Cited by 2,517Meta-analysis and systematic reviewsStatistical Methods in EpidemiologyDelphi Technique in Research

Quantitative Synthesis in Systematic Reviews

Lau, Joseph, Ioannidis, John P.A., Schmid, Christopher H.·1997·Annals of Internal Medicine

Abstract

The final common pathway for most systematic reviews is a statistical summary of the data, or meta-analysis. The complex methods used in meta-analyses should always be complemented by clinical acumen and common sense in designing the protocol of a systematic review, deciding which data can be combined, and determining whether data should be combined. Both continuous and binary data can be pooled. Most meta-analyses summarize data from randomized trials, but other applications, such as the evaluation of diagnostic test performance and observational studies, have also been developed. The statistical methods of meta-analysis aim at evaluating the diversity (heterogeneity) among the results of different studies, exploring and explaining observed heterogeneity, and estimating a common pooled effect with increased precision. Fixed-effects models assume that an intervention has a single true effect, whereas random-effects models assume that an effect may vary across studies. Meta-regression analyses, by using each study rather than each patient as a unit of observation, can help to evaluate the effect of individual variables on the magnitude of an observed effect and thus may sometimes explain why study results differ. It is also important to assess the robustness of conclusions through sensitivity analyses and a formal evaluation of potential sources of bias, including publication bias and the effect of the quality of the studies on the observed effect.

Integrity signals

Registry

Crossref · 10.7326/0003-4819-127-9-199711010-00008

Cites

60 of 60 references verified0 citation errors · 0 not checkable

Statistics

Recomputing p-values needs the paper’s full text.

Claims the paper makes

This paper’s full text isn’t openly available, so its claims and self-check weren’t run. The registry and citation checks still apply.

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

  • Thirty-nine articles (42%) made recommendations on the assessment of quality and risk of bias of individual studies, with publication dates between 1994 and 2016 (median year 2009).

    “Thirty-nine articles (42%) made recommendations on the assessment of quality and risk of bias of individual studies, [12, 16, 29, 30, 32–35, 37, 38, 43, 47–50, 52–54, 56, 64–66, 70, 71, 73, 75, 77, 82, 84–94] with publication dates between 1994 and 2016 (median year of publication 2009).”
  • Twenty (22%) articles reported on item "publication bias", with publication dates between 1994 and 2012 (median year 2001).

    “Twenty (22%) articles reported on item “publication bias” [16, 29, 33, 34, 36, 46, 48–50, 55, 56, 61, 70, 78, 85, 95–99], with publication dates between 1994 and 2012 (median year of publication 2001).”
  • The need for complete reporting of all relevant methodological aspects as a prerequisite to assessing the risk of bias of studies was highlighted in 10 articles, and was the focus of recommendations in a publication from the MOOSE group.

    “The need for complete reporting of all relevant methodological aspects as a prerequisite to assessing the risk of bias of studies was highlighted in 10 articles, [16, 33, 35, 47, 50, 56, 86, 87, 90, 91] and was the focus of recommendations in a publication from the Meta-analysis Of Observational Studies in Epidemiology (MOOSE) group [12].”
  • Thirty-nine articles (42%) provided recommendations on heterogeneity and its assessment, with publication dates between 1994 and 2016 (median year 2006).

    “Thirty-nine articles (42%) provided recommendations on heterogeneity and its assessment, [8, 12, 16, 29, 30, 33, 34, 36, 38, 39, 41–44, 47, 49–52, 54–56, 61, 66, 70, 72, 77, 82, 85, 87, 98, 100–107] with publication dates between 1994 and 2016 (median year of publication 2006).”
  • Fifty-two articles (56%) provided recommendations on the statistical analysis in the context of meta-analysis of observational studies, with publication dates between 1994 and 2016 (median year 2006).

    “Fifty-two articles (56%), provided recommendations on the statistical analysis in the context of meta-analysis of observational studies [8, 12, 16, 29–31, 33–36, 38, 41–44, 47, 48, 52, 54–56, 61, 66, 70, 72, 73, 77, 79–83, 85, 87, 96, 100, 101, 103–105, 107–118], with publication dates between 1994 and 2016 (median year of publication 2006).”

Cites

Flagged first, then by citations

Cited by

Sorted by citations

Authors
Lau, J., Ioannidis, J. P. A., & Schmid, C. H.
Affiliations
Tufts University (US)
Corresponding author
Lau, Joseph
Venue
Annals of Internal Medicine
Published
1997-11-01
Volume
127
Issue
9
Pages
820-826
ISSN
  • 0003-4819 (ISSN-L)
  • 1539-3704
OpenAlex
W1982228885
Cited by
2,517
References
60
Citations by year
2026: 40 · 2025: 48 · 2024: 67 · 2023: 45 · 2022: 96 · 2021: 104 · 2020: 112 · 2019: 131 · 2018: 126 · 2017: 165 · 2016: 161 · 2015: 202 · 2014: 186 · 2013: 181 · 2012: 134
FWCI
36.48
Type
journal article
Language
en
Open access
closed
Locations
  • Annals of Internal Medicine (publishedVersion)
  • PubMed (publishedVersion)
Funders
  • Agency for Healthcare Research and Quality (R01-HS07782, R01 HS08532)
  • National Institute of Allergy and Infectious Diseases (T32AI07389, T32AI007389)
Topics
  • Social Sciences › Decision Sciences › Statistics, Probability and Uncertainty › Meta-analysis and systematic reviews
  • Physical Sciences › Mathematics › Statistics and Probability › Statistical Methods in Epidemiology
  • Social Sciences › Sociology and Political Science › Delphi Technique in Research
Keywords
meta-analysis, systematic reviews, heterogeneity, fixed effects model, random effects models, meta-regression, sensitivity analysis, publication bias, study quality, diagnostic performance, observational studies
MeSH terms
Humans, Research Design, Statistics as Topic, Meta-Analysis as Topic, Bias, Randomized Controlled Trials as Topic

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review