Creating a "Culture of Triage": A Dual-Perspective Study of Interruptions During ICU Rounds

Abstract

Objectives: To explore interrupters' and interruptees' experiences of interruptions occurring during morning rounds in a PICU in an attempt to understand better how to limit interruptions that threaten patient safety.

Design: Qualitative ethnographic study including observations, field interviews, and in-depth interviews.

Setting: A 55-bed PICU in a free-standing, quaternary-care children's hospital.

Subjects participants: Attending physicians, fellow physicians, frontline clinicians (resident physicians and nurse practitioners), and nurses working in the PICU.

Interventions: Data collection occurred in two parts: 1) field observations during morning rounds with brief field interviews conducted with participants involved in an observed interruption and 2) in-depth interviews conducted with selected participants from prior field observations.

Measurements and main results: Data were coded using a constant comparative method with thematic analysis, clustering codes into groups, and subsequently into themes. We observed 11 rounding encounters (17 hr of observation and 48 patient encounters), conducting 25 field interviews and eight in-depth interviews. Themes included culture of interruption triage, interruption triage criteria, and barriers to interruption triage. Interruptees desired forming a culture of triage, whereby less-urgent interruptions were deferred until later or addressed through an asynchronous method; this desire was misaligned with interrupters who described ongoing interruption triage based on clinical changes, time-sensitivity, and interrupter comfort, despite not having a formal triage algorithm. Barriers to interruption triage included a lack of situational awareness and experience among interrupters and interruptees.

Conclusions: Interrupters and interruptees did not have a shared understanding of the culture of triage within the PICU. Although interrupters attempted to triage interruptions, no formal triage algorithm existed and interruptees did not perceive a triaging culture. Using data from this study, we created a triage algorithm that could inform future studies, potentially decrease unnecessary interruptions, and optimize information sharing for essential interruptions.