Test Cases Urged In Usability Recommendations

Rather than study electronic health record (EHR) usability in the abstract, “a few high-value use cases, particularly those that have patient safety implications,” should be examined, according to recommendations developed by the HIT Policy Committee’s Adoption and Certification Workgroup.

Those recommendations — based on a day of hearings held by the workgroup — were outlined in a letter to National Coordinator Farzad Mostashari, M.D.

In the letter, authors Marc Probst, CIO at Intermountain Healthcare and co-chair of the workgroup, and Larry Wolf, senior consulting application/data architect at Kindred Healthcare and a workgroup member, broke down different facets of usability before suggesting the test-case approach.

“A repeated analogy we heard in the hearing is that of the automobile: steering wheel, directional signals, accelerator and brake pedals have a high degree of consistency from vehicle to vehicle. Other aspects of the vehicle’s controls and indicator dials vary somewhat but not all that much. Similarly, road signage is consistent for stop signs, informational and warning signs,” they wrote.

The duo then noted that each EHR vendor applies different “control” and custom “gauges” to their applications, making it difficult for clinicians to move from one system in their practice, for example, to another in the hospital setting. One suggested improvement was defining a limited set of icons that would be consistent across applications, as well as consistent terminology for actions taken within the EHR.

Probst and Wolf also wrote that medical schools and clinicians have a role to play in improving usability by creating and undergoing “user training that addresses how to develop a ‘technology-enabled’ bedside manner.”

Vendors, of course, are largely responsible for improving usability by continually developing their products. The letter states applications need to better manage the “cognitive load” clinicians must bear when navigating the systems. For example, are related bits of data accessible on the same screen or must users go to different sections to find all the information they need at any one time? “This sometimes results in information being written down by the clinician as they find it,” the letter warns.

In what may be a cause for concern to policymakers considering the Meaningful Use timeline, the letter states that deep usability often requires customization, and customization takes time.

“The complex development and implementation process requires multiple feedback loops: between the users and the implementers within an organization, between the implementers and the vendors to improve the implementation, and within the vendor organization to improve the product. All of these changes are areas where getting the details right is the difference between a highly successful implementation and a failure.”

Probst and Wolf explain that while “regulation drives change in systems,” such change can be positive or negative.

“Regulation can increase the complexity of an application and can also shift the work from one group of individuals to another. Many regulations change the data needed for subsequent activity and these place a burden on the care process. For example, additional detail to support quality measures and to document the level of service for billing increases the burden on clinicians,” they caution.

The letter concludes with a suggestion to develop the aforementioned “few, high-value” use cases. “This will provide the context for measuring usability, allowing usability testing and process improvement to proceed.”