A Primer on Systematic Review and Meta-analysis in Diabetes Research.

ABSTRACT

A systematic review is a rigorous process that involves identifying, selecting, and synthesizing available evidence pertaining to an a priori-defined research question. The resulting evidence base may be summarized qualitatively or through a quantitative analytic approach known as meta-analysis. Systematic review and meta-analysis (SRMAs) have risen in popularity across the scientific realm including diabetes research. Although well-conducted SRMAs are an indispensable tool in informing evidence-based medicine, the proliferation of SRMAs has led to many reviews of questionable quality and misleading conclusions. The objective of this article is to provide up-to-date knowledge and a comprehensive understanding of strengths and limitations of SRMAs. We first provide an overview of the SRMA process and offer ways to identify common pitfalls at key steps. We then describe best practices as well as evolving approaches to mitigate biases, improve transparency, and enhance rigor. We discuss several recent developments in SRMAs including individual-level meta-analyses, network meta-analyses, umbrella reviews, and prospective meta-analyses. Additionally, we outline several strategies that can be used to enhance quality of SRMAs and present key questions that authors, editors, and readers should consider in preparing or critically reviewing SRMAs.