Incorporating published univariable associations in diagnostic and prognostic modeling
Abstract Background Diagnostic and prognostic literature is overwhelmed with studies reporting univariable predictor-outcome associations. Currently, methods to incorporate such information in the construction of a prediction model are underdeveloped and unfamiliar to many researchers. Methods This...
Published in: | BMC Medical Research Methodology |
---|---|
Main Authors: | , , , , , |
Format: | Article in Journal/Newspaper |
Language: | English |
Published: |
Springer Science and Business Media LLC
2012
|
Subjects: | |
Online Access: | http://dx.doi.org/10.1186/1471-2288-12-121 http://link.springer.com/content/pdf/10.1186/1471-2288-12-121.pdf http://link.springer.com/article/10.1186/1471-2288-12-121/fulltext.html http://link.springer.com/content/pdf/10.1186/1471-2288-12-121 https://link.springer.com/content/pdf/10.1186/1471-2288-12-121.pdf |
id |
crspringernat:10.1186/1471-2288-12-121 |
---|---|
record_format |
openpolar |
spelling |
crspringernat:10.1186/1471-2288-12-121 2023-05-15T16:29:32+02:00 Incorporating published univariable associations in diagnostic and prognostic modeling A Debray, Thomas P Koffijberg, Hendrik Lu, Difei Vergouwe, Yvonne Steyerberg, Ewout W M Moons, Karel G 2012 http://dx.doi.org/10.1186/1471-2288-12-121 http://link.springer.com/content/pdf/10.1186/1471-2288-12-121.pdf http://link.springer.com/article/10.1186/1471-2288-12-121/fulltext.html http://link.springer.com/content/pdf/10.1186/1471-2288-12-121 https://link.springer.com/content/pdf/10.1186/1471-2288-12-121.pdf en eng Springer Science and Business Media LLC http://www.springer.com/tdm BMC Medical Research Methodology volume 12, issue 1 ISSN 1471-2288 Health Informatics Epidemiology journal-article 2012 crspringernat https://doi.org/10.1186/1471-2288-12-121 2022-01-04T10:55:27Z Abstract Background Diagnostic and prognostic literature is overwhelmed with studies reporting univariable predictor-outcome associations. Currently, methods to incorporate such information in the construction of a prediction model are underdeveloped and unfamiliar to many researchers. Methods This article aims to improve upon an adaptation method originally proposed by Greenland (1987) and Steyerberg (2000) to incorporate previously published univariable associations in the construction of a novel prediction model. The proposed method improves upon the variance estimation component by reconfiguring the adaptation process in established theory and making it more robust. Different variants of the proposed method were tested in a simulation study, where performance was measured by comparing estimated associations with their predefined values according to the Mean Squared Error and coverage of the 90% confidence intervals. Results Results demonstrate that performance of estimated multivariable associations considerably improves for small datasets where external evidence is included. Although the error of estimated associations decreases with increasing amount of individual participant data, it does not disappear completely, even in very large datasets. Conclusions The proposed method to aggregate previously published univariable associations with individual participant data in the construction of a novel prediction models outperforms established approaches and is especially worthwhile when relatively limited individual participant data are available. Article in Journal/Newspaper Greenland Springer Nature (via Crossref) Greenland BMC Medical Research Methodology 12 1 |
institution |
Open Polar |
collection |
Springer Nature (via Crossref) |
op_collection_id |
crspringernat |
language |
English |
topic |
Health Informatics Epidemiology |
spellingShingle |
Health Informatics Epidemiology A Debray, Thomas P Koffijberg, Hendrik Lu, Difei Vergouwe, Yvonne Steyerberg, Ewout W M Moons, Karel G Incorporating published univariable associations in diagnostic and prognostic modeling |
topic_facet |
Health Informatics Epidemiology |
description |
Abstract Background Diagnostic and prognostic literature is overwhelmed with studies reporting univariable predictor-outcome associations. Currently, methods to incorporate such information in the construction of a prediction model are underdeveloped and unfamiliar to many researchers. Methods This article aims to improve upon an adaptation method originally proposed by Greenland (1987) and Steyerberg (2000) to incorporate previously published univariable associations in the construction of a novel prediction model. The proposed method improves upon the variance estimation component by reconfiguring the adaptation process in established theory and making it more robust. Different variants of the proposed method were tested in a simulation study, where performance was measured by comparing estimated associations with their predefined values according to the Mean Squared Error and coverage of the 90% confidence intervals. Results Results demonstrate that performance of estimated multivariable associations considerably improves for small datasets where external evidence is included. Although the error of estimated associations decreases with increasing amount of individual participant data, it does not disappear completely, even in very large datasets. Conclusions The proposed method to aggregate previously published univariable associations with individual participant data in the construction of a novel prediction models outperforms established approaches and is especially worthwhile when relatively limited individual participant data are available. |
format |
Article in Journal/Newspaper |
author |
A Debray, Thomas P Koffijberg, Hendrik Lu, Difei Vergouwe, Yvonne Steyerberg, Ewout W M Moons, Karel G |
author_facet |
A Debray, Thomas P Koffijberg, Hendrik Lu, Difei Vergouwe, Yvonne Steyerberg, Ewout W M Moons, Karel G |
author_sort |
A Debray, Thomas P |
title |
Incorporating published univariable associations in diagnostic and prognostic modeling |
title_short |
Incorporating published univariable associations in diagnostic and prognostic modeling |
title_full |
Incorporating published univariable associations in diagnostic and prognostic modeling |
title_fullStr |
Incorporating published univariable associations in diagnostic and prognostic modeling |
title_full_unstemmed |
Incorporating published univariable associations in diagnostic and prognostic modeling |
title_sort |
incorporating published univariable associations in diagnostic and prognostic modeling |
publisher |
Springer Science and Business Media LLC |
publishDate |
2012 |
url |
http://dx.doi.org/10.1186/1471-2288-12-121 http://link.springer.com/content/pdf/10.1186/1471-2288-12-121.pdf http://link.springer.com/article/10.1186/1471-2288-12-121/fulltext.html http://link.springer.com/content/pdf/10.1186/1471-2288-12-121 https://link.springer.com/content/pdf/10.1186/1471-2288-12-121.pdf |
geographic |
Greenland |
geographic_facet |
Greenland |
genre |
Greenland |
genre_facet |
Greenland |
op_source |
BMC Medical Research Methodology volume 12, issue 1 ISSN 1471-2288 |
op_rights |
http://www.springer.com/tdm |
op_doi |
https://doi.org/10.1186/1471-2288-12-121 |
container_title |
BMC Medical Research Methodology |
container_volume |
12 |
container_issue |
1 |
_version_ |
1766019242325442560 |