According to the PRISMA statement, you should both rate the risk of bias in the individual studies (e.g. using the Cochrane Risk of Bias Tool or another relevant tool) and across studies. Is the later the same as using the GRADE approach? I'm in doubt because in the same PRISMA statement it says, that the GRADE approach is used in item 24 " summary of evidence".
GRADE is used to rate the overall body of evidence rather than individual studies
You cannot do more than just looking for heterogeneity and then guess whether the low or the high study is more likely to be wrong, judged by its methods and its general presentation. And keep in mind that the heterogeneity may be a fact and not an artifact.
Dear @Michael Dan Arvig,
Please check methods of our recently published meta-analysis. I hope it will help you.
Prevalence of Type 2 Diabetes Mellitus in Adult Population of Pakistan A Meta-Analysis of Prospective Cross-Sectional Surveys
January 2020Annals of Global Health 86(1):7
http://doi.org/10.5334/aogh.2679
- Badges
- Science topic
- Similar topics
- Microbiology
- Assays