 |
 |
Hi Thais, As written, the reviewer is wrong, that is, randomise can definitely take more than just two covariates. That said, the whole context suggests that the reviewer didn't mean what they wrote. They seem to be concerned with the fact that age, education and interval between scans is different between groups, which reduces power to investigate group differences. Another concern between the lines is that interval between scans could in fact not be considered a confound, but an explanatory variable of interest, and having it in the model as nuisance can further reduce power. Finally, for me it isn't clear whether age was entered in the model as a constant for both timepoints (e.g., baseline age or average age), as opposed to age at the time of scan, in which case it would be confounded with interval between scans. If both age, interval, and the subject-specific intercepts are all nuisance, this isn't really a problem, but if one of these (such as interval between scans) became a variable of interest, then jointly age at scan and subject-specific intercepts would wash any effect away. So yes, the reviewer is wrong regarding the number of covariates, but maybe there is something they are getting at that may need attention... Hope this helps! All the best, Anderson On 24 February 2018 at 09:55, Thais Minett <[log in to unmask]> wrote: > Dear Experts, > > I used TBSS to compare the differences between 3 groups of patients with > Parkinson's disease at baseline and change at follow-up. In my data > analyses, when performing the FSL randomisation I used age, sex, education, > and levo-dopa equivalent dose and when appropriate, interval between scans > as covariates. I have submitted the paper for publication and one of the > reviewers criticised that approach by saying that FSL randomisation can > only take 2 covariates (see next): > > "The three study groups were different in terms of age, education, > interval between scans. Thus, these variables have been added as nuisance > covariates, along with gender and levodopa equivalent daily dose. This may > severely limit the consistency of findings, as the permutation test > performed with randomise FSL tool can be weaken by using more than two > covariates." > > Is that true? Do we have evidence for that? What is the best approach for > me to respond to that? > > Regards, > > Thais Minett >