Predictive value derived from likelihood ratios: a superior technic to interpret quantitative laboratory results.

The predictive value of a quantitative test value calculated from a likelihood ratio is compared with the predictive value of a positive result for computer-simulated laboratory data. A significant loss of information about the magnitude of a quantitative test value occurs when calculating the predictive value of a positive result. This is because of the selection of an arbitrary critical value that classifies the quantitative result as positive or negative. The likelihood ratio (the probability of having disease divided by the probability of not having disease for a given test result) and the prevalence of disease is used to calculate predictive values. Logistic regression is used to predict the likelihood ratio of a test value.