METHOD
A multicenter observational and descriptive study carried out in southwestern part of Nigeria among adults PWE after due ethical clearance has been obtained. The estimation of sample size was done by using the single population proportional formulation by taking 5% margin of error, 95% confidence level and proportion of reported adverse events as 0.5%. Using convenient sampling method, we included 940 consenting participants aged 16 years and above attending epilepsy clinics for period of 5years with diagnosis confirmed by International League against Epilepsy(ILAE) criteria and supported by Electroencephalography(EEG).(Brodie et al., 2018; Fisher, 2017; Hirsch et al., 2013; Scheffer et al., 2017) The case records of all patients attending/that attended neurology clinics in the last 5 years with diagnosis of epilepsy in the selected hospitals was reviewed with the use of a semi structured questionnaire, to extract information about clinical and socio demographic characteristics. These include age, age of onset, seizure characteristics, type, and duration of AEDs.
LEAP, a previously validated instrument was used to assess ADR. It is a 21- item self-reported questionnaire, designed to measure AEDs side effect which covers both Central Nervous System (CNS) and Non-CNS related adverse effects. The total score of LAEP ranges from 19 to 76, with higher score indicating more adverse effects burden. Scores ≤45 indicates mild to moderate adverse effects and reaches the toxic level if the total score exceeds 45.(Fadare et al., 2018). (Fadare et al., 2018; Sakuma et al., 2014; Yang et al., 2014). MMAS a previously validated and extensively used adherence instrument was used to measure medication adherence. There are five expected responses in MMAS namely; never, rarely, sometimes, often and always respectively which are scored as 0, 1, 2, 3, 4 respectively and subsequently subdivided into 0 (high adherence), 1-2 (medium adherence) and >2 (poor adherence). These instruments were administered by Neurologist and trainees who are qualified Doctors at participating Centers. Data were cleaned, coded, and analyzed using the IBM Statistical Package of Social Sciences Version 23. Socio-demographic and clinical characteristics of patients were presented as frequency (percentage). Using Pearson chi square test. Independent T-test was used to examine the possible association between ADR with gender, AED therapy (monotherapy and polytherapy) and medication adherence. One-way ANOVA test was used to assess the association between type of seizure, duration of epilepsy and epileptiform pattern. The significance level of statistical measures was set at p<0.05.