induwara.lk
induwara.lkClinical · Cardiology

HAS-BLED Score Calculator — Bleeding Risk in Atrial Fibrillation

Score a patient's 1-year major-bleeding risk on anticoagulation. Tick the nine HAS-BLED criteria for the 0–9 score, the estimated bleeding rate, the low/high risk band, and which factors are modifiable. Free, no signup, sources cited below.

By Induwara AshinsanaUpdated Jul 12, 2026
HAS-BLED score/ 9
Cross-checked
Bleeding-risk criteria — tap to toggle (1 point each)
Scenarios
Total score
3/9
HAS-BLED
Major bleeds / 100 patient-yrs
3.74%
1-year risk (Pisters 2010)
Risk band
High risk
Anticoagulate with caution + review

High bleeding risk — anticoagulate with caution and review regularly. A HAS-BLED score of 3 or more flags a high 1-year major-bleeding risk. The 2020 ESC guideline uses this to prompt correction of modifiable factors and earlier, more-frequent review — not to withhold anticoagulation.

1 modifiable factor present

The 2020 ESC guideline says a HAS-BLED score should trigger action on reversible factors rather than stopping anticoagulation. Reconsider: Drugs predisposing to bleeding.

Where the points came from

LetterCriterionPoints
SStroke history+1
EElderly (age > 65)+1
DDrugs predisposing to bleeding+1
Total HAS-BLED3

Score-to-bleeding-rate reference

ScoreMajor bleeds / 100 patient-years
01.13%
11.02%
21.88%
3You3.74%
48.70%
512.50%
6–9Very high (no stable rate)

Criteria & 1-year major-bleeding rates: Pisters 2010 (Euro Heart Survey derivation cohort). High-risk threshold (≥ 3) & modifiable-factor framing: 2020 ESC Atrial Fibrillation Guidelines. Educational decision-support only — a high score is not a contraindication to anticoagulation, and this tool does not replace clinical judgement.

Educational decision-support only. This calculator does not diagnose, prescribe, or replace clinical assessment. A high HAS-BLED score is not a contraindication to anticoagulation — the guideline uses it to flag reversible risks to correct. Always decide with the treating team and full clinical context.

How it works

HAS-BLED is a simple additive point score published by Pisters and colleagues in Chest in 2010, derived from 3,978 patients with atrial fibrillation in the Euro Heart Survey. It estimates the risk of major bleeding within one yearfor a patient taking oral anticoagulation. “Major bleeding” in the derivation was defined as intracranial bleeding, a hospitalisation for bleeding, a haemoglobin drop of more than 2 g/dL, and/or bleeding needing a transfusion.

The mnemonic stands for Hypertension, Abnormal renal/liver function, Stroke, Bleeding, Labile INR, Elderly, and Drugs/alcohol. Although there are seven letters, the two “A” items (renal and liver) and the two “D” items (drugs and alcohol) each contribute a separate point, giving nine one-point criteria and a maximum score of 9. The calculation is:

  1. Award 1 pointfor each of the nine criteria the patient meets. Each uses a specific threshold — for example, hypertension counts only when uncontrolled with systolic BP above 160 mmHg, and “elderly” means age over 65.
  2. Sum the points. The total ranges from 0 to 9. There is no weighting or further arithmetic — HAS-BLED is a plain count.
  3. Map the total to the 1-year major-bleeding rate from the Pisters 2010 cohort: 1.13% at score 0, rising through 1.02% (1), 1.88% (2), 3.74% (3), 8.70% (4) and 12.50% at score 5. Scores 6–9 had too few patients for a stable rate and are treated as at least as high as score 5.
  4. Read the risk band. Under the 2020 ESC atrial fibrillation guidelines, a score of 3 or more is high bleeding risk; 0–2 is low-to-intermediate.
  5. Act on modifiable factors. Four criteria are potentially reversible — uncontrolled hypertension, labile INR, concomitant antiplatelets/NSAIDs, and harmful alcohol use. The guideline recommends correcting these and reviewing more often, rather than withholding anticoagulation.

Because the score is a pure count, this tool cross-checks it two ways — summing the per-criterion breakdown and independently re-counting the raw inputs — and shows a “cross-checked” badge only when the two derivations agree.

Worked examples

Example 1 — score 3, high risk

72-year-old man, new AF, prior TIA, BP 150/90, on daily aspirin, on a DOAC

  1. Elderly (age > 65) = 1
  2. Stroke history (TIA counts) = 1
  3. Drugs (aspirin, an antiplatelet) = 1
  4. Hypertension: 150/90 is not > 160 systolic → 0
  5. Renal/liver/bleeding/labile-INR/alcohol → 0
  6. Total = 3 → High risk; ~3.74 major bleeds / 100 patient-years
  7. Modifiable: reconsider the aspirin with the cardiology team

Example 2 — score 5, high risk

80-year-old woman on warfarin, TTR ~45%, BP 172/95, on haemodialysis, previous GI bleed

  1. Hypertension (SBP 172 > 160) = 1
  2. Abnormal renal function (dialysis) = 1
  3. Bleeding history (prior GI bleed) = 1
  4. Labile INR (TTR < 60%) = 1
  5. Elderly (age > 65) = 1
  6. Liver/stroke/drugs/alcohol → 0
  7. Total = 5 → High risk; ~12.50 major bleeds / 100 patient-years
  8. Modifiable: control the BP and improve INR stability (or switch to a DOAC)

Example 3 — score 2, intermediate (edge case at the band boundary)

68-year-old man on warfarin with occasional out-of-range INRs, no other risk factors

  1. Elderly (age > 65) = 1
  2. Labile INR (unstable INRs, TTR < 60%) = 1
  3. All other criteria → 0
  4. Total = 2 → Intermediate risk; ~1.88 major bleeds / 100 patient-years
  5. Sits one point below the high-risk threshold of 3 — still below 'high'
  6. Modifiable: improving INR stability could drop this to a score of 1

Frequently asked questions

Sources & references

The criteria, rate table, and high-risk threshold on this page were last cross-checked against the Pisters 2010 paper and the 2020 ESC guidelines on 2026-07-12. The page is reviewed whenever the ESC atrial fibrillation guideline is updated.

Related tools

Rate this tool
Be the first to rate

Comments & feedback

Spotted a bug or want an improvement? Tell us — our team reviews every comment, and good ideas get built. Comments are public and anonymous.

Found a bug, edge case, or want to suggest an improvement?

Email me at [email protected] — most fixes ship within 24 hours.