Abstracts

Rationale: To examine AED adherence and to explore the relationship between episodic memory performance and adherence among ethnically-diverse people with epilepsy (PWE).Methods: Data were derived from an ongoing study at our institution on the neurobehavioral correlates of AED adherence. Ethnically diverse PWE (N=58; 57% Female; 53% African-American, 24% Caribbean-American, 20% Latino) had AED adherence electronically monitored for one month (Medication Event Monitoring System; MEMS). As part of a comprehensive assessment, patients completed the Morisky 8-item scale of medication adherence and rated their ability to take AEDs as prescribed over the past week, two weeks, and one month. Episodic memory was assessed using Hopkins Verbal Learning Test – Revised (HVLT-R). Adherence results and HVLT-R performance were characterized, and their relationships were examined.Results: On average, participants took AEDs as prescribed 72% of the time (percent days of correct dosing based on MEMS). Self-reported adherence on the Morisky significantly correlated with electronically monitored adherence (r = -0.60, p < 0.01). Patients’ self-ratings of their abilities to take AEDs as prescribed over the one-month monitoring period also significantly correlated with MEMS data (r = 0.56, p < 0.01). Twenty-eight percent of patients rated their adherence over the preceding month as “good,” “very good,” or “excellent,” but were less than 80% adherent based on MEMS data (i.e., over-reporters). Similarly, 31% over-reported adherence for the previous one week and two weeks. HVLT-R performance indicated borderline impaired episodic memory (mean age-corrected Immediate Recall T=32, SD=10; Retention T=33, SD=13). When relationships between adherence, its reporting, and episodic memory performance were explored, the difference between accurate and inaccurate reporters on HVLT-R Retention did not reach significance (p = 0.07). Lower scores on HVLT-R Immediate Recall (age-corrected) were significantly associated with higher frequency of extra doses taken (r = -0.28, p < 0.05). The association between Retention and MEMS percent days of correct dosing did not reach significance (r = 0.26, p = 0.06); however, this may be due to small sample size.Conclusions: Preliminary findings indicated significant relationships between self-reported and electronically monitored AED adherence, although many patients over-reported their adherence. PWE in our sample had borderline impaired episodic memory, which did not differentiate accurate and inaccurate reporters. Findings did not suggest an optimal interval for retrospective ratings of adherence. Poor immediate recall was significantly related to higher frequency of extra doses taken, whereas the relationship between retention of information and adherence did not reach statistical significance. The role of episodic memory requires further investigation with a larger sample, as understanding its effects on adherence could inform behavioral interventions.