I'm currently doing a meta-analysis and I came across a study in which there's a cohort with 1201 patients, 107 were on a particular drug and 1094 were not on that drug. Then they further did 2, 1:1 case-control analysis matching. In total, we have 3 datasets one from the original cohort and 2 from the 1:1 matches so can we use all these 3 datasets separately in our analysis?
By using all the datasets from the single cohort study haphazardly, you will problematically duplicate or perhaps triplicate the statistical weight contribution of that study in your systematic review. It is your responsibility to assess which of the three analyses - the cohort, case-control #1 or case-control #2 - is uniquely relevant for your review. It cannot be all three, especially when they were all drawn from the same pool of study participants. The more important component of the term "meta-analysis" is "analysis" - don't pool data just for the sake of it.
If you intend to find a Cohen D effect size you will need the SD and mean from the Control and experimental group.For the same these set of results are only one set or one paper or one value. You would need some 20-30 papers with the same control and experiemntal group for this same drug to do the meta analysis. You would need the SD and Mean of all these papers from both control and experimental sets
- Badges
- Science topic