Diagnostic errors are common, deadly, and costly. Twelve million Americans annually experience diagnostic
error in ambulatory care, including in Emergency Department (EDs), over half of these with potential for harm.
ED clinical practice is especially prone to diagnostic error as a sociotechnical work system that is fast-paced,
high-stakes, highly adaptive and complex. The 2016 National Academy of Medicine (NAM) report was an
urgent call for more research regarding diagnostic safety, making particular reference to the ED. ED diagnosis
is cognitively-intense work, distributed across team members who may or may not be co-located. There is very
limited understanding of the salient `real-time' details of the ED diagnostic process and associated
performance shaping factors on the work system. Without structured in-depth analysis of ED diagnosis
occurring as part of `real-time ED work,' that is “work-as-done,” we will continue the struggle with the design of
effective, sustainable interventions to improve diagnostic safety. Accordingly, we are proposing a 3-year, multi-
site, multi-method field study in the ED based on a sociotechnical systems approach and a macrocognition
framework, which is the study of cognitive tasks that characterize how people think in natural settings. We
have 3 specific aims: (1) AIM 1. To understand provider (physician and advanced practice provider) work
involved in ED diagnosis and identify associated performance shaping factors. (2) AIM 2. To understand
collaborative (team-oriented) work involved in ED diagnosis and identify associated performance shaping
factors. (3) AIM 3. To conduct a proactive risk assessment of the diagnostic process in the ED.
AIM 1 and AIM 2 will be achieved by conducting in-depth qualitative studies using a variety of data
collection methods (observations, interviews) and cognitive task analyses techniques. Data analysis will
produce a range of outputs such as process maps, macrocognitive and procedural tasks involved in diagnosis,
information flow diagrams, role network graphs, among others. AIM 3 will use two complementary proactive
risk assessment methods to assess failure modes and performance shaping factors and to identify possible
interventions to improve ED diagnostic safety: (1) Health Care Failure Mode and Effect Analysis (HFMEA); (2)
Functional Resonance Analysis Method (FRAM) Based “What-if” Risk Analysis. Additionally, we will develop a
research methods compendium/guide for those interested in conducting similar research on diagnostic safety.
The study will be conducted in 3 different EDs (urban, suburban, rural) that serve patients from 6 AHRQ priority
population groups. The research team is interdisciplinary, composed of internationally known experts in patient
safety, human factors, systems engineering, cognitive psychology, communication, emergency medicine, and
nursing. The study is innovative due to its lens on ED diagnostic process as a whole, its use of human factors-
based conceptual approaches, its investigation of the ED team's role in the diagnosis, and its use of a variety
of cognitive task analysis techniques and proactive risk assessment methods.
