Base-rate error in the interpretation of immunohistochemistry

Area of Science:

• Pathology
• Medical Diagnostics
• Cognitive Bias

Background:

• Diagnostic test interpretation is crucial in pathology.
• Base-rate neglect, a cognitive bias, can lead to misinterpreting test results when prior probabilities are not adequately considered.

Purpose of the Study:

• To investigate whether pathologists are susceptible to base-rate neglect when evaluating unexpected immunohistochemical results.
• To assess the impact of base-rate information on diagnostic probability estimations by pathologists.

Main Methods:

• A scenario-based study was conducted with consultant pathologists.
• Participants estimated the post-test probability of a tumor (Tumor B) given an unexpected positive immunostaining result (Antibody X).
• The prior probability strongly favored Tumor A (99%) over Tumor B (1%).

Main Results:

• Pathologists' probability estimates for Tumor B ranged widely (0-100%), with a mean of 29.7%.
• The calculated Bayesian probability for Tumor B was significantly lower at 7.5%.
• This indicates a substantial overestimation of the diagnostic significance of the unexpected result.

Conclusions:

• Pathologists may exhibit base-rate neglect, leading to overestimation of diagnoses based on unexpected test findings.
• Failing to account for the base rate of disorders can significantly skew diagnostic probability assessments.
• Further education on cognitive biases in diagnostic reasoning is warranted for pathologists.