Well-trained, competent radiologists make significant errors in reading radiographs about 30% (sic thirty per cent) of the time[1]. This figure has not varied much in studies over the last 25 years and does not improve with professional experience[2].
| PRACTICE
POINT
Competent and experienced experts frequently make errors in reading radiographs. |
Taking longer to read a radiograph does not help: in one study, erroneous readings averaged 147 seconds,correct readings 113 seconds[2(A)]. The same radiologist reading the same radiograph at a later time will disagree with his own previous reading 21% of the time[3].
If the radiologist has a full clinical history and/or laboratory data, his error rate may improve. Sometimes, however, this information will cause him to draw erroneous conclusions and overlook important radiographic features. Review of films and reports of previous similar studies of the same patient may lead him to copy errors or commit new ones[4].
Having radiographs checked by a second radiologist reduced the errors by more than a third, and increased the correct interpretation by 18%. However, the financial costs increased by 19%, and the number of false-positive diagnoses doubled[5]. Indeed, two radiologists reading the same radiograph disagree with one another more than half the time[6].
Copyright © 2009 Electronic Handbook of Legal Medicine