Feasibility of Using Wearables to Monitor Physical Activity and Sleep During Substance Use Disorder Treatment: Insights from a Longitudinal Study.

BACKGROUND:Wearable technologies enable objective measurement of physical activity (PA) and sleep in substance use disorder (SUD) treatment research, yet the feasibility of deploying wearables in real-world SUD service settings is poorly understood. Given high rates of treatment discontinuation (30%-50%), housing instability, and transience among individuals with SUDs, wearable loss may be substantially higher than in other populations. METHODS:We prospectively tracked SENS Motion wearables deployed to 63 participants across 8 SUD treatment and recovery service sites in Western Australia, spanning residential, continued-care, and community-based services, as part of an ongoing longitudinal study. Wearable loss was treated as a key feasibility outcome. An advisory group comprising individuals with lived experience of SUD recovery (n = 2), treatment service stakeholders (n = 3), and researchers (n = 4) contextualised observed challenges. Advisory group data were analysed using inductive qualitative content analysis. FINDINGS:Thirteen wearables (20.6%) were permanently lost, compared with approximately 1% in general population studies. Loss occurred almost exclusively in residential care settings (14.3%-42.9% across facilities) and coincided with participant exit from treatment. No losses were attributable to technical malfunction or deliberate non-compliance. Five higher-order categories were identified: treatment transitions as drivers of loss, diffuse responsibility for wearable return, ethical tensions around accountability, impact on future research, and practical mitigation strategies. CONCLUSIONS:Wearable-based PA and sleep measurement in SUD treatment settings is feasible but requires structural and procedural safeguards, including integration of wearable retrieval into discharge workflows and clarification of responsibility between research teams and service partners that explicitly anticipate high rates of unplanned treatment discontinuation.