Barcelona 2013
Menu
AIDSImpact.com
Conference Details
Authors
International Committee
Plenary Speakers
Presenting Speakers
Programme
Sessions
Scientific Committee
Acknowledgements
| Abstract #580 - Theory and Overview | ||||
|
Session: 14.5: Theory and Overview (Parallel) on Monday @ 14.30-16.00 in Raval Chaired by Richard Harding, Sheana Bull Authors: Presenting Author: Mr. David Finitsis - University of Connecticut/Center for Health, Intervention, and Prevention, United States |
||||
| Additional Authors: | ||||
| Aim: The continued success of antiretroviral therapy (ART) to reduce HIV/AIDS-related morbidity and mortality hinges upon the behavior of medication adherence. In spite of the risks and promises involved, many do not optimally adhere. Providers and researchers alike require effective means to monitor adherence. Visual analogue scales (VASs) appear to offer a way to measure medication adherence that is economical, fast, and simple to deploy. In the last 10 years, researchers have examined the use of VAS to measure ART adherence, but this literature appears mixed, particularly in recent years. The purpose of this paper is to provide a systematic quantitative review of studies that test the concordance of the VAS with other measures of medication adherence. | ||||
| Method / Issue: Searches were conducted using multiple electronic databases. The same combinations of Boolean search terms were run in all searches. Studies were included if they (1) reported on medication adherence in an HIV sample using VAS and (2) compared VAS to at least one other measure, biomarker or clinical outcome of medication adherence. Studies were excluded if data did not include a statistical test of VAS scores in comparison with at least one other adherence or outcome measure. Studies were coded for participant and design characteristics. Effect sizes (ESs)were calculated using correlation coefficients and were transformed to Fisher's Zr for the purpose of meta-analysis. Inverse variance weighted ESs were used to calculate mean effect sizes under random effects assumptions. | ||||
| Results / Comments: Database searches yielded 171 articles. After screening for the presence of inclusion/exclusion criteria, a final set of 14 studies remain that examine the use of VAS to measure ART adherence in HIV samples. Studies compared VAS with other self-report measures, pill count, electronic drug monitoring, pharmacy data, viral load, CD4 count, and plasma or urinary drug levels. Included studies report on research conducted in 8 countries: 38% of participants (N=2007) were living in East or Southeast Asia (n=765) and 27% (n=536) were from Sub-Saharan Africa. The majority of the sample was male (54.5%) and the mean age was 37 years. The majority of participants met poverty guidelines for their respective countries. The overall mean ES indicated a large strength association between VAS and other measures of adherence behavior (ES=.5235; 95%CI: .3308, .7163; p<.0001). Of studies reporting viral load data (k=5) there was an aggregate medium strength association between VAS scores and viral load (ES= -.3855; 95%CI: -.6237, -.1472; p=.0015). Both mean ESs upheld the assumption of homogeneity as assessed with Cochrane's Q statistic (p >.9). | ||||
| Discussion: Visual Analogue Scales have sufficient concordance with other self-report and objective measures of adherence to recommend their continued use. Moreover, these measures are capable of predicting viral load in HIV samples. Deceptively simple in appearance, these measures can be confusing to patients unfamiliar with them; administrators must provide adequate instruction to promote accurate results. Normalizing the experience of missed/late doses for the respondent also appears to increase the accuracy of the VAS. These psychological, social, and educational components may account for a proportion of the variance observed in the literature. | ||||
| Go Back | ||||