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

242 Abstracts 14297-14301

12th World AIDS Conference

Results: The BSS covered these sentinel groups in both sites: female
sex workers (classified as establishment-based and freelancers - streetwalkers/brothel-based), clients of female sex workers, and men who have sex with
other men. For female sex workers, BSS should identify sub-groups, and cover all
establishments and cruising areas which need exhaustive inventory and mapping
as data collection sites for future surveys. For female sex workers, BSS should
focus mainly on risk reduction practices such as condom use and condom negotiation, and other risk factors such as drug use. For the male sentinel groups,
the inclusion criteria needs to be reviewed, and allow for a bigger sample size to
establish prevalence of risky behaviors. For all sentinel groups, STD incidence
and health seeking behavior should be retained in the BSS. Some of the BSS
data collected have no direct relevance to the content and focus of STD/AIDS
prevention programs.
Lessons learned: To be an effective tool, the BSS needs to focus on generating relevant information which will enable HIV/AIDS prevention programs to
monitor effects and gaps of interventions for specific population groups.
14297 Issues in studying high risk populations in diverse
locations: Puerto Ricans in New York and Puerto Rico
Sherry Deren1, R. Robles2, M. Beardsley1, J. Andia1, H. Colon2,
D. Oliver-Velez1, A. Finlinson2. 1NDRI 2 Worl Trade Center 16th Fl. New York,
NY, USA; 2Universidad Central Del Caribe/POB 60327 Bayamon, Puerto Rico
Objectives: To assess factors influencing differences in risk behaviors between
Puerto Rican drug users (who are at high risk of HIV) in Puerto Rico and New
York.
Design: Prospective dual-site study.
Methods: Qualitative methods (mapping, field observations, focus groups, indepth interviews) and quantitative methods (structured interviews) are being used
in Bayamon PR and East Harlem NY to assess differences in influences on risk
behaviors for injection drug users (IDUs) and crack smokers (CS).
Results: Analysis of prior survey data indicated significant differences in risk behaviors between the two sites: PR IDUs in PR inject more frequently than those in
NY (188 vs 108 times/month); PR CS in NY use crack more frequently (194 vs 127
times/month). Injection-related risk behaviors were generally higher in PR (sharing
needles, 40% vs 23%). Some sex risks were higher in NY (trading drugs/money
for sex), others were higher in PR (unprotected sex). (All comparisons significant
at p <.001). Qualitative methods, utilized to understand site differences, indicate
that environmental and social factors can influence risk behaviors. These include
access to: sterile syringes (needle exchange and pharmacy availability), drug
treatment (differences in methadone treatment availability), and to harm reduction materials (bleach kits, condoms), as well as stability of shooting galleries
(in response to police crack-downs), composition of family support and networks
(as influences on homelessness and relationship between sex and injection risk
networks) and availability of health care and other medical/social services.
Conclusion: Multi-site studies, especially those involving groups of similar cultures, benefit from the use of both qualitative and quantitative research methods
to identify differential influences on risk behaviors. Assessment of social, cultural,
and environmental influences on risk behaviors, as well as individual factors, is
helpful in developing locally optimal interventions.
14298 Recently diagnosed HIV-infected patients: Outcome of
partner notification - European Concerted Action Study
Anne-Marie Worm1, E. Smith1, P. Holmstr6m2, S.L. Valle3, S. Gore4. 1Statens
Serum Institut, Copenhagen, Denmark, 2National Public Health Institute,
Helsinki; 3Helskinki University, Helskinki, Finland; 4MRC Biostatistics Unit,
Cambrigde, England
Objectives: To describe practices and outcome of HIV partner notification (PN)
in Belgium, Denmark, Finland, Greece, Norway, Scotland and Sweden.
Design: Prospective study with standardized questionnaires.
Methods: Each country defined a setting from which all newly HIV diagnosed
index patients (iPs) were recruited within a period of up to twelve months. The
PN process was reported after a nine months follow-up period.
Results: A total of 694 iPs (517 males and 177 females) were recruited. Further information about the PN process was obtained in 397/592 (67%) of the iPs
from six countries (England and Sweden excepted). Attention to a few keydates
especially date of last negative HIV test would allow PN period to be defined for
more than half of the iPs. A major subset of 356 iPs had sexual transmission
as most likely mode of HIV infection, and reported a total of 200 current and
508 non-current partners. PN was attempted for 83% of current and 24% of
non-current partners. PN problems of a general nature were more likely if iP was
non-white and from a non-Nordic country, PN problems due to uncontactable
partners centred on iPs from non-Nordic countries and PN problems due to no
re-attending on older iPs. PN outcome was similar whether the iP was heterosexually or homosexually infected. Of the 96 current and 34 non-current partners
for whom pre-PN HIV status was negative or unknown 29% (28/96 and 10/34
respectively) were found HIV positive on post-PN testing. The major difference

lay in the informative loss-rate in partners not already known to be HIV positive,
being 28% for current and 68% for non-current partners.
Conclusion: PN for HIV infection can result in disclosing HIV infection in
partners previously unaware or at least undiagnosed. The study highlights the difficulties encountered in the identification, notification and follow-up of non-current
sexual partners.

S14299 Method in the madness: The efficacy of narrative-based
interviewing for sexuality research
Stephen Samis', K. Whyte. AIDS Vancouver Island, 1704-1035 Belmont
Avenue, Victoria, BC, Canada
Issue: The effectiveness of narrative research methodology for issues relating to
sexual identity, sexual relationships and safer sex amongst men who have sex
with men (MSM)
Project: Sexuality is complex, emotional and at the heart of lived experience.
Research on sexuality must capture this complexity. Narrative (unstructured and
conversational) interviewing facilitates this process. A federally funded, community-based research project at AIDS Vancouver Island in Victoria, British Columbia,
Canada used the narrative research method in 80 interviews and 2 focus groups
with MSM between July 1996 and September 1997.
Results: The narrative-based research method produced research "data" that
is a tangled web of deep meaning and insight. While the method presented
analytical challenges, it enabled both the researchers and participants to come to
new understandings about issues related to sexual identity, sexual relationships
and safer sex.
Lessons Learned: Narrative-based methodology challenges the traditional
power dynamic between researcher and participant and facilitates a more "democratic" research process. It facilitates an exploration of issues that are deeply
emotional. Feelings, beliefs and experiences associated with safer sex often have
little to do with condoms and a lot to do with life history and experience. There is
a need for more of this research on sexuality and HIV/AIDS-related issues.
14300 Methodological issues in reducing bias in assessing HIV
related risk behaviour
Joseph Tak Fai Lau1, Joe Thomas2. 1Centre for Clinical Trials and
Epidemiological Research, Flat 7B, 7/F, Block B, Staff Quaters, Prince of Wales
Hospital, Shatin; 2The Chinese University of Hong Kong, Hong Kong, China
Objectives: To assess the efficacy of asking sensitive AIDS related questions
by an interactive call and key-in telephone method using mobile phones in
comparison to two other traditional methods.
Method: Male adults (18-60) were interviewed on the same questions at the
border area on their return trip from Southern China using the new method (Ml;
n = 1240) and two other traditional methods, self-administered (M2; n = 528) and
postal survey method (M3; n = 549). Structured questions on risk behaviours
including commercial sexual networking, use of condom and history of STD were
asked. Anonymity is maximised by the new method as the respondents listened
to the prerecorded questions from the mobile phone and key-in their response
confidentially.
Results: The background and the KAPB data of the three groups do not differ.
A high percentage of respondents reported high-risk behaviour. Overall 11% of
the respondents reported to have sex with a CSW during the recent trip (M1 =
11.3%, M2 = 9.2%, M3 = 14.6%). About 10% of the respondents reported to have
sex with a commercial sex worker without condom. (M1 = 10.3%, M2 = 6.4%, M3
= 14.4%). Method 1 & 3 were less subjected to presentation bias as compared
to M 2, for instance, the percentages reporting STD history were M1 = 20%, M2
= 6.5%, M3 = 18.2%. However, M3 had the lowest response rate. Item response
rates were also compared for three methods.
Conclusion: Presenting method (M1) is a less intrusive, yields a reasonable
response rate and less likely to subjected to presentation bias. The traditional
self-administered method suffered from the possibility of a high level of presentation bias while the postal survey method has too lower response rate. The
innovative method presented in this study should be further tested and refined in
other contexts for its potential applicability.
14301 Developing an appropriate methodology to study the
sexual behaviour of a rural population in India
Elizabeth Kurian. Operations research Group, Madhu Hans, 263 DR Ab Road,
Prabhadevi, Bombay, India
Issue: As the strategies that address HIV/AIDS have now shifted focus on the
general population from the high-risk behaviour groups, it is imperative to formulate an appropriate research methodology to understand the sexual behaviour of
the general population in rural areas.
Project: Given that behaviour change is the most effective means of curtailing the spread of HIV/AIDS, a qualitative research approach was devised
to understand the sexual behaviour of married women in rural Gujarat. After
an exhaustive household listing and community mapping exercise in 8 villages
around Baroda city, 69 women reporting reproductive health problems (RHPs)
were purposively selected for repeated (4-6) indepth interviews. The research
was concentrated on women reporting RHPs in order to give a strong rationale for
inquiry into sexual experiences. In all, 350 interviews were conducted by a team
of 5 experienced qualitative research professionals, each assisted by a research
assistant. Besides, focus group discussions and key informant interviews were

also conducted.
Results: Initially, due to the sensitive nature of the issues discussed, considerable animosity was expressed by certain sections of the community. This also
lead to a number of women "dropping out" from the study and hence, additional
women had to be selected via the "snowball" technique. However, gradually, over
repeated meetings, a rapport was established not only with the repondent women