Clinical Audit Logs as Multi-Axial Traces of Care Delivery
Computers and Society
Summary
The authors explain that electronic health record audit logs track many actions from different viewpoints, like a doctor's work, a patient's history, a team's efforts, and common routines. They suggest these logs can be seen as streams with multiple layers happening at once. This idea helps guide how computer models should learn useful patterns from the logs, how those patterns should be tested, and how their use should be controlled. The work highlights the complexity and potential of audit logs for understanding healthcare processes.
Authors
Braden Eberhard, Nate Apathy, Kevin Johnson
Abstract
Electronic health record audit logs record timestamped actions through which clinical work is carried out. Generated as operational metadata, they now support research on clinician effort, patient outcomes, care-team coordination, and workflow structure. This Perspective explains that breadth by articulating audit logs as multi-axial event streams and drawing implications for representation learning, evaluation, and governance. Each logged action belongs simultaneously to multiple clinically meaningful relations: a clinician's work, a patient's trajectory, a team's activity, and a recurring workflow. This structure motivates foundation-model pretraining to learn reusable representations over the raw stream. Reading audit logs as multi-axial traces specifies what such representations must preserve, how their value should be tested, and how their use should be governed.