HAS-BLED Score Calculator
Estimate bleeding risk in people on (or being considered for) anticoagulation, and see which contributing factors can actually be modified.
—
—
What the Score Measures
HAS-BLED estimates the risk of major bleeding in people with atrial fibrillation who are taking — or being considered for — oral anticoagulation. Each letter contributes one point, except renal and liver function, drugs and alcohol, which each count separately, giving a maximum of nine:
- H — uncontrolled hypertension (systolic above 160 mmHg): 1
- A — abnormal renal function: 1 · abnormal liver function: 1
- S — prior stroke: 1
- B — bleeding history or predisposition: 1
- L — labile INR (time in therapeutic range below 60 %): 1
- E — age over 65: 1
- D — drugs that predispose to bleeding: 1 · alcohol (eight or more drinks per week): 1
How the Result Is Used
The conventional threshold is three or more points, which marks a high bleeding risk and prompts closer follow-up rather than a change in anticoagulation. This is the point most often misunderstood: guidelines are explicit that a high HAS-BLED score is not a reason to withhold anticoagulation, because the stroke risk these patients carry is usually higher still. The score is meant to start a conversation about modifiable factors, and to justify more frequent review.
Modifiable Versus Fixed Factors
Several components can be changed, which is where the score earns its place:
- Blood pressure control — bringing systolic pressure below 160 mmHg removes a point
- Concurrent antiplatelets or NSAIDs — stopping unnecessary ones removes a point and often more real risk than the score implies
- Alcohol intake — reduction removes a point
- Time in therapeutic range — better INR control removes the labile-INR point for people on warfarin
- Renal and liver function, age, prior stroke and bleeding history are fixed; they inform monitoring intensity instead
Limitations
- Discrimination is modest: the score separates groups, not individuals, and many patients with a high score never bleed.
- It was derived and validated largely in warfarin-treated cohorts; its performance with direct oral anticoagulants is less well established.
- It does not capture all relevant factors, including concomitant use of SSRIs, thrombocytopenia, or the specific anatomical site of a previous bleed.
- It should be interpreted alongside a stroke-risk score such as CHA₂DS₂-VASc, never instead of one.
Frequently Asked Questions
Does a high HAS-BLED score mean I should stop my blood thinner?
No. Guidelines state that a high score is not a reason to withhold anticoagulation, because untreated atrial fibrillation usually carries a greater stroke risk. It should prompt a review of modifiable factors and closer follow-up.
How does it relate to CHA₂DS₂-VASc?
They answer different questions. CHA₂DS₂-VASc estimates stroke risk and drives the decision to anticoagulate; HAS-BLED estimates bleeding risk and shapes monitoring. Both are usually documented together.
Is labile INR relevant if I take a DOAC?
No — that point applies to warfarin, where INR control is measured. People on direct oral anticoagulants do not accrue it.