Skip to main content

Table 1 Characteristics of the study population

From: Validation of a deep-learning-based retinal biomarker (Reti-CVD) in the prediction of cardiovascular disease: data from UK Biobank

 

Reti-CVD score

  

Characteristics

Low risk

Moderate risk

High risk

No. of participants

19,304

26,543

2413

Cardiovascular outcomes

 Nonfatal and fatal CVD events (QRISK)

545 (2.8%)

1900 (7.2%)

321 (13.3%)

QRISK3

 QRISK3 score, mean (SD)

3.0 (2.5)

7.0 (4.1)

10.6 (4.6)

 QRISK3 score

   

 ≥ 0 to < 5

16,282 (84.3%)

9646 (36.3%)

208 (8.6%)

 ≥ 5 to < 10

2572 (13.3%)

11,393 (42.9%)

967 (40.1%)

 ≥ 10 to < 15

384 (2.0%)

4408 (16.6%)

876 (36.3%)

 ≥ 15 to < 20

55 (0.3%)

880 (3.3%)

272 (11.3%)

 > 20

11 (0.1%)

216 (0.8%)

90 (3.7%)

Clinical biomarkers

 Age, mean (SD)

50.8 (7.0)

59.8 (6.7)

64.4 (4.5)

 Gender

   

  Female, n (%)

12,702 (65.8%)

13,622 (51.3%)

741 (30.7%)

  Male, n (%)

6602 (34.2%)

12,921 (48.7%)

1672 (69.3%)

Other factors

 Antihypertensive medication, n (%)

1637 (8.5%)

5642 (21.3%)

846 (35.1%)

 Stage 1 hypertension, n (%)

4199 (21.8%)

7178 (27.0%)

589 (24.4%)

 Pre-diabetes and diabetes, n (%)

269 (1.4%)

677 (2.6%)

86 (3.6%)

 Statin, n (%)

694 (3.6%)

1923 (7.2%)

170 (7.0%)

 Current smoker, n (%)

6451 (33.4%)

11,221 (42.3%)

1219 (50.5%)

  1. Data are presented as n, n (% of participants), mean (standard deviation [SD]), or median (interquartile range [IQR]). CVD cardiovascular disease. Reti-CVD deep-learning-based retinal CVD biomarker