Treatment Non-Adherence Bias in Clinical Machine Learning: A Real-World Study on Hypertension Medication

Zhongyuan Liang, Arvind Suresh, Irene Y. Chen
Proceedings of the sixth Conference on Health, Inference, and Learning, PMLR 287:430-442, 2025.

Abstract

Machine learning systems trained on electronic health records (EHRs) increasingly guide treatment decisions, but their reliability depends on the critical assumption that patients follow the prescribed treatments recorded in EHRs. Using EHR data from 3,623 hypertension patients, we investigate how treatment non-adherence introduces implicit bias that can fundamentally distort both causal inference and predictive modeling. By extracting patient adherence information from clinical notes using a large language model, we identify 786 patients (21.7%) with medication non-adherence. We further uncover key demographic and clinical factors associated with non-adherence, as well as patient-reported reasons including side effects and difficulties obtaining refills. Our findings demonstrate that this implicit bias can not only reverse estimated treatment effects, but also degrade model performance by up to 5% while disproportionately affecting vulnerable populations by exacerbating disparities in decision outcomes and model error rates. This highlights the importance of accounting for treatment non-adherence in developing responsible and equitable clinical machine learning systems.

Cite this Paper

BibTeX
@InProceedings{pmlr-v287-liang25a, title = {Treatment Non-Adherence Bias in Clinical Machine Learning: A Real-World Study on Hypertension Medication}, author = {Liang, Zhongyuan and Suresh, Arvind and Chen, Irene Y.}, booktitle = {Proceedings of the sixth Conference on Health, Inference, and Learning}, pages = {430--442}, year = {2025}, editor = {Xu, Xuhai Orson and Choi, Edward and Singhal, Pankhuri and Gerych, Walter and Tang, Shengpu and Agrawal, Monica and Subbaswamy, Adarsh and Sizikova, Elena and Dunn, Jessilyn and Daneshjou, Roxana and Sarker, Tasmie and McDermott, Matthew and Chen, Irene}, volume = {287}, series = {Proceedings of Machine Learning Research}, month = {25--27 Jun}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v287/main/assets/liang25a/liang25a.pdf}, url = {https://proceedings.mlr.press/v287/liang25a.html}, abstract = {Machine learning systems trained on electronic health records (EHRs) increasingly guide treatment decisions, but their reliability depends on the critical assumption that patients follow the prescribed treatments recorded in EHRs. Using EHR data from 3,623 hypertension patients, we investigate how treatment non-adherence introduces implicit bias that can fundamentally distort both causal inference and predictive modeling. By extracting patient adherence information from clinical notes using a large language model, we identify 786 patients (21.7%) with medication non-adherence. We further uncover key demographic and clinical factors associated with non-adherence, as well as patient-reported reasons including side effects and difficulties obtaining refills. Our findings demonstrate that this implicit bias can not only reverse estimated treatment effects, but also degrade model performance by up to 5% while disproportionately affecting vulnerable populations by exacerbating disparities in decision outcomes and model error rates. This highlights the importance of accounting for treatment non-adherence in developing responsible and equitable clinical machine learning systems.} }
Endnote
%0 Conference Paper %T Treatment Non-Adherence Bias in Clinical Machine Learning: A Real-World Study on Hypertension Medication %A Zhongyuan Liang %A Arvind Suresh %A Irene Y. Chen %B Proceedings of the sixth Conference on Health, Inference, and Learning %C Proceedings of Machine Learning Research %D 2025 %E Xuhai Orson Xu %E Edward Choi %E Pankhuri Singhal %E Walter Gerych %E Shengpu Tang %E Monica Agrawal %E Adarsh Subbaswamy %E Elena Sizikova %E Jessilyn Dunn %E Roxana Daneshjou %E Tasmie Sarker %E Matthew McDermott %E Irene Chen %F pmlr-v287-liang25a %I PMLR %P 430--442 %U https://proceedings.mlr.press/v287/liang25a.html %V 287 %X Machine learning systems trained on electronic health records (EHRs) increasingly guide treatment decisions, but their reliability depends on the critical assumption that patients follow the prescribed treatments recorded in EHRs. Using EHR data from 3,623 hypertension patients, we investigate how treatment non-adherence introduces implicit bias that can fundamentally distort both causal inference and predictive modeling. By extracting patient adherence information from clinical notes using a large language model, we identify 786 patients (21.7%) with medication non-adherence. We further uncover key demographic and clinical factors associated with non-adherence, as well as patient-reported reasons including side effects and difficulties obtaining refills. Our findings demonstrate that this implicit bias can not only reverse estimated treatment effects, but also degrade model performance by up to 5% while disproportionately affecting vulnerable populations by exacerbating disparities in decision outcomes and model error rates. This highlights the importance of accounting for treatment non-adherence in developing responsible and equitable clinical machine learning systems.
APA
Liang, Z., Suresh, A. & Chen, I.Y.. (2025). Treatment Non-Adherence Bias in Clinical Machine Learning: A Real-World Study on Hypertension Medication. Proceedings of the sixth Conference on Health, Inference, and Learning, in Proceedings of Machine Learning Research 287:430-442 Available from https://proceedings.mlr.press/v287/liang25a.html.

Related Material