Most of the world’s cases of HIV infection/AIDS, including nearly 70%–80% of all HIV/AIDS-related deaths, have occurred in sub-Saharan Africa. 1 Although access to antiretroviral treatment has been severely limited in Africa, it is now expanding. Several policy leaders and investigators have raised concerns that adherence to HIV antiretroviral therapy may be difficult in settings of extreme poverty. 2,3 Others have suggested that a rapid expansion of antiretroviral treatment access in Africa may lead to widespread resistance. 4 It is widely recognized that adherence is the primary predictor of viral suppression, 5–9 progression to AIDS, and death. 10–12 As such, it is critical to address adherence to antiretroviral therapy in this dynamic environment. Although assessments of adherence to highly active antiretroviral therapy in Africa are emerging, 12–15 the validity of adherence measurement strategies developed in resource-rich settings for use in resource-limited settings has not been investigated. We evaluated the association between various adherence measures and their effect on viral load suppression in antiretroviral-naive Ugandans initiating Triomune therapy, a generic coformulation of lamivudine, stavudine, and nevirapine (CIPLA, Ltd., Mumbai, India) administered as 1 pill twice a day.
Antiretroviral-naive patients initiating Triomune therapy were recruited from the pharmacy at the Joint Clinical Research Center in Kampala, Uganda. Patients were receiving medical care from the Joint Clinical Research Center, the Mulago Hospital Infectious Disease Clinic, Nsambya Hospital, or other community health centers. The correspondence between 4 adherence measures was compared: 3-day structured self-report (number of doses reported missed/doses prescribed over the prior 3 days), 16 30-day visual analog scale (percent of pills reported missed over the last 30 days), 17 electronic medication monitoring (number of pill bottle openings registered/number of doses prescribed), 18 and unannounced monthly pill counts (1 - number pills missing between counts/number pills prescribed between counts). 18 Adherence data were collected monthly at the participant’s usual place of residence over 3 months. Participants were informed that assessments would take place between 2 and 4 weeks; however to preserve the “unannounced” nature of the pill count, the exact date of the encounter was not provided. Adherence was analyzed as a continuous variable. Mean adherence over the 3 measures was calculated for each individual. Average adherence determinations of >100% (due to ingesting more pills than prescribed, extra cap openings, or lost pills) were truncated at 100%. The population mean and 95% confidence interval were then calculated for the study population. Individual mean adherence estimates were also used to examine the association between individual measures and log HIV-1 load at 12 weeks using the Pearson correlation coefficient. Individual measure differences were tested using a paired t test. Statistical Package for the Social Sciences version 8 for Windows (SAS Institute) was used for all analyses. HIV-1 load was determined at baseline and 12 weeks using the Amplicor assay (Roche).
Between September 2002 and July 2003, 36 eligible individuals were approached to participate in the study. Thirty-four participants (94%) consented and completed adherence assessments over 12 weeks. The study sample characteristics are shown in Table 1. Of note, 53% of participants had a monthly income of less than US $50 per month (median Ugandan income, US $250 per year). Small business owners, police officers, tailors, teachers, and housemaids were the most common forms of employment.
One hundred percent of the scheduled 102 adherence assessments were completed. Median adherence for electronic medication monitoring and pill count was 99%. Median adherence for 3-day self-report and 30-day visual analog scale was 100%. Mean adherence (95% confidence interval) for each measure was as follows: 90.9% (85.4%–96.3%), electronic medication monitoring; 93.7% (88.7%–98.7), pill count; 94.4% (89.3%–95.6%), 3-day self-report; and 93.5% (88.7%–98.4%), 30-day visual analog scale. There were no significant differences between any 2 measures, and each measure was closely associated with each other (P < 0.001;Table 2).
Twenty-six patients (76.5%) had a viral load of <400 copies/mL at 12 weeks. Twelve-week viral load was significantly associated with all adherence measures (electronic medication monitoring, R = −0.34, P = 0.04; pill count, R = −0.41, P = 0.01; visual analog scale, R = −0.36, P = 0.03; and 3-day self-report, R = −0.42, P = 0.01).
Using multiple measures of adherence, our findings suggest that patients purchasing generic HIV antiretroviral therapy take, on average, 91%–94% of their prescribed therapy. This high level of adherence is consistent with the high rate of viral suppression demonstrated in this study. The high level of adherence and the high rate of viral load suppression are also consistent with findings of early reports in resource-limited settings using clinic-based pill count and self-report in African cohorts receiving both subsidized and unsubsidized highly active antiretroviral therapy. 13–15,19 This level of adherence is higher than in most studies using objective measures in resource-rich settings, where average adherence is 70%. 20 Only 1 study in a resource-rich setting found a comparable level of adherence. 17 Collectively, these data suggest that adherence to HIV antiretroviral therapy in this sample of people purchasing generic HIV antiretroviral therapy in a resource-limited setting was equal to or better than adherence previously assessed in resource-rich settings.
We found a high degree of correspondence between all measures of adherence. All adherence measures were also associated with HIV-1 load at 12 weeks. The close correlation between each of the measures as well as their correspondence with the 12-week HIV-1 load supports the validity of each of the measures.
Unlike resource-rich settings where patients overreport adherence by 15%–20%, 5,6,8,21–23 the patients we studied did not significantly overreport adherence. Whether this difference is due to differences in social desirability bias, the relative ease of accurately remembering missed doses on a regimen of 1 pill twice a day, or some other reason is unclear. Nonetheless, self-reported adherence with either a 3-day structured self-report or 30-day visual analog scale appears to accurately reflect objectively measured adherence in this setting. The close correspondence between 30-day visual analog scale and objective measures is consistent with the findings of Walsh et al 17 and Giordano et al 24 in domestic settings. Although there was little difference in the performance between the 3-day self-report and the 30-day visual analog scale, the latter is far simpler to administer. The 3-day self-report requires participants to recall their medication-taking behavior day by day for the last 3 days. The 30-day visual analog scale, however, simply asks the patient to draw a line between 0 and 100% to indicate the percentage of pills taken over the past month. The relative ease of administration of the 30-day visual analog scale over the 3-day self-report, given equivalent performance, suggests that this may be the preferred method.
There are several limitations to our study. Our study population was composed of a small number of patients purchasing highly active antiretroviral therapy. Despite many of these patients’ moderate level of poverty, their mere ability to secure funds for treatment limits the generalizability of our findings. In addition, it is possible that our intensive adherence measurements altered patients’ adherence. Although it is difficult to exclude this possibility, studies with intensive measurement in resource-rich settings have not had a significant impact on adherence. 25 Because Triomune is prescribed as 1 pill twice daily, it is possible that levels of adherence to more complicated regimens might be lower. Finally, full validation of self-reported measures in resource-limited settings will require study of patients with a wider range of adherence.
Accurate adherence assessments are critical to appropriately describe patterns of adherence and, more importantly, to understand the unique barriers to adherence. As such, valid adherence measurement is a necessary first step to fully optimize HIV treatment outcomes in resource-poor settings.
The authors acknowledge Annet Kawuma and Mary Kasango for assistance with participant recruitment, data collection, and data management.
1. UNAIDS. AIDS Epidemic Update.
Geneva: Joint United Nations Programme on HIV/AIDS (UNAIDS) and World Health Organization (WHO); December 2001.
2. Popp D, Fisher JD. First, do no harm: a call for emphasizing adherence and HIV prevention interventions in active antiretroviral therapy programs in the developing world. AIDS
3. Donnelly J. Prevention urged in AIDS fight; Natsio says funds should spend less on HIV treatment. Boston Globe.
June 7, 2001.
4. Harries AD, Nyangulu DS, Hargreaves NJ, et al. Preventing antiretroviral anarchy in sub-Saharan Africa. Lancet
5. Bangsberg DR, Hecht FM, Charlebois ED, et al. Adherence to protease inhibitors, HIV-1 viral load, and development of drug resistance in an indigent population. AIDS
6. Arnsten JH, Demas PA, Farzadegan H, et al. Antiretroviral therapy adherence and viral suppression in HIV-infected drug users: comparison of self-report and electronic monitoring. Clin Infect Dis
7. Paterson DL, Swindells S, Mohr J, et al. Adherence to protease inhibitor therapy and outcomes in patients with HIV infection. Ann Intern Med
8. McNabb J, Ross JW, Abriola K, et al. Adherence to highly active antiretroviral therapy predicts virologic outcome at an inner-city human immunodeficiency virus clinic. Clin Infect Dis
9. Gifford AL, Bormann JE, Shively MJ, et al. Predictors of self-reported adherence and plasma HIV concentrations in patients on multidrug antiretroviral regimens. [In Process Citation]. J Acquir Immune Defic Syndr
10. Bangsberg DR, Perry S, Charlebois ED, et al. Non-adherence to highly active antiretroviral therapy predicts progression to AIDS. AIDS
11. Hogg R, Heath K, Bangsberg DR, et al. Intermittent use of triple combination therapy is predictive of mortality at baseline and after one year of follow-up AIDS. AIDS
12. Garcia de Olalla P, Knobel H, Carmona A, et al. Impact of adherence and highly active antiretroviral therapy on survival in HIV-infected patients. J Acquir Immune Defic Syndr
13. Laurent C, Diakhate N, Gueye NF, et al. The Senegalese government’s highly active antiretroviral therapy initiative: an 18-month follow-up study. AIDS
14. Byakika-Tusiime J, Oyugi J, Tumwikirize W, et al. Ability to purchase and secure stable therapy are significant predictors of non-adherence to antiretroviral therapy in Kampala, Uganda. 10th Conference on Retroviruses and Opportunistic Infections; Boston; 2003.
15. Weiser S, Wolfe W, Bangsberg D, et al. Barriers to antiretroviral adherence for patients living with HIV and AIDS in Botswana. JAIDS
. 2004;(in press).
16. Chesney MA, Ickovics JR, Chambers DB, et al. Self-reported adherence to antiretroviral medications among participants in HIV clinical trials: the AACTG adherence instruments. Patient Care Committee & Adherence Working Group of the Outcomes Committee of the Adult AIDS Clinical Trials Group (AACTG). AIDS Care
17. Walsh JC, Mandalia S, Gazzard BG, et al. Responses to a 1 month self-report on adherence to antiretroviral therapy are consistent with electronic data and virological treatment outcome. AIDS
18. Bangsberg D, Hecht F, Charlebois E, et al. Comparing objective measures of adherence to HIV antiretroviral therapy: electronic medication monitors and unannounced pill counts. AIDS and Behavior.
19. Orrell C, Bangsberg D, Badri M, et al. Adherence is not a barrier to delivery of HIV antiretroviral therapy in resource-poor countries. AIDS
20. Bangsberg DR, Deeks SG. Is average adherence to HIV antiretroviral therapy enough?J Gen Intern Med
21. Liu H, Golin CE, Miller LG, et al. A comparison study of multiple measures of adherence to HIV protease inhibitors. Ann Intern Med
22. Miller LG, Hays RD. Measuring adherence to antiretroviral medications in clinical trials. HIV Clin Trials
23. Miller LG, Liu H, Hays RD, et al. How well do clinicians estimate patients’ adherence to combination antiretroviral therapy?J Gen Intern Med
24. Giordano T, Guzman D, Clark R, et al. Measuring adherence to antiretroviral therapy in a diverse population using a visual analogue scale. HIV Clin Trials
. 2004;(in press).
25. Wagner GJ, Ghosh-Dastidar B. Electronic monitoring: adherence assessment or intervention?HIV Clin Trials