Systematic Selection of Key Logistic Regression Variables for Risk Prediction Analyses: A Five-Factor Maximum Model

Timothy Hewett, Kate E. Webster, Wendy J. Hurd

Research output: Contribution to journalArticle

8 Citations (Scopus)

Abstract

GENERAL AND CRITICAL REVIEW FORMAT: The evolution of clinical practice and medical technology has yielded an increasing number of clinical measures and tests to assess a patient's progression and return to sport readiness after injury. The plethora of available tests may be burdensome to clinicians in the absence of evidence that demonstrates the utility of a given measurement. OBJECTIVE: Thus, there is a critical need to identify a discrete number of metrics to capture during clinical assessment to effectively and concisely guide patient care. DATA SOURCES: The data sources included Pubmed and PMC Pubmed Central articles on the topic. Therefore, we present a systematic approach to injury risk analyses and how this concept may be used in algorithms for risk analyses for primary anterior cruciate ligament (ACL) injury in healthy athletes and patients after ACL reconstruction. MAIN RESULTS: In this article, we present the five-factor maximum model, which states that in any predictive model, a maximum of 5 variables will contribute in a meaningful manner to any risk factor analysis. CONCLUSIONS: We demonstrate how this model already exists for prevention of primary ACL injury, how this model may guide development of the second ACL injury risk analysis, and how the five-factor maximum model may be applied across the injury spectrum for development of the injury risk analysis.

Original languageEnglish (US)
Pages (from-to)78-85
Number of pages8
JournalClinical journal of sport medicine : official journal of the Canadian Academy of Sport Medicine
Volume29
Issue number1
DOIs
StatePublished - Jan 1 2019

Fingerprint

Logistic Models
Wounds and Injuries
PubMed
Statistical Factor Analysis
Anterior Cruciate Ligament Reconstruction
Information Storage and Retrieval
Primary Prevention
Athletes
Patient Care
Technology
Anterior Cruciate Ligament Injuries

ASJC Scopus subject areas

  • Orthopedics and Sports Medicine
  • Physical Therapy, Sports Therapy and Rehabilitation

Cite this

@article{bef78117cf8b4b928eb843c71d4056d4,
title = "Systematic Selection of Key Logistic Regression Variables for Risk Prediction Analyses: A Five-Factor Maximum Model",
abstract = "GENERAL AND CRITICAL REVIEW FORMAT: The evolution of clinical practice and medical technology has yielded an increasing number of clinical measures and tests to assess a patient's progression and return to sport readiness after injury. The plethora of available tests may be burdensome to clinicians in the absence of evidence that demonstrates the utility of a given measurement. OBJECTIVE: Thus, there is a critical need to identify a discrete number of metrics to capture during clinical assessment to effectively and concisely guide patient care. DATA SOURCES: The data sources included Pubmed and PMC Pubmed Central articles on the topic. Therefore, we present a systematic approach to injury risk analyses and how this concept may be used in algorithms for risk analyses for primary anterior cruciate ligament (ACL) injury in healthy athletes and patients after ACL reconstruction. MAIN RESULTS: In this article, we present the five-factor maximum model, which states that in any predictive model, a maximum of 5 variables will contribute in a meaningful manner to any risk factor analysis. CONCLUSIONS: We demonstrate how this model already exists for prevention of primary ACL injury, how this model may guide development of the second ACL injury risk analysis, and how the five-factor maximum model may be applied across the injury spectrum for development of the injury risk analysis.",
author = "Timothy Hewett and Webster, {Kate E.} and Hurd, {Wendy J.}",
year = "2019",
month = "1",
day = "1",
doi = "10.1097/JSM.0000000000000486",
language = "English (US)",
volume = "29",
pages = "78--85",
journal = "Clinical Journal of Sport Medicine",
issn = "1050-642X",
publisher = "Lippincott Williams and Wilkins",
number = "1",

}

TY - JOUR

T1 - Systematic Selection of Key Logistic Regression Variables for Risk Prediction Analyses

T2 - A Five-Factor Maximum Model

AU - Hewett, Timothy

AU - Webster, Kate E.

AU - Hurd, Wendy J.

PY - 2019/1/1

Y1 - 2019/1/1

N2 - GENERAL AND CRITICAL REVIEW FORMAT: The evolution of clinical practice and medical technology has yielded an increasing number of clinical measures and tests to assess a patient's progression and return to sport readiness after injury. The plethora of available tests may be burdensome to clinicians in the absence of evidence that demonstrates the utility of a given measurement. OBJECTIVE: Thus, there is a critical need to identify a discrete number of metrics to capture during clinical assessment to effectively and concisely guide patient care. DATA SOURCES: The data sources included Pubmed and PMC Pubmed Central articles on the topic. Therefore, we present a systematic approach to injury risk analyses and how this concept may be used in algorithms for risk analyses for primary anterior cruciate ligament (ACL) injury in healthy athletes and patients after ACL reconstruction. MAIN RESULTS: In this article, we present the five-factor maximum model, which states that in any predictive model, a maximum of 5 variables will contribute in a meaningful manner to any risk factor analysis. CONCLUSIONS: We demonstrate how this model already exists for prevention of primary ACL injury, how this model may guide development of the second ACL injury risk analysis, and how the five-factor maximum model may be applied across the injury spectrum for development of the injury risk analysis.

AB - GENERAL AND CRITICAL REVIEW FORMAT: The evolution of clinical practice and medical technology has yielded an increasing number of clinical measures and tests to assess a patient's progression and return to sport readiness after injury. The plethora of available tests may be burdensome to clinicians in the absence of evidence that demonstrates the utility of a given measurement. OBJECTIVE: Thus, there is a critical need to identify a discrete number of metrics to capture during clinical assessment to effectively and concisely guide patient care. DATA SOURCES: The data sources included Pubmed and PMC Pubmed Central articles on the topic. Therefore, we present a systematic approach to injury risk analyses and how this concept may be used in algorithms for risk analyses for primary anterior cruciate ligament (ACL) injury in healthy athletes and patients after ACL reconstruction. MAIN RESULTS: In this article, we present the five-factor maximum model, which states that in any predictive model, a maximum of 5 variables will contribute in a meaningful manner to any risk factor analysis. CONCLUSIONS: We demonstrate how this model already exists for prevention of primary ACL injury, how this model may guide development of the second ACL injury risk analysis, and how the five-factor maximum model may be applied across the injury spectrum for development of the injury risk analysis.

UR - http://www.scopus.com/inward/record.url?scp=85057726332&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85057726332&partnerID=8YFLogxK

U2 - 10.1097/JSM.0000000000000486

DO - 10.1097/JSM.0000000000000486

M3 - Article

C2 - 28817414

AN - SCOPUS:85057726332

VL - 29

SP - 78

EP - 85

JO - Clinical Journal of Sport Medicine

JF - Clinical Journal of Sport Medicine

SN - 1050-642X

IS - 1

ER -