Since the initiation of the NIST EHR usability program, I have noticed a range of reactions to its supposed intent. Everything from “the federal government should not dictate EHR design” to “it’s about time someone did something about EHR usability.” It appears that much of the discrepancy in opinion may be due to the varying ways the word “usability” is employed. Many clinicians think of usability in terms of productivity and workflow issues (i.e. how easy it is to use an EHR to complete common tasks). To some extent, the NIST framework helps in these areas. However, the stated goals of the EHR Usability Protocol (EUP) given in the document, Technical Evaluation, Testing, and Validation of the Usability of Electronic Health Records, point to a more narrow focus.
Usability is defined as:
The extent to which a product can be used by specified users to achieve specified goals with effectiveness, efficiency, and satisfaction in a specified context of use.
The executive summary of this document clearly states the goals of the EUP:
The purpose of this proposed usability protocol is to encourage user-centered development processes focused on safety by facilitating the design of EHR interfaces with good usability. The authors of this document seek to make EHRs safer by providing methods to measure and validate user performance prior to deployment. Moreover, the authors hope to encourage system developers to apply human factors best practices and incorporate user-centered design processes into the development and deployment of EHR systems. Such practices and processes have a proven record in industries such as aviation, military systems, transportation, and nuclear power.
Given the above statements and the contents of the protocol, it should be clear that the focus of the EUP is making EHRs safer for clinicians to use. It attempts to accomplish this by decreasing the likelihood that EHRs could engender errors that harm patients. It is not an attempt to create a “government-sanctioned” user interface and, thereby, stifle innovation.
Because the focus of the EUP is mainly patient safety, it does not substantially address clinicians’ concerns about workflow and productivity issues. Correcting workflow issues in any software product requires altering design elements. At best, those are interface components that can be changed easily; at worst, it may mean considerable reorganization or rewriting of major internal components.
Even without considering design changes, fixing workflow issues is difficult because every clinician is different. Any rigid workflow pattern built into an EHR will invariably be problematic for some subset of clinicians. I believe the ultimate key to addressing workflow issues is providing EHRs that have sophisticated end-user configuration tools. Such systems would permit clinicians to tweak the system until it matches their work habits and cognitive needs. Unfortunately, EHRs like this are far easier to describe than they are to build; however, I do believe they represent the future. In the meantime, the best way to improve the workflow hassles of a specific product is via feedback to the vendor or voting with one’s feet and changing products. In my experience, vendors are not only open to feedback, but welcome it. (Of course, like mileage, individual experiences will vary.)
The NIST usability initiative is designed to make EHRs safer, not to increase clinicians’ productivity or force a standard user-interface on all EHRs. If properly utilized, the EUP should help to remove obvious safety problems from EHRs. However, for clinicians looking at longer hours and more work since implementing an EHR, the EUP doesn’t have much to offer. Until highly configurable EHRs are available, the best way to avoid buying an EHR that works against you is to spend the time required to select the one that best matches your work habits.