Retrospective modeling study of ChatGPT (GPT 4) warfarin dose adjustment in patients with INRs outside the therapeutic range
Research article published in Digital health (2026)
Abstract
OBJECTIVE: This retrospective, modeled study evaluates the accuracy of ChatGPT (GPT-4)-based warfarin dose adjustments compared to clinician recommendations at the Cardiology Clinic of Konya City Hospital, focusing on patients with international normalized ratio (INR) values outside the therapeutic range (2-3). We hypothesized that ChatGPT could provide reliable, consistent dose guidance. METHODS: We reviewed the records of warfarin-treated patients from 1 June 2022 through 24 November 2024. Clinical data used by physicians (e.g. baseline INR, warfarin indication, comorbidities, and current dose) were provided to ChatGPT to generate hypothetical weekly dose recommendations. ChatGPT's impact on INR normalization was modeled using standard dose-response assumptions and compared with actual outcomes under physician-guided therapy. RESULTS: A total of 180 patients met the inclusion criteria. ChatGPT's recommended doses were within ±1 mg/week of physician prescriptions in 74% of cases and within ±2 mg/week in 84%. The mean physician dose was 28.0 ± 6.1 mg/week versus ChatGPT's 27.5 ± 5.9 mg/week (p = .12). Seventy-two percent of patients achieved therapeutic INR under physician-managed dosing, while the model suggested a 69% success rate for ChatGPT-guided dosing (p = .15). Real-world adverse events were infrequent under physician management (1.1% major bleeding, 0.6% thrombotic events). CONCLUSION: In this retrospective, exploratory analysis with modeled outcomes, ChatGPT's weekly dose suggestions showed high concordance with clinician dosing. These findings are hypothesis-generating and do not establish clinical efficacy or safety; prospective, physician-supervised trials-potentially integrated with home INR monitoring-are required for validation.
Abstract sourced from PubMed (NCBI) for the cited record. See the original publication for the authoritative version.
Summary
Peer-reviewed research on anticoagulant and antithrombotic drug development relevant to leech-derived therapeutics. Indexed in PubMed and verified against the NCBI record.
Why This Matters for Hirudotherapy
This retrospective, modeled analysis of 180 warfarin-treated cardiology patients tested whether ChatGPT (GPT-4) could reproduce clinician weekly-dose adjustments for INRs outside the 2-3 therapeutic range; the model's doses fell within +/-1 mg/week of physician prescriptions in 74% of cases and within +/-2 mg/week in 84%, with no statistically significant difference in modeled INR-normalization rates, while real-world adverse events under physician care were infrequent (1.1% major bleeding, 0.6% thrombotic). For ASH this sits in the anticoagulation-management context that frames medicinal-leech practice: warfarin is the vitamin K antagonist clinicians titrate against bleeding-versus-clotting risk, the same balance that defines when an externally applied biological anticoagulant like the leech is appropriate, and it underscores how delicate INR control is in the populations leech therapy often touches (post-surgical, congested-flap patients). The strict caveat is that this is an exploratory, retrospective study with MODELED rather than measured outcomes: the authors themselves state it is hypothesis-generating and does not establish efficacy or safety, and it concerns a software dosing aid, not hirudotherapy, so it informs surrounding anticoagulation thinking only and carries no direct clinical claim for leeching.
Citation
Retrospective modeling study of ChatGPT (GPT 4) warfarin dose adjustment in patients with INRs outside the therapeutic range.
Tezcan et al. · Digital health, 2026
Added to ASH library: May 28, 2026 · Site last updated: June 18, 2026