A modified test for small‐study effects in meta‐analyses of controlled trials with binary endpoints

Statistics in Medicine - Tập 25 Số 20 - Trang 3443-3457 - 2006
Roger Harbord1, Matthias Egger2,1, Jonathan A C Sterne1
1MRC Health Services Research Collaboration, Department of Social Medicine, University of Bristol, U.K.
2Department of Social and Preventive Medicine, University of Berne, Switzerland

Tóm tắt

AbstractPublication bias and related bias in meta‐analysis is often examined by visually checking for asymmetry in funnel plots of treatment effect against its standard error. Formal statistical tests of funnel plot asymmetry have been proposed, but when applied to binary outcome data these can give false‐positive rates that are higher than the nominal level in some situations (large treatment effects, or few events per trial, or all trials of similar sizes). We develop a modified linear regression test for funnel plot asymmetry based on the efficient score and its variance, Fisher's information. The performance of this test is compared to the other proposed tests in simulation analyses based on the characteristics of published controlled trials. When there is little or no between‐trial heterogeneity, this modified test has a false‐positive rate close to the nominal level while maintaining similar power to the original linear regression test (‘Egger’ test). When the degree of between‐trial heterogeneity is large, none of the tests that have been proposed has uniformly good properties. Copyright © 2005 John Wiley & Sons, Ltd.

Từ khóa


Tài liệu tham khảo

10.1200/JCO.1986.4.10.1529

10.2307/2982993

10.1001/jama.1990.03440100097014

10.1016/0140-6736(91)90201-Y

10.1136/bmj.315.7109.640

10.4159/9780674040243

10.1136/bmj.315.7109.629

Egger M, 2003, How important are comprehensive literature searches and the assessment of trial quality in systematic reviews? Empirical study, Health Technology Assessment, 7, 10.3310/hta7010

10.1016/S0895-4356(00)00242-0

10.1136/bmj.323.7304.101

DeeksJJ HigginsJPT AltmanDG. Analysing and presenting results. In Cochrane Reviewers' Handbook 4.2.2 Alderson P Green S Higgins J (eds) [updated March 2004];http://www.cochrane.org/resources/handbook/handbook.pdf(accessed 17 January 2005)

10.2307/2533446

Irwig L, 1998, Bias in meta‐analysis detected by a simple, graphical test. Graphical test is itself biased (Letter), British Medical Journal, 316, 470

10.1002/sim.698

10.1002/sim.1224

10.1111/j.1469-1809.1955.tb01348.x

10.1002/sim.1761

SilagyC LancasterT SteadL MantD FowlerG.Nicotine replacement therapy for smoking cessation. The Cochrane Database of Systematic Reviews Issue 3 2004.

10.1002/sim.1186

10.1136/bmj.327.7414.557

10.1016/S0895-4356(01)00377-8

10.1002/sim.4780070807

10.1080/00401706.1988.10488400

Steichen TJ, 1998, sbe19.1: tests for publication bias in meta‐analysis, Stata Technical Bulletin, 44, 3

10.1016/S0895-4356(99)00204-8

10.1002/sim.4780101105

Whitehead J, 1992, The Design and Analysis of Sequential Clinical Trials

10.1016/S0033-0620(85)80003-7

Maldonado G, 1997, The importance of critically interpreting simulation studies, Epidemiology, 8, 453

10.1002/sim.1461

Hilbe J, 1995, sg44: random number generators, Stata Technical Bulletin, 28, 20

Hilbe J, 1998, sg44.1: corrections to random number generators, Stata Technical Bulletin, 41, 23

StataCorp, 2003, Stata Statistical Software: Release 8.0

10.1016/0197-2456(86)90046-2

Sackett DL, 1996, Down with odds ratios!, Evidence‐Based Medicine, 1, 164

10.1046/j.1365-2168.2002.02122.x

10.1002/(SICI)1097-0258(19991030)18:20<2693::AID-SIM235>3.0.CO;2-V

10.1007/978-1-4899-3242-6

10.1136/bmj.315.7122.1610

10.1214/ss/1177013012

10.1214/ss/1177011363

10.1177/096228020101000402

10.1111/j.0006-341X.2000.00455.x

10.1080/01621459.2000.10473905

SterneJAC EggerM. High false positive rate for trim and fill method (electronic letter).http://bmj.bmjjournals.com/cgi/eletters/320/7249/15742000(accessed 17 January 2005)

10.1093/ije/31.1.1

10.1002/sim.4780090307

10.1016/S0140-6736(04)16136-9

10.1001/jama.285.15.1987

Thông tin
Thông tin xuất bản

Statistics in Medicine

Tập 25 Số 20

3443-3457

Thông tin tác giả