This thesis develops and tests a way to read emergency department (ED) performance indicators as direct evidence of resilient performance during disruptions, rather than as disconnected “better or worse” numbers. It focuses on how concrete work adaptations during COVID-19— like new isolation protocols, rapid assessment areas, and point-of-care testing—changed ED performance, and how those changes can be systematically translated into resilient performance profiles and quality trade-off narratives.
Background and problem
Emergency Departments operate under constant pressure to deliver fast, safe, and efficient care with finite resources. These pressures intensified during the COVID-19 pandemic, when EDs had to adjust their operations repeatedly while still maintaining core care functions. Performance indicators (PIs) such as length of stay, waiting times, and left-without-being-seen rates are widely used to monitor quality, but they are usually treated as isolated metrics or crude targets. Resilience Engineering and Safety-II emphasize how systems adapt under stress, yet existing tools typically produce qualitative capability profiles that are weakly linked to day-to-day operational performance.
The thesis identifies a central gap: there is no widely adopted method that uses routinely observed ED performance data to make resilience measurable, interpretable, and comparable, including its implications for the quality of care. As a result, resilience assessments often remain abstract, and they struggle to show concretely how disruptions and work adaptations affect real operations.
Research objective and questions
To address this gap, the thesis develops the Performance Indicator Resilience Assessment (PI–RA) framework, which links work adaptations to observable changes in performance indicators and to their associated quality trade-offs. PI–RA translates heterogeneous case evidence into a transparent read-out of resilient performance. In this thesis, resilient performance is interpreted using the resilience curve in Figure 1: a disruption pushes ED performance away from its usual level, after which the system may stabilize in a degraded state, recover back towards the baseline, or even improve beyond it. PI–RA can not measure the exact depth of the drop, but it uses before–after patterns in performance indicators to classify where the ED ends up on this curve—whether required operations remain degraded, move back onto a recovery trajectory, or improve with limited trade-offs—and what this implies for the quality of care.