KLAS: Epic Signal Data Cannot Predict EHR Satisfaction, Clinician Burnout

Signal, the provider-efficiency tracking tool from Epic, has several valuable uses, but it is not a meaningful or predictive measure for EHR satisfaction or clinician burnout, according to a KLAS report.

The report relies on information from 16 organizations that have measured their clinicians’ EHR experience with the Arch Collaborative in the past three years and expressed interest in understanding how their Signal data correlates with aspects of the clinician experience.

Healthcare organizations leverage Signal data to understand the ways end-users interact with the EHR. Often, practices use the data to pinpoint providers who may be struggling and could benefit the most from EHR training.

Signal data cannot accurately predict EHR satisfaction, clinician burnout, or turnover. However, KLAS identified five key principles for successfully leveraging Signal data to drive provider EHR satisfaction:

1. Use Signal to identify outliers among healthcare providers and determine how optimization and training can improve efficiency.
2. Use Signal data to identify areas requiring future training or optimization efforts through informaticists or trainers.
3. Approach providers in a non-punitive way to understand EHR usage and address pain points
4. Provide training on best practices to enhance EHR efficiency and help end-users personalize workflows.
5. Measure progress and take an iterative approach to improvement efforts.

As already noted, Signal data is not a good predictor of a provider’s EHR satisfaction as measured by the Collaborative’s EHR Experience Survey. For instance, many providers whose Signal data suggests they have low satisfaction are very satisfied, and vice versa.

“Part of this disconnect is rooted in the inherent differences in how the two sources of data are collected and what they measure,” the KLAS authors explained. “Epic Signal data is closely tied to operational understanding—when and how the EHR is being used, what features are being used, and the nature of provider workloads.”

Therefore, using a Signal metric for how much time users spend in the EHR makes a disengaged user appear similar to someone who is highly efficient in the EHR. In comparison, Arch Collaborative data measures overall EHR perception and is not dependent on specific time frames like Epic Signal data.

Still, Signal data can be useful as a stepping-off point to improve the EHR experience. In the report, high-performing Arch Collaborative members shared ways they use Signal data to spark conversations and guide interactions with providers.

For instance, Penn Medicine uses Signal data and provider surveys to tailor EHR personalization. The data helps determine what should be covered during the session, considering the provider’s specialty, role, and needs.

Additionally, Lee Health uses Signal data to analyze specific EHR metrics, compare providers to peers, and identify opportunities for EHR efficiency improvement.