Kempainen RR, Migeon MB, Wolf FM. Understanding our mistakes: a primer on errors in clinical reasoning. Med Teach. Mar;2003:25(2);177-181.
Many hospitalists are involved in processes to reduce errors in the hospital. Given the dozens of clinical decisions hospitalists make each day, errors in clinical reasoning are worth exploring. However, few physicians are familiar with the terminology and classes of clinical reasoning errors as described in the literature.
This article outlines two models of clinical reasoning and explains common biases that distort clinical reasoning. Although the examples used to illustrate these errors draw from primary care internal medicine, they are easily recognizable.
Biases are defined as inaccurate beliefs that affect decision-making.
When generating a differential diagnosis, the bias of availability (aka recall bias) involves a clinician being influenced by what is easily recalled, creating a false sense of prevalence. This is especially common with less experienced clinicians, residents, and medical students.
The bias of representativeness (aka judging by similarity) involves the physician choosing a diagnosis based solely on signs and symptoms, while neglecting the prevalence of competing diagnoses.
The bias of confirmation involves the clinician using additional testing to confirm a suspected diagnosis, but failing to test competing hypotheses.
The bias of anchoring and adjustment involves the clinician inadequately adjusting the differential in light of new data.
The bias of bounded rationality (aka search satisfying) involves the clinician stopping the search for additional diagnoses after the anticipated diagnosis has been made.
Outcome bias involves judging a clinical decision retrospectively based on the outcome, rather than on logic and evidence supporting the original decision.
Omission bias involves placing undue emphasis on avoiding the adverse effect of a therapy, leading to under-utilization of a beneficial treatment.
Most of the work in clinical reasoning errors is published in the cognitive psychology and education literature, which most hospitalists do not regularly read. By becoming familiar with the concepts and terminology, hospitalists can more readily engage in discussions of clinical reasoning errors and how to avoid them. TH