I have definitely committed my fair share of cognitive errors resulting in missed diagnoses. The first step is to be aware of these cognitive biases so we can avoid them. I have listed some of the most common ones below and broken them down into sections.
Over-attachment to a specific diagnosis
Anchoring- Fixating on specific features of a presentation too early in the diagnostic process and subsequent failure to adjust
Confirmation bias- The tendency to look for confirming evidence to support the hypothesis while overlooking and evidence that refutes it
Premature closure- Accepting a diagnosis before it has been fully verified
Failure to consider alternative diagnoses
Sutton’s slip- Fixation on the most obvious answer
Search satisfaction- The tendency to stop searching once something is found and not considering additional diagnoses (i.e. the first positive finding was a red herring).
Representativeness restraint- Not considering a particular diagnosis for a patient because the presentation is not representative enough, i.e. it is not a “classic” presentation
Error due to inheriting someone else’s thought process
Triage cueing- A predisposition toward a diagnosis as a result of a judgment made by the triage physician, whose care may have been brief and early in the care process
Diagnosis momentum- The tendency for a particular diagnosis to become established in spite of other evidence
Framing effect- A decision being influenced by the way in which the scenario is presented or ‘‘framed’
Ascertainment effect- When thinking is preshaped by expectations. The alcoholic is just drunk (but may actually be herniating from ICH)
Errors in prevalence estimation
Availability bias- The tendency for things to be thought of and placed on the differential more frequently if they come to the mind more easily
Base-rate neglect- Failing to accurately take into account the prevalence of a particular disease
Gambler’s fallacy- Belief the same thing won’t happen again
Playing the odds- Deciding a patient doesn’t have a disease based on low likelihood and prevalence
Posterior probability error- Having a decision unduly influenced by a previous case
Errors involving patient characteristics
Gender bias- When the decision made is influenced unduly by the patient’s gender or the gender of the decision maker
Psych out error- A variety of biases associated with the health care provider’s perception of the psychiatric patient and blaming new organic disease on chronic psychiatric illness
Yin-yang out- Presumption that extensive prior investigation has ruled out any serious diagnosis on the current presentation. Beware of dismissing high utilizers.
Errors associated with physician affect or personality
Order effects- Focusing on information given at the beginning or end of a history and missing key information in the middle
Commission bias- Tendency toward action rather than inaction (over investigation, over intervention etc…)
Omission bias- Tendency toward inaction rather than action (under investigation etc…)
Outcome bias- Choosing a course of action according to a desired outcome and avoiding diagnoses that could lead to an undesirable outcome.
Visceral bias- Making decisions influenced by personal (positive or negative) feelings toward patients
Overconfidence/under-confidence- Being overconfident in or under-confident in the efficacy of decisions
Sunk costs- Unwillingness to give up a diagnosis in which considerable time and effort has been invested
Zebra retreat- Not willing to pursue rare diagnoses for a variety of reasons (delay in departmental flow, time intensive workup etc…)