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

658 Abstracts 33333-33337

12th World AIDS Conference

playing, distribution of educational materials and condoms. Consistent condom
use were emphasized in all activities; and 5) an end-project evaluation survey.
Results: 119 prostitutes in Chengjiang, 128 in Ruili and 66 in Longchuan participated baseline survey and 134 in Chengjiang, 112 in Ruili and 51 in Longchuan
participated evaluation survey. After 3 months intervention, the knowledge of
three transmission routes of HIV increased from 18% to 98% in Chengjiang, 44%
to 88% in Ruili and 0% to 59% in Longchuan, the knowledge of condom reducing
the risk of STD/HIV infection increased from 58% to 99% in Chengjiang, 55%
to 91% in Ruili and 55% to 90% in Longchuan. Condom use in last 3 sexual
intercourse increased from 31% to 77% in Chengjiang, 51% to 75% in Ruili and
39% to 41% in Longchuan. Intervention was a strong independent contributor to
the behavior change based on modeling.
Conclusion: In natural settings, it is feasible to implement intervention programs among prostitutes and to achieve objectives of increasing AIDS knowledge
and increasing condom use rate in Yunnan, China. We recommend that multiple
strategies should be used to target establishment owners as well as prostitutes in
future programs. Due to high mobility and low respectability of prostitutes, health
workers and establishment owners should play more significant roles than peers
in interventions.
S33333 1 Religion and HIV risk perception among African
immigrants living in New York City
Mulusew Bekele, Kim Nichols. African Services Committee, Inc. 28 East 35
Street, New York, NY, 10016, USA
Objectives: to compare HIV risk perception between Christian and Moslem
African immigrants to New York City.
Design: Cross-sectional survey.
Methods: As part of a structured outreach survey, a three-point scale of HIV
risk perception and a comprehensive set of demographic variables were administered over a one-year period to a sample of 3,655 African immigrants residing
in the five boroughs of New York City. Group mean scores for three HIV risk
perception items on a five point likert-scale (1 = strongly disagree; 5 = strongly
agree) were recoded, where higher scores had an inverse relationship with lower
risk perception.
Results: A total of 1,019 (28.3%) Christian, 2,393 (66.4%) Muslim, 186 (5.2%)
atheist or agnostic, and 5 (0.1%) other were interviewed. Moslems (M = 2.48,
SD = 1.02) had higher mean group score than Christians (M = 2.24, SD = 0.88),
indicating lower risk perceptions among Moslems (t = -6.78, df = 2079, two tailed
p <.05). In addition, a one-way ANCOVA showed that when educational level
and length of stay in the United States were covaried out, the difference between
risk perception between Moslems and Christians was significant (F1,3014 = 34.92,
p < 0.05). Educational level and years in the United States had no significant
effect on risk perception.
Conclusions: The data suggests that religious background influences risk
perception, a crucial component in behavioral interventions. Consequently, educational interventions designed to impact on risk perception of African immigrants
should take into account religious background.
[33334 Success of HIV/AIDS program through inclusion in a
community managed child survival project
Nalini Abraham. Plan International, 8/16 Sarva Priya Vihar, New Delhi, India
Issue: A relatively small effort devoted to HIV/AIDS can be effective if integrated
with a community managed child survival project.
Project: A community managed Child Survival project, with USAID funding was
implemented in an urban slum in New Delhi, India. A Reproductive Health project
with a HIV/AIDS prevention component, funded by MacArthur Foundation, was
also implemented in the same community. The interventions targeted mothers of
children in the age group 0-23 months, couples in the reproductive age group
and adolescents. The strategies employed included Outreach through Community Health Guides, social marketing of condoms, prevention and treatment of
Reproductive Tract Infections (RTIs) and STDs and AIDS Awareness.
Results: Within 2 years of the project 73% of eligible women and 33% of men
had received training on Reproductive Health and AIDS. Communication skills
were imparted to 800 adolescents for spreading AIDS awareness. Awareness of
HIV increased from 26% to 76%. Knowledge of HIV prevention through condom
use increased from 39% to 60%. The HIV/AIDS awareness program has also
created a huge demand for treatment of RTIs.
Lessons Learned: Successful incorporation of AIDS prevention and control
interventions depends on the existence of an ongoing health & development program, availability of existing community based organisations, and trained health
& development workers from the community. Reproductive and child health
interventions provide an appropriate entry point for AIDS programs.

409*/ 33335 The impact results of a peer-driven intervention to
combat HIV among drug injectors
Robert Broadhead, D.D. Heckathorn, H. Madray, R.J. Mills, J. Hughes. Dept. of
Sociology, Univ of Connecticut, Storrs, CT USA
Background: Since 1985, community outreach efforts to combat AIDS among
injection drug users (IDUs) in the United States (US) have overwhelmingly depended on a provider-client model that relies on staffs of professional outreach

workers. In this model, IDUs are regarded as recipients of services provided by
outreach workers.
Objectives: We report on a multi-year empirical comparison, funded by the
US National Institute on Drug Abuse (R01 DA08014), of a traditional outreach
model with an innovative social-network model, termed a peer-driven intervention
(PDI). The PDI provides IDUs with guidance and structured rewards to play a
much more active role in the outreach process than is allowed in the traditional
model, thereby harnessing peer pressure on behalf of HIV-prevention efforts. In
the PDI, IDUs are relied on to carry out the core activities provided by professional
outreach workers.
Methods: We compare the performance of a traditional outreach intervention
and a peer-driven intervention that were implemented in medium-sized towns
in eastern and central Connecticut. Comparisons are based on the number and
representativeness of IDUs recruited at each site, the effectiveness of HIV-prevention education, compliance rates with AIDS risk reduction recommendations, and
relative cost. The analyses are based on 552 initial interviews and 190 six-month
follow-up interviews conducted during the first two years of each intervention's
operation.
Results: Both interventions produced significant reductions in HIV risk behaviors, as measured using self-reports. The peer-driven intervention outperformed
the traditional intervention with respect to the number of IDUs recruited, the
ethnic and geographic representativeness of the recruits, and the effectiveness of
prevention education. In addition, the costs of recruiting IDUs into the intervention
and educating them about HIV in the community was only one thirtieth as much
in the peer-driven intervention as in the traditional intervention.
Conclusions: The findings demonstrate that given guidance and nominal
incentives, IDUs can play a far more extensive and active role in community
outreach efforts than the traditional model allows. The findings also demonstrate
that both interventions reduce HIV-associated risk behaviors, but the peer-driven
intervention reaches a larger and more diverse set of IDUs, and does so at much
less expense.
133336 Religion and HIV knowledge among African immigrants
living in New York City
Mulusew Bekele, Kim Nichols. 1African Services Committee, Inc, 28 East 35
Street, New York, NY, 10016, USA
Objectives: to compare HIV knowledge between Christian and Moslem African
immigrants to New York City.
Design: Cross-sectional survey.
Methods: As part of a structured outreach survey, a four item HIV knowledge
construct was used to measure level of HIV knowledge in the two groups, and a
comprehensive set of demographic variables were administered over a one-year
period to a sample of 3,655 African immigrants residing in the five boroughs of
New York City. Mean sum knowledge scores for four HIV knowledge items were
computed by recoding correct responses into 1' and incorrect or "do not know"
responses into '0'. A four point mean sum scoring scale was used, where, '4';
indicated an 'all correct' sum mean score and 0' indicated an 'all incorrect' sum
mean score.
Results: A total of 1,019 (28.3%) Christian, 2,393 (66.4%) Muslim, 186 (5.2%)
atheist or agnostic, and 5 (0.1%) other were interviewed. Christians (M = 3.39,
SD = 1.01) had higher mean sum score than Moslems (M = 2.89, SD = 1.22),
indicating a higher level of HIV knowledge (t = 12.09, df = 2333, two tailed p <
0.05). In addition, a one-way ANCOVA showed that when educational level and
length of stay in the United States were covaried out, the difference in knowledge
scores between Moslems and Christians was significant (F1,3145 = 27.8, p <
0.05). Educational level and years in the United States had no significant effect
on knowledge scores for the two groups.
Conclusion: Educational/Behavioral interventions designed to impact on HIV
knowledge of African immigrants should take into account religious background.
33337] Willingness to participate in a vaccine preparatory study
in Rio de Janeiro, Brazil
Paulo Longo1, S.N. Silva1, M. Schechter2, L.H. Harrison3. 1Projeto Praca Onze
Avenida Presidente Vargas 2863 Cidade Nova-Rio de Janeiro; 2 Universidade
Federal do Rio de Janeiro Rio de Janeiro RJ, 3 University of Pittsburgh
Pittsburgh, Brasil
Objectives: 1) To investigate the effectiveness of different strategies for recruiting
volunteers to participate in a vaccine preparatory study; 2) To understand the
motivation of volunteers to participate in such a study.
Methods: Potential participants in a HIV seroincidence study with men who
have sex with men were recruited utilizing different strategies. Volunteers were
interviewed by a multi professional team, between August/95 and May/1997.
Results: 856 volunteers were recruited. The most successful strategy was
recruitment at anonymous testing sites (34%), followed by snow ball (34%) and
direct contact in places frequented by gay men (24%). Other recruiting strategies,
such as announcements in the media, account only for 8%. Considering the

motivations to participate in vaccine trials, 95% stated that they want to help to
stop the HIV pandemic; 84% to help their community and 43% to honor friends
or others affected by HIV/AIDS.
Conclusions: Indirect strategies, such as media campaigns, have not contributed much to recruitment of volunteers, suggesting the importance of direct
contact between the recruitment team, those who are already volunteers and
potential volunteers. Main stated reasons for joining the study appear to be re