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

Comparing in-patient classification systems: a problem of non-nested regression models.

A R Willan1, W Ross, T A Mackenzie

  • 1Department of Clinical Epidemiology and Biostatistics, McMaster University, Hamilton, Ontario, Canada.

Statistics in Medicine
|July 1, 1992
PubMed
Summary

This study introduces a new statistical test to compare healthcare classification systems like Diagnosis Related Groups (DRG). It helps select the best system for predicting patient resource use, reducing payment errors.

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

The use of methylprednisolone after third molar surgery. A systematic review and meta-analysis of randomized controlled trials.

Medicina oral, patologia oral y cirugia bucal·2024
Same author

Incidence of Parkinson disease in North America.

NPJ Parkinson's disease·2022
Same author

Cesarean delivery or induction of labor in pre-labor twin gestations: a secondary analysis of the twin birth study.

BMC pregnancy and childbirth·2020
Same author

Reported Weight Change in Older Adults and Presence of Frailty.

The Journal of frailty & aging·2020
Same author

Planned Cesarean or planned vaginal delivery for twins: secondary analysis of randomized controlled trial.

Ultrasound in obstetrics & gynecology : the official journal of the International Society of Ultrasound in Obstetrics and Gynecology·2019
Same author

Association of Obesity and Frailty in Older Adults: NHANES 1999-2004.

The journal of nutrition, health & aging·2019

Area of Science:

  • Health economics
  • Health services research
  • Biostatistics

Background:

  • Hospitals in the US use prospective payment systems, such as Diagnosis Related Groups (DRG), for Medicare admissions since 1983.
  • Reimbursement is based on patient classification, adjusted for hospital characteristics.
  • There is ongoing interest in refining existing DRG systems or developing new classification methods.

Purpose of the Study:

  • To develop a method for comparing the effectiveness of different patient classification systems in explaining resource consumption variability.
  • To address the challenge of model selection between non-nested regression models used for comparing classification systems.
  • To provide a statistical test that quantifies the probability of a false positive when selecting a classification system.

Main Methods:

Related Experiment Videos

  • Utilized a simple measure of fit to develop a symmetric statistical test.
  • The test evaluates the null hypothesis that two classification systems explain resource consumption variability equally well.
  • Compared the proposed method to Akaike's Information Criterion (AIC) for model selection.

Main Results:

  • The developed method allows for the quantification of the probability of a false positive.
  • This enables a controlled approach to selecting between classification systems, avoiding selection when systems perform equally.
  • Offers an advantage over methods like AIC by providing a direct measure of error probability.

Conclusions:

  • The proposed statistical test provides a robust and interpretable method for comparing healthcare classification systems.
  • It aids in selecting the most appropriate system for resource consumption prediction, thereby improving healthcare payment accuracy.
  • This approach helps limit erroneous choices between classification systems that do not offer superior explanatory power.