Intervals for posttest probabilities: A comparison of 5 methods

Background. Several medical articles discuss methods of constructing confid ence intervals for single proportions and the likelihood ratio, but scant a ttention has been given to the systematic study of intervals for the poster ior odds, or the positive predictive value, of a test. Methods. The authors describe 5 methods of constructing confidence intervals for posttest proba bilities when estimates of sensitivity, specificity and the pretest probabi lity of a disorder are derived from empirical data. They then evaluate each method to determine how well the intervals' coverage properties correspond to their nominal value. Results. When the estimates of pretest probabiliti es, sensitivity, and specificity are derived from more than 80 subjects and are not close to 0 or 1, all methods generate intervals with appropriate c overage properties. When these conditions are not met, however, the best pe rforming method is an objective Bayesian approach implemented by a simple s imulation using a spreadsheet. Conclusion. Physicians and investigators can generate accurate confidence intervals for posttest probabilities in small -sample situations using the objective Bayesian approach.