Hi,

We are currently working on a SDM meta-analysis of VBM studies and would need some clarifications regarding the exclusion criteria. Our analysis will include the raw SPM from most of the studies included in the meta-analysis

Based on this, would it be accurate to have the following exclusion criteria: 1) studies that did not include a control group, 2) did not report whole-brain results using stereotactic coordinates and did not use significance thresholds corrected for multiple comparisons, or uncorrected with spatial extent thresholds, (3) used different significance or extent thresholds throughout the whole brain, (4) included duplicated data sets, and (5) used less than 10 participants per group.

These are some of the exclusion criteria that we have come across in the literature, but we are not sure about criteria 2, 3 and 5). the last one seems a bit arbitrary.

Given that we will have raw SPM for most, but not all, studies, are criteria 2 and 3 necessary? Finally should criteria 2) be broken down into a) did not report whole-brain results using stereotactic coordinates, b)did not use significance thresholds corrected for multiple comparisons or uncorrected with spatial extent thresholds.

Thank you in advance for your help!

Stephane

Dear Stephane,

In criterion 2), I think there is not a clear need to discard findings found using uncorrected height thresholds without spatial extent thresholds; indeed, I would reduce point 2) to the a) half. Point 5) seems reasonable, the number 10 may seem arbitrary but I feel that any other number would also indeed be somehow arbitrary...

Hope this helps,

Joaquim