February 26, 1997

Radiologists' Interpretations: The Problem of Context Bias-Reply

Author Affiliations

Yale University New Haven, Conn

JAMA. 1997;277(8):628. doi:10.1001/jama.1997.03540320030026

This article is only available in the PDF format. Download the PDF to view the article, as well as its associated figures and tables.


In Reply.  —The availability of a clinical history certainly affects the interpretations of radiologists. We agree with Dr Vermont that informing the radiologists in our study that patients belonged to either high-risk or low-risk groups might have had a significant impact on their diagnoses. We believe, however, that his proposed scenario would simply replicate the effect of adding the history. Our results showed that, even when such influences are avoided, radiologists' diagnoses are not based solely on each set of films. "Context bias" demonstrates the effect of particular sequences or combinations of films. These influences of history, risk, and context complicate the objective assessment of radiologic tests, although their relation to the outcomes of patient care remains unexplored.