Implementation of probabilistic decision rules improves the predictive values of algorithms in the diagnostic management of ectopic pregnancy

摘要: Current algorithms for the diagnosis of ectopic pregnancy do not take into account heterogeneity in patient profiles. Such can lead to differences pre-test probability pregnancy. In patients with clinical symptoms, example, presence an is higher than symptom-free patients. Any additional tests should then be interpreted differently, depending on probability. We present a diagnostic algorithm that uses probabilistic decision rules evaluation women suspected flexible cut-off levels test positivity compare it general fixed levels. Fictitious cohorts, varying prevalence were put together, using data obtained from cohort >800 inflexible algorithm, was diagnosed whenever could visualized at transvaginal sonography, or where serum human chorionic gonadotrophin (HCG) exceeded rigid level; rejected if intrauterine seen when HCG decreased. post-test after each test, probabilities and test-based likelihood ratios. Ectopic 95%, whereas this calculated fell below 1%. For both algorithms, sensitivity specificity as well predictive values calculated. At prevalence, associated 93% 97%. contrast, flexible, individualized depended Consequently, varied strongly used, they much more stable algorithm. five possible valuations false positive negative diagnoses, reduced expected disutility, compared It concluded clinicians incorporate used