[R-meta] How to assess the amount of change within group through the percentage change from baseline?

Viechtbauer, Wolfgang (SP) wo||g@ng@v|echtb@uer @end|ng |rom m@@@tr|chtun|ver@|ty@n|
Sun May 23 19:03:56 CEST 2021

Dear Florencio,

Say x1 and x2 are the variables corresponding to the measurements at baseline and after the treatment and m1 and m2 are the respective means. So you have m1 but what exactly do you mean by 'percentage change'? In other words, do you have the mean of 100(x2-x1)/x1 or do you have 100(m2-m1)/m1? I assume you have the former. If so, then you could just meta-analyze those means directly. In other words, just use:

escalc(measure="MN", mi=mean_of_the_percentage_change_values,


>-----Original Message-----
>From: R-sig-meta-analysis [mailto:r-sig-meta-analysis-bounces using r-project.org] On
>Behalf Of Florencio Sousa
>Sent: Wednesday, 19 May, 2021 19:05
>To: r-sig-meta-analysis using r-project.org
>Subject: [R-meta] How to assess the amount of change within group through the
>percentage change from baseline?
>We are performing a meta-analysis utilizing metafor package and we have one doubt.
>We intent to assess the amount of change within individual group (before and after
>a treatment), but all studies only report at the pre-test (baseline) the absolute
>values (continuo variable – bone mineral density g/cm2), and the percentage change
>from baseline (with respective SD) at the post-test.
>Can we use the percentage change as mean change? In other words, can we use the
>"MC" as for the measure argument, utilize in the m1i the percentage change values,
>and both sd2i and ri to 0? Because ri are not reported, should we set as 0.5, and
>then perform a sensitive analysis (eg. test ri from 0.2 to 0.8) to confirm the
>results robustness?
>Any comments are welcome.
>Best regards.

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