In a study in this issue, Perez-Rossello, et al, investigate the accuracy of whole-body MRI in detecting skeletal and soft tissue injury among infants suspected of being physically abused. They report on the sensitivity and specificity of whole-body MRI compared to the standard of skeletal survey, and they conclude that although the specificity was high, because of its low sensitivity, the MRI cannot replace the skeletal survey in this situation.

If a test has very high sensitivity, is it most useful to rule disease in or to rule it out? The somewhat counter-intuitive answer to this is that a test with high sensitivity (eg, most of the patients with the disease have a positive test) is most useful to rule disease out, and one with high specificity (eg, most of the patients without the disease will have a negative test) is most useful to rule disease in. This is because...

You do not currently have access to this content.