Study protocol Effects of Philips Visual Patient Avatar on vital sign deviations and audible alarm burden in perioperative care: a dual-centre, quasi-experimental pre-post big-data study protocol (NewYork-Presbyterian/Weill Cornell and University Hospital Zurich)

Introduction

Perioperative patient monitoring requires clinicians to integrate multiple physiological data streams under time pressure and frequent interruptions. Conventional monitors predominantly present vital signs as separate numerical values and waveforms, which must be sequentially interpreted and mentally integrated, imposing substantial cognitive demands. Audible alarms are intended to enhance safety but contribute to alarm fatigue and increased workload. Time spent outside predefined safe ranges for key physiological variables and excessive alarm burden are associated with adverse outcomes, motivating approaches that support earlier detection and improved situation awareness without increasing cognitive load.

The Philips Visual Patient Avatar is an avatar-based visualisation technology displayed on the patient monitor that supports clinicians’ situation awareness by integrating multiple vital signs and sensor states into a single animated virtual patient, while retaining conventional numerical displays.

Although laboratory, simulation and qualitative studies suggest benefits of avatar-based monitoring, its impact on objective monitoring outcomes has not been systematically quantified.

Methods and analysis

This investigator-initiated, dual-centre, quasi-experimental pre–post study evaluates the clinical introduction of VPA using routinely collected perioperative monitoring data. The study is conducted at NewYork-Presbyterian Hospital / Weill Cornell Medical Center (New York, USA) and University Hospital Zurich (Zurich, Switzerland).

At each centre, data are collected across three phases: pre-implementation, a sink-in (adaptation) phase during staged installation and training, and post-implementation. The unit of analysis is the individual anaesthesia case. Outcomes are analysed retrospectively.

The primary outcome is vital-sign deviation time, normalised to 60 minutes of case time, defined as cumulative time outside predefined upper or lower thresholds for each vital sign. Secondary outcomes include a time-based area-under-the-curve metric capturing magnitude and duration of deviations, as well as audible alarm burden (alarm count, alarm time and alarm reaction time). Multivariable regression models will estimate associations between VPA implementation phase and outcomes while adjusting for relevant patient- and procedure-related covariates. Analyses will be performed separately for each centre, with sensitivity analyses as appropriate.

Ethics and dissemination

The study was approved by the Institutional Review Board of Weill Cornell Medicine. For Zurich, a declaration of non-jurisdiction was issued for the analysis of anonymised routinely collected data. At the time of protocol submission, data collection is complete at University Hospital Zurich and ongoing at NewYork-Presbyterian Hospital / Weill Cornell Medical Center; no analyses have been performed. Findings will be disseminated through peer-reviewed publications and scientific conferences.

Trial registration

Not applicable (observational analysis of routinely collected data).

Strengths and limitations of this study

• This dual-centre study leverages large volumes of routinely collected perioperative monitoring data from two tertiary academic hospitals on different continents, enhancing external validity across diverse clinical settings.

• The quasi-experimental pre–post design enables evaluation of the Visual Patient Avatar under routine clinical conditions but does not permit causal inference.

• A dedicated sink-in phase was included to mitigate learning and adaptation effects; however, residual confounding by time and unmeasured factors cannot be fully excluded.

• Differences in alarm configurations and visualisation thresholds between centres reflect local standards of care and limit direct standardisation, while allowing assessment of the intervention across heterogeneous real-world alarm environments.

• Outcomes focus on process-level monitoring and alarm metrics rather than hard patient outcomes; however, deviations in vital signs and alarm burden are supported by prior evidence as clinically meaningful indicators of patient safety. If improvements are observed, this study will provide the basis for future investigations evaluating patient-centred clinical outcomes.