Jove
Visualize
Contact Us

Related Experiment Videos

Matching

M C Costanza1

  • 1Medical Biostatistics/Biometry Facility, University of Vermont, Burlington 05405, USA.

Preventive Medicine
|September 1, 1995
PubMed
Summary

Matching in observational studies can yield unexpected results, particularly in case-control designs. Proper statistical analysis accounting for matching is crucial for valid effect estimates, not just matching itself.

Related Experiment Videos

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

Undocumented migrants in Switzerland: geographical origin versus legal status as risk factor for tuberculosis.

Journal of immigrant and minority health·2009
Same author

Twelve-year trends and correlates of dietary salt intakes for the general adult population of Geneva, Switzerland.

European journal of clinical nutrition·2007
Same author

Cancer risks in LKB1 germline mutation carriers.

Gut·2006
Same author

[Determinants of hypercholesterolemia in Geneva: should the focus be on genes or environment?].

Revue medicale suisse·2005
Same author

Impact of 'Mad Cow Disease' publicity on trends in meat and total vitamin A consumption in Geneva between 1993 and 2000.

European journal of clinical nutrition·2003
Same author

Transcranial Doppler versus angiography in patients with vasospasm due to a ruptured cerebral aneurysm: A systematic review.

Stroke·2001
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

Area of Science:

  • Epidemiology
  • Biostatistics

Background:

  • Matching is a common design strategy in observational studies to control confounding.
  • Its effectiveness and implications for validity are often assumed rather than critically examined.

Purpose of the Study:

  • To critically evaluate the role and impact of matching on confounding control.
  • To assess the validity of effect estimates (odds ratios) in cohort and case-control studies.

Main Methods:

  • Illustrative examples are used to demonstrate the effects of matching.
  • Analysis focuses on the validity of point estimates of the odds ratio.

Main Results:

  • Matching generally performs as expected in cohort studies.
  • Matching can produce unexpected and potentially invalid results in case-control studies.

Conclusions:

  • Matching alone does not automatically control confounding in case-control studies.
  • Valid effect estimates in case-control studies require both matching and appropriate statistical analysis that accounts for the matching.