Appendix Table 5.
C-statistics from Studies Comparing Scores of Interest for Prediction of Major Bleeding Riska
| Study | BRI | HEMORR2HAGES | HAS-BLED | ATRIA | ABC |
|---|---|---|---|---|---|
| Patients on warfarin | |||||
| Apostolakis, 201282d | – | 0.60 (95% CI 0.51 to 0.69) |
0.65 (95% CI 0.56 to 0.73) |
0.61 (95% CI 0.51 to 0.70) |
– |
| Barnes, 201486 | – | 0.66 (95% CI 0.61-0.74) |
0.69 (95% CI 0.63-0.75) |
0.67 (95% CI 0.61 to 0.74) |
– |
| Fang, 201176d,f | Categorical: 0.59 (95% CI 0.58 to 0.61) Continuous: 0.68 (95% CI 0.65 to 0.70) |
Categorical: 0.67 (95% CI 0.65 to 0.70) Continuous: 0.71 (95% CI 0.69 to 0.73) |
– | Categorical: 0.69 (95% CI 0.66 to 0.71) Continuous: 0.74 (95% CI 0.72 to 0.76) |
– |
| Friberg, 201218d | – | 0.63 (95% CI 0.61 to 0.64) |
0.61 (95% CI 0.59 to 0.62) |
– | – |
| Gage, 200679b,c | 0.65 (SE 0.03) | 0.67 (SE 0.04) | – | – | – |
| Hijazi, 201693 | – | – | ARISTOTLE: 0.61 (0.58 to 0.63) RE-LY: 0.60 (0.56 to 0.64) |
– | ARISTOTLE: 0.68 (0.65 to 0.70) RE-LY: 0.65 (0.61 to 0.70) |
| Jaspers Focks, 201688 | – | Major bleeding = 0.57 (95% CI 0.50 to 0.63) Clinically relevant bleeding = 0.53 (95% CI 0.50 to 0.57) Any bleeding = 0.53 (95% CI 0.50 to 0.57) |
Major Bleeding = 0.57 (95% CI 0.50 to 0.63) Clinically relevant bleeding = 0.50 (95% CI 0.47 to 0.54) Any bleeding = 0.51 (95% CI 0.47 to 0.54) |
Major Bleeding = 0.58 (95% CI 0.51 to 0.64) Clinically relevant bleeding = 0.52 (95% CI 0.49 to 0.56) Any bleeding= 0.53 (95% CI 0.50 to 0.57) |
– |
| Lip, 201177d | 0.56 (95% CI 0.51 to 0.60) |
0.61 (95% CI 0.56 to 0.65) |
0.66 (95% CI 0.61 to 0.70) |
– | – |
| Lip, 201285g | Categorical: 0.56 (95% CI 0.53 to 0.59) Continuous: 0.60 (95% CI 0.56 to 0.63) |
Categorical: 0.53 (95% CI 0.50 to 0.57) Continuous: 0.59 (95% CI 0.56 to 0.62) |
Categorical: 0.58 (95% CI 0.55 to 0.61) Continuous: 0.61 (95% CI 0.58 to 0.65) |
Categorical: 0.55 (95% CI 0.52 to 0.59) Continuous: 0.60 (95% CI 0.56 to 0.63) |
– |
| Lip, 201794 | – | – | 0.58 (95% CI 0.57-0.59) | 0.59 (95% CI 0.57-0.60) |
– |
| Olesen, 201181d | – | Categorical: 0.78 (95% CI 0.75 to 0.82) Continuous: 0.77 (95% CI 0.73 to 0.81) |
Categorical: 0.80 (95% CI 0.76 to 0.83) Continuous: 0.80 (95% CI 0.76 to 0.83) |
– | – |
| Pisters, 20109d,e | – | 0.64 (95% CI 0.53 to 0.75) |
0.69 (95% CI 0.59 to 0.80) |
– | – |
| Roldan, 201284h | – | – | Categorical: 0.68 (95% CI 0.65 to 0.71) Continuous: 0.71 (95% CI 0.68 to 0.74) |
Categorical: 0.59 (95% CI 0.55 to 0.62) Continuous: 0.68 (95% CI 0.65 to 0.71) |
– |
| Senoo, 201691 | – | – | 0.65 (95% CI 0.56 to 0.73) |
0.61 (95% CI 0.51 to 0.70) |
– |
| Patients on aspirin alone | |||||
| Friberg, 201218e | – | 0.60 (95% CI 0.59 to 0.61) |
0.59 (95% CI 0.58 to 0.60) |
– | – |
| Gage, 200679b,c | 0.69 (SE 0.05) | 0.72 (SE 0.05)b | – | – | – |
| Pisters, 20109d,e | – | 0.83 (95% CI 0.68 to 0.98) |
0.91 (95% CI 0.83 to 1.00) |
– | – |
| Patients off antithrombotic therapy | |||||
| Friberg, 201218e | – | 0.69 (95% CI 0.67 to 0.70) |
0.66 (95% CI 0.65 to 0.68) |
– | – |
| Gage, 200679b,c | 0.65 (SE 0.03) | 0.66 (SE 0.04) | – | – | – |
| Lip, 201177d | 0.50 (95% CI 0.44 to 0.57) |
0.62 (95% CI 0.52 to 0.72) |
0.66 (95% CI 0.55 to 0.74) |
– | – |
| Lip, 201285f | Categorical: 0.58 (95% CI 0.54 to 0.62) Continuous: 0.60 (95% CI 0.56 to 0.64) |
Categorical: 0.55 (95% CI 0.50 to 0.59) Continuous: 0.59 (95% CI 0.54 to 0.63) |
Categorical: 0.60 (95% CI 0.54 to 0.64) Continuous: 0.60 (95% CI 0.56 to 0.64) |
Categorical: 0.47 (95% CI 0.42 to 0.51) Continuous: 0.59 (95% CI 0.55 to 0.64) |
– |
| Olesen, 201181d | – | Categorical: 0.77 (95% CI 0.74 to 0.80) Continuous: 0.79 (95% CI 0.73 to 0.79) |
Categorical: 0.82 (95% CI 0.79 to 0.84) Continuous: 0.81 (95% CI 0.78 to 0.83) |
– | – |
| Pisters, 20109d,e | – | 0.81 (95% CI 0.00 to 1.00) |
0.85 (95% CI 0.00 to 1.00) |
– | – |
C-statistics given are for categorical risk scores unless otherwise noted.
Derivation study for HEMORR2HAGES.
P-value for 2-way between-score comparison not provided.
P-value for between-score comparison not provided.
Derivation study for HAS-BLED.
Derivation study for ATRIA.
P-values for all between-score comparisons >0.05 (not specified as <0.05 in source article).
P=0.035 for comparison of between-score categorical c-statistics and p=0.356 for comparison of between-score continuous c-statistics.
Abbreviations: ABC=Age, biomarkers, clinical history; ATRIA=Anticoagulation and Risk Factors in Atrial Fibrillation; BRI=Bleeding Risk Index; CI=confidence interval; HAS-BLED=Hypertension, Abnormal renal/liver function, Stroke, Bleeding history or predisposition, Labile international normalized ratio, Elderly (> 65 years), Drugs/alcohol concomitantly; HEMORR2HAGES=Hepatic or renal disease, Ethanol abuse, Malignancy, Older (age >75 years), Reduced platelet count or function, Re-bleeding risk (2 points), Hypertension (uncontrolled), Anemia, Genetic factors, Excessive fall risk, Stroke; SE=standard error