Two procedures of meta-analysis in clinical trials and interlaboratory studies

Rok, strany: 2003, 155 - 168

The random effects model designed to combine information from several clinical trials is related to an analysis of data from interlaboratory studies. The primary goal of this work is to establish the relationship between the estimator proposed by DerSimonian and Laird [ Meta-analysis in clinical trials Controlled Clinical Trials 7 (1986), 177–188] and the Mandel–Paule method [ Consensus values and weighting factors J. Res. Nat. Bureau Standards 87 (1982), 377–385] used in collaborative studies. Both procedures estimate the overall treatment effect via a moment-type estimator and give approximate formulas for their standard errors. Theoretical properties of these estimates are compared in the assymptotic setting. To this end, a class of weighted means statistics is investigated for situations where the number of sources is large. It is shown that both the Mandel–Paule rule and the DerSimonian–Laird estimator of the between-trials variance are inconsistent in this setting. The results of numerical comparison of mean squared errors of these estimators for a special distribution of within-trials variances are also reported.

ISO 690:

Rukhin, A. 2003. Two procedures of meta-analysis in clinical trials and interlaboratory studies. In Tatra Mountains Mathematical Publications, vol. 26, no.1, pp. 155-168. 1210-3195.

APA:

Rukhin, A. (2003). Two procedures of meta-analysis in clinical trials and interlaboratory studies. Tatra Mountains Mathematical Publications, 26(1), 155-168. 1210-3195.