# Backdoor Adjustment of Confounding by Provenance for Robust Text Classification of Multi-institutional Clinical Notes

arXiv: https://arxiv.org/abs/2310.02451 · HTML: https://mlsanity.com/paper/2310.02451/

## Key facts

- Cited by 5 papers in the map (data through 14 Oct 2026).
- Placed in Safety alignment, refusal and guardrails, one of 12 directions in LLM safety and security.
- Builds on 7 of its 28 references that are in the map.
- First posted on arXiv 3 Oct 2023.

## Placed in

- LLM safety and security / [Safety alignment, refusal and guardrails](https://mlsanity.com/safety/safety-alignment-refusal-and-guardrails/)

## Builds on (most cited first, 7 of 7)

- [BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://mlsanity.com/paper/1810.04805/) (2018)
- [Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks](https://mlsanity.com/paper/1908.10084/) (2019)
- [XGBoost: A Scalable Tree Boosting System](https://mlsanity.com/paper/1603.02754/) (2016)
- [Deconfounded Image Captioning: A Causal Retrospect](https://mlsanity.com/paper/2003.03923/) (2020)
- [Annotating Social Determinants of Health Using Active Learning, and Characterizing Determinants Using Neural Event Extraction](https://mlsanity.com/paper/2004.05438/) (2020)
- [The 2022 n2c2/UW Shared Task on Extracting Social Determinants of Health](https://mlsanity.com/paper/2301.05571/) (2023)
- [A Causally Formulated Hazard Ratio Estimation through Backdoor Adjustment on Structural Causal Model](https://mlsanity.com/paper/2006.12573/) (2020)

## Built on by (most cited first, 5 of 5)

- [Mitigating Confounding in Speech-Based Dementia Detection through Weight Masking](https://mlsanity.com/paper/2506.05610/) (2025)
- [Enhancing Robustness of Foundation Model Representations under Provenance-related Distribution Shifts](https://mlsanity.com/paper/2312.05435/) (2023)
- [When Prompts Interact: Assessing Prompt Arithmetic for Deconfounding under Distribution Shift](https://mlsanity.com/paper/2605.03096/) (2026)
- [DeconDTN-Toolkit: A Library for Evaluation and Enhancement of Robustness to Provenance Shift](https://mlsanity.com/paper/2605.11237/) (2026)
- [The record is part of the task: matched-record evaluation of text classifiers across maintenance, safety and recall reporting](https://mlsanity.com/paper/2609.16267/) (2026)

## Abstract (arXiv)

Natural Language Processing (NLP) methods have been broadly applied to clinical tasks. Machine learning and deep learning approaches have been used to improve the performance of clinical NLP. However, these approaches require sufficiently large datasets for training, and trained models have been shown to transfer poorly across sites. These issues have led to the promotion of data collection and integration across different institutions for accurate and portable models. However, this can introduce a form of bias called confounding by provenance. When source-specific data distributions differ at deployment, this may harm model performance. To address this issue, we evaluate the utility of backdoor adjustment for text classification in a multi-site dataset of clinical notes annotated for mentions of substance abuse. Using an evaluation framework devised to measure robustness to distributional shifts, we assess the utility of backdoor adjustment. Our results indicate that backdoor adjustment can effectively mitigate for confounding shift.