Have you ever meta-analysis you didn't like?

S. N. Goodman

Research output: Contribution to journalArticlepeer-review

Abstract

Meta-analysis is a relatively new discipline whose techniques are unfamiliar to many physicians. A meta-analysis should combine the careful thought and synthesis of a good review with the scientific rigor of a good experiment. In the report of a meta-analysis, the data from the combined studies should be displayed so that clinicians can see at a glance the spectrum of results. The justification for combining the results and the sensitivity to underlying assumptions must be made clear. Regardless of the summary number, meta-analyses should shed light on why trials differ, raise research and editorial standards by calling attention to the strengths and weaknesses of the body of research in an area, and provide an objective view of what the research literature says. When a sufficient number of similar studies addresses a topic, a meta-analysis may indicate a quantitative 'truth'; however, the computing of weighted averages is a comparatively small part of the process and should not be seen as its most important contribution.

Original languageEnglish (US)
Pages (from-to)244-246
Number of pages3
JournalAnnals of Internal Medicine
Volume114
Issue number3
StatePublished - 1991

Keywords

  • Clinical trials
  • Meta-analysis
  • Outcome and process assessment (health care)
  • Review literature
  • Statistics

ASJC Scopus subject areas

  • Medicine(all)

Fingerprint Dive into the research topics of 'Have you ever meta-analysis you didn't like?'. Together they form a unique fingerprint.

Cite this