Meta-analyses and meta-regressions: principles and pitfalls through a review of recently published research works

Meta-analysis is a statistical procedure that combines several published studies on a specific subject and infers to a universal outcome. Meta-regression is a method that can be implemented following the traditional meta-analysis and can be regarded as an extension to it. In the present thesis, we e...

Full description

Saved in:
Bibliographic Details
Main Authors: Ευθυμίου, Βασιλική, Efthymiou, Vasiliki
Other Authors: Καραγρηγορίου, Αλέξανδρος
Language:en_US
Published: 2023
Subjects:
Online Access:http://hdl.handle.net/11610/24941
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Meta-analysis is a statistical procedure that combines several published studies on a specific subject and infers to a universal outcome. Meta-regression is a method that can be implemented following the traditional meta-analysis and can be regarded as an extension to it. In the present thesis, we examine the principles and pitfalls of these methods through a review of recently published research works in health science. Our review for meta-analyses and meta-regressions was limited in the two last years (2021 and 2022) and attributed 1,991 studies from which we studied extensively 100 randomly selected published works. The characteristics and pitfalls that we paid attention was; impact factor, number of published studies in metaanalysis, range of the years of the published studies, drug intervention existence, type of analysis, overfitting, visual presentation, method, weighted meta-regression, ecological fallacy and underlying risk. Finally, several techniques are suggested, in order to overcome the pitfalls.