A Safety-II approach to learn from practice variation in the diagnostic process in the emergency department: an action research study

Objectives: 

The focus on improving patient safety has mainly been by learning from errors and near misses (Safety-I). We applied a novel Safety-II approach to identify and learn from practice variations in the diagnostic process in the emergency department (ED), and subsequently design and implement practice changes. 

Methods: 

In this single-center study, we used action research (cycles of micro-experiments to study and improve processes with active stakeholder involvement) in the diagnostic process. We used three subsequent observation cycles of the following six steps: observations, gathering of follow-up data, analyses, a co-creation session with involved stakeholders, sharing of best practices, and updating the observation form. The observations and analyses focused on identifying practice variation in everyday practice rather than on what went right or wrong. During co-creation sessions, stakeholders discussed whether practice variations were reflective of possible improvements in the diagnostic process. Promising best practices were identified, and subsequently implemented as practice changes aiming to improve the diagnostic process. Implemented practice changes were evaluated in subsequent cycles. 

Results: 

Forty diagnostic processes were observed. We identified practice variations that reflected the resilience and adaptability of clinicians, as well as variations revealing opportunities to improve the diagnostic process. Five identified best practices were implemented as practice changes: a template for the documentation of ED preannouncements, a document with relevant digital information resources for residents, face-to-face supervision by Internal Medicine consultants during office hours, blood sampling at triage, and adding lipase to the standard ED blood tests. These changes were well-received by stakeholders, also shown by an adoption-rate of 67–100 % of observed cases after implementation. 

Conclusions: 

A Safety-II approach with action research and direct observations of the diagnostic process in the ED can be successfully applied to identify and learn from practice variation, and can lead to well-received practice changes.