Detection of lacunar infarction in brain CT-scans: no evidence of bias from accompanying patient information.

Interobserver agreement in assessing brain CT-scans is, in general, high. The extent, however, to which such agreement is caused by bias through knowledge of other clinical details remains uncertain. The hypothesis that observers are somehow prejudiced before assessing ambiguous CT-scans in this particular situation was tested. Sixteen neurologists and 16 radiologists volunteered to interpret two ambiguous brain CT-scans, with regard to the presence or absence of a lacunar infarct in the region of the internal capsule. The scans were accompanied by "patient" information that was or was not suggestive of a stroke. These scans were camouflaged by a variety of other scans, to be assessed in the same way, to mask the purpose of the study. It was assumed that the observers, in their assessments of the scans, would somehow let their ratings of the likelihood of a lacunar infarction in or near the internal capsule be subject to the accompanying information. Results showed lower ratings produced by neurologists (i.e., less likelihood of an infarction) than by radiologists in the majority of all assessments, but no bias by the accompanying information.