|
Abstract:
|
In evaluating the accuracy of diagnostic tests, imperfect gold standard bias occurs when an imperfect reference test is used as a gold standard. This bias arises in estimating accuracy measures of chest computed tomography (CT) for the diagnosis of coronavirus disease 2019 (COVID-19) since many existing studies treated reverse transcriptase polymerase chain reaction (RT-PCR) test which was known for producing false negative results as the gold standard. Meanwhile, verification bias occurs in some studies where only highly suspected cases based on results of chest CT, blood tests and epidemiological history were verified by RT-PCR test resulting in a non-representative subsample of the original population. Consequently, the accuracy estimates would be biased and lead to invalid inference. Motivated by the problems in the accuracy evaluation of COVID-19 tests, in this talk, we propose a new method for correcting both imperfect gold standard and verification bias in estimating the sensitivity and specificity of two diagnostic tests. The application of our method to the motivating example gives unbiased estimates of accuracy measures for RT-PCR test and chest CT, which can be served a
|
