Skip to main content

Table 4 Discrimination performance of the existing models externally validated across the OHDSI datasets

From: Feasibility and evaluation of a large-scale external validation approach for patient-level prediction in an international data network: validation of models predicting stroke in female patients newly diagnosed with atrial fibrillation

 

Database AUROC (95% CIs)

Target Populationa

Model

CCAE

MDCD

MDCR

Optum claims

Optum EHR

CUMC

AUSOM

STRIDE

T1: Females aged 65+ with atrial fibrillation no prior stroke or anticoagulants

ATRIA

0.57 (0.55–0.58)

0.63 (0.62–0.64)

0.61

0.62

0.64 (0.61–0.68)

0.60 (0.33–0.87)

0.49 (0.40–0.58)

CHADS2

0.54 (0.53–0.56)

0.60 (0.59–0.61)

0.59

0.60

0.60 (0.57–0.64)

0.51 (0.27–0.75)

0.48 (0.39–0.57)

CHA2DS2VASc

0.55 (0.53–0.57)

0.60 (0.59–0.61)

0.59

0.62

0.61 (0.58–0.65)

0.53 (0.32–0.74)

0.52 (0.42–0.62)

Framingham

0.56 (0.54–0.57)

0.62 (0.61–0.63)

0.59

0.61

0.63 (0.60–0.66)

0.58 (0.33–0.83)

0.61 (0.52–0.70)

Q-Stroke

0.53 (0.52–0.55)

0.56 (0.55–0.57)

0.55

0.56

0.55 (0.51–0.59)

0.56 (0.29–0.84)

0.50 (0.41–0.59)

T2: Females with atrial fibrillation no prior stroke or anticoagulants

ATRIA

0.62 (0.60–0.64)

0.58 (0.56–0.59)

0.65

0.65

0.66 (0.62–0.69)

0.73 (0.58–0.89)

0.52 (0.44–0.60)

CHADS2

0.61 (0.59–0.62)

0.56 (0.55–0.57)

0.62

0.63

0.63 (0.60–0.66)

0.63 (0.43–0.83)

0.50 (0.42–0.57)

CHA2DS2VASc

0.63 (0.61–0.65)

0.58 (0.56–0.59)

0.64

0.65

0.64 (0.61–0.67)

0.73 (0.60–0.85)

0.55 (0.47–0.62)

Framingham

0.62 (0.60–0.64)

0.57 (0.56–0.59)

0.64

0.65

0.65 (0.62–0.68)

0.70 (0.53–0.86)

0.61 (0.53–0.69)

Q-Stroke

0.61 (0.59–0.63)

0.54 (0.53–0.56)

0.57

0.58

0.56 (0.53–0.60)

0.63 (0.39–0.88)

0.51 (0.43–0.59)

Sensitivity T1: Females aged 65+ with atrial fibrillation no prior stroke or anticoagulants (no anticoagulants during 1 year time-at-risk)

ATRIA

0.56 (0.55–0.58)

0.63 (0.62–0.64)

0.61 (0.61–0.62)

0.63 (0.61–0.64)

0.65 (0.62–0.69)

0.69 (0.43–0.95)

0.55 (0.47–0.62)

CHADS2

0.54 (0.53–0.56)

0.61 (0.60–0.62)

0.59 (0.58–0.60)

0.61 (0.59–0.62)

0.62 (0.58–0.65)

0.61 (0.36–0.85)

0.51 (0.38–0.63)

CHA2DS2VASc

0.55 (0.54–0.57)

0.61 (0.60–0.62)

0.59 (0.58–0.60)

0.63 (0.61–0.64)

0.63 (0.59–0.66)

0.64 (0.45–0.83)

0.55 (0.42–0.67)

Framingham

0.55 (0.54–0.57)

0.62 (0.61–0.63)

0.59 (0.59–0.60)

0.62 (0.61–0.63)

0.64 (0.61–0.68)

0.68 (0.44–0.93)

0.64 (0.53–0.74)

Q-Stroke

0.53 (0.52–0.55)

0.57 (0.55–0.58)

0.55 (0.54–0.56)

0.57 (0.55–0.58)

0.56 (0.52–0.60)

0.61 (0.30–0.92)

0.47 (0.35–0.58)

Sensitivity T2: Females with atrial fibrillation no prior stroke or anticoagulants (no anticoagulants during 1-year time-at-risk)

ATRIA

0.63 (0.61–0.66)

0.58 (0.56–0.59)

0.67

0.67

0.67 (0.64–0.70)

0.79 (0.63–0.94)

0.53 (0.43–0.63)

CHADS2

0.62 (0.60–0.65)

0.56 (0.55–0.58)

0.64

0.65

0.64 (0.61–0.68)

0.72 (0.53–0.91)

0.51 (0.41–0.62)

CHA2DS2VASc

0.65 (0.62–0.67)

0.58 (0.56–0.59)

0.65

0.67

0.66 (0.63–0.69)

0.81 (0.71–0.90)

0.55 (0.44–0.65)

Framingham

0.64 (0.61–0.66)

0.57 (0.56–0.59)

0.65

0.66

0.66 (0.63–0.69)

0.76 (0.59–0.93)

0.62 (0.51–0.72)

Q-Stroke

0.62 (0.60–0.64)

0.55 (0.53–0.56)

0.58

0.6

0.57 (0.53–0.61)

0.68 (0.42–0.94)

0.47 (0.36–0.57)

  1. Discrimination performance of the existing models across the datasets. The AUROC 95% confidence intervals were only calculated when the outcome count was less than 1000. a See section ‘Participants’ for full inclusion/exclusion criteria