Bei-Hung Chang1, David C Hoaglin. 1. Department of Quantitative Health Sciences, University of Massachusetts Medical School, Worcester, MA.
Abstract
BACKGROUND: Many systematic reviews of randomized clinical trials lead to meta-analyses of odds ratios (ORs). The customary methods of estimating an overall OR involve weighted averages of the individual trials' estimates of the logarithm of the OR. That approach, however, has several shortcomings, arising from assumptions and approximations, that render the results unreliable. Although the problems have been documented in the literature for many years, the conventional methods persist in software and applications. A well-developed alternative approach avoids the approximations by working directly with the numbers of subjects and events in the arms of the individual trials. OBJECTIVE: We aim to raise awareness of methods that avoid the conventional approximations, can be applied with widely available software, and produce more-reliable results. METHODS: We summarize the fixed-effect and random-effects approaches to meta-analysis; describe conventional, approximate methods and alternative methods; apply the methods in a meta-analysis of 19 randomized trials of endoscopic sclerotherapy in patients with cirrhosis and esophagogastric varices; and compare the results. We demonstrate the use of SAS, Stata, and R software for the analysis. RESULTS: In the example, point estimates and confidence intervals for the overall log-odds-ratio differ between the conventional and alternative methods, in ways that can affect inferences. Programming is straightforward in the 3 software packages; an appendix, Suppemental Digital Content 1 (http://links.lww.com/MLR/B335) gives the details. CONCLUSIONS: The modest additional programming required should not be an obstacle to adoption of the alternative methods. Because their results are unreliable, use of the conventional methods for meta-analysis of ORs should be discontinued.
BACKGROUND: Many systematic reviews of randomized clinical trials lead to meta-analyses of odds ratios (ORs). The customary methods of estimating an overall OR involve weighted averages of the individual trials' estimates of the logarithm of the OR. That approach, however, has several shortcomings, arising from assumptions and approximations, that render the results unreliable. Although the problems have been documented in the literature for many years, the conventional methods persist in software and applications. A well-developed alternative approach avoids the approximations by working directly with the numbers of subjects and events in the arms of the individual trials. OBJECTIVE: We aim to raise awareness of methods that avoid the conventional approximations, can be applied with widely available software, and produce more-reliable results. METHODS: We summarize the fixed-effect and random-effects approaches to meta-analysis; describe conventional, approximate methods and alternative methods; apply the methods in a meta-analysis of 19 randomized trials of endoscopic sclerotherapy in patients with cirrhosis and esophagogastric varices; and compare the results. We demonstrate the use of SAS, Stata, and R software for the analysis. RESULTS: In the example, point estimates and confidence intervals for the overall log-odds-ratio differ between the conventional and alternative methods, in ways that can affect inferences. Programming is straightforward in the 3 software packages; an appendix, Suppemental Digital Content 1 (http://links.lww.com/MLR/B335) gives the details. CONCLUSIONS: The modest additional programming required should not be an obstacle to adoption of the alternative methods. Because their results are unreliable, use of the conventional methods for meta-analysis of ORs should be discontinued.
Authors: Joellen M Schildkraut; Lauren C Peres; Traci N Bethea; Fabian Camacho; Deanna Chyn; Emily K Cloyd; Elisa V Bandera; Alicia Beeghly-Fadiel; Loren Lipworth; Charlotte E Joslin; Faith G Davis; Patricia G Moorman; Evan Myers; Heather M Ochs-Balcom; Veronica Wendy Setiawan; Malcolm C Pike; Anna H Wu; Lynn Rosenberg Journal: Cancer Causes Control Date: 2019-06-24 Impact factor: 2.506
Authors: Heather M Ochs-Balcom; Courtney Johnson; Kristin A Guertin; Bo Qin; Alicia Beeghly-Fadiel; Fabian Camacho; Traci N Bethea; Lauren F Dempsey; Will Rosenow; Charlotte E Joslin; Evan Myers; Patricia G Moorman; Holly R Harris; Lauren C Peres; V Wendy Setiawan; Anna H Wu; Lynn Rosenberg; Joellen M Schildkraut; Elisa V Bandera Journal: Br J Cancer Date: 2022-09-22 Impact factor: 9.075
Authors: Carissa van den Berk-Clark; Scott Secrest; Jesse Walls; Ellen Hallberg; Patrick J Lustman; F David Schneider; Jeffrey F Scherrer Journal: Health Psychol Date: 2018-05 Impact factor: 4.267
Authors: Ana Cecília de Sena Oliveira; Bruno da Silva Athanasio; Flávia Cristina de Carvalho Mrad; Monica Maria de Almeida Vasconcelos; Maicon Rodrigues Albuquerque; Débora Marques Miranda; Ana Cristina Simões E Silva Journal: Pediatr Nephrol Date: 2021-05-19 Impact factor: 3.714
Authors: Calista Leung; Julia Pei; Kristen Hudec; Farhud Shams; Richard Munthali; Daniel Vigo Journal: J Med Internet Res Date: 2022-06-15 Impact factor: 7.076
Authors: Colette P Davis; Elisa V Bandera; Traci N Bethea; Fabian Camacho; Charlotte E Joslin; Anna H Wu; Alicia Beeghly-Fadiel; Patricia G Moorman; Evan R Myers; Heather M Ochs-Balcom; Lauren C Peres; Will T Rosenow; Veronica W Setiawan; Lynn Rosenberg; Joellen M Schildkraut; Holly R Harris Journal: Cancer Epidemiol Biomarkers Prev Date: 2021-06-21 Impact factor: 4.254