CoChat
Sign Up
Cited by 85Meta-analysis and systematic reviewsAdvanced Causal Inference TechniquesStatistical Methods in Epidemiology

Methods to Address Confounding and Other Biases in Meta-Analyses: Review and Recommendations

Open access
Mathur, Maya B., VanderWeele, Tyler J.·2022·Annual Review of Public Health
DOI 10.1146/annurev-publhealth-051920-114020PMC PMC8959012
Open this paper in CoChat — free account

Abstract

Meta-analyses contribute critically to cumulative science, but they can produce misleading conclusions if their constituent primary studies are biased, for example by unmeasured confounding in nonrandomized studies. We provide practical guidance on how meta-analysts can address confounding and other biases that affect studies’ internal validity, focusing primarily on sensitivity analyses that help quantify how biased the meta-analysis estimates might be. We review a number of sensitivity analysis methods to do so, especially recent developments that are straightforward to implement and interpret and that use somewhat less stringent statistical assumptions than do earlier methods. We give recommendations for how these newer methods could be applied in practice and illustrate using a previously published meta-analysis. Sensitivity analyses can provide informative quantitative summaries of evidence strength, and we suggest reporting them routinely in meta-analyses of potentially biased studies. This recommendation in no way diminishes the importance of defining study eligibility criteria that reduce bias and of characterizing studies’ risks of bias qualitatively.

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.1146/annurev-publhealth-051920-114020

Cites

59 of 59 references verified0 citation errors · 0 not checkable

Statistics

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

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?

  • The causal interpretation of the estimated treatment effects relies on standard identification assumptions used in the program-evaluation literature.

    “The causal interpretation of the estimated treatment effects relies on the standard identification assumptions used in the program-evaluation literature [16, 20, 21].”

    From Health and Socio-Economic Impacts of Climate-Related Displacement in Bangladesh’s Chars: Causal Evidence From a Household Survey (2026) · cites this paper as [21]

    The section on eligibility criteria says causal conclusions from nonrandomized studies depend on temporal ordering and adequate control of confounding. That supports the general point that causal interpretation requires assumptions, but the paper does not identify the cited paper’s 'standard identification assumptions' for estimated treatment effects or frame them as program-evaluation assumptions. No page number is available in our copy. (LLM, conf=0.78)

    Experimental — accuracy not yet measured

  • Both models account for potential biases and confounding factors.

    “Both models account for potential biases and confounding factors [20, 21].”

    From Health and Socio-Economic Impacts of Climate-Related Displacement in Bangladesh’s Chars: Causal Evidence From a Household Survey (2026) · cites this paper as [21]

    The section ‘Biases other than confounding’ says the reviewed methods assess sensitivity to unmeasured confounding and that some also accommodate other biases. It does not establish that both models referred to in the citing sentence account for both, and it qualifies the broader point as applying to only some methods. (LLM, conf=0.78)

    Experimental — accuracy not yet measured

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

The self-check needs the paper’s full text.

Cites

Flagged first, then by citations

Research works printed but not linked: 4

  1. [4]Sv Buuren, Groothuis-Oudshoorn K. 2010. mice: Multivariate imputation by chained equations in R. Journal of Statistical Software: 1– 68
  2. [13]Greenland S, O’Rourke K. 2008. Meta-analysis. In Modern epidemiology. Lippincott Williams & Wilkins, 3rd ed.
  3. [23]Lash TL. 2021. Bias analysis. In Modern epidemiology. Wolters Kluwer, 4th ed.
  4. [53]VanderWeele TJ. 2021. Causal inference with time-varying exposures. In Modern epidemiology. Wolters Kluwer, 4th ed.

Cited by

Sorted by citations

Related

Authors
Mathur, M. B., & VanderWeele, T. J.
Affiliations
  • Stanford University (US)
  • Harvard University (US)
ORCID
Venue
Annual Review of Public Health
Publisher
Annual Reviews
Published
2022-04-05
Volume
43
Issue
1
Pages
19-35
ISSN
  • 0163-7525 (ISSN-L)
  • 1545-2093
DOI
10.1146/annurev-publhealth-051920-114020
PMC
PMC8959012
OpenAlex
W3199520066
Cited by
85
References
59
Citations by year
2026: 16 · 2025: 17 · 2024: 21 · 2023: 18 · 2022: 12 · 2021: 1
FWCI
0.54
Type
journal article
Language
en
Open access
Licence
Locations
  • Annual Review of Public Health (publishedVersion, cc-by)
  • PubMed (publishedVersion)
  • PubMed Central (acceptedVersion)
Funders
  • National Institutes of Health (UL1TR003142)
  • National Cancer Institute (R01CA222147, P30CA124435)
  • National Institute of Diabetes and Digestive and Kidney Diseases (P30DK116074)
  • U.S. National Library of Medicine (R01LM013866)
  • National Center for Advancing Translational Sciences
Topics
  • Social Sciences › Decision Sciences › Statistics, Probability and Uncertainty › Meta-analysis and systematic reviews
  • Physical Sciences › Mathematics › Statistics and Probability › Advanced Causal Inference Techniques
  • Physical Sciences › Mathematics › Statistics and Probability › Statistical Methods in Epidemiology
Keywords
meta-analysis bias, confounding, sensitivity analysis, internal validity, non-randomized studies, risk of bias, bias correction, reporting recommendations, quality of evidence
MeSH terms
Humans, Bias

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