A model for assessing the sensitivity and specificity of tests subject to selection bias

Diamond, George A.; Rozanski, Alan; Forrester, James S.; Morris, Denise; Pollock, Brad H.; Staniloff, Howard M.; Berman, Daniel S.; Swan, H. J. C.

doi:10.1016/0021-9681(86)90119-0

Cited by 91 publications

(13 citation statements)

References 31 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Diamond et al , 20 in a recent study, suggested that the sensitivity of radionuclide ventriculography estimated in previous reports may have been biased by the inclusion of a preponderance of patients with coronary disease of sufficient severity to justify cardiac catheterization. They estimated that its true sensitivity may be nearer 63 % when an attempt is made to exclude such selection bias.…”

Section: Discussionmentioning

confidence: 99%

The application of radionuclide ventriculography to cardiac screening

Lindsay

Milner

Chandeysson

et al. 1989

Clinical Cardiology

View full text Add to dashboard Cite

Summary:Screening asymptomatic individuals for latent coronary disease often requires sequential testing because exercise electrocardiography typically produces more false positive than true positive results in a population with a low prevalence of coronary disease. Cardiac scintigraphy is a technique that may be employed as a confirmatory test in lieu of coronary arteriography to further evaluate the significance of a positive exercise electrocardiogram.Radionuclide ventriculography was employed in 98 asymptomatic individuals who were considered to be at moderate risk of heart disease after risk factor analysis and exercise electrocardiography. Seventeen (17 %) patients had an abnormal study and underwent cardiac catheterization. Seven had coronary artery disease, two had cardiomyopathy , and eight were normal. Eighty-one (83%) patients had a normal study. Because the sensitivity of radionuclide ventriculography is 63-80%, it was postulated that 2 to 5 individuals with disease were missed. Thus, from a population with an 1*1-14% prevalence of disease, two subsets were identified. A large subset in which a prevalence of 2-6% could be estimated was separated from a much smaller one in which a prevalence of approximately 50 % was demonstrated.

show abstract

Section: Discussionmentioning

confidence: 99%

The application of radionuclide ventriculography to cardiac screening

Lindsay

Milner

Chandeysson

et al. 1989

Clinical Cardiology

View full text Add to dashboard Cite

show abstract

“…The most relevant aspects of this approach are 1) that a ground truth is assumed to exist and be known and 2) that at some point there has to be a defining test or the next best thing (e.g. a gold standard) 6 .…”

Section: Predicting the Taxonomymentioning

confidence: 99%

Medical Image Acquisition and Processing: Clinical Validation

Goris¹

2014

OJMI

View full text Add to dashboard Cite

The validation of medical imaging (processing and acquisition) can be achieved in multiple ways, somewhat influenced by the context. There are three traps to avoid: First reliance on ground truth requires the knowledge of it before the end of the trial, second comparison to gold standards cannot show improvement and finally one needs to deal with confirmation bias. In this paper we discuss those topics and alternative validation schemes.

show abstract

“…Indeed, when the conditions leading to test independence are understood, the utility of testing in a low prevalence disease context can often be dramatically enhanced by a simple random replication of a testing process that samples from the variables contributing to error. to a more accurate "gold standard" testing methodology [2,3]. Although methods for estimating disease prevalence with the use of multiple tests, none of which are gold standard, have been developed [4], and latent class analysis has been used to estimate prevalence, specificity and sensitivity in the absense of a gold standard [5][6][7], there is little guidance for revising diagnostic predictions for a specific patient when evaluating multiple fallible test results.…”

Section: Introductionmentioning

confidence: 99%

Using Generalizability Theory to Evaluate the Applicability of a Serial Bayes Model in Estimating the Positive Predictive Value of Multiple Psychological or Medical Tests

Kreiter¹

2010

PSYCH

View full text Add to dashboard Cite

Introduction: It is a common finding that despite high levels of specificity and sensitivity, many medical tests are not highly effective in diagnosing diseases exhibiting a low prevalence within a clinical population. What is not widely known or appreciated is how the results of retesting a patient using the same or a different medical or psychological test impacts the estimated probability that a patient has a particular disease. In the absence of a 'gold standard' special techniques are required to understand the error structure of a medical test. Generalizability can provide guidance as to whether a serial Bayes model accurately updates the positive predictive value of multiple test results. Methods: In order to understand how sources of error impact a test's outcome, test results should be sampled across the testing conditions that may contribute to error. A generalizability analysis of appropriately sampled test results should allow researchers to estimate the influence of each error source as a variance component. These results can then be used to determine whether, or under what conditions, the assumption of test independence can be approximately satisfied, and whether Bayes theorem accurately updates probabilities upon retesting. Results: Four hypothetical generalizability study outcomes are displayed as variance component patterns. Each pattern has a different practical implication related to achieving independence between test results and deriving an enhanced PPV through retesting an individual patient. Discussion: The techniques demonstrated in this article can play an important role in achieving an enhanced positive predictive value in medical and psychological diagnostic testing and can help ensure greater confidence in a wide range of testing contexts.

show abstract

A model for assessing the sensitivity and specificity of tests subject to selection bias

Cited by 91 publications

References 31 publications

The application of radionuclide ventriculography to cardiac screening

The application of radionuclide ventriculography to cardiac screening

Medical Image Acquisition and Processing: Clinical Validation

Using Generalizability Theory to Evaluate the Applicability of a Serial Bayes Model in Estimating the Positive Predictive Value of Multiple Psychological or Medical Tests

Contact Info

Product

Resources

About