The use of wearable devices in the community to detect symptom severity and relapse of psychosis: A systematic review

BACKGROUND: Psychosis is a debilitating illness, marked by frequent relapses that are difficult to manage. Community mental health care can be reactive and inconsistent, but wearable devices offer the potential to monitor clinical status using biometric data (e.g. sleep-wake cycle, physical activity, heart rate). This study evaluates the feasibility of using wearable devices and examines if they can detect early warning signs of clinical deterioration and relapse in people with psychosis in the community setting. METHODS: A systematic review following Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines, consisting of 18- to 65-year-olds living in the community with psychosis. Only studies using a wearable device (research- or consumer-grade) with direct skin contact measuring biometric markers (e.g. actigraphy, electrodermal activity, heart rate variability) were included. RESULTS: Twenty-seven studies met the inclusion criteria, with 18 unique studies. Schizophrenia spectrum disorders were the predominant diagnoses, and the Positive and Negative Syndrome Scale was the most used clinical outcome measure. Most studies (12/18) were short-term (<28 days). Physical activity via actigraphy (n = 18) was most commonly examined, followed by sleep-wake cycle (n = 16) and heart rate (n = 13), with some integrating smartphone-based ecological momentary assessment (n = 8). Three unique studies attempted to detect relapse by combining wearable data with clinical data. Overall, most studies carried a serious risk of bias, primarily due to confounding and selective reporting. CONCLUSION: The use of wearable devices in community settings is feasible for monitoring the severity of psychotic symptoms and may detect relapse, though findings are limited by heterogeneity in wearable and clinical measures and study design.