We are conducting a meta-analysis that includes primary studies which compare different support surfaces for the prevention of pression ulcers. We have come across a cross-over trial and we are not quite sure of how to introduce data into Revman. We are thinking of adding up results from both periods and making the study similar to a parallel trial but we don´t know if this approach is correct.
see the revman handbook, user guide. Not sure if this is still the same terminology as they may have renamed it from when I looked at it years ago and I maybe forgetting its precise name but there is a large user guide for this software which I've used previously also
I don't think you can simply tabulate the pre cross and post-cross periods separately as this takes away from the actual effect of the intervention in both groups. If secular trends or unmeasured confounders that varied by time affected one period, you would have mismatched groups and may introduce substantial heterogeneity in your meta-analysis. Because this is such a different study compared to what you are pooling, my advice would be to exclude it from the meta-analytical component but keep it in the systematic review of the literature.
Generally it is discouraged to add up the two periods and treat cross-over trials as a parallel study since the confidence intervals are likely to be too wide, and the study will receive too little weight. However the error will be on the conservative side (the cross-over studies will be under-weighted). Elbourne et al discuss methods for including cross-over studies in reviews.
Article Meta-analyses involving cross-over trials: Methodological issues
Article Meta-analyses involving cross-over trials: Methodological issues
- Badges
- Science method
- Similar topics
- Mathematics
- Statistics
- Statistical Techniques
- Meta-Analysis