Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Sequential Counterfactual Inference for Temporal Clinical Data: Addressing the Time Traveler Dilemma

Jingya Cheng, Alaleh Azhir, Jiazi Tian, Hossein Estiri

TL;DR

The Sequential Counterfactual Framework is introduced, which respects temporal dependencies in electronic health records by distinguishing immutable features from controllable features and modeling how interventions propagate through time, yielding clinically actionable insights grounded in biological plausibility.

Abstract

Counterfactual inference enables clinicians to ask "what if" questions about patient outcomes, but standard methods assume feature independence and simultaneous modifiability -- assumptions violated by longitudinal clinical data. We introduce the Sequential Counterfactual Framework, which respects temporal dependencies in electronic health records by distinguishing immutable features (chronic diagnoses) from controllable features (lab values) and modeling how interventions propagate through time. Applied to 2,723 COVID-19 patients (383 Long COVID heart failure cases, 2,340 matched controls), we demonstrate that 38-67% of patients with chronic conditions would require biologically impossible counterfactuals under naive methods. We identify a cardiorenal cascade (CKD -> AKI -> HF) with relative risks of 2.27 and 1.19 at each step, illustrating temporal propagation that sequential -- but not naive -- counterfactuals can capture. Our framework transforms counterfactual explanation from "what if this feature were different?" to "what if we had intervened earlier, and how would that propagate forward?" -- yielding clinically actionable insights grounded in biological plausibility.

Sequential Counterfactual Inference for Temporal Clinical Data: Addressing the Time Traveler Dilemma

TL;DR

The Sequential Counterfactual Framework is introduced, which respects temporal dependencies in electronic health records by distinguishing immutable features from controllable features and modeling how interventions propagate through time, yielding clinically actionable insights grounded in biological plausibility.

Abstract

Counterfactual inference enables clinicians to ask "what if" questions about patient outcomes, but standard methods assume feature independence and simultaneous modifiability -- assumptions violated by longitudinal clinical data. We introduce the Sequential Counterfactual Framework, which respects temporal dependencies in electronic health records by distinguishing immutable features (chronic diagnoses) from controllable features (lab values) and modeling how interventions propagate through time. Applied to 2,723 COVID-19 patients (383 Long COVID heart failure cases, 2,340 matched controls), we demonstrate that 38-67% of patients with chronic conditions would require biologically impossible counterfactuals under naive methods. We identify a cardiorenal cascade (CKD -> AKI -> HF) with relative risks of 2.27 and 1.19 at each step, illustrating temporal propagation that sequential -- but not naive -- counterfactuals can capture. Our framework transforms counterfactual explanation from "what if this feature were different?" to "what if we had intervened earlier, and how would that propagate forward?" -- yielding clinically actionable insights grounded in biological plausibility.
Paper Structure (24 sections, 9 equations, 2 figures, 2 tables, 1 algorithm)

This paper contains 24 sections, 9 equations, 2 figures, 2 tables, 1 algorithm.

Table of Contents

  1. Introduction
  2. Background and Related Work
  3. Counterfactual Explanations in Machine Learning
  4. Counterfactual Inference in Healthcare
  5. Temporal Modeling in Clinical Data
  6. Causal Inference and Counterfactuals
  7. Long COVID and Cardiovascular Complications
  8. Summary: The Gap We Address
  9. Methods
  10. Feature Taxonomy for Temporal Data
  11. Temporal Dependency Structure
  12. Intervention Pathway Specification.
  13. Plausibility Constraints
  14. Violation Detection.
  15. Counterfactual Quality Metrics
  16. ...and 9 more sections

Figures (2)

  • Figure 1: Temporal persistence of chronic conditions. Probability of diagnosis in Last period given presence (red) or absence (blue) in History. The 6--13$\times$ differences demonstrate temporal dependencies.
  • Figure 2: Cardiorenal cascade. Top: Flow diagram showing CKD$\rightarrow$AKI (RR=2.27)$\rightarrow$HF (RR=1.19). Bottom: Naive vs. sequential approaches.

Theorems & Definitions (5)

  • definition 1: Temporal Feature Representation
  • definition 2: Feature Taxonomy
  • definition 3: Temporal Dependency Graph
  • definition 4: Plausible Counterfactual
  • definition 5: Propagation Operator