Abstracts

Rationale: HOBSCOTCH is an evidence-based self-management intervention which targets cognitive dysfunction in adults with epilepsy. The program is delivered by a Cognitive Coach via telehealth. In the HOBSCOTCH-3 randomized controlled trial a smartphone App is incorporated to encourage medication adherence, seizure tracking and use of learned memory strategies. Our aim is to drive improvements in participants’ health behaviors with the support of a mobile app and observe how use relates to key health outcomes.

Methods: Adults with epilepsy (n = 106) were enrolled in the HOBSCOTCH 3 RCT and were trained by Cognitive Coaches to submit logs daily using the HOBSCOTCH mobile app that allows for self-reporting seizure frequency, medication adherence, memory strategy use, and well-being. To encourage daily submissions, the mobile app reminded participants to submit logs every day via notifications, and study coordinators provided technical app assistance as needed. Before and after completing the HOBSCOTCH intervention, participants completed surveys measuring QOL (QOLIE-31), depression (PHQ-9), cognitive function (Neuro-QOL), epilepsy self-management behaviors (AESSMI), and medication adherence (MARS).

Results: Participants’ app activity during the 9-week intervention revealed 46.2% of participants submitted logs for at least 90% of days during their intervention (i.e., intervention period) and 85.6% of participants submitted logs for at least 50% of their intervention period. An average of 49.0 daily logs (SD 19.8), out of a possible 56 days, were submitted by participants during their intervention period (Figure 1). A correlation matrix calculating Spearman rho was calculated using participant post-intervention outcome survey data, app activity, and app data (self-reported responses) (Figure 2). Of specific note, self-reported seizures was correlated with PHQ-9 positively [rho = 0.32, p = 0.003] and QOLIE-31 negatively [rho = -0.35, p = 0.001], while self-reported mood was correlated with PHQ-9 negatively [rho = -0.54, p < 0.001] and QOLIE-31, and NEURO-QOL positively [rho = 0.46, p < 0.001], [rho = 0.37, p = 0.001]. Self-reported medication was positively correlated with AESSMI [rho = 0.33, p = 0.002] and MARS [rho = 0.24, p = 0.029], app activity was positively correlated with self-reported memory strategy usage [rho = 0.25, p = 0.023], and self-reported mood was negatively correlated with self-reported seizures [rho = -0.33, p = 0.002].

Conclusions: This analysis supports that user engagement with the HOBSCOTCH app during intervention is high. App and outcome measure correlations demonstrate that more frequent use of the app encouraged the use of memory strategies learned during the HOBSOTCH intervention. Responses to app logs correlated to QOL, depression severity, cognitive function, epilepsy self-management behaviors, and medication adherence in this cohort of study participants. These preliminary results help validate both HOBSCOTCH and the mobile app’s ability to promote healthy self-management behaviors and improved outcomes.

Funding: Centers for Disease Control and Prevention U48 DP006377-01-00.