CoChat
Sign Up
Cited by 48Statistical Methods in Clinical TrialsMeta-analysis and systematic reviewsAdvanced Causal Inference Techniques

A review of cluster randomized trials found statistical evidence of selection bias

Bolzern, Jaime, Mnyama, Nandi, Bosanquet, Katharine, Torgerson, David J.·2018·Journal of Clinical Epidemiology
DOI 10.1016/j.jclinepi.2018.03.010
Open this paper in CoChat — free account

Abstract

Objectives To assess markers of selection bias risk in a sample of recently published cluster randomized controlled trials compared with individually randomized trials. Study design and setting We used OVID Medline and the online archives of the Journal of the American Medical Association to search for cluster randomized trials published between January 2015 and June 2017 from four high-impact journals and compared them to a matched sample of individually randomized trials. Results We identified 23 cluster trials: 57% (n = 13) described a robust allocation method and 17% (n = 4) recruited all participants before randomization. Four (17%), eight (35%), and 11 (48%) were classified as at low, medium, and high bias risk, respectively. Meta-analysis showed significant age imbalance (-0.05, 95% CI = -0.057 to -0.043, I 2 = 93.2%) in cluster trials, while the matched individually randomized trials showed no imbalance (0.005, 95% CI = -0.026 to 0.035, I 2 = 0%). Cluster trials finding a statistically significant outcome in their primary measure showed a larger age imbalance (0.082, 95% CI = -0.091 to -0.073, I 2 = 87%) than trials finding a nonstatistically significant outcome (0.022, 95% CI = 0.008 to 0.035, I 2 = 83%). Conclusions There is strong evidence in this sample of an effect of selection bias seen in an imbalance in baseline participant age, something not seen in a comparable sample of individually randomized trials.

Integrity signals

Registry

Crossref · 10.1016/j.jclinepi.2018.03.010

Cites

12 of 12 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?

  • Cluster-randomized designs are known to be at increased risk of bias compared to the individually randomized design.

    “Cluster-randomized designs are known to be at increased risk of bias compared to the individually randomized design [1], [2], [7], [10], [13], [16], [28], [37].”

    From Cluster randomized trials of individual-level interventions were at high risk of bias (2021) · cites this paper as [1]

  • Recruitment and identification biases are a unique source of bias under cluster randomization, with trials being particularly vulnerable to this bias when it is necessary to identify or recruit individuals into the study after randomization.

    “Recruitment and identification biases are a unique source of bias under cluster randomization, with trials being particularly vulnerable to this bias when it is necessary to identify or recruit individuals into the study after randomization [1,11].”

    From Cluster randomized trials of individual-level interventions were at high risk of bias (2021) · cites this paper as [1]

  • Knowledge of treatment condition at the time of patient recruitment is known to be a risk factor for differential identification and recruitment of participants across treatment arms, unless recruitment and identification are conducted by someone blind to the treatment allocation or the inclusion criteria are broad.

    “Knowledge of treatment condition at the time of patient recruitment is known to be a risk factor for differential identification and recruitment of participants across treatment arms [1,14,16,37], unless recruitment and identification are conducted by someone blind to the treatment allocation or the inclusion criteria are broad [2,11,14].”

    From Cluster randomized trials of individual-level interventions were at high risk of bias (2021) · cites this paper as [1]

  • Methodological reviews have identified that many cluster trials are at risk of these identification and recruitment biases because they recruit participants with knowledge of allocated treatment and this often manifests in baseline imbalances.

    “Methodological reviews have identified that many cluster trials are at risk of these identification and recruitment biases because they recruit participants with knowledge of allocated treatment and this often manifests in baseline imbalances [1,2,28].”

    From Cluster randomized trials of individual-level interventions were at high risk of bias (2021) · cites this paper as [1]

  • In a review of recent randomized trials, cluster trials were reported to be more likely to have a significant baseline imbalance on age, whereas individually randomized trials were not.

    “in a review of recent randomized trials, cluster trials were reported to be more likely to have a significant baseline imbalance on age, whereas individually randomized trials were not [1].”

    From Cluster randomized trials of individual-level interventions were at high risk of bias (2021) · cites this paper as [1]

Cites

Flagged first, then by citations

Cited by

Sorted by citations

Related

Authors
Bolzern, J., Mnyama, N., Bosanquet, K., & Torgerson, D. J.
Affiliations
University of York (GB)
Corresponding author
Bolzern, Jaime
ORCID
Venue
Journal of Clinical Epidemiology
Publisher
Elsevier BV
Published
2018-07
Volume
99
Pages
106-112
ISSN
0895-4356 (ISSN-L)
DOI
10.1016/j.jclinepi.2018.03.010
OpenAlex
W2792037397
Cited by
48
References
12
Citations by year
2026: 4 · 2025: 4 · 2024: 1 · 2023: 11 · 2022: 10 · 2021: 8 · 2020: 7 · 2019: 2
FWCI
3.54
Type
journal article
Language
en
Open access
closed
Licence
www.elsevier.com/tdm/userlicense/1.0/ (tdm)
Locations
  • Journal of Clinical Epidemiology (publishedVersion)
  • PubMed (publishedVersion)
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 › Advanced Causal Inference Techniques
Keywords
selection bias, allocation concealment, baseline imbalance, risk of bias assessment
MeSH terms
Age Factors, Humans, Risk, Selection Bias, Randomized Controlled Trials as Topic, Patient Selection

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