Comparison of self-report and pharmacy pill count as measures of adherence to antiretroviral therapy in children at Kenyatta National Hospital

Abstract

Background Assessment of adherence is an essential component of monitoring antiretroviral therapy. In Kenyatta National Hospital, self7caregiver reported adherence is used to assess adherence. It is not known how this subjective method of measuring adherence compares with the more objective pharmacy pill count method. Objective The objective of this study was to compare caregiver reported adherence and pharmacy pill count adherence and to correlate both methods with the patients’ clinical and immunological outcome and to describe the characteristics (sociodemographic, economic and clinical) associated with adherence. Study Methods: This was a cross sectional study whereby caretakers of children who presented to the routine paediatric HIV clinic from November 2007 to February 2008 were subjected to a

structured interview to establish self-reported adherence for the two weeks prior to the clinic day. The patients’ pharmacy pill count records for the prior month were abstracted from the MSH-ART dispensing tool and these were used to compute adherence rates. CD4 counts, percentages and anthropometric measurements were abstracted from the patients’ clinical records. Adherence rates by the two methods were compared and then correlated with the patients’ response to treatment.

Results Mean adherence rate by pharmacy pill count for 1 month was 61% and 87% by caregiver reported adherence by use of 2-week recall. There was a significant difference when the two adherence rates by the two methods were compared ip <0.001). We did not find a difference in the clinical and immunologic response when patients were classified as adherent and non-adherent by either pharmacy pill count or self report. The caregiver and child’s sociodemographic characteristics did not influence adherence. However we did draw firm conclusions on our findings on associations of adherence and patient response and on the correlates of adherence because this study did not have adequate power to answer these questions. Conclusions Caregiver report gave higher adherence rates compared to pharmacy pill count. There was no difference in clinical and immunological parameters between patients who were adherent and those who were not adherent. Recommendation There is need to incorporate pharmacy pill count as a method of assessing adherence during the routine clinical care of HIV-infected children at the KNH paediatric CCC.