[R-meta] Meta Analysis with HR values

Awais Wahab @||gner@79 @end|ng |rom gm@||@com
Wed Feb 13 13:41:39 CET 2019

Dear Doctor Guido!
Thank you very much for your much valuable e-mail.

I will start working on it and will let you know if there is any concern.

best regards,

On Wed, Feb 13, 2019 at 1:44 PM Guido Schwarzer <sc using imbi.uni-freiburg.de>

> Dear Awais,
> Survival data using the hazard ratio as effect measure are typically
> meta-analysed using the inverse variance method. The following general R
> packages for meta-analysis provide this method: *epiR*, *meta*, *metafor*,
> *rmeta* (see https://cran.r-project.org/web/views/MetaAnalysis.html).
> The inverse variance method needs, for each study, the *log* hazard ratio
> and its standard error as input. You write that you have hazard ratios and
> CIs which is sufficient as standard errors can be calculated from CIs (see
> Parmar et al., 1999).
> Please find attached R code using R package *meta* which conducts a
> meta-analysis for survival data from a Cochrane review I was involved.
> Note, the minor differences in the results of the two meta-analyses can be
> explained by rounding hazard ratios and confidence limits to only two
> significant digits; we would get identical results with more significant
> digits.
> Best wishes, Guido
> Reference: Parmar MK, Torri V, Stewart L. Extracting summary statistics to
> perform meta-analyses of the published literature for survival endpoints.
> Statistics in Medicine. 1998;17(24):2815–34.
> --
> Dr. Guido Schwarzer
> Institute of Medical Biometry and Statistics,
> Faculty of Medicine and Medical Center - University of Freiburg
> Postal address: Stefan-Meier-Str. 26, D-79104 Freiburg
> Phone: +49/761/203-6668
> Mail: sc using imbi.uni-freiburg.de
> Homepage: http://www.imbi.uni-freiburg.de

	[[alternative HTML version deleted]]

More information about the R-sig-meta-analysis mailing list