Secondary Logo

Journal Logo

Clinical Trial Results

Sustained Benefit From a Long-Term Antiretroviral Adherence Intervention

Results of a Large Randomized Clinical Trial

Mannheimer, Sharon B MD*; Morse, Edward PhD; Matts, John P PhD; Andrews, Laurie RN, MPH§; Child, Carroll RN, MSc; Schmetter, Barry BS; Friedland, Gerald H MD§ for the Terry Beirn Community Programs for Clinical Research on AIDS

Author Information
JAIDS Journal of Acquired Immune Deficiency Syndromes: December 1, 2006 - Volume 43 - Issue - p S41-S47
doi: 10.1097/
  • Free


The critical role of adherence in the treatment of HIV has been demonstrated in clinical trials and clinical care settings. Antiretroviral (ARV) adherence is a strong predictor of biologic (virologic and immunologic)1-7 and clinical outcomes in HIV, including quality of life, HIV progression, hospitalizations, and death.8-13

To date, studies of interventions for improving ARV adherence have been limited in design and execution. Comprehensive, multifaceted, and repetitive interventions have achieved the best results in improving medication adherence for HIV and other conditions, although effect sizes are often modest.14-17 Successful ARV adherence interventions tested in randomized clinical trials have included pharmacists,18,19 education,19-22 adherence counselors,23,24 and incentives.25 Few trials showed improved biologic outcomes (eg, improved virologic suppression)19,23 or a durable effect, however. Larger effect sizes for adherence and biologic outcomes have been demonstrated in pilot studies of directly observed therapy (DOT) for ARV therapy,26-29 but widespread use of DOT may be limited by cost and labor intensity.30

Given the frequency of suboptimal ARV adherence,31 there is a clear need for reliable, sustainable, and generalizable adherence interventions. It is likely that long-term adherence interventions are needed for a durable effect, particularly in chronic diseases such as HIV. The Community Programs for Clinical Research on AIDS (CPCRA) conducted an adherence study (CPCRA 062) to test the effectiveness of 2 long-term ARV adherence interventions in a randomized clinical trial. The CPCRA is a community-based HIV clinical trials network with 17 units and 160 collaborating HIV primary care sites throughout the United States.



HIV-infected persons enrolling in CPCRA 058 (the Flexible Initial Retrovirus Suppressive Therapies [FIRST] study) were eligible to coenroll in the CPCRA Adherence Study from November 1999 to January 2002. Informed consent was obtained from all FIRST study participants for study procedures and data collection. The FIRST study32 compared 3 ARV strategies for ARV-naive patients: (1) protease inhibitor (PI)- + nucleoside reverse transcriptase (NRTI)-containing regimens (2-class ARV) (2) nonnucleoside reverse transcriptase inhibitor (NNRTI)- + NRTI-containing regimens (2-class ARV), and (3) PI- + NNRTI- + NRTI-containing regimens (3-class ARV). The FIRST study opened in January 1999 and closed enrollment in January 2002; participant follow-up ended in September 2005. Patients enrolling in the FIRST study at clinical sites authorized to carry out the CPCRA Adherence Study were offered the option of participating in the Adherence Study intervention that the site was randomized to administer. Informed consent for the Adherence Study was obtained from FIRST study participants who agreed to participate. Only participants who signed the Adherence Study informed consent form received the assigned adherence intervention. Adherence Study follow-up continued through June 2003.


The interventions evaluated in this study are 2 different approaches to promoting ARV adherence. Both were long term, starting before or at the time of ARV therapy initiation and continuing throughout follow-up.

The medication manager (MM) intervention involved a trained research staff member who worked individually with study participants to provide tailored adherence support over time in a standardized protocol-guided manner, identifying and addressing each participant's information, motivation, and skills for ARV adherence using detailed questionnaires. This multifaceted intervention was based on health behavioral theory, including the information, motivation, and behavioral skills model of behavior change.33,34

The second intervention was the electronic medication reminder system. The study used a small portable alarm (A Little Reminder [ALR]; Timely Devices Incorporated, Edmonton, Alberta, Canada) individually programmed to sound and flash at times of all ARV doses. The ALR addressed the most common reason for missed ARV doses reported at the time the study was developed, forgetfulness.35


This study used a 2 × 2 factorial design of MM versus no MM and ALR versus no ALR. The 4 study group assignments were (1) MM, (2) ALR, (3) MM + ALR, and (4) usual care (control); primary comparisons were MM versus no MM and ALR versus no ALR.

Cluster randomization was used, in which a site or group of sites at a CPCRA unit were randomized with a 1:1:1:1 allocation into 1 of 4 study arms (rather than randomizing individual participants). Sites were excluded if they used interventions similar to MM or ALR for most of their patients.


For clusters randomized to the MM intervention, MMs were identified from existing CPCRA staff at clinical sites-in most cases, a research nurse. MMs received intensive 1- to 2-day centralized training conducted by CPCRA Adherence Study investigators, which included didactic and experiential sessions on the behavioral rationale for intervention, importance of adherence, predictors of nonadherence, identifying and addressing adherence barriers, use of study questionnaires and tools, and strategies for long-term adherence support.

MMs followed structured specific procedures for regular contact with participants. A standardized MM baseline questionnaire (available at:, developed by the protocol team and administered to participants who consented to the Adherence Study, was used to obtain an in-depth assessment of potential barriers to adherence related to a lack of information, motivation, and behavioral skills. MMs conducted baseline assessments within 45 days before ARV initiation in the FIRST study and used the information to develop individually tailored adherence support plans. Participants also met with MMs after randomization to a FIRST study arm for further adherence support, including a review of their ARV regimen, development of a plan for timing of ARV doses and possible side effects, and instruction on pillbox use. Adherence support also was provided through weekly telephone contacts for weeks 1, 2, and 3 after randomization, followed by telephone contacts at least monthly thereafter. A medication review, a 1-day adherence self-report form, and a Morisky adherence questionnaire36 were used by MMs to assess adherence during contacts. Follow-up face-to-face assessments, coinciding with FIRST study follow-up visits, occurred at 1 and 4 months after randomization and every 4 months thereafter. At the 1-month visit, participants demonstrated how they used their pillboxes and completed 3-day adherence and Morisky adherence questionnaires. At 4-month study visits, standardized MM follow-up questionnaires were used to identify and address new or ongoing adherence barriers. Contacts were increased to weekly for 4 weeks if ARV medications were changed or virologic failure occurred.

CPCRA staff (mostly research nurses) at clusters randomized to the ALR intervention received initial training about the use and programming of ALRs and were responsible for distributing, programming, and monitoring the use of ALR devices. ALRs were programmed to correspond to the timing of all ARV medication doses for each participant. A standardized instrument was used to monitor participants' ALR use at 1 and 4 months after randomization and every 4 months thereafter. Battery function was checked at the same time points; new batteries or reprogramming was available as needed.

Data Collection

All data for analysis were collected through the FIRST study, including baseline demographic characteristics and laboratory data (CD4 lymphocyte cell count and HIV RNA level by the ultrasensitive Roche (Alameda, CA) 1.0 Amplicor polymerase chain reaction [PCR] assay). Follow-up data collected through the FIRST study at months 1 and 4 and every 4 months thereafter included CD4 cell counts and HIV RNA levels, ARV medications, self-reported ARV adherence using a validated questionnaire,1 and quality of life. Grade 4 adverse events were collected when reported.


The primary endpoint was first virologic failure, defined as first plasma HIV RNA level >2000 copies/mL occurring on or after the 4-month follow-up visit (regardless of whether undetectable HIV RNA levels were achieved before that time). This endpoint was chosen to match the planned primary endpoint of the parent ARV strategy trial, the FIRST study. Secondary planned endpoints included other measures of plasma HIV RNA level, CD4 cell count, adherence, ARV regimen changes, ARV resistance, grade 4 adverse events, and quality of life.


Sample Size

The assumptions used in the FIRST study regarding time to first HIV RNA level >2000 copies/mL were utilized to estimate the event rate.32 We chose a conservative effect size of 15% based on a meta-analysis adherence intervention used in other chronic medical conditions.14 Other CPCRA data were used to examine the intraunit correlation of HIV RNA level. Most values were small; only 1 exceeded 0.05. To be conservative, we assumed an intracluster correlation of 0.05. Losses to follow-up were assumed to be 3% per year. The sample size assumed that 40 clusters would be randomized. A 2-sided level of significance of 0.05 was used, along with a power of 0.80. Incorporating the intracluster correlation,37,38 the total sample size was estimated to be 948 patients.

Data Analysis

The primary data analysis, involving all patients enrolled in the CPCRA FIRST study after the patient's site was authorized to begin the Adherence Study, compared participants in clusters randomized to MM versus those in clusters with no MM and those in clusters randomized to ALR versus those with no ALR. Because the randomization unit was a cluster, this intent-to-treat (ITT) analysis included all patients within each cluster (from the parent FIRST study) whether or not these FIRST study participants consented to participate in the Adherence Study and their cluster's adherence intervention. All data used in the ITT analysis were collected through the FIRST study. The ITT analysis was based on the assumption that participants may benefit from being in the environment of the site's adherence intervention whether or not they received the actual intervention. A test for interaction between interventions was taken into account before assessing the main effects. Cox proportional hazards life table analysis, which takes into account the clustering, was used to compare the interventions.

Secondary analyses were performed using the same methods utilized for primary analysis to address time to event endpoints. Mixed models with clusters as random components were used to analyze the continuous variables cross sectionally and longitudinally. Additional analyses examined specific effects of baseline determinants of adherence (age, race, prior AIDS, baseline CD4 count, baseline HIV RNA level, and injection drug use [IDU] history) and effects of the randomized FIRST study ARV arm on the intervention effects. Additional “as-treated” analyses were conducted for the primary endpoint, evaluating only FIRST study participants who consented to participate in the adherence intervention assigned to their CPCRA clinical site.


Baseline Participant Characteristics

A total of 928 FIRST study participants (98% of target) were eligible for enrollment into the CPCRA Adherence Study, and data from these participants were used in the main ITT analyses. Participants were distributed into study groups by cluster randomization as follows: 10 clusters (256 patients) in the MM arm, 10 clusters (254 patients) in the ALR arm, 9 clusters (196 patients) in the MM + ALR arm, and 9 clusters (222 patients) in the control (usual care) arm. Baseline characteristics of the 928 FIRST study participants eligible for participation in the Adherence Study are summarized in Table 1. Characteristics were similar across study arms, including distribution of randomized ARV strategies (data not shown). Of the 928 FIRST study participants in the ITT analyses, 823 (89%) consented to participate in the Adherence Study and to receive the adherence intervention assigned to their clinical site; data from these Adherence Study participants were used in as-treated analyses. Participants who consented to participate in the Adherence Study were similar to those who declined except that refusers were less likely to be injection drug users and had lower baseline CD4 counts (159 vs. 205 cells/mm3; P = 0.02).

Participants' Baseline Characteristics*

Follow-up continued through June 2003; the median follow-up was 30 months. Follow-up HIV RNA levels and CD4 cell counts were obtained at 89% and 91%, respectively, of expected follow-up visits. For 1% of participants, all follow-up specimens were missing.

Medication Manager Results

Primary Endpoint

In the main ITT analysis, the MM participants had a 13% lower rate of the primary endpoint (ie, first virologic failure on or after the 4-month visit), but this was not statistically significant (P = 0.13; Fig. 1A; Table 2). The effect size was 14% after adjusting for baseline characteristics associated with adherence (P = 0.13). The as-treated analysis (evaluating only FIRST study participants who consented to participate in the Adherence Study) of the virologic failure primary endpoint yielded similar results to the ITT analysis (relative risk [RR] = 0.84 in the MM group; P = 0.10). Significant findings, however, were seen in a planned subgroup analysis of the effect of the randomized FIRST study ARV strategy (P = 0.03). MM participants randomized to PI-based ARV had significantly lower rates of virologic failure (vs. no MM), with a trend favoring MM for those randomized to NNRTI-based ARV treatment (see Table 2). Similar results were seen when comparing the 2-class ARV strategies with the 3-class ARV strategy, with a 28% lower risk of virologic failure in MM versus no-MM participants on 2-class ARV (28.6 vs. 41.2 per 100 person-years, RR = 0.72; P = 0.01) and no benefit from MM for participants on 3-class ARV. Of note, the 3-class arm had a significantly higher rate of discontinuing ARVs because of toxicity than the 2-class strategy arms (17.6 vs. 11.2 per 100 person-years; P < 0.01).

A, MM effect on primary endpoint of virologic failure (defined as first HIV RNA level >2000 copies/mL at or after the 4-month visit). Participants receiving the MM intervention had a 13% lower risk of achieving virologic failure than participants not receiving the MM intervention (No-MM) (P = 0.13). B, MM effect on CD4 cell count change from baseline. Participants randomized to receive the MM intervention achieved a higher mean increase in CD4 cell count from baseline compared with participants randomized to a no-MM intervention (difference = 22.5 cells/mm3; P = 0.01). C, Effect of MM on 100% adherence over time. Participants randomized to the MM intervention had a higher rate of reporting 100% adherence over time compared with participants randomized to a no-MM intervention (OR = 1.42; P < 0.001).
Comparison of Primary and Secondary Endpoint Rates for MM and Non-MM Groups

Secondary Endpoints

Significant long-term (over entire study duration) benefits associated with MM included a sustained higher mean increase in CD4 cell count from baseline (difference = 22.5 cells/mm3; P = 0.01; see Fig. 1B) and a sustained higher rate of reporting 100% adherence (odds ratio [OR] = 1.42; P < 0.001; see Fig. 1C). No significant differences were noted between MM and no-MM groups for other secondary endpoints, including change in any ARV drug at any time (60.7% vs. 62.5%, respectively), grade 4 adverse events (10.8 vs. 12.2 per 100 person-years, respectively), and quality of life. When the primary endpoint was analyzed among the subgroup that first achieved an HIV RNA level <2000 copies/mL and then later rebounded above that level, results were similar to those for the entire group, with a 15% reduced risk of virologic rebound in MM groups (P = 0.11). No benefit was seen in all MM participants when virologic failure was defined as an HIV RNA level >50 copies/mL (P = 0.16). Further analyses of ARV resistance and its relation to the adherence intervention strategies are to be conducted when relevant data are available through the CPCRA FIRST study.

Electronic Reminder Results

The rate of first virologic failure was 25% higher in ALR groups (37.8 vs. 29.2 per 100 person-years; P = 0.02; Fig. 2). The adjusted effect size was 22% (P = 0.06). The as-treated analysis (evaluating only the participants who consented to participate in the Adherence Study) yielded similar results, with a 25% higher rate of virologic failure in ALR groups (P = 0.03). Among those who first achieved an HIV RNA level ≤2000 copies/mL in the first 8 months and then rebounded, the rate of first virologic failure was 11% higher in ALR groups than in no-ALR groups (P = 0.34). All randomized FIRST study ARV treatment strategies had higher rates of virologic failure in ALR versus no-ALR groups, although a significantly worse outcome was only seen in 3-class subgroup (40.1 vs. 22.8 per 100 person-years; P < 0.01).

Electronic medication reminder (ALR) effect on primary endpoint. Participants randomized to the ALR electronic reminder device had a 25% higher rate of first virologic failure (the primary endpoint) compared with the no-ART group (37.5 vs. 29.2 per 100 person-years for the ALR vs. no-ALR groups; P = 0.02).

No significant differences were seen between the ALR and no-ALR groups for any long-term secondary endpoint, including proportion over time with an HIV RNA level <50 copies/mL, log HIV RNA level over time, CD4 change over time, adherence, changes in ARV drugs, grade 4 adverse events, and quality of life.

Individual Group Comparisons

There were a total of 445 first virologic failure endpoints reached during the follow-up period. The first virologic failure rate per 100 person-years was lowest in the MM (28.2), highest in the ALR (41.6), and intermediate in the ALR + MM (33.3) and control (no MM, no ALR) arms (30.4). Compared with the control group, the RR of first virologic failure was 1.08 (P = 0.49) in the ALR + MM group, 0.91 (P = 0.53) in the MM group, and 1.34 (P = 0.05) in the ALR group. No significant interaction was seen between MM and ALR interventions (P = 0.51).


This study, the largest randomized controlled adherence trial, evaluated 2 approaches to enhancing ARV adherence: a comprehensive labor-intensive MM support intervention aimed at identifying and addressing participants' barriers to adherence and a technologic medication reminder intervention (ALR device). A strength of the study was the use of adherence and biologic endpoints as study outcomes. Significant reductions in virologic failure were seen for MM participants receiving 2-class (PI- or NNRTI-based) ARV when analyzed separately by PI or NNRTI arms or pooled, although the primary endpoint of first virologic failure did not reach statistical significance in all MM intervention participants. The 3-class ARV arm did not benefit from the MM intervention, possibly related to the higher rate of medication toxicity experienced by those randomized to this strategy. The significant reductions in virologic failure seen in MM participants on 2-class ARV therapy are more relevant for current HIV therapy, because 3-class ARV is not recommended as standard of care.39 In addition, other statistically significant benefits seen with the MM intervention included sustained CD4 lymphocyte count increase and better long-term adherence.

The effect sizes seen in this study are similar to those achieved with other successful behavioral interventions.14,16 In addition, the results of the MM intervention are consistent with adherence literature showing that multifaceted, comprehensive, and repetitive adherence support interventions are the most effective. The results may be related to the more frequent contact and social support provided in the MM intervention compared with the alarm intervention. Both interventions were long term, offered for a median of 30 months. Of note, the MM intervention and benefits followed a single carefully planned training session without additional training over the course of the study. The MM intervention differs from other recent nursing-based ART adherence support interventions40 in its structured frequent protocol-directed interactions with patients, its theory-driven identification and management of adherence barriers, and the tailoring of the intervention to individual needs.

Although the effect size for the primary endpoint was smaller than that hypothesized in the study design, the MM intervention may still be cost-effective in similar ARV-naive populations with advanced HIV disease, particularly given the other significant findings in this study.41 The MM intervention itself was not costly, because no additional staff were hired at the sites to perform the MM duties; sites utilized research nurses already in place for the FIRST study. The research staff underwent only a single initial centralized MM training session. The intervention itself required regular telephone contact but did not require additional face-to-face visits beyond those required for the FIRST study. If the MM intervention was applied in nonresearch clinical practice settings, however, the cost of nursing staff time would likely need to be evaluated.

The cause of the negative effect noted in association with the ALR intervention is unclear. Research staff indicated problems associated with the ALR device, including confidentiality concerns and battery issues and/or malfunctioning, which are factors that may have limited its effectiveness. Although reminders have been shown to benefit those with impaired cognitive function,42 they may not be an appropriate “one-size-fits-all” intervention for most persons receiving ARV therapy. In addition, although the ALR served as a reminder, it did not address other potential adherence barriers such as depression, substance use, and medication side effects, which are issues more likely to be handled by a more comprehensive intervention such as the MM intervention. Unfortunately, data on such factors were not collected, representing a limitation of the study.

A notable feature of this study was its cluster randomization. This design allowed for the administration of the same intervention for all participants at participating sites, thereby eliminating cross-contamination among study personnel and between participants, which can occur when a behavioral intervention is randomized within a clinical site. Cluster randomization assumed that patients would benefit from the interventions whether or not they actually agreed to participate and also reduced the risk of selection bias from differentially favoring participation by site-randomized condition. By including all FIRST study participants at a specific research site in analyses irrespective of their enrollment in the Adherence Study, this study reflects the broad spectrum of persons encountered in clinical practice rather than only those agreeing to participate in an adherence trial, thus adding to the generalizability of the trial results. The generalizability of this and other past adherence intervention trials, however, could be limited by differences in ARV regimen complexity from the time of the trial to the current standard of care.

Several additional study limitations must be noted. The potency and adverse effects of the FIRST study ARV regimens used, reflecting the drug availability and standard of care at the time, may have limited virologic responses for some study participants. For example, there was only limited use of boosted PIs. Other factors that may have affected MM results include MM staff turnover and the use of only 1 initial training session. Another limitation is that because of the multifaceted nature of the MM intervention, we were unable to analyze which specific components of the strategy were most effective. Finally, this study used self-report as the only adherence measure. The instrument used was previously validated in HIV.1 Self-report was chosen because of its ease of use for this large community-based clinical trial.

The results of this large randomized clinical trial suggest that an interpersonal, multifaceted, long-term adherence support intervention is effective in producing long-term improvements in HIV medication adherence and HIV virologic and immunologic outcomes for ARV-naive persons with HIV disease initiating ARV therapy.


The authors thank the study participants and the hard work of the many CPCRA MMs who conducted the intervention.


1. 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.
2. 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. 2001;33:1417-1423.
3. 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. 2000;14:357-366.
4. Paterson DL, Swindells S, Mohr J, et al. Adherence to protease inhibitor therapy and outcomes in patients with HIV infection. Ann Intern Med. 2000;133:21-30.
5. Gifford AL, Bormann JE, Shively MJ, et al. Predictors of self-reported adherence and plasma HIV concentrations in patients on multidrug antiretroviral regimens. J Acquir Immune Defic Syndr. 2000;23:386-395.
6. Haubrich RH, Little SJ, Currier JS, et al. The value of patient-reported adherence to antiretroviral therapy in predicting virologic and immunologic response. California Collaborative Treatment Group. AIDS. 1999;13:1099-1107.
7. Parienti JJ, Massari VR, Descamps D, et al. Predictors of virologic failure and resistance in HIV-infected patients treated with nevirapine or efavirenz-based antiretroviral therapy. Clin Infect Dis. 2004;38:1311-1316.
8. Mannheimer SB, Matts J, Telzak E, et al. Quality of life in HIV-infected individuals receiving antiretroviral therapy is related to adherence. AIDS Care. 2005;17:10-22.
9. Bangsberg DR, Perry S, Charlebois ED, et al. Non-adherence to highly active antiretroviral therapy predicts progression to AIDS. AIDS. 2001;15:1181-1183.
10. Knobel H, Carmona A, Guelar A, et al. Adherence in patients treated with HAART: influence in hospital admissions. Presented at: XIII International AIDS Conference; 2000; Durban.
11. 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. 2002;30:105-110.
12. Hogg RS, Heath K, Bangsberg D, et al. Intermittent use of triple-combination therapy is predictive of mortality at baseline and after 1 year of follow-up. AIDS. 2002;16:1051-1058.
13. Wood E, Hogg RS, Yip B, et al. Effect of medication adherence on survival of HIV-infected adults who start highly active antiretroviral therapy when the CD4+ cell count is 0.200 to 0.350 × 10(9) cells/L. Ann Intern Med. 2003;139:810-816.
14. Roter DL, Hall JA, Merisca R, et al. Effectiveness of interventions to improve patient compliance: a meta- analysis. Med Care. 1998;36:1138-1161.
15. Ickovics JR, Meade CS. Adherence to HAART among patients with HIV: breakthroughs and barriers. AIDS Care. 2002;14:309-318.
16. Haynes RB, McKibbon KA, Kanani R. Systematic review of randomised trials of interventions to assist patients to follow prescriptions for medications. Lancet. 1996;348:383-386.
17. Simoni JM, Frick PA, Pantalone DW, et al. Antiretroviral adherence interventions: a review of current literature and ongoing studies. Top HIV Med. 2003;11:185-196.
18. Rathbun RC, Farmer KC, Stephens JR, et al. Impact of an adherence clinic on behavioral outcomes and virologic response in treatment of HIV infection: a prospective, randomized, controlled pilot study. Clin Ther. 2005;27:199-209.
19. Knobel H, Carmona A, Grau S, et al. Strategies to optimise adherence to highly active antiretroviral treatment. Presented at: 12th World AIDS Conference; 1998; Geneva.
20. Gifford A, Bormann J, Shively M, et al. Effects of group HIV patient education on adherence to antiretrovirals: a randomized controlled trial. Presented at: Eighth Conference on Retroviruses and Opportunistic Infections; 2001; Chicago.
21. Goujard C, Bernard N, Sohier N, et al. Impact of a patient education program on adherence to HIV medication: a randomized clinical trial. J Acquir Immune Defic Syndr. 2003;34:191-194.
22. Wyatt GE. A randomized clinical trial of a medical adherence and risk reduction intervention with HIV positive men and women. Presented at: XIII International AIDS Conference; 2000; Durban.
23. Tuldra A, Fumaz CR, Ferrer MJ, et al. Prospective randomized two-arm controlled study to determine the efficacy of a specific intervention to improve long-term adherence to highly active antiretroviral therapy. J Acquir Immune Defic Syndr. 2000;25:221-228.
24. Pradier C, Bentz L, Spire B, et al. Efficacy of an educational and counseling intervention on adherence to highly active antiretroviral therapy: French prospective controlled study. HIV Clin Trials. 2003;4:121-131.
25. Rigsby MO, Rosen MI, Beauvais JE, et al. Cue-dose training with monetary reinforcement: pilot study of an antiretroviral adherence intervention. J Gen Intern Med. 2000;15:841-847.
26. Altice FL, Mezger J, Hodges J, et al. Developing a directly administered antiretroviral therapy intervention for HIV-infected drug users: implications for program replication. Clin Infect Dis. 2004;38(Suppl 5):S376-S387.
27. Clarke S, Keenan E, Ryan M, et al. Directly observed antiretroviral therapy for injection drug users with HIV infection. AIDS Read. 2002;12:305-307, 312-316.
28. Conway B, Prasad J, Reynolds R, et al. Directly observed therapy for the management of HIV-infected patients in a methadone program. Clin Infect Dis. 2004;38(Suppl 5):S402-S408.
29. Stenzel MS, McKenzie M, Mitty JA, et al. Enhancing adherence to HAART: a pilot program of modified directly observed therapy. AIDS Read. 2001;11:317-319, 324-328.
30. Mitty JA, Stone VE, Sands M, et al. Directly observed therapy for the treatment of people with human immunodeficiency virus infection: a work in progress. Clin Infect Dis. 2002;34:984-990.
31. Stone VE. Strategies for optimizing adherence to highly active antiretroviral therapy: lessons from research and clinical practice. Clin Infect Dis. 2001;33:865-872.
32. MacArthur RD, Chen L, Mayers DL, et al. The rationale and design of the CPCRA (Terry Beirn Community Programs for Clinical Research on AIDS) 058 FIRST (Flexible Initial Retrovirus Suppressive Therapies) trial. Control Clin Trials. 2001;22:176-190.
33. Fisher JD, Fisher WA, Williams SS, et al. Empirical tests of an information-motivation-behavioral skills model of AIDS-preventive behavior with gay men and heterosexual university students. Health Psychol. 1994;13:238-250.
34. Fisher JD, Fisher WA, Misovich SJ, et al. Changing AIDS risk behavior: effects of an intervention emphasizing AIDS risk reduction information, motivation, and behavioral skills in a college student population. Health Psychol. 1996;15:114-123.
35. Chesney MA. Factors affecting adherence to antiretroviral therapy. Clin Infect Dis. 2000;30(Suppl 2):S171-S176.
36. Morisky D, Green L, Levine D. Concurrent and predictive validity of a self-reported measure of medication adherence. Med Care. 1986;24:67-74.
37. Donner A, Birkett N, Buck C. Randomization by cluster, sample size requirements and analysis. Am J Epidemiol. 1981;114:906-914.
38. Simon R. Composite randomization designs for clinical trials. Biometrics. 1981;37:723-731.
39. DHHS Panel on Antiretroviral Guidelines for Adults and Adolescents-A Working Group of the Office of AIDS Research Advisory Council (OARAC). Guidelines for the use of antiretroviral agents in HIV-1-infected adults and adolescents, May 4, 2006. Available at:
40. Collier AC, Ribaudo H, Mukherjee AL, et al. A randomized study of serial telephone call support to increase adherence and thereby improve virologic outcome in persons initiating antiretroviral therapy. J Infect Dis. 2005;192:1398-1406.
41. Goldie SJ, Paltiel AD, Weinstein MC, et al. Projecting the cost-effectiveness of adherence interventions in persons with human immunodeficiency virus infection. Am J Med. 2003;115:632-641.
42. Andrade AS, McGruder HF, Wu AW, et al. A programmable prompting device improves adherence to highly active antiretroviral therapy in HIV-infected subjects with memory impairment. Clin Infect Dis. 2005;41:875-882.

adherence; alarm; antiretroviral therapy; behavior; HIV; intervention study

© 2006 Lippincott Williams & Wilkins, Inc.