Understanding Systematic Reviews and Meta‐Analyses

Abstract

Systematic reviews should be distinguished from narrative reviews. In the latter, an editor asks an expert to sum up all of the information that is known about a particular topic. However, the expert is under no constraints regarding what he or she does, or does not, choose to include in the review. As a result, his or her bias can influence the final message. A systematic review, which may or may not be written by experts, typically asks a narrower question, and then answers it using the entirety of the medical literature. The systematic review process includes computer searches to identify the pertinent literature, a statement of the inclusion and exclusion criteria for identified studies, a list of items of interest to extract from each study, a method to assess the quality of each study, a summary of the evidence that has been found (which may or may not involve attempts to combine data), a discussion of the evidence and the limitations of the conclusions, and suggestions for future research efforts. If the data are combined, that process is called meta-analysis. In meta-analysis, an estimate of the reliability of each study is made, and those that appear to be more reliable are weighed more heavily when the data are combined. While systematic reviews depend on a more preplanned method and thus, unlike narrative reviews, contain sections on method, they can be easily read once the reader becomes familiar with the vocabulary.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 13 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.