Empirical evaluation suggests Copas selection model preferable to trim-and-fill method for selection bias in meta-analysis

被引:81
作者
Schwarzer, Guido [1 ,2 ]
Carpenter, James [1 ,3 ]
Ruecker, Gerta [1 ,2 ]
机构
[1] Univ Med Ctr, Inst Med Biometry & Med Informat, D-79104 Freiburg, Germany
[2] Univ Med Ctr, German Cochrane Ctr, D-79104 Freiburg, Germany
[3] London Sch Hyg & Trop Med, Med Stat Unit, London WC1, England
关键词
Meta-analysis; Publication bias; Funnel plot asymmetry; Copas selection model; Trim-and-fill method; Empirical evaluation; PUBLICATION BIAS;
D O I
10.1016/j.jclinepi.2009.05.008
中图分类号
R19 [保健组织与事业(卫生事业管理)];
学科分类号
摘要
Objective: Meta-analysis yields a biased result if published studies represent a biased selection of the evidence. Copas proposed a selection model to assess the sensitivity of meta-analysis conclusions to possible selection bias. An alternative proposal is the trim-and-fill method. This article reports an empirical comparison of the two methods. Study Design and Setting: We took 157 meta-analyses with binary outcomes, analyzed each one using both methods, then performed an automated comparison of the results. We compared the treatment estimates, standard errors, associated P-values, and number of missing studies estimated by both methods. Results: Both methods give similar point estimates, but standard errors and P-values are systematically larger for the trini-and-fill method. Furthermore, P-values from the trim-and-fill method are typically larger than those from the usual random effects model when no selection bias is detected. By contrast, P-values from the Copas selection model and the usual random effects model are similar in this setting. The trim-and-fill method reports more missing studies than the Copas selection model, unless selection bias is detected when the position is reversed. Conclusions: The assumption that the most extreme studies are missing leads to excessively conservative inference in practice for the trim-and-fill method. The Copas selection model appears to be the preferable approach. (C) 2010 Elsevier Inc. All rights reserved.
引用
收藏
页码:282 / 288
页数:7
相关论文
共 25 条
  • [1] [Anonymous], 2007, R News
  • [2] CARPENTER J, 2009, COPAS R PAC IN PRESS
  • [3] Empirical evaluation showed that the Copas selection model provided a useful summary in 80% of meta-analyses
    Carpenter, James R.
    Schwarzer, Guido
    Ruecker, Gerta
    Kuenstler, Rita
    [J]. JOURNAL OF CLINICAL EPIDEMIOLOGY, 2009, 62 (06) : 624 - 631
  • [4] What works?: selectivity models and meta-analysis
    Copas, J
    [J]. JOURNAL OF THE ROYAL STATISTICAL SOCIETY SERIES A-STATISTICS IN SOCIETY, 1999, 162 : 95 - 109
  • [5] Copas J, 2000, Biostatistics, V1, P247, DOI 10.1093/biostatistics/1.3.247
  • [6] A sensitivity analysis for publication bias in systematic reviews
    Copas, JB
    Shi, JQ
    [J]. STATISTICAL METHODS IN MEDICAL RESEARCH, 2001, 10 (04) : 251 - 265
  • [7] METAANALYSIS IN CLINICAL-TRIALS
    DERSIMONIAN, R
    LAIRD, N
    [J]. CONTROLLED CLINICAL TRIALS, 1986, 7 (03): : 177 - 188
  • [8] A nonparametric "trim and fill" method of accounting for publication bias in meta-analysis
    Duval, S
    Tweedie, R
    [J]. JOURNAL OF THE AMERICAN STATISTICAL ASSOCIATION, 2000, 95 (449) : 89 - 98
  • [9] Trim and fill: A simple funnel-plot-based method of testing and adjusting for publication bias in meta-analysis
    Duval, S
    Tweedie, R
    [J]. BIOMETRICS, 2000, 56 (02) : 455 - 463
  • [10] Duval S, 2005, PUBLICATION BIAS IN META-ANALYSIS: PREVENTION, ASSESSMENT AND ADJUSTMENTS, P127, DOI 10.1002/0470870168.ch8