A review of solutions for diagnostic accuracy studies with an imperfect or missing reference standard

OBJECTIVE: In diagnostic accuracy studies, the reference standard may beimperfect or not available in all patients. We systematically reviewed theproposed solutions for these situations and generated methodological guidance.STUDY DESIGN AND SETTING: Review of methodological articles.RESULTS: We categorized the solutions into four main groups. The first groupincludes methods that impute or adjust for missing data on the referencestandard. The second group consists of methods that correct estimates of accuracyobtained with an imperfect reference standard. In the third group a referencestandard is constructed by combining multiple test results through a predefinedrule, based on a consensus procedure, or through statistical modeling. In thefourth group, the diagnostic accuracy paradigm is abandoned in favor ofvalidation studies that relate index test results to relevant clinical data, suchas history, future clinical events, and response to therapy.CONCLUSION: Most of the methods try to impute, adjust, or construct a referencestandard. In situations that deviate only marginally from the classicaldiagnostic accuracy paradigm, these are valuable methods. In cases where anacceptable reference standard does not exist, the concept of clinical testvalidation may provide an alternative paradigm to evaluate a diagnostic test.