University of California, San Francisco Logo

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

Electronic measurement of adherence to pediatric antiretroviral therapy in South Africa
Global Health Sciences Literature Digest
Published July 10, 2008
Journal Article

Muller A, Bode S, Myer L, Roux P, von Steinbuchel N. Electronic measurement of adherence to pediatric antiretroviral therapy in South Africa. AIDS 2008;27:257-62.

In Context

Adherence to antiretroviral therapy (ART) is necessary for optimal effect,(1) and poor adherence has been associated with drug resistance and virologic failure.(2) Studies of ART adherence among children in resource-constrained areas have not been evaluated previously.


To evaluate adherence to a liquid formulation of ART using the Medication Event Monitoring System (MEMS) among outpatient children, to compare adherence measured through MEMS and through self-report from caregivers, and to correlate both adherence measures with viral loads

Study Design

Prospective cohort study.


An outpatient HIV clinic in Cape Town, South Africa that serves a predominantly indigent population.


Children aged <10 years who were attending the clinics and who were receiving liquid formulation of ART and each child's caregiver. Children who had more than one caregiver or who lived in an orphanage were excluded.

Main Outcome Measure

Median adherence measured by MEMS and caregiver report, the proportion of children who were >95%, >90%, >80%, and <60% adherent, sensitivity and specificity rates for detecting suppressed viral load for both adherence measures, and predictors of adherence.


All eligible children and their caregivers seen at the clinic between February and April 2006 were invited to participate. The characteristics of the children were obtained from their medical records. Caregiver characteristics were obtained from a self-administered questionnaire conducted at baseline. ART bottles were fitted with MEMS caps and supplied to participants at the pharmacy. Caregivers received written instructions regarding use of MEMS bottles and were directed to return the bottles monthly. Caregivers were asked to rate adherence using a Visual Analogue Scale (VAS) on a scale of zero to 100%. CD4% and viral load were measured at baseline and six months. Viral loads <50 copies/mL (log<1.49) were defined as undetectable.


Seventy-eight consecutive eligible patients were selected. The results from five patients were excluded because of transfer of care and technical problems with the MEMS. The children's median age was 48 months and median duration of ART use was 28 months. Mean baseline CD4% was 24% (SD 0.9) and mean log viral load was 3.29 (SD 0.2) as measured from 72 of the patients (two additional children had insufficient specimen).

MEMS monitoring lasted for a mean of 81 days (interquartile range [IQR] 69-87) and median MEMS adherence was 87.5% (IQR 69-97). MEMS adherence during the first, second, and last month of monitoring did not differ substantially. Mean adherence reported by caregivers using VAS was 98.6% (SD 0.3%) with a median of 100%. The mean difference between MEMS and VAS adherence was 15% (SD 2.3%). Adherence of >95% from MEMS and VAS differed (36% of patients compared to 91% of patients, respectively). Sixty-five percent of viral loads were undetectable at the end of the study. Adherence as measured through MEMS was significantly correlated with HIV suppression (P=0.01). When MEMS adherence was >95%, the sensitivity of virologic suppression was 48% (95% confidence interval [CI] 32-63) and the specificity was 83% (95% CI 62-95). Sensitivity increased and specificity decreased when adherence levels were lower.


Adherence to ART in children in South Africa is comparable to adherence rates in developed countries. Caregiver report of adherence is not a reliable measure of ART adherence in children.

Quality Rating

This study is of very good quality in that it compared two measures of adherence with one (MEMS) regarded as the gold standard. Adherence using MEMS does not allow for measurement bias by investigators in that it is computerized. Follow-up was sufficient to measure the outcome although longer studies of adherence would be beneficial. A large proportion of enrolled subjects completed the study. However, the cohort was derived from a single clinic in Cape Town, South Africa and may not be generalizable.

Programmatic Implications

This study found that adherence to ART in children was as high as has been reported among children in the United States(3) and among adults in resource limited areas.(4) Adherence to ART is an essential component of HIV care. Providers carefully consider whether ART should be prescribed to persons who are not likely to be compliant because the development of resistant HIV strains may outweigh any clinical benefit if adherence is not sufficiently high. The authors were able to demonstrate that self-report of ART adherence by children's caregivers was highly overestimated and is not likely to be reliable.


  1. Watson DC, Farley JJ. Efficacy of and adherence to highly active antiretroviral therapy in children infected with human immunodeficiency virus type 1. Pediatr Infect Dis J 1999 Aug;18(8):682-9.
  2. Chesney MA, Ickovics J, Hecht FM, Sikipa G, Rabkin J. Adherence: a necessity for successful HIV combination therapy. AIDS 1999;13 Suppl A:S271-8.
  3. Steele RG, Anderson B, Rindel B, et al. Adherence to antiretroviral therapy among HIV-positive children: examination of the role of caregiver health beliefs. AIDS Care 2001 Oct;13(5):617-29.
  4. Orrell C, Bangsberg DR, Badri M, Wood R. Adherence is not a barrier to successful antiretroviral therapy in South Africa. AIDS 2003 Jun 13;17(9):1369-75.