The effect of clinical bias on the interpretation of myelography and spinal computed tomography.

Spinal computed tomograms and myelograms of 107 patients with sciatica or low back pain were interpreted with and without knowledge of clinical history. A significant number of interpretations was changed by knowledge of the clinical history. More studies were interpreted correctly without the clinical history than with it. Knowledge of the clinical history increased the number of false-positive and decreased the number of false-negative diagnoses. This study suggests a tendency of observers to interpret questionable myelographic or computed tomographic findings as positive when they correlate with clinical findings.