I want to know if it is acceptable to combine cross-sectional studies (with the prevalence from the case group and control group) and case-control studies in a meta-analysis. I have seen several meta-analyses which combined the prevalence odds ratio from cross-sectional studies and odds ratio from case-control studies. I want to know if this is accepted in the field?
It depends on two issues: conceptual and technical. You can combine those odds ratio if both studies consider the same outcome. See the attached article, particularly the "Combining different metrics" section.
Article COSMOS-E: Guidance on conducting systematic reviews and meta...
Thanks, Md. Belal Hossain! Yes, they have similar outcomes with different study designs.
The prevalence odds ratio (POR) is calculated in the same way as the odds ratio (OR). So it is ok to pool these two.
- Badges
- Science method
- Similar topics
- Mathematics
- Statistics
- Statistical Techniques
- Meta-Analysis