Identifying risk factors for drug use recurrence with ecological momentary assessment, wearable technologies, and machine learning: a feasibility trial of peer recovery support specialist intervention - Summary - MDSpire
Advertisement
Identifying risk factors for drug use recurrence with ecological momentary assessment, wearable technologies, and machine learning: a feasibility trial of peer recovery support specialist intervention
To investigate the use of wearable technologies and ecological momentary assessment (EMA) to predict drug use recurrence (DUR) and explore opportunities for peer recovery support specialist (PRSS) interventions based on these predictions, including physiological and behavioral biomarkers.
Approach:
Participants and Data Collection: 229 participants were recruited from various settings and provided with a wearable device (Oura ring) and daily EMA prompts to assess mood and cravings over 90 days.
Randomization and Monitoring: Participants were randomized into Standard of Care (SoC) or PRSS Intervention arms and monitored for two additional 90-day phases, with alerts sent based on machine learning anomaly detection.
Key Findings:
108 participants provided EMA and Oura data for at least one of the 90 days.
The PRSS intervention arm showed significant decreases in anxiety, stress, depression, and maximum craving compared to the SoC arm (all p's < 0.001 for anxiety, stress, and depression; p = 0.011 for maximum craving).
The PRSS made 483 call attempts for unique alerts, averaging about 20 calls per participant.
Interpretation:
The study suggests potential benefits of using wearable devices and EMA for predicting DUR and the effectiveness of PRSS interventions in response to alerts.
Limitations:
Patient compliance and attrition were challenges that need to be addressed to optimize the approach, potentially affecting the generalizability of the findings.
Only a subset of participants provided sufficient data for analysis, limiting the robustness of the conclusions.
Conclusion:
The study highlights the feasibility of using technology to predict DUR and the role of PRSS in intervention, while identifying areas for improvement in future research.
by James J. Mahoney III, Victor S. Finomore, Jennifer L. Marton, Lucinda J. England, Sara McFoy, Danielle Romanoff, Jad Ramadan, Anahita Zarei, Amer Mahyoub, Jessie Crooks, James H. Berry, Steven D. Shirk, Manish Ranjan, Ali R. Rezai
In a survey of 420 Italian adults, psychological distress showed stronger associations than autistic traits with problematic internet and mobile phone use, although both were associated with higher digital-use scores.
