The purpose of a diagnostic test is to confirm or rule out disease or to increase or decrease the probability of disease. Only a few tests can separate all patients into those with and without a disease (true positive and true negative test). Usually there will be some false test results (false positive and false negative). Traditionally, the four test results are given in a 2 by 2 table, and the terms sensitivity, specificity, and predictive values defined. The influence of the prevalence of the disease in question on the relative distribution of the four test results is not obvious. This technical note brings a new illustration of the relative distribution of the four test results at prevalence from 0-1. The figure facilitates the understanding of the impact prevalence has on the predictive values of a clinical test.
Scandinavian Journal of Clinical and Laboratory Investigation, 2010, Vol 70, Issue 6, p. 458-61