A refined method for the meta‐analysis of controlled clinical trials with binary outcome

By J. Hartung, Guido Knapp. Statistics in Medicine, 2001

672 citations recorded by the source corpus.

View Paper: https://www.semanticscholar.org/paper/42e816702485d953c4e92ce79b9bd4b499d0f10b

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Education

Citations in the corpus

On Constructing Confidence Intervals for a Standardized Mean Difference in Meta-analysis (2003, background): “…i¼1 ŵi : ð2:10Þ Here V̂S is an alternative variance estimate for ̂, and can be viewed as the weighted extension of the usual sample variance estimate for ̂, based on the sample effect measures d1, . . . , dk in the random effects model (Sidik and Jonkman, 2002; Hartung and Knapp, 2001a, 2001b).”
Meta-Analysis in Clinical Trials Revisited (2015, methodology): “Several alternative approaches have been proposed, including a weighted least squares approach [29] and using the ‘robust’ variance estimator [27].”
The Hartung-Knapp-Sidik-Jonkman method for random effects meta-analysis is straightforward and considerably outperforms the standard DerSimonian-Laird method (2014, methodology): “The results reported by Hartung, Knapp and Makambi [4-6,8,9] imply that for meta-analyses of three, six or twelve studies the DL error rates for studies with similar sizes were closer to 5% than for studies of different sizes, and that the HKSJ method performed much better than DL in the latter situ”
Simple heterogeneity variance estimation for meta‐analysis (2005, methodology): “In addition, Hartung and Knapp (2001a) studied tests of the null hypothesis of a zero overall log-odds ratio and showed that the normal theory test based on the DerSimonian and Laird (1986) estimator of τ2 is not satisfactory in terms of the size of the test.”
A re-evaluation of random-effects meta-analysis (2009, methodology): “…researchers have investigated the distribution of the Wald statistic T = μ̂=ŜE.μ̂/, with suggestions to use tk−4 (Berkey et al., 1995), tk−2 (Raghunathan and Ii, 1993) or tk−1 (Follmann and Proschan, 1999), or to use a modified Wald statistic with a tk−1-distribution (Hartung and Knapp, 2001a, b).”