QUESTION:

I was wondering whether SDM has the ability to produce results for two groups at two time points (time X group effects)?


ANSWER:

SDM uses the general linear model and should thus potentially be able to conduct, with some modifications or not, many analyses other than the standard ones. However, this may specially depend on what are the data.

For example, if the studies you are going to meta-analyse already report pre-post differences, you can treat these maps as standard maps. Then you can compare groups with SDM. Or more straighforward: if the studies already report the difference between groups in pre-post differences, you can just do the "Mean".

Conversely, I think that things may be much more difficult if the studies report the differences between patients and controls, separately for different time-points. Apart from modifying SDM codes, I think that you would need to know the intra-patient and intra-control covariances or correlations.