A flexible approach to sequential prediction under intervention

TL;DR

This paper introduces a causal predictive framework for estimating health risks under various preventative interventions, emphasizing models that handle mediators and interventions flexibly while ensuring causal consistency.

Contribution

It presents novel models for risk prediction under interventions, including the Unexposed Mediator Model, Modifiable Risk Factor Model, and a combined Two Component Model, advancing causal prediction methods.

Findings

Models enable evaluation of arbitrary interventions within a causal framework

Framework provides causally consistent risk estimates across repeated visits

Illustrated with primary prevention of cardiovascular disease

Abstract

We propose a causal predictive framework for estimating risk under preventative interventions. The Unexposed Mediator Model maintains mediators that are also predictors at their unexposed level, removing double counting of intervention effects at followup visits. The Modifiable Risk Factor Model handles multiple interventions flexibly by modelling their effects via mediators that are also predictors, assuming a known causal structure. The Two Component Model combines a predictive baseline model with an intervention model to improve predictive performance. We illustrate the framework in primary prevention of cardiovascular disease. The proposed models allow arbitrary interventions to be evaluated within a prediction under intervention framework, with causally consistent risk estimates across repeated visits. Limitations include reliance on predictor values from an arbitrary first visit, requirements for causal structural knowledge, and a consistency assumption, that interventions with identical effects on predictors have identical effects on outcomes, which warrant further investigation.