Share this article on:

Patterns of Individual and Population-Level Adherence to Antiretroviral Therapy and Risk Factors for Poor Adherence in the First Year of the DART Trial in Uganda and Zimbabwe

Muyingo, Sylvia K MSc*; Walker, A Sarah PhD; Reid, Andy MD; Munderi, Paula MD*; Gibb, Diana M MD; Ssali, Francis MD§; Levin, Jonathan PhD*; Katabira, Elly MD; Gilks, Charlie PhD¶#; Todd, Jim MSc*the DART Trial Team

JAIDS Journal of Acquired Immune Deficiency Syndromes: August 1st, 2008 - Volume 48 - Issue 4 - p 468-475
doi: 10.1097/QAI.0b013e31817dc3fd
Epidemiology and Social Science

Background: Good adherence is essential for successful antiretroviral therapy (ART) provision, but simple measures have rarely been validated in Africa.

Methods: This was an observational analysis of an open multicenter randomized HIV/AIDS management trial in Uganda and Zimbabwe. At 4-weekly clinic visits, ART drugs were provided and adherence measured through pill usage and questionnaire. Viral load response was assessed in a subset of patients. Drug possession ratio (percentage of drugs taken between visits) defined complete (100%) and good (≥95%) adherence.

Results: In 2957 patients, 90% had pill counts at every visit. Good adherence increased from 87%, 4 weeks after ART initiation, to 94% at 48 weeks, but only 1454 (49%) patients achieved good adherence at every visit in the first year. Complete adherence was associated with 0.32 greater reduction in log10 viral load (95% confidence interval 0.05, 0.60 P = 0.02) and was independently associated with higher baseline CD4 count, starting ART later in the trial, reporting a single regular sexual partner, clinical center, and time on ART.

Conclusions: Population level adherence improved over time suggesting an association with clinical experience. Most patients had at least one visit in the year on which they reported not having good adherence, showing the need for continued adherence interventions.

From the *MRC Uganda Research Unit on AIDS, Entebbe, Uganda; †Medical Research Council Clinical Trials Unit, London, United Kingdom; ‡University of Harare, College of Health Sciences, Harare, Zimbabwe; §Joint Clinical Research Centre, Kampala, Uganda; ‖Makerere University, Infectious Diseases Institute, Kampala, Uganda; ¶Division of Medicine, Imperial College, London, United Kingdom; and the #World Health Organization, Geneva, Switzerland.

Received for publication November 14, 2007; accepted April 24, 2008.

Names of researchers in the DART Trial Team given in the acknowledgments.

The DART trial is funded by the UK Medical Research Council, the UK Department for International Development, and the Rockefeller Foundation. First-line drugs are provided by GlaxoSmithKline, Gilead Sciences, and Boehringer Ingelheim.

Presented at the 1st International Workshop on HIV Treatment, Pathogenesis and Prevention Research in Resource-Poor Settings, May 30 to June 2, 2007, Kampala, Uganda.

Correspondence to: Sylvia K. Muyingo, MSc, MRC Uganda Research Unit on AIDS, Box 49, Entebbe, Uganda (e-mail:

Back to Top | Article Outline


Since 1995, the introduction of antiretroviral therapy (ART) has led to substantial reductions in HIV-associated mortality and morbidity in industrialized countries.1-3 There is a strong relationship between adherence, effective viral suppression, and survival.4-6 Although some patients maintain viral suppression with moderate levels of adherence to newer boosted protease inhibitor and nonnucleoside reverse transcriptase inhibitors,7,8 the potential for the emergence of resistance supports maximal adherence for optimal long-term outcomes.

Recent expansion of ART in resource-limited settings has led to similar improvements in mortality. Early reports suggest high levels of adherence in Cote d'Ivoire, Senegal, Uganda, and South Africa.9-12 However, most reports have considered adherence at a population level, that is, what proportion of the population on treatment is adherent at various times after initiation of therapy, and have shown this population level adherence declining or unchanging as time on ART increases.13,14 Analysis at an individual level is of more interest, whether there is a small subset of patients consistently adhering poorly or more patients have poor adherence for a limited period of time. The challenge is maintaining good adherence over the longer term at the individual level. This may be difficult to sustain without cultural and social support in resource-limited settings, support which may not be scalable to the estimated 4.6 million people needing ART in Africa alone.15 In addition, good adherence is more than compliance to medication but includes following instructions for prescriptions and attendance at scheduled appointments.16

There is no gold standard for measuring adherence. In developed countries, self-report and pill count (PC) seem to overestimate adherence but have been significantly associated with viral load suppression.13 Given the absence of viral load monitoring in most African settings, measurement of adherence is even more important to manage response to treatment and provide interventions for patients having difficulties with medication. There have been few reports of validation of adherence measurements in Africa. Nonadherence has been associated with the drug regimen, personal factors, stigma, side effects, and travel away from home.10,17,18

This paper describes adherence to ART in HIV-infected subjects in Uganda and Zimbabwe enrolled in the Development of Antiretroviral Therapy in Africa (DART) trial over the first year on ART,19 comparing different measures of adherence, their relationship with viral load suppression in a subset of participants, and assessing risk factors for poor adherence.

Back to Top | Article Outline


Study Design

DART is an open, multicenter, randomized trial, comparing management approaches relevant to resource-limited settings: clinical monitoring only versus laboratory plus clinical monitoring.19,20 Eligible subjects were ART-naive adults (18 years and older), with documented advanced HIV infection [World Health Organization (WHO) clinical stage 2, 3, or 4] and CD4 count <200 cells/μL in 2 centers in Uganda (1 with an additional satellite site) and 1 center in Zimbabwe. Subjects were ineligible if they had current acute infections, were receiving chemotherapy for malignancies, or were pregnant or breast-feeding. At enrollment, all participants received counseling about medication adherence and drug side effects from a nurse or doctor and had group counseling sessions. This counseling was also reinforced at each clinic visit. All participants initiated first-line triple drug therapy with coformulated zidovudine/lamivudine (Combivir) and tenofovir disoproxil fumarate (DF) (3 pills/d), nevirapine (4 pills/d), or abacavir (4 pills/d) as recommended in World Health Organization guidelines at the time the trial started.21

Clinic visits were scheduled at 2 and 4 weeks after enrollment and then every 4 weeks. Participants were asked to return to the clinic if they felt unwell at any time. At each clinic visit, participants were given a new 4-week supply of drugs (no extra pills for late attendance), unused pills from the previous period were counted and recorded, and a structured adherence questionnaire was completed. Every 12 weeks, all participants were seen by a doctor and had a routine full blood count, tests of liver and kidney function, and measurement of lymphocyte subsets.

Viral loads were performed at baseline and at weeks 4, 12, 24, 36, and 48 on a subset of subjects taking zidovudine, lamivudine, and tenofovir. Shortly after the trial started, 100 consecutive subjects were selected from each of the Entebbe, Harare, and Kampala centers divided equally between those with CD4 cell counts 0-99 and 100-199 cells/mm3 at ART initiation.22

Back to Top | Article Outline

Adherence Definitions

The 2 primary measures of adherence were based on the total number of unused pills (for all ART drugs prescribed, including substitutions for toxicity) at each 4-week visit documented by the nurse in the clinic. The PC adherence was defined as 1 minus the proportion of pills dispensed that were returned. Drug possession ratio (DPR) was defined as the days' supply of drugs delivered minus the days' supply of drugs returned divided by the number of days between clinic visits, assuming that ART was used continuously throughout the period between the clinic visits.6,23 DPR takes into account the date of the clinic visits and adjusts for late return to clinic (meaning pills had been missed) and early return to clinic (meaning the patient should have pills to return). For both measures, we defined complete adherence as 100% adherence and good adherence as at least 95% adherence.24

Secondary measures of adherence were taken from the structured adherence questionnaire, namely, reporting late for the scheduled clinic visit, missing any ART doses in the 4 days before the clinic visit, missing an ART dose in the past month, or forgetting the dose at weekends.

Back to Top | Article Outline

Selection of Participants

This analysis considers the first 52 weeks after ART initiation, excluding those lost to follow-up or dying in the first year. For each participant, the number and proportion of clinic visits that achieved complete (100%) adherence or good (≥95%) adherence were calculated per quarter and over the whole year for each measure of adherence, treating missing observations as not achieving complete/good adherence.

Back to Top | Article Outline

Statistical Analyses

Data were analyzed in STATA 9.0. The effect of each adherence measure summarized over the preceding 12 weeks on viral load suppression (<400 or <50 copies/mL) at 12, 24, 36, and 48 weeks was estimated using generalized estimating equations adjusting for time on ART. Similar models using normal interval regression25 were used to investigate the effect of different measures of adherence on absolute log10 HIV-1 RNA viral load adjusted for baseline viral load and CD4 count.

Generalized estimating equations (GEE) with exchangeable correlation structure were used to estimate associations between complete (100%) DPR adherence and baseline social, demographic, and clinical characteristics across study visits. The effect of time on ART was estimated using 2 linear slopes with a change-point at 12 weeks. Multivariable models were selected based on backward elimination (P = 0.2). Odds ratios (ORs), adjusted odds ratios (aORs), and 95% confidence intervals (95% CIs) are presented.

Back to Top | Article Outline


From January 5, 2003 to October 28, 2004, the DART trial enrolled 3316 ART-naive adults, 2156 (65%) women, and 1160 (35%) men.19 For this analysis focusing on cumulative adherence to ART during the first year after initiation, we excluded 137 (4%) participants who entered a pilot structured treatment interruption study at 28 weeks, 3 participants who subsequently reported ART use before enrollment, 171 (5%) who died in the first year (85 died in the first 3 months), and 48 (1%) who were permanently lost to follow-up before the end of the first year. There were no significant differences in age, sex, center of recruitment, or WHO clinical stage between those excluded and included in the subsequent analysis. However, 53% of those who died or were lost to follow-up had a CD4 count <50 cells/mL at enrollment.

Of the 2957 participants still under follow-up at 52 weeks, 252 individuals (8.5%) missed a total of 393 of the 38,441 scheduled 4-weekly clinic visits during the course of the year (range 0.9%-1.7% over visit weeks). PCs were not performed at 455 (1.2%) of the 38,048 visits that occurred (range 0.6-2.1%), leaving PCs observed at 37,593 clinic visits: missing PCs and/or visits was treated as nonadherent (see Methods). Structured adherence questionnaires were not completed at 282 visits (0.7%), leaving 37,766 responses for analysis. Nine hundred seventy-nine (2.5%) visits occurred one or more days late (range 1.9%-3.7%). The 2957 patients initiated ART with zidovudine/lamivudine (as Combivir) plus tenofovir DF (2161, 73%), abacavir (285, 10%), or nevirapine (511, 17%).

Back to Top | Article Outline

Adherence Measures

Across all 38,048 PCs, 28,652 (75%) showed 100% adherence, 35,377 (93%) ≥95% adherence, and only 1475 (4%) with <80% adherence. Figure 1 demonstrates that the proportion of patients with ≥95% and 100% DPR adherence over the previous 4 weeks increased progressively over the first year on ART. At week 4, 2581 (87%) participants had ≥95% adherence, increasing to 2745 (93%) at 28 weeks and 2785 (94%) at week 52. Similar increases were seen for other measures of adherence.



Table 1 summarizes the adherence data for participants across 13 clinic visits in the first year. Although good (≥95%) DPR adherence was seen at 93% of all visits in the first year, only 1454 (49%) participants achieved good adherence at every clinic visit (13/13), 76% at at least 12 of the 13 visits, and 87% at at least 11 visits. From the structured questionnaire, during the first year, 1852 (62%) of patients never reported being late for a clinic visit, 1418 (48%) never reported missing a dose in the preceding 4 days, 1938 (66%) never reported forgetting to take ART at weekends, and 816 (28%) never reported missing any dose in the previous month. Two thousand one hundred thirty-seven participants reported 14,595 reasons for missing doses, the most common being: forgot 1072 (N, 50% of participants reporting reasons), away from home 953 (N, 45%), ran out of drugs 678 (N, 32%), too busy 425 (N, 20%), sick 380 (N, 18%), avoiding side effects 282 (N, 13%), change in routine 248 (N, 12%), slept through dose 235 (N, 11%), and too many pills 187 (N, 9%).



Back to Top | Article Outline

Association Between Adherence and Viral Load

Viral load data were available for 278 (9.4%) of the 2957 subjects at baseline. In each subsequent 12-week period, complete adherers were defined as those achieving 100% DPR at all 3 clinic visits preceding the viral load measurement, with 73/267 (27%), 115/274 (42%), 136/270 (50%), and 151/273 (55%) achieving this at 12, 24, 36, and 48 weeks, respectively (denominators changing because of missed visits/no samples). Differences in viral load suppression <50 and <400 copies/mL (the primary outcomes of the virology substudy) between those with 100% DPR and those with lower adherence in the last 12 weeks widened with increasing time on ART, reaching formal statistical significance at week 48 for <50 copies/mL (P = 0.01) but not for <400 copies/mL (P = 0.11) (Fig. 2). Pooling data across all 4 quarters, the unadjusted GEE analysis showed greater viral suppression in complete adherers for <50 copies/mL [OR = 1.34, 95% CI (1.05, 1.71), P = 0.02; adjusted for baseline HIV-1 RNA and CD4 aOR = 1.29 (0.97, 1.70) P = 0.08] but not for <400 copies/mL [OR = 1.09 (0.83, 1.44), P = 0.54; aOR = 1.13 (0.88, 1.45), P = 0.33]. There was no evidence that the effect of complete adherence on viral load varied over time (heterogeneity P = 0.97 and 0.29, respectively). None of the other adherence measures were significantly associated with viral load suppression to either <400 or <50 copies/mL.



In an exploratory analysis with greater power, more sensitive normal interval regression models for change in HIV-1 RNA viral load from baseline, we found independent associations for complete DPR and 2 other measures of adherence over the preceding 3 clinic visits, not missing pills at the weekend and not missing pills in the last month (Table 2). The magnitude of the association between log viral load and 100% DPR was similar between the first 6 visits and the later 7.



Back to Top | Article Outline

Association Between Different Measures of Adherence

Across individual visits, the kappa agreement between 100% DPR and patient reporting not missing any dose in the last month was 83% (κ = 0.44) suggesting moderate agreement, but kappa <0.15 for other adherence measures.

From the 4 questions in the structured questionnaire, the only independent predictor of 100% DPR was not missing doses in the last month, which had a specificity of 97% at identifying complete adherence (100% DPR) and a sensitivity of 40% at identifying poor adherence (<100% DPR adherence). Defining nonadherence by reporting missed doses in last month, OR at weekends did not change specificity/sensitivity (97%, 41%, respectively).

Back to Top | Article Outline

Effect of Baseline Characteristics on Adherence Using DPR

Independently, complete (100%) DPR adherence was significantly higher in Center C (OR = 1.89) and in Center D (OR = 1.68) compared with Center A (P < 0.001, Table 3). Patients initiating ART in 2004 were significantly more likely to achieve complete DPR adherence than those initiating ART in 2003, and the size of this effect increased after adjustment for center, CD4 count, and time since ART. Although lower adherence was observed in Center A, exclusion of this center did not change the overall results.



Complete DPR adherence increased over time, with a 9% increase every 4 weeks over the first 12 weeks [aOR = 1.09, 95% CI (1.08, 1.10), P < 0.001] and a 2% increase from weeks 12 to 52 [aOR = 1.02, (1.02, 1.03), P < 0.001] independently of adjustment for other factors. Complete DPR adherence was also significantly more frequent in those with higher CD4 counts at enrollment [adjusted aOR = 1.08/100 cells higher, (1.01, 1.16), P = 0.025].

Complete DPR adherence was not significantly associated with education, the number of child dependents, or the length of the relationship with the current partner. Although univariably there were significant associations between greater adherence and higher previous educational attainment, female gender not disclosing HIV status and not having been admitted to hospital in the last year (P < 0.05), these did not persist after adjustment for center, baseline CD4 count, and year of randomization (P > 0.2). The only independent social predictor of complete DPR adherence was reporting other sexual partners in the 3 months before starting ART, associated with lower rates of 100% DPR adherence [aOR = 0.72, 95% CI (0.60, 0.87), P = 0.001].

Back to Top | Article Outline


This paper shows excellent clinic attendance and high adherence to ART over the first year of treatment in a large cohort of previously untreated African individuals similar to other studies in sub-Saharan Africa.17,26 In contrast to other studies, population level adherence increased during the course of the first year. Although 90% of clinic visits showed good adherence, only 49% of participants achieved this level of adherence at every clinic visit over the first year, highlighting the fact that individuals may find it difficult to maintain good adherence for long periods of time, as demonstrated from chronic disease studies.27 Our results demonstrate that high levels of adherence reported at a population level may mask considerably more variable adherence at the individual level and emphasize the need for better assessment and promotion of adherence by health care workers in Africa.

Accurate measurement of adherence is difficult: electronic-based monitoring of pill bottle opening is often considered the proxy gold standard but is expensive and intrusive, and its intrinsic effect on underlying adherence is also unknown.28 Clinic-based PCs,24,29 drug refills, and self-reported measures of adherence have therefore generally been used to identify poor adherence patterns, and meta-analysis has demonstrated that self-reported measures can distinguish between virologically meaningful patterns of medication use.30 There are several ways to measure adherence to ART medication; as our aim was to consider convenient measures of adherence to use in routine ART clinics, we used DPR, the proportion of drugs possessed by the patient between clinic visits, and responses to structured questions. In common with other studies, we validated these measures against viral load outcomes, which were associated with 100% DPR adherence but not 95% adherence. Some have argued that lack of virologic suppression may be an inadequate indicator of nonadherence, as other factors such as mutations, regimen potency, and pharmacokinetics can affect virological suppression.30 Thus, adherence measures associated with viral load suppression may not provide very accurate determinants of adherence behavior. We only investigated viral load in patients on Combivir + tenofovir DF, but the association could be different for other drugs.

We found that reductions in HIV RNA viral load were significantly greater in patients with 100% DPR. Although only 9% of subjects had HIV viral load measurements, these patients were selected according to baseline CD4 and clinical center22 and thus should not bias the assessment of any differences in adherence level. The numbers provide more than 90% power to detect a difference of 20% in those with undetectable viral load, between different adherence levels. Responses to simple questions about when the last dose was missed and missing doses at weekends independently predicted viral load changes, and reporting missing a dose in the last month also predicted 100% DPR. We recommend that these 2 questions, as the best self-reported measures, are used by African health care workers on a regular (monthly) basis when the DPR cannot be calculated6,31 and viral load monitoring is not available.

Our finding that self-report of missing pills at the weekend predicted viral load is intriguing in the light of recent observations that 90% of missed doses are treatment interruptions of at least 48 hours. These interruptions are more likely associated with drug resistance, at least when using fixed-dose combinations.32 Nevertheless, the majority of patients, even without good adherence, achieved viral load suppression in agreement with previous reports of excellent virological suppression with adherence to NNRTIs as low as 54%.33

In contrast to other studies in sub-Saharan Africa,34 all measures used in this study showed adherence improving over the first year of ART, with the most significant improvement seen in the first 12 weeks. Possible explanations include increased social support and health improvements, making adherence and clinic attendance easier with returning strength, and increased ability to manage drug side effects. This accords with the observation that adherence was significantly better in patients with higher CD4 counts at ART initiation, highlighting the need for prompt diagnosis of HIV-positive individuals in resource-limited countries to enable them to access ART as soon as they reach thresholds recommended in national guidelines.35

We observed a strong learning effect of calendar time, likely explained by improved experience of counseling and support for patients within the DART trial. This highlights the importance of prioritizing adherence counseling when setting up new ART clinics in resource-limited settings. The best adherence was seen in an urban center and lowest in a rural center, which may indicate that urban, easily accessible clinics are an important component to maintaining adherence. Support from the regular partner and family is also likely to be important, and any lack of trust may impinge negatively on adherence, perhaps explaining why adherence was significantly lower in those who had another sexual partner other than their main/regular partner 3 months before initiating ART.

The reasons for missing pills reported by patients reflected 2 main problems. The first of these was drug related, when patients felt sick, depressed, or side effects from ART. Other reasons were more frequent and came from their personal circumstances when they forgot to take the drugs or were unable to take the drugs correctly because of absence from home or being asleep. These reasons are similar to previous studies in both resource-rich and -limited settings.36,37 Some patients said they did not adhere because they had run out of pills, and this could reflect poor attendance at the clinic or pill sharing.

In this study, clinical care, ART drugs, and transport were provided without cost to study patients. Other studies have shown that cost of drugs and clinical care are important predictors of adherence in sub-Saharan Africa,17,18 and therefore, adherence may be higher in our study than in other settings. Our findings are based on clinic-based PCs and self-reported responses to questionnaires: We did not verify whether or not the patients actually took their drugs or if they had problems with dosing interval or dietary requirements. There are also inevitable errors in our estimates of adherence-for example, extra pills were occasionally given to patients who knew beforehand that they would be late for the next clinic visit, but this was not recorded systematically. Other studies have reported that some patients gave or sold drugs to others or threw away their drugs if they experienced difficulties or side effects38: Such activities would lead us to overestimate adherence. Patient retention in care is also higher (90%) in DART than reported in other ART programs in sub-Saharan Africa,39 perhaps due to the greater resources available in this study.

These data show that overall adherence to ART in this setting is good and improved over the first year of treatment. Further analysis of the longer term adherence of patients in the DART trial is ongoing to explore whether good adherence continues up to 4 years of follow-up. In the first year, all patients were on first-line drugs but adherence may be more complicated with second-line treatment regimens. Despite high adherence at the population level, many patients have periods when they do not achieve good adherence. Patients with these problems are associated with lower baseline CD4 counts, side effects, or lack of support mechanisms. DPR and simple questions on missed doses could be useful tools in identifying when patients have problems with adherence and are feasible for large numbers of patients attending routine clinics. Accurate, valid measures of adherence are needed from regular ART clinics to reinforce the ongoing monitoring of adherence and to evaluate interventions to enhance successful use of ART.

Back to Top | Article Outline


Thanks to Ugandan Ministry of Health, Medical Research Council, Uganda Virus Research Institute, the Tropical Epidemiology Group at the London School of Hygiene and Tropical Medicine & Scholarship. Others working on the study: The AIDS Support Organisation, JCRC, Academic Alliance, Harare. We are grateful to David Bangsberg for his comments on this paper. We thank the reviewers for their helpful and interesting comments. We thank all the patients and staff from all the centers participating in the DART trial. Joint Clinical Research Centre, Kampala, Uganda: P. Mugyenyi, C. Kityo, F. Ssali, D. Tumukunde, T. Otim, L. Namale, A. Mukose, A. Muhwezi, G. Kabuye, G. Mulindwa, D. Atwine, H. Kyomugisha, A. Drasiku, C. Tumusiime, J. Sabiiti, C. Zawedde, J. Komugyena, J. Okiror, R. Byaruhanga, P. Ocitti, T. Bakainyaga Grace, H. Katabira, G. Barungi, D. Masiira, A. Atwine, S. Murungi, J. Tukamushaba, L. Mugarura, P. Mwesigwa; University of Zimbabwe, Harare, Zimbabwe: A. Latif, J. Hakim, A. Reid, A. Jamu, S. Makota, T. Mupudzi, G. Musoro, N. Ngorima, M. Pascoe, F. Taziwa, L. Chakonza, E. Chidziva, H. Chirairo, S. Chitsungo, F. Mapinge, A. Mawora, C. Muvirimi, G. Tinago, J. Chimanzi, J. Machingura, C. Maweni, S. Mutsai, R. Warara, M. Matongo, N. Mdege, S. Mudzingwa, M. Jangano, I. Machingura, K. Moyo, L. Vere, E. Chigwedere, M. Phiri; MRC/UVRI Uganda Research Unit on AIDS, Entebbe, Uganda: H. Grosskurth, P. Munderi, K. Wangati, G. Kibuye, H. Byomire, B. Amuron, D. Nsibambi, R. Kasirye, E. Zalwango, M. Nakazibwe, B. Kikaire, G. Nassuna, R. Massa, M. Aber, M. Namyalo, A. Zalwango, L. Generous, P. Khauka, N. Rutikarayo, W. Nakahima, A. Mugisha, A. Ruberwantari, J. Nakiyingi-Miiro, J. Levin, S. Muyingo, J. Todd, P. Hughes; Academic Alliance, Mulago Hospital, Uganda: E. Katabira, A. Ronald, A. Kambungu, J. Martin, R. Nalumenya, R. Nairubi, E. Bulume, M. Teopista, C. Twijukye, F. Sematala, H. Byakwaga, E. Nabankema, F. Lutwama, A. Muganzi, J. Walusimbi, A. Nanfuka, C. Twijukye, P. Nabongo, J. Wanyama, I. Namata, T. Oketta, R. Kulume, E. Namara; The AIDS Support Organisation, Uganda: A. Coutinho, B. Etukoit; Imperial College, London, UK: C. Gilks, K. Boocock, C. Puddephatt, D. Winogron; MRC Clinical Trials Unit: J. Darbyshire, D. M. Gibb, A. Burke, D. Bray, A. Babiker, A. S. Walker, H. Wilkes, M. Rauchenberger, S. Sheehan, L. Peto, K. Taylor; Trial Steering Committee: I. Weller (Chair), A. Babiker (Trial Statistician), S. Bahendeka, M. Bassett, A. Chogo Wapakhabulo, J. Darbyshire, B. Gazzard, C. Gilks, H. Grosskurth, J. Hakim, A. Latif, C. Mapuchere, O. Mugurungi, P. Mugyenyi; Observers: C. Burke, S. Jones, C. Newland, J. Rooney, W. Snowden, J-M Steens; Data and Safety Monitoring Committee: A. Breckenridge (Chair), A. McLaren (Chair, deceased), C. Hill, J. Matenga, A. Pozniak, D. Serwadda.

Back to Top | Article Outline


1. Palella FJ Jr, Delaney KM, Moorman AC, et al. Declining morbidity and mortality among patients with advanced human immunodeficiency virus infection. HIV Outpatient Study Investigators. N Engl J Med. 1998;338:853-860.
2. Hogg RS, Heath KV, Yip B, et al. Improved survival among HIV-infected individuals following initiation of antiretroviral therapy. JAMA. 1998;279:450-454.
3. Lima VD, Hogg RS, Harrigan PR, et al. Continued improvement in survival among HIV-infected individuals with newer forms of highly active antiretroviral therapy. AIDS. 2007;21:685-692.
4. Bangsberg DR, Perry S, Charlebois ED, et al. Non-adherence to highly active antiretroviral therapy predicts progression to AIDS. AIDS. 2001;15:1181-1183.
5. Nachega JB, Hislop M, Dowdy DW, et al. Adherence to highly active antiretroviral therapy assessed by pharmacy claims predicts survival in HIV-infected South African adults. J Acquir Immune Defic Syndr. 2006;43:78-84.
6. Gross R, Yip B, Re VL III, et al. A simple, dynamic measure of antiretroviral therapy adherence predicts failure to maintain HIV-1 suppression. J Infect Dis. 2006;194:1108-1114.
7. Shuter J, Sarlo JA, Kanmaz TJ, et al. HIV-infected patients receiving lopinavir/ritonavir-based antiretroviral therapy achieve high rates of virologic suppression despite adherence rates less than 95%. J Acquir Immune Defic Syndr. 2007;45:4-8.
8. Nachega JB, Hislop M, Dowdy DW, et al. Adherence to nonnucleoside reverse transcriptase inhibitor-based HIV therapy and virologic outcomes. Ann Intern Med. 2007;146:564-573.
9. Katzenstein D, Laga M, Moatti JP. The evaluation of the HIV/AIDS drug access initiatives in Cote d'Ivoire, Senegal and Uganda: how access to antiretroviral treatment can become feasible in Africa. AIDS. 2003;17 (Suppl 3):S1-S4.
10. Laniece I, Ciss M, Desclaux A, et al. Adherence to HAART and its principal determinants in a cohort of Senegalese adults. AIDS. 2003;17 (Suppl 3):S103-S108.
11. Gill CJ, Hamer DH, Simon JL, et al. No room for complacency about adherence to antiretroviral therapy in sub-Saharan Africa. AIDS. 2005;19:1243-1249.
12. Orrell C, Bangsberg DR, Badri M, et al. Adherence is not a barrier to successful antiretroviral therapy in South Africa. AIDS. 2003;17:1369-1375.
13. Mannheimer S, Friedland G, Matts J, et al. The consistency of adherence to antiretroviral therapy predicts biologic outcomes for human immunodeficiency virus-infected persons in clinical trials. Clin Infect Dis. 2002;34:1115-1121.
14. Howard AA, Arnsten JH, Lo Y, et al. A prospective study of adherence and viral load in a large multi-center cohort of HIV-infected women. AIDS. 2002;16:2175-2182.
15. World Health Organisation Report. Towards Universal Access: Scaling Up Priority HIV/AIDS Interventions in the Health Sector. Geneva, Switzerland: The World Health Organisation; 2006.
16. Simoni JM, Frick PA, Pantalone DW, et al. Antiretroviral adherence interventions: a review of current literature and ongoing studies. Top HIV Med. 2003;11:185-198.
17. Weiser S, Wolfe W, Bangsberg D, et al. Barriers to antiretroviral adherence for patients living with HIV infection and AIDS in Botswana. J Acquir Immune Defic Syndr. 2003;34:281-288.
18. Byakika-Tusiime J, Oyugi JH, Tumwikirize WA, et al. Adherence to HIV antiretroviral therapy in HIV+ Ugandan patients purchasing therapy. Int J STD AIDS. 2005;16:38-41.
19. Reid A, Munderi P, Kityo C, et al. A randomised trial of monitoring practice and structured treatment interruptions in the management of antiretroviral therapy in adults with HIV infection in Africa. Paper presented at: 15th World AIDS Conference; July 11-16, 2004; Bangkok, Thailand. Abstract TuPeB4495.
20. Ssali F, Stohr W, Munderi P, et al. Prevalence, incidence and predictors of severe anaemia with zidovudine-containing regimens in African adults with HIV infection within the DART trial. Antivir Ther. 2006;11:741-749.
21. WHO. World Health Organization Scaling Up Antiviral Therapy in Resource Limited Settings: Guidelines for a Public Health Approach. Geneva, Switzerland; 2002.
22. DART Virology Group and Trial Team. Virological response to a triple nucleoside/nucleotide analogue regimen over 48 weeks in HIV-1-infected adults in Africa. AIDS. 2006;20:1391-1399.
23. Sikka R, Xia F, Aubert RE. Estimating medication persistency using administrative claims data. Am J Manag Care. 2005;11:449-457.
24. Mills EJ, Nachega JB, Buchan I, et al. Adherence to antiretroviral therapy in sub-Saharan Africa and North America: a meta-analysis. JAMA. 2006;296:679-690.
25. Marschner IC, Betensky RA, DeGruttola V, et al. Clinical trials using HIV-1 RNA-based primary endpoints: statistical analysis and potential biases. J Acquir Immune Defic Syndr Hum Retrovirol. 1999;20:220-227.
26. Oyugi JH, Byakika-Tusiime J, Charlebois ED, et al. Multiple validated measures of adherence indicate high levels of adherence to generic HIV antiretroviral therapy in a resource-limited setting. J Acquir Immune Defic Syndr. 2004;36:1100-1102.
27. Reid SE, Reid CA, Vermund SH. Antiretroviral therapy in sub-Saharan Africa: adherence lessons from tuberculosis and leprosy. Int J STD AIDS. 2004;15:713-716.
28. Chesney MA. The elusive gold standard. Future perspectives for HIV adherence assessment and intervention. J Acquir Immune Defic Syndr. 2006;43 (Suppl 1):S149-S155.
29. Bangsberg DR, Hecht FM, Clague H, et al. Provider assessment of adherence to HIV antiretroviral therapy. J Acquir Immune Defic Syndr. 2001;26:435-442.
30. Nieuwkerk PT, Oort FJ. Self-reported adherence to antiretroviral therapy for HIV-1 infection and virologic treatment response: a meta-analysis. J Acquir Immune Defic Syndr. 2005;38:445-448.
31. Glass TR, De Geest S, Weber R, et al. Correlates of self-reported nonadherence to antiretroviral therapy in HIV-infected patients: the Swiss HIV Cohort Study. J Acquir Immune Defic Syndr. 2006;41:385-392.
32. Oyugi JH, Byakika-Tusiime J, Ragland K, et al. Treatment interruptions predict resistance in HIV-positive individuals purchasing fixed-dose combination antiretroviral therapy in Kampala, Uganda. AIDS. 2007;21:965-971.
33. Bangsberg DR. Less than 95% adherence to nonnucleoside reverse-transcriptase inhibitor therapy can lead to viral suppression. Clin Infect Dis. 2006;43:939-941.
34. Laurent C, Diakhate N, Gueye NF, et al. The Senegalese government's highly active antiretroviral therapy initiative: an 18-month follow-up study. AIDS. 2002;16:1363-1370.
35. Wood E, Hogg RS, Yip B, et al. The impact of adherence on CD4 cell count responses among HIV-infected patients. J Acquir Immune Defic Syndr. 2004;35:261-268.
36. Ammassari A, Trotta MP, Murri R, et al. Correlates and predictors of adherence to highly active antiretroviral therapy: overview of published literature. J Acquir Immune Defic Syndr. 2002;31 (Suppl 3):S123-S127.
37. Bartlett JA. Addressing the challenges of adherence. J Acquir Immune Defic Syndr. 2002;29 (Suppl 1):S2-S10.
38. Weidle PJ, Wamai N, Solberg P, et al. Adherence to antiretroviral therapy in a home-based AIDS care programme in rural Uganda. Lancet. 2006;368:1587-1594.
39. Rosen S, Fox MP, Gill CJ. Patient retention in antiretroviral therapy programs in sub-Saharan Africa: a systematic review. PLoS Med. 2007;4:e298.

HIV; Africa; adherence; ART

© 2008 Lippincott Williams & Wilkins, Inc.