I am trying to calculate the change in standard deviation for my metaanalysis.
I have the following data available:
- mean for control group at baseline and endpoint
- mean for intervention group at baseline and endpoint
- 95% Confidence interval for control group at baseline and endpoint
- 95% Confidence interval for intervention group at baseline and endpoint
- Number of subjects in control and intervention group
I would like to calculate the difference in standard deviation for control group (sd_baseline and sd_endpoint) and intervention group (sd_baseline and sd_endpoint). I would like to use the cochrane handbook as reference: https://handbook-5-1.cochrane.org/chapter_16/16_1_3_2_imputing_standard_deviations_for_changes_from_baseline.htm
It is stated that "When there is not enough information available to calculate the standard deviations for the changes, they can be imputed."
Does this mean we need to impute the correlation coefficient for all the studies for every different outcomes separately?