Exploring the Impact of In Basket Metrics on the Adoption of a New Electronic Health Record System Among Specialists in A Tertiary Hospital in Alberta: A Descriptive Study (Preprint)

Abstract

Health care organizations implement electronic health record (EHR) systems with the expectation of improved patient care and enhanced provider performance. However, while these technologies hold the potential to create efficiencies, they can also lead to unintended negative consequences, such as patient safety issues, communication problems, and provider burnout.

The current study aimed to document in-basket metrics (Time in In Basket per Day; Time in In Basket per Appointment; In Basket Messages Received per Day; and Turnaround Time) of the EHR implemented by Alberta Health Services (AHS), the province-wide health delivery system, called Connect Care. Also included Appointments per Day for comparison purposes (workload versus usage).

Based on the objectives of this study, a descriptive study was conducted. As research using Connect Care data is still in its infancy, this study design was deemed as appropriate. Due to the diversity of specialties, the providers were grouped into Medical and Surgery based on previous similar studies. The participants were then further sub-grouped based on their self-reported clinical Full Time Equivalent (FTE). This resulted in 3 sub-groups for analysis: medical FTE < 0.5, medical FTE > 0.5, and surgical (all of whom reported FTE > 0.5).

Seventy-two participants from 19 different specialties enrolled in this study. The providers had on average 8.31 appointments per day during the reporting periods. The providers received on average 21.93 messages per day and spent 7.61 minutes on average in Time in In Basket per Day and 1.84 minutes on average in Time in In Basket per Appointment. The time for the providers to mark messages as done (Turnaround Time), was on average 11.45 days during the reporting period. Although the surgical group had, on average, about twice as many appointments per scheduled day, they spent considerably less “connected time” (based on almost all of the time metrics) than the medical group. However, the surgical group took much longer to mark message as done (Turnaround time).

Although there were increasing and decreasing patterns among the included metrics, there were no obvious patterns across metrics and among groups. Therefore, there does not seem to be evidence of a ‘learning curve’ which would have shown consistent reduction in time spent on the system over time due to familiarity and experience. While this study does not show how the included metrics could be used as predictors of providers’ satisfaction or feeling of burnout, the usage trends could be used to start discussions about future Canadian studies needed in this area.