My systematic review has 9 unique RCT studies and 4 additional reports which are secondary analysis of the parent RCT. Considering that the reports are from the same parent studies, do I need to show them individually in the quality analysis table?
I chose Cochrane rob2 as my quality assessment tool of choice.
Hello Zafar Khan you can use a table to present the results of each study and a text summarizing all results together. For these 4 reports that are part of a unique RCT, I think you should only present the RCT that is going to be included in the review. This post hoc analysis should not be included and can be one specific objective of your review.
I hope this helps
Zafar Khan In general where you have 1 RCT reported in X papers you present one risk of bias assessment for the entire study, so that is an accepted approach. However, the risk of bias may vary for different outcomes reported in different papers (or indeed within the same paper) - for example, because methods of outcome assessment or follow-up rates differed. In this case, it may be an entirely appropriate thing to do a per-outcome risk of bias assessment - but that also applies when you have several outcomes reported in one paper. This is the approach recommended for assessing the risk of bias in observational studies. However, it could all get rather tedious - my suggestion is to use your judgement and do what best reflects the quality of the evidence you have and if a single risk of bias assessment reasonably reflects overall bias (and your primary outcome in particular) that is OK.
- Badges
- Science method
- Similar topics
- Mathematics
- Statistics
- Statistical Techniques
- Meta-Analysis