Disease/Test Relation Calculators

 File Name (Click the ".xls" file, hold mouse button, select "save as" to download) Calculator Name Function of Calculator Contingency table (test by disease): counts from probabilities You specify base rate, sensitivity, specificity, it produces counts in 2 by 2 table plus other statistics. Contingency table (test by disease): probs. from counts You put counts in cells of 2 by 2 table, it produces sensitivity, specificity, etc. 2x2p&ci.xls Contingency table (test by disease) with confidence intervals on each probability As with 2x2probs, you put counts in cells of 2 by 2 table; it produces the list of probabilities, and gives the confidence interval for each. Decision Tree: Bayes' Theorem You specify base rate, specificity, and sensitivity. It demonstrates "inversion" of a tree. Bayes' Theorem in the Odds/Likelihood ratio form You specify Prior Odds the patient has disease, and likelihood ratio of seeing test results, given patient has the disease or does not have the disease; it calculates posterior odds. Log Odds Version of Bayes’ Theorem You specify prior probability, sensitivity, specificity, and test result for a series of tests. It graphs Bayes' Theorem calculations in Log(odds) scale. Bayesian graph You specify sensitivity and specificity of a test, it prints posterior probability as a function of any prior probability. baysgrmg.xls Bayesian graph for diseases with low prior probability You specify sensitivity and specificity of a test, for a disease with a low prior probability (less than 10%). It prints posterior probability as a function of prior probability in the 0 to 0.1 range. Bayesian graph with thresholds As with Bayesian graph; plus you specify a treatment threshold and it specifies and graphs a No-test/Test Threshold and a Test/Treat Threshold. Bayesian graph with thresholds based on utilities As with Bayesian graph with thresholds; plus you specify the utilities upon which a threshold would be based. Screening perspective calculator You specify the base rate for a condition, the number of people screened, and the sensitivity and specificity of a screening test, and it displays a tree showing the numbers with/without the disease and with true and false screen results. Explore the use of “negative entropy” to measure uncertainty You specify prior probability of a disease, and the sensitivity and specificity of a test. It shows probability of a positive or negative test, post test result probabilities, the pretest uncertainty, and the expected posttest uncertainty. Explore implications of 2 diagnostic tests - are they independent or correlated? You specify the characteristics of a disease (prior probability) and of 2 diagnostic tests for it (sensitivities and specificities, and correlation between them when disease is present and when absent). It constructs the corresponding data set, and gives post test probabilities for all combinations of test results. Explore implications of 3 diagnostic tests - are they independent or correlated? You specify the characteristics of a disease (prior probability) and of 3 diagnostic tests for it (sensitivities and specificities, and correlation between them when disease is present and when absent). It constructs the corresponding data set, gives post test probabilities for all combinations of test results.