PROJECT SUMMARY
The Coronavirus disease 2019 (COVID-19) pandemic dramatically altered healthcare delivery in persistent ways.
To adhere to the physical distancing guidelines and to provide continuity of patient care, organizations shifted
the primary modality of ambulatory care to telemedicine-based virtual encounters. This transition has altered the
structure, management, and delivery of patient care—with large potential changes to patient-provider
communication and availability of clinical information. For example, during telemedicine encounters clinicians
can no longer conduct physical exams or obtain vitals that inform clinical reasoning and decision making in
routine ambulatory evaluation and management (E&M) encounters. In turn, these factors may contribute to
clinical uncertainty, and thereby alter how the clinician leverages the electronic health record (EHR). They may
need to engage in additional chart review to fill information gaps, enter patient-generated health data, or send
more follow-up communications. One or more of these changes serves to intensify EHR-based cognitive load
as EHR activities and activity switching increase, both at the encounter level and cumulatively over the course
of a workday. In turn, greater EHR-based cognitive load could contribute to suboptimal clinical decisions (e.g.,
more diagnostic tests or referrals) and more errors (e.g., wrong-patient orders).
In the proposed Assessing the Effect of Telemedicine on Physician EHR Work, Cognition, and Process
Outcomes (ASPIRE) project, we investigate the primary hypothesis that ambulatory telemedicine encounters in
the COVID context are associated with increased EHR-based cognitive load among clinicians, and downstream
suboptimal clinical decisions and more frequent errors. We leverage novel, EHR-based audit log data from a 3-
year period spanning pre- and COVID-periods (March 2019 – February 2022) to directly measure clinicians' EHR
activities in telemedicine and face-to-face encounters at two large academic medical centers (Washington
University in St Louis and University of California, San Francisco). Using the COVID-19 pandemic as a natural
experiment that dramatically increased ambulatory telemedicine encounters (more than 25-fold at the two health
systems), our first aim will characterize the differences in EHR-based activities between face-to-face and
telemedicine encounters. We will then construct a derivative measure of EHR activity switches (within and across
encounters) as a proxy measure of cognitive load and evaluate the downstream impact of cognitive load on
clinical decision making and wrong-patient errors. The findings from these aims will be leveraged in our final
“design” aim that uses frontline clinician interviews and a national expert eDelphi process to elicit the EHR-based
factors impacting telemedicine encounters and to identify potential design strategies to address associated
challenges. The eDelphi process will focus on translating and prioritizing the identified design strategies into
pragmatic goals to improve EHR support for telemedicine encounters.