Holdouts set for safe predictive model updating

TL;DR

This paper introduces a holdout set approach for updating predictive risk scores to maintain accuracy over time without bias, demonstrating its effectiveness with a pre-eclampsia risk score example.

Contribution

It proposes a method for selecting an optimal holdout set size to improve risk score updates, with algorithms for estimating this size and practical application to pre-eclampsia risk prediction.

Findings

Optimal holdout size around 10,000 individuals for pre-eclampsia score

Method reduces adverse outcomes asymptotically to the best possible level

Algorithms for estimating holdout size are effective and practical

Abstract

Predictive risk scores for adverse outcomes are increasingly crucial in guiding health interventions. Such scores may need to be periodically updated due to change in the distributions they model. However, directly updating risk scores used to guide intervention can lead to biased risk estimates. To address this, we propose updating using a `holdout set' - a subset of the population that does not receive interventions guided by the risk score. Balancing the holdout set size is essential to ensure good performance of the updated risk score whilst minimising the number of held out samples. We prove that this approach reduces adverse outcome frequency to an asymptotically optimal level and argue that often there is no competitive alternative. We describe conditions under which an optimal holdout size (OHS) can be readily identified, and introduce parametric and semi-parametric algorithms for OHS estimation. We apply our methods to the ASPRE risk score for pre-eclampsia to recommend a plan for updating it in the presence of change in the underlying data distribution. We show that, in order to minimise the number of pre-eclampsia cases over time, this is best achieved using a holdout set of around 10,000 individuals.