Artificial intelligence surrogates for treatment effect estimation with before-and-after data
Abstract
Estimating the causal effects of medical treatments is difficult when clinically important outcomes are costly to measure or require long follow-up. Short-term or inexpensive surrogate outcomes offer a potential alternative, but surrogate biomarkers may be unavailable or difficult to identify. Advances in artificial intelligence (AI) have enabled increasingly accurate prediction of clinical outcomes from inexpensive, high-dimensional measurements, which creates an opportunity to use AI predictions themselves as surrogates. To this end, we develop a framework for estimating treatment effects from paired measurements obtained before and after treatment for each treated individual. A pretrained AI model is applied to the before and after measurements, and our estimator compares the resulting outcome predictions. We characterize the technical assumptions under which this within-person contrast identifies the average treatment effect on the treated, even when clinical outcomes are never observed for treated individuals. When these assumptions cannot be justified, we use prediction-powered inference to correct bias using a small number of observed clinical outcomes and obtain valid inference. Synthetic and real-world cardio-oncology experiments demonstrate the validity and accuracy of the approach.
Related
- Efficient Deep Gaussian Process Surrogates via Mini-Batch Latent MAP Estimation
- Causal Mediation Analysis with High-Dimensional Surrogates: Identification, Estimation and Inference under Latent Factor Models
- Causal Inference with Generative Artificial Intelligence: Application to Texts as Treatments
- Differentially Private Average Treatment Effect Estimation by Propensity Score Blocking
- The Akaike Information Criterion and Artificial Intelligence
BibTeX
@misc{dean2026artificial,
title = {{Artificial intelligence surrogates for treatment effect estimation with before-and-after data}},
author = {Frances Dean and Anna Neufeld and Joshua Barrios and Geoffrey H Tison and Ahmed Alaa},
howpublished = {arXiv},
year = {2026},
month = sep,
doi = {10.48550/arxiv.2609.27180},
eprint = {2609.27180},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2609.27180},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - Artificial intelligence surrogates for treatment effect estimation with before-and-after data
AU - Frances Dean
AU - Anna Neufeld
AU - Joshua Barrios
AU - Geoffrey H Tison
AU - Ahmed Alaa
PY - 2026
DA - 2026/09/23
PB - arXiv
DO - 10.48550/arxiv.2609.27180
UR - https://arxiv.org/abs/2609.27180
AB - Estimating the causal effects of medical treatments is difficult when clinically important outcomes are costly to measure or require long follow-up. Short-term or inexpensive surrogate outcomes offer a potential alternative, but surrogate biomarkers may be unavailable or difficult to identify. Advances in artificial intelligence (AI) have enabled increasingly accurate prediction of clinical outcomes from inexpensive, high-dimensional measurements, which creates an opportunity to use AI predictions themselves as surrogates. To this end, we develop a framework for estimating treatment effects from paired measurements obtained before and after treatment for each treated individual. A pretrained AI model is applied to the before and after measurements, and our estimator compares the resulting outcome predictions. We characterize the technical assumptions under which this within-person contrast identifies the average treatment effect on the treated, even when clinical outcomes are never observed for treated individuals. When these assumptions cannot be justified, we use prediction-powered inference to correct bias using a small number of observed clinical outcomes and obtain valid inference. Synthetic and real-world cardio-oncology experiments demonstrate the validity and accuracy of the approach.
ER -