Accounting for baseline differences in meta-analysis.

The article offers an overview of meta-regression models and possible assumptions that can be employed in mental health disorder trials to account for differences in baseline characteristics of participants between the studies of a meta-analysis. Two datasets of antidepressant trials including fluoxetine, venlafaxine, nefazodone and parosetine that applied the meta-regression models versus placebo are presented. The paper investigates the underlying risks for developing depression.