ABSTRACT
Meta-analysis of diagnostic tests aims to combine and contrast evidence from multiple independent studies on test accuracy parameters, such as sensitivity and specificity, and positive and negative predictive values. In this chapter, we present an overview of statistical methods for meta-analysis of diagnostic tests under three scenarios: (1) when the reference test can be considered as a gold standard; (2) when the reference test cannot be considered as a gold standard; and (3) when studies are a mixture of case-control and cohort studies in which partial verification bias may be present when the selection of samples to be evaluated by a reference test is affected by the results of a diagnostic test. For each scenario, we first provide an overview of commonly used statistical approaches, followed by a detailed case study. Finally, recommendations for further research and statistical software packages are provided.
