Empirical Evidence That There Is No Such Thing As A Validated Prediction Model

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: van Leeuwen, Florian D., Steyerberg, Ewout W., van Klaveren, David, Wessler, Ben, Kent, David M., van Zwet, Erik W.
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866913389853278208
author van Leeuwen, Florian D.
Steyerberg, Ewout W.
van Klaveren, David
Wessler, Ben
Kent, David M.
van Zwet, Erik W.
author_facet van Leeuwen, Florian D.
Steyerberg, Ewout W.
van Klaveren, David
Wessler, Ben
Kent, David M.
van Zwet, Erik W.
contents Background: External validations are essential to assess clinical prediction models (CPMs) before deployment. Apart from model misspecification, differences in patient population and other factors influence a model's AUC (c-statistic). We aimed to quantify variation in AUCs across external validation studies and adjust expectations of a model's performance in a new setting. Methods: The Tufts-PACE CPM Registry contains CPMs for cardiovascular disease prognosis. We analyzed the AUCs of 469 CPMs with a total of 1,603 external validations. For each CPM, we performed a random effects meta-analysis to estimate the between-study standard deviation $τ$ among the AUCs. Since the majority of these meta-analyses has only a handful of validations, this leads to very poor estimates of $τ$. So, we estimated a log normal distribution of $τ$ across all CPMs and used this as an empirical prior. We compared this empirical Bayesian approach with frequentist meta-analyses using cross-validation. Results: The 469 CPMs had a median of 2 external validations (IQR: [1-3]). The estimated distribution of $τ$ had a mean of 0.055 and a standard deviation of 0.015. If $τ$ = 0.05, the 95% prediction interval for the AUC in a new setting is at least +/- 0.1, regardless of the number of validations. Frequentist methods underestimate the uncertainty about the AUC in a new setting. Accounting for $τ$ in a Bayesian approach achieved near nominal coverage. Conclusion: Due to large heterogeneity among the validated AUC values of a CPM, there is great irreducible uncertainty in predicting the AUC in a new setting. This uncertainty is underestimated by existing methods. The proposed empirical Bayes approach addresses this problem which merits wide application in judging the validity of prediction models.
format Preprint
id arxiv_https___arxiv_org_abs_2406_08628
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Empirical Evidence That There Is No Such Thing As A Validated Prediction Model
van Leeuwen, Florian D.
Steyerberg, Ewout W.
van Klaveren, David
Wessler, Ben
Kent, David M.
van Zwet, Erik W.
Methodology
Background: External validations are essential to assess clinical prediction models (CPMs) before deployment. Apart from model misspecification, differences in patient population and other factors influence a model's AUC (c-statistic). We aimed to quantify variation in AUCs across external validation studies and adjust expectations of a model's performance in a new setting. Methods: The Tufts-PACE CPM Registry contains CPMs for cardiovascular disease prognosis. We analyzed the AUCs of 469 CPMs with a total of 1,603 external validations. For each CPM, we performed a random effects meta-analysis to estimate the between-study standard deviation $τ$ among the AUCs. Since the majority of these meta-analyses has only a handful of validations, this leads to very poor estimates of $τ$. So, we estimated a log normal distribution of $τ$ across all CPMs and used this as an empirical prior. We compared this empirical Bayesian approach with frequentist meta-analyses using cross-validation. Results: The 469 CPMs had a median of 2 external validations (IQR: [1-3]). The estimated distribution of $τ$ had a mean of 0.055 and a standard deviation of 0.015. If $τ$ = 0.05, the 95% prediction interval for the AUC in a new setting is at least +/- 0.1, regardless of the number of validations. Frequentist methods underestimate the uncertainty about the AUC in a new setting. Accounting for $τ$ in a Bayesian approach achieved near nominal coverage. Conclusion: Due to large heterogeneity among the validated AUC values of a CPM, there is great irreducible uncertainty in predicting the AUC in a new setting. This uncertainty is underestimated by existing methods. The proposed empirical Bayes approach addresses this problem which merits wide application in judging the validity of prediction models.
title Empirical Evidence That There Is No Such Thing As A Validated Prediction Model
topic Methodology
url https://arxiv.org/abs/2406.08628