From: Dealing with missing standard deviation and mean values in meta-analysis of continuous outcomes: a systematic review