I have just finished rereading the NIST Draft EHR Usability Protocol (EUP). I am even more impressed with its quality now that I have a new software project underway. When developing software, creating an actionable set of requirements is difficult. Working with users only solves part of the problem because naïve users (those who have never used a system of the type under development) have only a vague idea of what a system should do. I have seen many HIT projects end in disaster because developers gave users what they thought they asked for. The workflow assumptions built into an EHR are a good example of translation mishaps.
Every clinician has his/her own personal work habits. Building a system by canvassing a group of clinicians for information about what they do, the steps involved, the information required at a given step, and other workflow-related issues will result in information that the developers must interpret in order to create a working system. However, since what emerges will be a distillation of the commonalities between users, the final system may not be ideal for anyone consulted. This is where the NIST information becomes helpful.
The EUP, along with the Guide to the Processes Approach for Improving the Usability of Electronic Health Records and Customized Common Industry Format Template for Electronic Health Record Usability Testing, provide developers with concrete data on common usability problems that can be used to complement and enrich data-gathering interactions with end users. The guidance in these documents allows developers to uncover many workflow issues during design phases, when they are more easily (and cheaply) corrected, instead of relying predominantly on end-user test results.
While these documents are a boon for developers, they also hold value for anyone buying an EHR. The expert review questions/checklists in the EUP are easily converted into screening tools for EHR purchasers trying to evaluate products. Even a small subset of the checklist questions, when morphed into an evaluation script, is likely to be a better screening tool than most potential buyers usually have.
It will be interesting to see how the federal push for EHR usability affects the marketplace. Knowing that a product has usability issues and fixing them are two very different matters. The usability evaluation protocol focuses on user interface issues. Poor usability at the user interface level could easily be a symptom of a deeper design issue. In such cases, fixing the usability issue might require rewriting significant portions of the program. Going back to our workflow example, consider how an EHR product with hard-coded workflow pathways differs from one that has a configurable workflow engine. Correcting usability issues in the former would require significant coding, whereas in the latter it might only require a change in configuration settings. Systems with layered, component-based architectures that allow for “separation of concerns” will have a competitive advantage.
Considering both the lag time before significant product changes may occur and the current push to have as many clinicians as possible using EHR systems by 2015, in three years will providers discover they’ve gone through expensive implementations with outdated products?
Whatever happens in the marketplace, NIST has provided EHR developers and EHR purchasers a goldmine of high quality resources for evaluating EHR products. If you are in either category, read them—you won’t regret it.