Bridging the Gap: Conference Record [Abstract book, International Conference on AIDS (12th: 1998: Geneva, Switzerland)]

12th World AIDS Conference

Abstracts 43292-43295 905

Methods: The Women and Infants Demonstration Project (WIDP) was a community intervention study conducted in Pittsburgh, Portland and Philadelphia
between 1993 and 1996. The WIDP aimed at changing the behaviors of women
who were at high risk for HIV, including sex traders, and used paid outreach
workers, peer volunteers, project-produced print materials and community networking to help women move toward the goal of using condoms consistently with
both main and non-main sex partners. A baseline survey in 1993 of 1836 women
included 312 women who reported trading sex. In 1996, following a 24 month
intervention in the treatment communities, a post-intervention cross-sectional
survey of 1889 women included 331 women who reported trading sex.
Results: Sex with non-main sexpartners: Among sex traders in the intervention communities, reported condom use during last sex increased from 52.7%
to 73.1% over the study period compared to a slight decrease from 64.6% to
59.3% among sex traders in the comparison communities. Also, among sex
traders in the intervention communities reported consistent (100%) condom use
for at least the past 3 months increased from 28.3% to 53.2% over the study
period compared to a modest increase from 30.5% to 33.0% among sex traders
in the comparison communities. Sex with main sexpartners: Among sex traders
in the intervention communities, reported consistent (100%) condom use for at
least the past 3 months increased from 4.3% to 14.3% compared to a slight decrease from 13.3% to 11.4% among sex traders in the comparison communities.
In each case, the improvement in safe sex behaviors was greater among sex
traders in intervention communities and was statistically significant at the p = 0.01
level.
Conclusion: In this study, a community-level intervention based on behavioral
science theory increased condom use with both main and non-main sex partners.
These changes may lead to reduced risk for HIV.
43292 1 Meta-analysis of sexual risk reduction studies among
men who have sex with men
Wayne D. Johnson1, G. Ramirez2, S. Semaan', L.R. Norman', E. Sogolow1,
M.D. Sweat3, R.M. Diaz4. ' CDC, 1600 Clifton Rd., NE Mailstop, Atlanta, GA;
2University of North Texas, Forth Worth, TX; 3Johns Hopkins University,
Baltimore, MD; 4Center for AIDS Prevention Studies, San Francisco, CA, USA
Background: The United States Centers for Disease Control and Prevention has
initiated a system to identify and characterize behavioral interventions to reduce
HIV transmission. In this presentation we synthesize results of research studies
that evaluate interventions for men who have sex with men (MSM), and report a
meta-analysis based on standardized program effect sizes.
Methods: We performed a comprehensive search for published and unpublished reports of HIV intervention studies, 1988 to present, and applied systematic
criteria to select studies that measured HIV sexual risk behavior outcomes among
MSM. Studies that assigned participants to intervention and comparison groups
without apparent bias were eligible. We extracted data on sexual risk behavior,
and used meta-analysis to compute a weighted average (WA) effect size and
95% confidence interval (CI) for the selected studies.
Results: We have obtained 53 reports of HIV behavioral interventions for
MSM; of these, 43 measured sexual risk outcomes, including only 14 studies that
assigned intervention groups without bias. Of the 14 studies, 8 reported outcome
data necessary to determine intervention effect sizes; all reported some measure
of unprotected sex. Among the 8 studies, 4 interventions were delivered through
small groups featuring education, development of skills and social support. Two
studies tested community-level interventions featuring peer outreach and media
campaigns, 1 evaluated individual HIV counseling with optional antibody testing,
and I featured analysis of self-justifications after sexual risk. Of the 8 studies, 4
were published after 1994. All 8 studies included more than 50 participants; 4 had
>200 participants. The summary effect (WA = 0.16) was positive and statistically
significant (95% CI, 0.06 to 0.26), indicating that intervention groups reported less
risk behavior than comparison groups.
Conclusions: Meta-analysis of rigorously evaluated interventions for MSM reveals a statistically significant reduction in unprotected sex. This weighted average
effect corresponds to an 8% reduction in HIV sexual risk behavior, and was statistically similar among the 8 eligible studies. The risk for this population, however,
has not been eliminated. Further research is needed to determine how various
study characteristics contribute to the effects of these prevention approaches,
and to develop and test new strategies that may yield greater potency.
43293 An evaluation of community-based STD/HIV prevention
work in a public sex environment (PSE)
Rebecca French', R. Power1, S. Mitchell2, M. Dockrell2. 'Sexually Transmitted
Diseases, University College London Medical School; 2Gay Men Fighting AIDS,
London, UK
Background: The most extensive year to year evaluation of PSE outreach work
in the UK has come from Gay Men Fighting AIDS' (GMFA) Hampstead Heath
Project, which is a peer-led HIV prevention and safer sex education campaign. It
is a nighttime service run by volunteers. Condom packs and safer sex literature

are distributed by volunteers or are available from illuminated 'glowboxes'. This
research was an external process evaluation of the intervention.
Methods: Data were collected through surveys of gay and bisexual men in a
variety of settings (n = 871). In-depth interviews (n = 38) were conducted with
PSE users, professionals working within the broad field of gay men and health
promotion, and GMFA volunteers and workers. Observational data were collected
during volunteer shifts.

Results: Hampstead Heath is predominantly used by gay-identified men aged
between 20-49 years. A wide range of health promotion needs were reported,
including: cruising safety, safer sex information and provision of condoms. There
was evidence of unprotected anal intercourse taking place. A total of 100 088
condoms were distributed over the intervention's five month period (at weekends
an average of 430 per night). The commitment of the volunteers played a vital role
in the delivery of the intervention. GMFA's established identity and peer emphasis
made it acceptable to the target group.
The Heath Project provided an efficient, valued and cost-effective health promotion service. The evaluation highlighted the need for such interventions in PSEs
as men were still putting themselves at risk of acquiring STDs and HIV infection,
despite being aware of health promotion messages. Consideration should be
given to the wider application of its model in other PSEs.
43294 A randomized clinical trial of three intervention models'
effectiveness in lowering STD/HIV risk of
substance-dependent adolescents
Janet St. Lawrence', M.H. Hennessy', T.L. Brasfield2. 1Centers for Disease
Control & Prevention 1600 Cliftonrdne, MS-E44, Atlanta, GA., 30333; 2Jackson
State University Jackson MS, USA
Objectives: To evaluate behavioral interventions for substance dependent adolescents and to assess whether behavioral changes sustained after a randomized
clinical trial of the interventions.
Methods: 161 drug-dependent adolescents were randomly assigned to a
health education curriculum (HE); a cognitive-behavioral risk reduction intervention (BST); or the cognitive-behavioral intervention plus a risk sensitization
component that included transformed images depicting how each youth might
appear in late stage AIDS (BST + RS). Youth participated in twelve 90-minute sessions and in baseline, post intervention, 6-, and 12-month follow-up assessments.
89% were retained through one-year.
Results: Multivariate analyses revealed significant changes over time in AIDSrelated knowledge, attitudes toward prevention and toward condoms, sensationseeking propensity, self efficacy, and risk perceptions (all p < 0.0001); acquisition
of social competency and problem-solving skills (all p < 0.01); less risky sexual
behavior (number of sex partners, number of intercourse occasions, and frequency of sex in high risk contexts such as trading sex for drugs; all p. 0.0001).
While the interventions usually did not differ on mediating variables, consistently
higher values on attitudes toward condoms and preventive attitudes appeared in
the BST and BST + RS groups. Significant changes also were present in the %
who were sexually-active:

% Engaging in Sex last 30 days
Baseline
6-month Follow-up
12-month Follow-Up

HE
82%
60%
51%

BST
82%
45%
55%

BST + RS
93%
360
45%

Conclusions: The interventions produced substantial changes in knowledge,
attitudes, and risk behavior for youth receiving all three interventions with greater
skill acquisition for the BST and BST + RS conditions and sustained risk reduction
across a one-year follow up period.
43295 Intervening to address "heat of the moment" thinking
that leads to unsafe sex
Ron Gold1'2, M.J. Skinner2, D.A. Rosenthal3. 'School of Psychology; 2Deakin
University, 221 Burwood Highway, Victoria; 3La Trobe University, Melbourne,
Australia
Problem: There is evidence to suggest that gay men who decide to engage in
unprotected anal intercourse (UAI) generally try to justify this decision to themselves at the time they make it. How might one address self-justifications for
unsafe sex in educational interventions?
Methods: Evidence is reported from:
(a) Three studies in which gay men who had recently engaged in UAI (n = 734)
described the relevant encounter in detail, including any self-justifications they
used.
(b) Two intervention studies, in which men who had recently had UAI were
either given very brief educational interventions or assigned to a no-intervention
control group. Sexual diaries were used to record behaviour. Study 1 (n = 109)
investigated the effect of a questionnaire requiring the men to recall an occasion
when they had UAI and to reflect on the self-justifications they used on that
occasion. Study 2 (n = 92) investigated the effect of exposure to posters that used
photos and thought bubbles to depict gay men employing self-justifications.
Results: (a) Data are reported on the extent to which various self-justifications
are associated with decisions to have UAI that are made earlier vs. later in the
encounter. A commonly used self-justification, involving a resolution to withdraw
before ejaculation, is particularly associated with "last minute" decisions to have
UAI. Much the same is true of some other common self-justifications.
(b) The Study 1 intervention produced a significant reduction in sexual risk-taking, but the Study 2 intervention did not.
Conclusion: The resolution to withdraw seems to derive mainly from "heat of
the moment" thinking, rather than from a stable belief that withdrawal reduces
risk to an acceptable level. Simply informing gay men of the risks involved in withdrawal is thus unlikely to be effective. A better way of addressing self-justifications
like this is to stimulate reflection on them, but this works only where the men see