Skip to main content
ABCT 2024 58th Annual Convention Main banner
Psychology Tools Ad
Final Program


Icon Legend
This session is not in your schedule.
This session is in your schedule. Click again to remove it.
Presentation Icons
Recorded Session
Recorded Session
Ticketed Session
Ticketed Session
Back

Symposia

Technology/Digital Health

Broadening the Reach of Passive Sensing: Insights and Lessons Learned from Clinical Researchers

3 - (SYM 10) Using Continuous Glucose Monitoring and Machine Learning to Passively Detect Naturalistic Binge Eating and Vomiting Among Adults with Binge-spectrum Eating Disorders: A Preliminary Investigation

Friday, November 15, 2024
8:30 AM – 10:00 AM EST
Location: Liberty B, Level 3, Marriott Downtown
Abstract Body
Binge eating and purging (e.g., self-induced vomiting) are common, transdiagnostic eating disorder (ED) symptoms associated with substantial morbidity. Efforts to understand these behaviors in research and clinical settings have historically relied on self-report measures, which may be biased and have limited ecological validity. It may be possible to passively detect binge eating and vomiting using data collected by continuous glucose monitors (CGMs; minimally invasive wearable sensors that measure blood glucose levels), as these behaviors yield characteristic glucose responses. The present study developed machine learning classification algorithms to detect engagement in binge eating and vomiting among 22 adults with binge-spectrum EDs (i.e., binge eating disorder or bulimia nervosa) using CGM data. Participants wore Dexcom G6 CGMs and reported all eating episodes and disordered eating symptoms using event-contingent ecological momentary assessment for two weeks. Group-level random forest models were generated to distinguish binge eating from typical meals and snacks and to detect instances of vomiting. The binge eating model had accuracy of 0.85 (95% CI: 0.80, 0.89), sensitivity of 0.65, and specificity of 0.92. The vomiting model demonstrated accuracy of 0.71 (95% CI: 0.53, 0.85), sensitivity of 0.71, and specificity of 0.71. Results suggest that CGM may be a promising avenue for passively detecting binge eating and vomiting, with implications for innovative research and clinical applications. Additional research is needed to further optimize algorithms, evaluate their performance in larger, more diverse, and more representative samples, and evaluate their utility for improving clinical outcomes via delivery of ecological momentary interventions.