Network-meta analysis made easy: detection of inconsistency using factorial analysis-of-variance models

BMC Medical Research Methodology

Hans-Peter Piepho¹ 

[1] Bioinformatics Unit, Institute of Crop Science, University of Hohenheim, Fruwirthstrasse 23, 70599 Stuttgart, Germany

关键词: Studentized residual;    PRESS residual;    Pairwise treatment contrast;    Network meta-analysis;    Linear mixed model;    Inconsistency;    Heterogeneity;    Baseline contrast;    Analysis of variance;   

Others  :  866346
DOI  :  10.1186/1471-2288-14-61

 received in 2014-02-08, accepted in 2014-04-09,  发布年份 2014

【 摘 要 】

Background

Network meta-analysis can be used to combine results from several randomized trials involving more than two treatments. Potential inconsistency among different types of trial (designs) differing in the set of treatments tested is a major challenge, and application of procedures for detecting and locating inconsistency in trial networks is a key step in the conduct of such analyses.

Methods

Network meta-analysis can be very conveniently performed using factorial analysis-of-variance methods. Inconsistency can be scrutinized by inspecting the design × treatment interaction. This approach is in many ways simpler to implement than the more common approach of using treatment-versus-control contrasts.

Results

We show that standard regression diagnostics available in common linear mixed model packages can be used to detect and locate inconsistency in trial networks. Moreover, a suitable definition of factors and effects allows devising significance tests for inconsistency.

Conclusion

Factorial analysis of variance provides a convenient framework for conducting network meta-analysis, including diagnostic checks for inconsistency.

【 授权许可】

   
2014 Piepho; licensee BioMed Central Ltd.

【 图 表 】

【 参考文献 】

• [1]Krahn U, Binder H, König J: A graphical tool for locating inconsistency in network meta-analysis. BMC Med Res Methodol 2013, 13:35. BioMed Central Full Text
• [2]Donegan S, Williamson P, D’Allesandro U, Smith CT: Assessing key assumptions of network meta-analysis: a review of methods. Res Synth Meth 2013, 4(4):291-323.
• [3]Lu G, Ades AE: Combining direct and indirect evidence in mixed treatment comparisons. Stat Med 2004, 23:3105-3124.
• [4]Higgins JPT, Jackson D, Barrett JK, Lu G, Ades AE, White IR: Consistency and inconsistency in network meta-analysis: concepts and models for multi-arm studies. Res Synth Meth 2012, 3:98-110.
• [5]White IR, Barrett JK, Jackson D, Higgins JPT: Consistency and inconsistency in network meta-analysis: model estimation using multivariate meta-regression. Res Synth Meth 2012, 3:111-125.
• [6]Piepho HP, Williams ER, Madden LV: The use of two-way mixed models in multi-treatment meta-analysis. Biometrics 2012, 68:1269-1277.
• [7]Senn S, Gavini F, Magrez D, Scheen A: Issues in performing network meta-analysis. Stat Methods Med Res 2013, 22(2):169-189.
• [8]Searle SR, Casella G, McCulloch CE: Variance Components. New York: John Wiley & Sons; 1992.
• [9]Wilkinson GN, Rogers CE: Symbolic description of factorial models for analysis of variance. Appl Stat 1973, 22:392-399.
• [10]McCullagh P, Nelder JA: Generalized Linear Models. 2nd edition. London: Chapman & Hall; 1989.
• [11]Piepho HP, Büchse A, Richter C: A mixed modelling approach to randomized experiments with repeated measures. J Agron Crop Sci 2004, 190:230-247.
• [12]Lu G, Welton NJ, Higgins JPT, White IR, Ades AE: Linear inference for mixed treatment comparison meta-analysis: A two-stage approach. Res Synth Meth 2011, 2:43-60.
• [13]Möhring J, Piepho HP: Comparison of weighting in two-stage analyses of series of experiments. Crop Sci 2009, 49:1977-1988.
• [14]Piepho HP, Möhring J, Schulz-Streeck T, Ogutu JO: A stage-wise approach for analysis of multi-environment trials. Biom J 2012, 54:844-860.
• [15]Whitehead A: Meta-Analysis of Controlled Clinical Trials. New York: Wiley & Sons; 2002.
• [16]Nelder JA: Linear models: back to basics. Stat Comput 1994, 4:221-234.
• [17]Kenward MG, Roger JH: Small sample inference for fixed effects from restricted maximum likelihood. Biometrics 1997, 53:983-997.
• [18]Belsley DA, Kuh E, Welsch RE: Regression Diagnostics; Identifying Influential Data and Sources of Collinearity. New York: John Wiley & Sons; 1980.
• [19]Cook RD, Weisberg S: Residuals and Influence in Regression. New York: Chapman and Hall; 1982.
• [20]Edwards D, Berry JJ: The efficiency of simulation-based multiple comparisons. Biometrics 1987, 43:913-928.
• [21]Piepho HP: A SAS macro for generating letter displays of pairwise mean comparisons. Commun Biometry Crop Sci 2012, 7:4-13.