Statistical considerations on prognostic models for glioma

Annette M. Molinaro, Margaret R. Wrensch, Robert Brian Jenkins, Jeanette E Eckel-Passow

Research output: Contribution to journalArticle

6 Citations (Scopus)

Abstract

Given the lack of beneficial treatments in glioma, there is a need for prognostic models for therapeutic decision making and life planning. Recently several studies defining subtypes of glioma have been published. Here, we review the statistical considerations of how to build and validate prognostic models, explain the models presented in the current glioma literature, and discuss advantages and disadvantages of each model. The 3 statistical considerations to establishing clinically useful prognostic models are: study design, model building, and validation. Careful study design helps to ensure that the model is unbiased and generalizable to the population of interest. During model building, a discovery cohort of patients can be used to choose variables, construct models, and estimate prediction performance via internal validation. Via external validation, an independent dataset can assess how well the model performs. It is imperative that published models properly detail the study design and methods for both model building and validation. This provides readers the information necessary to assess the bias in a study, compare other published models, and determine the model's clinical usefulness. As editors, reviewers, and readers of the relevant literature, we should be cognizant of the needed statistical considerations and insist on their use.

Original languageEnglish (US)
Pages (from-to)609-623
Number of pages15
JournalNeuro-Oncology
Volume18
Issue number5
DOIs
StatePublished - May 1 2016

Fingerprint

Glioma
Decision Making
Therapeutics
Population
Datasets

Keywords

  • glioma
  • model building
  • prognostic models
  • statistics
  • validation

ASJC Scopus subject areas

  • Cancer Research
  • Oncology
  • Clinical Neurology

Cite this

Statistical considerations on prognostic models for glioma. / Molinaro, Annette M.; Wrensch, Margaret R.; Jenkins, Robert Brian; Eckel-Passow, Jeanette E.

In: Neuro-Oncology, Vol. 18, No. 5, 01.05.2016, p. 609-623.

Research output: Contribution to journalArticle

Molinaro, Annette M. ; Wrensch, Margaret R. ; Jenkins, Robert Brian ; Eckel-Passow, Jeanette E. / Statistical considerations on prognostic models for glioma. In: Neuro-Oncology. 2016 ; Vol. 18, No. 5. pp. 609-623.
@article{43c6a50ad4194325ab492e1b51618a09,
title = "Statistical considerations on prognostic models for glioma",
abstract = "Given the lack of beneficial treatments in glioma, there is a need for prognostic models for therapeutic decision making and life planning. Recently several studies defining subtypes of glioma have been published. Here, we review the statistical considerations of how to build and validate prognostic models, explain the models presented in the current glioma literature, and discuss advantages and disadvantages of each model. The 3 statistical considerations to establishing clinically useful prognostic models are: study design, model building, and validation. Careful study design helps to ensure that the model is unbiased and generalizable to the population of interest. During model building, a discovery cohort of patients can be used to choose variables, construct models, and estimate prediction performance via internal validation. Via external validation, an independent dataset can assess how well the model performs. It is imperative that published models properly detail the study design and methods for both model building and validation. This provides readers the information necessary to assess the bias in a study, compare other published models, and determine the model's clinical usefulness. As editors, reviewers, and readers of the relevant literature, we should be cognizant of the needed statistical considerations and insist on their use.",
keywords = "glioma, model building, prognostic models, statistics, validation",
author = "Molinaro, {Annette M.} and Wrensch, {Margaret R.} and Jenkins, {Robert Brian} and Eckel-Passow, {Jeanette E}",
year = "2016",
month = "5",
day = "1",
doi = "10.1093/neuonc/nov255",
language = "English (US)",
volume = "18",
pages = "609--623",
journal = "Neuro-Oncology",
issn = "1522-8517",
publisher = "Oxford University Press",
number = "5",

}

TY - JOUR

T1 - Statistical considerations on prognostic models for glioma

AU - Molinaro, Annette M.

AU - Wrensch, Margaret R.

AU - Jenkins, Robert Brian

AU - Eckel-Passow, Jeanette E

PY - 2016/5/1

Y1 - 2016/5/1

N2 - Given the lack of beneficial treatments in glioma, there is a need for prognostic models for therapeutic decision making and life planning. Recently several studies defining subtypes of glioma have been published. Here, we review the statistical considerations of how to build and validate prognostic models, explain the models presented in the current glioma literature, and discuss advantages and disadvantages of each model. The 3 statistical considerations to establishing clinically useful prognostic models are: study design, model building, and validation. Careful study design helps to ensure that the model is unbiased and generalizable to the population of interest. During model building, a discovery cohort of patients can be used to choose variables, construct models, and estimate prediction performance via internal validation. Via external validation, an independent dataset can assess how well the model performs. It is imperative that published models properly detail the study design and methods for both model building and validation. This provides readers the information necessary to assess the bias in a study, compare other published models, and determine the model's clinical usefulness. As editors, reviewers, and readers of the relevant literature, we should be cognizant of the needed statistical considerations and insist on their use.

AB - Given the lack of beneficial treatments in glioma, there is a need for prognostic models for therapeutic decision making and life planning. Recently several studies defining subtypes of glioma have been published. Here, we review the statistical considerations of how to build and validate prognostic models, explain the models presented in the current glioma literature, and discuss advantages and disadvantages of each model. The 3 statistical considerations to establishing clinically useful prognostic models are: study design, model building, and validation. Careful study design helps to ensure that the model is unbiased and generalizable to the population of interest. During model building, a discovery cohort of patients can be used to choose variables, construct models, and estimate prediction performance via internal validation. Via external validation, an independent dataset can assess how well the model performs. It is imperative that published models properly detail the study design and methods for both model building and validation. This provides readers the information necessary to assess the bias in a study, compare other published models, and determine the model's clinical usefulness. As editors, reviewers, and readers of the relevant literature, we should be cognizant of the needed statistical considerations and insist on their use.

KW - glioma

KW - model building

KW - prognostic models

KW - statistics

KW - validation

UR - http://www.scopus.com/inward/record.url?scp=84965157590&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84965157590&partnerID=8YFLogxK

U2 - 10.1093/neuonc/nov255

DO - 10.1093/neuonc/nov255

M3 - Article

C2 - 26657835

AN - SCOPUS:84965157590

VL - 18

SP - 609

EP - 623

JO - Neuro-Oncology

JF - Neuro-Oncology

SN - 1522-8517

IS - 5

ER -