A clinician's guide for developing a prediction model: a case study using real-world data of patients with castration-resistant prostate cancer

Kevin M Veen; Isabel B de Angst; Mostafa M Mokhles; Hans M Westgeest; Malou Kuppen; Carin A Uyl-de Groot; Winald R Gerritsen; Paul J M Kil; Johanna J M Takkenberg

doi:10.1007/s00432-020-03286-8

A clinician's guide for developing a prediction model: a case study using real-world data of patients with castration-resistant prostate cancer

Kevin M Veen, Isabel B de Angst, Mostafa M Mokhles, Hans M Westgeest, Malou Kuppen, Carin A Uyl-de Groot, Winald R Gerritsen, Paul J M Kil, Johanna J M Takkenberg

Research output: Contribution to journal › Review article › peer-review

Abstract

PURPOSE: With the increasing interest in treatment decision-making based on risk prediction models, it is essential for clinicians to understand the steps in developing and interpreting such models.

METHODS: A retrospective registry of 20 Dutch hospitals with data on patients treated for castration-resistant prostate cancer was used to guide clinicians through the steps of developing a prediction model. The model of choice was the Cox proportional hazard model.

RESULTS: Using the exemplary dataset several essential steps in prediction modelling are discussed including: coding of predictors, missing values, interaction, model specification and performance. An advanced method for appropriate selection of main effects, e.g. Least Absolute Shrinkage and Selection Operator (LASSO) regression, is described. Furthermore, the assumptions of Cox proportional hazard model are discussed, and how to handle violations of the proportional hazard assumption using time-varying coefficients.

CONCLUSION: This study provides a comprehensive detailed guide to bridge the gap between the statistician and clinician, based on a large dataset of real-world patients treated for castration-resistant prostate cancer.

Original language	English
Pages (from-to)	2067-2075
Number of pages	9
Journal	Journal of Cancer Research and Clinical Oncology
Volume	146
Issue number	8
DOIs	https://doi.org/10.1007/s00432-020-03286-8
Publication status	Published - Aug 2020
Externally published	Yes

Keywords

Clinical Decision-Making
Decision Support Systems, Clinical
Humans
Male
Models, Statistical
Netherlands
Proportional Hazards Models
Prostatic Neoplasms, Castration-Resistant/therapy
Registries
Regression Analysis
Retrospective Studies

Access to Document

10.1007/s00432-020-03286-8

Cite this

Veen, K. M., de Angst, I. B., Mokhles, M. M., Westgeest, H. M., Kuppen, M., Groot, C. A. U., Gerritsen, W. R., Kil, P. J. M., & Takkenberg, J. J. M. (2020). A clinician's guide for developing a prediction model: a case study using real-world data of patients with castration-resistant prostate cancer. Journal of Cancer Research and Clinical Oncology, 146(8), 2067-2075. https://doi.org/10.1007/s00432-020-03286-8

@article{802ce25234744dcf8acdb93bb450289f,

title = "A clinician's guide for developing a prediction model: a case study using real-world data of patients with castration-resistant prostate cancer",

abstract = "PURPOSE: With the increasing interest in treatment decision-making based on risk prediction models, it is essential for clinicians to understand the steps in developing and interpreting such models.METHODS: A retrospective registry of 20 Dutch hospitals with data on patients treated for castration-resistant prostate cancer was used to guide clinicians through the steps of developing a prediction model. The model of choice was the Cox proportional hazard model.RESULTS: Using the exemplary dataset several essential steps in prediction modelling are discussed including: coding of predictors, missing values, interaction, model specification and performance. An advanced method for appropriate selection of main effects, e.g. Least Absolute Shrinkage and Selection Operator (LASSO) regression, is described. Furthermore, the assumptions of Cox proportional hazard model are discussed, and how to handle violations of the proportional hazard assumption using time-varying coefficients.CONCLUSION: This study provides a comprehensive detailed guide to bridge the gap between the statistician and clinician, based on a large dataset of real-world patients treated for castration-resistant prostate cancer.",

keywords = "Clinical Decision-Making, Decision Support Systems, Clinical, Humans, Male, Models, Statistical, Netherlands, Proportional Hazards Models, Prostatic Neoplasms, Castration-Resistant/therapy, Registries, Regression Analysis, Retrospective Studies",

author = "Veen, {Kevin M} and {de Angst}, {Isabel B} and Mokhles, {Mostafa M} and Westgeest, {Hans M} and Malou Kuppen and Groot, {Carin A Uyl-de} and Gerritsen, {Winald R} and Kil, {Paul J M} and Takkenberg, {Johanna J M}",

note = "Publisher Copyright: {\textcopyright} 2020, The Author(s).",

year = "2020",

month = aug,

doi = "10.1007/s00432-020-03286-8",

language = "English",

volume = "146",

pages = "2067--2075",

journal = "Journal of Cancer Research and Clinical Oncology",

issn = "0171-5216",

publisher = "Springer-Verlag",

number = "8",

}

Veen, KM, de Angst, IB, Mokhles, MM, Westgeest, HM, Kuppen, M, Groot, CAU, Gerritsen, WR, Kil, PJM & Takkenberg, JJM 2020, 'A clinician's guide for developing a prediction model: a case study using real-world data of patients with castration-resistant prostate cancer', Journal of Cancer Research and Clinical Oncology, vol. 146, no. 8, pp. 2067-2075. https://doi.org/10.1007/s00432-020-03286-8

A clinician's guide for developing a prediction model: a case study using real-world data of patients with castration-resistant prostate cancer. / Veen, Kevin M; de Angst, Isabel B; Mokhles, Mostafa M et al.
In: Journal of Cancer Research and Clinical Oncology, Vol. 146, No. 8, 08.2020, p. 2067-2075.

Research output: Contribution to journal › Review article › peer-review

TY - JOUR

T1 - A clinician's guide for developing a prediction model

T2 - a case study using real-world data of patients with castration-resistant prostate cancer

AU - Veen, Kevin M

AU - de Angst, Isabel B

AU - Mokhles, Mostafa M

AU - Westgeest, Hans M

AU - Kuppen, Malou

AU - Groot, Carin A Uyl-de

AU - Gerritsen, Winald R

AU - Kil, Paul J M

AU - Takkenberg, Johanna J M

PY - 2020/8

Y1 - 2020/8

N2 - PURPOSE: With the increasing interest in treatment decision-making based on risk prediction models, it is essential for clinicians to understand the steps in developing and interpreting such models.METHODS: A retrospective registry of 20 Dutch hospitals with data on patients treated for castration-resistant prostate cancer was used to guide clinicians through the steps of developing a prediction model. The model of choice was the Cox proportional hazard model.RESULTS: Using the exemplary dataset several essential steps in prediction modelling are discussed including: coding of predictors, missing values, interaction, model specification and performance. An advanced method for appropriate selection of main effects, e.g. Least Absolute Shrinkage and Selection Operator (LASSO) regression, is described. Furthermore, the assumptions of Cox proportional hazard model are discussed, and how to handle violations of the proportional hazard assumption using time-varying coefficients.CONCLUSION: This study provides a comprehensive detailed guide to bridge the gap between the statistician and clinician, based on a large dataset of real-world patients treated for castration-resistant prostate cancer.

AB - PURPOSE: With the increasing interest in treatment decision-making based on risk prediction models, it is essential for clinicians to understand the steps in developing and interpreting such models.METHODS: A retrospective registry of 20 Dutch hospitals with data on patients treated for castration-resistant prostate cancer was used to guide clinicians through the steps of developing a prediction model. The model of choice was the Cox proportional hazard model.RESULTS: Using the exemplary dataset several essential steps in prediction modelling are discussed including: coding of predictors, missing values, interaction, model specification and performance. An advanced method for appropriate selection of main effects, e.g. Least Absolute Shrinkage and Selection Operator (LASSO) regression, is described. Furthermore, the assumptions of Cox proportional hazard model are discussed, and how to handle violations of the proportional hazard assumption using time-varying coefficients.CONCLUSION: This study provides a comprehensive detailed guide to bridge the gap between the statistician and clinician, based on a large dataset of real-world patients treated for castration-resistant prostate cancer.

KW - Clinical Decision-Making

KW - Decision Support Systems, Clinical

KW - Humans

KW - Male

KW - Models, Statistical

KW - Netherlands

KW - Proportional Hazards Models

KW - Prostatic Neoplasms, Castration-Resistant/therapy

KW - Registries

KW - Regression Analysis

KW - Retrospective Studies

U2 - 10.1007/s00432-020-03286-8

DO - 10.1007/s00432-020-03286-8

M3 - Review article

C2 - 32556680

SN - 0171-5216

VL - 146

SP - 2067

EP - 2075

JO - Journal of Cancer Research and Clinical Oncology

JF - Journal of Cancer Research and Clinical Oncology

IS - 8

ER -

A clinician's guide for developing a prediction model: a case study using real-world data of patients with castration-resistant prostate cancer

Abstract

Keywords

Access to Document

Fingerprint

Cite this