In Part I of this series, we argued that physicians need tools they can use to evaluate EMR usability. One reason for this is that the ONC’s soon-to-be-released EMR certification criteria don’t address usability, even though usability has a dramatic effect on productivity and the quality of care. Part I also presented a partial list of the dimensions of EMR usability.
In today’s post, we complete the list and introduce some principles that serve as the foundation for creating a tool physicians can use to evaluate EMR usability on their own. Readers who wish to explore these topics in more detail would do well to begin with the fine review by a HIMSS Task Force on the subject.
Dimensions of EMR Usability (Continued)
We had previously mentioned simplicity, naturalness, consistency and minimizing cognitive load. The others are:
Efficient Interactions
Two ways to facilitate efficient user interactions are to minimize the number of steps required to complete tasks and to provide shortcuts for use by frequent and/or experienced users. Other ways to assure efficient interactions are autotabbing, good default values, and lists and text boxes that contain enough information to limit scrolling and minimize the need for frequent switching between keyboard and mouse.
Forgiveness and Feedback
The former involves allowing the user to explore the application without worrying about disastrous results. The latter, feedback, augments forgiveness by informing the user about the consequences of the actions she appears ready to implement.
Effective Use of Language
All language in the EMR should be concise, unambiguous, familiar and meaningful to the user. Only commonly understood abbreviations and acronyms should be used in the EMR, yet many more of these should be recognized by the EMR when users attempt to enter data or engage the search function.
Proper Use of Color
This contributes to a pleasing user interface, but more importantly it conveys meaning to the user. For color to convey meaning, there cannot be a larger number of colors in play than the user can remember, and the colors must be used consistently throughout the application.
Readability
This usually implies a text size of at least 12‐point for key content and nothing smaller than 9‐point, with customer-modifiable sizes for those with visual impairment. San serif fonts are easier to read than serif fonts in computer displays. Maximal contrast between text and background is preferred, with black on white being the best.
Preservation of Context
This involves minimizing screen changes and visual interruptions during any particular task. The classic visual interrupter is the dialog box. These should be should deployed judiciously when they are used, they should be displayed adjacent to or below the prompt that triggered it.
Designing an EMR Usability Tool
To translate this list into a tool that can be used to evaluate EMR usability, it is necessary to identify a dozen or so tasks that are performed using an EMR and then develop metrics to assess performance against these tasks. Ideally, the tasks are either performed frequently (e.g. scheduling a patient) or known to be associated with quality issues and patient safety (e.g. ordering a drug). As for the assessments themselves, they should include both objective metrics and subjective assessments of satisfaction.
(Note: Since physicians can sign-up for and access the Practice Fusion EMR in minutes and at no cost, physicians will find it easy to assess the usability of this EMR. Other EMR vendors typically do not make it easy for physicians to engage in these kinds of self-evaluations—they might insist that physicians be trained first, for example. This is in itself not a good sign when it comes to usability. Regardless, physicians interested in assessing the usability of other EMRs may find it necessary to observe others carrying out these tasks instead of doing it themselves.)
It turns out to be fairly easy to develop objective measures of EMR performance. For example, the efficiency with which a task is performed using an EMR can be measured as the:
• Time required to complete the task,
• Number of key strokes or screen visits required to complete the task,
• Number of back-button presses required while completing a task.
Similarly, the effectiveness with which a task is performed on an EMR can be measured as the:
• Number or rate of errors generated during x runs of the task,
• Severity of errors generated during x runs of the task,
• Number of requests for help while completing the task.
The most common subjective assessment of EMR usability is user satisfaction, which refers to one’s overall impression of the experience using an EMR. Satisfaction is linked to efficiency and effectiveness, and to other factors including cost and quality of support, as well. The simplest way to assess satisfaction is to use the familiar 10-point Likert-scale.
In Part III of this series (appearing Friday), we’ll present an actual measurement tool that physicians can use to assess EMR usability. Stay tuned!
Glenn Laffel, MD, PhD
Sr. VP Clinical Affairs
Practice Fusion EMR
