Co-construction of Dynamic Temporal Visualization Tools in the Electronic Health Record to Enhance User Experience and Clinical Decision-Making

Core Concepts

The ERIOS project aims to co-create components of the Electronic Health Record (EHR) with end-users, focusing on developing dynamic temporal visualization tools to improve user experience and support clinical decision-making.

Abstract

The ERIOS project is a collaborative initiative between a healthcare software company (Dedalus), a university hospital (CHU de Montpellier), and the University of Montpellier. The project focuses on integrating research and development (R&D) within the hospital setting to co-create components of the Electronic Health Record (EHR) with end-users.
The project started with two initial use cases:

"IsoPsy" for monitoring therapeutic isolation in psychiatry
"AtbViz" for managing anti-infective treatment sequences

Through a participatory design approach, the team analyzed user needs and academic recommendations on user engagement, human-computer interaction, and data visualization. This led to the development of dynamic temporal visualization components integrated into specific dashboards.
For the "IsoPsy" use case, the team created a timeline-based visualization that allows users to easily track tasks, identify upcoming deadlines, and anticipate the need for task completion. The dashboard also includes features for task prioritization and filtering by user role.
For the "AtbViz" use case, the team developed synchronized timeline and graph components to visualize a large amount of heterogeneous data (prescriptions, laboratory results, clinical observations) in a thematic and temporally-aligned manner, supporting users in tasks such as monitoring therapies, evaluating effectiveness, and assessing treatment tolerance.
The application of user-centered design principles and academic recommendations significantly enhanced the team's ability to meet user needs and improve the usability of the EHR system.

Stats

A hospital physician spends an average of 5.3 hours per day using the EHR, compared to only 1.3 hours in direct patient contact, accounting for about 37% of their daily effective work time.
The current EHR design often requires users to navigate through numerous tabs and screens to correlate relevant data, leading to information fragmentation and cognitive fatigue.

Quotes

"The current work process is inefficient, mainly due to a fragmented and dispersed task tracking across multiple communication channels."
"To evaluate a patient's situation and decide on an anti-infective therapeutic approach, an infectious disease specialist must navigate through an average of 70 screens in the EHR."

Key Insights Distilled From

ERIOS

by Loui... at arxiv.org 04-02-2024

https://arxiv.org/pdf/2404.00061.pdf

Deeper Inquiries

How can the dynamic temporal visualization tools be extended to support other clinical workflows and decision-making processes beyond the two initial use cases?

The dynamic temporal visualization tools developed in the ERIOS project can be extended to support other clinical workflows and decision-making processes by following a systematic approach. Firstly, conducting thorough user research and needs analysis across different clinical specialties will help identify common patterns and requirements for dynamic visualization. This will involve engaging with healthcare professionals from various departments to understand their workflow challenges and information needs.
Secondly, the co-creation methodology used in the ERIOS project can be replicated in other healthcare settings to involve end-users in the design and development process. By collaborating closely with clinicians, nurses, and other healthcare staff, the dynamic visualization tools can be tailored to specific clinical workflows and decision-making contexts. This approach ensures that the tools are user-friendly, intuitive, and aligned with the needs of the end-users.
Furthermore, the integration of the dynamic visualization tools with existing Electronic Health Record (EHR) systems and clinical decision support systems is crucial for seamless adoption and interoperability. By leveraging interoperability standards such as HL7 FHIR, the visualization tools can access data from multiple sources and present a comprehensive view of patient information. This integration will enable healthcare providers to make informed decisions based on real-time data and trends, improving patient outcomes and operational efficiency.

What are the potential challenges and limitations in scaling the co-creation approach to other healthcare organizations and EHR systems?

Scaling the co-creation approach to other healthcare organizations and EHR systems may face several challenges and limitations. One of the primary challenges is the variability in organizational culture, workflows, and information systems across different healthcare settings. Each organization may have unique requirements and processes, making it challenging to implement a one-size-fits-all co-creation approach.
Another challenge is the resistance to change and the need for extensive training and support for healthcare professionals to adopt new tools and workflows. Co-creation requires active participation and collaboration from end-users, which may be hindered by time constraints, workload pressures, and competing priorities in busy healthcare environments.
Moreover, the scalability of the dynamic visualization tools to different EHR systems may be limited by the interoperability and compatibility issues between systems. Integrating the tools with legacy EHR systems or proprietary software may require significant customization and development efforts, increasing the complexity and cost of implementation.
Additionally, ensuring data security, privacy, and regulatory compliance when sharing patient information across multiple systems and organizations is a critical consideration. Healthcare organizations must adhere to strict data protection regulations such as HIPAA in the United States or GDPR in Europe, which may pose challenges in sharing and accessing patient data for co-creation activities.

How can the dynamic visualization tools be further integrated with other data sources and clinical decision support systems to provide a more comprehensive and holistic view of patient information?

To enhance the integration of dynamic visualization tools with other data sources and clinical decision support systems, several strategies can be employed. Firstly, leveraging interoperability standards such as HL7 FHIR (Fast Healthcare Interoperability Resources) can facilitate seamless data exchange between different systems. By adopting FHIR APIs, the visualization tools can access patient data from diverse sources, including EHRs, laboratory systems, imaging systems, and wearable devices.
Secondly, implementing data aggregation and normalization processes can help harmonize data from disparate sources into a unified format for visualization. This involves mapping and transforming data elements to ensure consistency and accuracy across different systems. By standardizing data structures and terminology, the visualization tools can provide a more comprehensive and holistic view of patient information.
Furthermore, integrating clinical decision support algorithms and predictive analytics models into the visualization tools can enhance their functionality. By incorporating evidence-based guidelines, risk stratification scores, and treatment recommendations, the tools can assist healthcare providers in making informed decisions and improving patient outcomes. Real-time alerts and notifications based on clinical rules can also be embedded in the visualization interface to support proactive decision-making.
Moreover, leveraging advanced technologies such as artificial intelligence and machine learning can enable the dynamic visualization tools to analyze large datasets, identify patterns, and generate actionable insights. By incorporating predictive modeling and trend analysis capabilities, the tools can empower healthcare providers to anticipate patient needs, optimize treatment plans, and personalize care delivery.