Editverse · Clinical Decision Tools
Free HAS-BLED Bleeding Risk Calculator (2026)
HAS-BLED is a 9-point bedside score that estimates the 1-year risk of major bleeding in patients taking oral anticoagulation, most often for atrial fibrillation. It scores hypertension, abnormal renal or liver function, prior stroke, a history of bleeding or bleeding predisposition, labile INR control, older age, and bleeding-prone drugs or alcohol use — each contributing one point, to a maximum of nine. It is built for clinicians who have already identified a patient’s stroke risk (typically with CHA₂DS₂-VASc) and now need to weigh that risk against the bleeding risk of anticoagulation, and — critically — to see which of that bleeding risk is modifiable before treatment starts. A high score is a prompt to fix what can be fixed, not a reason to withhold anticoagulation outright.
Clinical decision-support aid, not a substitute for clinical judgment. These tools implement published scoring criteria for education and workflow support. They are not a medical device and have not been reviewed by the FDA or any regulatory body. Every result should be interpreted by a qualified clinician in the context of the full clinical picture, not used as a standalone diagnostic or treatment decision.
Bleeding risk on anticoagulation
HAS-BLED
Estimates 1-year risk of major bleeding in patients on oral anticoagulation for atrial fibrillation, and — just as importantly — flags which of that risk is modifiable before treatment starts.
Pisters R, Lane DA, Nieuwlaat R, de Vos CB, Crijns HJGM, Lip GYH. A novel user-friendly score (HAS-BLED) to assess 1-year risk of major bleeding in patients with atrial fibrillation: the Euro Heart Survey. Chest. 2010;138(5):1093–1100.
Worked example
Consider a 70-year-old patient with uncontrolled hypertension (systolic blood pressure consistently above 160 mmHg despite treatment) who takes a daily over-the-counter NSAID for chronic joint pain, and has no history of renal or hepatic impairment, no prior stroke, no bleeding history, a stable INR, and no heavy alcohol use.
Age over 65 contributes 1 point (elderly), uncontrolled hypertension contributes 1 point, and the NSAID use contributes 1 point under the “drugs predisposing to bleeding” criterion. The remaining six criteria are all absent. HAS-BLED = 1 + 1 + 1 = 3, placing this patient in the high-risk category. The correct clinical response is not to withhold anticoagulation on the basis of this score — it is to address the modifiable factors the score just identified: get the blood pressure under control and stop the NSAID, then reassess. Two of this patient’s three points are fixable; only “elderly” is not. Check “Hypertension”, “Elderly”, and “Drugs predisposing to bleeding” in the calculator above to reproduce this result.
Methods
Bleeding risk was estimated using the HAS-BLED score (1 point each for hypertension, abnormal renal function, abnormal liver function, prior stroke, bleeding history or predisposition, labile INR, age >65 years, and concomitant bleeding-prone drugs or alcohol use, to a maximum of 9 points) as derived by Pisters et al. in the Euro Heart Survey on atrial fibrillation cohort (Pisters R, et al. Chest. 2010;138(5):1093–1100).
Assumptions and when not to use this tool
HAS-BLED was designed to surface modifiable bleeding risk, not to serve as a gatekeeper for anticoagulation. A high score should prompt closer monitoring and correction of what can be corrected — blood pressure control, review of concomitant antiplatelet or NSAID use, alcohol reduction, tighter INR management if the patient is on a vitamin K antagonist — rather than a decision to withhold anticoagulation from a patient whose stroke risk (per CHA₂DS₂-VASc or CHA₂DS₂-VA) otherwise warrants it.
Several of its inputs overlap with stroke-risk criteria (age, prior stroke, hypertension), which is expected: bleeding and thrombotic risk share underlying drivers, and the two scores are meant to be read together, not as competing verdicts. The labile-INR criterion only applies to patients on a vitamin K antagonist such as warfarin; it should be scored as absent (0 points) for patients on a direct oral anticoagulant, where INR is not a monitoring parameter.
Does a high HAS-BLED score mean I should stop anticoagulation?
No. HAS-BLED was designed to flag modifiable bleeding risk so it can be addressed, not to justify withholding anticoagulation from a patient whose stroke risk otherwise warrants it. A score of 3 or more calls for closer monitoring and correction of reversible factors — blood pressure control, review of NSAID or antiplatelet use, alcohol reduction — alongside, not instead of, the stroke-risk decision.
How is HAS-BLED different from CHA₂DS₂-VASc?
They answer two different questions about the same decision. CHA₂DS₂-VASc (or CHA₂DS₂-VA) estimates the stroke risk that anticoagulation is meant to reduce; HAS-BLED estimates the bleeding risk that anticoagulation introduces. Neither score is meant to be used alone — a high stroke risk and a high bleeding risk commonly coexist in the same patient, since age, hypertension, and prior stroke drive both scores, and the clinical task is to manage the bleeding risk down rather than let it override the stroke-risk decision.
What counts as a “labile INR” if a patient isn’t taking warfarin?
It scores as absent (0 points). The labile-INR criterion — unstable or supratherapeutic INR values, or a time-in-therapeutic-range below 60% — only applies to patients on a vitamin K antagonist like warfarin, where INR is the relevant monitoring parameter. Patients on a direct oral anticoagulant are not assessed on this criterion.
Primary references
- Pisters R, Lane DA, Nieuwlaat R, de Vos CB, Crijns HJGM, Lip GYH. A novel user-friendly score (HAS-BLED) to assess 1-year risk of major bleeding in patients with atrial fibrillation: the Euro Heart Survey. Chest. 2010;138(5):1093–1100.
Studying bleeding risk, anticoagulation safety, or drug-related adverse events using national survey data? Editverse’s NHANES Study service supports the full analysis pipeline for NHANES-based research — from cohort and variable selection through weighted statistical analysis and manuscript-ready output.