Individual patient meta-analysis using JASP
Hello forum,
I was wondering if anyone has previously worked with individual patient meta-analysis using JASP. Individual patient meta-analysis can be done with both the one stage or two stage method. I meta looks at biochemical properties of the protein of interest involving rare disease. For my meta, I think are both are appropriate since I have a mix of aggregate(10%) and individual data (90%). I see that there is a feature for generalized linear mixed models, I was wondering if it can be used for individual patient meta. My data does not have any standard deviation or standard error reported. Does anyone have any recommendation.
Any guidance would be really helpful
Priyesh99
Comments
Hi Priyesh99,
Usually the SE can be obtained (or approximated) from the sample size (and an estimate of the effect size), or it can be inferred from a combination of an estimate and a test statistic. There are many good sources and text books on effect sizes and their standard errors. For example: https://bookdown.org/MathiasHarrer/Doing_Meta_Analysis_in_R/es-calc.html
Thank you for the reply but I don't think that SE approximation can be used for studies with individual patient data unless the two step method is used.