Methods for the joint meta‐analysis of multiple tests

Research Synthesis Methods - Tập 5 Số 4 - Trang 294-312 - 2014
Thomas A Trikalinos1,2, David C. Hoaglin1,3,4, Kevin Small5, Norma Terrin6, Christopher H. Schmid1,7
1Center for Evidence-based Medicine, School of Public Health Brown University Providence RI USA
2Department of Health Services, Policy & Practice, School of Public Health Brown University Providence RI USA
3Consulting Statistician Sudbury MA USA
4Department of Quantitative Health Sciences, University of Massachusetts Medical School, Worcester, MA, USA
5Office of Portfolio Analysis National Institutes of Health Bethesda MD USA
6Research Design Center/Biostatistics Research Center, Clinical and Translational Science Institute Tufts University, and Institute for Clinical Research and Health Policy Studies, Tufts Medical Center Boston MA USA
7Department of Biostatistics, School of Public Health Brown University Providence RI USA

Tóm tắt

Existing methods for meta‐analysis of diagnostic test accuracy focus primarily on a single index test. We propose models for the joint meta‐analysis of studies comparing multiple index tests on the same participants in paired designs. These models respect the grouping of data by studies, account for the within‐study correlation between the tests' true‐positive rates (TPRs) and between their false‐positive rates (FPRs) (induced because tests are applied to the same participants), and allow for between‐study correlations between TPRs and FPRs (such as those induced by threshold effects). We estimate models in the Bayesian setting. We demonstrate using a meta‐analysis of screening for Down syndrome with two tests: shortened humerus (arm bone), and shortened femur (thigh bone). Separate and joint meta‐analyses yielded similar TPR and FPR estimates. For example, the summary TPR for a shortened humerus was 35.3% (95% credible interval (CrI): 26.9, 41.8%) versus 37.9% (27.7, 50.3%) with joint versus separate meta‐analysis. Joint meta‐analysis is more efficient when calculating comparative accuracy: the difference in the summary TPRs was 0.0% (−8.9, 9.5%; TPR higher for shortened humerus) with joint versus 2.6% (−14.7, 19.8%) with separate meta‐analyses. Simulation and empirical analyses are needed to refine the role of the proposed methodology. Copyright © 2014 John Wiley & Sons, Ltd.

Từ khóa


Tài liệu tham khảo

10.1148/radiology.173.2.2529580

10.1148/radiology.193.1.8090881

10.7863/jum.1992.11.9.449

10.1097/00006250-199102000-00012

10.1002/pd.1970140603

10.1046/j.1469-0705.1997.10050321.x

Brumfield CG, 1989, Sonographic measurements and ratios in fetuses with Down syndrome, Obstetrics and Gynecology, 73, 644

10.1159/000263928

ChengW SchmidCH TrikalinosTA GatsonisCA.2013.Network meta‐analysis modeling for diagnostic accuracy studies.Society for Research Synthesis Methodology Annual Meeting Providence RI USA.

10.1016/j.jclinepi.2006.06.011

10.1111/j.1471-0528.1989.tb06297.x

Dicke JM, 1989, Fetal biometry as a screening tool for the detection of chromosomally abnormal pregnancies, Obstetrics and Gynecology, 74, 726

10.1214/ss/1177011136

10.1016/0002-9378(90)90686-2

10.1016/0002-9378(95)90162-0

10.1148/radiology.174.3.2137635

10.1016/j.jclinepi.2007.03.016

10.1093/biostatistics/kxl004

Hill LM, 1989, The current role of sonography in the detection of Down syndrome, Obstetrics and Gynecology, 74, 620

10.7326/0003-4819-120-8-199404150-00008

10.1016/0002-9378(93)90620-X

10.1016/0002-9378(95)91484-6

Kliegman RM, 2007, Nelson Textboook of Pediatrics

10.1002/sim.5909

10.7863/jum.1989.8.12.657

10.1016/S0140-6736(97)08468-7

10.1177/0272989X09336144

10.1016/S0002-9378(87)80059-5

10.1016/0002-9378(93)90280-V

10.1177/0272989X09340584

10.1093/biostatistics/kxp032

Lynch L, 1989, Ultrasound detection of Down syndrome: is it really possible?, Obstetrics and Gynecology, 73, 267

10.1016/j.jclinepi.2003.12.019

10.1016/0002-9378(90)90636-L

10.1016/0002-9378(95)91322-X

10.1046/j.1469-0705.1998.12010008.x

10.1016/0002-9378(90)90028-6

10.1016/0002-9378(93)90487-4

10.1038/gim.2011.73

10.1007/BF00140873

10.1016/j.jclinepi.2005.02.022

10.1016/0002-9378(91)90468-7

Rutter CM, 1995, Regression methods for meta‐analysis of diagnostic test data, Academic Radiology, 2, S48

Shah YG, 1990, Biparietal diameter/femur length ratio, cephalic index, and femur length measurements: not reliable screening techniques for Down syndrome, Obstetrics and Gynecology, 75, 186

10.1001/jama.285.8.1044

10.7326/0003-4819-158-7-201304020-00006

10.7326/0003-4819-142-12_Part_2-200506211-00004

10.1007/s11606-012-2029-1

Trikalinos TA, 2012, Simulation‐based Comparison of Methods for the Meta‐analysis of Proportions and Rates

Trikalinos TA, 2013, Towards a Framework for Communicating Confidence in Methodological Recommendations for Systematic Reviews and Meta‐analyses

Trikalinos TA, 2013, Evaluating Practices and Developing Tools for Comparative Effectiveness Reviews of Diagnostic Test Accuracy: Methods for the Joint Meta‐analysis of Multiple Tests

10.1111/j.1471-0528.1998.tb09352.x

10.1016/0029-7844(96)00058-0

Thông tin
Thông tin xuất bản

Research Synthesis Methods

Tập 5 Số 4

294-312

Thông tin tác giả