The fundamental problem of risk prediction for individuals: health AI, uncertainty, and personalized medicine

TL;DR

This paper investigates the various sources of uncertainty in individual health risk predictions, revealing that model and applicability uncertainties often surpass estimation uncertainty, impacting personalized medicine.

Contribution

It empirically demonstrates the dominance of model and applicability uncertainty over estimation uncertainty in individual health risk predictions, emphasizing the need for cautious interpretation.

Findings

Estimation uncertainty can be overshadowed by model and applicability uncertainty.

Increasing training data reduces estimation uncertainty but not model/applicability uncertainty.

Individual risk estimates are often more uncertain than commonly assumed.

Abstract

Background: Clinical prediction models for a health condition are commonly evaluated regarding performance for a population, although decisions are made for individuals. The classic view relates uncertainty in risk estimates for individuals to sample size (estimation uncertainty) but uncertainty can also be caused by model uncertainty (variability in modeling choices) and applicability uncertainty (variability in measurement procedures and between populations). Methods: We used real and synthetic data for ovarian cancer diagnosis to train 59400 models with variations in estimation, model, and applicability uncertainty. We then used these models to estimate the probability of ovarian cancer in a fixed test set of 100 patients and evaluate the variability in individual estimates. Findings: We show empirically that estimation uncertainty can be strongly dominated by model uncertainty and applicability uncertainty, even for models that perform well at the population level. Estimation uncertainty decreased considerably with increasing training sample size, whereas model and applicability uncertainty remained large. Interpretation: Individual risk estimates are far more uncertain than often assumed. Model uncertainty and applicability uncertainty usually remain invisible when prediction models or algorithms are based on a single study. Predictive algorithms should inform, not dictate, care and support personalization through clinician-patient interaction rather than through inherently uncertain model outputs. Funding: This research is supported by Research Foundation Flanders (FWO) grants G097322N, G049312N, G0B4716N, and 12F3114N to BVC and/or DTi, KU Leuven internal grants C24M/20/064 and C24/15/037 to BVC and/or DT, ZoNMW VIDI grant 09150172310023 to LW. DT is a senior clinical investigator of FWO.