Building a Composable Electronic Health Records System using Fast Healthcare Interoperability Resources (FHIR) web components (Preprint)

Abstract

In the last decade, Electronic Health Records systems (EHRs) have been widely adopted in the United States because of the government's efforts. These efforts have led to the broad and rapid adoption of the EHR but with numerous usability and interoperability challenges. The disconnect between the clinical users' expectations from their EHRs and what is available in the EHRs is well-known.

This study aimed to evaluate (1) the feasibility of a user-composable user interface (UI) for an EHR system using FHIR web components, (2) the user's ease of learning and ease of use of the assembled EHR's UI.

We developed web components for the FHIR patient resource using the Polymer 3 library. The components were assembled by 19 subjects with a clinical background and past EHR use experience utilizing a drag and drop UI designer. We measured latent variables of "ease of learning" and "ease of use" through a 33-questions pre-validated survey from these subjects. Further, we conducted semi-structured interviews as per the sequential explanatory research design. The quantitative and qualitative data obtained from the survey and the semi-structured interviews were analyzed using the convergent parallel mixed method design. The quantitative data derived through the survey estimated the Structural Equation Modeling (SEM) latent variables tagged "ease of learning" and "ease of use" of the user interface.

About 88 percent (16 of 18) of the participants were confident or very confident about assembling the user components to form a user interface after watching a workflow video of the process, and 63 percent (12 of 19) of participants eventually completed the assembly process. A partial least square structural equation modeling (PLS-SEM) analysis of the survey results showed that there is a significant positive effect (p =0.00) on the ease of learning of the assembling process and on ease of use of the system and vice versa. The qualitative analysis of the interview showed that more' easy to use' and 'easy to learn' themes appeared in the participants' responses (40.1%) than the "difficult to learn," "difficult to use," "too complex," and "cumbersome" themes combined appearing in 11.2 % of the combined themes.

It seems that the non-technical users can assemble an EHR user interface without little or no technical support. Moreover, a user-composed EHR user interface will improve EHR users' experience. Finally, web components based on FHIR resources can be developed and implemented into an EHR user interface.