Incompatibility between levels of reliability of physical examination techniques and their clinical usefulness produces an enormous chasm, into which any hope of accurate diagnosis falls and flounders. Bridging this chasm will require validation of the examination techniques along more than just observer variability lines. Greater attention to the use of gold standards is recommended, even though they themselves must be critically evaluated for the errors which they contain and for technicalities involved in developing elaborate and accurate standards.
This abstract is reproduced with the permission of the publisher. Article only available in print.