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

12th World AIDS Conference

Abstracts 14162-14166 215

importance of parents' knowledge, skill, and comfort when discussing sexuality
and sexual risk may be a useful public health strategy to reduce HIV infection in
young adults.
14162 Perceptions and practices related to condoms and
awareness of AIDS among married women in Bombay,
India
Nilesh Chatterjee1, S. Shankaranarayan2, D.G. Gill3. CHPRD RAs-W904 PO
Box 20186 Univ. Texas-School Public HLTH Houston TX 77225; 2LTMMC
Bombay University Bombay; 3Univ of Maryland-Baltimore County Baltimore MD,
USA
Objectives: To examine perceptions and practices related to condoms and condom-use, and awareness of AIDS among married women, a group increasingly
at risk of HIV-infection, in Bombay, India.
Design: Cross-sectional survey.
Methods: Three hundred fifty married women were randomly selected for
face-to-face interviews from among all women accompanying patients to 3 municipal hospitals in northeast Mumbai (Bombay) from June through September
1994. The instrument used was adapted from the Phase II KABP Questionnaire
on HIV/AIDS and the Partner Relations Interview Schedule of the WHO/Global
Programme on AIDS.
Results: Although 75% (n = 266) of all 350 respondents had heard of male
condoms, only 66% (n = 236) were aware of AIDS, 38% (n = 134) knew that
condoms could prevent AIDS and 25% (n = 86) had ever-used condoms. The
women who had ever-used male condoms had significantly more years of schooling, family income, media exposure (all at p <.01 level) and knowledge about
AIDS (p <.05), and fewer living children (p <.01) and pregnancies (p <.05)
compared with those who had never used condoms. More women in both groups,
80% of ever-users and 62% of never-users, mentioned the effectiveness of male
condoms in preventing pregnancy compared with only 56% of ever-users and
37% of never-users who said condoms prevented STDs. About 81% of ever-users
said condoms were easy to use compared with 43% of never-users. Comparable
percentages, less than 20% in both groups, stated that condoms were expensive,
condom-use was against their religion, caused a man to lose potency or climbed
up the womb. Of all ever-users, 72% thought condoms were most appropriate for
used with regular partners and 67% for use with casual partners compared with
34% and 35% respectively among non-users. Only 11% (n = 37) of all women
had heard of the female condom and after explanation, 20% (n = 71) said they
would try it.
Conclusion: Although, 3 of every 4 women had heard of male condoms,
only 1 in 4 had used them. Education, family income and media exposure were
positively associated with condom-use, yet, awareness of their role in pregnancy
prevention is higher than that in STD-prevention. There are few mis-perceptions
about condoms and a comprehensive and targeted health-education program
could effectively inform more women about female condoms and male-condom
use in STD prevention. The barriers to more widespread condom-use needs
further examination.
14163 Theory based predictors of condom use intention and
behavior among women of different ethnic groups
Danuta Kasprzyk, K. Broglio, D. Montano. Battelle, 4000 NE 41 ST Street,
Seattle, WA 98105, USA
Background: The study had two objectives: 1) To identify predictors of women's
condom use intention and behavior with regular and casual partners, and 2) To
determine whether these predictors are different for Black, White and Hispanic
women.
Methods: A theory driven questionnaire was developed to measure components of an integrated behavioral theory. Intention, attitude, norms, facilitating
conditions, perceived control, susceptibility, and alternative protective strategies
were measured, with respect to condom use with regular and casual partners. A
prospective design was used; the survey was administered twice, three months
apart, to 395 women at high risk for HIV. Correlational analyses were conducted
to identify Time 1 measures that were most strongly associated with intention at
Time 1 and with condom use behavior at Time 2.
Results: Summed model construct scores (attitude, norms, facilitating conditions, and perceived control) varied in their correlations with intention and
behavior. Attitude and facilitating conditions were universally, while norms and
perceived control were differentially associated with condom use intention and
behavior across partner types and ethnicities. For all groups of women, these
relationships were weaker with casual than regular partners, and weaker with
behavior than intention; less so among Whites. Key attitude item predictors varied
across ethnic groups. One norm (best friend) predicted behavior across all groups
and partner types, with other key norms varying by group. Two facilitating conditions (availability and partner open to condom use) predicted behavior across
all groups. Perceived susceptibility and alternative protective strategies were not
predictive of intention or behavior.

Conclusions: Intervention messages should vary by ethnic group and for
different partner types, differentially targeting each model construct and key predictors based on study findings. In addition, constructs predictive of intentions are
not necessarily predictive of behavior. Interventions addressing constructs highly
associated with condom use behavior are likely to have more impact in changing
behavior than interventions focusing on constructs associated with intention.

S14164 Socio-demographic influences on condom use: rural,
chronic drug users and their main and other sexual
partners
Sherril Aversa l2, H.V. McCoy2, L. Randall3, Duane C. McBridge4. 11400 N. W
10th Aveneu Room 1109B Miami, Florida 33136; 2University of Miami School of
Medicine, Miami, FL; 3Florida International University, North Miami FL;
4University of Illinois Urbana-Champaign, Champaign IL; 5Andrews University,
Berrien Springs MI, USA
Background: Widespread alcohol and drug use and the frequent use of sex
workers contribute to a high risk profile for migrant and agricultural workers. This
study examines condom use with different partner types among a sample of rural
men and women in an agricultural community who use crack cocaine (chronic
drug users).
Methods: Interviews (n = 242) were conducted in 1997 with sexually active,
subjects. Cocaine use was verified by urine test. Stepwise logistic regression
analysis was utilized to measure the relationship between social/demographic
factors (gender, ethnicity, living arrangements, employment, income,) and cultural/psychological barriers to condom use on use of condoms. The dependent
variable was measured dichotomously as the perceived likelihood that the respondent would use a condom with various types of sexual partners - main,
casual, and commercial (sex for money or drugs).
Results: Despite reports of high levels of sexual activity, the expectation that
condoms would be used was not universal. Condom use was greater with casual
or commercial partners, compared to condom use with loved partners. Analysis
revealed that (1) for main partners, women were half as likely as men to report
using a condom; Hispanic ethnic identity also reduced the reported likelihood
of condom use; (2) for casual sexual relationships, Hispanic subjects were half
as likely to use condoms than were African Americans or whites; (3) for commercial partners, overall likelihood that a condom would be used was reduced
dramatically for Hispanics but was markedly higher among subjects who had
been employed steadily during the preceding month.
Conclusions: This study of crack cocaine users in a rural agricultural community demonstrates that ethnicity and the nature of sexual relationships play
important roles in determining whether subjects anticipate condom use, even in
the presence of multiple partners and a widely perceived epidemic of AIDS and
other sexually transmitted diseases. Cultural beliefs and values, gender norms,
and beliefs and attitudes about condoms appear to affect the likelihood that
subjects will use condoms with various partner types.
14165 HIV sexual risk reduction: An empirically derived
intervention for urban and rural crack users
Anne Bowen1, Mark Williams2, M.L.W. Williams2, H.V. McCoy3, C.B. McCoy3.
SBox 3415 Dept. of Psychology Laramie, WY 82071; 2Behavioral Research
Group NOVA Research Houston TX; 3 University of Miami Miami FL, USA
Objectives: To empirically identify an intervention to reduce sexual risks for HIV
with loved and casual partners among urban and rural not-in-treatment crack
users.
Methods: A quantitative questionnaire was developed based on 18 months of
ethnographic interviews using Stages of Change, Information-Motivation-Behavior and Theory of Reasoned Action as organizing tools. Developed scales for
psychosocial variables included self-efficacy, norms, outcomes expectancies, behavioral beliefs, partner beliefs, and specific condom beliefs. A Processes of
Change scale was developed using the Stages of Change Model, to identify
therapeutic processes that impact different levels of the psychosocial variables.
The questionnaire was administered to 594 not-in-treatment crack users. Scales
were factor analyzed. Logistic regressions were used to predict higher levels of
intention, with psychosocial variables as predictors. Linear regression was used to
predict psychosocial variables, with processes of change as predictor variables.
Results: Predictors across the two partner types were similar. Outcome expectancy was the best predictor for participants who reported that it was "unlikely"
they would use a condom vs. "extremely unlikely". Normative beliefs and sexual
refusal were the best predictors for reporting that condom use was "likely" rather
than "unlikely." Self-efficacy was the best predictor for reporting that condom
use was "extremely likely" vs. "likely". The therapy process variables that best
predicted the psychosocial variables were Social Liberation and Making Condoms Romantic (self-efficacy), Behavioral Management and Making Condoms
Romantic (outcome expectancies), Social Liberation, Behavioral and Condom
Awareness (Normative beliefs), and Condom Awareness (sexual refusal).
Conclusion: Results indicate that psychosocial variables defined by Stages
of Change, IMB, and TRA are useful for predicting increased intention to use
condoms. Therapy process variables that may be most helpful for changing
antecedent cognitions are related to making condoms romantic, within a social
context, that includes having participants use behavioral techniques such as
reinforcement and stimulus control to strengthen the changes.

S14166 Patterns of risk behaviour among street children of Delhi
Upma Sharma1, P. Majumdar2, I. Haque1, S. Chowdhury2. 1AHEAD, New Delhi;
2NGO AIDS Cell AIIMS, New Delhi, India
Issue: To study the sexual risk behavious and drug use pattern among street
children of Delhi.
Project/Research: A large number of street children are at risk of using drugs
and being infected by HIV and STDs, hence the need for this study came in