IMPROVE Bleeding Risk Score
Evaluates bleeding risk at hospital admission.
Creator
Dr. Hervé Décousus, MDInstructions
Utilize upon admission to 14 days after admission or when initiating anticoagulation therapy, primarily for venous thromboembolism (VTE) prophylaxis, in hospitalized patients.
Risk factors at admission
IMPROVE Bleeding Risk Score
The IMPROVE Bleeding Risk Score estimates the risk of in-hospital bleeding in acutely ill, hospitalized medical patients. Utilize upon admission to 14 days after admission or when initiating anticoagulation therapy, primarily for venous thromboembolism (VTE) prophylaxis, in hospitalized patients.
Development
The score was derived by Decousus, Tapson, and the IMPROVE Investigators (Chest, 2011) using data from the International Medical Prevention Registry on Venous Thromboembolism (IMPROVE), a multinational observational study that enrolled 15,156 acutely ill medical patients across 12 countries. The strongest independent predictors of in-hospital bleeding identified were active gastroduodenal ulcer (OR 4.15), prior bleeding (OR 3.64), and low platelet count (OR 3.37), with additional contributions from age, hepatic or renal failure, ICU/CCU stay, central venous catheter, rheumatic disease, cancer, and male sex.
The 11 Components
The IMPROVE Bleeding Risk Score is calculated by the addition of the selected points:
Variable | Points | |
Age, years | <40 | 0 |
40-84 | 1.5 | |
≥85 | 3.5 | |
Gender | Female | 0 |
Male | 1 | |
Renal function (GFR), mL/min/m2 | ≥60 | 0 |
30-59 | 1 | |
<30 | 2.5 | |
Current cancer (evidence of active malignancy within the last 6 months) | No | 0 |
Yes | 2 | |
Rheumatic disease | No | 0 |
Yes | 2 | |
Central venous catheter | No | 0 |
Yes | 2 | |
ICU/CCU | No | 0 |
Yes | 2.5 | |
Evidence of hepatic failure (INR >1.5) | No | 0 |
Yes | 2.5 | |
Platelet count, cells/L | ≥50 x 109 | 0 |
<50 x 109 | 4 | |
Bleeding in 3 months before admission | No | 0 |
Yes | 4 | |
Active gastroduodenal ulcer | No | 0 |
Yes | 4.5 |
Facts & Figures
IMPROVE Score | Risk of Bleeding |
<7 | No increased risk of bleeding
|
≥7 | Increased risk of bleeding
|
*Major bleeding: A bleeding event contributing to death, bleeding associated with a ≥2 g/dL fall in Hg, or leading to transfusion of ≥2 units of packed red blood cells, or within a critical organ (including intracranial, retroperitoneal, intraocular, adrenal gland, spinal, or pericardial bleeding).
*Non-major but clinically relevant bleeding: Overt gastrointestinal bleeding (except for insignificant hemorrhoidal bleeding), gross hematuria (macroscopic and lasting longer than 24 hours), substantial epistaxis that requires intervention and is recurrent and/or lasts ≥5 minutes, extensive hematoma or bruising (>5 cm in diameter), intra-articular bleeding (documented by aspiration), menorrhagia or metrorrhagia (increased quantity or duration), other bleeding important enough to be recorded on the hospital chart.
Management Considerations
Consider using results in tandem with the IMPROVE or IMPROVEDD Risk Score for VTE when assessing the use of thromboprophylaxis during hospitalization.
Consider addressing modifiable risk factors for bleeding.
Risk and benefits of anti-coagulation should be considered in patients, to include shared decsion-making and informed consent, prior to initiating therapy.
Clinical judgement and best practice guidelines supersedes adjunct standalone decision tools.
Important Limitations
Developed specifically for acutely ill medical inpatients, not surgical patients — applying it outside that population (as some validation studies have tested) is an extrapolation beyond its original design
Not all 11 risk factors have been independently re-validated in every subsequent study — the score functions better as a structured framework for decision-making than as a precise individual probability estimate
Bleeding risk should be reassessed throughout the hospital stay rather than treated as fixed at admission — a patient who was high-risk at admission (e.g., recovering from a recent GI bleed) may become low-risk after a period of clinical stability, at which point pharmacologic prophylaxis can be reconsidered
All questions & possible results
Risk factors at admission
Age
Select one option:
- <40 years
- 40-84 years
- ≥85 years
Gender
Select one option:
- Female
- Male
Renal function (GFR, mL/min/m²)
Select one option:
- ≥60
- 30-59
- <30
Current cancer
Evidence of active malignancy within the last 6 months
Select one option:
- No
- Yes
Rheumatic disease
Select one option:
- No
- Yes
Central venous catheter
Select one option:
- No
- Yes
ICU or CCU
Select one option:
- No
- Yes
Evidence of hepatic failure (INR >1.5)
Select one option:
- No
- Yes
Platelet count (cells/L)
Select one option:
- ≥50x10⁹
- 50x10⁹
Bleeding in 3 months before admission
Major bleeding: A bleeding event contributing to death, bleeding associated with a ≥2 g/dL fall in Hg, or leading to transfusion of ≥2 units of packed red blood cells, or within a critical organ (including intracranial, retroperitoneal, intraocular, adrenal gland, spinal, or pericardial bleeding).
Non-major but clinically relevant bleeding: Overt gastrointestinal bleeding (except for insignificant hemorrhoidal bleeding), gross hematuria (macroscopic and lasting longer than 24 hours), substantial epistaxis that requires intervention and is recurrent and/or lasts ≥5 minutes, extensive hematoma or bruising (>5 cm in diameter), intra-articular bleeding (documented by aspiration), menorrhagia or metrorrhagia (increased quantity or duration), other bleeding important enough to be recorded on the hospital chart.
Select one option:
- No
- Yes
Active gastroduodenal ulcer
Select one option:
- No
- Yes
Possible results
IMPROVE Bleeding Risk Score: No increased risk of bleeding (<7pts)
A score <7 points identifies patients at low risk of in-hospital major or clinically important bleeding.* Anticoagulation prophylaxis may be considered according to the patient's thrombotic risk and local protocol.
Consider using results in tandem with the IMPROVE or IMPROVEDD Risk Score for VTE when assessing the use of thromboprophylaxis during hospitalization.
Consider addressing modifiable risk factors for bleeding.
Risk and benefits of anti-coagulation should be considered in patients, to include shared decsion-making and informed consent, prior to initiating therapy.
Clinical judgement and best practice guidelines supersedes adjunct standalone decision tools.
*Major bleeding: A bleeding event contributing to death, bleeding associated with a ≥2 g/dL fall in Hg, or leading to transfusion of ≥2 units of packed red blood cells, or within a critical organ (including intracranial, retroperitoneal, intraocular, adrenal gland, spinal, or pericardial bleeding).
*Non-major but clinically relevant bleeding: Overt gastrointestinal bleeding (except for insignificant hemorrhoidal bleeding), gross hematuria (macroscopic and lasting longer than 24 hours), substantial epistaxis that requires intervention and is recurrent and/or lasts ≥5 minutes, extensive hematoma or bruising (>5 cm in diameter), intra-articular bleeding (documented by aspiration), menorrhagia or metrorrhagia (increased quantity or duration), other bleeding important enough to be recorded on the hospital chart.
IMPROVE Bleeding Risk Score: Increased risk of bleeding (≥7pts)
A score of ≥7 identifies patients at increased risk of in-hospital major or clinically important bleeding.* Avoid anticoagulants where possible and consider non-pharmacologic interventions. Weigh bleeding risk against thrombotic risk before anticoagulation.
Consider using results in tandem with the IMPROVE or IMPROVEDD Risk Score for VTE when assessing the use of thromboprophylaxis during hospitalization.
Consider addressing modifiable risk factors for bleeding.
Risk and benefits of anti-coagulation should be considered in patients, to include shared decsion-making and informed consent, prior to initiating therapy.
Clinical judgement and best practice guidelines supersedes adjunct standalone decision tools.
*Major bleeding: A bleeding event contributing to death, bleeding associated with a ≥2 g/dL fall in Hg, or leading to transfusion of ≥2 units of packed red blood cells, or within a critical organ (including intracranial, retroperitoneal, intraocular, adrenal gland, spinal, or pericardial bleeding).
*Non-major but clinically relevant bleeding: Overt gastrointestinal bleeding (except for insignificant hemorrhoidal bleeding), gross hematuria (macroscopic and lasting longer than 24 hours), substantial epistaxis that requires intervention and is recurrent and/or lasts ≥5 minutes, extensive hematoma or bruising (>5 cm in diameter), intra-articular bleeding (documented by aspiration), menorrhagia or metrorrhagia (increased quantity or duration), other bleeding important enough to be recorded on the hospital chart.
Citation
Decousus H, Tapson VF, Bergmann JF, Chong BH, Froehlich JB, Kakkar AK, Merli GJ, Monreal M, Nakamura M, Pavanello R, Pini M, Piovella F, Spencer FA, Spyropoulos AC, Turpie AG, Zotz RB, Fitzgerald G, Anderson FA; IMPROVE Investigators. Factors at admission associated with bleeding risk in medical patients: findings from the IMPROVE investigators. Chest. 2011 Jan;139(1):69-79.
Literature
Decousus H, Tapson VF, Bergmann JF, Chong BH, Froehlich JB, Kakkar AK, Merli GJ, Monreal M, Nakamura M, Pavanello R, Pini M, Piovella F, Spencer FA, Spyropoulos AC, Turpie AG, Zotz RB, Fitzgerald G, Anderson FA; IMPROVE Investigators. Factors at admission associated with bleeding risk in medical patients: findings from the IMPROVE investigators. Chest. 2011 Jan;139(1):69-79.
https://pubmed.ncbi.nlm.nih.gov/20453069/Rosenberg DJ, Press A, Fishbein J, Lesser M, McCullagh L, McGinn T, Spyropoulos AC. External validation of the IMPROVE Bleeding Risk Assessment Model in medical patients. Thromb Haemost. 2016 Aug 30;116(3):530-6.
https://pubmed.ncbi.nlm.nih.gov/27307054/Hostler DC, Marx ES, Moores LK, Petteys SK, Hostler JM, Mitchell JD, Holley PR, Collen JF, Foster BE, Holley AB. Validation of the International Medical Prevention Registry on Venous Thromboembolism Bleeding Risk Score. Chest. 2016 Feb;149(2):372-9.
https://pubmed.ncbi.nlm.nih.gov/26867833/
- Creator
- Provided by
EVAL Foundation
- Contributor · Reviewer
Jennifer Glen, DNP, FNP-BCEVAL Health, Chief Medical Officer EVAL Foundation
Related evaluations
VTE-BLEED Score
Assess bleeding risk during stable-phase anticoagulation.
Dr. Frederikus A. Klok, MD, PhDIMPEDE-VTE
Predicts risk of venous thromboembolism in multiple myeloma
Dr. Kristen M. Sanfilippo, MD, MPHSATRIA Bleeding Risk Score
Determines bleeding risk for patients being considered for warfarin anticoagulation
Dr. Daniel Singer, MDISTH-SSC Bleeding Assessment Tool (ISTH-SCC BAT)
Evaluates bleeding symptoms in patients with inherited bleeding disorders.
Dr. Francesco Rodeghiero, M.D.Aspirin Deprescribing Guidelines
Patient-specific recommendations to change or discontinue aspirin that may lead to fewer adverse events
Padua Prediction Score for Risk of VTE
Determines anticoagulation need in hospitalized patients by risk of VTE.
Dr. Sofia Barbar, MDVenous Thromboembolism Evaluation & Management in the Emergency Department
App for the evaluation, diagnosis, & management of ED patients with VTE to stratify for In- vs. Out-patient treatment.
OSU Center for Health SciencesRIETE Predictive Score for Bleeding
Determine the risk of major bleeding during anticoagulant therapy.
Dr. Nuria Ruíz-Giménez Arrieta, MDPulmonary Embolism Severity Index (PESI)
Predicts 30-day outcome of patients with pulmonary embolism using 11 clinical criteria.
Dr. Drahomir Aujesky, MD, MSSimplified PESI (Pulmonary Embolism Severity Index: sPESI)
Estimates the 30-day mortality risk following acute pulmonary embolism.
OSU Center for Health SciencesIMPROVE Risk Score for VTE
Predicts 3-month risk of VTE in hospitalized patients.
Dr. Alex C. Spyropoulos, MD, FACP, FCCP, FRCPCIMPROVEDD Risk Score for VTE
Predicts risk of VTE in hospitalized patients, adding D-dimer to IMPROVE Risk Score.
Dr. C. Michael Gibson, MS, MDAlso from EVAL Foundation
Alcohol Use Disorders Identification Test - C (AUDIT-C)
A 3-item tool to screen for unhealthy alcohol use in patients who may not be alcohol-dependent.
Dr. Katharine Bradley, MD, MPHAlcohol Use Disorders Identification Test (AUDIT)
A 10-item tool to screen for hazardous alcohol use, harmful use, and alcohol dependence.
World Health Organization Collaborative ProjectAmerican Shoulder and Elbow Surgeons (ASES) Shoulder Score
15-item assessment of shoulder pain and functional limitations in activities of daily living.
Dr. Robin R. Richards, MD, FRCSC, FAAOS, FAOABehavioral Activity Rating Scale (BARS)
A 1-item rating scale to quantify a patient's level of agitation in acute care or psychiatric settings.
Drug Abuse Screening Test (DAST-10)
A patient self-report questionnaire to screen for problematic drug use (excluding alcohol/tobacco).
Dr. Harvey A. Skinner, Ph.D.Foot and Ankle Ability Measure (FAAM)
Assesses functional limitations and disability related to foot and ankle conditions in daily activities and sports.
Dr. Jesse Van Swearingen, PhD, PT, FAPTAFoot and Ankle Disability Index (FADI)
Assess functional limitations of those with a variety of foot and ankle conditions
Dr. RobRoy Martin, PhD, PT, CSCSGeneral Anxiety Disorder (GAD-7)
Measure severity of anxiety
Dr. Robert L. Spitzer, MD