sdm-help-list > Double counting of control groups
Showing 1-4 of 4 posts
Display:
Results per page:
Jul 5, 2016  12:07 PM | Chandan Shah
Double counting of control groups
Dear Dr. Radua,

First of all I would like to thank you once again for your previous suggestion. I again would like to request you for a suggestion as I was asked by one the reviewers to take your expert opinion before they consider for the publication of my paper.
I did a meta-analysis using ES-SDM in which I included several studies including that of Guo et.al 2013 (Duration of untreated psychosis is associated with temporal and occipitotemporal gray matter volume decrease in treatment naive schizophrenia. PLoS One 8, e83679), and Deng et.al 2009 ( A naturalistic study of grey matter volume increase after early treatment in anti-psychotic naive, newly diagnosed schizophrenia. Psychopharmacology (Berl) 206, 437-46). Both of the studies used similar numbers of control groups but had different patient groups. The reviewer has concerns about double counting of control groups as he believes that in standard meta-analyses this is sometimes fixed by having a penaly for the control group and halfing this. I therefore request you to kindly put a light on this as I was not sure whether it would effect my results or not.
Thanking you in anticipation and hoping to hearing from you soon.
Sincerely yours,
Dr. Chandan Shah
Jul 6, 2016  04:07 PM | Joaquim Radua
RE: Double counting of control groups
Dear Shah,

I think one option could consist in repeating the meta-analysis excluding one of the studies, in order to check whether the results are the same. We have used this approach on several occasions.

Best wishes,

Joaquim
Jul 20, 2016  08:07 AM | Chandan Shah
RE: Double counting of control groups
Dear Dr. Radua,
Thank you for the suggestion. I however have one more question. What happens if (in case ) there is a change in result. How do we approach then? 
Thanking you in anticipation and hoping to hearing from you soon. 

Sincerely yours, 
Dr. Chandan Shah
Sep 16, 2016  12:09 PM | Joaquim Radua
RE: Double counting of control groups
Dear Shah,

In that case I would suggest to report it and note that this result may be weaker.

Best wishes,

Joaquim