Thesis (MSc(Med)(Pharmacy)) -- University of Limpopo, 2010. / SUMMARY Introduction: The Human Immunodeficiency Virus (HIV) pandemic has become a global "monster" and much effort and funds have been channelled by various stake holders to change the deadly course of this threatening disease. Adherence has been identified as a critical element in optimal treatment of the disease with antiretroviral (ARV) medicines. A literature review was conducted on relevant facets of the HI virus, Acquired Immune Deficiency Syndrome (AIDS), disease prevention and treatment with ARVs, treatment obstacles, the importance of a sustained adherence level of at least 95% and the treatment interventions to promote medication adherence. Objectives: The objectives of this study were to evaluate the e-MuM electronic monitoring system (electronic microprocessor, reminder unit [a specialised wristwatch] and software program) for practicality, impact and effectiveness on ARV adherence, to test the ease of use for the patient and pharmacist, to get feedback from patients and clinic staff and finally to make recommendations concerning possible adaptations and the ideal use of the e¬MuM system. Method: The design was a four phase, prospective, randomised experimental, longitudinal study, conducted at the Tshepang Clinic of the Dr George Mukhari Hospital in Garankuwa using 210 volunteer patients.. After signing a consent form, patients were randomized into test and control groups, balanced according to gender and time on treatment. At the end of the baseline phase (only written adherence data collection), both groups were given their stavudine tablets in an electronic monitoring (e-MuM) container (from the end of the second month). The test group received interventions in the form of a alarm watch reminder unit (end of Phase 1), followed by visual (based on e-MuM generated graphs) and verbal feedback (end of Phases 2, 3 and 4). Other adherence tools used to evaluate and compare adherence in this study included a self-assessment questionnaire for gathering quantitative and qualitative data, visual analogue scale (VAS), 2-day and 7-day recall, tablet counts and the biological markers of the patients at the start and end of the study Results: There was no statistically significant difference between demographic data of the two groups at any point during the study. The mean time on treatment of the test group patients increased relative to baseline by the end of the study, which follows a reported trend that patients who have been on treatment for longer, tend to remain in studies. The e-MuM system revealed a large scattering of adherence results in both groups. Medication taken within an hour of the correct time was regarded as "strict" adherence and that taken at any time on the correct day as "lenient" adherence. The most significant increase in mean strict adherence was from Phase 2 to Phase 4 (after two verbal and visual feedback sessions). The adherence increase for the test group was 18.8% and 14.3% for the control group. The mean strict adherence level was 36.1 % for the test group and 29.8% for the control group for the full period. The mean lenient adherence for the full period was 45.5% for the test group and for the control group it was 36.6%. The difference between the two groups in adherence increase over time, did not reach the statistically significant level of P<0.05. One of the difficulties in the study data interpretation of the self-reported adherence was due to patieilts' decanting habits. Patients were asked to state whether and how often they removed their tablets from the original container into another container (decanting). Decanting habit options included daily, weekly, no decanting and unspecified decanting habits. Patients' decanting habits varied greatly. From decanting data available for a total of 209 patients, a majority (145) reported at least two different types of decanting habits for the study period, while five patients reported every decanting habit. Patients statements on their decanting habits were compared with the data recorded by the e-MuM system, but did not correlate at an individual level and were found not reliable enough to permit adjustment of the adherence levels that were calculated by the e-MuM system. Although self-assessment adherence measures exhibited a high degree of correlation, this was in stark contrast with the data obtained from the e-MuM. The mean adherence according to the Visual Analogue Scale (VAS) for the test group and control group was 87.7% and 88.4% respectively. Some of the study participants (11%) marked their adherence out of bounds on the 10cm solid line used for VAS. Despite frequent explanations from staff, some patients were not able to understand the abstract nature of the VAS. This observation may be related to patient educational level, as the majority of study participants (82%) had an incomplete secondary education or lower. In addition, clinic staff and data collectors commented that patients did not want to admit to being non¬ adherent. The mean stated adherence with the 7-day recall was 93.6% for the test group and 92.8% for the control group patients. The 2-day recall was omitted at the end of Phase 2. Adherence measured with tablet counts could not be used as it was only available in 60% of visits by test group patients and 64% of control group patients, as a result of patients not returning their remaining tablets at follow-up visits. Although positive tendencies in biological markers (CD4 and viral load [VL]) were evident towards the end of the study, differences between the groups did not reach statistical significance. The mean increase in CD4 count in the test group over the full period was 76.2 cells/mm3 and the number of patients in the test group with VL < 400 copies/ml increased from 72% to 89%. The mean increase in CD4 count in the control group was 72.2 cells/mm3 and the number of patients with VL < 400 copies/ml increased from 65% to 75%. Conclusion: The results of the study illustrated that the e-MuM system could be integrated in a normal clinic routine but additional staff and facilities (hardware) would be needed. The e-MuM system could be particularly helpful with new or suspected non¬ adherent patients. The disadvantage of the electronic monitoring system is its ability to monitor only one drug per container. It could be used with a fixed combination single tablet regimen. Patients were positive about the reminder unit as a tool to improve adherence. Doctors had mixed opinions of the usefulness of feedback graphs in monitoring adherence. Some patients disagreed with feedback results and this may reflect the anomalies caused by the range of decanting habits. The test group reached higher average rates of adherence than the control group, as jUdged bye-MuM recorded events, but differences were not statistically significant. The e¬ MuM data do not reflect adherence as such, merely container opening patterns, which makes it difficult to interpret results. In focus groups, patients and staff expressed their views about the use of the e-MuM system. No difficulty in the ease of use was reported by patients or staff. Notwithstanding this, a large portion of patients did decant tablets for various reasons, which made calculation of true adherence rates very difficult. A group of patients suggested routine use of the e-MuM system, to keep them conscious of their medication regimen while staff suggested limited use for new patients and non-adherent patients. The size of the container was the biggest obstacle according to patients and clinic staff and a small, more portable container was suggested. The sturdiness of the e-MuM lid with the microprocessor used for the study was questionable. Recommendations: Based on the results of this study, suggestions to increase adherence and utilise the e-MuM system are offered. The ideal medication dosage interval for patients, whose adherence patterns are being monitored electronically, would be once daily. For the e-MuM system to be practical, the device (container with embedded micro chip) has to be small, portable and sturdy. Patients will have to be educated to take every dose directly from the e-MuM container. For optimal e-MuM data interpretation, patient medication taking behaviour, including decanting of tablets must be accurately identified. A trained, dedicated, sensitive person has to interpret data and give feedback to patients.