HAS-BLED Score Calculator for Bleeding Risk Assessment in AF
The HAS-BLED score is a clinical tool used to assess 1-year risk of major bleeding in patients with atrial fibrillation who are being considered for anticoagulation therapy.
About
The HAS-BLED bleeding risk score (Hypertension, Abnormal Renal/Liver Function, Stroke, Bleeding History or Predisposition, Labile INR, Elderly, Drugs/Alcohol) was developed by Pisters et al. and published in Chest in 2010 using data from 3,978 patients in the Euro Heart Survey on Atrial Fibrillation. The score was specifically designed to assess 1-year risk of major bleeding in patients with atrial fibrillation who are being considered for anticoagulation therapy. It comprises nine clinical variables, each contributing 1 point for a maximum score of 9: uncontrolled hypertension (systolic blood pressure >160 mmHg), abnormal renal function (chronic dialysis, renal transplantation, or serum creatinine >2.6 mg/dL or 200 µmol/L), abnormal liver function (cirrhosis or bilirubin >2 times the upper limit of normal with AST/ALT/ALP >3 times the upper limit of normal), prior stroke (particularly lacunar stroke which is associated with increased bleeding risk), prior major bleeding history or predisposition (anemia, bleeding diathesis), labile international normalized ratio (unstable or high INRs with poor time in therapeutic range <60% for patients on warfarin), elderly age (≥65 years), drug therapy predisposing to bleeding (antiplatelet agents, nonsteroidal anti-inflammatory drugs), and excessive alcohol consumption (≥8 drinks per week). The score performs well for bleeding risk prediction with a c-statistic of approximately 0.72 in the original derivation cohort. It has been externally validated in numerous AF populations and is now recommended by the ESC, AHA/ACC, and NICE guidelines for routine bleeding risk assessment before initiating anticoagulation. Its evidence level is Grade B.
Formula
Hypertension (1) + Abnormal Renal Function (1) + Abnormal Liver Function (1) + Stroke (1) + Bleeding History (1) + Labile INR (1) + Elderly ≥65 (1) + Drugs (1) + Alcohol (1)
The HAS-BLED score is calculated by summing 1 point for each of nine clinical variables present at the time of assessment, yielding a total ranging from 0 to 9. Hypertension is defined as uncontrolled systolic blood pressure >160 mmHg — this differs from the hypertension definition in CHADS-VASc (which uses >140/90 mmHg) and requires active elevation rather than a history of treated hypertension. Abnormal renal function (1 point) encompasses chronic dialysis, renal transplantation, or serum creatinine >2.6 mg/dL (200 µmol/L). Abnormal liver function (1 point) includes cirrhosis or biochemical evidence of significant hepatic impairment (bilirubin >2 times upper limit of normal with transaminases or alkaline phosphatase >3 times upper limit of normal). Prior stroke (1 point) captures the paradoxical observation that patients with prior stroke are at increased risk for both ischemic events AND hemorrhage, particularly cerebral hemorrhage. Bleeding history or predisposition (1 point) includes prior major bleeding (requiring hospitalization or transfusion), anemia (hemoglobin <13 g/dL in men, <12 g/dL in women), or known bleeding diathesis. Labile INR (1 point) applies to patients on warfarin with unstable anticoagulation control (time in therapeutic range [TTR] <60%). Elderly age ≥65 years (1 point) recognizes age as an independent risk factor for bleeding. Drug use (1 point) includes concurrent antiplatelet agents (aspirin, clopidogrel) or NSAIDs that synergistically increase bleeding risk with anticoagulation. Alcohol excess (1 point) is defined as ≥8 standard drinks per week. The score is interpreted dichotomously: a score of 0-2 indicates low bleeding risk (1-year major bleeding rate 1.0-3.7%), while a score of ≥3 indicates high bleeding risk (1-year major bleeding rate 4.0-12.5%). Importantly, a high score should prompt identification and correction of modifiable risk factors (hypertension control, avoidance of NSAIDs, reducing alcohol intake, improving INR control) rather than automatic withholding of anticoagulation.
Score Interpretation
The HAS-BLED score is the most widely validated bleeding risk assessment tool in atrial fibrillation and is recommended by the 2020 ESC Guidelines for atrial fibrillation, the 2024 AHA/ACC Guideline for AF management, and the NICE Guideline NG196 for routine assessment before initiating oral anticoagulation. Its primary clinical impact is not to deny anticoagulation to high-risk patients but rather to guide risk factor modification, inform the choice of anticoagulant, and tailor the intensity of follow-up. The critical message emphasized by all major guidelines is that a high HAS-BLED score (≥3) should NOT lead to withholding of anticoagulation in patients who need it, because the net clinical benefit of stroke prevention with anticoagulation in AF almost uniformly outweighs the bleeding risk. Instead, a high score should trigger a structured approach to identify and address modifiable bleeding risk factors: optimizing blood pressure control (targeting systolic BP <140 mmHg), avoiding unnecessary antiplatelet or NSAID co-prescription, reducing or eliminating alcohol consumption, and improving INR stability with better warfarin management or switching to a DOAC. DOACs have consistently demonstrated lower rates of major bleeding (particularly intracranial hemorrhage) compared to warfarin, and this risk reduction is maintained even in patients with high HAS-BLED scores, making DOACs the preferred choice in this population. The HAS-BLED score has also been validated in DOAC-treated patients, confirming its utility across all anticoagulation strategies. Additionally, the score performs well in predicting major bleeding events beyond gastrointestinal bleeding, including intracranial hemorrhage (c-statistic 0.65-0.70), which carries the highest morbidity and mortality. The HAS-BLED score complements the CHADS-VASc score in what is now regarded as the standard shared decision-making framework for AF management: assess stroke risk with CHADS-VASc, assess bleeding risk with HAS-BLED, and engage the patient in a discussion of individualized risks and benefits.
Low Bleeding Risk — 0–2
1-year major bleeding risk: 1.0-3.7%. Safe to proceed with anticoagulation.
Management: Proceed with anticoagulation as indicated. Reassess bleeding risk annually and when clinical status changes.
High Bleeding Risk — 3–9
1-year major bleeding risk: 4.0-12.5%. Caution with anticoagulation.
Management: Consider alternative to OAC if possible (e.g., LAA occlusion). If anticoagulation required, use lower DOAC dose. Correct reversible bleeding risk factors. Close follow-up and repeat HAS-BLED in 1-3 months.
Reference Ranges
| Population | Normal Range | Notes |
|---|---|---|
| AF patients on anticoagulation (ESC guidelines) | 0-9 points | Score ≥3 indicates high risk. Modifiable risk factors (H, A, L, D) should be addressed. |
Dr. Khaled Hassan
Dr. Khaled is a cardiology consultant with 20 years of experience in managing cardiovascular patients.
View medical review board & editorial policy →Example Calculation
A 72-year-old woman with permanent atrial fibrillation (diagnosed 2 years ago), hypertension (treated with amlodipine 10 mg daily, recent office BP 162/94 mmHg), type 2 diabetes, and a prior lacunar stroke 1 year ago with no residual deficit. She has stable coronary artery disease treated with aspirin 81 mg daily. Her most recent laboratory studies show serum creatinine 1.5 mg/dL (estimated CrCl 38 mL/min using Cockcroft-Gault) and normal liver function tests. She reports drinking 2 glasses of wine most evenings (approximately 14 drinks per week). Her INR has been labile during previous warfarin trials with time in therapeutic range of only 52%. HAS-BLED score calculation: Hypertension — uncontrolled SBP >160 mmHg (1 point), Abnormal renal function — CrCl <50 mL/min (1 point), Prior stroke — lacunar stroke (1 point), Age ≥65 (1 point), Drugs — concurrent aspirin use (1 point), Labile INR — TTR <60% (1 point), Alcohol excess — >8 drinks/week (1 point) = Total Score 7 out of 9, indicating high bleeding risk. Rather than withholding anticoagulation for this patient with a clear need for stroke prevention (CHADS-VASc score 4), the clinician addresses modifiable risk factors: (1) intensifies antihypertensive therapy by adding chlorthalidone 12.5 mg daily, (2) discusses discontinuation of aspirin given the lack of clear benefit for stable CAD with concurrent anticoagulation, (3) advises reducing alcohol consumption to <8 drinks per week, and (4) switches from warfarin to apixaban 2.5 mg twice daily (dose-adjusted for age ≥80, weight ≤60 kg, or CrCl <25 mL/min; this patient meets the renal criterion for dose reduction at CrCl 38 mL/min). Close follow-up is scheduled in 4 weeks to reassess blood pressure, renal function, and adherence.
Related Conditions
Related Medications
Common Mistakes
Withholding anticoagulation solely because of a high HAS-BLED score
A high HAS-BLED score should prompt risk factor optimization and closer monitoring, not automatic withholding of anticoagulation. The net clinical benefit of OAC in AF is nearly always positive. Focus on correctable risk factors: uncontrolled BP, NSAID use, alcohol excess, labile INR.
Counting both renal AND liver abnormalities as a single combined variable
Renal function and liver function are two separate variables. If both are abnormal, each contributes 1 independent point toward the total score. Do not combine them into a single "organ dysfunction" score.
Applying the HAS-BLED hypertension definition (SBP >160) interchangeably with CHADS-VASc hypertension (any history of treated HTN)
HAS-BLED requires uncontrolled SBP >160 mmHg at the time of assessment. Well-controlled hypertension (SBP <160 on medications) does not score a point for HAS-BLED, whereas CHADS-VASc scores 1 point for any history of hypertension regardless of control.
Not recognizing that labile INR only applies to warfarin users
The Labile INR (L) variable specifically refers to unstable warfarin anticoagulation with TTR <60%. This variable does not apply to patients on DOACs. In DOAC-treated patients, the maximum HAS-BLED score is 8 (excluding the L variable).
Assuming the score is only validated for warfarin-treated patients
HAS-BLED was originally validated in warfarin-treated patients but has since been extensively validated in DOAC-treated patients as well. The same bleeding risk factors apply regardless of which anticoagulant is used, though absolute bleeding rates are lower with DOACs.
Frequently Asked Questions
Should anticoagulation be withheld if the HAS-BLED score is ≥3?
What is the maximum score on HAS-BLED?
Does HAS-BLED apply to patients on DOACs as well as warfarin?
How often should HAS-BLED be reassessed?
Can HAS-BLED be used in non-AF patients on anticoagulation?
Does the HAS-BLED score predict intracranial hemorrhage specifically?
References
- Pisters R, Lane DA, Nieuwlaat R, et al. 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. PubMed
- Lip GY, Frison L, Halperin JL, et al. Comparative validation of a novel risk score for predicting bleeding risk in anticoagulated patients with atrial fibrillation. J Am Coll Cardiol. 2011;57(2):173-180. PubMed
- Hindricks G, Potpara T, Dagres N, et al. 2020 ESC Guidelines for the diagnosis and management of atrial fibrillation. Eur Heart J. 2021;42(5):373-498. PubMed
- January CT, Wann LS, Calkins H, et al. 2024 AHA/ACC Guideline for the Management of Patients With Atrial Fibrillation. Circulation. 2024;149(1):e1-e156. PubMed
- NICE Guideline NG196. Atrial fibrillation: diagnosis and management. 2021.
- Ruff CT, Giugliano RP, Braunwald E, et al. Comparison of the efficacy and safety of new oral anticoagulants with warfarin in patients with atrial fibrillation: a meta-analysis. Lancet. 2014;383(9921):955-962. PubMed
- Yao X, Abraham NS, Sangaralingham LR, et al. Effect of adherence to oral anticoagulants on risk of stroke and major bleeding among patients with atrial fibrillation. J Am Heart Assoc. 2016;5(2):e003074. PubMed