Transcript Predictors of Change in HIV Risk Factors for Adolescents

Predictors of Change in HIV Risk
Factors for Adolescents Admitted to
Substance Abuse Treatment
Passetti, L. L., Garner, B. R., Funk, R.,
Godley, S. H., & Godley, M. D.
Chestnut Health Systems
JMATE 2008
Acknowledgements
Preparation of this presentation was supported by
funding from the following sources:
– Center for Substance Abuse Treatment (Strengthening
Communities-Youth project grant no. TI 13356)
– National Institute on Drug Abuse (grant no. DA
018183)
– National Institute on Alcohol Abuse and Alcoholism
(grant no. AA 010368).
HIV Infection in Adolescents
• Estimated 5,322 adolescents living with AIDS in
the U.S.
– 46.7% increase since 2001 (CDC, 2005)
• Average of 10 years from HIV infection to
development of AIDS
– Many young adults likely infected as teenagers
(National Institute of Allergy and Infectious Disease,
2000)
HIV Risk in Adolescents Presenting
to Substance Abuse Treatment
• Sexual activity at early age
• Injection drug use
• Unprotected sex
• Sex with injection drug users
• Multiple partners
• Victimization
• Sex under the influence
• Multiple risk behaviors
(Ammon et al., 2005; Deas-Nesmith et al., 1999; Jainchill et al., 1999; Malow et
al., 2001; Tapert et al., 2001)
Purpose
• For adolescents admitted to substance abuse
treatment, identify variables that most
strongly predict the transition from:
Presence
of any
HIV risk
factor
Follow-up Interview
Absence of
HIV risk
factors
Next Follow-up Interview
Sample
• 283 adolescents
– Strengthening Communities - Youth (SCY)
• n=113
• Admitted to outpatient substance abuse
treatment
– Assertive Continuing Care (ACC-2)
• n=170
• Admitted to residential substance abuse
treatment
Participant Characteristics at
Intake (n=283)
• Average Age: 16
• Caucasian: 70%
• Male: 65%
• Main substances of
choice: marijuana,
alcohol
• Average years of
education: 9
• In school: 83%
• Employed: 39%
• Involved with criminal
justice system: 78%
Measurement
• Global Appraisal of Individual Needs (GAIN)
– Administered at intake and quarterly follow-up
intervals
• 3, 6, 9, and 12 months post-intake for SCY
• 3, 6, 9, and 12 months post-discharge for ACC-2
– Follow-up rates ranged from 90% to 96%
Analysis
• Step One - Univariate logistic regression
– Identify variables that predict the transition from:
Presence
of any
HIV risk
factor
Follow-up Interview
Absence of
HIV risk
factors
Next Follow-up Interview
(i.e., from 3 to 6 months, 6 to 9 months, 9 to 12 months)
Analysis
• Step Two - Multivariate mixed nominal
regression
– Identify strongest predictors of transition
– Enter significant predictors from univariate
analysis simultaneously
Unit of Analysis
• 283 adolescents
– 477 observations in which adolescents reported
at least one risk factor for HIV infection
Predictors
• Intake Variables
–
–
–
–
–
–
Age
Gender
Minority (Yes/No)
Years of education
Symptoms of internalizing disorder (Yes/No)
Symptoms of externalizing disorder (Yes/No)
Predictors
• Follow-up Variables (During the past 90 days)
–
–
–
–
–
–
In school (Yes/No)
Employed (Yes/No)
Involved with the criminal justice system (Yes/No)
Substance Frequency Scale (SFS) – 8 items
Substance Problem Scale (SPS) – 16 items
Recovery Environment Risk Index (RERI) – 13 items
Predictors
• Follow-up Variables (During the past 90 days)
–
–
–
–
–
–
–
Social Risk Index (SRI) – 6 items
Treatment Motivation Index (TMI) – 5 items
Treatment Resistance Index (TRI) – 4 items
Problem Orientation Scale (POS) – 5 items
Weeks in substance abuse treatment
Weeks in mental health treatment
Weeks in a controlled environment
Outcome Measure
– HIV Risk Status (Yes/No)
• Endorsed any of the following HIV risk factors
during the past 90 days:
– Needle use
– Sex with a needle user
– Sex while adolescent or partner was high on alcohol or
drugs
– Unprotected sex
– Multiple sex partners (two or more)
– Trading sex for drugs/money
– Victimized (sexually, physically, or emotionally)
Transition Period: 3 to 6 months
67%
Presence
33%
Absence
Presence
(n = 117)
HIV Risk Status:
3 Months
HIV Risk Status:
6 Months
Transition Period: 6 to 9 months
71%
Presence
29%
Absence
Presence
(n = 174)
HIV Risk Status:
6 Months
HIV Risk Status:
9 Months
Transition Period: 9 to 12 months
61%
Presence
39%
Absence
Presence
(n = 186)
HIV Risk Status:
9 Months
HIV Risk Status:
12 Months
Results
Univariate Logistic Regression
Odds Ratio
95% CI
p=
Age
0.83
(0.71, 0.99)
0.03
Female
0.72
(0.48, 1.07)
0.11
Minority
0.99
(0.65, 1.50)
0.96
Years of
Education
0.82
(0.71, 0.94)
0.01
Symptoms of
internalizing
disorder
0.80
(0.55, 1.18)
0.26
Symptoms of
externalizing
disorder
1.37
(0.90, 2.08)
0.15
Intake Variables
Results
Univariate Logistic Regression
Odds Ratio
95% CI
p=
In School
1.15
(0.77, 1.71)
0.50
Employed
0.87
(0.59, 1.27)
0.47
Involved with CJS
1.82
(0.87, 1.90)
0.22
Substance
Frequency Scale
0.83
(0.69, 0.98)
0.03
Substance Problem
Scale
0.81
(0.68, 0.97)
0.21
Follow-up
Variables
Results
Univariate Logistic Regression
Odds Ratio
95% CI
p=
Recovery Environment Risk
Index
0.82
(0.67, 0.99)
0.03
Social Risk Index
0.79
(0.65, 0.97)
0.02
Treatment Motivation Index
1.00
(0.82, 1.23)
0.99
Treatment Resistance Index
0.79
(0.66, 0.95)
0.01
Problem Orientation Scale
0.90
(0.73, 1.09)
0.28
Weeks in SA Treatment
1.10
(1.06, 1.51)
0.00
Weeks in MH Treatment
1.02
(0.99, 1.04)
0.07
Weeks in a Controlled
Environment
1.09
(0.99, 1.19)
0.06
Follow-up Variables
Results
Multivariate Mixed Nominal Regression
β
Odds
Ratio
95% CI
p=
Intercept
1.65
Age
-1.95
0.79
(0.63, 1.00)
0.05
Recovery
Environment Risk
Index
-2.10
0.78
(0.62, 0.98)
0.04
Treatment Resistance
Index
-2.12
0.79
(0.64, 0.98)
0.03
Conclusions
• In this sample, the strongest predictors of
transitioning to the absence of any HIV risk
factors were:
– Younger age
– Lower recovery environment risk
– Lower treatment resistance
Strengths
– Few studies examining change in HIV risk
factors over time
– Adolescents in OP and residential treatment
– High follow-up rates
Limitations
– Self-report
– No measure of HIV risk interventions received
during or after treatment
Implications
• Interventions with this population may be
developed and tested that are tailored by:
• Age
• Level of risk in the recovery environment
• Level of treatment resistance
Implications
• While 1/3 of the analyzed transitions demonstrated
improvement in HIV risk, 2/3 represented the
same or greater levels of risk
• Longer and/or repeated assessments and
interventions may be required to initiate and
sustain a reduction in HIV risk