ANSI-approved standard defines requirements to help developers, healthcare organizations and evaluators assess health IT usability
By the EHR Usability Functional Profile Co-Facilitators: John Ritter, FHL7, Project Lead; David Schlossman, MD, PhD, Project Lead; Gary Dickinson, FHL7, EHR Standards Consulting; Don Mon, PhD; and Mitra Rocca, Dipl. Medical Informatics, AI
Usability matters whenever clinicians interact with health information technology. Systems designed around users’ needs can help them find information, complete tasks more efficiently, and make informed decisions.
The EHR-S FM Release 2.1: Usability Functional Profile, Edition 2 has been approved by the American National Standards Institute (ANSI) and is now available for industry use. Based on the HL7 Electronic Health Record System Functional Model (EHR-S FM), the profile defines functional requirements and associated conformance criteria to help stakeholders evaluate system usability.
User-centered design principles and best practices in human-computer interaction design are critical to developing more usable healthcare information technology (health IT) systems that help clinicians work more efficiently, reason more accurately, reduce errors, improve care quality, improve clinician and patient experience, and reduce per capita healthcare costs. The HL7/ANSI EHR-S FM Release 2.1 Usability Functional Profile (UFP), Edition 2 standard defines functional requirements and associated conformance criteria to help stakeholders evaluate system usability.
The UFP is designed to assist multiple stakeholder groups, including:
Usability in the Age of Clinical AI
Currently, the healthcare field is undergoing revolutionary changes by incorporating large-language model, generative artificial intelligence systems (LLM AI) into clinical care and clinical decision support. These AI tools aim to help clinicians organize and synthesize the rapidly growing body of clinical and biomedical knowledge to support better clinical decision-making. However, performance on synthetic benchmarks alone does not establish how effectively clinicians can use LLM AI tools.
These tools can still increase burden, or even cause harm if clinicians misunderstand the output, over-trust it, under-trust it, cannot see when it is wrong, and cannot efficiently correct errors. Incorporating AI in a way that maximizes benefits and prevents harm or misuse is a very significant health IT challenge. Addressing them will require new approaches to usability and new human-computer interaction models.
The UFP project team hopes to address some of these issues in future releases.
Access the Usability Functional Profile
The UFP is available on the HL7 website at: https://www.hl7.org/implement/standards/product_brief.cfm?product_id=611.