Meta‐Analysis of Diagnostic Tests

Abstract

Abstract Meta‐analysis can summarize the performance of a diagnostic test based on all available high quality studies, investigate reasons for variation in test performance, and compare the performance of two or more tests. Strategies are outlined for finding all relevant primary studies, and assessing the quality of studies. Study quality, study design factors, and patient characteristics should be assessed and explored as possible sources of heterogeneity in diagnostic performance. Summary ROC analysis (SROC) is the most common meta‐analytic method for diagnostic studies. In addition, linear regression is used to model test accuracy as a function of implicit test threshold and other study level covariates. The SROC approach shows the trade‐off in sensitivity and specificity as the threshold for a positive test result varies. The more complex, hierarchical SROC model, takes account of both within and between study variability, and can be used to explore variation in both accuracy and threshold.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 11 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.