Prognosis and prognostic research: validating a prognostic model

Prognostic models are of little clinical value unless they are shown to work in other samples. Douglas Altman and colleagues describe how to validate models and discuss some of the problems

Saved in:
Bibliographic Details
Published in:BMJ 2009-05, Vol.338 (7708), p.1432-1435
Main Authors: Altman, Douglas G, Vergouwe, Yvonne, Royston, Patrick, Moons, Karel G M
Format: Article
Language:English
Subjects:
Acute Disease
Calibration
Case studies
Child
Classification
Colleagues
Colorectal surgery
Colorectal Surgery - mortality
Cough - complications
Data sampling
Datasets
Future events
Health care access
Heart surgery
Humans
Life tables
Logistic regression
Medical practice
Modeling
Models, Biological
Mortality
Patients
Performance evaluation
Positive-Pressure Respiration - mortality
Predictive Value of Tests
Primary health care
Prognosis
Reproducibility of Results
RESEARCH METHODS & REPORTING
Risk Factors
Stratification
Thoracic Surgical Procedures - mortality
Citations: Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Prognostic models are of little clinical value unless they are shown to work in other samples. Douglas Altman and colleagues describe how to validate models and discuss some of the problems
ISSN:0959-8138
0959-535X
1468-5833
1756-1833
DOI:10.1136/bmj.b605