Accepted for/Published in: JMIR Mental Health
Date Submitted: Dec 10, 2020
Date Accepted: May 10, 2021
Date Submitted to PubMed: Aug 12, 2021
Clinical Utility of Wearable Sensors and Patient-Reported Surveys in Patients With Schizophrenia: A Non-Interventional, Observational Study
ABSTRACT
Background:
Relapse in schizophrenia may be preceded by early warning signs of biological, sensory and clinical status. Early Detection of warning signs may facilitate intervention and prevent relapses.
Objective:
This study aimed to investigate the feasibility of using wearable devices and self-reported technologies to identify symptom exacerbation correlates and relapse in patients with schizophrenia.
Methods:
In this observational study, patients with schizophrenia were provided with remote-sensing devices to continuously monitor activity (Garmin vivofit) and sleep (Philips Actiwatch); smartphones were used for recording patient-reported outcomes. Clinical assessments on symptoms (Positive and Negative Syndrome Scale [PANSS], Brief Psychiatric Rating Scale [BPRS] were performed bi-weekly; other clinical scales on symptoms (Clinical Global Impression–Schizophrenia [CGI-SCH]), Calgary Depression Scale [CDS]), psychosocial functioning, physical activity (Yale Physical Activity Survey [YPAS]) and sleep (Pittsburgh Sleep Quality Index [PSQI]) were assessed every 4 weeks. Patients were observed for 4 months. Correlations between clinical assessments and aggregated device metrics data were assessed using mixed-effect model. Elastic net model was used for predicting clinical symptoms based on the device features.
Results:
Of 40 patients enrolled, one patient relapsed after being stable with evaluable post-baseline data. Across the cohort, biweekly clinical assessments on symptoms (PANSS and BPRS) showed a strong correlation (r = 0.97) while CGI-SCH did not strongly correlate with PANSS or BPRS total scores. Weekly patient reported outcomes moderately correlated with psychiatric symptoms (BPRS total score, r = 0.29; CDS total score, r = 0.37; PANSS total score, r = 0.3). In the elastic net model, sleep and activity features derived from Philips ActiGraph and Garmin were predictive of the sitting index of YPAS and sleep duration component of PSQI.
Conclusions:
The study demonstrated that wearable devices and smartphones could be effectively deployed and potentially utilized to monitor patients with schizophrenia. Metrics-based prediction models could assist in detecting earlier signs of symptom changes. The operational learnings from this study may provide insights to conduct future studies. Clinical Trial: NCT02224430
Citation
Request queued. Please wait while the file is being generated. It may take some time.
Copyright
© The authors. All rights reserved. This is a privileged document currently under peer-review/community review (or an accepted/rejected manuscript). Authors have provided JMIR Publications with an exclusive license to publish this preprint on it's website for review and ahead-of-print citation purposes only. While the final peer-reviewed paper may be licensed under a cc-by license on publication, at this stage authors and publisher expressively prohibit redistribution of this draft paper other than for review purposes.