HSS 4303 Lecture Notes - Lecture 9: Likelihood Ratios In Diagnostic Testing, False Positives And False Negatives, Mammography
10 Oct 2018
School
Department
Course
Professor
Document Summary
Positive predictive value: the proportion of patients with positive test results who are correctly diagnosed. Positive test are people who have been diagnosed as positive for the disease. Likelihood ratio: the value of the test for increasing certainty about a positive diagnosis. Likelihood ratio is independent of prevalence i. e mammography. Sensitivity = 14/15 = 93%; specificity = 91/99 = 92% Similar thing can be done with lr-; mostly look at lr+ Lr+ 11. 63; lr- = 0. 08 i. e performance yield. Sensitivity = a / (a + c) = 400 / (400 + 100) = 80%; specificity = d / (b + d) = 98905 / (995 : = 99% Lr+ = sensitivity / (1-specificity) = 0. 8 / (1-0. 99) = 80. Lr+ = 80 a positive test result increases your odds of having the disease by 80 fold. Reliability: extent to which screening test will produce the same or very similar results each time it is administered.
