Study: Adopting New EHR System Doubled Clinicians’ Cognitive Workloads

Adopting a new electronic health records (EHR) system doubled the amount of cognitive effort healthcare workers expended to perform their work during the first six months after implementation, according to a study that took place at two urgent care clinics.

In a collaborative study with Carle Health Systems, an integrated health system in Urbana, Ill., researchers examined how the urgent care clinics’ transition to the new EHR system affected staff members’ workloads as well as their perceptions of the old and new systems’ usability. Researchers found that the increase in cognitive workload also persisted for at least 30 months. The team reported its findings in a study published in the journal Applied Ergonomics.

“What we found surprised us a little,” the study’s first author, acknowledged Karen Dunn Lopez, Ph.D., R.N., the director of the Center for Nursing Classification and Clinical Effectiveness at the University of Iowa. “After two and a half years of continuous use, clinicians’ cognitive workload remained very high, and they still found the new EHR system more difficult to use than the prior paper- and computer-based hybrid system.” Dunn Lopez began working on the project during graduate school at the University of Illinois Chicago, where Morrow was teaching a course at the time. The team collected the data used in the study from 2011 to 2014.

With the hybrid system, clinicians used paper forms for tasks such as prescribing, writing clinical notes and medical orders. Seldom used while delivering patient care, the computer system then was primarily a repository for doctors’ notes after they were dictated and transcribed, and for documents that were scanned and uploaded after seeing the patients, noted co-author Daniel G. Morrow, Ph.D., an educational psychology professor at the University of Illinois Urbana-Champaign.

While the hybrid system evolved with the collaboration of Carle’s IT staff and clinicians, an external vendor developed the new EHR system, the researchers said. “Any time you change technology, you tend to change the whole work system to some extent. It’s important to take a user-centered approach to designing these systems and have users’ perspectives in mind so you can forecast how their work processes will change,” said Morrow.

Researchers noted that under the federal Health Information Technology for Economic and Clinical Health (HITECH) Act, healthcare providers were offered financial incentives to adopt EHR systems with certain features that constituted “Meaningful Use”—or in other words ensuring that their certified EHR technology connects in a manner that provides for the electronic exchange of health data to improve the quality of care.

However, according to Dunn Lopez, “this changeover was fraught with complaints that the new systems’ complex interfaces turned routine tasks into protracted multi-step processes, decreasing efficiency and forcing clinicians to devote ‘pajama time’– their off-duty hours–to entering data from patient visits.”

There have been plenty of studies over recent years showing that providers are spending more time in their EHRs than they would like. One commonly cited 2016 study by healthcare stakeholders, published in the Annals of Internal Medicine,  found many clinicians spend half their day on a computer navigating EHRs and other systems.

Clinical staff surveyed on cognitive workload

Study participants rated their perceived workloads under the old and new EHR systems using the NASA-TLX survey, which asks respondents to rate work tasks on a 0 to 100 scale based on their mental, physical and temporal demands, as well as the levels of effort and frustration involved and workers’ perceptions of their own performance.

Clinical staff completed the surveys prior to the new EHR system’s implementation and again at six to eight months and 30 to 32 months post-implementation. The clinicians included 22 members of the nursing staff, such as registered nurses and medical assistants, and 11 members of the provider staff, such as physicians, physician assistants and advanced practice nurses, researchers explained.

The researchers also observed the participants performing tasks with the new EHR interface, reviewed the system’s pages and compared its design to standard usability guidelines. Although the providers’ and nurses’ scores were similar on the pre-implementation surveys, there were evident differences after using the new system for six months. Compared with the nurses, the providers reported greater increases in cognitive workload, including higher mental and temporal demands and levels of frustration, the researchers said.

Part of the increased workload occurred because, under the Meaningful Use standards, some tasks previously performed by the nursing staff shifted to the provider staff, contended co-author Kayla Banks, Ph.D., R.N., Carle Health Systems’ vice president of women’s health and children’s services.

However, since the provider staff’s scores increased on all the survey’s subscales except that for physical activity, the researchers said the data suggested that some of the increased cognitive workload resulted from their performing tasks with the new EHR system while they were with patients rather than afterward.

What’s more, Morrow said that “minor design flaws such as slow response times when clinicians clicked on buttons and nonstandard labeling of tools negatively impacted their perceptions of the system’s usability and caused mounting frustration during their workdays.” The negative usability ratings were trending downward at 30 months’ post-implementation, however, and might have declined to the pre-implementation levels if the study had continued several months longer, he added.

Although usability may improve over time, healthcare organizations “can’t say ‘mission accomplished’ and withdraw support quickly after adopting new EHR systems,” said Banks. “There’s an ongoing need to modify workflows to accommodate increased workloads and look for iterative improvements. Nurses especially may be less likely to report that systems are not user-friendly or if their cognitive workload is significantly worsened.”