Abstracts Vol. 1 [International Conference on AIDS (11th: 1996: Vancouver, Canada)]

Mo.D.1799 - Mo.D.1803
associated with increases in receiving street outreach services among on the street risk
populations. Specfic enhancements, including use of peer outreach workers nd nol
vans can, have a measurable impact on the amount of services delivered to hI gh rsk, outof treatmenorit populations, and in some settings can be associated with beh ior h irfe
F'luting the effect of outreach programs on beh vior is difficult and reiqui-i.......
J.E. Ande son, DSF[)D E 44, C(enters For Disease Controind Prevention, A i:t, A,
30333, USA. Telephone: [phone] Fax: [phone] emrail: jeal(cpsstdL.en. ov
Mo.D. 1799
RECENT IMMIGRANT HISPANICS AND HIV/AIDS PREVENTIVE EDUCATION:
PROGRAM AND IMPACT EVALUATION THAT WORKS
Londero Miia T., Daorond, Marietta F American Nrtional Red Cross,i Washi Ito, DeC, USA.
Issue: It is a challete in tire U.S. to provide effective HIV/AIDS education to recent
Htispanic imrmigrants w th ittle or no formal education. Developing culturally and educationally -appropriate progr ams wa s the first tep in addressing this issue. Evaluation methods
must also be developed to reliably measure the impact of these programs.
Project: The Amiiercan Red Cro ass Hispani HIV/ AIDS Education PrIogram is ba sed on the
educational methodolog sy of Paul o Freire th rough which learners beconme empowee aby
actsely participatn in the ir nin process. Group dialogue and visua irages are used to
engage partiipanti.The American Red Cross also developed a culturally and educationallyapproptrate evaluation method. Eighty-three recent Hispanic immigrants in the Washingiton,
D.C. area participated in in American Red Cross Hispanic HIV/AIDS community session.
Participants were surveyed immediately before and after the session to assess the impact of
the progra rn ial questions and illustrations were used by instructors to help participants
identify the riskinress of certain behaviors and their willingness to engage in certain behaviors. Numbers under the mages were used by participants to respond to survey questions.
Results: Relabihty of the evaluation methodology was tested through inter-item reliability
testig, through qualitative analysis with instructors, and through individual item analysis. All
analyses indicated that this format was effective with this audience, including inter item reliabilities which pr oduced coefficient alphas of.67 and above. Outcome analysis was conducted by way of paired t tests. Results showed that participants exhibited a significant increase
(p C 0.001 ) from pre to post-test in knowledge of the ways that HIV cannot be transmitted
is well is knowledge about ways to prevent HIV. Participants exhibited a significant increase
(p <_ 0.001) from pre- to post test (p < 0.001) for a scale that measured the behavioral
intention (willingness) to engage in these same behaviors.
Lessons Learned: This study demonstrated that a method for pre/post-testing recent
Hispanic immigrants with ittle or no formal education can be developed and used reliably It
also demonstrated that the American Red Cross Hispanic HIV/AIDS program can increase
HIV/AIDS knowledge and encourage positive behavioral intent in this population.
Londero, tM i T -Americin Red Cross, Health and Safety Services, 8 I I IGatehouse Road,
Falls ChurcI,VA 22042; telephone: [phone]; FAX: [phone];
email: lionderom(ared cross.usa.org
Mo.D. 1800
VARIATIONS AMONG DEMOGRAPHIC GROUPS IN RESPONSE TO A COMMUNITYLEVEL INTERVENTION
johnson, Wiyne D., ofuenther-Gre yC.A., Higgins, D.L.* The AIDS Corunity
Demonstration Projects. Centers for Disease Control and Prevention, Atlanta.
Objective: To examine vari ations among African Americans, whites, and Hispanics in the
effectiveness of a multisite comnunity-level intervention promoting consistent condom use
with non-main.artners.
Methods: In July 1991, a theory based behavioral intervention promoting several HIV risk
reduction behaviors was introduced to 10 US communities. Consistent condom use with
non-main partners was targeted in 8 communities: injection drug users in two cities, sex
workers in two cities, men who have sex with men in one city street youth in one city and
two neighboorhods with high rates of sexually transmitted disease. Peer volunteers distrib
uted brochures featuring true stories of local residents who were making progress toward
consistent condom use. Repeated randomized cross-sectional samrples accessed in street
settings yielded 711 responses from these communities and matched comparison areas
from February 1991 to June 1994. Progress toward consistent condom use was measured
on the stae of change (SOC) scale from I (no intention to adopt consistent condom use)
to (consistent condom use foir more than 6 months). For each race/ethnic group, linear
regression models were used to measure changes in mean SOC among respondents across
time (the intervention and comparison trends), and the divergence between these trends
(the intervention effect).
Results: Most respondents (58%) were African American; 25% were white, and I I% were
Hispanic. Six percent were Native American, Asian, or other race, and are not included inthis comparison. Mean 50C increased significantly across time in the interveticon areas
among Afncan Americans (p<0.01), whites (p<0.0S), and Hispanics (p<0.001), but not in
the comparison areas. The rnte-vention trend was significantly greater thn the conmparison
trend in all s groups [(p<0.001) (p<0.05), and (p<0.0 I) respectively. These 3 positive
intervenrion effects were not statistically different from each other (p-0.2866), indicating
that the interve-tion was equally successful among African Americans, whites, and Hispanics.
Conclusions: Amon/ the three race/ethnic groups most heavily represented in these 8
communities, this community level intervention succeeded in promoting consistent condom
use with non-main partners.Variations by gender, age, and type of commurity, as well as
arong other race/ethnic yrouips, are also being investigated.
WD Johnson, BIRB, 3HA3 CFDC 1S F 44, I600 Cihfton Road, Aitnta, GA, 30333. USA
Teleehonei ir-104 639 8400 Fax-lf10463 9-9622 email:[email protected]
Mo.D. I80 I
DESIGNING EVALUATIONS OF COMMUNITY-LEVEL HIV PREVENTION PLANS
Westoveti B., Nowaik, Clen,'* Rupp, D., Kasprcyk,m e Miiihell, K. i5C* -Centers foi

Dissee C cntr e and Prevention, Atlanta, GA, USA; *CDC and University of Georgia, GA,
USA: iBattelle, Seattle, WA, USA,;* ** Battelle, ArlingtonVA, USA
Issue: Community bmsed heailth promotion strategies have been essential for addressing a
vnei ty of health related issues, and much has been learned. Community-based programs

Monday, July 8, 1996

also have beep I, -,d for HIV/AI[)S prevention but there are few rigorous evaluations.The cii's n- f dral, state, and local organizations is to determine the best strategies for eva ii - n- ii Ibaised HIV prevention programs.
Project: A - -,. - i.feievco f more than 75 evaluation studies was conducted to learn
what has b lly to evaluate comrnur nty health promotion 1cti ities. The prm ry purp ',,..-ert and cail al the evaluation methods used in community
level ve - fective and tfersible evaluat on designs aso were identified
Results: t he I i rt. s' revealed: I) large-scale st es that rely solely on quantitative
methods ac ofII-. C -, - to detect the true effects of rommunitry-level programs; 2) scientifc rigor ' an p r rtrni -tudy design consideration, but strnct experime tal designs have
often shown disppoint, r esults; and 3) effective evalations of communty-levei HIV intervention programs r equire: a) developing a b1ehav ora ar d/or social science basis for the
evaluat ion durigioC'ram design phase; ) docume no g delver of the ter venton; c)
usirgiresearc h Jr/s,and methods that enable evaluator to identiy the mechanisms of
behav ior change(s) nd)ing i well-defined, theory -Fbased outcome measures that capture
both community and individual-evel effects.
Lessons Learned: A sessi o the impact of cormmunity--level HIV prevention prog rams
requires multitaceted evaluation plans that use multiple methods and measures to oting-l
for alternatve hypotheses. BotI qualitative and quantitative methods need to be used.
Program exposure and partricpation are often best assessed using quantitatroe methods,
while qualitative methods enable evaluators to document the fidelity of interventions, triangulate findings, add needed context inform.ition, and erhance the peneralizablity of findings.
Glen J. Nowak, CDC/NCHSTP/DHAP 600 Clifton Road, M ilstop E-59, Atlanta, GA 30333
Telephone: [phone] Fax: [phone]
Mo.D. 1802
UTILIZATION OF NEEDLE EXCHANGE BY INJECTION DRUG USERS NEWLY
ADMITTED TO METHADONE MAINTENANCE
Calsyn Donald A.'-., Saxon A",Wells F (1I, Jackson TP I. *Veterans Affairs Puget Sound
Health Care System; University of Washington School of Medicine; IEvergreen Treatment
Services; Seattle, WA, USA.
Objective: To measure the utilization of a government sanctioned needle exchange (NE) by
injection drug users (IDU) newly admitted to methadone maintenance (MM)
Methods: One hundred ninety eight male and one hundred twenty-three female IDU con
secutively admitted to MM were assessed by structured interview for HIV risk behaviors
and needle exchan ge utilization at admission and at a 1 2 month follow up.
Results: Using the NE in the 6 months prior to admission was reported by 202 (62.6%). Of
these 125 (61.9%) visited the NE at least weekly NE utilizers did not differ from non-utiliz
ers as a function of age, race or sex. Non utilizers were more likely to be married. Most NE
utilizers (N= 133, 61. I%) reported obtaining new needles primarily from the INE, whereas
most nn utilizers (N -7, 70.2%) obtain needles primarily from i pharma cy. here wars no
difference between utilizers and non-utilizers in the nu ibe of people with whom needles
were shared. At the 12 month follow up 146 (45.5%) reported using the NE in the prior 6
months. Of these 67 (45.9%) visited at least weekly There were no demographic differences
between NE utilizers and non utilizers at follow up, however;n utihzers were involved in illegal
activities more days in the prior 30 (M 5.8 vs. 2.3). NE utuilizers and non- utilizers did not
differ on frequency of using nclean injection equnpmert and number of people with whom
equipment was shared during the 6 eorths pir-Ithe follow up assessment.
Conclusions: The mijo ty ca new admissions to MM had utilized the NE. NE utilization
decreased over time. NE utilizers and non utilizers dd,at dnrffermn injection risk behavior.
Failure to find such difference is probabl related to most non utiizer s obtaining needles
fom pharmacies. Relationship between tr eatment retentior, self reported drug, risky sexual
behavior and NE utilization wi also be examined
Donald A. CalsynVA Puget Sound Health Care System (I I6DD tP), 660 S. Columbian
Way Seattle, WA 98 08, CISA;Telephone: [phone]. Fax: [phone],
email: Calsyn,Donald Ati Seattle.VA.GOV
Mo.D. 1803
OUTLOUD:A COMMUNITY LEVEL INTERVENTION TARGETING HOMOSEXUALLY
ACTIVE MEN
Wood Robert VW., Hartfield, K*, Lucchese, KC*, Lightnern, R", ittle, E,Welch, - O.
"Seattle King County Dept of Public Health: '"*NW AIDS Foundation; I YouthCare;,(People
of Color AC inst AiD)S Network, Seattle, Washington. CSA.
Issue: In Seattle, Washington, one-third of men who have sex with men (MSM) continue to
engage in unprotected anal sex despite high levels of IV knowledge.
Project: OutLOUD is a collaborative mult agency inte vention targeting rnen who have
sex with men. Five community-based organizations and the local health department employ
outreach for specific target audiences (including African Amerncan, Asian/Pacific, and Iatino
MSM, gay youth, and nor n-gay-identifying MSM) using a Dehavoral theory-based communitylevel intervention itested bI the Centers for Diasase Control AIDS Community
Demonstraton I Prjeots. CiiOutLOUD includes: newslettrers fer-atuin first-hand role rhodel
stories of comrunty memters who have reduced the r risk of acquinng HIV; six specifically
targeted versions of the newsletter distributed ever 8 weeks; distribution of mateials
through networks of sperally trained outreach workers, peer educators and communty
merchants; speciic reinforcenent of risk reduction mesages; and access to OutLOUD stories and related rnformation via the World Wide Web. The staff and volunteer training cur
riculum includes modules on building rapport, overcon ng obections, setting outreach
goals, and "closing the srle." Community satisfaction with OutL( 31)is,ssessed via periodic
surveys, focus yroups and mcteractive Inte nt feedb of
Results: In 19)5. mc devo opod 24 vonrsnons of the neoxsletter W~e ris tinned 123 outreach workers and volunteers from 14 agencies, provi: ing more than 12,000 encounters in
parks, bars, bathhouses and adult vudeo arcades. FIoe thousand persons accessed the
Internet coponent i n its Ersi two mom if ve-misI questions suesu ih m eany non gay
ieniying M1SM are accessing OmutLOUlD our tee ier riet. Clirert satusFi 1t00n wiin
OutLOUD is hugh. Dtreach wo kes bport tht the ew Nsk Iearned in trn are
enhancing their-cot eaci and g vng ithe the ub- ty totn ri Caron sit ons ro
succStiesflnea Ito -s.

191