metaSEM (version 1.2.4)

BCG: Dataset on the Effectiveness of the BCG Vaccine for Preventing Tuberculosis

Description

This dataset includes 13 studies on the effectiveness of the Bacillus Calmette-Guerin (BCG) vaccine for preventing tuberculosis (see van Houwelingen, Arends, & Stijnen (2002) for details).

Usage

data(BCG)

Details

A list of data with the following structure:

Trial

Number of the trials

Author

Authors of the original studies

Year

Year of publication

VD

Vaccinated group with disease

VWD

Vaccinated group without the disease

NVD

Not vaccinated group with disease

NVWD

Not vaccinated group without the disease

Latitude

Geographic latitude of the place where the study was done

Allocation

Method of treatment allocation

ln_OR

Natural logarithm of the odds ratio: log((VD/VWD)/(NVD/NVWD))

v_ln_OR

Sampling variance of ln_OR: 1/VD+1/VWD+1/NVD+1/NVWD

ln_Odd_V

Natural logarithm of the odds of the vaccinated group: log(VD/VWD)

ln_Odd_NV

Natural logarithm of the odds of the not vaccinated group: log(NVD/NVWD)

v_ln_Odd_V

Sampling variance of ln_Odd_V: 1/VD+1/VWD

cov_V_NV

Sampling covariance between ln_Odd_V and ln_Odd_NV: It is always 0

v_ln_Odd_NV

Sampling variance of ln_Odd_NV: 1/NVD+1/NVWD

References

Berkey, C. S., Hoaglin, D. C., Mosteller, F., & Colditz, G. A. (1995). A random-effects regression model for meta-analysis. Statistics in Medicine, 14, 395--411.

van Houwelingen, H. C., Arends, L. R., & Stijnen, T. (2002). Advanced methods in meta-analysis: Multivariate approach and meta-regression. Statistics in Medicine, 21, 589--624.

Viechtbauer, W. (2010). Conducting meta-analyses in R with the metafor package. Journal of Statistical Software, 36(3), 1--48. http://www.jstatsoft.org/v36/i03/.

Examples

# NOT RUN {
data(BCG)

## Univariate meta-analysis on the log of the odds ratio
summary( meta(y=ln_OR, v=v_ln_OR, data=BCG,
              x=cbind(scale(Latitude,scale=FALSE),
              scale(Year,scale=FALSE))) )

## Multivariate meta-analysis on the log of the odds
## The conditional sampling covariance is 0
bcg <- meta(y=cbind(ln_Odd_V, ln_Odd_NV), data=BCG,
            v=cbind(v_ln_Odd_V, cov_V_NV, v_ln_Odd_NV))
summary(bcg)

plot(bcg)
# }