Cargando…

Calibration Belt for Quality-of-Care Assessment Based on Dichotomous Outcomes

Prognostic models applied in medicine must be validated on independent samples, before their use can be recommended. The assessment of calibration, i.e., the model's ability to provide reliable predictions, is crucial in external validation studies. Besides having several shortcomings, statisti...

Descripción completa

Detalles Bibliográficos
Autores principales: Finazzi, Stefano, Poole, Daniele, Luciani, Davide, Cogo, Paola E., Bertolini, Guido
Formato: Texto
Lenguaje:English
Publicado: Public Library of Science 2011
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3043050/
https://www.ncbi.nlm.nih.gov/pubmed/21373178
http://dx.doi.org/10.1371/journal.pone.0016110
_version_ 1782198594265153536
author Finazzi, Stefano
Poole, Daniele
Luciani, Davide
Cogo, Paola E.
Bertolini, Guido
author_facet Finazzi, Stefano
Poole, Daniele
Luciani, Davide
Cogo, Paola E.
Bertolini, Guido
author_sort Finazzi, Stefano
collection PubMed
description Prognostic models applied in medicine must be validated on independent samples, before their use can be recommended. The assessment of calibration, i.e., the model's ability to provide reliable predictions, is crucial in external validation studies. Besides having several shortcomings, statistical techniques such as the computation of the standardized mortality ratio (SMR) and its confidence intervals, the Hosmer–Lemeshow statistics, and the Cox calibration test, are all non-informative with respect to calibration across risk classes. Accordingly, calibration plots reporting expected versus observed outcomes across risk subsets have been used for many years. Erroneously, the points in the plot (frequently representing deciles of risk) have been connected with lines, generating false calibration curves. Here we propose a methodology to create a confidence band for the calibration curve based on a function that relates expected to observed probabilities across classes of risk. The calibration belt allows the ranges of risk to be spotted where there is a significant deviation from the ideal calibration, and the direction of the deviation to be indicated. This method thus offers a more analytical view in the assessment of quality of care, compared to other approaches.
format Text
id pubmed-3043050
institution National Center for Biotechnology Information
language English
publishDate 2011
publisher Public Library of Science
record_format MEDLINE/PubMed
spelling pubmed-30430502011-03-03 Calibration Belt for Quality-of-Care Assessment Based on Dichotomous Outcomes Finazzi, Stefano Poole, Daniele Luciani, Davide Cogo, Paola E. Bertolini, Guido PLoS One Research Article Prognostic models applied in medicine must be validated on independent samples, before their use can be recommended. The assessment of calibration, i.e., the model's ability to provide reliable predictions, is crucial in external validation studies. Besides having several shortcomings, statistical techniques such as the computation of the standardized mortality ratio (SMR) and its confidence intervals, the Hosmer–Lemeshow statistics, and the Cox calibration test, are all non-informative with respect to calibration across risk classes. Accordingly, calibration plots reporting expected versus observed outcomes across risk subsets have been used for many years. Erroneously, the points in the plot (frequently representing deciles of risk) have been connected with lines, generating false calibration curves. Here we propose a methodology to create a confidence band for the calibration curve based on a function that relates expected to observed probabilities across classes of risk. The calibration belt allows the ranges of risk to be spotted where there is a significant deviation from the ideal calibration, and the direction of the deviation to be indicated. This method thus offers a more analytical view in the assessment of quality of care, compared to other approaches. Public Library of Science 2011-02-23 /pmc/articles/PMC3043050/ /pubmed/21373178 http://dx.doi.org/10.1371/journal.pone.0016110 Text en Finazzi et al. http://creativecommons.org/licenses/by/4.0/ This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are properly credited.
spellingShingle Research Article
Finazzi, Stefano
Poole, Daniele
Luciani, Davide
Cogo, Paola E.
Bertolini, Guido
Calibration Belt for Quality-of-Care Assessment Based on Dichotomous Outcomes
title Calibration Belt for Quality-of-Care Assessment Based on Dichotomous Outcomes
title_full Calibration Belt for Quality-of-Care Assessment Based on Dichotomous Outcomes
title_fullStr Calibration Belt for Quality-of-Care Assessment Based on Dichotomous Outcomes
title_full_unstemmed Calibration Belt for Quality-of-Care Assessment Based on Dichotomous Outcomes
title_short Calibration Belt for Quality-of-Care Assessment Based on Dichotomous Outcomes
title_sort calibration belt for quality-of-care assessment based on dichotomous outcomes
topic Research Article
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3043050/
https://www.ncbi.nlm.nih.gov/pubmed/21373178
http://dx.doi.org/10.1371/journal.pone.0016110
work_keys_str_mv AT finazzistefano calibrationbeltforqualityofcareassessmentbasedondichotomousoutcomes
AT pooledaniele calibrationbeltforqualityofcareassessmentbasedondichotomousoutcomes
AT lucianidavide calibrationbeltforqualityofcareassessmentbasedondichotomousoutcomes
AT cogopaolae calibrationbeltforqualityofcareassessmentbasedondichotomousoutcomes
AT bertoliniguido calibrationbeltforqualityofcareassessmentbasedondichotomousoutcomes