Message-ID: <7455d7b1-3715-ea34-56ee-3f52c105a43e@imbi.uni-freiburg.de>
Date: 2021-10-05T15:59:21Z
From: Guido Schwarzer
Subject: [R-meta] meta-analysis of mean differences with only the difference and its SD
In-Reply-To: <DM5PR2001MB0907761942FA8480EF36CC80ACAF9@DM5PR2001MB0907.namprd20.prod.outlook.com>
Am 05.10.21 um 16:00 schrieb Martin Lobo:
> Hi Gerta, I'm sorry, I was later.
>
> I need to do a mean difference analysis. For example, there was a decrease in LDL with a drug versus a placebo. But I do not have the average value of each branch with its standard deviations. I have the difference of each branch with its confidence intervals. [...]
You can use metagen() from R package meta if you have mean differences
and CI limits. Note, you should know how the original confidence
intervals were calculated, i.e. based on normal approximation or
t-distribution, and provide the corresponding information (see attached
R code).
Best, Guido
-------------- next part --------------
An embedded and charset-unspecified text was scrubbed...
Name: metagen09-ci.R
URL: <https://stat.ethz.ch/pipermail/r-sig-meta-analysis/attachments/20211005/ae72d06b/attachment.ksh>