Abstracts

Rationale: Medication nonadherence has been identified as a risk factor for morbidity and mortality in children with epilepsy and is one of the only modifiable risk factors for SUDEP. Despite these compelling reasons to assess for non-adherence, our only screening approach involved a single yes/no question embedded in our electronic health record (EHR) Epilepsy Visit SmartForm. Over 6 months including 1,039 visits, providers documented a 96% rate of adherence. Given the existing literature and our personal experiences, we suspected our screening method was inadequate.

Methods: This is a quality improvement project using the A3 framework to improve screening for medication nonadherence and improve intervention for patients and families reporting barriers or nonadherence. To assess the current state, we surveyed 36 providers and 100% wanted a screening tool. Thus, we implemented the 18-item Adherence Screening Tool created by the Epilepsy Learning Healthcare System (ELHS). Given the exclusive use telemedicine due to the COVID pandemic, we initially sent the screener through our EHR before appointments to patients with epilepsy taking anti-seizure medications. We tracked responses and implemented countermeasures. Our main intervention was transitioning to paper forms at some sites as in-person visits resumed. We also aimed to improve provider counseling at encounters in which barriers to medication non-adherence were reported. Our main intervention was implementation of the ELHS Barriers Toolkit which offers targeted counseling for specific barriers. We tracked repeat visits to obtain longitudinal data.

Results: We improved our rate of formal screening from 0% to 39% primarily by sending the barriers screener to patients through the EHR as a patient message. Implementation of paper screeners at in-person visits further improved our screening rate to 45% (Figure 1a). Introduction of the Barriers Toolkit increased provider counseling rate from 41% to 56% (Figure 1b). Of surveyed patients, 41% endorsed some form of nonadherence (compared to 3% in our baseline data with use of a single yes/no question). Additionally, 57% reported potential barriers. (Figure 2).

Conclusions: Implementation of the formal ELHS Adherence Screening Tool identified a much higher proportion of patients with barriers to nonadherence than previously recognized with our prior informal screening methods. By identifying specific barriers, providers could counsel in a targeted manner. We have documented individual cases of improvement in adherence and reduction in barriers throughout our study period, although further longitudinal data is needed. Future Plan-Do-Study-Act (PDSA) cycles include implementation of an automated EHR survey that links to counseling toolkits to help efficiently embed screening in clinic workflow.

Funding: Please list any funding that was received in support of this abstract.: N/A.