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

Bounding causal effects under uncontrolled confounding using counterfactuals.

Richard F MacLehose1, Sol Kaufman, Jay S Kaufman

  • 1Department of Epidemiology, University of North Carolina School of Public Health, Chapel Hill, NC 27599-7435, USA. maclehose@unc.edu

Epidemiology (Cambridge, Mass.)
|June 14, 2005
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

Influence of Gestational Age Measurement Technique on the Associations Between Maternal Characteristics and Birth Outcomes.

Paediatric and perinatal epidemiology·2026
Same author

Pregnancy Length Measurement Error: A Comparison of Last Menstrual Period and Ultrasonography with Ovulation-based Estimation.

Epidemiology (Cambridge, Mass.)·2025
Same author

The exposure potential restriction rule revisited.

American journal of epidemiology·2025
Same author

[Strengthening the Reporting of Observational Studies in Epidemiology (STROBE): explanation and elaboration - a Korean translation].

Ewha medical journal·2025
Same author

Evaluating behavioral goals for eating timing, frequency, and distribution of daily carbohydrate consumption among youth with type 1 diabetes (MyPlan): A single arm pilot and feasibility study.

Diabetes research and clinical practice·2025
Same author

Hazard Ratios and Alternative Effect Measures: An Applied Illustration.

Pharmacoepidemiology and drug safety·2024

This study introduces new methods for causal effect analysis, providing bounds for unmeasured confounding. These bounds offer a range for the true causal effect, improving sensitivity analysis in observational studies.

Area of Science:

  • Causal inference and observational studies
  • Statistical methodology and sensitivity analysis

Background:

  • Standard sensitivity analysis methods for unmeasured confounders yield corrected point estimates.
  • These methods rely on specifying values for unknown parameters of unmeasured confounders.

Purpose of the Study:

  • To review alternative methods for generating deterministic nonparametric bounds on causal effect magnitude.
  • To utilize linear programming and potential outcomes models for bounding causal effects.
  • To demonstrate reducing bound widths through assumptions on potential outcomes.

Main Methods:

  • Employs linear programming methods and potential outcomes models.
  • Generates bounds on causal effect magnitude using only observed data.
  • Applies the approach to data from the Cooperative Cardiovascular Project.

Related Experiment Videos

Main Results:

  • Deterministic nonparametric bounds for causal effect magnitude are generated from observed data.
  • The width of these bounds can be reduced by making specific assumptions about potential outcomes.
  • The linear programming approach provides a range within which the causal effect must lie.

Conclusions:

  • The proposed bounds on causal effect under uncontrolled confounding complement existing sensitivity analyses.
  • This method offers a robust range for causal effects when unmeasured confounding is present.
  • It enhances the interpretation of causal effects in observational research.