We thank Volk, Spann, and Mold for their comments on our recently published article.1 They raise several important issues.
We agree with their comment that it should not be surprising that the specificity of serum PSA determinations should be greater in the general population than in urologic practices. This is precisely one of the reasons we undertook this analysis. More importantly, we hoped that we could draw attention to this discrepancy so that future discussions of the accuracy of the PSA test would take into account the reason for and setting in which the test is being ordered.
Similarly, we agree with their statement that the predictive value is a function of both the prior probability of disease (prevalence) and the accuracy of the test. Several additional comments are warranted. The utility of the test will depend on the perspective from which the assessment is made. Volk et al