Policy Learning with Observational Data: The Case of Hepatitis C Treatment for HIV/HCV Co-Infected Patients

Abstract

Decision-makers frequently must choose a single action from a finite set of alternatives -- for example, physicians selecting a treatment, investors choosing a portfolio risk level, or judges determining sentences. To improve outcomes, policymakers often issue policy rules or guidelines to inform such choices. In this paper, I show how to generally derive policy rules from observational data in a multi-action framework under relatively weak assumptions about the underlying structure of the heterogeneous sampled population. Conditional average treatment effects (CATEs) are consistently estimated via a weighted K-means algorithm, assuming the outcome model is correctly specified within each homogeneous subgroup. I then use standard decision trees to implement feasible policy rules, allowing for both perfect and imperfect adherence to treatment. The methodology is applied to treatment options for Hepatitis C (HCV) among patients co-infected with human immunodeficiency virus (HIV), a setting in which no uniform guideline exists for modern pharmaceutical therapies. The results identify a subgroup of patients with approximately an 80% probability of spontaneous HCV clearance without treatment. Estimation results also show that reallocating treatments among treated patients could have reduced total treatment costs by CAN$2.5-4.5 million while increasing aggregate health benefits relative to the status quo. These findings demonstrate that the proposed approach can generate improved, data-driven treatment guidelines for the management of HIV/HCV co-infected patients.

Threshold Geometry, Bifurcation, and Data-Driven Analysis of a Vaccination-Treatment Model for Hepatitis B
Mustaq Ahmad, A. S. Bhadauria
arXiv9 Oct 2026PreprintCC BY 4.0Full textDynamical systemsarXiv
Regarding the inference of causality in observational non-controlled studies. Comment on “Assessment of sexual function and its association with quality of life and disease severity in patients with atopic dermatitis”
Vanessa Rigoni Marcato, Anna Carolina Miola, Paula Rosa Coutinho Goulart Borges Mariottoni +1
Anais Brasileiros de Dermatologia3 Nov 2026Peer-reviewedCC BYFull textStatisticsDOAJ
Treatment persistence drives estimator performance in longitudinal causal inference based on observational data: A simulation study
Sergio Gaiotti, Sara Poletto, Enrico Longato +2
arXiv4 Sep 2026PreprintarXiv licenceFull textStatisticsarXiv
Bayesian additive regression trees for evaluating treatment benefit predictors using observational data
Yuan Xia, Mohsen Sadatsafavi, Jeenat Mehareen +1
arXiv17 Sep 2026PreprintarXiv licenceFull textStatisticsarXiv
Learning Optimal Prescriptive Trees from Observational Data
Nathanael Jo, Sina Aghaei, Andrés Gómez +1
Management Science30 Jul 2026Peer-reviewedLicence not statedFull text8 citedStatisticsOpenAlex

Cite

BibTeX RIS

BibTeX

@misc{langevin2026policy,
  title = {{Policy Learning with Observational Data: The Case of Hepatitis C Treatment for HIV/HCV Co-Infected Patients}},
  author = {Raphaël Langevin},
  howpublished = {arXiv},
  year = {2026},
  month = oct,
  doi = {10.48550/arxiv.2605.16593},
  eprint = {2605.16593},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2605.16593},
  note = {CC BY 4.0},
}

RIS

TY  - UNPB
TI  - Policy Learning with Observational Data: The Case of Hepatitis C Treatment for HIV/HCV Co-Infected Patients
AU  - Raphaël Langevin
PY  - 2026
DA  - 2026/10/21
PB  - arXiv
DO  - 10.48550/arxiv.2605.16593
UR  - https://arxiv.org/abs/2605.16593
AB  - Decision-makers frequently must choose a single action from a finite set of alternatives -- for example, physicians selecting a treatment, investors choosing a portfolio risk level, or judges determining sentences. To improve outcomes, policymakers often issue policy rules or guidelines to inform such choices. In this paper, I show how to generally derive policy rules from observational data in a multi-action framework under relatively weak assumptions about the underlying structure of the heterogeneous sampled population. Conditional average treatment effects (CATEs) are consistently estimated via a weighted K-means algorithm, assuming the outcome model is correctly specified within each homogeneous subgroup. I then use standard decision trees to implement feasible policy rules, allowing for both perfect and imperfect adherence to treatment. The methodology is applied to treatment options for Hepatitis C (HCV) among patients co-infected with human immunodeficiency virus (HIV), a setting in which no uniform guideline exists for modern pharmaceutical therapies. The results identify a subgroup of patients with approximately an 80% probability of spontaneous HCV clearance without treatment. Estimation results also show that reallocating treatments among treated patients could have reduced total treatment costs by CAN$2.5-4.5 million while increasing aggregate health benefits relative to the status quo. These findings demonstrate that the proposed approach can generate improved, data-driven treatment guidelines for the management of HIV/HCV co-infected patients.
ER  -

Abstract

Related

Cite