Letter to the Editor: Retrospective Modeling Study of ChatGPT (GPT-4) Warfarin Dose Adjustment in Patients with INRs Outside the Therapeutic Range
By
Mehmet U. Çalışkan
Gökhan Keskin
May 27, 2026
Clinical Scorecard: Analysis of ChatGPT (GPT-4) for Warfarin Dose Modification in Patients with INRs Deviating from the Therapeutic Range
At a Glance
Category Detail
Condition Key Mechanisms Use of large language model (LLM) for simulated INR monitoring and dose adjustment, considering patient-specific needs. Target Population Care Setting
Key Highlights
Study highlights the potential role of ChatGPT in warfarin dose adjustment. INR target range of 2.0–3.0 is uniformly defined, which may not reflect patient-specific needs. Practical constraints exist in translating calculated doses into available tablet formulations. Clinician-integrated AI-supported clinical decision support systems (CDSS) are emphasized as a promising direction. Importance of patient-specific INR targets in dose adjustment. Guideline-Based Recommendations
Diagnosis
Management
Warfarin dose adjustment should consider individual patient clinical context and treatment indication. Incorporate patient-specific INR targets for dose adjustment. Monitoring & Follow-up
Risks
Patient & Prescribing Data
Weekly doses derived from target INR may not be feasible with available tablet strengths; practical dosing considerations are essential.
Clinical Best Practices
Incorporate patient-specific INR targets for dose adjustment. Consider formulation-related constraints when adjusting warfarin therapy. Integrate AI with clinician oversight to address practical constraints in dosing. Related Resources & Content
