Path-specific harm decomposition: A partial identification framework
Abstract
A central goal when designing treatment policies is often to "do no harm", that is, to avoid interventions that improve average outcomes while worsening outcomes for some individuals. A widely used notion for harm is the fraction of negatively affected (FNA), defined as the probability that an intervention decreases an individual's outcome. However, in many applications, treatments operate through mediators, and a single "total" FNA can obscure whether harm arises primarily through direct pathways or indirect (mediator-induced) pathways. In this work, we introduce a path-specific analogue of the FNA. For this, we disentangle total harm into direct and indirect harm in causal mediation settings. However, these quantities depend on joint distributions of potential outcomes that are not point-identified even in randomised controlled trials. As a remedy, we develop a novel partial identification framework for direct and indirect FNA. In our framework, we (i) derive sharp Makarov bounds for the FNA, and (ii) propose a semiparametrically efficient estimator with valid confidence intervals for these bounds under mild margin conditions. We demonstrate our framework across various numerical experiments. To the best of our knowledge, we are the first to study path-specific decomposition of causal harm and to develop an orthogonal inference framework for its analysis.
Related
- Efficient Numerical Evaluation of Arbitrary High-order Partial Derivatives via Algebraic Decomposition
- Freeness for the G-circulant Decomposition of the Partial Transpose of Random Matrices
- Gaussian BROJA Partial Information Decomposition in Closed Form
- A Quantum Path to Partial Differential Equations
- Solving Fractional Partial Differential Equations Using the Double Natural Transform Decomposition Method
BibTeX
@misc{yan2026pathspecific,
title = {{Path-specific harm decomposition: A partial identification framework}},
author = {Ruizi Yan and Dennis Frauen and Maresa Schröder and Stefan Feuerriegel},
howpublished = {arXiv},
year = {2026},
month = sep,
doi = {10.48550/arxiv.2609.29938},
eprint = {2609.29938},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2609.29938},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - Path-specific harm decomposition: A partial identification framework
AU - Ruizi Yan
AU - Dennis Frauen
AU - Maresa Schröder
AU - Stefan Feuerriegel
PY - 2026
DA - 2026/09/24
PB - arXiv
DO - 10.48550/arxiv.2609.29938
UR - https://arxiv.org/abs/2609.29938
AB - A central goal when designing treatment policies is often to "do no harm", that is, to avoid interventions that improve average outcomes while worsening outcomes for some individuals. A widely used notion for harm is the fraction of negatively affected (FNA), defined as the probability that an intervention decreases an individual's outcome. However, in many applications, treatments operate through mediators, and a single "total" FNA can obscure whether harm arises primarily through direct pathways or indirect (mediator-induced) pathways. In this work, we introduce a path-specific analogue of the FNA. For this, we disentangle total harm into direct and indirect harm in causal mediation settings. However, these quantities depend on joint distributions of potential outcomes that are not point-identified even in randomised controlled trials. As a remedy, we develop a novel partial identification framework for direct and indirect FNA. In our framework, we (i) derive sharp Makarov bounds for the FNA, and (ii) propose a semiparametrically efficient estimator with valid confidence intervals for these bounds under mild margin conditions. We demonstrate our framework across various numerical experiments. To the best of our knowledge, we are the first to study path-specific decomposition of causal harm and to develop an orthogonal inference framework for its analysis.
ER -