Adherence to HAART therapy measured by electronic monitoring in newly diagnosed HIV patients in Botswana
EUROPEAN JOURNAL OF CLINICAL PHARMACOLOGY
Leiden University, VU University Amsterdam, Ardex Ltd, CHDR, Nyangabgwe Hosp, Pharmion Syst Ltd
Aims This pilot study was designed to evaluate the feasibility and benefits of electronic adherence monitoring of antiretroviral medications in HIV patients who recently started Highly Active Anti Retroviral Therapy (HAART) in Francistown, Botswana and to compare this with self-reporting.
Methods Dosing histories were compiled electronically using Micro Electro Mechanical Systems (MEMS) monitors to evaluate adherence to prescribed therapies. Thirty patients enrolled in the antiretroviral treatment program were monitored over 6 weeks. These patients were all antiretroviral (ARV) naive. After each visit (mean three times) to the pharmacy, the data compiled by the monitors were downloaded. Electronic monitoring of adherence was compared to patient self-reports of adherence.
Results The mean individual medication adherence level measured with the electronic device was 85% (range 21-100%). The mean adherence level measured by means of self-reporting was 98% (range 70-100%). Medication prescribed on a once-a-day dose base was associated with a higher adherence level (97.9% for efavirenz) compared with a twice-a-day regimen (88.4% for Lamivudine/Zidovudine).
Conclusions It is feasible to assess treatment adherence of patients living in a low resource setting on HAART by using electronic monitors. Adherence, even in the early stages of treatment, appears to be insufficient in some patients and may be below the level required for continuous inhibition of viral replication. This approach may lead to improved targeting of counselling about their medication intake of such patients in order to prevent occurrence of resistant viral strains due to inadequate inhibition of viral replication. In this pilot study a significant difference between the data recorded through the electronic monitors and those provided by self-reporting was observed.