Statistically Valid Post-Deployment Monitoring Should Be Standard for AI-Based Digital Health

TL;DR

This paper advocates for the adoption of statistically valid, label-efficient testing frameworks as a standard for post-deployment monitoring of clinical AI, to ensure safety, reliability, and regulatory compliance in real-world healthcare settings.

Contribution

It introduces a formal, statistically rigorous approach to post-deployment monitoring, framing change detection as hypothesis testing to improve reliability and reproducibility.

Findings

Current monitoring practices are manual and reactive.

Statistically valid testing provides explicit error guarantees.

Framework supports reproducibility and formal inference.

Abstract

This position paper argues that post-deployment monitoring in clinical AI is underdeveloped and proposes statistically valid and label-efficient testing frameworks as a principled foundation for ensuring reliability and safety in real-world deployment. A recent review found that only 9% of FDA-registered AI-based healthcare tools include a post-deployment surveillance plan. Existing monitoring approaches are often manual, sporadic, and reactive, making them ill-suited for the dynamic environments in which clinical models operate. We contend that post-deployment monitoring should be grounded in label-efficient and statistically valid testing frameworks, offering a principled alternative to current practices. We use the term "statistically valid" to refer to methods that provide explicit guarantees on error rates (e.g., Type I/II error), enable formal inference under pre-defined assumptions, and support reproducibility--features that align with regulatory requirements. Specifically, we propose that the detection of changes in the data and model performance degradation should be framed as distinct statistical hypothesis testing problems. Grounding monitoring in statistical rigor ensures a reproducible and scientifically sound basis for maintaining the reliability of clinical AI systems. Importantly, it also opens new research directions for the technical community--spanning theory, methods, and tools for statistically principled detection, attribution, and mitigation of post-deployment model failures in real-world settings.