Redundancy of single diagnostic test evaluation

Diagnostic research and diagnostic practice frequently do not cohere. Studi es commonly evaluate whether a single test discriminates between disease pr esence and absence, whereas in practice a test is always judged in the cont ext of other information. This study illustrates drawbacks of single-test e valuation and discusses principles of diagnostic research. We used data on 140 patients suspected of pulmonary embolism who had an inconclusive ventil ation-perfusion lung scan. We evaluated three tests: partial pressure of ox ygen in arterial blood (PaO2), x-ray film of the thorax, and leg ultrasound . On the basis of single-test evaluations, ultrasound was most informative. Given a prior probability of 0.27, it had a much better combination of pos itive and negative predictive value (0.71 and 0.21, respectively) relative to thorax x-ray (0.33 and 0.11) and PaO2 (0.35 and 0.27). The combination o f positive and negative likelihood ratio was also more promising for ultras ound (7.3 and 0.7) than for thorax x-ray (1.3 and 0.3) and PaO2 (1.3 and 0. 9). As the tests are always performed after the history and physical, we ju dged their added. value using multivariable logistic modeling with receiver operating characteristic (ROC) analyses. The ROC areas of the model, inclu ding history and physical, with additional PaO2, thorax x-ray, or ultrasoun d, were 0.75, 0.77, 0.81, and 0.81, respectively, which indicated similar a dded value of thorax x-ray and ultrasound. Application of the models to pat ient subgroups also yielded added predictive value for thorax x-ray film. T hus, the results of single-test evaluations may be very misleading. As no d iagnosis is based on one test, single-test evaluations have limited value i n diagnostic research and only have relevance in the context of screening a nd the initial phase of test development. Diagnostic research should always apply an approach of constructing, extending, and validating diagnostic mo dels in agreement with routine clinical work up using logistic regression a nalyses.