Patients’ Adherence to Antiretroviral Therapy in Lagos State University Teaching Hospital (LASUTH), Ikeja, Lagos, Nigeria.

Aderemi-Williams, R.I (2014)

A Thesis Submitted to the School of Postgraduate Studies, University of Lagos.


Evidence abound that successful management of HIV and AIDS using Highly Active Antiretroviral Therapy (HAART) requires not less than 95% adherence. Longitudinal studies measuring adherence using four different tools have not been documented in Africa. The use of different tools to determine and predict factors that may affect adherence in Nigeria has not been documented to the best of our knowledge. A study (comprising of 3-month retrospective and 45-month prospective) was carried out in Lagos State University Teaching Hospital (LASUTH) between December, 2006 to December, 2010. Validated questionnaires and monitoring of patients’ clinical progress were used. Adherence was assessed using self-report, unannounced pill counts, doctor’s appointment attendance, and pharmacy refill record. Patients with 95% and above adherence levels were considered adherent. Data collected from 248 of the 294 (84.35%) patients eligible for the study were analysed. ANOVA was used to compare and determine the adequacy of the adherence assessment tools. Predictors of adherence were determined with logistic regression models and paired sample t-test was used to test effect on patients’ CD4 count. Social and clinical demographic data analysis of study patients show that 134 (54.00%) were married, 58 (23.40%) were single; 148 (59.70%) females, 100 (40.30%) males; 211 (85.10%) were Christians, 31 (12.50%) were Muslims; 106 (42.70%) had secondary school education, 67 (27.00%) had tertiary education, 43 (17.3%) had primary education and 139 (56.05%) were employed. Their mean age was 40.39 ±8.78 years and mean baseline CD4 cell count was 143.46 ±92.72 cell/µL. At baseline, 66.10% of the patients were on 12- hourly regimen while 33.90% were on 24-hourly regimen. Unannounced pill count (mean = 96.97±6.02) measure was found to be the best measure for patient’s adherence, followed by pharmacy refill (mean = 96.64±6.95), self-report (mean = 95.26±15.51) and doctor’s appointment attendance (mean = 74.70±14.93) measure. None of the variables studied was a predictor of adherence using self-report as a measure of adherence. Using unannounced pill counts: younger age (18-39 years), being Muslim and patients’ knowledge of transmission were predictors of adherence. In doctor’s appointment attendance measure of adherence; knowledge of management, knowledge of adherence, knowledge of HIV pathogenesis and patients’ habits predicted adherence. The following are predictors of adherence when measured with pharmacy refill record: marital status, Muslim religion, tertiary education, being a professional, skilled worker, knowledge of drug management, knowledge of adherence and presence of social support. The effect of highly active antiretroviral therapy (HAART) on improving the immunological status of patients was significant. Adherence was dynamic over the study period. The study was also able to identify many possible predictors of adherence with pharmacy refill record identifying most predictors. Due to identified gaps in patients’ CD4 count monitoring, the difference in immunological status between adherent and non-adherent patients was not significant.