Don’t crush the sensitive snout. Reply to: Sensitivity is not enough

Editor—We entirely agree with Wilson's comments on our paper¹ that sensitivity and positive predictive value (PPV) cannot be equated,² and appreciate the opportunity to clarify this point. Indeed, it is important to highlight that a sensitive test often has a low PPV (Table 1).³ In the editorial, we did not in fact liken sensitivity to PPV, but rather related sensitivity to negative predictive value (NPV), using the acronym SNOUT (a negative result in a sensitive test rules out the disease). Table 1NPV, negative predictive value. PPV, positive predictive value. This table illustrates that a sensitive test (e.g.≥90%) has a high NPV (e.g.≥80%), except when the prevalence of a disorder is high and the specificity of the test is lowPrevalence %Sensitivity %Specificity %PPV %NPV %190908.399.9190501.899.8190201.199.520909069.297.320905031.095.220902022.088.950909090.090.050905064.383.350902052.966.7

http://ift.tt/2ASoHG8