Clinical world models respond differently to whole treatment bundles
Intervention Granularity Matters: Coherent Treatment Bundles in Counterfactual Simulation with Clinical World Models
Machine Learning
Summary
When predicting how a patient might respond to a change in treatment, it's important to know if the change affects just one part or the entire group of related treatments. The authors found that changing a whole treatment bundle, rather than a single setting, causes bigger predicted effects in their clinical model. This means that looking at treatments as complete groups can help make better predictions about patient outcomes. Ignoring this could make a model less sensitive to real treatment impacts.
What this means in practice
- •For hospital data teams: Improve clinical decision support by using treatment bundles rather than single settings to simulate patient responses more accurately.
- •For medical device engineers: Design better simulation tools that consider complete device parameter bundles for ventilators to predict patient outcomes more reliably.
Tested on one dataset.
Authors
Fangzhou Wang, Yixuan Yang, Camilla Balzarotti, Rishikesan Kamaleswaran
Abstract
Counterfactual simulation with a clinical world model means fixing a patient's history, changing the treatment, and reading off the predicted response. Doing so requires deciding what counts as one intervention. In clinical settings, interventions are documented as bundles: a co-occurrence audit of 945,707 patient-hours from MIMIC-IV shows groups of components, such as every parameter of a dialysis circuit, that never appear apart, so an edit that changes one component on its own describes an hour that never occurs in the data. We hypothesize that the granularity at which an intervention is edited changes how a world model responds, and test this with Clin-JEPA, a latent world model of patient trajectories conditioned on hourly treatment text. At 1,019 documented onsets of invasive ventilation, we keep the patient's history and other treatments fixed and compare editing one ventilator setting with editing the complete configuration recorded for a real patient with the most similar recent trajectory. The complete bundle moves the predicted next state further than any single setting, consistently across all five settings, and the difference remains after accounting for how much each edit changes the model's input. Intervention granularity therefore materially affects the response of a clinical world model: single-component edits may understate treatment sensitivity, and bundle-aware editing may offer a better-supported basis for counterfactual treatment simulation.