Maintenance Notice

Due to necessary scheduled maintenance, the JMIR Publications website will be unavailable from Wednesday, July 01, 2020 at 8:00 PM to 10:00 PM EST. We apologize in advance for any inconvenience this may cause you.

Who will be affected?

Accepted for/Published in: JMIR Cardio

Date Submitted: Jul 7, 2023
Date Accepted: Nov 24, 2024

The final, peer-reviewed published version of this preprint can be found here:

Application of Dragonnet and Conformal Inference for Estimating Individualized Treatment Effects for Personalized Stroke Prevention: Retrospective Cohort Study

Lolak S, Attia J, McKay GJ, Thakkinstian A

Application of Dragonnet and Conformal Inference for Estimating Individualized Treatment Effects for Personalized Stroke Prevention: Retrospective Cohort Study

JMIR Cardio 2025;9:e50627

DOI: 10.2196/50627

PMID: 39780350

PMCID: 11735012

Warning: This is an author submission that is not peer-reviewed or edited. Preprints - unless they show as "accepted" - should not be relied on to guide clinical practice or health-related behavior and should not be reported in news media as established information.

Causal and Individual treatment effect in Stroke: A Retrospective Cohort Study

  • Sermkiat Lolak; 
  • John Attia; 
  • Gareth J. McKay; 
  • Ammarin Thakkinstian

ABSTRACT

Introduction: Stroke is a major cause of death and disability. This study aims to determine the causal effects and individual treatment effects (ITEs) using real-world data. Method: This study included high-risk patients treated at Ramathibodi Hospital, Thailand between 2010-2020. Hospital records were used to identify risk factors including hypertension (HT), diabetes (DM), dyslipidemia (DLP), and atrial fibrillation (AF). Ischemic/ hemorrhagic stroke was the main outcome. Machine learning and conventional methods were used to estimate causal effect, while weighted split-conformal quantile regression conformal inference (CI) was used to calculate ITEs.

Results:

AF, HT, and DM were significant stroke risk factors with average causal (risk) effect ranging from 0.075-0.097, 0.017-0.025, and 0.006-0.01. Estimated causal (risk) ratios from Dragonnet associated with these corresponding factors were 4.56 (4.56,4.57), 2.44 (2.41,2.46), and 1.41 (1.21,1.60) respectively. Mean ITEs indicated that there were several patients with DM or DM with HT who were not currently receiving antiplatelet treatment and would be more likely to benefit if they had received it. Conclusion: This study provides causal estimates of AF, HT, and DM on stroke. This study improves our understanding of stroke risk and highlights the need for further research to inform treatment options for high-risk patients.


 Citation

Please cite as:

Lolak S, Attia J, McKay GJ, Thakkinstian A

Application of Dragonnet and Conformal Inference for Estimating Individualized Treatment Effects for Personalized Stroke Prevention: Retrospective Cohort Study

JMIR Cardio 2025;9:e50627

DOI: 10.2196/50627

PMID: 39780350

PMCID: 11735012

Download PDF


Request queued. Please wait while the file is being generated. It may take some time.

© The authors. All rights reserved. This is a privileged document currently under peer-review/community review (or an accepted/rejected manuscript). Authors have provided JMIR Publications with an exclusive license to publish this preprint on it's website for review and ahead-of-print citation purposes only. While the final peer-reviewed paper may be licensed under a cc-by license on publication, at this stage authors and publisher expressively prohibit redistribution of this draft paper other than for review purposes.