[Skip to Navigation]
March 2004

A Note on Meta-analysis of Diagnostic Test Studies

Arch Dermatol. 2004;140(3):363-364. doi:10.1001/archderm.140.3.363-b

The meta-analysis by Rosado et al1 contains some methodologic pitfalls. Actually, the aim of a meta-analysis is to providea summary estimate of diagnostic accuracy more precise than the estimate of a single study. However, it is well known that the heterogeneity of estimatesof primary studies should be explored before they are summarized. A pooled estimate can be justified only if primary studies are not heterogeneous.2 In fact, if heterogeneity is identified, pooledestimates are not sensible, and the analysis should remain only at a qualitative level. Explaining heterogeneity would be, in this case, more relevant thansummarizing.