Intervention Description
Intervention Package Information
Evaluation Study and Results
References and Contact Information

Intervention Description

Target Population
Out-of-treatment active crack and injection drug users

Goals of Intervention

• Eliminate or reduce sex risk behaviors
• Eliminate or reduce drug risk behaviors

Theoretic Basis

• Health Belief Model
• Theory of Protection Motivation
• Transtheoretical Stages of Change Model

Intervention Duration
Nine sessions over a 4-6 month period

Intervention Settings
Public area/community, storefront offices, and local community recreational facilities

Deliverer
Counselors, outreach staff, and peer social support buddies

Delivery Methods

• Counseling
• Exercise
• Games
• Goal setting/plan
• Group Discussion
• Role play
• Video

Go to top

Intervention Package Information

The intervention package and training are available through CDC’s Diffusion of Effective Behavioral Interventions (DEBI) project

Go to top

Evaluation Study and Results

The original evaluation was conducted in Long Beach, California, between January 1992 and December 1996. This was one of the 23 cooperative agreement studies under the NIDA Cooperative Agreement for AIDS Community-Based Outreach/Intervention Research Program.

Key Intervention Effects

• Reduced drug injection
• Reduced sharing works

Study Sample
The analytic sample of 726 drug users is characterized by:

• 47% African American, 28% White, 20% Hispanic, 4% Native American, 1% Asian
• 67% Male, 33% Female
• Mean age of 39 years
• 59% completed high school education

Recruitment Settings
Street outreach in public areas and the community

Eligibility Criteria
Men and women were eligible if they were at least 18 years old, reported injecting drugs or using crack cocaine in the past 30 days, provided confirmation of recent drug use through urine based drug testing or track marks, and were not in drug treatment in past 30 days.

Assignment Method
The community from which participants were sampled was divided into two comparable geographic regions, each containing 3 zip codes. One region was randomly selected to receive the intervention initially, with one crossover midway through the study. A total of 1,362 eligible drug users were then assigned to 1 of 2 groups based on their “hangout” zip code: Safety Counts enhanced intervention (n = 687) or NIDA standard comparison (n = 675)

Comparison Group
The NIDA standard intervention was delivered in public areas and communities to individuals, or small groups or pairs of drug users for outreach. The intervention was delivered in two 20-30 minute sessions by an indigenous peer outreach worker and counselor, and included counseling, skills building, drug-focused prevention education as mandated by NIDA to review basic HIV/AIDS information, optional HIV testing, and referral to other services.

Relevant Outcomes Measured and Follow-up Time

• Sex behaviors during past 30 days include: having any sex, percentage of times used condoms, percentage of times always used condoms, having 2 or more sex partners, exchanged sex for drugs, and having sex with an IDU.
• Needle-related risk behaviors during past 30 days include: injecting any drugs, number of times injected, percentage of times did not use own works, and percentage of times used unclean needles.
• Sex and drug outcomes were measured at 5 to 9 months after baseline, which translates to 1 to 5 months after intervention.

Participant Retention

• Enhanced Safety Counts Intervention:

  74% retained at 1-5 months after intervention

• Standard Intervention:

  76% retained at 1-5 months after intervention

Significant Findings

• The participants in the Enhanced intervention were significantly less likely to report injecting drugs (p < 0.05) than those in the Standard at 1 to 5 months after intervention.
• Among injectors only, the percentage of times people did not use their own works was significantly lower in the Enhanced intervention compared to the Standard at 1 to 5 months after intervention (p < 0.05).

Considerations

• This intervention fails to meet the best-evidence criteria due to assigning groups of individuals to study conditions while analyzing at the individual level, a small number of participants being excluded from analyses after assignment, and a short follow-up time.
• There were three significant baseline demographic differences. The standard intervention group included more Hispanics (23% vs. 19%), fewer Asians (0.6% vs. 2%), and fewer married people (8% vs. 12%) than the Enhanced intervention group.
• Of the 687 participants assigned to the Enhanced intervention group, 462 (67%) did not receive all 9 sessions as allocated, whereas only 61 (9%) of the 675 participants assigned to the Standard intervention participants did not receive the full 2 sessions as allocated.
• Among those that completed the intervention as allocated, participants in the Enhanced group were significantly less likely to report having sex at follow-up compared to those in the standard group (p < .05). This finding does not satisfy GOOD-EVIDENCE efficacy criteria due to a potentially biased restriction based on complete exposure.
• Among those that completed at least 7 out of 9 sessions, participants in the Enhanced group were significantly more likely to report an increase in condom use from baseline to follow-up as compared to those in the Standard group (p = .01). This finding does not satisfy GOOD-EVIDENCE efficacy criteria due to a potentially biased restriction based on complete exposure.
• Among injectors that completed at least 7 out of 9 sessions, participants in the Enhanced group were significantly more likely to report decreases in high-risk drug behaviors from baseline to follow-up – stopped injecting drugs, p < .001, decreased number of days injected drugs, p = .001, decreased frequency of injecting drugs, p < .001 – as compared to those in the Standard group. These findings do not satisfy GOOD-EVIDENCE efficacy criteria due to a potentially biased restriction based on complete exposure.

Go to top

References and Contact Information

• Hershberger, S.L., Wood, M.M., & Fisher, D.G. (2003). A cognitive-behavioral intervention to reduce HIV risk behaviors in crack and injection drug users. AIDS and Behavior, 7, 229-243.
• Rhodes, F., Wood, M.M., & Hershberger, S.L. (2000). A cognitive-behavioral intervention to reduce HIV risks among active drug users: Efficacy study. Sacramento, CA: CA Department of Health Services, Office of AIDS, 23-33.
• Rhodes, F., & Woods, M.M. (2000). A cognitive-behavioral intervention to reduce HIV risks among active drug users. Sacremento, CA: CA Department of Health Services, Office of AIDS, 2-22.
• Wood, M.M., & Rhodes, F.R. (2000). A cognitive-behavioral intervention to reduce HIV risks among active drug users: Implementation issues. Sacramento, CA: CA Department of Health Services, Office of AIDS, 35-51.

  Researcher: Dr. Scott Hershberger
  Department of Psychology, California State University, Long Beach
  1250 Bellflower Blvd.
  Long Beach, CA 90840
  email: scotth@csulb.edu
  
  Dr. Fen Rhodes, UCLA
  243 Park Avenue
  Long Beach, CA 90803
  email: fenrhodes@gmail.com

Go to top