Interpretation of tests of heterogeneity and bias in meta‐analysis

Journal of Evaluation in Clinical Practice - Tập 14 Số 5 - Trang 951-957 - 2008
John P. A. Ioannidis1
1Department of Hygiene and Epidemiology, University of Ioannina School of Medicine, Ioannina, Greece. [email protected]

Tóm tắt

AbstractStatistical tests of heterogeneity and bias, in particular publication bias, are very popular in meta‐analyses. These tests use statistical approaches whose limitations are often not recognized. Moreover, it is often implied with inappropriate confidence that these tests can provide reliable answers to questions that in essence are not of statistical nature. Statistical heterogeneity is only a correlate of clinical and pragmatic heterogeneity and the correlation may sometimes be weak. Similarly, statistical signals may hint to bias, but seen in isolation they cannot fully prove or disprove bias in general, let alone specific causes of bias, such as publication bias in particular. Both false‐positive and false‐negative signals of heterogeneity and bias can be common and their prevalence may be anticipated based on some rational considerations. Here I discuss the major common challenges and flaws that emerge in using and interpreting statistical tests of heterogeneity and bias in meta‐analyses. I discuss misinterpretations that can occur at the level of statistical inference, clinical/pragmatic inference and specific cause attribution. Suggestions are made on how to avoid these flaws, use these tests properly and learn from them.

Từ khóa


Tài liệu tham khảo

10.1136/bmj.312.7023.71

10.1001/jama.293.19.2362

10.1016/S0140-6736(97)08468-7

Sutton A., 2000, Methods for Meta‐analysis in Medical Research

10.1093/aje/kwj069

10.1001/jama.1990.03440100097014

10.1016/0140-6736(91)90201-Y

10.1002/0470870168

10.2307/2533446

10.1136/bmj.315.7109.629

10.1080/01621459.2000.10473905

10.1001/jama.295.6.676

10.1002/sim.2380

10.1503/cmaj.060410

10.1016/j.jclinepi.2005.01.006

10.2307/3001666

10.7326/0003-4819-127-9-199711010-00008

10.1002/sim.1186

10.1136/bmj.327.7414.557

10.1002/sim.2692

10.1136/bmj.39343.408449.80

10.1002/1097-0258(20000715)19:13<1707::AID-SIM491>3.0.CO;2-P

10.1016/j.jclinepi.2006.02.013

10.1001/jama.298.10.1209

10.1371/journal.pmed.0020124

10.1037/1082-989X.11.2.193

10.1002/sim.698

10.1136/bmj.333.7568.597

10.1002/sim.1224

10.7326/0003-4819-116-1-78

10.1056/NEJMsr077003

10.1001/jama.298.8.880

10.1191/1740774505cn087oa

10.1093/aje/kwf031

Ioannidis J. P., 2005, Publication Bias in Meta‐analysis: Prevention, Assessment and Adjustments

10.1016/S0197-2456(97)00097-4

10.1093/oxfordjournals.aje.a009590

10.1016/S0895-4356(97)00149-2

10.1002/(SICI)1097-0258(19960830)15:16<1713::AID-SIM331>3.0.CO;2-D

10.1002/sim.4780070807

10.1002/sim.1221

Patsopoulos N. A. Evangelou E.&Ioannidis J. P.(2008)Sensitivity of between‐study heterogeneity in meta‐analysis: proposed metrics and empirical evaluation.International Journal of Epidemiology(in press) [Epub ahead of print 18 April 2008].

10.1016/j.jclinepi.2003.12.001

10.1016/S0895-4356(01)00414-0

10.1002/sim.1187

10.1002/sim.1752

10.1136/bmj.38356.424606.8F

10.1001/jama.291.20.2457

10.1371/journal.pmed.0040079

10.1177/1740774507079441

10.1016/S0140-6736(07)60894-0

10.1016/0895-4356(94)00106-Z

10.1073/pnas.98.3.831

10.1016/S0895-4356(98)00159-0

10.1016/S0895-4356(01)00377-8

10.1016/S0895-4356(99)00204-8

10.1371/journal.pmed.0040026

Thông tin
Thông tin xuất bản

Journal of Evaluation in Clinical Practice

Tập 14 Số 5

951-957

Thông tin tác giả