Wearable-Based Detection and Prediction of Depression in Autistic Individuals Using Physical Activity and Sleep Data: A Scoping Review

Depression is one of the most prevalent and debilitating mental health comorbidities in autistic individuals, with estimated prevalence rates ranging from 23% to 54% — significantly higher than in the general population. Despite this, depression in autistic individuals remains critically underdiagnosed. Standard self-report tools (e.g., PHQ-9) are often poorly suited to this population due to alexithymia, atypical emotional expression, and differences in social communication, creating a significant clinical gap in early detection. Wearable devices (e.g., actigraphs, smartwatches, fitness trackers) offer a promising, non-intrusive approach to passively capturing behavioral signals associated with depression — including physical activity levels, sedentary behavior, sleep disturbances, and circadian rhythm disruption. While wearable-based depression detection models have advanced rapidly in the general population, their applicability, validity, and clinical utility in autistic individuals remains largely unknown. This scoping review aims to systematically map the current evidence on wearable-derived physical activity and sleep data for depression detection, prediction, and monitoring in autistic individuals of any age. Following JBI scoping review methodology and reported according to PRISMA-ScR guidelines, we will search six major databases (MEDLINE, PsycINFO, CINAHL, Embase, IEEE Xplore, Web of Science) with no date restriction. Specifically, this review will: (1) identify which physical activity and sleep features have been examined as digital biomarkers of depression in autistic individuals; (2) evaluate the types and performance of detection and prediction models applied; (3) examine whether and how detection outputs have been linked to digital interventions (e.g., just-in-time adaptive interventions, JITAIs); and (4) map key research gaps to guide future primary studies. Expected outcomes include a comprehensive evidence map, identification of the most promising digital biomarkers for autism-specific depression detection, and a set of prioritized research recommendations to support the development of personalized, wearable-based mental health interventions for autistic individuals.