fixedMetaAnalMD {CRTSize}R Documentation

Fixed Effects Meta-Analysis for Mean Differences

Description

This function performs a fixed effects meta-analysis of mean differences (a continuous effect measure) from a matrix of mean differences, lower and upper confidence limits.

Usage

fixedMetaAnalMD(data, alpha=0.05)

Arguments

data

A matrix with individual studies in each row. The first column contains the estimate of the Mean Difference (Treatment - Control), the second column contains the lower 95 % confidence limit and the third contains the 95 % upper limit. For simplicity, this function assumes that each of the inputed lower and upper limits that are 95 % confidence limits.

alpha

The desired type I error rate for calculation of confidence limits for the fixed effects mean difference.

Details

This function performs an elementary fixed effects meta-analysis of continous outcome measures from an input matrix of mean differences and their respective confidence limits for any number of studies. This function is designed for use with n4meansMeta to provide sample size estimation based on an updated meta-analysis. Note that it is assumed that the 95 % confidence limits are correctly calculated, and that this function, being elementary in nature only accepts the mean difference as the effect measure. Additional utilities for a meta-analysis or meta-regression can be found in Viechbauer (2006).

Value

data

The data matrix is returned.

thetaF

The fixed effects outcome measure, the mean difference.

lF

The 100(1 - α) % lower limit of the pooled mean difference.

uF

The 100(1 - α) % upper limit of the pooled mean difference.

Var

The variance of the fixed effects treatment mean difference.

Sig

Does this show a statistically significant benefit or harm? (Binary: zero (nonsignificant) or one (significant)).

alpha

The desired type I error rate.

Author(s)

Michael Rotondi, mrotondi@yorku.ca

References

Matthews JNS. Introduction to Randomized Controlled Clinical Trials (2nd Ed.) Chapman & Hall: New York, 2006.

Donner A and Klar N. Design and Analysis of Cluster Randomization Trials in Health Research. Arnold: London, 2000.

White IR and Thomas J. Standardized mean differences in individually-randomized and cluster-randomized trials, with applications to meta-analysis. Clinical Trials 2005. 2:141-151.

Viechtbauer, W. Metafor: A meta-analysis package for R. 2010. Available at: http://www.metafor-project.org/

See Also

fixedMetaAnalRROR, n4meansMeta

Examples

fixedMetaAnalMD(data=rbind(c(100, 50, 150), c(25, -100, 150), c(-90, -190, 10), 
c(-125, -200, -50)), alpha=0.05);

[Package CRTSize version 1.0 Index]