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

A program to implement a search method for identification of clinical subgroups

R J Marshall1

  • 1Department of Community Health, University of Auckland, New Zealand.

Statistics in Medicine
|December 30, 1995
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

Systematic review of the prevalence of faecal incontinence.

The British journal of surgery·2016
Same author

Putting it bluntly: communication skills in the Iliad.

Medical humanities·2013
Same author

Mortality risk stratification in severely anaemic Jehovah's Witness patients.

Internal medicine journal·2012
Same author

Clinical benefits and cost-effectiveness of allogeneic red-blood-cell transfusion in severe symptomatic anaemia.

Vox sanguinis·2011
Same author

Determining levels of fecal incontinence in the community: a New Zealand cross-sectional study.

Diseases of the colon and rectum·2011
Same author

Complicated hyperthyroidism.

Bulletin of the University of Minnesota Hospitals and Minnesota Medical Foundation. University of Minnesota. Hospitals·2010

Search Partition Analysis (SPAN) identifies homogeneous subgroups using predictor variables. This method aids in understanding outcomes like low birth weight and impaired glucose tolerance for better screening.

Area of Science:

  • Biostatistics
  • Data Mining
  • Predictive Analytics

Background:

  • Identifying homogeneous subgroups is crucial for understanding complex outcomes.
  • Existing methods may not efficiently partition data based on multiple predictors.

Purpose of the Study:

  • To introduce Search Partition Analysis (SPAN), a novel method for identifying homogeneous subgroups.
  • To enable the partitioning of observations into two groups with maximal homogeneity of a numeric outcome variable.

Main Methods:

  • SPAN formulates subgroup identification using a numeric outcome variable (y) and predictor variables (x).
  • It employs Boolean expressions of predictors to create binary partitions, optimizing for outcome homogeneity.
  • The method incorporates complexity penalizing and defined search strategies for optimal partition selection.

Related Experiment Videos

Main Results:

  • SPAN effectively identifies subgroups with uniform 'low' or 'high' outcome variable values.
  • The approach was illustrated using analyses of predictors for low birth weight and impaired glucose tolerance.
  • The method allows for the definition and representation of identified subgroups.

Conclusions:

  • SPAN provides a robust procedure for subgroup identification based on predictor variables.
  • This method enhances the understanding of factors influencing health outcomes.
  • SPAN is applicable for screening purposes, aiding in targeted interventions.