CoChat
Sign Up
Cited by 894Statistical Methods in Clinical TrialsMeta-analysis and systematic reviewsStatistical Methods and Bayesian Inference

A tutorial on sensitivity analyses in clinical trials: the what, why, when and how

Open access
Thabane, Lehana, Mbuagbaw, Lawrence, Zhang, Shiyuan, Samaan, Zainab, Marcucci, Maura, Ye, Chenglin, Thabane, Marroon, Giangregorio, Lora +9 more·2013·BMC Medical Research Methodology
DOI 10.1186/1471-2288-13-92PMC PMC3720188
Open this paper in CoChat — free accountOpen PDF

Abstract

BACKGROUND: Sensitivity analyses play a crucial role in assessing the robustness of the findings or conclusions based on primary analyses of data in clinical trials. They are a critical way to assess the impact, effect or influence of key assumptions or variations--such as different methods of analysis, definitions of outcomes, protocol deviations, missing data, and outliers--on the overall conclusions of a study.The current paper is the second in a series of tutorial-type manuscripts intended to discuss and clarify aspects related to key methodological issues in the design and analysis of clinical trials. DISCUSSION: In this paper we will provide a detailed exploration of the key aspects of sensitivity analyses including: 1) what sensitivity analyses are, why they are needed, and how often they are used in practice; 2) the different types of sensitivity analyses that one can do, with examples from the literature; 3) some frequently asked questions about sensitivity analyses; and 4) some suggestions on how to report the results of sensitivity analyses in clinical trials. SUMMARY: When reporting on a clinical trial, we recommend including planned or posthoc sensitivity analyses, the corresponding rationale and results along with the discussion of the consequences of these analyses on the overall findings of the study.

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 · 10.1186/1471-2288-13-92

Cites

76 of 76 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?

  • A post-hoc sensitivity analysis involving multiple imputation by chained equations was performed for variables that showed a significant change from T1 to T3 (for any group), using the mice package in R.

    “To test the robustness of the data analysis strategy (given missing data associated with participant attrition), for variables that showed a significant change from T1 to T3 (for any group), a post-hoc sensitivity analysis involving multiple imputation by chained equations was performed using the mice package in R [85, 87].”

    From Optimizing parenting and child outcomes following parent–child interaction therapy – toddler: a randomized controlled trial (2024) · cites this paper as [85]

    The Discussion’s ‘Types of sensitivity analyses’ section recommends using a different missing-data approach to test the robustness of a primary analysis and lists multiple imputation as an option. It does not describe chained equations, the mice package, or the citing study’s T1-to-T3 selection criterion; those details are not supported by this source. (LLM, conf=0.84)

    Experimental — accuracy not yet measured

  • A complete-case analysis was conducted as a sensitivity analysis on observed data to corroborate the assumption of data missing at random.

    “Additionally, a complete-case analysis was conducted as a sensitivity analysis on observed data to corroborate the assumption of data missing at random [94].”

    From Lessons Learned from an Attempted Pragmatic Randomized Controlled Trial for Improvement of Chronic Pain-Associated Disability in Green Professions: Long-Term Effectiveness of a Guided Online-Based Acceptance and Commitment Therapy  (PACT-A) (2022) · cites this paper as [94]

    The Discussion’s ‘Types of sensitivity analyses’ section lists complete-case analysis as an option for sensitivity analysis of missing data. It does not say that this analysis corroborates a missing-at-random (MAR) assumption; it says that ignoring missing data implicitly assumes missing completely at random (MCAR), while complete-case estimates can be consistent under MAR if predictors of missing observations are included. No page number is available in our copy. (LLM, conf=0.88)

    Experimental — accuracy not yet measured

  • The generality of a researcher's results under alternative relevant approaches can be checked via robustness analyses.

    “The generality of a researcher's results under alternative relevant approaches can be checked via robustness analyses (e.g., Thabane et al., 2013, also called sensitivity or multiverse analyses) that indicate how research results vary as a function of reasonable data exclusions and alternative sensible data coding, aggregation, and analysis approaches.”

    From Does preregistration improve the credibility of research findings? (2020) · cites this paper as Thabane et al., 2013

  • Missing data will be handled following the authors' recommendations, using the most appropriate method depending on the reasons for the missing data and the sensitivity analysis principles.

    “Missing data will be handled following the authors’ recommendations, using the most appropriate method depending on the reasons for the missing data and the sensitivity analysis principles.57”

    From Effectiveness of a projection-based augmented reality exposure system in treating cockroach phobia: Study protocol of a randomised controlled trial (2023) · cites this paper as [57]

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: 1

  1. Goldsmith CH Gafni A Drummond MF Torrance GW Stoddart GL Sensitivity Analysis and Experimental Design: The Case of Economic Evaluation of Health Care Programmes. Proceedings of the Third Canadian Conference on Health Economics 1986 1987 Winnipeg MB: The University of Manitoba Press

Cited by

Sorted by citations

Related

Authors
Thabane, L., Mbuagbaw, L., Zhang, S., Samaan, Z., Marcucci, M., Ye, C., Thabane, M., Giangregorio, L., Dennis, B., Kosa, D., Debono, V. B., Dillenburg, R., Fruci, V., Bawor, M., Lee, J., Wells, G., & Goldsmith, C. H.
Affiliations
  • Population Health Research Institute (CA)
  • St. Joseph’s Healthcare Hamilton (CA)
  • McMaster University (CA)
  • GlaxoSmithKline (Canada) (CA)
  • University of Waterloo (CA)
  • Korea University (KR)
  • University of Ottawa (CA)
  • Simon Fraser University (CA)
Corresponding author
Thabane, Lehana
ORCID
Venue
BMC Medical Research Methodology
Publisher
Springer Science and Business Media LLC
Published
2013-07-16
Volume
13
Issue
1
Pages
92-92
ISSN
1471-2288 (ISSN-L)
DOI
10.1186/1471-2288-13-92
PMC
PMC3720188
OpenAlex
W2095437242
Cited by
893
References
76
Citations by year
2026: 76 · 2025: 113 · 2024: 107 · 2023: 109 · 2022: 90 · 2021: 84 · 2020: 90 · 2019: 52 · 2018: 57 · 2017: 43 · 2016: 31 · 2015: 28 · 2014: 11 · 2013: 1
FWCI
20.82
Type
journal article
Language
en
Open access
Licence
Locations
  • BMC Medical Research Methodology (publishedVersion, cc-by)
  • Europe PMC (PubMed Central) (publishedVersion, cc-by)
  • PubMed Central (publishedVersion, cc-by)
  • PubMed (publishedVersion)
  • uO Research (University of Ottawa) (submittedVersion, public-domain)
  • Archivio Istituzionale della Ricerca (Universita Degli Studi Di Milano) (submittedVersion, other-oa)
Topics
  • Physical Sciences › Mathematics › Statistics and Probability › Statistical Methods in Clinical Trials
  • Social Sciences › Decision Sciences › Statistics, Probability and Uncertainty › Meta-analysis and systematic reviews
  • Physical Sciences › Mathematics › Statistics and Probability › Statistical Methods and Bayesian Inference
Keywords
sensitivity analysis, clinical trials, missing data, outliers, protocol deviations, outcome definitions
MeSH terms
Clinical Trials as Topic, Humans, Outcome and Process Assessment, Health Care, Research Design, Sensitivity and Specificity

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