[Q] Does a network meta-analysis confidence interval rely on the assumption that the effect estimates are all being taken from independent samples? from r/statistics