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

XIV International AIDS Conference

Abstracts WePeB5864-WePeB5868 47

degree to which individuals rated concerns relative to perceived need (NCD tl=
-2.49; p<0.05; NCD t2=-2.52;p<0.01).
Conclusions: This ongoing study shows that beliefs about HAART predict adherence. It illuminates patients' perspectives of HAART and suggests that interventions to support adherence will be more effective if they take account of patients'
beliefs about HAART as well as the practicalities of taking it. It also informs the
design of these interventions by identifying the key beliefs that should be targeted.
Presenting author: Rob Horne, Centre for Health Care Research, University
of Brighton, 1 Great Wilkins, Falmer, Brighton. BN1 9PH, United Kingdom, Tel.:
[phone], Fax: [phone], E-mail: [email protected]
WePeB5864 Twelve-week outcome of a cue-controlled paging
system to increase ART adherence
S.A. Safren', E.S. Hendriksen2, S. Boswell2, N. DeSousa2, K.H. Mayer3.
1 Fenway Community Health, MGH - Harvard Medical School, Research and
Evaluation Dept, Fenway Community Health, 7 Havi/land St, Boston, MA 02115,
United States; 2 Fenway Community Health, Boston, United States; 3Fenway
Community Health and Brown University Boston, United States
Background: To help with one of the most frequent reasons for non-adherence,
"forgetting", the present study sought to test the feasibility and efficacy of an
internet-based paging system (MediMom) to increase and maintain adherence
to ART over 12 weeks.
Methods: After a two-week monitoring period with an electronic pill cap, seventyone participants with HIV (64 men and 18 women) with less than 90% adherence
were randomized to continue monitoring or to receive a pager, and followed for
two and twelve weeks.
Results: Among the participants who qualified after the monitoring period, adherence assessed by pill cap at baseline was very low: 56%. Sixty participants
arrived at the two-week assessment and 44 at the 12-week assessment. The
group who received the paging system had greater adherence than those who
continued to monitor adherence only as evidenced by a significant interaction of
group assignment and assessment point for both the two-week (F(1,58) = 8.24;
70% and 56% adherence) and twelve week assessments (F(1,43) = 5.79, p <
0.05; 64% and 52% adherence). However, even after improvement, adherence
was less than optimal.
Conclusions. While the provision of a reminder system helped improve adherence, it is likely that a more intensive intervention would be required for patients
with moderate to severe adherence problems.
Presenting author: Steven Safren, Research and Evaluation Dept, Fenway
Community Health, 7 Haviland St, Boston, MA 02115, United States, Tel.: +1
617 927 5057, Fax: [phone], E-mail: [email protected]
WePeB5865 A comparison of resistance propensity between
NNRTIs and PIs (protease inhibitors) in VA New
Jersey Health Care System
R.H.K. Eng, L.L. Dever, M. Sasanpour, J. Meltzer. VAMC, VAMC-/nfect Dis, 385
Tremont Avenue, Rm 5-197, East Orange, NJ, United States
Background:Both NNRTI and PI combination therapies are popular and effective choices of therapy From past work on resistance, we expect the threshold
for resistance to NNRTIs to be lower than for the PIs. This may be overcome by
the easier dosing schedule and tolerance to the NNRTIs. The VANJHCS has a
record of all the HIV prescriptions dispensed and in the last 12 months we have
a record of genotype results for the patients failing their therapy with viral RNA of
>2000 c/mL. We examined the above issue in a retrospective manner.
Methods: The prescriptions were retrieved from the "Prescription" file in the VA's
computer system -VISTA. The genotype results were retrieved from the "Lab
Data" file. The extracted results were analyzed in both Microsoft ACCESS and
EXCEL. Statistics were done using chi-square tables and t-test.
Results: The medications used in the NNRTI group include Efaviranz (>90%),
Nevirapine, and delavirdine. The medications used in the PI group include
Saquinavir, indinavir, nelfinavir, and amprenavir. Ritonavir was used but only as a
booster. The results are as follows:
NNRTIs PIs
#pts 471 716
#Prescriptions 3437 9400
Average months on treatment 7 14
Resistance/1000 pt-Rx-days 0.42 0.20 p<0.05
Conclusions: The results from our institution would indicate that the expected
greater adherence to medication by the NNRTI regimens did not overcome the
lower threshold for resistance. The end result in our population is that NNRTI
combinations have a higher propensity (twice the risk) for resistance than the PI
combinations.

Presenting author: Robert Eng, VAMC-Infect Dis, 385 Tremont Avenue, Rm 5 -197, East Orange, NJ, United States, Tel.: [phone], Fax: [phone]
7093, E-mail: [email protected]

WePeB5866 Determinants of adherence to antiretroviral
therapy in a developing country
S.B. May, G. Cardoso, P.F. Barroso. Universirty Hospital Clementino Fraga
Fi/ho/Federal University of Rio de Janeiro, Rio de Janeiro, Brazil
Background: Long-term adherence to antiretrovirals is critical for a sustained
response to HIV therapy. In Brazil all HIV-infected subjects have free access to
antiretroviral therapy (ART). As a result of this policy, a reduction in mortality and
hospitalizations rates in recent years have been observed. Around 100.000 patients are receiving ARV drugs, and the expenditures in purchasing antiretroviral
therapy in 2000, were estimated to be US$ 332 millions. The purpose of this
study was to determine the adherence rates and its determinants in a cohort of
HIV infected subjects in a large academic HIV reference center in Rio de Janeiro,
Brazil.
Methods: We conducted a cross-sectional study of a randomly selected sample
among 1,300 patients currently receiving ART. Adherence to therapy was measured by self-report. Trained interviewers used a structured questionnaire. Percent adherence to ARV was defined as the number of pills taken divided by the
total number of pill prescribed in the last three days. Adherence was dichotomized
into non-adherence (80% or less) and adherence (>80%).
Results: We report results in 97 patients with a median of 60 months of known
HIV-seropositivity. Fifty-eight (60%) were of male gender and median age was 40
years. Mean self-reported therapy was 90% (95% CI: 0.84-0.96) and 84 (87%) of
patients reported more than 80% adherence to therapy None of the demographic
or clinic variables were associated with adherence to ART.
Conclusions: A high rate of adherence to ART may be achievable in a resource
poor setting. Efforts to increase access to ART may diminish the impact of the
HIV epidemic in developing countries.
Presenting author: Silvia Beatriz May, Av Epitacio Pessoa 2530/ 803 Lagoa, Rio
de Janeiro/ RJ, CEP- 22-471-001, Brazil, Tel.: +55+21-2522-9316, Fax: +55+ 21
- 2590-5422, E-mail: [email protected]
WePeB5867 Efficacy of a counseling intervention on
adherence to HAART: results of a French
prospective controlled study
L. Bentz', C. Pradier', M. Morin2, B. Spire3, M. Rebillon4, M. Souville2,
J.G. Fuzibet', A. Pescel, P. Dellamonical, C. Tourefte-Turgis5, J.P. Moatti3.
INice University Hospital, CISIH, Hdpital de IArchet, BP 3079 06202, Nice
cedex 3, France; 2University of the Mediterranean, INSERM U 379, Marseilles,
France; 31NSERM U 379, Marseilles, France; 4Comment Dire, Paris, France;
5Rouen University, Rouen, France
Background: To evaluate the efficacy of a counseling intervention on adherence
and viral load (VL), and identify psycho-social factors contributing to changes in
adherence to HAART among HIV-infected patients.
Methods: 244 patients, receiving HAART since at least 1 month, were included
in year 2000 in a prospective, controlled, randomised trial comparing a group
who received a counseling intervention (IG), in addition to ordinary clinical followup (n=123) versus a control (CG) (n=121). Patients in the IG received 3 interactive counseling sessions focused on HAART regimens and on cognitive,
emotional, social and behavioural determinants known to affect adherence by
specially trained nurses. The main outcome measures were proportions of patients achieving 100% adherence at 6 months follow-up (M6), evolution in VL
between inclusion (MO) and M6, and factors of adherence analysed through a
self-administered questionnaire.
Results: The proportion of 100% adherent patients was similar in both groups at
MO (58% vs 63%, p=0.59) but became higher in the IG (75% vs 61%, p=0.04),
while mean differences in VL decreased significantly in the IG vs the CG at M6
(p<0.01). The analysis of the determinants affecting adherence found that patients received more information on HIV regimen in the IG (p=0.04), developed
specific skills in their daily routines, were more likely to use pill boxes (p=0.05).The
IG followed dietary regimen (p=0.06) and got more involved in health care holistic approaches (complementary medicine services, relaxation, massages,...)
(p=0.05).No difference was found between IG and CG regarding scores of depression, use of family networks, negative emotional effects of adverse events
such as lipodystrophies.
Conclusions: The study brings evidence of the efficacy of a structured counseling intervention to enhance adherence to HAART, and illustrates that a counseling intervention has an impact on the abilities of HIV people to develop self-care
management skills.
Presenting author: Laurence Bentz, CISIH, H6pital de I'Archet, BP 3079 06202,
Nice cedex 3, France, Tel.: [phone], Fax: [phone], E-mail:
bentz.I @chu-nice.fr
WePeB5868 Sustained antiretroviral adherence following a
short-term directly observed therapy intervention:

The S.T.A.R.T. Program
J. Myers, C. Farthing, J.C. Ricaurte, P.T. Korthuis. AIDS Healthcare Foundation,
Los Angeles, CA, United States
Background: Management of HIV-infected persons non-adherent to antiretroviral
therapy (ARV) is a challenging aspect of HIV care. Many have barriers rendering
them refractory to usual adherence tools. We developed a directly observed ther