Abstract Book Vol. 2 [International Conference on AIDS (14th: 2002: Barcelona, Spain)]

XIV International AIDS Conference

Abstracts WePeB5847-WePeB5850 43

Results: A total of 195 patients participated in the study, 56.9% reported >=
95% adherence in the previous two days. HIV-1 RNA levels was <500 copies/ml
in 67.5% of the adherent group. In univariate analysis, the odds of adherence increased with self-efficacy to take medications as prescribed (OR=3.50
C195%1.90-6.55), decreased with perception of negative affect and physical concerns (OR=0.71, C195% 0.53-0.95), was lower for taking antiretroviral medications >4 times a day (OR=0,44, C195% 0.20-0.94) and greater for 8 years of education (OR=2.28, C195% 1.12-4.66). Outcome expectations of treatment, stages
of HIV/AIDS (CDC clinical categories A, B,C), time in treatment, age, gender and
income were not associated with adherence. Perception of adverse effects was
associated with adherence (OR=0.57, C195% 0.31-1.03), but the statistical evidence was weak. In multivariate analysis, self-efficacy (OR=3.33, C195% 1.69 -6.56) and >4 times a day (OR= 0.34, C195%0.14-0.80) were independently associated with adherence.
Conclusion: Self-efficacy was the most important predictor of adherence, followed by number of times per day antiretroviral medication. Among sociodemographic and clinical, only years of education was associated with adherence
Presenting author: Cezar Arthur Tavares Pinheiro, parque Dom Antonio Zattera
379/201 Pelotas RS, Brazil, Tel.: [phone], Fax: [phone], E-mail:
[email protected]
WePeB5847 HIV medication adherence assessment: a novel
i ilow literacy self-report tool compared to electronic
Medication Event Monitoring System (MEMS) data
L.M. Kunches, E. Coakley, K. Neill, D. Marquis, D. Dean, J. DeCristofaro,
L. Ruhlmann, J. Musolino. John Snow, Inc., John Snow, Inc., 44 Farnsworth
Street, Boston, MA 02210, United States
Background: Because strict adherence to antiretroviral regimens is necessary
for maximizing clinical response, current attention is focused on approaches to
adherence measurement. Client self-report instruments and clinician interviews
are most widely used, but measurement complexity and recall or reporting biases
pose limitations to usual methods. We designed a simple adherence measurement tool (AMT) to provide a mechanism for patients with limited reading skills to
answer a written self-report.
Methods: To evaluate the consistency between patient responses using the AMT
(for recall of dosing over the previous 2 days) with data captured on MEMS caps
(applied to the most frequently taken drug), a community sample of volunteers on
stable multidrug HIV regimens for at least 4 weeks was enrolled. Over a 3mo. followup period, AMT's were completed at 3 scheduled and 5 random unscheduled
times, capturing 16 days of medication taking. Dose events and times for these
days were compared for the two methods.
Results: Participants (n=52) were 75% male, 60% white, with mean age 44 yrs.
One quarter had not attended college including 4 (8%) who had completed <12
yrs education. At baseline, 56% thought they were excellent at taking their HIV
medications; 38% rated themselves good and 6% fair. AMT responses and MEMS
openings were significantly correlated (p<0.001 Pearson) for random, scheduled
and total days. Reporting accuracy of dosing times (~1 hr) was only 48% overall.
Contrary to expectations, accuracy of reporting on scheduled visits (45%) was
not better than for random days (50%). Details of temporal patterns and analyses
of missed doses will be presented.
Conclusions: In this real world sample, dosing patterns of HIV medications are
inconsistent for many patients. Assuming MEMS caps accurately record dosing
events, the majority of patients over-estimate their adherence using a simple selfadministered tool.
Presenting author: Laureen Kunches, John Snow, Inc., 44 Farnsworth Street,
Boston, MA 02210, United States, Tel.: [phone], Fax: [phone], Email: [email protected]
WePeB5848 Adherence to antiretroviral therapy in Kampala,
Uganda
C. Kityo', M. Rabkin2, D. Atwine', G. Mulindwa', A. Kebba', G. Kabuye',
P. Mugyenyi'. 'Joint Clinical Research Centre, Kampala, Uganda; 2 Columbia
University 622 West 168th street, Vanderbilt Clinic 2-205, New York, New York,
10032, United States
Background: While nonadherence to antiretroviral (ARV) medication regimens is
known to be a problem in resource-rich settings, little is known about adherence
to ARVs in resource-limited settings.
Methods: The Joint Clinical Research Centre (JCRC) in Kampala, Uganda has
long been a regional center of excellence in HIV/AIDS care, and is among the
oldest and largest providers of ARVs in sub-Saharan Africa. A retrospective chart
review was performed on a convenience sample of 577 patients enrolled at JCRC
between January 1998 and June 2001. At each follow-up visit, adherence with
ARVs was assessed via patient self-report; patients were then characterized as
adherent or nonadherent.

Results: Of the 577 patients registered at the JCRC during this period, 39 percent returned for a second visit. Of those patients who returned at least once,
56 percent were classified as "adherent" at every follow-up visit. 66 percent were
classified as "adherent" on 80 percent or more of their return visits.
Conclusion: Loss to follow-up and nonadherence to antiretroviral therapy are
significant barriers to care in Uganda, as they are in resource-rich settings. Interventions to increase adherence are underway at the JCRC.

Presenting author: Miriam Rabkin, 622 West 168th street, Vanderbilt Clinic 2 -205, New York, New York, 10032, United States, Tel.: [phone], Fax: [phone], E-mail: [email protected]
WePeB5849 Social support and social networks: The challenge
for adherence to HIV/AIDS antiviral treatment in a
sample of patients in Puerto Rico
I. Anduiar-Bello', J. Toro-Alfonso1, R. Amico2, J.D. Fisher2. 1University of
Puerto Rico, PO Box 23174, San Juan, Puerto Rico; 2University of Connecticut,
Storrs, United States
Background: Treatment adherence has become a major task for public health
in HIV antiviral treatment. The possibility of resistance and re-appearance of opportunistic infections is a common threat for people living with HIV/AIDS. Access
to treatment, attitudes towards adherence, family support, behavioral skills, and
motivation has been identified as important variables in adherence. Social networks and social supports might represent a key factor for adherence. There is
not much research done on social support and adherence.
Method: We interviewed 200 participants older than 18 years, living with
HIV/AIDS in Puerto Rico. Participants were males and females receiving treatment at community and Health Department Clinics in Puerto Rico. This was a
structured two-hour interview done by trained interviewers.
Results: This is the first attempt in Puerto Rico to formally study treatment adherence in a sample of people living with HIV/AIDS and we participants reported
high levels of adherence (70%) and high levels (77%) of social support. The sample was composed of 78% people younger than 45 years; 67% were males, 71%
identified as heterosexuals, 81% reported an income of less than US $10,000
per year, 35% had been diagnosed with HIV for more than 10 years, and 25%
reported undetectable levels of viral load. More than half of the sample (52%)
reported a neutral attitude towards adherence, while 32% informed a positive attitude. Sixty-nine percent (69%) informed high levels of behavioral intentions for
adherence.
Conclusions: The sample reported high levels of adherence and high levels of
social support. Social support was identified as family and friends. Difference on
level of adherence and level of social support was found among participants by
gender and mode of exposure. It seems that there is no relation between adherence and social support among these participants and, there might be more
influence of their intentions to adhere to treatment.
Presenting author: IvAn Andtjar-Bello, PO Box 23174, San Juan, Puerto Rico,
Tel.: [phone] x-7883, Fax: [phone], E-mail: [email protected]
WePeB5850 Adherence to HAART, HIV-1 RNA level, and
genotypic resistance among injection drug users
in Baltimore, Maryland, USA
M.K. Lin', A. Sethi', A.W. Wu2, S.A. Strathdeel, D.D. Celentanol, D. Vlahov3.
' Johns Hopkins School of Public Health, 30 Dulaney Hills Court, Cockeysville,
Maryland, 21030, United States; 2Johns Hopkins Schools of Public Health and
Medicine, Baltimore, Maryland, United States; 3New York Academy of Medicine,
New York, New York, United States
Objective: To examine the association between adherence to HAART and treatment outcome, as measured by viral load (VL) and drug resistance, among HIVinfected injection drug users (IDUs).
Methods: From a community cohort study, HIV-infected IDUs completed semiannual visits that included self-report of medications and adherence, and
venipuncture for VL and genotypic resistance testing. HAART is defined using
IAS guidelines, and eligibility is based on CD4 < 500/mm3.
Results: There were 281 person-visits between 7/99 and 12/00. During 86% of
these visits, subjects were HAART-eligible. Among these HAART-eligible subjects,
44% were on HAART, 34% were on non-HAART, and 22% received no therapy. In
the study cohort, 93% were African-American, 71% were male, 44% were actively
injecting, and the median age was 44.
Of the subjects on HAART, 45% were completely adherent. High adherence was
associated (p<0.05) with age <40 years, having access to health care, and nonuse of alcohol, cocaine, speedball, or heroin. There was no association with having a single health care provider or methadone treatment. Those completely adherent to HAART had lower VL (p=0.02) than those who did not.
Of the person-visits, 54% were associated with no clinically significant genotypic
resistance with 15% wild type strain present. In 31%, 17%, and 8% of personvisits, subjects were resistant to at least one NRTI, 3 NNRTIs, and at least one PI,
respectively. The most frequent reverse transcriptase mutations were at codons
184, 103, and 108; the most frequent protease mutations were at codons 63, 36,
and 71. Resistance was not associated with duration of therapy.
Conclusions: HAART utilization for IDUs has increased since earlier reports,
but remains non-universal. Self-reported adherence is associated with a lower
VL and more stable lifestyle. The modest levels of drug resistance supports the

availability of ART options in this HIV-infected IDU population.
Presenting author: Maria Lin, 30 Dulaney Hills Court, Cockeysville, Maryland,
21030, United States, Tel.: [phone], Fax: [phone], E-mail: mlin
@jhsph.edu