Dear All, I?m fairly new to R and would like to meta-analyze data where I only have available the differences between the outcomes for the experimental groups and the control groups, as well as the SEMs and SDs of those delta values. However it appears that metafor, as well as other meta-analysis packages I?ve looked at, expect the mean effect sizes of the two groups and their SDs as input. Is it possible to use the delta values mentioned above as input for a meta-analysis using the metafor package? And if not, does anyone know of an R package with which this can be done? Best regards Felix
[R-meta] Meta-analysis with R, use treatment effect as input instead of 2 groups?
3 messages · Felix M, Guido Schwarzer
Am 17.10.17 um 17:57 schrieb Felix M:
Dear All, I?m fairly new to R and would like to meta-analyze data where I only have available the differences between the outcomes for the experimental groups and the control groups, as well as the SEMs and SDs of those delta values.
I assume that you have the following information for each study: delta = mean difference between experimental and control group SEM = standard error of delta SD = standard deviation of delta In order to conduct a meta-analysis you only need delta and SEM. You can use, for example, metagen() in R package meta in the following way: metagen(delta, SEM) Similarly, you can use rma() in R package metafor: rma(yi = delta, sei = SEM) See the corresponding help pages for more information, e.g. help(metagen). Best wishes, Guido
Dr. Guido Schwarzer (sc at imbi.uni-freiburg.de) Institute for Medical Biometry and Statistics Stefan-Meier-Str. 26, D-79104 Freiburg | Phone: +49 (0)761 203 6668 http://www.imbi.uni-freiburg.de | Fax: +49 (0)761 203 6680
Dear Guido, thank you very much for the swift reply. Yes, that is indeed the information I have available for each study. I will start testing this right away! I may have more detailed questions later. I'll study the help pages as well of course. I just didn't know where to start on this particular issue, and I figured it would be best to just ask people who already have lots of experience with the software. I'm glad this mailing list exists! Best regards, Felix 2017-10-17 20:18 GMT+02:00 Guido Schwarzer <sc at imbi.uni-freiburg.de>:
Am 17.10.17 um 17:57 schrieb Felix M:
Dear All, I?m fairly new to R and would like to meta-analyze data where I only have available the differences between the outcomes for the experimental groups and the control groups, as well as the SEMs and SDs of those delta values.
I assume that you have the following information for each study: delta = mean difference between experimental and control group SEM = standard error of delta SD = standard deviation of delta In order to conduct a meta-analysis you only need delta and SEM. You can use, for example, metagen() in R package meta in the following way: metagen(delta, SEM) Similarly, you can use rma() in R package metafor: rma(yi = delta, sei = SEM) See the corresponding help pages for more information, e.g. help(metagen). Best wishes, Guido -- Dr. Guido Schwarzer (sc at imbi.uni-freiburg.de) Institute for Medical Biometry and Statistics Stefan-Meier-Str. 26, D-79104 Freiburg | Phone: +49 (0)761 203 6668 http://www.imbi.uni-freiburg.de | Fax: +49 (0)761 203 6680