University of California, San Francisco Logo

University of California, San Francisco | About UCSF | Search UCSF | UCSF Medical Center

Randomized control trial of peer-delivered, modified directly observed therapy for HAART in Mozambique
Global Health Sciences Literature Digest
Published January 24, 2008
Journal Article

Pearson CR, Micek MA, Simoni JM, Hoff PD, Matediana E, Martin DP et al. Randomized control trial of peer-delivered, modified directly observed therapy for HAART in Mozambique. J Acquir Immune Defic Syndr. 2007 Oct 1;46(2):238-44.


To assess the efficacy of a peer-delivered intervention to promote short-term (6-month) and long-term (12-month) adherence to HAART in a Mozambican clinic population.

Study Design

A 2-arm randomized controlled trial.


The public HIV care clinic in the Beira Central Hospital, providing free HIV care and ART to all affected Mozambicans in Beira and the surrounding area.


Clinic attendees initiating HAART between October 2004 and March 2005 were eligible for enrollment if they were at least 18 years old, lived near the clinic, and were free of severe mental illness or dementia. 658 patients were referred to the study; 433 were eligible of whom 350 (54% female) were randomized.


Participants were randomly assigned to either peer-delivered modified directly observed therapy (mDOT) or standard of care. The mDOT intervention consisted of 6 weeks of peer supported DOT during which peers provided education about treatment and adherence and sought to identify and mitigate adherence barriers. The standard of care included no-cost medications, clinical and lab follow-up, psychosocial adherence support by a trained social worker, and referral to community-based peer support group. Patients were also encouraged to identify treatment partners.

Primary Outcomes

The primary outcome was self-reported number of HAART medication doses taken in the last 7 days and in the last 30 days at 6 months after HAART initiation. Other outcomes included the same adherence measures at 12 months, CD4 count, and mortality. Data were analyzed with intention to treat both with missing observations imputed as 0% of doses taken and without imputation.


No demographic or clinical differences were identified between the two study arms. Ninety-seven percent of participants were taking an NNRTI combination pill. Adherence Outcomes: Mean adherence was greater than 85% of prescribed doses taken across all groups, decreasing slightly from 6 weeks to 6 months, and increasing slightly at 12 months. Both 7-day and 30-day measures showed greater adherence in the intervention group at 6 months, but results differed somewhat by the method of analysis. Imputing 0% for missing observations adherence comparing the intervention and usual care groups was 97.5% vs. 90.7% (7-day) and 97.7% vs. 92.8% (30-day), both statistically significant, and without imputation adherence was 95.2% vs. 90.1% (7-day) and 96.3% vs. 92.3% (30-day), but the latter was not statistically significant. Further, the adherence difference at 12 months was significant on analysis with imputation on both measures but neither measure differed at 12 months on intention to treat analysis without imputation. Mean adherence at 6 months and 12 months was significantly higher among participants who had attended a community support group (97% vs. 95% at 6 months, and 99% vs. 96% at 12 months), and met with a peer at least once post intervention (97% vs. 91% at 6 months, 98% vs. 93% at 12 months). At 12 months, increased adherence was also associated with self-efficacy and social support. Change in adherence over the 12-month period differed significantly between the two arms (p<0.05), however the between arm difference at each assessment point was less than the 10% considered to be clinically meaningful. mDOT participants were more likely than the standard-care group to achieve ≥90% adherence level at 6 months (92% vs. 85% for the 7-day measure) though this difference was not significant. Clinical Outcomes: mDOT participants were more likely than the control participants to get a CD4 count, but mean CD4 count among those with data did not differ between arms at baseline, 6 or 12 months. Mortality also did not differ between the arms at either 6 or 12-months. Potential Mediators: While mDOT participants were more likely at the 6-month follow-up to attend community support groups, visit a peer post-intervention, and discuss adherence and treatment issues, none of these differences remained significant at 12 months.


The authors conclude that a peer-delivered mDOT program may be an effective strategy to promote long-term adherence among persons initiating HAART in resource-poor settings, but note that further effectiveness research that incorporates community support resources may help to demonstrate how best to disseminate an mDOT strategy in resource-constrained settings.

Quality Rating

Using the Jadad Criteria this study scores 3 out of a possible 5 points--adequate quality. The nature of the study precluded blinding and this may have affected their results. In addition, they acknowledge that the use of self-report for adherence may have biased their results. Furthermore, the recent introduction of free HAART in Mozambique may have led to study participants with unusually strong motivation. Finally, their results are somewhat dependent on the method of analyzing subjects lost to follow-up.

In Context

As levels of adherence <90% have been shown to be associated with decline in increases of HIV-1 RNA viral load and CD4 count,(1) and non-adherence is strongly associated with mortality,(2) establishing the most effective means for high adherence levels is of primary import. While reviews have shown some successful methods for increasing adherence,(3) the most effective intervention components for implementation in real-world settings are yet to be determined, especially in low-resource settings. mDOT has been demonstrated to be successful for HAART in Haiti and among injection drug users.(4,5) The adherence differences in the intervention arm reported here were small (3% to 7%) compared to the 10% difference felt to be clinically significant; however, one study reported a difference of just 5% may decrease the emergence of drug-resistant HIV with an NNRTI regimen.(6)

Programmatic Implications

The authors note that mDOT for HAART has the potential to increase adherence, and while they recommend its implementation, they note that substantial resources may be required. For example, restructuring of clinics to provide more patient-centered care through the provision of food and social services, and sustained commitment at the institutional level may be prerequisites. At the same time, the authors note that costs can be reduced by using peers for counseling and potentially by shortening the 6-week duration of the mDOT course. Less encouraging is that sustaining the improvement seen at 6 months to 12 months may have been an artifact of their method of analysis and no data were available for longer follow-up times beyond 12 months.


  1. Knobel H, Guelar A, Carmona A, Espona M, Gonzalez A, Lopez-Colomes JL et al. Virologic outcome and predictors of virologic failure of highly active antiretroviral therapy containing protease inhibitors. AIDS Patient Care STDS 2001 Apr;15(4):193-9.
  2. Bangsberg DR, Acosta EP, Gupta R, Guzman D, Riley ED, Harrigan PR et al. Adherence-resistance relationships for protease and non-nucleoside reverse transcriptase inhibitors explained by virological fitness. AIDS 2006 Jan 9;20(2):223-31.
  3. Simoni JM, Pearson CR, Pantalone DW, Marks G, Crepaz N. Efficacy of interventions in improving highly active antiretroviral therapy adherence and HIV-1 RNA viral load. A meta-analytic review of randomized controlled trials. J Acquir Immune Defic Syndr 2006 Dec 1;43 Suppl 1:S23-35.
  4. Farmer P, Leandre F, Mukherjee J, Gupta R, Tarter L, Kim JY. Community-based treatment of advanced HIV disease: introducing DOT-HAART (directly observed therapy with highly active antiretroviral therapy). Bull World Health Organ 2001;79(12):1145-51.
  5. McCance-Katz EF, Gourevitch MN, Arnsten J, Sarlo J, Rainey P, Jatlow P. Modified directly observed therapy (MDOT) for injection drug users with HIV disease. Am J Addict 2002 Fall;11(4):271-8.
  6. Gulick RM. Adherence to antiretroviral therapy: how much is enough? Clin Infect Dis 2006 Oct 1;43(7):942-4.