Print

Print
Hi Catherine,

Yes, this looks correct. Only note that, if you are interested in "determining if there are significant differences between the groups at post-intervention", then the input files aren't the differences, but just the post-interventional (i.e., the longitudinal component would be lost, which doesn't seem what you actually want).

Also, for FA and MD you'd want to correct for the fact that there are two non-independent modalities, which is available in PALM (option -corrmod).

All the best,

Anderson


On 30 October 2017 at 13:55, Catherine Mewborn <[log in to unmask]> wrote:
Hello,

I have posted a couple of questions before and received helpful answers. However, I am still confused about some aspects of my DTI analysis. I have a longitudinal investigation tracking white matter changes in older adults following a year-long nutrition intervention. I am interested in determining if there are significant differences between the groups at post-intervention. I calculated change scores for each group that calculated the difference from pre-to-post intervention and now I want to set up contrasts to use in Randomise that will compare the two groups. I am interested in knowing if there are areas where the intervention group improved (increased FA, decreased RD) significantly more than the control group as well as if there are areas where the intervention group declined (decreased FA, increased RD) significantly more than the control group. I would also like to test whether there are any areas where the control group improved significantly more than the intervention group or decreased significantly more than the intervention group. In other words, I want to look at significant changes in the groups following intervention and determine if there are areas where the intervention group improved or declined significantly more than the control group and vice versa.

I have set up a design file to use in Randomise which looks like this:

Intervention    Control
1               0
1               0
1               0
1               0
1               0
1               0
1               0
1               0
1               0
1               0
1               0
1               0
1               0
1               0
1               0
1               0
1               0
1               0
1               0
1               0
1               0
1               0
1               0
1               0
1               0
1               0
1               0
1               0
1               0
1               0
1               0
1               0
1               0
0               1
0               1
0               1
0               1
0               1
0               1
0               1
0               1
0               1
0               1
0               1
0               1
0               1
0               1

Would the following be the appropriate contrasts to look at 1) increases or decreases in the experimental group significantly greater than control group and 2) increases or decreases in the control group significantly greater than experimental group?

1)     1  -1
2)     -1   1

Thank you for your help!

Best,
Catherine Mewborn