Currently submitted to: JMIR Formative Research
Date Submitted: Aug 25, 2026
Open Peer Review Period: Sep 9, 2026 - Nov 4, 2026
(currently open for review)
Warning: This is an author submission that is not peer-reviewed or edited. Preprints - unless they show as "accepted" - should not be relied on to guide clinical practice or health-related behavior and should not be reported in news media as established information.
Merging Visualizations, Statistics, and Clinical Interpretation in Digital Biomarker Development for Delirium: An Observational Study
ABSTRACT
Background:
In hospitalized older patients, delirium is highly prevalent and harmful. Delirium prevention efforts are limited by our restricted ability for early detection and incomplete understanding of delirium as a process. Digital biomarkers promise a deeper and more continuous understanding of delirium and its precursor signals, however several challenges from diversity in clinical presentation to confounding by medication limit progress.
Objective:
In this registered single center study we aim to investigate the suitability of wearables for delirium research in hospitalized patients using a prespecified three level analytic approach including group, individual, and visual analysis.
Methods:
Inpatients over 65 years were prospectively recruited upon hospital admission to an internal medicine ward, equipped with wrist-worn actigraphy devices, and followed until discharge. Delirium Observation Scale (DOS) scores were assessed three times per day, and delirium diagnosis was based on retrospective chart review.
Results:
Final analyses included 11 patients who developed delirium and 24 who did not (n=35). Patients with delirium were older and stayed longer in the hospital. Sensor data completeness was 60.7% in the delirium group and 57.4% in the non-delirium group. Group-level comparisons did not reveal any significant differences between the investigated delirious and non-delirious patients using digital biomarkers. Within-person comparison of delirium and non-delirium phase showed differences in some markers, though they varied between patients. Visual analysis was able to exemplify characteristics of a delirium patient journey.
Conclusions:
Group-level comparisons of wearable sensor based digital biomarkers were limited in their ability to characterize delirium in older patients. Visual analysis was better able to integrate digital biomarkers and medication information, with fewer requirements regarding data availability and circadian rhythm length. While there are clear downsides regarding the certainty of observed patterns and generalizability in visual analysis, they are valuable and should receive more attention in digital biomarker development, especially in observational studies of complex conditions. Clinical Trial: NCT05677646
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.