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

Reasons for caution when evaluating health care interventions using non-randomised study designs.

B C Reeves1

  • 1Department of Public Health and Policy, London School of Hygiene and Tropical Medicine, Keppel Street, London WC1E 7HT, Great Britain. barney.reeves@lshtm.ac.uk

Forschende Komplementarmedizin Und Klassische Naturheilkunde = Research in Complementary and Natural Classical Medicine
|September 9, 2004
PubMed
Summary

Non-randomised studies (NRS) are prone to selection bias and confounding by clinical indication (CCI). Residual confounding in NRS introduces uncertainty bias, impacting effect size estimates.

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

Peri-operative corticosteroid supplementation for patients on therapeutic glucocorticoids: a national survey.

Anaesthesia·2020
Same author

Randomized trial of near-infrared spectroscopy for personalized optimization of cerebral tissue oxygenation during cardiac surgery.

British journal of anaesthesia·2017
Same author

Adaptation of the By-Band randomized clinical trial to By-Band-Sleeve to include a new intervention and maintain relevance of the study to practice.

The British journal of surgery·2017
Same author

Cost-effectiveness of community versus hospital eye service follow-up for patients with quiescent treated age-related macular degeneration alongside the ECHoES randomised trial.

BMJ open·2016
Same author

Are lower levels of red blood cell transfusion more cost-effective than liberal levels after cardiac surgery? Findings from the TITRe2 randomised controlled trial.

BMJ open·2016
Same author

The design and implementation of a study to investigate the effectiveness of community vs hospital eye service follow-up for patients with neovascular age-related macular degeneration with quiescent disease.

Eye (London, England)·2015

Area of Science:

  • Epidemiology
  • Biostatistics
  • Clinical Research Methodology

Background:

  • Randomised controlled trials (RCTs) minimize selection bias, unlike non-randomised studies (NRS).
  • NRS are susceptible to confounding by clinical indication (CCI) due to patient selection factors.
  • Selection bias and CCI can compromise the validity of effect size estimates from NRS.

Purpose of the Study:

  • To highlight the inherent susceptibility of NRS to selection bias and CCI.
  • To discuss methods for mitigating bias in NRS design and analysis.
  • To emphasize the consequences of residual confounding and uncertainty bias in NRS.

Main Methods:

  • Comparison of susceptibility to selection bias between RCTs and NRS.
  • Identification of confounding by clinical indication (CCI) as a key issue in NRS.

Related Experiment Videos

  • Discussion of bias mitigation strategies in NRS (e.g., matching, regression modeling).

    • Main Results:

    • NRS are inherently more prone to selection bias and CCI than RCTs.
    • Despite mitigation efforts, residual confounding is inevitable in NRS.
    • Unpredictable CCI can lead to "uncertainty bias" beyond statistical error.

    Conclusions:

    • Awareness of residual confounding and uncertainty bias is crucial for researchers and users of evidence from NRS.
    • Methodological limitations in NRS contribute to unavoidable bias in effect size estimation.
    • The interpretation of findings from NRS requires careful consideration of potential biases.