A METAANALYTIC METHOD FOR SUMMARIZING DIAGNOSTIC-TEST PERFORMANCES - RECEIVER-OPERATING-CHARACTERISTIC SUMMARY POINT ESTIMATES

The authors have devised a meta-analytic method to summarize diagnosti c test performances, which they describe along with a clinical example wherein they analyze the performances of real-time ultrasonography in eight studies of the detection of proximal deep venous thrombosis of the lower extremity, selected on the basis of specific inclusion crite ria. To evaluate the evidence for fitting a summary receiver operating characteristic (ROC) curve, a test of correlation between the estimat es of true-positive rate (TPR) and false-positive rate (FPR) is perfor med. A high positive correlation argues for summarizing the studies wi th an ROC curve. In the absence of such correlation, a test of homogen eity is applied separately to the estimates of sensitivity (TPR) and s pecificity (TNR) to evaluate whether differences among studies are due to chance alone. If the estimates are homogeneous, a summary point es timate and confidence intervals (CIs) are calculated. As a final step, subgroup analyses can be performed to assess alternative explanations of variability in TPR and FPR. Within groups defined by the presence or absence of symptoms of venous thrombosis, a negative correlation be tween TPR and FPR and homogeneity among studies were found. For sympto matic patients, the summary TPR was 0.97 with a 95% CI of (0.94, 0.99) and the summary TNR was 0.97 (0.95, 0.99). For asymptomatic patients, the summary TPR was 0.66 (0.50, 0.81), and the summary TNR was 0.96 ( 0.90, 0.99). The difference in TPR between symptomatic and asymptomati c patients was statistically significant. There was no evidence that t est referral bias or lack of independent interpretation of test result s influenced these findings. It is concluded that this meta-analytic m ethod represents a quantitative approach to summarizing the performanc es of diagnostic tests. It provides criteria for deciding among the ap proaches of fitting a summary ROC curve, calculating a summary point e stimate, or reporting ranges of TPRs and TNRs. A summary point estimat e is introduced for use when TPR and FPR are not positively correlated but index test characteristics are homogeneous among studies. This me thod also allows examination of alternative explanations of difference s in test performance characteristics such as differences among subgro ups of patients or biases in study design.