Title | A composite likelihood method for bivariate meta-analysis in diagnostic systematic reviews. |
Publication Type | Journal Article |
Year of Publication | 2017 |
Authors | Chen, Yong, Yulun Liu, Jing Ning, Lei Nie, Hongjian Zhu, and Haitao Chu |
Journal | Stat Methods Med Res |
Volume | 26 |
Issue | 2 |
Pagination | 914-930 |
Date Published | 2017 Apr |
ISSN | 1477-0334 |
Keywords | Analysis of Variance, Biostatistics, Computer Simulation, Diagnostic Tests, Routine, Humans, Likelihood Functions, Linear Models, Melanoma, Meta-Analysis as Topic, Odds Ratio, Sensitivity and Specificity, Skin Neoplasms |
Abstract | Diagnostic systematic review is a vital step in the evaluation of diagnostic technologies. In many applications, it involves pooling pairs of sensitivity and specificity of a dichotomized diagnostic test from multiple studies. We propose a composite likelihood (CL) method for bivariate meta-analysis in diagnostic systematic reviews. This method provides an alternative way to make inference on diagnostic measures such as sensitivity, specificity, likelihood ratios, and diagnostic odds ratio. Its main advantages over the standard likelihood method are the avoidance of the nonconvergence problem, which is nontrivial when the number of studies is relatively small, the computational simplicity, and some robustness to model misspecifications. Simulation studies show that the CL method maintains high relative efficiency compared to that of the standard likelihood method. We illustrate our method in a diagnostic review of the performance of contemporary diagnostic imaging technologies for detecting metastases in patients with melanoma. |
DOI | 10.1177/0962280214562146 |
Alternate Journal | Stat Methods Med Res |
Original Publication | A composite likelihood method for bivariate meta-analysis in diagnostic systematic reviews. |
PubMed ID | 25512146 |
PubMed Central ID | PMC4466215 |
Grant List | P30 CA077598 / CA / NCI NIH HHS / United States U54 MD008620 / MD / NIMHD NIH HHS / United States R21 AI103012 / AI / NIAID NIH HHS / United States R03 HS022900 / HS / AHRQ HHS / United States P01 CA142538 / CA / NCI NIH HHS / United States R01 AI130460 / AI / NIAID NIH HHS / United States |
A composite likelihood method for bivariate meta-analysis in diagnostic systematic reviews.
Project: