A model for assessing the sensitivity and specificity of tests subject to selection bias. Application to exercise radionuclide ventriculography for diagnosis of coronary artery disease

摘要: A probabilistic model was developed which allows one to estimate sensitivity and specificity of diagnostic tests for coronary artery disease without reference angiography. The feasibility the evaluated first in a series computer simulations, then applied assessment ejection fraction 933 patients prior myocardial infarction who underwent exercise radionuclide ventriculography. In 196 were referred angiography, conventional abnormal criterion--an absolute rise less than 0.05 with exercise--had 79% 68% when referenced 737 not angiography analyzed instead by our model, 63% (p = 0.004 compared that 119 angiographically diseases patients) 0.036 77 angiographic normals). Both higher lower catheterized are consistent preferential referral positive test responders negative away from distortion results this selection bias can be circumvented substituting ascertainment.