[690] SELF-REPORTED ADHERENCE MEASURES ARE FEASIBLE AND VALID COMPARED TO MULTIPLE OBJECTIVE MEASURES IN KAMPALA, UGANDA

2nd International AIDS Society Conference on HIV Pathogenesis and Treatment

Paris, France - July 13 - 16, 2003

[TITLE:] SELF-REPORTED ADHERENCE MEASURES ARE FEASIBLE AND VALID COMPARED TO MULTIPLE OBJECTIVE MEASURES IN KAMPALA, UGANDA

[AUTHOR(S):] JH Oyugi^1,2, J Byakika-Tusiime³, P Mugyenyi^2,4, R Mugerwa^2,3, D Guzman¹, E Charlebois¹ and DR Bangsberg^1,2
¹University of California-San Francisco, San Francisco, USA; ²Academic Alliance for AIDS Care and Prevention in Africa, Kampala, Uganda; ³Makerere University, Kampala, Uganda; and ⁴Joint Clinical Research Center, Kampala, Uganda

IAS Conf HIV Pathog Treat 2003 Jul 13-16;2nd: Abstract No. 690
Antiviral Therapy 2003; 8(Suppl. 1):S374-S375

[ABSTRACT:] Background: Despite increasing access to antiretroviral therapy (ART) in sub-Saharan Africa, evaluations of the feasibility and validity of adherence measurements are lacking.

Methods: ART naïve patients initiating Triomune® (generic D4T/3TC/NVP) were recruited from the Joint Clinical Research Center in Kampala, Uganda. Adherence was assessed monthly using: structured 3-day self-report, visual analog scale (VAS) of % doses taken in 30 days, electronic medication monitors (MEMS; with and without adjustment for extra openings or pocket doses) and unannounced pill counts (UPC) conducted in participants' homes (Bangsberg AIDS-Beh. 2000). Mean adherence estimates were calculated per individual to examine the association between individual measures (Pearson's R). Individual measure differences were tested using a paired T-test.

Results: 28 patients were followed for 1-6 months (mean 3.1). Patient characteristics: 75% female, 37 years (mean), pre-treatment CD4 118 (mean), and pre-treatment log VL 5.2 (mean). Mean adherence: 88% (adjusted MEMS), 95% (unadjusted MEMS), 90% (UPC), 91% (3- day self-report), and 87% (30-day visual analogue scale). There were no significant differences between any two measures. All measures were highly correlated with each other (R=0.56-0.95). Each measure had the following association with UPC: adjusted MEMS R=0.70 P<0.0001; unadjusted MEMS R=0.73 P<.0001; 3-day self-report R=0.76 P<0.0001; VAS R=0.95 PP<0.0001.

Conclusions: Measuring adherence is feasible in a resource-limited setting. The close association amongst all measures supports their validity. Unlike resource-rich settings, patients did not significantly over report adherence. This may be due, in part, to the ceiling effect of high adherence levels seen in this population.

030714
690