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

The intermediate endpoint effect in logistic and probit regression.

D P MacKinnon1, C M Lockwood, C H Brown

  • 1Department of Psychology, Arizona State University, Tempe, Arizona 85287-1104, USA. davidpm@asu.edu

Clinical Trials (London, England)
|October 19, 2007
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

We need to make progress on blunted affect: A commentary.

Schizophrenia research·2024
Same author

Reply to Explaining variations in survival in breast cancer in the Eastern Region of England.

Annals of oncology : official journal of the European Society for Medical Oncology·2020
Same author

The impact of phone calls on follow-up rates in an online depression prevention study.

Internet interventions·2018
Same author

Association of intraoperative changes in brain-derived neurotrophic factor and postoperative delirium in older adults.

British journal of anaesthesia·2017
Same author

Oxytocin Neurones: Intrinsic Mechanisms Governing the Regularity of Spiking Activity.

Journal of neuroendocrinology·2015
Same author

Reproductive Regulation of Gene Expression in the Hypothalamic Supraoptic and Paraventricular Nuclei.

Journal of neuroendocrinology·2015
Same journal

A statistical evaluation of decision-making methods and the efficiency of Bayesian multi-arm multi-stage trials.

Clinical trials (London, England)·2026
Same journal

Accounting for non-adherence: A re-analysis of the Liraglutide Effect and Action in Diabetes: Evaluation of Cardiovascular Outcome Results trial.

Clinical trials (London, England)·2026
Same journal

Phase I design for partially ordered groups with late-onset toxicity.

Clinical trials (London, England)·2026
Same journal

Trial informed consent forms, the Declaration of Helsinki and the SPIRIT 2025 statement.

Clinical trials (London, England)·2026
Same journal

17th Annual University of Pennsylvania Conference on statistical issues in clinical trials - Covariate adjustment in randomized clinical trials: New methods and applications (Morning panel discussion).

Clinical trials (London, England)·2026
Same journal

17th Annual University of Pennsylvania Conference on statistical issues in clinical trials - Covariate adjustment in randomized clinical trials: New methods and applications (Afternoon panel discussion).

Clinical trials (London, England)·2026
See all related articles

A new method using the product of coefficients accurately assesses intermediate endpoint effects in clinical studies. This approach avoids the distorted conclusions often seen with widely used difference in coefficients methods, improving research reliability.

Area of Science:

  • Biostatistics
  • Clinical Trial Design
  • Epidemiology

Background:

  • Intermediate endpoints, also known as surrogate or mediating variables, are crucial in causal pathways between independent and dependent variables.
  • Clinical studies often aim to influence ultimate endpoints by modifying intermediate endpoints, frequently employing logistic or probit regression for binary outcomes.

Purpose of the Study:

  • To identify limitations in common methods for evaluating intermediate endpoint effects.
  • To propose an alternative, more accurate method for assessing intermediate endpoint effects using products of coefficients.

Main Methods:

  • Describing the intermediate endpoint model for binary outcomes, considering both true binary and dichotomized latent continuous outcomes.
  • Utilizing plots of true values and a simulation study to compare different assessment methods.

Related Experiment Videos

Main Results:

  • Widely used methods can yield distorted estimates of intermediate endpoint effects and lead to incorrect conclusions.
  • The proportion of effect explained by an intermediate endpoint, a proposed measure, is also susceptible to these inaccuracies.
  • A solution is provided based on the relationship between latent variable modeling and regression models.

Conclusions:

  • Researchers should adopt intermediate endpoint methods based on the product of regression coefficients for more accurate results.
  • Commonly used difference in coefficient methods can produce misleading conclusions about intermediate effects.