CoChat
Sign Up
Cited by 39Advanced Causal Inference TechniquesStatistical Methods in EpidemiologyMeta-analysis and systematic reviews

The Importance of Making Assumptions in Bias Analysis

Open access
MacLehose, Richard F., Ahern, Thomas P., Lash, Timothy L., Poole, Charles, Greenland, Sander·2021·Epidemiology
DOI 10.1097/ede.0000000000001381PMC PMC8318561
Open this paper in CoChat — free account

Abstract

Quantitative bias analyses allow researchers to adjust for uncontrolled confounding, given specification of certain bias parameters. When researchers are concerned about unknown confounders, plausible values for these bias parameters will be difficult to specify. Ding and VanderWeele developed bounding factor and E-value approaches that require the user to specify only some of the bias parameters. We describe the mathematical meaning of bounding factors and E-values and the plausibility of these methods in an applied context. We encourage researchers to pay particular attention to the assumption made, when using E-values, that the prevalence of the uncontrolled confounder among the exposed is 100% (or, equivalently, the prevalence of the exposure among those without the confounder is 0%). We contrast methods that attempt to bound biases or effects and alternative approaches such as quantitative bias analysis. We provide an example where failure to make this distinction led to erroneous statements. If the primary concern in an analysis is with known but unmeasured potential confounders, then E-values are not needed and may be misleading. In cases where the concern is with unknown confounders, the E-value assumption of an extreme possible prevalence of the confounder limits its practical utility.

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 · 10.1097/ede.0000000000001381

Cites

36 of 36 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?

  • If the actual risk ratio for the effect was 1.4 (the risk ratio of the combined AESI analysis), it could be quite difficult to unambiguously replicate it with observational data given concerns about systematic as well as random errors.

    “If the actual risk ratio for the effect was 1.4 (the risk ratio of the combined AESI analysis), it could be quite difficult to unambiguously replicate it with observational data given concerns about systematic as well as random errors. [33-35].”

    From Serious adverse events of special interest following mRNA COVID-19 vaccination in randomized trials in adults (2022) · cites this paper as [34]

    The Introduction (page not available) identifies uncontrolled confounding as a central concern in observational studies, and Methods > Bounding Factors vs Probabilistic Bias Analysis incorporates random error into an example estimate. These passages support the general concern about systematic and random error, but the paper does not assess whether a true risk ratio of 1.4 could be unambiguously replicated in observational data. (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: 6

  1. [12]Robins J, Rotnitzkey A, Scharfstein D. Sensitivity analysis for selection bias and unmeasured confounding in missing data and causal inference models. In: Halloran E, Berry D, eds. Statistical Models in Epidemiology. Springer; 1999: 1– 92.
  2. [13]Greenland S, Lash TL. Bias Analysis. In: Rothman KJ, Greenland S, Lash TL, eds. Modern Epidemiology. 3rd ed. Lippincott Williams & Wilkins; 2008: 345– 380.
  3. [14]Lash TL, Fox MP, Fink AK. Applying Quantitative Bias Analysis to Epidemiologic Data. Springer; 2009.
  4. [17]Greenland S An argument against E-values for assessing plausibility that associations would be explained away by residual confounding. International Journal of Epidemiology. Published online 2020. 32808028
  5. [34]Trinquart L, Galea S. E-Values and quantitative epistemology. American journal of epidemiology. Published online 2019.
  6. [36]Zhang B, Tchetgen EJT. A Semiparametric Approach to Model-based Sensitivity Analysis in Observational Studies. arXiv preprint arXiv:191014130. Published online 2019. PMC9916707 36777968

Cited by

Sorted by citations

Related

Authors
MacLehose, R. F., Ahern, T. P., Lash, T. L., Poole, C., & Greenland, S.
Affiliations
  • University of Minnesota (US)
  • University of Vermont (US)
  • Emory University (US)
  • University of North Carolina at Chapel Hill (US)
  • University of California, Los Angeles (US)
Corresponding author
MacLehose, Richard F.
ORCID
Venue
Epidemiology
Publisher
Ovid Technologies (Wolters Kluwer Health)
Published
2021-06-24
Volume
32
Issue
5
Pages
617-624
ISSN
1044-3983 (ISSN-L)
DOI
10.1097/ede.0000000000001381
PMC
PMC8318561
OpenAlex
W3181011780
Cited by
39
References
36
Citations by year
2026: 4 · 2025: 4 · 2024: 10 · 2023: 10 · 2022: 7 · 2021: 4
FWCI
5.78
Type
journal article
Language
en
Open access
Licence
Locations
  • Epidemiology (publishedVersion, cc-by-nc-nd)
  • PubMed Central (publishedVersion, cc-by-nc-nd)
  • PubMed (publishedVersion)
  • Carolina Digital Repository (University of North Carolina at Chapel Hill) (submittedVersion)
  • UNC Libraries
Funders
  • National Institute on Aging (R01AG056479)
  • National Institute of General Medical Sciences (P20GM103644)
  • U.S. National Library of Medicine (R01LM013049)
Topics
  • Physical Sciences › Mathematics › Statistics and Probability › Advanced Causal Inference Techniques
  • Physical Sciences › Mathematics › Statistics and Probability › Statistical Methods in Epidemiology
  • Social Sciences › Decision Sciences › Statistics, Probability and Uncertainty › Meta-analysis and systematic reviews
Keywords
E-value, unmeasured confounding, bias assessment, bias bounds
MeSH terms
Humans, Bias, Confounding Factors, Epidemiologic

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