Sensitivity Analysis for Trials with Irregular and Informative Assessment Times: Marginal Mean Models and Sensitivity Parameter Selection
Abstract
In randomized trials, outcome assessment times often vary across participants and may depend on the underlying outcome process, so outcomes observed near a target time need not represent the target population. Whether assessment is informative cannot be determined from the observed data alone. We extend the sensitivity analysis framework of Smith et al.\ (2024), which is anchored in an explainable assessment assumption and was restricted to continuous outcomes through an identity-link marginal mean model, to a broader class of link functions that accommodates binary and count outcomes. Within this framework, we develop an influence function-based augmented inverse intensity-weighted estimator that is root-n consistent under standard regularity conditions. We also propose a data-driven procedure for selecting the range of the sensitivity parameter from plausible bounds on the target mean outcome and for constructing a confidence set for this range, which accounts for uncertainty in sensitivity parameter selection. We illustrate the method with data from the ARC study and evaluate its finite-sample performance in simulation studies.
Related
- Turtles All the Way Down: Parameter Sensitivity in Metaheuristic Hyperparameter Optimisation for Machine Learning Models
- Sensitivity-Guided Parameter Estimation of Bubbly Liquid via a Nonlinear Wave Equation and Physics-InformedNeuralNetworks
- Kernelized Parameter Sensitivity Atlases for Amortized Conditional Interpretability
- Topological sensitivity analysis for some nonlinear PDE systems
- Global Sensitivity Analysis for Studies Extending Inferences From a Randomized Trial to a Target Population
BibTeX
@misc{gao2026sensitivity,
title = {{Sensitivity Analysis for Trials with Irregular and Informative Assessment Times: Marginal Mean Models and Sensitivity Parameter Selection}},
author = {Yujing Gao and Shu Yang and Ravi Varadhan and Andrew Redd and Andrea J. Apter and Daniel O. Scharfstein},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2610.25454},
eprint = {2610.25454},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2610.25454},
}
RIS
TY - UNPB
TI - Sensitivity Analysis for Trials with Irregular and Informative Assessment Times: Marginal Mean Models and Sensitivity Parameter Selection
AU - Yujing Gao
AU - Shu Yang
AU - Ravi Varadhan
AU - Andrew Redd
AU - Andrea J. Apter
AU - Daniel O. Scharfstein
PY - 2026
DA - 2026/10/29
PB - arXiv
DO - 10.48550/arxiv.2610.25454
UR - https://arxiv.org/abs/2610.25454
AB - In randomized trials, outcome assessment times often vary across participants and may depend on the underlying outcome process, so outcomes observed near a target time need not represent the target population. Whether assessment is informative cannot be determined from the observed data alone. We extend the sensitivity analysis framework of Smith et al.\ (2024), which is anchored in an explainable assessment assumption and was restricted to continuous outcomes through an identity-link marginal mean model, to a broader class of link functions that accommodates binary and count outcomes. Within this framework, we develop an influence function-based augmented inverse intensity-weighted estimator that is root-n consistent under standard regularity conditions. We also propose a data-driven procedure for selecting the range of the sensitivity parameter from plausible bounds on the target mean outcome and for constructing a confidence set for this range, which accounts for uncertainty in sensitivity parameter selection. We illustrate the method with data from the ARC study and evaluate its finite-sample performance in simulation studies.
ER -