Close
About
FAQ
Home
Collections
Login
USC Login
Register
0
Selected
Invert selection
Deselect all
Deselect all
Click here to refresh results
Click here to refresh results
USC
/
Digital Library
/
University of Southern California Dissertations and Theses
/
Stage of readiness and learning styles -- a tailored HIV/AIDS prevention intervention for youth detained in the Los Angeles County Juvenile Justice system
(USC Thesis Other)
Stage of readiness and learning styles -- a tailored HIV/AIDS prevention intervention for youth detained in the Los Angeles County Juvenile Justice system
PDF
Download
Share
Open document
Flip pages
Contact Us
Contact Us
Copy asset link
Request this asset
Transcript (if available)
Content
STAGE OF READINESS AND LEARNING STYLES -
A TAILORED HIV/AIDS PREVENTION INTERVENTION FOR
YOUTH DETAINED IN THE
LOS ANGELES COUNTY JUVENILE JUSTICE SYSTEM
by
Nikki Shipley
______________________________________________________________________
A Dissertation Presented to the
FACULTY OF THE GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
PREVENTIVE MEDICINE--HEALTH BEHAVIOR
December 2008
Copyright 2008 Nikki Shipley
ii
DEDICATION
This is dedicated to the men I love,
Bobby and Taylor -
my tenacity and inspiration.
iii
ACKNOWLEDGEMENTS
During my doctoral training I have had many friends and colleagues who have
encouraged and supported me throughout the years. If not for Dr. Carl Anderson
Johnson who gave me the opportunity and Phyllis Paxton and Mark Casanova who
supported this effort by giving me the flexibility to further my academic aspirations
while limiting the financial burden to my family, I would never have been able to fulfill
my goal of obtaining my doctorate.
I can’t say enough about how appreciative I am of the members of my
Dissertation Committee, Drs., Thomas Valente, Donna Spruijt-Metz, Sheila Murphy,
and especially Lourdes Baezconde-Garbanati along with Kim Reynolds for all their
assistance and guidance in developing this manuscript.
Without the expertise, wisdom, and advice of my colleagues, and friends Dr.
Selena Nguyen-Rodriguez, Dianne Jackson, and Marny Barovich I would never have
seen this to fruition.
To my family, Beverly Dyck, the Goldsteins, the Foremans, the Melameds, the
Hawks, the Stapletons, and Cheryl and Eric who were always ready to have us over for
Holidays or just for a nice evening while I was inundated with this work, I thank you.
Finally to my son Taylor and husband Bobby, I offer one promise, to always
support you in whatever endeavor you choose to partake as you two have done for me.
I assure you that I will turn the limitations I brought to our family over the past years
into abundance in the future.
iv
TABLE OF CONTENTS
Dedication ii
Acknowledgements iii
List of Tables v
List of Figures vi
Abstract vii
Chapter 1 INTRODUCTION 1
Chapter 2 LITERATURE REVIEW 4
Chapter 3 METHODS 26
Chapter 4 RESULTS 48
Chapter 5 DISCUSSION and CONCLUSION 71
Bibliography 85
APPENDICES 93
Appendix 1 Intervention Instruments for Data Collection 94
Appendix 2 Comparison Group Instruments For Data Collection 132
Appendix 3 Rankings of HIV Prevention Interventions for Incarcerated
Youth 143
Appendix 4 Results of the Validity Study of the Intervention Curriculum 145
v
LIST OF TABLES
Table 2-1 List of Identified Prevention Programs Targeting Incarcerated Youth 11
Table 2-2 Summary of Evaluation Outcome of the Fourteen Behavior Change
Programs for Incarcerated Youth 14
Table 2-3 Self-Reported Risky Behaviors Among Los Angeles County
Incarcerated Youth 15
Table 2-4 Stages Of Change (SOC) 18
Table 2-5 The Learning Preference Inventory 21
Table 3-1 Common Items in Intervention and Comparison Pre & Post Surveys 42
Table 3-2 Schedule of Evaluation Data Collection for the Two Study Groups 44
Table 3-3 Sample Size Calculations 45
Table 3-4 Total Variance Explained by the Factors 45
Table 3-5 Rotated Factor Pattern 46
Table 4-1 Response Rates of Study Groups 48
Table 4-2 Demographic Characteristics of the Study Groups 50
Table 4-3 Difference in Mean Changes Within Each Study Group from Pre to
Post 59
Table 4-4 Differences Between Study Groups Changed Means from Pre to Post 61
Table 4-5 Association Between Change in Outcomes and SOC at Baseline 65
Table 4-6 Predictors of Decrease In Reporting An STD Since Intervention At
Post Intervention 66
Table 4-7 Predictors of Increase in Condom Behaviors at Post-Intervention 66
Table 4-8 Predictors of Reduction in Risk Behaviors at Post-Intervention 67
Table 4-9 Predictors of Increase in Knowledge at Post-Intervention 67
vi
LIST OF FIGURES
Figure2-1 The Intervention Model 24
Figure 3-1 Research Design 26
Figure 3-2 Algorithm Used in Intervention Group to Stage Each Youth’s
“Stage of Change” in Sessions 1, 3, & 6 32
Figure 3-3 Distribution of Youth in The Juvenile Justice System of Los
Angeles County 35
Figure 3-4 Intervention Participant Selection 37
Figure 4-1 Differences in Response Between the Two Study Groups –
Baseline Condom Use 51
Figure 4-2 Differences in Response Between the Two Study Groups
- Baseline Knowledge 52
Figure 4-3 Differences in Response Between the Two Study Groups
- Baseline HIV/AIDS Risk Behaviors 53
Figure 4-4 Differences in Response Between the Two Study Groups
- Baseline Stage of Change 54
Figure 4-5 Differences in Response Between the Two Study Groups
- Post-Intervention Condom Use 55
Figure 4-6 Differences in Response Between the Two Study Groups
- Post-Intervention Knowledge 56
Figure 4-7 Differences in Response Between the Two Study Groups
- Post-Intervention HIV/AIDS Risk Behaviors 57
Figure 4-8 Differences in Response Between the Two Study Groups
- Post-Intervention Stage of Change 58
Figure 4-9 Differences in Mean Change of SOC Between the Two
Study Groups 63
Figure 5-1 The Revised Intervention Model 74
vii
ABSTRACT
The prevalence of HIV/AIDS is very low among incarcerated youth (~ 1%)
compared to the incarcerated adult population (4% to 6%). However, these youth report
practicing many behaviors, such as unprotected sex with multiple partners and sharing
of tattoo needles that puts them at risk for infection. To reduce the risk-taking
behaviors of youth detained by the Los Angeles County Juvenile Justice system an
HIV/AIDS prevention intervention tailored to learning styles and stages of change
(SOC) was developed. This study used an observational, prospective parallel group
design with historical comparisons and a randomized intervention sample to test the
effectiveness of this program. The magnitude and consistency of the results make it an
important contribution to research on HIV/AIDS prevention among incarcerated youth.
Results showed that a participant in the intervention program was 8 times more
likely to report using a condom last time he had sex (p < .001) and 10 times more likely
to report not having sex in the last 3 - 6 months since the program (p < .001) than a
participant in the comparison group. This reflects a substantial increase in protective
behaviors by decreasing sexual exposure, the primary form of transmitting HIV in this
population.
The intervention program included proactive strategies that focused efforts on
the individual’s current SOC thereby maximizing opportunities to assist youth in
reducing their risk behaviors. Results confirm the effectiveness of this type of strategy:
viii
there was a significantly larger increase in the proportion of intervention participants in
the “action” stage than in the comparison group at post-intervention. Importantly, this
may be the first data to explore the possible mediation of SOC in a population of
incarcerated youth at significantly increased risk for exposure to HIV. It provides
support for using an intervention tailored to individual learning style and stage of
change to teach youth who may not otherwise have been approachable, to change their
behaviors thereby reducing their risk of becoming HIV+ in the future.
1
CHAPTER ONE: INTRODUCTION
An estimated 2.3 million youth (Snyder & Sickmund, 2006) in the United States
are detained by the Juvenile Justice system each year of which over 39,000 reside in
Los Angeles County. The majority practice numerous HIV risk behaviors such as
unprotected sex and multiple sex partners on a regular basis. During their detention
many report contemplating lifestyle changes. But, even when they seem to understand
that certain activities put them at high risk of HIV infection, they often lack the skills
and psychosocial support to change. Still more disconcerting is the challenge of
maintaining these changes once released to return to the same community from which
they came. Therefore, developing effective interventions for reducing and sustaining
risk reduction behaviors is crucial for this population.
It has been repeatedly demonstrated that HIV/AIDS knowledge is necessary but
not sufficient to motivate adolescents to adopt preventive measures (Anderson, Kann,
et. al., 1990; Diclemente, 1991; Fisher & Fisher, 1992b; Stott, Kinnersley, & Rollnick,
1994). Therefore, prevention intervention programs must do more than increase
knowledge. Reports of the incarcerated population are in agreement with these
observations; these youth are knowledgeable about HIV/AIDS but that knowledge does
not seem to translate into safer sex practices (Forrest & Tambor, 2000). In spite of this,
a study conducted by the National Institute of Justice and the CDC found that the
majority of the juvenile facilities report HIV/AIDS information as part of general
health education rarely included behavior change strategies (Flores, 2003).
In 1999, the results of a needs assessment (Table 1-1) portrayed the lifestyle of
youth detained by the Los Angeles County Juvenile Justice system (Morris, 1999).
2
These findings agreed with the above studies; the basic information being provided to
them was not enough to cause these youth to take preventive measures to protect
themselves from being infected with HIV. The local community-based organization
providing the program decided that they needed to implement an intervention using a
theoretical model of change. Two important constructs were used to “tailor” the
intervention:
Stages of Change (SOC): Prochaska and DiClemente's "Stages of Change Theory"
suggests that people do not change their behaviors at once but move successively
through five stages (DiClemente, Prochaska, Fairhurst, & Velicer, 1991; Prochaska,
DiClemente, & Norcross, 1992; Prochaska, 1994; Prochaska & Norcross, 2001).
Yet, many intervention programs are implicitly designed for individuals who are not
prepared to change their lifestyle behaviors (Bandura, 1984). The most effective
strategies should be tailored differently for people in different stages. Therefore,
throughout the intervention SOC was used to determine each youth’s readiness to
adopt safer sexual practices as defined by consistent condom use. The goal was to
move them along the continuum into the next stage toward maintenance.
Learning Preferences Style - To assess each participant’s educational strengths and
preferred methodology for learning the Hanson, Silver & Strong Learning
Preference Inventory (LPI) (Silver & Hanson, 1992) was used. All of the learning
preferences were integrated into the intervention activities to ensure that each
participant received the information in their preferred context.
The purpose of this study was to test the efficacy of the tailored prevention
intervention compared to the original basic HIV/AIDS education program in increasing
3
knowledge, and reducing risk-taking behaviors among participating youth incarcerated
in Los Angeles County. This study will determine whether the intervention was more
efficacious then the original (comparison).
RESEARCH HYPOTHESES AND QUESTIONS
H
1
: There will be a greater level of HIV knowledge at post-test among the
intervention group when assessed against the comparison group;
H
2
: There will be a greater reduction of HIV risk-taking behaviors from pre- to
post-test among the intervention group when assessed against the comparison
group;
H
3
: There will be a greater reduction of HIV risk-taking behaviors from pre- to
post-test among the intervention group regardless of Stage of Change when
assessed against the comparison group;
RQ
1
: What are the complexities and disadvantages of implementing behavior change
interventions within Juvenile Justice Facilities?
RQ
2
: What policy changes could be made to enable and enhance the impact of such
behavior change interventions within Juvenile Justice Systems?
4
CHAPTER TWO: LITERATURE REVIEW
On any given day, there are some 3,600 correctional facilities housing more than
110,000 juvenile offenders throughout the United States (Sickmund, 2002). It is
estimated that 2.3 million youth are detained by the U.S. Juvenile Justice System each
year (Sickmund, 2002). The purpose of these facilities, defined as “residences built
with construction fixtures and/or staffing models designed to restrict the movements
and activities of juveniles or other individuals” is to house youth during judgment and
after the Court’s conviction of a criminal offense (Sickmund, 2002). However, nothing
other than their purpose is uniform in the U.S. youth detention system; discrepancies
range from the size of the facilities, to the type and location of the building structure,
the administrative responsibilities and the type of programming. Some are very large
“military-style boot camps” while others are smaller occupational skill-building schools
and still others are small family-size group homes. These differences can be found not
only between states but also within a state or jurisdiction.
A. Incarcerated Youth Throughout the U.S. - A Population at Risk of HIV/AIDS
More than 18% of all new cases of AIDS reported in the U.S. were between 20
to 29 years of age. Since the average incubation period from HIV infection is 10 years,
the majority of these individuals were, in all probability, infected during their
adolescence. And, although the prevalence of HIV/AIDS (~1%) is low among
incarcerated youth compared to the adult jailed population (~4% to 6%), a number of
studies have depicted a lifestyle that includes the common practice of many behaviors
that put them at high risk of HIV exposure. The regularity of unprotected sex with
multiple partners and survival sex, along with sharing of needles for tattooing and body
5
piercing in conjunction with substance and alcohol abuse reported among this group
puts them at high risk for becoming infected in the future as they venture beyond their
immediate neighborhoods into other communities (Braithwaite, Robillard, Woodring,
Stephens, & Arriola, 2001; DiClemente, Lanier, Horan, 1991; Lanier, DiClemente, &
Horan, 1999; Morris et al., 1995; Morrison et al., 1994; Robertson & Levin, 1999).
Youth detained or incarcerated by the Juvenile Justice system are struggling
with many vital and life threatening issues including, but not limited to, being at high
risk for HIV infection. However since the majority of this population report being
gang-affiliated or gang-affected they worry most about dying while still young. When
asked what age "young meant,” they answered "less than 21 years of age” (Deering,
1993; Herman-Shipley, Morris, Jackson, & Valente, 2006).
1. Risk Behaviors
The majority of studies pertaining to incarcerated youth and HIV/AIDS are
epidemiological in nature and are used to determine the level of HIV/AIDS knowledge,
attitudes toward condom use, and the prevalence of sexual behaviors and substance
abuse among this population. They describe a group that is in distinct contrast to
mainstream youth. The largest such study was conducted at 39 correctional facilities
located throughout the U.S. (Morris et al., 1995). This large scale survey of detained
youth reported high rates of behaviors that placed them at risk of injury and disease
including early onset of sexual intercourse, irregular or no condom use, alcohol and
drug use during sex, and sharing of tattoo needles.
Sexual Behaviors: Early onset of sexual intercourse was prevalent throughout
the studies on incarcerated youth whose age ranged from as early as 11.9 years (Morris
6
et al., 1995) to 13 years (Harwell, Trino, Rudy, Yorkman, & Gollub, 1999). This
equates to early exposure to STIs including HIV. In addition, once sexually active,
these youth report multiple sex partners averaging 2 partners/year; 8 to 17.9 lifetime
partners during their adolescence along with high rates of anal intercourse (Braithwaite
et al., 2001; DiClemente, Lanier & Horan, 1991; Lanier, DiClemente, Horan, 1999;
Morris et al., 1995; Morrison, 1994; Robertson & Levin, 1999). One of these studies
compared incarcerated youth with a public school sample in San Francisco and found
similar levels of HIV/AIDS knowledge but the incarcerated youth were less likely to
know about prevention strategies particularly the protective values of condoms
(DiClemente, 1991).
Use of condoms is sporadic and inconsistent among the incarcerated youth
population (Canterbury et al., 1995; Gillmore, Morrison, Lowery, & Baker, 1994;
Harwell et al., 1999; Morrison et al., 1994). In one study, only 20% of the sexually
active detained youth in Philadelphia reported “using a condom 50% of the time” while
14% said they “had never used one.” Although there was some variation between the
sexes; girls are more likely to report using condoms when compared to boys. In fact,
incarcerated youth in a study conducted in Seattle, Washington reported that the
negative outcomes of unprotected intercourse were overshadowed by the perceived
costs of safer sex practices such as, reduced sexual pleasure, loss of spontaneity, or
partner rejection (Morrison et al., 1994).
Other studies report that these adolescents do not perceive themselves at risk of
HIV, so do not deem it necessary to take precautions (Robertson & Levin, 1999). But a
study of condom use among criminally-involved adolescents in New York City found
7
an association between the sexual partner and condom use; condoms were used more
frequently with a casual partner and rarely for oral or anal sex (Magura, Shapiro, &
Kang, 1994). However, Robertson et al (1999) did find that condom use at first
intercourse was a strong predictor of condom use during later sexual activity among
juvenile offenders in Mississippi. This indicates that early intervention, prior to sexual
initiation, might be a useful prevention strategy for this group.
A number of health markers substantiate these claims of risky sexual behaviors.
A prevalence study of STIs conducted on incarcerated male youth in Birmingham,
Alabama agreed with these patterns of unprotected sex; some of the highest rates of
gonorrhea and/or Chlamydia (18%) were diagnosed among the incarcerated youth
population 14 to 18 years of age(Godin et al., 2003). A study of adolescents in juvenile
detention facilities in Chicago detected similar findings. Moreover, the females in this
study were 3 times more likely to have an STI than the males.
Alcohol and Drug Abuse: In the National study by Morris, and others (1995),
alcohol use was universal by age 15; with no difference in amount of drinking between
correctional facilities or genders. In another study conducted with Los Angeles County
incarcerated youth (Morris, Baker, Valentine, & Pennisi, 1998) found that type of drug
use was highly race specific, with the exception of alcohol and marijuana which were
commonly used across all races. A more recent study by Castrucci and Martin (2002)
found no significant difference between genders in the prevalence of substance use
except for regular cocaine use among incarcerated youth in North Carolina; Females
were four times more likely to use cocaine on a regular basis than males. For the group
as a whole it was found that those reporting regular use of two or more substances were
8
three times more likely to report inconsistent condom use and 11 times more likely to
report having sex with multiple partners. In general, these studies agree that the
frequent use of drugs and alcohol in conjunction with sexual activity make condom use
less likely. Therefore, effective interventions must address the association between
sexual behaviors and substance use.
Needle Sharing: While substance use is very high among incarcerated youth the
prevalence of injection drug use (IDU) is very low (Morris, 1999; Morris et al., 1998)
thereby minimizing the concern for sharing of dirty needles. However the popularity of
piercing and tattooing is often ignored. Over 21% of incarcerated youth in Atlanta,
Georgia reported that they had gotten a tattoo by a non-professional (Braithwaite et al.,
2001). In addition, it was reported that substance use was significantly associated with
piercing by needle sharing while under the influence of hallucinogens (i.e., LSD, acid),
speed, uppers, downers, tranquilizers and PCP. They also found that gang members are
more likely to have a non-professional tattoo while females are more likely to have
body piercing done by a non-professional. Even though this population’s knowledge
pertaining to the transmission of HIV is typically high (Lanier, DiClemente, & Horan.,
1999; Lanier, Pack, & DiClemente, 1999) they are often unable to make the association
between one behavior that they know is high risk, such as sharing needles for IV drug
use (a behavior few practice) with another behavior practiced by many - sharing needles
for tattoos (Braithwaite et al., 2001; Herman-Shipley et al., 2006)
2. Barriers To Change
A substantial portion of this population understands that many of their behaviors,
especially having unprotected sex with multiple partners, put them at high risk for HIV
9
infection. But studies report that they lack the skills and psychosocial support necessary
to change and sustain risk-reduction behaviors (Deering, 1993; Grimley et al., 2000;
Hair & Jager, 2002). Even when health information reaches these youth and motivates
them to initiate positive lifestyle changes during their incarceration, there are many
barriers that make it difficult to sustain these changes once released back into their
community with the daily threats of the environment.
First, they lack both the parental guidance and family support for avoiding risky
behaviors; many of their parents are involved in these behaviors themselves; abuse of
alcohol and intravenous (IV) drug abuse is prevalent among incarcerated youths’
parents (Deering, 1993). Second, many youth find themselves with large amounts of
unsupervised time in communities with limited access to healthy after-school activities
such as team sports (Flores, 2003). Informed about the consequences of risky
behaviors, but deficient in the interpersonal skills needed for navigating peer pressure,
many youth digress from their resolution to change. Should they decide to seek
assistance, the lack of resources and services in the community to support their
determination make it almost impossible to sustain the change (Deering, 1993).
Therefore, programs that include psychosocial supportive interaction could not only
offer them skills to help reduce their risk taking behaviors but could provide a
constructive experience for many facets of their lives.
B. HIV Prevention Interventions for Incarcerated Youth
1
HIV/AIDS knowledge is necessary but not adequate to motivate youth to
increase preventive behavior (Anderson, Kann, et. al., 1990; DiClemente, Lanier, &
Horan, 1991; DiClemente, Zorn, & Temoshok, 1986; Fisher & Fisher, 1992b).
10
HIV/AIDS knowledge is no longer believed to be strongly related to risk-taking
behaviors (Greico, 1987; Roscoe & Kruger, 1990; Ross & Rosser, 1989; Stevenson &
DeBord, 1988), but only weakly related (Anderson et al., 1990; DiClemente., 1991). A
study by Morrison suggested that there is a threshold of knowledge beyond which
additional education is of little benefit (Morrison, Baker, & Gillmore, 1994). Therefore,
an HIV/AIDS program is not likely to produce reductions in risk-taking behaviors
unless it does more than increase knowledge. Reports on HIV/AIDS prevention
programs for the incarcerated population show agreement with these observations;
although these youth are very knowledgeable about HIV/AIDS that knowledge does not
translate into safer practices (Forrest & Tambor, 2000).
In spite of this, a study conducted by the National Institute of Justice and the
Centers for Disease Control and Prevention found that of the 3,600 Juvenile
Correctional Institutes throughout the U.S., only 2% (87) responded and reported some
type of an HIV/AIDS program as part of their facility’s school curriculum (Flores,
2003). The majority of these facilities reported that this information was part of their
general health education curriculum and rarely included behavior change strategies.
Table 2-1 depicts the limited number (14 of 87 or 16%) of programs that
published/reported integrating theories of behavior change. Furthermore, only 11 of the
87 interventions (13%) included any type of evaluation.
11
Table 2-1. List of Identified Prevention Programs Targeting Incarcerated Youth
1
First
Author Citation
Lanier M. Lanier and B. McCarthy (1989). AIDS awareness and the impact of AIDS
education in juvenile corrections. Criminal Jus & Behav; vol 16(4):395-411.
JWCH
Institute
Communication with JWCH Institute, Inc. HIV/AIDS Video Project; JWCH (ongoing
program since 1990): unpublished data.
JWCH
Institute
Communication with JWCH Institute, Inc. Project IMPACT-Incarcerated Minors
Preventing AIDS Compassionately Together (1993); Unpublished.
Magura S. Magura, S. Kang, and J. Shapiro (1994). Outcomes of intensive AIDS
education for male adolescent drug users in jail. J of Adolesc Hlth; vol 15: 457-
463.
Horan P.F. Horan and D.J. Barthlow (1995). Adolescent Peers – Peer programs for HIV
Prevention by and for incarcerated adolescents. Peer Facilitator Quarterly; vol
13(1) Fall: 29-32.
JWCH
Institute
R. Morris (1996). Peer Vs. Facilitator - HIV Prevention. Presented at the Int’l.
AIDS Conference.
Gillmore M.R. Gillmore, D.M. Morrison, C.A. Richey, M.L. Balassone, L. Gutierrez, M. Farris
(1997). Effects of a skill-based intervention to encourage condom use among high
risk heterosexually active adolescents. AIDS Ed & Prev; vol 9 (supp A): 22-43.
St.
Lawrence
J.S. St Lawrence, R.A. Crosby, L. Belcher, N. Yazdani, and T.L. Brasfield (1999).
Sexual risk reduction and anger management interventions for incarcerated male
adolescents: A randomized controlled trial of two interventions. J of Sex Ed and
Therapy; vol 24(1): 9-17.
Schlapman N. Schlapman and P.S. Cass (2000). Project: HIV prevention for incarcerated
youth in Indiana. J Community Hlth Nursing; vol 17(3); 151-158.
Clark J. N. Clark, RN Van Eck, A. King, B. Glusman, A. McCain Williams, S. Van Eck, F.
Beech (2000). HIV/AIDS education among incarcerated youth. J of Criminal
Justice; vol 28:415-433.
Shelton D. Shelton (2001). AIDS and drug use prevention intervention for fined youthful
offenders. Iss Mental Hlth Nursing; vol 22(2):159-172.
Peres C.A. Peres, R.A. Peres, F. de Silveira, V. Paiva, E. S. Hudes, and N. Hearst
(2002). Developing an AIDS prevention intervention for incarcerated male
adolescents in Brazil. AIDS Ed & Prev; vol 14(supp B): 36-44.
Godin G. Godin, F. Michaud, M. Alary, J. Otis, B. Masse, C. Fortin, M-P Gagnon. H
Gagnon (2003). Evaluation of an HIV and STD prevention program for
adolescents in juvenile rehabilitation centers. Hlth Ed & Behav; vol 30 (5): 601-
614.
Herman-
Shipley
N.A. Herman-Shipley, R. Morris, D.V. Jackson T.W. Valente, and P.A. Paxton
(2005). A Tailored HIV Intervention for Incarcerated Youth – Did It Work? In
preparation for submission.
Studies for the literature review were retrieved using the major descriptor key
words: health behavior, health education or health promotion AND the minor descriptor
12
key words: incarcerated youth, high risk youth, incarcerated adolescents, incarcerated
teens AND HIV, AIDS, Acquired Immune Deficiency Syndrome; AND identifiers:
intervention, prevention, risk reduction, risk-taking behaviors (1985 to 2006) from:
1. Computerized literature searches of AHSearch, AltPressIndex,
AltPressIndexArchive, APA PsycARTICLES Direct, ArticleFirst, ASTA,
BooksInPrint, CINAHL, Cochrane Library, ConsumerIndex, Current Contents
Connect, CWI, Dissertations, Ebooks, EducationAb, ERIC ,EssayGenLit,
Expanded Academic ASAP, FactSearch, GPO, HealthSTAR, INSPEC, LEXIS-
NEXIS, MEDLINE, OJJDP Publications, PAISArchive, PaperFirst, PerAbs,
Proceedings, PsychINFO, PsychARTICLES, ReadersGuideAbs, Science
Citation Index Expanded, Sociological Abstracts, Web of Science,
WilsonSelectPlus and WorldCat; and,
2. Reference sections from empirical and review articles; and,
3. Unpublished programs presented at conferences.
The following describes the 14 behavior change programs:
Program Development: All but one of the fourteen programs identified using
accepted behavioral theories and/or models such as, Stages of Change, Theory of
Reasoned Action, the Health Belief Model, or the AIDS Risk-Reduction Model
(ARRM), to guide the program design. Of these fourteen, six (43%) utilized modified
curriculums originally developed either for teen pregnancy prevention (2 programs),
HIV prevention for incarcerated adults (4 programs), or promoted abstinence (2
programs). Seven of the fourteen (50%) programs certified participating youth as AIDS
13
educators who could “spread the word” to their peers both within the correctional
facility and once released, continue into their community.
Program Participants: The number of participants in the fourteen programs
ranged from 36 to 925 with an average of about 334 incarcerated youth participating in
some or all of a behavior change intervention. The incarcerated youth were selected for
the fourteen programs based on their status as being detained at a particular facility.
Five studies randomized participants into different intervention groups: three were able
to identify a control group; two had a comparison group that received another
intervention. All but one collected baseline data at pre-intervention and some
evaluation measure at post-intervention. Half were able to collect some measure of
follow-up between one and six months after release. However, the majority of these are
offered on such a very limited basis that only a small handful of the millions of youth in
the correctional system each year are benefiting from such programs.
Program Implementation: The designs of these programs were widespread;
program duration ranged from a 1.5 hour play to 40 hours; Almost all were offered 2 to
3 times per week, although three were held over full days (1 day, 3 days, and 5 days)
while another extended over a 10-week period. The majority of the programs (78%)
were facilitated by “outside staff”; only three trained correctional facility staff to
conduct the program; Seven included peer (trained incarcerated youth) or peer-like
(hired staff) facilitators; two delegated a large portion of the facilitator’s initial work on
the study was spent on site getting acquainted and building-trust with the program
participants within the correctional setting prior to program implementation.
14
Evaluation: As shown in the Summary Table 2-2 below, the majority of the
data analysis was based on the change in means or mean scores from pre- to post-
intervention. All 10 programs (71%) that collected knowledge measures reported
finding an increase from pre- to post-test. Many of these studies attributed this to a
reduction in misconceptions concerning HIV/AIDS. There was an increase in attitude
toward, intent to use, and/or actual condom use among the eleven studies (79%) that
collected these data. However, while 50% of these studies found participants had a
significant increase in their “attitude toward condoms” only one found a significant
increase in “intention to use condoms regularly” but 75% did report a significant
increase in “condom use” at follow-up.
Table 2-2. Summary of Evaluation Outcomes of the
Fourteen Behavior Change Programs for
Incarcerated Youth
Behavior
# of Studies
Attempted
To Change
the Behavior
How
Measured*
Success
Knowledge 10 studies
7 Mean Scores
3 Means
1 +
9 ++
Perceived self-efficacy 5 studies
3 Mean Scores
2 Means
1 -
4++
Attitude 6 studies
4 Mean Scores
2 Means
1 -
2 +
3 ++
Intention to Use Condoms 4 studies
1 Mean Score
3 Means
2 -
1 +
1 ++
Perceived Risk 5 studies
1 Mean Score
4 Means
1 -
1+
3++
Condom Use 4 studies 4 Means
1 -
3++
# of Sex partners 4 studies 4 Means
1 -
1 +
2 ++
* Difference from pre to post + (not significant)
- (no change) ++ (significant)
15
Common Components of the Successful Programs: All programs sought to
increase knowledge, improve attitudes, and change behaviors. The most common
constructs (See Appendix A and B) found among 43% of these interventions were:
stages of change, perceived self-efficacy and perceived social norms.
C. Preliminary Data
In 1998, the results of a needs assessment collected by staff of the Los Angeles
County Juvenile Justice Health Department in conjunction with JWCH Institute, Inc.
(Table 2-3) portrayed the lifestyle of youth detained by the Los Angeles County
Juvenile Justice system.
Over 98% reported being sexually active.
90% were sexually active by 13 years of age
Less than 5% used condoms with every sexual encounter
51% of the females & 12% of the males have had an STI
2 males have had sex with an HIV+ partner
32% of the females & 2% of the males were sexually assaulted
More than 80% reported using drugs or alcohol on a regular basis.
55% of females & 58% of males are drug users
5% of females & 7% of males trade sex for drugs/money
11% of females & 7% of males use injection drugs
* These data were collected from a convenient sample of ~1500 youth detained by Los
Angeles County Probation Department prior to the intervention.
These findings were cause for further study to gain a keener insight to the factors
influencing these youth’s risk-taking behaviors (Unpublished Report, 1999). Four focus
groups were conducted with a convenience sample of Los Angeles County incarcerated
youth. They were asked, "what do you worry about?” All except one said they worried
about getting shot. Another student added, "At least HIV is not a bullet ripping through
your body." The majority of this population reported being gang-affiliated or gang-
Table 2-3. Self-Reported Risky Behaviors Among
16
affected with 43% worrying about dying while still young: when asked what age
"young meant" they answered "under 21 years.” Although this population's distress
about the immediate future was great and they actually were at high risk of being killed
by gang warfare, an even greater percentage (48%) reported "worrying about getting
AIDS.” However, only 29% said that they had actually had an HIV antibody test. In
fact, this number may be higher than the actual one, because a large proportion of
incarcerated youth said that they thought they had been tested, without their consent,
when blood was drawn during their medical physical exam completed during their
initial intake at Juvenile Hall.
Though this population said they worried about contracting HIV and were able to
understand that their behaviors, especially that of having unsafe sex with multiple
partners, put them at high risk, they showed a lack of the skills needed to change their
behaviors. Eighty-nine percent (89%) reported they had had vaginal, anal, or oral sex
with a minimum of two sex partners in the last three months. Yet, only 34% reported
that they "always use a condom.” Repeatedly, the reasons given for inconsistent
condom use was the lack of skills to negotiate with their partners and friends about safer
sex practices: fifty-one percent (51%) reported not feeling comfortable asking their
partner(s) if they could use a condom and 54% reported not feeling comfortable talking
to their friends about HIV and safer sex.
Although intravenous (IV) drug use is not typical among this population (98%
reported that they did not use drugs intravenously), non-injection drugs and alcohol use
is prevalent. A large proportion of the youth said they have had sex while under the
17
influence of drugs and/or alcohol; while 62% reported using alcohol or drugs at least 3
times per week.
While these youth average 65% on HIV prevention knowledge measures, they
are often unable to make the connection between one behavior which they know is high
risk, such as sharing needles for IV drug use (a behavior they themselves do not
practices) with sharing needles for tattoos (a behavior practiced by any). Gang-
affiliated youth give each other tattoos using home-made machines: 39% reported
"using dirty needles for tattooing purposes.”
D. The Tailored-HIV/AIDS Prevention Intervention
The Los Angeles County preliminary data described above agreed with the other
studies portraying incarcerated youth across the nation; basic HIV/AIDS information
being provided to the youth detained in the Los Angeles County Juvenile Justice
facilities was not enough to cause these youth to take preventive measures to protect
themselves from being infected with HIV. The local community-based organization
providing the educational program
1
in Los Angeles decided that they needed to develop
a theoretical intervention model.
1. Stages of Change (SOC)
Many intervention programs are implicitly designed for individuals who are
ready to-take action; that is to say, they are prepared to change their lifestyle behaviors
(Bandura, 1977). Prochaska and DiClemente's "Stages of Change Theory"
(DiClemente, Prochaska, et al., 1991; Prochaska & DiClemente, 1984; Prochaska, et al.,
1
JWCH Institute, Inc of Los Angeles California in conjunction with Executive staff of
the Los Angeles County Juvenile Court Health Services.
18
1992; Prochaska, 1994; Prochaska & Norcross, 2001) suggests (Table 2-4.) that people
do not change their behaviors at once but move successively through five stages: pre-
contemplation (not intending to change), contemplation (intending to change within six
months), preparation (actively planning change), action (making changes) and
maintenance (sustaining change, avoiding relapse).
Most people learn from their relapse experiences and recycle again through the
Table 2-4. STAGES OF CHANGE (SOC)
PrecontemplationContemplation Preparation Action Maintenance
No intention to
change
Intention to
change within 6
months
Actively
planning
change
Making
changes
Sustaining
change,
avoiding
relapse
5 Stages
No intention to
change behavior
in the
foreseeable
future. Many
individuals in this
stage are
unaware or
under-aware of
their problems.
Aware that a
problem exists
and are
seriously
thinking about
overcoming it
but have not
yet made a
commitment to
take action.
Combines
intention and
behavioral
criteria.
Individuals in
this stage are
intending to
take action in
the next
month and
have
unsuccessfully
taken action in
the past year
Individuals
modify their
behavior,
experiences,
or
environment
in order to
overcome
their
problems.
Action
involves the
most overt
behavioral
changes and
requires
considerable
commitment
of time and
energy.
People work
to prevent
relapse and
consolidate
the gains
attained
during action.
For addictive
behaviors
this stage
extends from
six months to
an
indeterminate
period past
the initial
action.
PROCESSES
OF
CHANGE
Consciousness
raising, dramatic
relief, self re-
evaluation,
environmental re-
evaluation, and
decisional
balance
Self re-
evaluation,
environmental
re-evaluation
decisional
balance, self-
liberation, self-
efficacy
Self-liberation,
self efficacy,
stimulus
control,
counter-
conditioning,
helping
relationships
Stimulus
control,
counter-
conditioning,
helping
relationships
and
reinforcement
management
Stimulus
control,
counter
conditioning,
helping
relationships,
reinforcement
management,
social
liberation.
19
stages. The most effective behavior change strategies should be tailored differently for
people in different stages; the goal should be to move people along the continuum into
the next stage. Therefore a successful intervention should help incarcerated youth who
are pre-contemplators to process information more clearly (consciousness-raising), to
increase their emotional awareness of the problem (dramatic relief), and to increase
their understanding of how their self-image is affected by risk reduction (self-
reevaluation). For those in later stages, techniques need to be used for reinforcing small
steps towards safer sex (reinforcement management), for substituting healthy
alternatives for high-risk alternatives (counter conditioning), and for developing social
support for living low-risk sexual lifestyles (helping relationships).
Previous studies using the SOC have identified several mechanisms that
facilitate change: greater motivation or readiness to change, lower temptation, and
higher confidence to perform the positive target behavior and the pros of adopting the
positive target behavior outweighing the cons (Prochaska, Velicer, DiClemente, &
Fava, 1988; Prochaska, Velicer, 1994). Most importantly, these variables are amenable
to change and interventions can be tailored to address the participant’s specific change
profile. In fact, two studies (Grimley, Williams, & Latasha, 2000; Hemphill & Howell,
2000) suggests that SOC may be as useful a construct among adolescent offenders as it
has been found for treatment completion and treatment outcomes among adults;
matching a person’s stage of readiness for change could be an effective intervention
approach.
20
2. Learning Preferences (Styles)
Realizing that the reading level of the youth detained in Los Angeles County
ranges as low as the 3
rd
grade and only as high as the 5
th
grade, the program staff sought
teaching strategies for improving each participant’s ability to grasp the intervention. A
presentation at a national conference for Prisoner Care (Silver, 1999) reported
differential learning styles among incarcerated adult males compared to the general
male population (Silver & Hanson, 1992) prompted program staff to develop a
curriculum that included materials and activities for all learning preferences. This
would increase the strength of the curriculum in changing a participant’s behavior
regardless of his/her learning preference.
Learning preferences are a composite of characteristic cognitive, affective, and
physiological factors that serve as relatively stable indicators of how a learner
perceives, interacts with, and responds to the learning environment (Keefe, 1979). It
defines how an individual perceives, thinks, remembers, and problem-solves. The
Hanson, Silver & Strong learning preference inventory (LPI) (Silver & Hanson, 1992)
was used to determine each participant’s learning preference. The individual is
determined to prefer one of four learning styles: sensing thinkers, sensing feelers,
intuitive thinkers, and intuitive feeler as defined in Table 2-5 below.
21
Table 2-5. The Learning Preference Inventory
The learning styles are categorized as:
The Businessman: very organized, everything by the book
Sensitive Thinker (ST)
NOT FLEXIBLE
The Psychologist: flexible, good collaborator/friend
Sensitive Feeler (SF)
TOO INVOLVED SO CAN’T PRIORITIZE
The Intellectual: Critical thinker, good debater
Intuitive Thinker (NT)
OVERLY CRITICAL AND FOCUSED
The Poet: a dreamer – goes outside the norm, fearless
Intuitive Feeler (NF)
CAN’T SEEM TO FOCUS
The instrument comprises 36 statements followed by four possible responses that are
assigned a number from 0 to 5 to show how the respondent feels that the answer is like
him/her. The score totaling 5 used the following criteria: 0=never, 1=almost never,
2=sometimes, 3=often, 4 =almost always, and 5 = always. The style with the highest
total score is considered the participant’s dominant style or learning preference.
A sensing feeler, for example, prefers to learn about things affecting peoples'
lives rather than facts or theories. Because personal relationships are of high value to
these type of learners, they relate what they are learning to what they already know, to
someone they know, or to themselves. Thus, an HIV+ speaker with similar life
experiences would have great impact with such a learner. Conversely, the intuitive
thinker prefers to work independently and is interested in consequences and
implications of new material; these learners would benefit from fact-filled
presentations. Intuitive feelers are more artistic and theatrical and would learn from
participating in role-play activities. Sensing thinkers, who are planning and results
oriented, respond well to challenges, and would learn from completing tasks, such as
taking a simple survey of peer dating attitudes.
22
3. Tailoring The Intervention
When describing tailoring a program a distinction should be made between
tailored, targeted, and personalized prevention messages:
Targeted prevention messages are intended for specific subgroups of the general
population usually based on a set of demographic characteristics such as age,
gender, and ethnicity. It assumes homogeneity among the group that may not be
true (Abad, Ramos, & Boyce, 1974). For instance, because public health problems
disproportionately affect specific subgroups then the general population researchers
“target” special programs or services. This is true for this study as well. It was
targeted for youth incarcerated by the Los Angeles County Juvenile Justice system
usually reside in South Central Los Angeles.
Personalized prevention messages use the person’s name and then conveys a
generic message (e.g., “Dr. Thomas Valente, you have been selected”) (Kreuter et
al., 2000).
Tailored prevention messages combine information and behavior change strategies
intended to reach one specific person based on characteristics that are unique to that
person, related to the outcome of interest, and derived from an individual
assessment (Kreuter et al., 2000). They describe the process to that of a tailor,
custom-fitting clothing: taking the measurements, asking about preference of color,
style, and fabric and then using this information to make a perfectly fit suit for the
individual.
Tailored messages are related to the outcome of interest and derived from an
“individual” assessment. In other words, there are no general group statements made.
23
Messages are derived from an assessment that drives the individualized focused
communication. If it were suggested that the program “target” a particular ethnicity,
age, gender, or language it would force the use of “group” rather than individual
characteristics making it less specific than tailoring. Again utilizing the example
above, it would be the difference between buying a size 8 red suit off-the-rack or going
into a tailor, choosing the fabric, and having it made from your specific measurements.
There have been a few studies that have shown targeted prevention interventions
to have some value in changing behavior (Davis, Cummings, Rimer, Sciandra, Stone,
1992; Gritz, & Berman, 1989; Kristeller, Merriam, Ockene, Ockene, Goldberg, 1993;
Morgan et al., 1996; Peppers & Rogers, 1993; Rimer & Glassman, 1994). However, all
of these were smoking-cessation studies. The characteristics being addressed when
targeting a certain group include personal values, cultural norms, and social networks;
these should be embedded within the various strategies of the tailored intervention. In
fact, a recent study conducted by Champion and others (2000) tested the effectiveness
of a “tailored interactive computer intervention with a targeted video and usual care” in
increasing mammography adherence among low income African American women.
Content of both the tailored and targeted materials were developed using a combination
of “stages of change” and the Health Belief model. Results indicated that stage of
mammography adoption did differ by intervention group (p <.04): 32% of the usual
care group were in Action at 3 months, while only 25% of the video group but 40% of
the tailored group, exhibiting the strength of a tailored versus targeted program
(Champion, Ray, Heilman, Springston, 2000).
24
In addition, ethnicity, age, and gender can be tested as possible covariates in the
analysis. For example, preliminary analysis of data collected from my study’s
intervention group (only) and reported in my empirical paper found that when
controlling for any differences that may have occurred due to age, gender or sex, we
still found that the intervention reduced their risky sexual behaviors, reduced their risky
needle behaviors, and moved one step closer to action.
4. The Intervention Model
The model in Figure 2-1 below describes the moderator–mediator relationship of
the intervention that was tested. In this model, the Learning Preference of each youth is
an antecedent and is treated in the analysis along with sex, age, ethnicity, camp,
facilitator, and group. In contrast, the participant’s stage of readiness mediates the
effectiveness of the program. A youth who enters the program in “preparation” is much
more likely to use a condom during his next sexual encounter at post-intervention than
one who is in pre-contemplation; this affects the program. The “change in stage” is the
outcome variable; you can enter the program in one stage and “move” to the next stage
toward action, characterized as a step – and numerically 1 thru 5.
Figure 2.1 – The Intervention Model
MEDIATOR
THE INTERVENTION
1. information
2. Decision-making skills
3. Motivation
4. Normative education
5. Refusal Skills
6. Personal Commitment
Age,
race/ethnicity,
camp, session
study group
Stage
of
Change
(SOC)
OUTCOMES
Change in knowledge
Change in Condom Use
Change in sexual Behaviors
Change in Stage of Change
RISK REDUCTION
25
The purpose of this study is to test the effectiveness of this model in reducing the HIV
risk behaviors of the program participants.
26
CHAPTER THREE: METHODS
This study used an observational, prospective parallel group design with
historical comparisons (See Figure 3-1). The main comparison of this study is between
incarcerated youth in the “comparison group” who received a basic HIV/AIDS
information program compared to youth in the “intervention group” who received an
HIV/AIDS curriculum tailored to learning styles and stages of change. Both programs
were presented as part of the L.A. County Probation detention facility’s regular school
curriculum.
A. The Two Comparison Groups
CDC through the Los Angeles County Office of AIDS Policy and Planning
(LAC-OAPP) funded a non-profit agency (JWCH Institute, Inc.) to develop and
implement both the comparison and intervention programs consecutively.
The Comparison Group (Basic HIV/AIDS Information): The comparison group
received a knowledge-based program implemented over 2 years (1995 – 1997). The
purpose of the program was to provide basic HIV/AIDS prevention education to youth
detained in four of the Los Angeles County Juvenile Justice system detention sites. The
program was provided for all youth attending classrooms in one of the four facilities in
which the teacher agreed to afford classroom time. The JWCH Executive Director (RN,
Phd
abd )
) who had extensive experience in program development targeting youth,
including teen pregnancy and smoking prevention, developed the curriculum. It was
Figure 3-1. Research Design
o
1
x
1
o
3
– Comparison group
o
2
x
2
o
4
– Randomly Selected Intervention
27
reviewed and approved by LAC-OAPP. Two females employed as assistant health
educators by the non-profit agency (JWCH Institute, Inc.) facilitated these sessions. An
effort was made to hire staff that the participants could relate to and feel comfortable
with hearing sensitive information about sexual behaviors:
One young Caucasian woman in her mid-20’s with a Masters of Public Health
(MPH) degree and Certified Health Educator (CHES) with the agency for the
duration of the comparison programs’ contract; and,
One young Latina woman, 25-30 years of age, with a Masters of Public Health
(MPH) degree who was employed by the agency for 2 years of the Comparison
program.
To enhance the program and maintain standardization of the curriculum, the two
health educators were required to complete: 1) a Confidential HIV/AIDS Counseling
and Testing certification training conducted by LAC-OAPP; and, 2) the JWCH
Institute’s program training and achieve at least 90% correct answers at post-test. This
Basic HIV/AIDS Information Program (Comparison) was offered as one session over
two-hours that included:
a. Definition of the terms; Explanation of how HIV is transmitted;
b. A list of safe practices (i.e., condoms, abstinence, and clean needles);
c. Explanation of how the virus works; and,
d. Finding out if you have HIV/AIDS – where to get confidential counseling and
testing.
The Intervention Group (Behavior Change Program): The intervention group
received a behavior change program implemented over 3 years (1997 – 2000).The
28
purpose of this program was to provide an effective intervention to reduce HIV/AIDS
risk behaviors to a randomly selected group of youth detained in six of the Los Angeles
County Juvenile Justice system detention sites. The JWCH Executive Director (RN,
PhD
abd
) who had crafted the comparison curriculum, and the JWCH Division Director
(MSN, MPH) responsible for overseeing program development and implementation
prepared a detailed curriculum (shown in Appendix C) integrating all SOCs (pre-
contemplation, contemplation, preparation, action, maintenance) with each of the four
learning preferences (sensitive feelers, sensitive thinkers, intuitive feelers, intuitive
thinkers). It was reviewed and approved by LAC-OAPP.
This program utilizes the Basic HIV/AIDS Information (comparison described
above) along with a behavior change intervention, tailored to each participant’s stage of
change; and included strategies addressing all learning preferences to ensure that each
participant received the information in the context of their own preferred learning style.
As detailed in Chapter 2, learning style was a strategy for overcoming the literacy
limitations of a population of adolescence whose average reading level was the fifth
grade. These preferences are a composite of characteristic cognitive, affective, and
physiological factors that serve as relatively stable indicators of how a learner
perceives, interacts with, and responds to the learning environment (Keefe, 1979). It
defines how an individual perceives, thinks, remembers, and problem-solves. The
Hanson, Silver & Strong learning preference inventory (LPI) (Silver & Hanson, 1992)
determined each participants’ learning preference as a sensing thinkers, sensing feelers,
intuitive thinkers, or intuitive feeler as defined in Table 2-4. The program’s focus was
29
on setting and meeting individuals’ small, realistic goals toward behavior change in
reducing or eliminating HIV/AIDS risk-taking behaviors.
Two trained and certified health educators facilitated the sessions. Results from
a study presented by R. Morris (1996) at the International AIDS Conference (discussed
in Chapter 2) found no difference in behavior change outcomes of the “Basic
HIV/AIDS” (comparison) by facilitator gender. However, a slight difference between
the facilitators’ race/ethnicity and a significant difference between their ages was
detected; Facilitators who appeared to be closer in age to the program participants were
more successful than older-looking ones. Therefore, caution was taken to ensure that
the intervention staff appeared youthful and their racial/ethnic makeup reflect that of the
predominantly Latino and African American incarcerated population. The two
facilitators implementing the intervention comprised:
The young Latina woman (25-30 years of age) with a Masters of Public Health
(MPH) degree who had implemented 2 years of the comparison group who also
facilitated intervention groups throughout the duration of the 3-year contract; and,
One young African American woman, about 20 years of age, who was attending
school in the evenings to complete her Bachelor of Science (BS) and was employed
by the non-profit agency beyond the duration of the intervention programs’ 3-year
contract under the direct supervision of the Latina facilitator.
To maintain the integrity of the intervention program curriculum across
sessions, facilitators were required to complete:
1. a Confidential HIV/AIDS Counseling and Testing certification training conducted
by LAC-OAPP;
30
2. the JWCH Institute’s curriculum training along with motivational interviewing
instruction and achieve at least 90% correct answers at post-test; and,
3. exhibit curriculum proficiency in adapting all sessions according to the participant’s
learning preference and current stage of change (SOC) to JWCH Institute’s Division
Director.
Special instruction in motivational interviewing strategies (Channon, Huws-
Thomas, Rollnick, & Gregory, 2005; DiClemente & Prochaska, 1985; Nick, 2005;
Rollnick, Kinnersley, & Stott, 1993; Rollnick, 1996) was required to enhance typical
encounters with youth. Staff were trained to integrate “Stages of Change” and
“Perceived Self-efficacy” within the curriculum to enable each to adapt all sessions
according to the participant’s learning preference and current stage of change (SOC).
This type of interviewing is a directive, client-centered counseling style that elicits
behavior change by helping each youth explore and resolve their ambivalence in
understanding the motivation of certain behaviors such as unprotected sex.
Characteristics of motivational interviewing style (Levesque, Prochaska, & Prochaska,
1999; Perz, DiClemente, & Carbonari, 1996) include:
Listening to understand the person's frame of reference;
Recognizing and reiterating their view;
Eliciting and reinforcing the client's own self-motivational statements of the
problem, concern, desire and intention to change, and ability to change;
Monitoring the client's degree of readiness to change, and ensuring that progress
does not generate resistance; and,
Reiterating the client's freedom of choice and self-direction.
31
This technique was used to change the course of participants that in the past may
have been deemed "unmotivated" or "not ready" for change. This non-confrontational,
non-threatening approach allowed program staff to embrace many who might not
otherwise have benefited from the program because of their low motivation to change
or intolerance of intense confrontational interactions.
The Intervention program comprised five sessions - three one-on-one and two
group sessions - within the following guidelines:
INTRODUCTION AND RECRUITMENT - First, the two-hour HIV/AIDS
Information session (Comparison program described above) was conducted. At the
conclusion of the informational session, the facilitator described the activities of another
program (Intervention Behavior Change Program) that could help further their
understanding of the risks of HIV/AIDS and asked each youth in the classroom if they
would like to participate. All interested youth completed a Permit to Participate
(described below in - Appendix D) and the Learning Preference Inventory (LPI) that
was administered (verbally).
SESSION #1 - ONE-ON-ONE COUNSELING: The first personal interaction
between the facilitator and the participant was to determine the youth’s current “stage of
change” (SOC) as determined by his/her last sexual encounter and use of condom
utilizing the staging matrix (Figure 3-2) illustrated below. SOC established the youth’s
readiness to adopt safer sexual practices, as defined by consistent condom use. The
youths’ SOC was determined before, during, after, and three- and/or six-month post-
intervention. It was used to measure the participant’s level of motivation to change
32
6 months or less =
ACTION
6 months or longer =
MAINTENANCE
ALWAYS
Before you were incarcerated, were you
using condoms EVERY time you had sex?
Do you intend to use a condom the next
time you have intercourse?
Less than always =
CONTEMPLATION
NO =
CONTEMPLATION
YES =
Ready-for-Action
YES
How often do you use
condoms when you have sex?
Do you use a condom when you
have vaginal or anal sex?
Aware of risks but
No commitment to action =
CONTEMPLATION
NO
Unaware of risks; denial =
PRE-CONTEMPLATION
Do you have vaginal and/or anal sex?
NO
Staging is not applicable
YES
Figure 3-2. Algorithm Used in Intervention Group to Stage Each
Youth’s “Stage of Change” in Sessions 1, 3, & 6
33
condom use behavior at baseline and as an assessment throughout the intervention and
at follow-up.
Using the various educational materials produced for the program (i.e. visuals,
worksheets and role-playing), the facilitator employed the appropriate protocol
according to the youth’s SOC determined by the algorithm (above) to explain the results
and utility of the LPI completed at the end of the Introduction/Recruitment session. The
two discussed the youth’s learning preference specifically in the context of his
interactions with other people, particularly sexual partners, and how it influences their
own HIV/AIDS risk-taking behaviors.
SESSION #2 - ONE-ON-ONE COUNSELING: The facilitator and participant
reviewed the results of the LPI discussed during session #1, what it means, and how to
integrate it into his daily life. Each youth then solidified their personal strategies for
HIV/AIDS prevention by describing their intended actions in various social settings.
For example, a youth might have pledged to discuss using a condom with his girlfriend
the next time they were together. The facilitator introduced various approaches the
youth could take that were comfortable for him considering his learning preference and
SOC.
SESSION #3- ONE-ON-ONE COUNSELING: A re-cap of session #2 occurred
along with an assessment of barriers the youth perceived still existed impeding his/her
risk reduction behavior. The SOC assessment was re-administered at the end of this
session.
SESSION #4: SMALL GROUPS of six to ten individuals who had completed
the first three sessions convened to role-play and complete skill-building activities in
34
the framework of the five stages of change. The emphasis was on understanding
peoples' learning preferences and how this influences their social interaction with
others.
SESSION #5: THE SAME GROUPS from Session #4 (described above) were
re-convened. Role-playing, based in part on individual variation in LP, behaviors, and
reactions were used to help participants plan and execute appropriate responses to risky
HIV/AIDS exposure situations and perceived social norms. A young HIV+ adult spoke
to the group describing his feelings of being invincible and his subsequent discovery of
being infected with HIV. At the conclusion, access to confidential HIV/AIDS
counseling and testing was offered. In addition, participants completed contact cards
for follow-up purposes and commitment cards establishing a pledge to reduce a
particular HIV/AIDS risk behavior. A toll-free telephone number was also provided to
contact staff and obtain condoms.
Continuing Quality Improvement: Both the Intervention and the Comparison
programs had a “Continuing Quality Improvement (CQI) Plan.” These were developed
by the JWCH Division Director, and approved by the agency’s CQI Committee
comprised of clinicians, health educators and community stakeholders and then by
LAC-OAPP. Standardized measurement tools were used by the Division Director on
random quarterly visits to program sessions, 100% data entry verification, and re-
contacting of participants to verify follow-up surveys were also included. These data
were not available by the end of this study due to the lapse of time between data
collection and this analysis. However, communication with the JWCH Institute’s
35
Division Director reported that her evaluations were always at least satisfactory or
above as were those by LAC-OAPP.
B. Sampling and Assessment
Youth, 10 - 20 years of age, 83% males, residing in Los Angles County,
convicted of a felony charge such as violent offenses (30%), property offenses (45%),
and drug offenses (10%) (Los Angeles County, 2004) and remanded to either one of
the three County Juvenile Halls or to one of the twenty-two (22) camps from 1995 to
2000 were eligible for study. Both the intervention and the comparison programs were
provided as part of the daily curriculum in the Community Education Center (CEC)
located within the detention facilities where youth must attend school each day as
indicated in Figure 3-3 below.
L.A. County Probation
Figure 3-3. Distribution of Youth in The Juvenile Justice
System of Los Angeles County
Incarcerated in 1 of 20 Camps
~2,200 youth
Average stay 5-6 months
State
Incarcerated at
Youth Correction
Facility
Released to
parent/guardian
into the community
~39 ,000 youth each year
11 to 19 years of age
15% Females 85% Males
Detained at 1 of 3 Juvenile Court Halls until Court Date
THE STUDY POPULATION
State
36
There were no restrictions as to participant’s age, race, or ethnic origin.
However, residence within the Los Angeles community (i.e., zip code), gang affiliation,
and reason for incarceration (i.e., severity of offense, drug related, or recidivism) did
influence the Judges’ decision on camp placement thereby affecting whether a youth
incarcerated during this time period would be placed in a facility selected for study.
In addition, camps with a high proportion of youth from South Central Los
Angeles were specifically targeted for the Intervention because the probation officers in
the area agreed to assist in locating youth for follow-up at post-release.
The Comparison Group (Historical): Participants in this group comprise youth
remanded to Camps Scudder, Afler/Paige, Challenger, and Kilpatrick in the Los
Angeles Juvenile Justice system from April 1995 through January 1997. All classrooms
who had not participated in the HIV/AIDS Information Program during the prior 12
months were eligible for study. Still, participation was determined by the teacher as to
whether she could afford the time in her classroom.
Intervention Group: Participants in this group comprise a quasi-experimental
group; Youth remanded to Camps Glen Rocky, Afler/Paige, Challenger, or Los
Padrinos and Central Juvenile Halls in the Los Angeles Juvenile Justice system from
October 1997 through February 2000 were eligible for the Behavior Change program.
Participant selection occurred as shown in Figure 3-4 below; all youth who had been
placed in the camp on one of the two randomly-selected days of the month were
identified and their assigned CEC classroom was solicited for recruitment into study.
Although all youth entering the camps during the designated periods were
eligible for participation, if his/her assigned classroom had participated in one of the
37
Randomly Selected
2-days of
Each Month
Randomly selected
one youth from all the
youth that entered the
camp on these 2-days
Study Staff
All Youth Assigned to
The Selected CEC Classroom
Complete the
HIV/AIDS Information Program
Must Complete a Permit to
Youth are asked to participate
PARTICIPANTS
COMPLETE
BEHAVIOR CHANGE
PROGRAM
POST- TEST
1. Knowledge
2. Stage of Change
3. Risk Behaviors
E Ev ve er ry yd da ay y
Y Yo ou ut th h a ar re e
A As ss si ig gn ne ed d t to o
C Ca am mp ps s b by y J Ju ud dg ge e
Juvenile Justice Facilities
LEARNING PREFERENCE INVENTORY
PRE- TEST
1. Knowledge
2. Stage of Change
3. Risk Behaviors
If this classroom has
NOT participated in
the HIV/AIDS
Information program
in the last 12 months
the teacher was
asked to participate
in the study.
Figure 3-4. Selection of Intervention
Participants
Y Yo ou ut th h i is s A As ss si ig gn ne ed d
t to o a a
C CE EC C C Cl la as ss sr ro oo om m
B By y C Ca am mp p S St ta af ff f
programs during the last year, or the teacher was unable to afford the time, it was
eliminated thereby excluding the new youth.
C. Consent and Confidentiality
Consent to access these youth was obtained from the Los Angeles County
Juvenile Court Judge who presides as the legal guardian over the entire population of
youth detained by the County of Los Angeles Department of Probation.
The Internal Review Board (IRB) at Harbor-UCLA Hospital reviewed the
Behavior Change Program (Intervention) including evaluation instruments and
concluded that this study was excluded from human subject review since it was part of
38
the school curriculum. However, the agency chose to maintain IRB standards that
included individual assent and confidentiality in order to ensure protection of human
subjects. As shown at the bottom of Figure 3-4, at the conclusion of the Basic
HIV/AIDS Information Program, youth were asked if they would like to participate in a
“Behavior Change” program. It was clearly explained to each student: What was
required to participate in the study; what benefits they could obtain by participating in
the program; and, what benefits they would not obtain by participating in the program.
Each youth given the opportunity to participate in the Behavior Change Program
and was told that anyone could refuse or withdraw from participation in any part of the
program, at anytime, without ramification. Additionally, staff portrayed the
confidentiality of all information pertaining to the program. It was clearly explained
“no one in Juvenile Justice System would have access to any data collected by the
program including Camp employees, teachers, and probation officers.” In fact, all data
collected by the programs were kept in locked files in the agency’s Offices located off-
site. Those willing to participate were required to sign a “Permit to Participate” form
(see Appendix 3). Note: Not one youth ever refused to participate or asked to
withdraw during the program.
D. The Total Sample
This was a quasi-experimental design, with a convenience sample for the
comparison group and a randomly selected sample of intervention participants who
completed baseline and follow-up surveys. Data were collected from 733 incarcerated
youth that were 12 to 20 years of age remanded to 9 facilities comprising 90% males
who reported as 50% Latino, 21% African American, 5% Caucasian, 5% Asian/Pacific
39
Islander, 4% Mixed and 9% Other. This was reflective of the general incarcerated
youth population in Los Angeles County at time of study. The limitations of such a
small sample of females caused the researcher to complete further analyses on the
MALES ONLY.
E. Measures
Date-of-birth, racial/ethnic group, gender, facility of program and date of
program session was recorded for both groups.
The Comparison Group: A self-administered knowledge, attitude, and behaviors
survey that included questions used by previous work with this population (Morris et
al., 1995; Morris et al., 1998) was used. One was completed prior to the initiation and
the other at the conclusion of the two-hour HIV/AIDS Information Program. The survey
included questions pertaining to current and past sexual behavior, use of condoms, and
knowledge about the transmission of HIV. Specific questions used in the analysis are
listed in Table 3-1 below.
The Intervention Group: The learning style preference was only collected at
baseline. A self-administered knowledge, attitude, and behaviors survey that included
questions used by previous work with high risk populations (Carbonari & DiClemente,
2000; Galavotti et al., 1995; Heather, Rollnick, & Bell, 1993; Lane et al., 2005;
Rollnick, Heather, Gold, & Hall, 1992) was completed at pre-intervention (baseline)
and post-intervention. The SOC measure (algorithm in Figure 3-3) was administered
three times throughout the program: baseline, post-intervention and at the follow-up
telephone survey completed at 1- and/or 3-months after the last session of the
intervention. The survey included questions pertaining to current and past sexual
40
behavior, use of condoms, and knowledge about the transmission of HIV. Specific
questions used in the analysis are listed in Table 3-1 below.
Learning Preference: The Hanson, Silver, and Strong Learning Preference
Inventory (LPI) (See Appendix 1) was completed by program participants to determine
his/her learning style and guide the facilitator’s strategies during the intervention. The
theoretical foundation for the LPI derives from Analytic Psychology – Carl G. Jung’s
assessment of how the mind is structured and how each person learns. The LPI is a
validated instrument (Rezler & Rezmovic, 1981) for helping youth, 8 to 18 years of age,
identify their own dominant learning style (preference) as either a Sensitive Thinker
(ST) - the businessman, a Sensitive Feeler (SF) – the psychologist, an Intuitive Thinker
(NT) – the intellectual, or an Intuitive Feeler (NF) – the poet. The instrument comprises
36 statements followed by six possible responses that are assigned a number from 0 to 5
to show how the respondent feels that the answer is like them. A score of 5 = always, 4
= almost always, 3 = often, 2 = sometimes, 1 = almost never, and 0 =never was used;
the number of points assigned must total 5. The highest scoring style is considered the
participant’s dominant style or learning preference.
Stages of Change (SOC): Previous studies using this construct identified
several mechanisms that facilitate change: greater motivation or readiness to change,
lower temptation, and higher confidence to perform the positive target behavior and the
pros of adopting the positive target behavior outweighing the cons (Prochaska et al.,
1988; Rollnick et al., 1993; Rollnick, 1996; Velicer, DiClemente, Prochaska, &
Brandenburg, 1985).
41
Evidence from Project MATCH (a multi-site clinical alcoholism Matched study)
indicated that several of the SOC constructs demonstrated strong outcome predictability
(Freyer et al., 2004) and that it is amenable to change through intervention.
Implications of these findings were twofold: first, SOC is predictive of a positive
outcome; and second, interventions can affect change in the variables that comprise
SOC. Most importantly, these variables can be tailored to address the youth’s specific
change profile (Prochaska, DiClemente, Velicer, & Rossi, 1993; Prochaska &
DiClemente, 1983).
SOC was used to measure the participant’s level of motivation to change at
baseline and as an assessment of behavior change throughout the intervention and at
follow-up. The facilitators were trained to use the staging algorithm (Figure 3-2)
throughout the intervention to determine the youth’s current stage of change. This
battery of questions has been used with adolescents, women, and college students.
Findings by Shew, and others suggest that self-report of condom use by adolescents is a
valid indicator of risk behavior (Shew & Remafedi, 1997).
Common Measures: Table 3-1 below, defines the 11 similar questions
common to both the intervention and comparison group assessments that were used to
measure: HIV/AIDS knowledge; HIV/AIDS risk behaviors – sexual partners, needle
sharing, and condom use; and, Stage of Change.
42
YES
Table 3-1. Common Items in Intervention and Comparison Pre & Post Surveys
Intervention Comparison
Common Variables
Used for Analysis
KNOWLEDGE Questions
Question #1
You have sex with a person who
says she or he is a virgin.
If no condom last time you
had sex -- why not?
Risk of sex with
someone who says they
are a “virgin”
Possible
Answers
incorrect NOT RISKY
SLIGHTLY RISKY
correct RISKY
Possible Reason:
MY PARTNER WAS A
VIRGIN
1 = one of the 2 correct
answers OR not as
a reason
0 = incorrect OR was
( ) checked as a
reason
Question #2
You have sex with a person you
know is clean.
If no condom last time
you had sex -- why not?
Risk of sex with
someone who says they
are “clean”
Possible
Answers
incorrect NOT RISKY
SLIGHTLY RISKY
correct RISKY
Possible Reason:
MY PARTNER WAS A
VIRGIN
1 = one of the 2 correct
answers OR not as
a reason
0 = incorrect OR was
( ) checked as a
reason
RISKY BEHAVIORS Questions
Question
#1
Have you ever shared
injection drug needles?
If you shoot drugs do you:
Ever shared drug
needles
Possible
Answers
YES
NO
SHARE NEEDLES 1 = has shared needles
0 = has not shared
needles
Question
#2
As far as you know, do any of
your sex partners shoot up
drugs?
Have you ever had sex
with someone who shoots
up drugs (or used to shoot
up)?
Ever had sex with a
partner who shoots
drugs
Possible
Answers
YES
NO
YES
NO
NOT SURE =
missing
1 = has had sex
w/partner who
shoots drugs
0 = has not had sex
with partner(s) who
shoots drugs
Question
#3
The last time you had sex,
were you using drugs, like pot,
primo, sherm, beer or wine?
How often are you drunk
or high when you have
sex?
Ever had sex when
drunk or high
Possible
Answers
YES
NO
ALWAYS
SOMETIMES
NEVER,
I DON’T GET
DRUNK OR HIGH
ON DRUGS
I DON’T HAVE SEX
1 = has had sex when
drunk or high
0 = has not had sex
when drunk or
high
NO
43
Table 3-1. Common Items in Intervention and Comparison Pre & Post Surveys
(cont.)
Intervention Comparison
Common Variables
Used for Analysis
RISKY BEHAVIOR Questions(cont.)
Question
#4
As far as you know, have you
ever had sex with a man who
has sex with other men?
Have you ever had sex
with a male who’s had
sex with another
male?
Ever had sex with a male
who’s had sex with another
male (MSM)
Possible
Answers
YES
NO
YES
NO
NOT SURE =
missing
1 = has had sex with MSM
0 = has never had sex with
MSM
Question
#5
Have you had sex in the last 6
months?
How many sex
partners have you had
in the last 3 months?
Have had sex in last 3
months
Possible
Answers
YES
NO
# OF PARTNERS:
> 1 = YES
0 = NO
1 = has had sex in last 3
months
0 = has not had sex in last
3 months
CONDOM USE Questions
Question
#1
Do you have vaginal and/or
anal sex?
Have you ever had
sex?
Ever had vaginal and/or
anal sex
Possible
Answers
YES
NO
YES, vaginal and/or
anal
NO
1 = has had vaginal and/or
anal sex
0 = has not had vaginal
and/or anal sex
Question
#2
Have you had a sexually
transmitted disease?
Have you ever had
any sexually
transmitted diseases
(clap, drip, pubic lice,
etc)?
Ever had a sexually
transmitted disease (STD)
Possible
Answers
YES
NO
YES
NO
NOT SURE =
Missing
1 = has had a STD
0 = has never had a STD
Question
#3
The last time you had vaginal
sex, was it with a condom?
The last time you had anal
sex, was it with a condom?
Did you use a condom
the last time you had
sex?
Did use a condom last time
had sex
Possible
Answers
YES – for either question
NO – for both questions
YES
NO
1 = used a condom last
time had sex
2 = did not use a condom
last time had sex
1
Both groups have demographic information about the youth including: date-of-birth, facility
completed program, session at facility, racial/ethnic group, and gender.
44
Table 3-2 below illustrates a comparison of the data collection schedules for the
two study groups. These eleven measures allowed analyses comparing the difference in
change of SOC, knowledge, and risk behaviors from baseline to post-test between
Intervention and Comparison groups. The results determined if the intervention
program was more efficacious then the comparison.
Table 3-2. Schedule of Evaluation Data Collection for the Two Study Groups
Program Pre-Test
Interim
Testing
Post-Test
Follow
-up
Change from
Pre to Post
COMPARISON
One 2-hour
HIV/AIDS
Information
Program
Knowledge
Attitude
Risky Behaviors
Knowledge
Attitude
Risky Behaviors
Extract data to
determine:
SOC
Knowledge
Risky Behaviors
Condom Use
INTERVENTION
5-sessions
HIV/AIDS
Behavior Change
Program
LPI
SOC
Risky Behaviors
Knowledge
Condom Use
SOC Knowledge
Risky Behaviors
Condom Use
SOC
SOC SOC
Knowledge
Risky Sexual Behaviors
Condom Use
F. Statistical Power
The sample size of both the intervention and the comparison groups were pre-
determined using data collected previously from this population. Prior to the
intervention, a power analysis was completed (a priori) to determine the needed effect
size for the sample. As shown in Table 3-3, the group sample sizes of 136 and 414
achieved 100% power to detect a mean difference of -11.7 points between the two
groups. The null hypothesis is that the mean difference is zero and the alternative
hypothesis is that the group means are at least -11.7 units apart. The estimated group
45
standard deviations are 16.70 and 16.70 and the significance level (alpha) is 0.05000
using a two-sided two-sample t-test.
Table 3-3. Sample Size Calculations
1
Power N1 N2 Ratio Alpha Beta Mean 1 Mean2 S1 S2
1.00 138 387 2.08 0.05 0.00 8.50 20.20 16.70 16.70
1
Using PASS Power Analysis
This equates to the adequacy of the population sample for this study: of the total
733 incarcerated youth, 483 participated in the intervention and 250 in the comparison
program. However, due to the mobility of the population not all youth completed the
programs; To compensate for this constantly changing population, all available data
were used for each of the eleven items. However, the smallest samples (i.e., youth who
“completed the program”) equated to 414 (86%) of the intervention and 136 (54%) of
the comparison groups.
G. Analyses
A factor analysis based on the (non-parametric) Spearman correlation matrix
and Promax rotation on the eight risk behavior variables was conducted. As shown in
Table 3-4, four factors were found among the baseline behaviors in the intervention
group, accounting for 65% of the variance. However, only three factors were found at
follow-up, accounting for 67% of the variance. The same eight items accounted for less
than 58% of the variance in the comparison group.
Table 3-4. Total Variance Explained by the Factors
Intervention Comparison
Baseline Follow-up Baseline Follow-up
Factor 1 1.28 17.97 1.77 25.31 1.83 30.46 1.36 22.64
Factor 2 1.18 16.86 1.27 18.12 1.13 18.81 1.12 18.65
Factor 3 1.06 15.19 1.11 15.80 1.04 17.27 1.02 16.98
Factor 4 1.04 14.88
TOTAL 65% 67% 59% 58%
46
As shown in Table 3-4, the first factor was based on two risk behavior
questions: 1) Sex while drunk or high and sex in last 6 months; and, 2) a condom
behavior question “used a condom during last sex.” These might have represented a
“risky sex” behavior dimension. However, only two of the questions were in the first
factor in the Comparison group. Since a factor must comprise at least three questions to
create a score, it is inappropriate to use scores in place of the individual response items
in these analyses
Data Analysis
Descriptive statistics were computed for all demographic variables and survey
items. t-tests and Chi-square tests were used to assess study group differences at
baseline and follow-up. Multi-level modeling (MLM) was used to test for intervention
effects. Because data were nested within sites, multi-level model multiple regression
was performed to determine if the intervention was significantly different in changing
outcomes relative to the comparison group. When data are nested, for example students
nested within schools, MLM is the appropriate method of analyses in order to control
Table 3-5. Rotated Factor Pattern
1 2 3 4
Interv Compar Interv Compar Interv Compar Interv
Variables
At Baseline
Base FU Base FU Base FU Base FU Base FU Base FU Base
Ever sex w/partner shoot
drugs .00 .26 -.02 -.02 .03 .53 .70 .01 -.03 -.14 .08 .97 .91
Ever have sex drunk/high .72 .49 .18 .57 -.04 .58 .54 .27 -.04 -.52 .50 -.11 .35
Sex in last 3-6 Months -.44 -.82 -.14 -.11 -.29 -.07 -.02 .68 .56 -.05 .82 .06 .23
Ever have an STD .09 -.29 .07 .00 .05 .71 .59 -.74 .82 .26 -.44 .05 -.09
Used condom at last sex .67 .85 -.66 -.68 .01 .12 .25 .10 .01 -.15 .02 .11 -.22
Sex safe w/clean partner -.10 .05 .79 .73 .75 .01 .29 -.26 .19 .87 -.01 .23 -.01
Sex safe w/ a virgin .11 .79 .73 .05 -.28 -.16 .04
47
for random effect of school (otherwise, the fundamental assumption of independence in
regression analysis is violated). Separate models for each of the eleven items by site of
program were tested; age, ethnicity, and session at the site were included in all models
as covariates.
48
CHAPTER FOUR: RESULTS
All males who participated in either of the programs were included in the
analyses regardless of their length of stay in the program. Since youth were coming to
and leaving the facilities daily, there are differential amounts of complete data for each
of the eleven items among the two study groups (Table 4-1) available for analyses and
were so noted.
Table 4-1 Response Rates of Study Groups
Intervention Comparison
# % # %
Response Rate – Completers 414 85.7 % 136 54.4%
Pre-Survey Only 15 3.1% 60 24.0%
Post-Survey Only 54 20.5% 54 21.6%
Total Participants 483 250
A. Comparison of Completers To Non-completers and Female Data
Data collected from non-completers (youth who were either moved to another
facility or released and female offenders) were used to determine if there was a
significant difference to age, SOC, knowledge, or risk behaviors at baseline between the
two groups. Of the original 733 cases, there were 70 females. Significant differences
between this group and the non-completers were found among the demographics
between the two groups: There were almost twice as many proportionately in the
comparison group (46%) than in the intervention group (24%) which was significant
with a χ
2
(1, N = 733) = 37.2, p < .001). There were significantly more youth among
the completers reporting their race as Latino (57% vs 34%), African American (25% vs
12%) and Caucasian (7% vs .4%). In contrast there were fewer completers (3%) than
49
non-completers (24%) reporting Other as race ( χ
2
(6, N = 733) = 2.068e
2
, p < .001).
When reviewing the mean age of the non-completers (16 years old) it was found that
they were on average about 4 months significantly younger than the completers (16.3
years old) (t (724) = 3.4, p = .00). There was a distinction to the sites the non-completers
had attended sessions; All participants (100%) from Camp Rocky and Central Juvenile
Hall; 60% of youth from Camp Challenger, and 54% of Camp Afler/Paige this was
significant at χ
2
(7, N = 733) = 2.521e
2
, p < .001).
Upon review of baseline knowledge and risk behaviors, the non-completers
were significantly more knowledgeable about the risk of not using a condom while
having sex with a partner who is clean ( χ
2
(1, N = 579) = 4.35, p = .05). Finally, it may
be of interest to note that there were no non-completers who reported ever sharing
needles at baseline.
B. Comparison of The Study Groups
The data shown in Table 4-2 below determined whether the two samples
(intervention and comparison groups) were comparable in the distribution of age,
racial/ethnic groups, and stage of change at baseline. No significant differences in
demographics were found between the two groups. The Table also demonstrates the
difference in the distribution of learning preferences among the intervention
incarcerated youth population as compared to the general U.S. youth population. There
were significantly more Intuitive Feelers (25%) and less Sensitive Feelers (20%) among
the Los Angeles youth then the general youth population (10% and 35%) (p < .001).
50
Table 4-2. Demographic Characteristics of the Study Groups
Comparison Intervention
ENTIRE SAMPLE
Gender
Males 100% 85%
Female 0 15%
MALES ONLY
Age
Mean Age 16.1 years 16.3 years
12 -14 years of age 13 ( 5.3%) 25 ( 6.1%)
15 -16 years of age 147 (60.5%) 178 (43.3%)
17 - 20 years of age 83 (34.2%) 208 (50.6%)
Race
Hispanic 133 (53.6%) 211 (54.1%)
African American 57 (23.0%) 88 (22.6%)
Caucasian 7 ( 2.8%) 20 ( 5.1%)
Asian/Pacific Islander 12 ( 4.8%) 27 ( 6.9%)
Mixed/Other 39 (15.7%) 44 (11.3%)
INTERVENTION GROUP ONLY
Learning Preferences
Incarcerated Youth
Intervention Group
*
General Youth
Population
Intuitive Feelers 25% 10%
Intuitive Thinkers 16% 20%
Sensitive Feelers 20% 35%
Sensitive Thinkers 39% 35%
*
P < .001
Figures 4-1 through 4-8 illustrate the results of chi-square analyses testing the
differences between the two study groups (comparison; intervention). These
demonstrate the significant difference found between the intervention and the
comparison groups for five of the eleven items at baseline.
51
Baseline: The first figure (4-1) above reveals that among the responders of the
condom behavior item: “used a condom last time had sex” there was a significantly
higher proportion of intervention participants (50%) reporting use of condom at
baseline than the comparison group participants (26%) ( χ
2
(1, N=135) = 23.76, p < .001).
Additionally, although both groups reported “having had sex in last 3-6 months before
the program,” there was a significantly higher number of youth in the comparison group
(97% vs 91%) ( χ
2
(1, N=24) = 5.944, p = .015)
The next two figures describe the difference in HIV/AIDS knowledge and risk
behaviors between the participants in the two groups.
Chart 4-1 . Differences Between Responses of the Two Study Groups
BASELINE CONDOM USE BEHAVIOR
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
Comparison 98% 97% 26% 5%
Intervention 96% 91% 50% 9%
Ever have anal and/or
vaginal sex
Had sex in last 3-6 months*
Used a condom last time
had sex*
Ever have an STD
Figure 4-1. Differences in Responses Between the Two Study Groups
BASELINE CONDOM USE
52
.
It seems that overall ( Figure 4-2), the intervention group was more
knowledgeable and less HIV/AIDS risk-takers at baseline then those in the comparison
group. However, analyses show that while there were significantly more youth in the
intervention group who answered correctly to the knowledge question: “not using a
condom because your partner is a virgin” is risky ( χ
2
(1, N=442) = 13.17, p < .001) there
were significantly less correctly answering “not using a condom because your partner is
clean” is risky ( χ
2
(1, N = 354) = 18.96, p < .001) compared to the comparisons.
Chart 4-2. Differences Between Responses of the Two Study Groups
BASELINE KNOWLEDGE
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
Comparison 75% 78%
Intervention 88% 60%
KNOWS to use a condom with a virgin* KNOWS to use a condom with clean partner*
Figure 4-2. Differences in Responses Between the Two Study Groups
BASELINE KNOWLEDGE
53
Among the risk behavior items at baseline (Figure 4-3), there was a significantly
greater number of youth reporting “ever having had sex while drunk or high” (84% vs
66%) ( χ
2
(1, N = 96) = 15.258, p < .001 in the comparisons then the intervention group.
Chart 4-3 . Differences Between Responses of the Two Study Groups
BASELINE HV/AIDS RISK BEHAVIORS
0%
10%
20%
30%
40%
50%
60%
70%
80%
Comparison 3% 6% 84% 2%
Intervention 1% 2% 66% 2%
Ever shared needles
Ever have a sex partner
who shoots drugs
Ever have sex while
drunk/high*
Ever have sex with MSM
Figure 4-3. Differences in Responses Between the Two Study Groups
BASELINE HIV/AIDS RISK BEHAVIORS
54
In regards to SOC, there was a statistically significant difference between the
comparison and the intervention groups at baseline. As shown in Figure 4-4 above,
there were significantly more Pre-contemplators among the comparison group (21% )
than the intervention group (4.5%); significantly fewer contemplators to Preparationors
(19%) in the comparison group then the intervention group (68%); significantly fewer
comparisons in Action (10%) then among the intervention group (22%); and,
Comparison Group
Figure 4-4. Differences in Responses Between the Two Study Groups
BASELINE STAGE OF CHANGE
Figure 4-4. Differences in Responses Between the Two Study Groups
0%
10%
20%
30%
40%
50%
60%
70%
Comparison 21% 19% 10% 50%
Intervention 4% 68% 22% 6%
Pre-contemplation
Contemplation
To
Preparation
Action Maintenance
BASELINE STAGE OF CHANGE
Coded: 1 = Precontemplation; 2 = Contemplation to Preparation; 3 = Action; 4 = Maintenance.
55
significantly more comparisons in Maintenance then among the intervention group
(6.4%) - ( χ
2
(3, N=517) =
1.955e
2
,
,
p < .001).
Change from Baseline: Chi-square results showed the differences in change
from baseline to post intervention between the two study groups. When looking at in
Figure 4-5 you find a reduction in risk behaviors. The decrease in the proportion of
youth who reported “having had sex in the last 3 - 6 months since the program ” was
significantly larger among the intervention group (91% to 59%) when compared to the
comparison group (97% to 90%); pre: χ
2
(1, N=24) = 5.94, p < .015; post: χ
2
(1, N=75) =
42.62, p < .001). The condom behavior item “used a condom last time had sex since
the program” shows a significantly larger proportion of those in the intervention group
Chart 4-5. Differences Between Responses of the Two Study Groups
POST-INTERVENTION CONDOM USE
-5%
5%
15%
25%
35%
45%
55%
65%
75%
85%
95%
Comparison 90% 29% 2%
Intervention 59% 75% 1%
Had sex in last 3-6 months since
program*
Used a condom last time had sex
since program*
Had an STD since program
Figure 4-5. Differences in Responses Between the Two Study Groups
POST-INTERVENTION CONDOM USE
56
reporting a change in use of condom (50% to 75%) at last sex than among the
comparison group (26% to 29%) (pre: χ
2
(1, N=135) = 23.76, p < .001; post: χ
2
(1, N=129)
= 56.41, p < .001).
Although the intervention group was significantly lower in responding correctly
to having sex without a condom with a clean partner question at baseline (Figure 4-6 as
explained above), they were significantly higher than the comparison group at
answering correctly at post-test (clean: χ
2
(1, N=455) = 14.58, p < .001). This was also
true for the knowledge question concerning condom use with a virgin ( χ
2
(1, N=473) =
74.37, p < .001).
Chart 4-6. Differences Between Responses of the Two Study Groups
POST-INTERVENTION KNOWLEDGE
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
Comparison 79% 82%
Intervention 100% 93%
Need to use a condom with a virgin* Need to use a condom with clean partner*
Figure 4-6. Differences in Responses Between the Two Study Groups
POST-INTERVENTION KNOWLEDGE
57
Both groups reported a decrease in HIV/AIDS risk behaviors; there was a
decrease in the number of participants “having sex while drunk or high since” from
baseline to post. As shown in Figure 4-7, the comparison group had a significant risk
reduction (84% to 28%) compared to the intervention group (66% to 39%); (pre: χ
2
(1,
N=95) = 15.27, p < .001; post: χ
2
(1, N=273) = 4.12, p < .042 ).
Chart 4-7. Differences Between Responses of the Two Study Groups
POST-INTERVENTION HIV/AIDS RISK BEHAVIORS
0%
5%
10%
15%
20%
25%
30%
35%
40%
45%
Comparison 2% 1% 28% 2%
Intervention 0% 2% 39% 0%
Shared needles since
program
Had a sex partner who
shoots drugs since
program
Had sex while drunk/high
since program*
Had sex with MSM since
program
Figure 4-7. Differences in Responses Between the Two Study Groups
POST-INTERVENTION HIV/AIDS RISK BEHAVIORS
58
Figure 4-8 provides a picture of the change in SOC among the two groups. The
comparison group reported a significantly larger proportion of youth in “readiness”
(Action - 10% and Maintenance- 50%) compared to the intervention group at baseline
(22% and 6% respectively). However, the intervention group had a significant increase
in “readiness” at post-intervention (Action - 49%, Maintenance -21%) while the
comparison group had a decrease in the number reporting “readiness to use condoms
every time they have sex” (Action - 17%, Maintenance -31%) (pre: χ
2
(3, N=517) =
1.955e
2
, p < .001 post: χ
2
(3, N=502) = 1.222e
2
, p < .001).
Chart 4-8. Differences in Responses Between the Two Groups
POST-INTERVENTION STAGE OF CHANGE
0%
10%
20%
30%
40%
50%
60%
Post Pre-contemplation* 31% 21%
Post Contemplation to Preparation* 26% 0%
Post Action* 26% 30%
Post Maintenance* 17% 49%
Comparison Group Intervention Group
Figure 4-8. Differences in Responses Between the Two Study Groups
POST-INTERVENTION STAGE OF CHANGE
59
T-tests: The results of the t-tests analyses in Table 4-3 below showed that there
was a greater change in knowledge, risk behaviors, condom use, and SOC among the
intervention group compared to the comparison group. T-tests indicated that the mean
change from baseline to post (post minus pre) within each of the two study groups
found major differences between the two groups. In the intervention group there was a
statistically significant change in two of the three Condom Use items - a decrease in
“having an STD” (t(118) = 2.98, p = .003) and an increase in “Used condom last time
had sex” (t(89) = 3.72, p< .001); there was a statistically significant change in two of
four Risk Behaviors
2
- a decrease in “having sex while drunk or high” (t(135) = 6.78, p
< .001) and “having sex in the last 3 - 6 months” (t(138) = 8.08, p < .001).
Table 4-3. Difference in Mean Changes Within Each Study Group
From Pre to Post
Comparison Intervention
Mean
Change
t DF
p*
Value
Mean
Change
t DF
p*
Value
Mean CONDOM USE (post-pre)
Ever had sex: Had sex .005 .58 192
.003 -1.00 322
Ever had an STD: Had an
STD .040 1.42 111 -.08 2.98 118 .003
Used condom last time sex .008 -.23 131 .24 3.72 89 < .001
Mean HIV/AIDS RISK BEHAVIORS (post-pre)
Shared needles -.03 2.02 134 .045 -.02 1.00 55
Sex partner who shoots
drugs .03 1.75 97 0 .00 112 < .001
Had sex while drunk/high -.57 12.03 128 < .001 -.32 6.78 135
Sex with MSM 0 1.00 109 .033 -- -- --
Sex in last 3-6 months -.05 2.15 129 -.36
8.08
138 < .001
Mean KNOWLEDGE (post-pre)
No condom if partner a virgin .01 -.22 134 .13 -6.94 323 < .001
No condom if partner clean .03 -.85 134 .34 -12.46 323 < .001
2
Had sex with a MSM” was too small a sample to analyze “
60
Table 4-3. Difference in Mean Changes Within Each Study Group
From Pre to Post (cont.)
Comparison Intervention
Mean
Change
t DF
p*
Value
Mean
Change
t DF
p*
Value
Mean STAGE OF CHANGE (post-pre) cont.
Stage Of Change* .35 3.05 132 .003 .61 -21.96 312 < .001
Pre-contemplation -.06 1.30 132 -.03 3.80 412 < .001
Contemplation to
Preparation .06
-1.52 130
-.29
11.19 412
< .001
Action -.06 .78 132 .21 -6.90 412 < .001
Maintenance -.16 3.54 132 .11 -7.10 412 < .001
*Ranges from 1=pre-contemplation; 2=contemplation to preparation; 3=action; 4=maintenance
# group too small to calculate.
There was also an overall increase in knowledge (a decrease in wrong answers)
that was statistically significant (no condom with virgin: t(323) = -6.94, p < .001) and
no condom with clean partner: t(323) = -12.457, p < .001) among the Intervention
group as well. Additionally, there was a shift in the proportion of participants SOC
(t(312) = -21.96, p< .001): to no Pre-contemplators at follow-up (t(412) = -3.80, p <
.001) with a larger number of youth in Action (t(412) = -6.90, p < .001) that were all
statistically significant (contemplation to preparation t(412) = 11.20, p < .001;
maintenance (t(412) = -7.10, p < .001).
When observing the mean changes from baseline to post within the Comparison
group there were fewer significant changes. Only three of the five risk behaviors were
found to be statistically significant from pre to post-intervention: a decrease in sharing
needles (t(134) = 2.02, p = .045), having sex while drunk or high (t(128) = 12.03, p <
.001) and having sex in the last 3 - 6 months (t(129) = 2.151, p = .033). A decrease in
SOC was also found to be significant (t(132) = 3.05, p = .003), particularly among those
reporting Maintenance (t(132) = 3.54, p < .001).
61
Independent samples t-tests were performed to compare the mean change of
each of the eleven items difference between the two study groups.
Table 4-4 below shows that there was a greater mean increase in reported
condom use last time had sex among the comparisons than the intervention participants
(t(220) = 3.73, p < .001).
The mean change for youth in the intervention group was significantly smaller
compared to the comparison group who reported “having sex while drunk or high”
(t(263) = 3.71, p < .001). However, in contrast, there was a greater mean change in
youth in the intervention group who reported “having sex in the last 3 - 6 months” when
in contrast to the comparison group (t(267) = 6.21, p < .001).
Table 4-4. Differences Between Study Groups Changed Means from Pre to Post
Changed Mean Outcome Items
(mean-post minus mean-pre)
Means (SD)
Mean**
Difference
t DF
P
Value
CONDOM USE
Comparison -.01 (.12)
Had sex
Intervention .01 (.06)
-.008 -1.03 514
Comparison -.04 (.27)
Had an STD
Intervention -.08 (.31)
.048 1.27 229
Comparison -.01 (.38) Used condom last time
Sex
Intervention .24 (.62)
.238 3.50 221 .001
HIV/AIDS RISK BEHAVIORS
Comparison -.03 (.17)
Shared needles
Intervention -.02 (.13)
-.012 -.46 189
Comparison .03 (.17) Sex partner who shoots
Drugs
Intervention .00 (.13)
-.031 -1.45 209
Comparison -.57 (.54)
Had sex while drunk/high
Intervention -.32 (.56)
-.250 -3.71 263 < .001
Comparison .00 (.19)
Sex with MSM
Intervention .00 ---
--- --- 116
Comparison .25 (.24)
Sex in last 3 - 6 months
Intervention .52 (.52) .314 6.21 267 < .001
62
The mean change for youth in the intervention group was significantly smaller
compared to the comparison group who reported “having sex while drunk or high”
(t(263) = 3.71, p < .001). However, in contrast, there was a greater mean change in
youth in the intervention group who reported “having sex in the last 3 - 6 months” when
compared to the comparison group (t(267) = 6.21, p < .001).
The mean change for youth in the intervention group was significantly lower
compared to the comparison group who reported “not having sex while drunk or high”
(t(263) = 3.71, p < .001). However, in contrast, there was a greater mean change in
youth in the intervention group who reported “having sex in the last 3 - 6 months”
compared to the comparison group (t(267) = 6.21, p < .001).
The mean change for youth in the intervention group was significantly lower
compared to the comparison group who reported “having sex while drunk or high”
(t(263) = 3.71, p < .001). However, there was a greater mean change in youth in the
intervention group who reported “having sex in the last 3 - 6 months” compared to the
comparison group (t(267) = 6.21, p < .001).
Table 4-4. Differences Between Study Groups Changed Means from Pre to Post
Changed Mean Outcome Items
(mean-post minus mean-pre)
Means (SD)
Mean**
Difference
t DF
P
Value
KNOWLEDGE (cont.)
Comparison .01 (.40) No condom because
partner a virgin
Intervention .13 (.34)
-.122 -3.36 457 < .001
Comparison .03 (.40) No condom because
partner clean
Intervention .34 (.49)
-.307 -6.46 457 < .001
STAGE OF CHANGE (SOC)*
Comparison -.35
(1.31)
Change in SOC
Intervention .61 (.49)
-.959 11.23 444 < .001
* Ranges from 1=pre-contemplation; 2=contemplation to preparation; 3=action; 4=maintenance
** Equal variances assumed
63
When reviewing the knowledge questions it was found that there was a
significantly greater increase in the proportion of youth in the intervention group
reporting the correct answer than among the comparison group participants: no condom
because partner is a virgin (t(457) = -3.36, p < .001) and no condom because partner is
clean ( (t(457) = -6.64, p < .001).
The most prominent difference is shown below in Chart 4-9 between the mean
changes of the two study groups and SOC: while the comparison group moved back
down the stages, the intervention group generally moved up (t(444) = 11.23, p < .001).
Figure 4-5. Test of Differences in Mean Change of
Stage Of Change (Post minus Pre)*
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
Mean Change in SOC (Post minus Pre)* 0.35 0.61
Comparison Intervention
Comparison Group Intervention Group Mean Change of SOC
(Post SOC – Pre SOC) 0.35 0.61
*Ranges from 1=pre-contemplation; 2=contemplation-preparation; 3=action; 4=maintenance
Figure 4-9. Differences in Mean Change of SOC*
Between the Two Study Groups
64
Multilevel Multivariate Regression: To test SOC as a potential mediator in
reducing risk behaviors and increasing knowledge and condom use, multilevel
multivariate regression models were employed. Baseline stage of change can be used to
predict distal follow-up condom use outcome in longitudinal data. Or, one can model
the dynamic change of stage of change pattern from baseline to follow-up. Either model
requires the inclusion of the study group membership (intervention/comparison) as well
as the interaction term of this study variable with baseline stage of change. By testing
the significance of this interaction term, one can address the question whether the
intervention group may mediate the association.
Table 4-5 reports the results of the regressions of the mean changes of each of
the eleven items (from baseline to follow-up) and the change associated to SOC while
controlling for covariates and random effect of site. Among the condom behavior items,
change in “had vaginal or anal sex” was only associated with Pre-Contemplation at
baseline (Std. β = 0.0213; p = 0.0564), whereas “using a condom last time had sex” was
associated with all SOCs at baseline: Pre-Contemplation at baseline (Std. β = 0.4470; p
= 0.0001), Contemplation to Preparation at baseline (Std. β = 0.2528; p = 0.0103) and
Action (Std. β = 0.2819; p = 0.0059).
There was no association between the change of any of the five risk behaviors
and SOC at baseline. However, the change (an increase) in both knowledge items was
associated with all SOCs at baseline: “No condom because partner was a virgin” was
significantly associated with being staged in Pre-Contemplation at baseline (Std. β = -
0.1389; p = 0.0047), Contemplation to Preparation at baseline (Std. β = -0.1383; p =
0.0003) and Action (Std. β =-0.1459; p = 0.0005); and, “No condom because partner
65
was clean” was significantly associated with baseline: Pre-Contemplation (Std. β = -
0.1362; p = 0.0327), Contemplation to Preparation (Std. β = -0.1261; p = 0.0102) and
Action (Std. β =-0.1847; p = 0.0007).
Table 4-5. Association Between Change in Outcomes and SOC at Baseline
Pre-Contemplation
Contemplation to
Preparation
Action MEAN CHANGE of
OUTCOME ITEMS
(post minus baseline)
Std.
β
t
p
Value
Std.
β
t
p
Value
Std.
β
t
p
Value
Ever sex; sex post- 0.02 1.91 0.06 0.01 1.43 0.15 0.01 1.49
0.15
Ever STD;STD post -0.02 -0.55 0.58 0.01 0.38 0.70 -0.01 -0.23
0.81
Condom last sex 0.45 3.90 0.00 0.25 2.59 0.01 0.28 2.78
0.01
Share needles 0.03 1.89 0.06 -0.01 -0.20 0.84 0.01 0.35 0.73
Partner shoots drugs -0.03 -0.77 0.44 -0.05 -0.17 0.86 -0.01 -0.30 0.77
Sex when drunk/high -0.12 -1.00 0.32 -0.05 -0.05 0.96 0.03 0.31 0.76
Sex w/MSM -0.07 -1.49 0.14 0.05 0.12 0.91 -0.02 -0.52 0.61
Sex last 3-6 months -0.02 -0.24 0.81 0.02 0.18 0.86 0.14 1.43 0.16
No condom; partner
virgin -0.14 -2.84 0.01 -0.14 -3.68 0.00 -0.15 -3.49 0.00
No condom; partner
clean -0.14 -2.14 0.03 -0.13 -2.58 0.01 -0.19 -3.41 0.00
Note: 1. All parameter estimates (standardized betas) are adjusted for age, ethnicity, study group, SOC at baseline,
Change in SOC, and random effect of facility.
2. SOC MAINTENANCE was used as the reference group.
Logistic Regression: To identify predictors of change in each of the three
outcome categories: condom behaviors, risk behaviors, and knowledge, logistic
regression analysis (SPSS 16.0) was performed. As detailed in Table 4-6, reporting
having had an STD sometime in the past was the only item that predicted a youth
reporting having had an STD since the program at post-test; if a person had reported
having an STD at pre-intervention, they were .12 times more likely not to report having
an STD at post-intervention (95% CI = .02 to .79; p = .03).
66
Table 4-6. Predictors of Decrease In Reporting An STD
Since Intervention At Post-Intervention
Predictors n
%
coded
1
Coefficient
B
Std.
Error
Wald
χ
2
p
Value
Odds
Ratio
95% CI
Ever have an STD?
1
YES, had STD-baseline 341 7% -2.09 .95 4.87 .03 .12 .02 - .79
Constant -- -- -2.14 .75 8.20 .004 .12 --
1
Yes, had an STD was coded 1.
NOTE: Variables removed from the equation during Stepwise procedure - age, site, study group,
classroom, Race, SOC at baseline, SOC at post-intervention.
Two items were found to be significant predictors of a youth reporting using a
condom last time he had sex at post-intervention (Table 4-7). If the youth was in the
intervention group he was almost 8 times more likely to report using a condom last time
he had sex since the program (95% CI = 3.87 to 15.3; p <.001). If a youth reported
NOT using a condom at baseline he was .12 times as likely to report using a condom at
post-intervention regardless of study group (95% CI = .06 to .24; p < .001).
Table 4-7. Predictors of Increase in Condom Use at Post-Intervention
Logistic regressions were then performed to determine if there were any
predictors of change in risk behaviors at post-intervention. As shown in Table 4-8
below, a youth who completed the intervention was almost 10 (95% CI = 4.5 to 21.9;
Predictors
(At Baseline)
n
%
coded
1
Coefficient
B
Std.
Error
Wald
χ
2
p
Value
Odds
Ratio
95% CI
Did you use a condom last time had sex?
1
Intervention group 363 47% 2.04 .35 33.87 .000 7.68 3.87 - 15.3
NO condom used-baseline 363 63% -2.15 .36 35.36 .000 .12 .06 -.24
Constant -- -- -.24 .27 .74 .389 .79 --
1
No condom last time had sex is coded 1
NOTE: Variables removed from the equation during Stepwise procedure - age, site, classroom, race,
SOC at baseline, SOC at post intervention.
67
p < .001) times more likely to report not having sex in the last 3 - 6 months since
the program than someone who completed the comparison program. It was also
found that a youth who had reported having had sex while drunk or high at baseline,
was .20 (95% CI = .08 to .52; p = .001) as likely to report not having sex while drunk
or high at post-intervention. There were not enough participants to analyze predictors
for sharing needles, and having sex with MSM.
Table 4-8. Predictors of Reduction in Risk Behaviors at Post-Intervention
Predictors
(At Baseline)
n
%
coded
1
Coefficient
B
Std.
Error
Wald
χ
2
p
Value
Odds
Ratio
95% CI
Had sex while drunk or high.
1
Age 314 -.01 .19 .00 .98 .99 .68 - 1.46
Other site 318 6.59 .47
Kirby 318 -1.35 1.70 .63 .43 .26 .009 - 7.21
Los Padrinos JH 318 -39.89 42108 .00 .99 .00
Camp Rocky 318 -21.84 12556 .00 .99 .00
Camp Afler/Paige 318 -1.20 .91 1.77 .18 .30 .05 - 1.77
Camp Scudder 318 .39 1.18 .11 .74 1.49 .15 - 14.90
Camp Challenger 318 -.98 1.25 .62 .43 .37 .03 - 4.32
Camp Kilpatrick 318 -20.68 8286 .00 .99 .00 .
Intervention Group 318 -20.98 12556 .00 .99 .00 .
Classroom 318 9.23 .90
Classroom 1 318 -18.72 12556 .00 .99 .00 .
Classroom 2 318 -20.02 12556 .00 .99 .00 .
Classroom 3 318 -18.99 12556 .00 .99 .00 .
Classroom 4 318 -20.44 12556 .00 .99 .00 .
Classroom 5 318 -19.48 12556 .00 .99 .00 .
Classroom 6 318 -18.14 12556 .00 .99 .00 .
Classroom 7 318 -19.72 12556 .00 .99 .00 .
Classroom 8 318 -19.10 12556 .00 .99 .00 .
Classroom 9 318 2.58 30582 .00 1.00 13.21 .
Classroom 10 318 -20.11 12556 .00 .99 .00 .
Classroom 11 318 -19.27 12556 .00 .99 .00 .
Classroom 12 318 1.43 1.06 1.82 .18 4.19 .52 - 33.62
Classroom 13 318 1.34 1.28 1.10 .29 3.83 .31 - 47.10
Classroom 17 318 .59 .88 .45 .50 1.80 .32 - 10.12
Classroom 18 318 .72 1.17 .38 .54 2.05 .21 - 20.15
YES, had sex
drunk/high 265 30% -1.61 .48 11.16 .001 .20 .08 - .52
Mixed Race 318 7.6% 2.10 .72
Latino 318 62.9% -.48 .71 .46 .49 .62 .15 - 2.48
African American 318 22.9% -.66 .73 .81 .37 .52 .12 - 2.16
Caucasian 318 1.9% -1.39 1.10 1.602 .21 .25 .03 - 2.15
68
Table 4-8. Predictors of Reduction in Risk Behaviors at Post-Intervention
(cont.)
Predictors
(At Baseline)
n
%
coded
1
Coefficient
B
Std.
Error
Wald
χ
2
p
Value
Odds
Ratio
95% CI
Had sex while drunk or high.
1
(cont.)
Asian/PI 318 4.8% -.13 .97 .018 .89 .88 .13 - 5.85
Pre-Contemp - Baseline 264 10.1% 2.37 .50
Cont to Prep - Baseline 264 55.1% .41 .66 .39 .53 1.51 .41 - 5.58
Action - Basline 264 12.4% .86 .59 2.06 .15 2.36 .73 - 7.62
Maint at Baseline 264 22.5% .73 .66 1.23 .27 2.07 .57 - 7.51
Pre-Cont -Post-Interv 316 14.3% 2.74 .43
Cont to Prep -PostInterv 316 21.9% -.88 .69 1.63 .20 .41 .11 - 1.6
Action - Post-Interv 316 35.2% -.94 .61 2.37 .12 .39 .12 - 1.29
Maint - Post-Intv 316 28.6% -.55 .58 .91 .34 .58 .19 - 1.79
Constant
21.02 12556
.00 .99
1.350E
9
Had sex during last 3- 6 months.
2,3
Intervention group 2.29 .40 32.04 .000 9.92 4.5 - 21.9
Constant .40 .18 5.33 .02 1.50
1
Yes, had sex while drunk or high was coded as 1.
2
Yes, had sex in last 3 – 6 months was coded as 1.
3
NOTE: Variables removed from the equation during Stepwise procedure – age, site, classroom,
race, SOC at baseline, SOC at post-intervention.
Upon review of the logistic regressions for the knowledge items reported in
Table 4-9 below, it was found a participant who answered the knowledge question
“having sex without a condom with a partner who is a virgin is risky” correctly at
baseline was slightly less likely (.01 (95% CI = .003 to .08; p < .001)) to answer it
correctly at post-intervention. Additionally, a youth staged in Contemplation to
Preparation at baseline was .16 (95% CI = .03 to .92; p = .04) times less likely to answer
this question correctly at post. However, a Caucasian participant was 61 times more
likely to respond correctly (95% CI = 1.9 to 1977; p = .02).
For the second question: “Is having sex without a condom with a partner who is
clean risky behavior” the analysis found that a youth in the intervention group, as was
69
someone staged in Contemplation to Preparation at baseline, was slightly less likely (.18
(95% CI = .07 to .43; p < .001) and .31 (95% CI = .11 to .84; p = .02) respectively) to
answer the question correctly at post-intervention. Additionally, a participant is .07
times more likely not to give the right answer if he gave the right answer at baseline
(95% CI = .03 to .16; p <.001).
Table 4-9. Predictors of Increase in Knowledge at Post-Intervention
Age 530 83% .74 .40 3.40 .06 2.09 .95 - 4.57
Other Site 442 22.6% 2.50 .96
Kirby 442 10.6% -.49 16617 .00 1.00 .62
Los Padrinos JH 442 0.2% 2.12 19894 .00 1.00 8.30
Camp Rocky 442 16.3% -.67 43294 .00 1.00 .51
Camp Afler/Paige 442 20.8% .16 21364 .00 1.00 1.18
Camp Scudder 442 11.3% -.39 12980 .00 1.00 .68
Camp Challenger 442 9.0% .64 12980 .00 1.00 1.89
Camp Kilpatrick 442 5.0% 1.66 12980 .00 1.00 5.25
Central JH 442 4.1% .75 12980 .00 1.00 2.11
Intervention 339 87.6% -21.19 19958 .00 .99 .00
Classroom 442 83.3% 1.57 1.00
Classroom 1 442 -1.75 24145 .00 1.00 .17
Classroom 2 442 .02 19133 .00 1.00 1.02
Classroom 3 442 -1.14 14956 .00 1.00 .32
Classroom 4 442 .96 21975 .00 1.00 2.60
Classroom 5 442 -.93 24836 .00 1.00 .39
Classroom 6 442 -.58 21222 .00 1.00 .56
Classroom 7 442 1.20 19702 .00 1.00 3.33
Classroom 8 442 -.77 19278 .00 1.00 .46
Classroom 9 442 1.83 19801 .00 1.00 6.23
Classroom 10 442 1.10 22554 .00 1.00 3.01
Classroom 11 442 -1.06 24868 .00 1.00 .35
Classroom 12 442 -.16 14625 .00 1.00 .86
Classroom 13 442 2.00 14800 .00 1.00 7.40
Classroom 14 442 -1.36 17199 .00 1.00 .26
Classroom 15 442 -.44 13705 .00 1.00 .64
Classroom 16 442 -1.62 19875 .00 1.00 .20
Classroom 17 442 -2.67 30801 .00 1.00 .07
Predictors
(At Baseline)
n
% coded
1
Coefficien
t
B
Std.
Error
Wald
χ
2
p
Value
Odds
Ratio
95% CI
Knows no condom with a partner who is a virgin is risky.
70
Table 4-9. Predictors of Increase in Knowledge at Post-Intervention (cont.)
Predictors
(At Baseline)
n
% coded
1
Coefficient
B
Std.
Error
Wald
χ
2
p
Value
Odds
Ratio
95% CI
Classroom 18 442 -.28 18922 .00 1.00 .76
Classroom 19 442 -1.15 19214 .00 1.00 .32
Classroom 20 442 -1.30 23992 .00 1.00 .27
Classroom 21 442 -1.72 20082 .00 1.00 .18
Classroom 22 442 -2.35 27695 .00 1.00 .09
Classroom 23 442 -4.03 23094 .00 1.00 .02
Classroom 24 442 -3.38 22988 .00 1.00 .03
Classroom 25 442 2.04 12540 .00 1.00 7.72
Classroom 26 442 .53 25338 .00 1.00 1.69
Classroom 28 442 1.14 1.71 .44 .51 3.11 .11 - 88.81
Classroom 29 442 -.30 1.88 .02 .87 .74 .02 - 29.79
Knows is risky-
baseline 532 83% -4.19 .85 24.42 .000 .01 .003 - .08
Mixed Race 442 6.6% -- -- 5.49 .24 -- --
Latino 442 56.8% 1.38 .96 2.08 .15 3.98 .61 - 25.92
African American 442 26.9% .77 1.06 .53 .46 2.17 .27 - 17.35
Caucasian 442 5.0% 4.11 1.78 5.34 .02 60.71 1.9 – 1977
Asian/PI 442 4.8% 1.53 1.69 .81 .37 4.60 .17 - 127
Pre-Contemp - Baseline 416 9.9% -1.26 .97 1.68 .19 .28 .04 - 1.91
Cont to Prep - Baseline 416 49.0% -1.85 .90 4.19 .04 .16 .03 - .92
Action - Basline 416 18.8% -2.33 1.29 3.25 .07 .10 .008 - 1.23
Maint at Baseline 416 22.4% 6.09 .11 --
Pre-Cont -Post-Interv 374 9.5% -1.65 .93 3.12 .08 .19 .03 - 1.2
Cont to Prep -PostInterv 374 27.8% -.69 1.01 .47 .49 .50 .07 - 3.63
Action - Post-Interv 374 39% .44 1.23 .13 .72 1.55 .14 - 17.25
Maint - Post-Intv 374 25.7% 4.77 .19 --
Constant -- -- 11.42 23808 .00 1.00 91581 --
Intervention group 533 64% -1.74 .45 14.81 .000 .18 .07 - .43
Knows is Risky 533 66% -2.63 .42 39.81 .000 .07 .03 - .16
Maint - post interv 291 29.2% -- -- 9.73 .02 -- --
Pre-Contem post interv 291 34.7% -.77 .65 1.39 .24 .46 .13 - 1.67
Cont to Prep-
PostInterv 291 26.5% -1.18 .51 5.29 .02 .31 .11 - .84
Action - Post interv 291 9.6% -.02 .58 .00 .97 .98 .31 - 3.05
Constant -- -- 4.74 .61 60.01 .000 114
1
Correct answer was coded 1.
Variables removed from the equation during Stepwise procedure - age, site, race, SOC at baseline.
71
CHAPTER FIVE: DISCUSSION AND CONCLUSION
This study provides a significant contribution to the HIV/AIDS prevention
literature, as it provides evidence of a tailored program for incarcerated youth that can
initiate sustained risk reduction behaviors. It also provides information about the
differential learning preferences among this unique population of incarcerated males 12
to 20 years of age in Los Angeles County. The magnitude and consistency of the
results of this study are worth noting. A participant in the Intervention program was 8
times more likely than a control to report using a condom last time he had sex (p <
.001); and, 10 times more likely to report not having sex in the last 3 - 6 months since
the program (p < .001) when compared to controls. This reflects a substantial increase
in protective behaviors by decreasing sexual exposure, the number one form of
transmitting HIV. Although the intervention group initially had a larger proportion of
youth reporting condom use at baseline, there was a significantly greater increase
reporting condom use last time had sex at post-intervention when compared to controls.
As expected, both groups were knowledgeable about HIV/AIDS at baseline.
However, the intervention group showed a significantly greater increase in both
questions than the comparisons “regardless of their learning style”. This means that the
intervention curriculum met the learning preference of each of the participants
regardless of age or race.
The analysis of the stages of change model as applied to these participants
provides the most comprehensive set of data to test the stage conceptualization of
HIV/AIDS risk reduction among incarcerated youth to date, and represents a unique
contribution to the scientific literature on this topic. The results overwhelmingly support
72
the process of change as described by the SOC patterns and stage-specific predictions of
condom use at post intervention.
The majority of the intervention group were staged at baseline in “pre-
contemplation to preparation”; this equates to over 70% of the intervention group
engaging in unprotected sex regularly. In contrast, 60% of the comparison group was in
the “action to maintenance” stages at baseline. This meant there was a significantly
smaller proportion of comparison participants (yet still a large number of youth)
reporting practicing unprotected sex prior to the intervention. At post-intervention,
however, the groups were reversed. There was a significantly larger increase in the
proportion of intervention participants in the “action” stage than in the comparison
group. The comparison group, in contrast, had a smaller proportion in maintenance and
a larger proportion in contemplation to preparation and action stages.
Some of the characteristics of the two study groups may explain this
transformation. First is the staging algorithm used by the intervention group. This
comprised a battery of four questions asked in the same manner at four different time
points throughout the curriculum; two of which were during one-on-one sessions. In
contrast, the comparison participants were staged using data collected from various
questions scattered throughout the survey. These were asked at baseline and
immediately post-intervention in a group-setting. The four questions used for staging a
number of populations for various behaviors including smoking cessation and weight
reduction have always been asked consecutively. Extracting the answers after they
were asked in a different context may not prompt same answers as collecting them
systematically in the algorithm.
73
Second, the movement in SOC particularly among the comparison participants
from “maintenance” to “pre-contemplation” may be in part, explained by the
relationship established between participants and the facilitator. By the completion of
the program, participants may have felt more comfortable and trustful in providing
truthful answers at post-intervention. Whereas in the intervention group, this trustful
relationship may have been established during the one-on-one sessions; where SOC was
initially obtained and remained consistent throughout the intervention sessions.
The overall multilevel regression analyses substantiated the model described in
Chapter 3; SOC at baseline exemplified mediation by the significant relationship found
with an increase in condom use and knowledge at post-interventions. That is to say, a
participant’s baseline SOC influenced whether he used a condom the next time he had
sex and answering the knowledge questions correctly after completing the program; the
further along the continuum of SOC at baseline (ie., toward action) the more likely
he/she will use protection (a condom) the next time he/she has sex. However, the
success of the intervention implies that the curriculum was effective in changing
program participants’ behaviors regardless of their SOC at baseline. In other words, the
curriculum addressed a participant within the appropriate context to his SOC. This does
not support the Intervention Model in Chapter One Figure 1; SOC is not a mediator.
Therefore, the Revised Model is:
74
.
Moreover, these data provided the opportunity to explore learning preferences in
a sample of incarcerated adolescents. Results showed the study group has a
significantly larger proportion (15% more) of intuitive feelers and smaller proportion
(15% fewer) of Sensitive Feelers then the general United States youth population. This
means there are more incarcerated youth in the sample of Los Angeles County’s
Juvenile Justice System who like to think outside the box, and need their freedom to
explore and use their imagination but tend to be fearless and feel as if they can do
anything. Along with these attitudes they have difficulty staying focused. This finding
in particular has important implications for school-settings both within correctional
facilities and in the community for out-of-mainstream youth at risk of or on probation.
Youth with these learning preferences find it challenging to learn in the typical
classroom setting. Feelings of confusion from their needs not being met and the stress
of being left-behind academically may account for some of these out-of-mainstream
youth’s troubles in school and with the law.
Figure 5.1 – The Revised Intervention Model
THE INTERVENTION
1. information
2. Decision-making skills
3. Motivation
4. Normative education
5. Refusal Skills
6. Personal Commitment
Age,
race/ethnicity,
camp, session,
study group
OUTCOMES
Change in knowledge
Change in Condom Use
Change in sexual Behaviors
Change in Stage of Change
RISK REDUCTION
75
In this study, we found that a tailored intervention led to a reduction of
HIV/AIDS high risk behaviors in an incarcerated adolescent population. Importantly,
this may be the first data to explore the possible mediation of SOC in a population of
youth at significantly increased risk for exposure to HIV. Therefore this study makes a
unique and important contribution to research on HIV prevention among incarcerated
youth. As expected we found a significant positive relationship between SOC and risk
reduction as previously found among college students and pregnancy prevention. It
provides support for being able to teach youth who may not otherwise have been
approachable to change their behaviors thereby reducing their risk of becoming HIV+.
A. Study Limitations
There are a number of considerations that must be addressed when approaching
the results of this study the majority of which entail the limitations of the comparison
group. Although there was a poor completion rate among comparison group
respondents in this study, the rate is not unusual for research conducted among
incarcerated youth. The instability of the general population of incarcerated youth
minimizes the likelihood of a participant completing the program, particularly among
the controls. As shown in Appendix 3, in the rankings of published studies, the response
rate was as low as 47% among one group. This attrition rate is very typical of a
program held in detention facilities because of the uncertainty associated with each
day’s activities. Incarcerated youth have many reasons for not attending class where the
program sessions are held, for instance:
they must be in attendance at all mandated court appearances;
they may be transferred to another camp; or,
76
they may be released into the community.
However, to avoid many of these attrition issues a number of precautions were
taken when the facility for the intervention group was selected:
the minimum stay of youth was 6 months;
the program sessions were combined – longer sessions in fewer days over
fewer number of weeks. For instance instead of two times per week over 3-
4 weeks it was completed 3 times a week over two weeks.
the majority of the youth were residents from South Central Los Angeles
(for follow-up purposes).
It is possible that the HIV/AIDS risk behaviors may have been over or
underreported. All data were self-reported and there were no bio-markers such as urine
tests to screen for Sexually Transmitted Infections (STIs) as an indicator of condom use
collected. We also did not use a Bogus pipeline (when the facilitators implies that there
will be a bio-marker specimen collection) to improve the chances of each youth
answering the surveys truthfully. Still, the use of previously validated instruments were
used which should yield legitimate results.
There is also some concern that the youth may have misconstrued program
participation as a useful commodity that could be used to improve their position either
at the facility or with the Judge. Even though it was clearly explained at program onset
that participation and survey response would have no affect on their treatment, a
participant may have felt compelled to participate or answer in a “socially desirable”
manner (i.e., yes, I will use condoms regularly when I have sex) in hopes that it would
be noticed and in turn affect his/her treatment and/or duration of incarceration.
77
To avoid this effect, two of the eighteen programs reviewed in Chapter Two
actually paid participants for completing pre-, post-intervention, and follow-up surveys.
These same two programs had staff spending a great deal of time developing a trusting
relationship with program participants to maximize the validity of their data (Magura,S,
1994; Gillmore, MR, 1997).
Measurement Validity and Reliability: Another limitation of this study is the
measures used by the two study groups. The comparison group comprises historical
data collected using the most efficient and affordable evaluation design: pre- and
immediate post-test measures. Additionally, although the survey questions for this
group had been used in the past for similar studies they were not validated. Both were
administered by health educators as part of the provision of services. Despite the
limited prior use of the control questions, these items were comparable enough to the
validated intervention questions to provide valuable information on HIV/AIDS risk
behaviors of incarcerated youth in Los Angeles County
Confounding Factors That Could Not Be Measured: As in any study
conducted outside the lab in the community there are a number of confounding factors
that may have occurred that the researcher had no control over including: prior
participation in an HIV/AIDS or STI prevention intervention, historical events (i.e.,
introduction of the cocktail treatment and Magic Johnson announcing testing HIV
positive), and the impact of incarceration itself on the participant, testing and retesting
78
B. Strengths of This Study
The U.S. Department of Education in conjunction with the Institute of Education
Sciences, and the National Center for Education Evaluation and Regional Assistance
(Baron, 2003) produced guidelines for evaluating evidence-based intervention
in schools or classrooms. Their investigations generally supported the value of
comparison-group designs with randomization in which the groups are closely matched
with the intervention group in prior test scores, demographics, time period in which
they are studied, and methods used to collect outcome data. They reported that, such
designs “yielded correct overall conclusions in most cases about whether an
intervention is effective, ineffective, or harmful.”
This study used an observational, prospective parallel group design with
historical comparisons and a randomized intervention sample. The comparison and
intervention groups were comparable in age and race; both were similar to the total
incarcerated youth population. Each of the groups had a sufficient number of
participants to detect the anticipated effect of the intervention as determined by the
sample size calculations a priori. Although group randomization of classrooms was
used rather than individual selection, Linear Multilevel Models were appropriately used
to take the correlation among subjects in the same setting into account. The use of two
standardized measurement tools, stage of change and the Learning Preference
Inventory, strengthens the utility of this study as well. When compared to the 18
HIV/AID intervention studies conducted with incarcerated youth that have been
published and reviewed in Chapter 2 this study ranks in the top 6.
79
CONCLUSION
The juvenile justice system seems to provide an ideal opportunity to convey
healthy practices to a captive audience; a large number of confined youth that could be
readily accessed for HIV/AIDS risk-reduction interventions. Then one must ask “why
are there so few?” But, in fact, they are a population of out-of-mainstream youth with
low literacy rates accompanied by high rates of risky behaviors who are being detained
temporarily. Therefore, the types of interventions and implementation strategies are
critical to the utility of any program targeting these youth. However, there are many
factors influencing the ability to provide a program in this setting. The majority of
barriers are associated with the infrastructure of the Juvenile Judicial system that entails
the capacity of each facility, the philosophy of the Judge presiding over the incarcerated
youth as well as the Administration of each facility.
A. Common Barriers and Limitations to HIV/AIDS Program Implementation
Of utmost importance is the direct conflict between the mission of youth
correctional facilities, that being of safety and conformity, and that of health care
professionals seeking an opportunity for eliciting risk reduction behavior change (R.
DiClemente, 1990; Hein, Cohen, & Litt, 1980). Repeatedly, study staff was told by
L.A. County Probation administration that even though they agreed that the program
would be useful for the detained youth, implementation time was very limited.
On the other hand, in order to adopt safe sex practices (i.e., protected sex),
program participants need access to condoms. However, only two Juvenile Justice
Facilities in the entire Country allow distribution of condoms; Los Angeles County is
not one of them. Still, many of the fourteen programs reviewed in Chapter Two
80
reported they could not even demonstrate condom use. While this study was able to
show proper condom use the participant had to call program staff after release (an 800
number) to request condoms to be mail to their current residence. This functioned to
maintain communication with participants and increased access for follow-up. Thus,
follow-up measures of these types of programs must be weighted by the barriers
participants confront in accessing condoms.
B. Difficulties with the Incarcerated Population
This study provides an example of how the instability of the population helped
determine the accessibility of these youth. The alarmingly low program completion
rates of the programs are a reflection the mobility of the population. Although the
program was conducted as part of the school curriculum, nothing is routine in these
classrooms; this is not a typical school-setting. Youth are constantly called out to go to
court or may suddenly be sent to another camp or released. Even more unpredictable
are the lock-downs and canceling of classes. Therefore, it is imperative to develop an
intervention that factors in length and number of sessions.
While the research staff was well aware of the benefits of peer educators in
preventions interventions the obstacles precluding program implementation for using
peers were too overwhelming. Mainly, as evidenced in this study, youth who have been
incarcerated in the past are not allowed to interact with other incarcerated/probationary
youth. Furthermore, in Los Angeles, gang affiliation and fear of crossing over
territories is yet another barrier to being a peer educator. Thus, the decision to use
young adults who were perceived as peers for this study was very successful in this
setting.
81
C. Policy Recommendations
A common assessment found throughout the various studies conducted as part
of this dissertation with the incarcerated youth population was the absence of
opportunities for positive experiences to empower and support changing behaviors and
in-turn reduce the risk of HIV infection. Initiating the strategies proposed here will
assist the U.S. Juvenile Justice system in ensuring that risk reduction programs are
provided to the high risk youth detained within their facilities. Findings from our study
lead us to conclude the following:
First, all correctional staff should be required to obtain and maintain HIV
certification; within one month of hire, attend training and yearly updates. This will
give each staff the information and tools needed to identify opportunities to initiate
discussions surrounding risk behaviors.
Second, each State’s Juvenile Justice System should establish a Task Force
comprising administrators, health educators, educators, health professionals, line staff,
and community-based providers who will be responsible for overseeing all risk-
reduction activities at juvenile detention facilities throughout the State. The focus of the
Task Force should include:
Identifying the most effective intervention models that encompass all learning
styles and stages of change that are peer-led, age-appropriate, culturally
competent, behavior change strategies;
Identifying standardized evaluation tools and assist in accessing program
participants after release for follow-up; and,
Overseeing program implementation to ensure uniformity and prevent
82
duplication of services.
Finally, to enable youth to sustain behavior change, each trained peer educator
should be required to spend a minimum of 60 hours (the required hours of community
service required to graduate High school ) working with field staff at post-release,
offering HIV information to youth within their community.
However, modifications of any government institution with all of its
complexities and multilevel systems, requires a similar approached to modifying
individual behaviors but on a much larger scale. This study determined that applying an
organizational integrative SOC model could modify juvenile detention facility staff to
integrate HIV prevention as part of their daily routine. This finding corroborates what
was found in other studies conducted by Levesque et al (2001), and Prochaska,
Prochaska, & Levesque (2001) who showed that a stage-matched change management
program, tailored to the organization's readiness to change was successful in minimizing
resistance and maximizing the likelihood of successful change.
83
IMPLICATIONS
More than 18% of all new cases of AIDS reported in the U.S. were among 20-29
years of age. Since the average incubation period from HIV infection is 10 years, the
majority of these individuals were in all probability infected during their adolescence.
And, although the prevalence of HIV/AIDS (~1%) is low among incarcerated youth
compared to the adult jailed population (~4%-6%), a number of studies have depicted a
lifestyle that includes the common practice of many behaviors that put them at high risk
of HIV exposure. This study agreed with past research of this population that described
the youth practicing many HIV/AIDS high risk behaviors including unprotected sex
with multiple partners in conjunction with substance and alcohol abuse and the sharing
needles for tattooing and body piercing. Continuation of these behaviors places them at
an increased risk of becoming infected in the future as they venture beyond their
immediate neighborhoods into other communities.
The limitations of implementing and evaluating behavior change interventions
are evidenced by the modest number of studies that have been published (18) and the
difficulties each program detailed. Upon review of this body of research for quality of
the study, places this study in the top 6.
In spite of its limitations, this study makes an important contribution to behavior
change research among youth detained in the juvenile justice system. The sample size is
adequate to generalize these findings to the incarcerated youth detained at the selected
facilities at time of study. Further work should test the utility of tailoring other
prevention interventions among this group to reduce teen pregnancy, other STIs, and
drug and alcohol use. In addition, data on the youth’s learning preference could be used
84
to modify teaching curriculum in general in Juvenile Justice settings, and Alternative
Schools. This could extend into other areas as well, for instance to youth having
difficulty in mainstream school.
The stages of change provide a substantial challenge for intervention
development. Intensity, duration, and type of intervention should be responsive to the
stage of change of the youth. This project involved proactive strategies that targeted the
intervention at their current SOC thereby maximizing opportunities to assist youth in
changing their risk behaviors. This confirms that risk reduction interventions may be
able to increase success rates by being sensitive to stage and by shifting strategies
depending on stage of change.
In conclusion, although the U.S. Juvenile Justice system comprises a large
number of detained youth with a very low prevalence rate of HIV/AIDS (~ 1%) their
documented widespread practice of risky behaviors along with their lack of health care
services puts them at high risk of future infection. Therefore, it is very important that
every opportunity to introduce risk reduction strategies be taken utilizing the most
effective prevention intervention models.
85
BIBLIOGRAPHY
Abad, V., Ramos, J., & Boyce, E. (1974). A role model for delivery of mental health
services to Spanish-speaking minorities. American Journal of Orthopsychiatry,
44(4), 584-595.
Anderson, J. E., Kann, L., Holtzman, D., Arday, S., Truman, B., & Kolbe, L. (1990).
HIV/AIDS knowledge and sexual behavior among high school students. Family
Planning Perspectives, 22(6), 252-255.
Bandura, A. (1984). Social foundations of thought and action: A social cognitive
approach. Englewood Cliffs, NJ: Prentice-All.
Bandura, A. (1982). Self-efficacy mechanism in human agency. American Psychologist,
37(2), 122-147.
Bandura, A. (1977). Self-efficacy: Toward a unifying theory of behavioral change.
Psychological Review, 84(2), 191-215.
Bandura, A., Barbaranelli, C., Caprara, G. V., & Pastorelli, C. (1996). Multifaceted
impact of self-efficacy beliefs on academic functioning. Child Development, 67(3),
1206-1222.
Baron, J. and the Coalition For Evidence-Based Policy (2003). Identifying And
Implementing Educational Practices Supported By Rigorous Evidence: A User
Friendly Guide. Prepared for the: Institute of Education Sciences; Grover J.
Whitehurst, Director.
Braithwaite, R., Robillard, A., Woodring, T., Stephens, T., & Arriola, K. J. (2001).
Tattooing and body piercing among adolescent detainees: Relationship to alcohol
and other drug use. Journal of Substance Abuse, 9(13(1-2)), 5-16.
Cahill, R., & Madigan, M. J. (1984). The influence of curriculum format on learning
preference and learning style. The American Journal of Occupational Therapy :
Official Publication of the American Occupational Therapy Association, 38(10),
683-686.
Canterbury, R. J., McGarvey, E. L., Sheldon-Keller, A. E., Waite, D., Reams, P., &
Koopman, C. (1995). Prevalence of HIV-related risk behaviors and STDs among
incarcerated adolescents. The Journal of Adolescent Health : Official Publication of
the Society for Adolescent Medicine, 17(3), 173-177.
86
Carbonari, J. P., & DiClemente, C. C. (2000). Using transtheoretical model profiles to
differentiate levels of alcohol abstinence success. Journal of Consulting and
Clinical Psychology, 68(5), 810-817.
Castrucci, B. C., & Martin, S. L. (2002). The association between substance use and
risky sexual behaviors among incarcerated adolescents. Maternal and Child Health
Journal, 6(1), 43-47.
Champion VL, Ray DW, Heilman DK, Springston JK. A tailored intervention for
mammography among African-American women. Journal of Psychosocial
Oncology. 2000; 18(4):1-13.
Channon, S., Huws-Thomas, M. V., Rollnick, S., & Gregory, J. W. (2005). The
potential of motivational interviewing. Diabetic Medicine : A Journal of the British
Diabetic Association, 22(3), 353.
Coates, T. J., & Greenblatt, K. (1989). In Holmes (Ed.), Behavioral change using
community level interventions in sexually transmitted diseases. New York:
McGraw-Hill.
Davis, S., Cummings, K., Rimer, B., & Sciandra, R' & Stone, JC. (1992). The impact of
tailored self-help smoking cessation guides on young mothers. Health Education
Quarterly, 19(4), 495-504.
Deering, M. J. (1993). Designing health promotion approaches to high-risk adolescents
through formative research with youth and parents. Public Health Reports,
108(Supplement 1), 68-75.
Deutsch, M., & Gerard, H. B. (1955). A study of normative and informational social
influences upon individual judgment. Journal of Abnormal Psychology, 51(3), 629-
636.
DiClemente, C. C. (1986). Self-efficacy and the addictive behaviors. Journal of Social
and Clinical Psychology, 4, 302-315.
DiClemente, C. C. (1981). Self-efficacy and smoking cessation maintenance: A
preliminary report. Cognitive Therapy and Research, 9, 181-200.
DiClemente, C. C., & Prochaska, J. O. (1985). Self-efficacy and the stages of self-
change of smoking. Cognitive Therapy and Research, 9, 181-200.
DiClemente, C. C., Prochaska, J. O., Fairhurst, S. K., & Velicer, W. F. (1991). The
process of smoking cessation: An analysis of precontemplation, contemplation, and
preparation stages of change. Journal of Consulting and Clinical Psychology, 59(2),
295-304.
87
DiClemente, R., Lanier, M., & Horan, P. (1991). Comparison of AIDS knowledge,
attitudes, and behaviors among incarcerated adolescents and a public school sample
in San Francisco. American Journal of Public Health, 81(5), 628-630.
Diclemente, R. J. (1991). Predictors of HIV-preventive sexual behavior in a high-risk
adolescent population: The influence of perceived peer norms and sexual
communication on incarcerated adolescents' consistent use of condoms. The Journal
of Adolescent Health: Official Publication of the Society for Adolescent Medicine,
12(5), 385-390.
DiClemente, R. J., Zorn, J., & Temoshok, L. (1986). Adolescents and AIDS: A survey
of knowledge, attitudes, and beliefs about AIDS in San Francisco. American
Journal of Public Health, 76(12), 1443-1445.
DiClemente, R. (1990). Adolescents and AIDS: Current research, prevention strategies,
and public policy. in psychological aspects of AIDS and HIV disease. Hillsdale,
NJ: Lawrence Eribaum.
Donaldson, S. (2001). Mediator and moderator analysis in program development in
handbook of program development for health behavior research & practice.
Thousand Oaks, CA: Sage Publications.
Feinstein, R. A., Lampkin, A., Lorish, C. D., Klerman, L. V., Maisiak, R., & Oh, M. K.
(1998). Medical status of adolescents at time of admission to a juvenile detention
center. The Journal of Adolescent Health : Official Publication of the Society for
Adolescent Medicine, 22(3), 190-196.
Fishbein, M. (1982). Social psychological analysis of smoking behavior. In R. E. Eiser
(Ed.), Social psychology and behavioral medicine (pp. 179-197). New York: Wiley.
Fisher, J. D., & Fisher, W. A. (1992a). A general social psychological model for
changing HIV risk behavior. Psychological Bulletin, 11(4), 55-47.
Fisher, J. D., & Fisher, W. A. (1992b). Changing AIDS-risk behavior. Psychological
Bulletin, 111(3), 455-474.
Flores, R. J. (2003). Juvenile justice bulletin. U.S. Department of Justice, Office of
Justice Programs,: Office of Juvenile Justice and Delinquency Prevention.
Forrest, C. B., & Tambor, E. (2000). The health profile of incarcerated male youth.
Pediatrics, 105(1 (Supplement)), 286-291.
88
Freyer, J., Tonigan, J. S., Keller, S., John, U., Rumpf, H., & Hapke, U. (2004).
Readiness to Change Versus Readiness to Seek Help for Alcohol Problems: The
Development of the Treatment Readiness Tool (TREAT) Alcohol Research
Documentation, Inc.
Galavotti, C., Cabral, R. J., Lansky, A., Grimley, D. M., Riley, G. E., & Prochaska, J.
O. (1995). Validation of measures of condom and other contraceptive use among
women at high risk for HIV infection and unintended pregnancy. Health
Psychology, 14(6), 570-578.
Gillmore, M. R., Morrison, D. M., Lowery, C., & Baker, S. A. (1994). Beliefs about
condoms and their association with intentions to use condoms among youths in
detention. The Journal of Adolescent Health : Official Publication of the Society for
Adolescent Medicine, 15(3), 228-237.
Gochman, D. S., & Saucier, J. F. (1982). Perceived vulnerability in children and
adolescents. Health Education Quarterly, 9(2-3), 142-155.
Godin, G., Michaud, F., Alary, M., Otis, J., Masse, B., & Fortin, C. et al. (2003).
Evaluation of an HIV and STD prevention program for adolescents in juvenile
rehabilitation centers. Health Education & Behavior : The Official Publication of
the Society for Public Health Education, 30(5), 601-614.
Greico, G. (1987). Cutting the risks for STDs. Medical Aspects of Human Sexuality, 21,
70-84.
Grimley, D. M., Williams, C. D., & Latasha, L. M. (2000). Stages of readiness for
changing multiple risk behaviors among incarcerated male adolescents. American
Journal of Health Behavior, 24(5), 361-369.
Gritz, ER & Berman, BA. (1989). Smoking and pregnancy. Journal of the American
Women's Association, 44(2), 57.
Hair, E., & Jager. (2002). Helping teens develop healthy social skills and relationships:
What the research shows about navigating adolescence -
Harwell, T. S., Trino, R., Rudy, B., Yorkman, S., & Gollub, E. L. (1999). Sexual
activity, substance use, and HIV/STD knowledge among detained male adolescents
with multiple versus first admissions. Sexually Transmitted Diseases, 26(5), 265-
271.
Heather, N., Rollnick, S., & Bell, A. (1993). Predictive validity of the readiness to
change questionnaire. Addiction (Abingdon, England), 88(12), 1667-1677.
89
Hein, K., Cohen, M. I., & Litt, I. F. (1980). Juvenile detention: Another boundary issue
for physicians. Pediatrics, 66(2), 239-45.
Hemphill, J. F., & Howell, A. J. (2000). Adolescent offenders and stages of change.
Psychological Assessment, 12(4), 371-381.
Herman-Shipley, N. A., Morris, R., Jackson, D. V., & Valente, T. (2006 - In Review).
A tailored HIV intervention for incarcerated youth - did it work? American Journal
of Public Health,
Jackson, D. V. (2003). HIV/AIDS Prevention Video Project JWCH Institute, Inc.
Kandel, D. B., Kessler, R. C., & Margulies, R. Z. (1978). Antecedents of adolescent
initiation into stages of drug use: A developmental analysis. Journal of Youth and
Adolescence, 7(1), 13-40.
Keefe, J. (1979). Learning style: An overview in national association of secondary
school principals' student learning styles: Diagnosing and prescribing programs.
Reston, VA: National Association of Secondary School Principal's.
Kreuter, M., Farrelll, D., Olevitch, L., & Brennan, L. (2000). Tailoring health messages
- customizing communication with computer technology. Mahwah, New Jersey:
Lawrence Erlbaum Associates.
Kristeller, J., Merriam, P., Ockene, J., & Ockene, IS & Goldberg, RJ. (1993). Smoking
intervention for cardiac patients: In search of more effective strategies. Cardiology,
82(5), 317-324.
Lane, C., Huws-Thomas, M., Hood, K., Rollnick, S., Edwards, K., & Robling, M.
(2005). Measuring adaptations of motivational interviewing: The development and
validation of the behavior change counseling index (BECCI). Patient Education and
Counseling, 56(2), 166-173.
Lanier, M. M., DiClemente, R. J., & Horan, P. F. (1999). HIV knowledge and behaviors
of incarcerated youth: A comparison of high and low risk locales. Journal of
Criminal Justice, 19, 257-262.
Lanier, M. M., Pack, R. P., & DiClemente, R. J. (1999). Changes in incarcerated
adolescents' human immunodeficiency virus knowledge and selected behaviors from
1988 to 1996. The Journal of Adolescent Health : Official Publication of the Society
for Adolescent Medicine, 25(3), 182-186.
Lawrance, L., & McLeroy, K. R. (1986). Self-efficacy and health education. Journal of
School Health, 56(8), 317-321.
90
Lesaffre, E. and Verbeke ,G. (2005). Clinical trials and intervention studies In: B.S.
Everitt and D.C. Howell, Ed. Encyclopedia of statistics in behavioral science Wiley,
Volume 1, 301–305.
Levesque, D. A., Prochaska, J. M., & Prochaska, J. O. (1999). Stages of change and
integrated service delivery. Consulting Psychology Journal: Practice and Research,
51(4), 226-241.
Levesque, D. A., Prochaska, J. M., Prochaska, J. O., Dewart, S. R., Hamby, L. S., &
Weeks, W. B. (2001). Organizational stages and processes of change for continuous
quality improvement in health care. Consulting Psychology Journal: Practice and
Research, 53(3), 139-153.
Levinger, G., & Schneider, D. J. (1969). Test of the "risk is a value" hypothesis. Journal
of Personality and Social Psychology, 11(2), 165-169.
Los Angeles County. (2004). Los Angeles County 2004 children's ScoreCard Los
Angeles County Children's Planning Council.
Lowe, C. S., & Radius, S. M. (1987). Young adults' contraceptive practices: An
investigation of influences. Adolescence, 22(86), 291-304.
Maddux, J. E., Brawley, L., & Boykin, A. (1995). Self-efficacy and healthy behavior:
Prevention, promotion, and detection. In J. E. Maddux (Ed.), Self-efficacy,
adaptation, and adjustment: Theory, research, and application. (173-202). Plenum
Press, New York, NY: US.
Magura, S., Shapiro, J. L., & Kang, S. -. (1994). Condom use among criminally-
involved adolescents. AIDS Care, 6(5), 595-603.
McKnight, A. J. (1986). Intervention in teenage drunk driving. Alcohol, Drugs &
Driving, 2(1), 17-28.
Morgan, G., Noll, E., Orleans, C., Rimer, B., Amfoh, K., & Bonney, G. (1996).
Reaching mid-life and older smokers: Tailored interventions for routine medical
care. Preventive Medicine, 25(3), 346-354.
Morris, R. E. (1999). California HIV Sero-Prevalence Study.
Morris, R. E., Baker, C. J., Valentine, M., & Pennisi, A. J. (1998). Variations in HIV
risk behaviors of incarcerated juveniles during a four-year period: 1989-1992.
Journal of Adolescent Health, 23(1), 39-48.
91
Morris, R. E., Harrison, E. A., Knox, G. W., Tromanhauser, E., Marquis, D. K., &
Watts, L. L. (1995). Health risk behavioral survey from 39 juvenile correctional
facilities in the United States. The Journal of Adolescent Health : Official
Publication of the Society for Adolescent Medicine, 17(6), 334-344.
Morrison, D. M., Baker, S. A., & Gillmore, M. R. (1994). Sexual risk behavior,
knowledge, and condom use among adolescents in juvenile detention. J.Youth
Adolesc., 23(2), 271-288.
Nick, H. (2005). Motivational Interviewing: Is it all our Clients Need? Taylor & Francis
Ltd.
Peppers, D., & Rogers, M. (1993). The One To One Future: Building Relationships
One Customer At A Time. New York: Doubleday.
Perz, C. A., DiClemente, C. C., & Carbonari, J. P. (1996). Doing the right thing at the
right time? The interaction of stages and processes of change in successful smoking
cessation. Health Psychology, 15(6), 462-468.
Prochaska, J. O. (1994). Strong and weak principles for progressing from
precontemplation to action on the basis of twelve problem behaviors. Health
Psychology, 13(1), 47-51.
Prochaska, J. O., & DiClemente, C. C. (1984). The transtheoretical approach: Crossing
traditional boundaries of therapy. Homewood, IL: Dow Jones Irwin.
Prochaska, J. O., DiClemente, C. C., & Norcross, J. C. (1992). In search of how people
change: Applications to addictive behaviors. American Psychologist, 47(9), 1102-
1114.
Prochaska, J. O., DiClemente, C. C., Velicer, W. F., & Rossi, J. S. (1993).
Standardized, individualized, interactive, and personalized self-help programs for
smoking cessation. Health Psychology, 12(5), 399-405.
Prochaska, J. O., & Norcross, J. C. (2001). Stages of change. Psychotherapy: Theory,
Research, Practice, Training, 38(4), 443-448.
Prochaska, J. O., Velicer, W. F., DiClemente, C. C., & Fava, J. (1988). Measuring
processes of change: Applications to the cessation of smoking. Journal of
Consulting and Clinical Psychology, 56(4), 520-528.
Prochaska, J. O., Velicer, W. F., Rossi, J. S., Goldstein, M. G., Marcus, B. H., &
Rakowski, W. et al. (1994). Stages of change and decisional balance for 12 problem
behaviors. Health Psychology, 13(1), 39-46.
92
Prochaska, J. M., Prochaska, J. O., & Levesque, D. A. (2001). A transtheoretical
approach to changing organizations. Administration and Policy in Mental Health,
28(4), 247-261.
Prochaska, J. O., & DiClemente, C. C. (1983). Stages and processes of self-change of
smoking: Toward an integrative model of change. Journal of Consulting and
Clinical Psychology, 51(3), 390-395.
Radius, S. M. (1980). Adolescent perspectives on health and illness. Adolescence,
15(58), 375-384.
Rezler, A. G., & Rezmovic, V. (1981). The learning preference inventory. Journal of
Allied Health, 10(1), 28-34.
Robertson, A., & Levin, M. L. (1999). AIDS knowledge, condom attitudes, and risk-
taking sexual behavior of substance-abusing juvenile offenders on probation or
parole. AIDS Education and Prevention, 11(5), 450-461.
Rollnick, S. (1996). Behaviour change in practice: Targeting individuals. International
Journal of Obesity and Related Metabolic Disorders : Journal of the International
Association for the Study of Obesity, 20 Suppl 1, S22-6.
Rollnick, S., Heather, N., Gold, R., & Hall, W. (1992). Development of a short
'readiness to change' questionnaire for use in brief, opportunistic interventions
among excessive drinkers. British Journal of Addiction, 87(5), 743-754.
Rollnick, S., Kinnersley, P., & Stott, N. (1993). Methods of helping patients with
behaviour change. BMJ (Clinical Research Ed.), 307(6897), 188-190.
Roscoe, B., & Kruger, T. L. (1990). AIDS: Late Adolescents' Knowledge and its
Influence on Sexual Behavior Libra Publishers, Inc.
Ross, M. W., & Rosser, B. S. (1989). Education and AIDS risks: A review. Health
Education Research, 4(3), 273-284.
Shew, M. L., & Remafedi. (1997). The validity of self-reported condom use among
adolescents. Sexually Transmitted Diseases, 24(9), 503-510.
Sickmund, M. (2002). Juvenile residential facility census: 2000. Washington, DC:
Office of Juvenile Justice and Delinquency Prevention.
Silver, H. F., & Hanson, J. R. (1992). Learning preference inventory (11th ed.).
Lincoln, NE: Hanson Silver Strong & Associates, Inc.
93
Snyder, H., & Sickmund, M. (2006). Juvenile offenders and victims: 2006 national
report No. 212906) US Department of Justice, Office of Justice Programs, Office of
Juvenile Justice and Delinquency Prevention.
Stevenson, M. R., & DeBord. (1988). Paper presented at The Society for Scientific Study
of Sex,
Stott, N. C., Kinnersley, P., & Rollnick, S. (1994). The limits to health promotion. BMJ
(Clinical Research Ed.), 309(6960), 971-972.
Teplin, L. A., Mericle, A. A., & McClelland, G. M. (2003). HIV and AIDS risk
behaviors in juvenile detainees: Implications for public health policy. American
Journal of Public Health, 93(6), 6-12.
Unpublished Report - A NEEDS ASSESSMENT OF PROBATIONARY YOUTH
Attending Community Education Centers And Participating in the JWCH Institute,
Inc.’s AIDS Video Project 1998-1999.
Velicer, W. F., DiClemente, C. C., Prochaska, J. O., & Brandenburg, N. (1985).
Decisional balance measure for assessing and predicting smoking status. Journal of
Personality and Social Psychology, 48(5), 1279-1289.
Velicer, W. F., Prochaska, J. O., Fava, J. L., Laforge, R. G., & Rossi, J. S. (1999).
Interactive versus non-interactive interventions and Dose–Response relationships
for stage-matched smoking cessation programs in a managed care setting. Health
Psychology, 18(1), 21-28.
Wallach, M. A., & Wing, C. W., Jr. (1968). Is risk a value? Journal of Personality and
Social Psychology, 9(1), 101-106.
94
Appendix 1
Intervention Instruments
For Data Collection
Facilitator Checklist
Participant Work Sheet
Permit to Participate
Learning Preference Inventory (LPI)
Example of a Youth’s LPI Report
Post-Test
Ability to Judge Risk Assessment
Stage of Change Instrument
Communication Do’s and Don’ts
Commitment Card
HIV Risk Assessment One Month Follow-up
HIV Risk Assessment Three Month Follow-up
95
Facilitator Checklist
96
Participant Work Sheet
97
98
Learning Preference Inventory (LPI)
99 99
100
101
102
103 103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
Example of a Youth’s LPI Report
119
120
121
122
123
124
125
126
127
128
129
130
131
132
Appendix 2
Comparison Group Instruments
For Data Collection
133
134
135
136
137
138
139
140
141
142
143
Appendix 3
Rankings of HIV Prevention Interventions for
Incarcerated Youth
The 24 articles/reports (including this intervention study) described in the literature
review section were scored and then ranked relative to the programs’
importance/significance to HIV prevention interventions among incarcerated youth.
The following questions were scored: 0 = missing; 1 = low; 2 = medium; 3 = high
1. Intervention has a clearly defined audience.
2. The Intervention has clearly defined goals and objectives.
3. The Intervention is based on sound behavioral and social science theory.
4. The Intervention is focused on reducing specific risk behaviors.
5. The Intervention provides opportunities to practice relevant skills.
6. The Intervention has an evaluation with outcome data.
7. Collected Follow-up data
8. Strength of study design:
o Experimental or not
o A comparison group
o Sample size
9. Controls for Confounding factors
o Internal
o External
10. Attrition
11. Validation of reports
144
The rankings were compiled into the Table below
Overview of National Programs
Author: Study Response Rate
Ranking
Score
Studies Inappropriate for scoring
Widom JJ Facilities
Hammett JJ Facilities
Greene: Health Promotion
Morrison-Beedy: Lessons Learned
Sanders: Developing a Program
Gelber: Developing a Program
Studies appropriate for scoring:
18. Cannon: Project IMPACT 100% (immed. Post) 13
17. Peres: Multiple Activities 13
16. Watson: SHIP 16
15. Lanier: AIDS Education 87% 17
14. Goldberg: Canada 22
13. Morris: Peers vs. Adults 22
12. Schlapman: Safe Choices 47% 22
11. Horan: A Peer-led Program Selection by length of stay 26
10. Jackson: Video Project 26
9. Magura: Intensive AIDS Education
72% @ post-test
66% Follow-up
26
8. Shelton: Peer & Leadership Training 100% (tx program) 26
7. Clark: Peer Training Curriculum 30
6. Gillmore: Group Skill-training 71% (paid $10) 35
5. Godin: Integrative Program
70% of all participants @
Post-test
58% of all participants @ FU
or 83% of all post-test
completers
35
4. Herman-Shipley: Tailored Intervention
86% intervention
54% controls
35
3. St. Lawrence: Risk Reduction Skills
Training
86% of completers @ follow-
up
35
2. Slonim-Nevo: Long-term 36
1. Slonim-Nevo: Residential Centers
92% interv group
85% control group
36
This means that out of the 18 studies that have been published and have the
appropriate information for ranking, my study ranks in the top 6.
145
Appendix 4
Results of the Validity Study of the
Intervention Curriculum
Using the attached script, matrix of the 8 sessions and explanations of the four Learning
Preferences (LPIs) and Stages of Change (SOC) nine (9) individuals with different
levels of expertise ranging from 20 years in Health Education to students in the
Masters’ of Public Health program agreed to participate in the study.
RESULTS
From the script:
91% identified the correct Stage of Change with the statement. However, only 57%
could identify the correct LP that the statement addressed.
For the sessions Matrix:
There was a general agreement that sessions 1,3, 4, and 5 (averages of 3.6, 4.0, 3.5, 3.6
respectively) attended to each of the 4 learning preferences (LPs). However, there was
a consensus that sessions 2 and 6 (averaged 3.33 and 3.00 respectively) only utilized 3
of the 4 LPs. Only 60% agreed that all 4 LPs were targeted by a particular activity
throughout all 8 sessions but 83% reported that at least 3 of the 4 LPs were attended by
an activity during each of the sessions.
Discussion:
The results tend to question the style of the survey administered rather than the validity
of the curriculum. The majority of the discrepancies have to do with the study subjects'
unfamiliarity with learning preferences whereas the stages of change are utilized
frequently and can be easily understood by someone who has never heard of it before.
The author believes that is why the LP inventory was interpreted by the creators as a
standardized test rather than subjectively as was asked of the study participants.
146
Please fill in the Stage of Change that is represented by the title statement and
the Learning Preference that each of the questions represents using the learning
preference matrix in the previous file and the Stages of Change matrix below.
147
Stages of Change
Five stages of change have been conceptualized for a variety of problem behaviors.
STAGES OF CHANGE (SOC)
Precontemplation
Contemplation
Ready-For-Action
Action -- Maintenance
No intention to
change
Intention to
change within 6
months
Actively planning
change
Making
changes
Sustaining
change, avoiding
relapse
5 Stages
Individuals
modify their
behavior,
experiences, or
environment in
order to
overcome their
problems.
Action
involves the
most overt
behavioral
changes and
requires
considerable
commitment of
time and
energy.
People work to
prevent relapse
and consolidate
the gains attained
during action. For
addictive
behaviors this
stage extends
from six months
to an
indeterminate
period past the
initial action.
Aware that a
problem exists
and are seriously
thinking about
overcoming it
but have not yet
made a
commitment to
take action.
Combines intention
and behavioral
criteria. Individuals
in this stage are
intending to take
action in the next
month and have
unsuccessfully taken
action in the past
year
No intention to
change behavior in
the foreseeable
future. Many
individuals in this
stage are unaware or
underaware of their
problems.
Stimulus
control,
counter-
conditioning,
helping
relationships
and
reinforcement
management
Self re-
evaluation,
environmental
re-evaluation
decisional
balance, self-
liberation, self-
efficacy
Stimulus control,
counter
conditioning,
helping
relationships,
reinforcement
management,
social liberation.
PROCESSES
OF
CHANGE
Consciousness
raising, dramatic
relief, self re-
evaluation,
environmental re-
evaluation, and
decisional balance
Self-liberation, self
efficacy, stimulus
control, counter-
conditioning,
helping relationships
148
References
Bacchetti P, Moss AR. (1989). Incubation period of AIDS in San Francisco. Nature.
Mar;338(6212):251-3
Braithwaite, R., Robillard, A., Woodring, T., Stephens, T., & Arriola, K. J. (2001).
Tattooing and body piercing among adolescent detainees: Relationship to alcohol
and other drug use. Journal of Substance Abuse, 9(13(1-2)), 5-16.
DiClemente, R., Lanier, M., & Horan, P. (1991). Comparison of AIDS knowledge,
attitudes, and behaviors among incarcerated adolescents and a public school sample
in San Francisco. American Journal of Public Health, 81(5), 628-630.
Lanier, M. M., Pack, R. P., & DiClemente, R. J. (1999). Changes in incarcerated
adolescents' human immunodeficiency virus knowledge and selected behaviors from
1988 to 1996. The Journal of Adolescent Health: Official Publication of the Society
for Adolescent Medicine, 25(3), 182-186.
Morris, R. E., Harrison, E. A., Knox, G. W., Tromanhauser, E., Marquis, D. K., &
Watts, L. L. (1995). Health risk behavioral survey from 39 juvenile correctional
facilities in the united states. The Journal of Adolescent Health : Official
Publication of the Society for Adolescent Medicine, 17(6), 334-344.
Morrison, D. M., Baker, S. A., & Gillmore, M. R. (1994). Sexual risk behavior,
knowledge, and condom use among adolescents in juvenile detention. J.Youth
Adolesc., 23(2), 271-288.
Robertson, A., & Levin, M. L. (1999). AIDS knowledge, condom attitudes, and risk-
taking sexual behavior of substance-abusing juvenile offenders on probation or
parole. AIDS Education and Prevention, 11(5), 450-461.
149
150
151
152
153
154
155
Abstract (if available)
Abstract
The prevalence of HIV/AIDS is very low among incarcerated youth (~ 1%) compared to the incarcerated adult population (4% to 6%). However, these youth report practicing many behaviors, such as unprotected sex with multiple partners and sharing of tattoo needles that puts them at risk for infection. To reduce the risk-taking behaviors of youth detained by the Los Angeles County Juvenile Justice system an HIV/AIDS prevention intervention tailored to learning styles and stages of change (SOC) was developed. This study used an observational, prospective parallel group design with historical comparisons and a randomized intervention sample to test the effectiveness of this program. The magnitude and consistency of the results make it an important contribution to research on HIV/AIDS prevention among incarcerated youth.
Linked assets
University of Southern California Dissertations and Theses
Conceptually similar
PDF
Factors and correlates of sexual behaviors among young adults from continuation high schools
PDF
Cultural risk and protective factors for tobacco use behaviors and depressive symptoms among American Indian adolescents in California
PDF
Trends and charateristics of causes of death among persons with AIDS in Los Angeles County, 1990-2002
Asset Metadata
Creator
Shipley, Nikki
(author)
Core Title
Stage of readiness and learning styles -- a tailored HIV/AIDS prevention intervention for youth detained in the Los Angeles County Juvenile Justice system
School
Keck School of Medicine
Degree
Doctor of Philosophy
Degree Program
Preventive Medicine (Health Behavior)
Publication Date
11/04/2008
Defense Date
08/21/2008
Publisher
University of Southern California
(original),
University of Southern California. Libraries
(digital)
Tag
high risk youth,HIV/AIDS prevention,incarcerated youth,learning styles,OAI-PMH Harvest,stages of change,tailored prevention intervention
Place Name
Los Angeles County
(counties)
Language
English
Contributor
Electronically uploaded by the author
(provenance)
Advisor
Valente, Thomas W. (
committee chair
), Baezconde-Garbanati, Lourdes (
committee member
), Murphy, Sheila (
committee member
)
Creator Email
hermansh@usc.edu,nikkishipley@sbcglobal.net
Permanent Link (DOI)
https://doi.org/10.25549/usctheses-m1743
Unique identifier
UC184800
Identifier
etd-Shipley-2444 (filename),usctheses-m40 (legacy collection record id),usctheses-c127-126882 (legacy record id),usctheses-m1743 (legacy record id)
Legacy Identifier
etd-Shipley-2444.pdf
Dmrecord
126882
Document Type
Dissertation
Rights
Shipley, Nikki
Type
texts
Source
University of Southern California
(contributing entity),
University of Southern California Dissertations and Theses
(collection)
Repository Name
Libraries, University of Southern California
Repository Location
Los Angeles, California
Repository Email
cisadmin@lib.usc.edu
Tags
high risk youth
HIV/AIDS prevention
incarcerated youth
learning styles
stages of change
tailored prevention intervention