Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Videos

Evaluating non-randomised intervention studies.

J J Deeks1, J Dinnes, R D'Amico

  • 1Centre for Statistics in Medicine, Institute of Health Sciences, Oxford, UK.

Health Technology Assessment (Winchester, England)
|September 23, 2003
PubMed
Summary
This summary is machine-generated.

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Are medical procedures that induce coughing or involve respiratory suctioning associated with increased generation of aerosols and risk of SARS-CoV-2 infection? A rapid systematic review.

The Journal of hospital infection·2021
Same author

CONSORT 2010 statement: extension checklist for reporting within person randomised trials.

The British journal of dermatology·2019
Same author

'Allocation concealment': the evolution and adoption of a methodological term.

Journal of the Royal Society of Medicine·2018
Same author

[Better Reporting of Interventions: Template for Intervention Description and Replication (TIDieR) Checklist and Guide].

Gesundheitswesen (Bundesverband der Arzte des Offentlichen Gesundheitsdienstes (Germany))·2018
Same author

GRIPP2 reporting checklists: tools to improve reporting of patient and public involvement in research.

Research involvement and engagement·2017
Same author

GRIPP2 reporting checklists: tools to improve reporting of patient and public involvement in research.

BMJ (Clinical research ed.)·2017
Same journal

Alpha2 agonists for sedation to produce better outcomes for adults with critical illness: a synopsis of the A2B RCT with cost-effectiveness and process evaluation.

Health technology assessment (Winchester, England)·2026
Same journal

Ethical implications of the use of AI-based technologies for medical image classification systems in screening: a qualitative systematic review.

Health technology assessment (Winchester, England)·2026
Same journal

Plain balloon angioplasty +/- bare metal stenting versus drug-coated balloon angioplasty +/- bare metal stenting versus primary drug-eluting stenting in patients with chronic limb-threatening ischaemia: BASIL-3, an open-label, three-arm, randomised, multi-centre, phase 3 trial.

Health technology assessment (Winchester, England)·2026
Same journal

Effectiveness of psychosocial interventions for adults with substance use disorder that have a co-occurring mental health disorder: an umbrella review and illustrative cost-effectiveness analysis.

Health technology assessment (Winchester, England)·2026
Same journal

Evaluating the effectiveness of simvastatin in slowing the progression of disability in secondary progressive multiple sclerosis: a synopsis of MS-STAT2, a multicentre, randomised controlled, double-blind, phase 3 clinical trial.

Health technology assessment (Winchester, England)·2026
Same journal

Placental growth factor (PLGF)-based testing to help diagnose suspected pre-eclampsia: a systematic review and economic evaluation.

Health technology assessment (Winchester, England)·2026
See all related articles

Evaluating bias in non-randomised studies is crucial. Standard adjustment methods often fail, leading to potentially misleading results and impacting healthcare policies. Further research is recommended.

Area of Science:

  • Medical Research Methodology
  • Biostatistics
  • Evidence-Based Medicine

Background:

  • Non-randomised studies are frequently used in medical research but are susceptible to bias.
  • Assessing the quality and bias of non-randomised studies is challenging.
  • Existing methods for bias adjustment in non-randomised studies may be inadequate.

Purpose of the Study:

  • To evaluate methods and evidence for assessing bias in non-randomised intervention studies.
  • To investigate the effectiveness of quality assessment tools for non-randomised studies.
  • To explore the impact of non-random allocation on study results and the efficacy of adjustment strategies.

Main Methods:

  • Conducted three systematic reviews on bias in non-randomised studies, quality assessment tools, and their application.

Related Experiment Videos

  • Performed new empirical investigations by generating non-randomised studies from two large randomised controlled trials (RCTs).
  • Resampled participants from RCTs based on allocated treatment, centre, and period to simulate non-randomised study conditions.

    • Main Results:

    • Meta-epidemiological studies revealed bias in non-randomised studies, potentially leading to over/underestimations of treatment effects.
    • Numerous quality assessment tools exist, but many omit key validity domains; few are suitable for systematic reviews.
    • Standard case-mix adjustment methods, including logistic regression and propensity scores, were largely ineffective in removing bias.
    • Bias could be substantial, leading to false conclusions of benefit or harm, and increased result variation.

    Conclusions:

    • Non-randomised studies can yield misleading results, even when groups appear similar on prognostic factors.
    • Current case-mix adjustment methods do not reliably eliminate selection bias.
    • Many quality assessment tools are insufficient for appraising non-randomised studies.
    • Non-randomised studies should be used cautiously, primarily when RCTs are infeasible or unethical.