YOUTH
Females included in this analysis were a sub-group of the cohorts presented throughout this
report. All of the females, like the males, had a need for one or more type of specialized
treatment. Because of the lower number of female youth in some of the specialized treatment
need groups, results are provided for all females as a group. Additionally, TYC provided no
gender-specific programming during the study period. Therefore, the results in this category are
for those girls who received any type of specialized treatment, rather than for gender-specific
services.
As presented in the methodology section, data on youth with an initial release from secure
confinement over two different five year periods were tracked to calculate measures at one and
three years after release. Characteristics of each youth in the female treatment sample
included: assessed as high need by TYC for specialized treatment and initial release from a
secure program during the established time frame. The total sample for measures at one year
post-release included 1,253 youth and for measures at three years post-release included 1,050
youth.
Female Enrollment Rates

Total Cohort
Total Enrolled
Enrollment Rate

One Year
Cohort
1,253
560
45%

Three Year
Cohort
1,050
475
45%

Forty-five percent of females were enrolled in specialized programs in both the one and threeyear cohorts. The enrolled group of females made up the study’s treatment group while the notenrolled group of females made up the group not receiving specialized treatment. Controlling for
statistical differences between the groups, calculations were made regarding the treatment
group’s expected outcomes, probabilities of the outcomes occurring by chance, and the
statistical significance of the differences between the actual and expected values. The following
table reflects summary results in this treatment category.
Female Specialized Treatment Results
No
Specialized
Treatment

Arrest for Any Offense: 1 Year

Treatment
Group

Expected
Value
Treatment
19
Group

Difference
between
Actual and
20
Expected

Probability
21

Level of
22
Significance

31.7%

36.3%

35.2%

3.1%

70.5%

ns

3.0%

3.2%

3.2%

0.0%

100.0%

ns

Incarceration for Any Offense: 1
Year

12.6%

18.9%

16.2%

16.7%

22.9%

ns

Incarceration for Any Offense: 3
Year

27.5%

28.4%

27.6%

2.9%

77.0%

ns

Arrest for Violent Offense: 1 Year

19

The expected outcome when differences between the youth receiving and not receiving specialized treatment were
controlled.
20
The difference is the percentage difference between the actual and expected rate. Positive values indicate that the
treatment group had a higher number than expected as the outcome.
21
The percent of time the effect would have occurred by chance with everything else held constant. Lower values
indicate that the difference was due to the intervention while higher values indicate a higher likelihood that the
difference would have occurred without the intervention.
22
*** p<.01 ** p<.05
* p<.10
ns = not significant at .10
Texas Youth Commission

22

FY2009 Review of Agency Treatment Effectiveness

Findings:
ƒ

The expected values that were calculated while controlling for differences between the
groups were not significantly different from those that occurred without controls. This
indicates that the differences in these outcomes could have occurred by chance at least
10 times in 100 had there been no specialized treatment effect. This means that the girls
released during the scope of the study who had received any type of specialized
treatment were about as likely to be arrested or incarcerated within one or three years as
girls who did not receive any specialized treatment.

Discussion
Youth who participated in the specialized treatment programs and were released after June
2008 were not included within the scope of this study because a full year of re-arrest and
incarceration data was not available. As a result, this study provides an analysis of pre-reform
programming. Girls received no gender-specific services during the study period. As stated
earlier, girls are included as a separate category due to interest in the effectiveness of treatment
for them as a group.
Study results continue to guide the agency as it builds and expands programs, and new genderspecific programming does not yet show results. Girls typically respond differently to treatment
than boys, and the agency continues to research the ways in which its specialized treatments
will be more effective for girls, who comprise a small percentage of the total youth population.
During 2008-09, the agency enhanced girls programming to improve female offender outcomes.
Highlights of gender-specific treatment enhancements for girls:
ƒ

In December 2008 Capital and Violent Offender programming began for the girls at Ron
Jackson Juvenile Correctional Complex Unit I.

ƒ

In February 2009 multi-disciplinary team meetings and individualized case planning were
enhanced to have a greater focus on criminogenic needs (dynamic risk and protective
factors), to increase collaboration between staff disciplines, the youth and the youth’s
family, and to increase individualization of programming for girls.

ƒ

In December 2009 staff at the female facilities was trained in the Girls Circle program.
Staff will begin implementing this ten week program for specifically identified girls
beginning in January 2010. The Girls Circle is a structured support group for girls that
focuses discussion on gender-specific topics designed to promote resiliency and selfesteem.

ƒ

In December 2009 staff at Ron Jackson Unit 1 received Gender Responsive training to
ensure that staff more effectively meets the special needs of this population.

ƒ

In December 2009, clinical staff at Ron Jackson Unit 1 received training in TraumaFocused Cognitive Behavioral Therapy (TF-CBT). This therapy is an evidence-based
treatment approach for assisting children, adolescents, and their caretakers to overcome
trauma-related challenges. It is designed to reduce negative emotional and behavioral
responses following child sexual abuse and other traumatic events. This training
enhanced the services already provided to girls to resolve trauma issues. In December
2009, eight girls were enrolled in a trauma resolution group.

Texas Youth Commission

23

FY2009 Review of Agency Treatment Effectiveness

CONCLUSION
Results of the FY 2009 Review of Agency Treatment Effectiveness provide important information as
the agency reshapes its specialized treatment programs for continuous and long term improvement
consistent with legislative intent for fundamental reform. Significant differences in recidivism rates
indicated that certain intensive specialized treatment programs reduced recidivism more than the
agency’s core treatment program alone, even when statistically controlling for initial differences
between the groups. Two specialized treatment programs resulted in significantly lower recidivism
rates for sexual offenders and for capital and serious violent offenders.
Although the scope of the report analyzed data from a pre-reform period and program changes are
well underway, the results are still useful in numerous ways. Pre-reform sex offender treatment was
effective, and the agency is building on those successful program elements to strengthen and
expand sex offender treatment services. The study also validated results from earlier reports that
showed pre-reform programming for alcohol and drug abuse, mental health needs, and non-genderspecific treatment for girls was not effective. As reflected in the earlier discussion section for each
result category, intensive and specific agency efforts to get the specialized services in these
treatment categories back on track have been underway since 2008.
For many reasons, recidivism is an imperfect measure of treatment effectiveness and public safety.
But it’s also the best one available for consistent data comparisons over long time periods. The
agency is exploring alternatives for supplementing recidivism data in the next annual report with
other measures, including educational achievement indicators, that reflect the effectiveness of the
different types of specialized treatment. Youth rehabilitation and public safety are intricately
intertwined, and the mixed results of this report highlight the agency’s continuing challenges.
The individual and social problems addressed by TYC specialized treatment programs are
historically intractable, complex, and typical of the issues in many juvenile justice systems. Most
TYC youth have had prior interventions through the juvenile justice or other state and local systems
that were not successful in preventing the offenses that resulted in TYC commitment. As TYC
continues to receive the State’s most serious offenders with increasingly complex treatment needs,
long term solutions are expected to emerge from evidence-based programs currently operating
across many jurisdictions. As refined data becomes increasingly available demonstrating what does
and does not work, the agency is hopeful about building on incremental program successes for
continuous improvement in each area. Although resource constraints will continue to be challenging
in a recovering economy, program excellence is still achievable within those limitations. The agency
strives to reclaim a national reputation for effective youth rehabilitation and public safety.
TYC will continue to focus on improving treatment effectiveness with the stability and predictability of
a three-legged stool. Ensuring that program models are well-matched to youth needs based on
periodic individual assessment data is one foundational leg. Another leg concentrates agency
initiatives on re-entry programming for youth preparing to be released from TYC. The third leg is
continuous investment in staff development to expand knowledge, skills, and abilities to manage
increasingly complex youth needs in an increasingly complex and fast changing correctional
environment.
During FY 2009 the Commission made significant progress in advancing reform efforts by evaluating
and enhancing its programs and operations. Many challenges and much work remain, but the
Commission is beginning to see improvements statewide. Demonstrating evidence-based results of
agency efforts will take time when measured on one-year and three-year recidivism performance
scales.

Texas Youth Commission

24