I would like to generate a pooled mean prevalence estimate (by applying a meta-analysis on prevalence studies) that is adjusted to important cofactors such as sample size, year of data collection, age and etc? Any guidance of step by step to do this with STATA or RevMan ? how can I pool the prevalence from different studies?, your recommendation will be highly appreciated
Thank you
Thérèse
When you pool the estimate, you may consider adjust it by using sample size or variance (less variance, more weight), in general. I think that using year of collection or age is rather clinical question, not a statistical one. You may check papers concerning similar co-factors.
Before combining information, I suggest conducting a homogeneous test with Q statistic or I statistic.
Thank you! that is useful...are you familiar of doing that with STATA?
Dear Ms Umuhoza, you may use OpenMeta, very simple user-friendly application. Available at: http://www.cebm.brown.edu/openmeta/download.html
Best regards!
Therese Umuhoza, I can do that in STATA for you, you will need to use the metaprop command to do that.
- Badges
- Science topic
- Similar topics
- Mathematics
- Statistics
- Sampling Design
- Sample Size