In a systematic review, if outcome information in some articles are not uniform (ex; some in mean, some in medians & some in proportions) then how to summarize it in forest plot?
Thanks Dr Doi, for your valuable suggestion. So, the better way is to convert them into effect size and then use for data pooling. I would appreciate if you add more on it.
Before converting to a common effect size, the fundamental question to be answered is related to heterogeneity. Assuming there is little heterogeneity among the studies, then conversion to a common effect size OR using a meta-analysis software package that can comine various outcome measures together are the only two options.