Table 1 Association of anthropometric features and cardiovascular risk factors in participants of the UKB with delta age.

From: Deep learning-derived cardiovascular age shares a genetic basis with other cardiac phenotypes

Covariate

Info

(\(N_{total}=36,349\))

Adjust for age, sex

Adjust for all

Effect size

\(P\)-value

Effect size

\(P\)-value

Sex (male)

17607 (48.4%)

−0.56 (−0.71, −0.41)

4.7e-13

−1.15 (−1.31, −0.98)

4.1e-42

Age

64.25 (±7.57)

−0.37 (−0.38, −0.36)

0.0e+00

−0.40 (−0.41, −0.39)

0.0e+00

Education

–

–

4.6e-18

–

2.4e-06

Secondary (ref. level)

14437 (40.1%)

–

–

–

–

Tertiary

19186 (53.3%)

−0.69 (−0.85, −0.53)

1.7e-17

−0.41 (−0.57, −0.25)

9.7e-07

Other

2343 (6.5%)

0.04 (−0.29, 0.36)

0.83

0.01 (−0.33, 0.34)

0.98

History of health problems:

Diabetes

1979 (5.5%)

0.81 (0.47, 1.15)

2.2e-06

−0.22 (−0.58, 0.14)

0.23

Hypertension

8419 (23.2%)

1.85 (1.67, 2.04)

3.7e-88

0.77 (0.56, 0.97)

2.1e-13

Angina

727 (2.0%)

0.88 (0.34, 1.42)

1.5e-03

0.11 (−0.48, 0.70)

0.72

Stroke

366 (1.0%)

1.47 (0.71, 2.23)

1.6e-04

0.99 (0.20, 1.78)

0.014

Heart attack

524 (1.4%)

1.49 (0.85, 2.13)

4.6e-06

1.43 (0.75, 2.12)

4.1e-05

Physiological measurements:

BMI

26.62 (±4.25)

0.24 (0.22, 0.25)

3.6e-149

0.16 (0.14, 0.18)

3.8e-55

MAP

81.11 (±8.89)

0.13 (0.12, 0.14)

9.9e-173

0.10 (0.09, 0.11)

3.0e-80

LDL [mM]

3.58 (±0.82)

0.15 (0.05, 0.24)

2.6e-03

0.03 (−0.07, 0.13)

0.52

Lifestyle:

Smoking

–

–

6.2e-11

–

1.6e-04

Never / rarely smoked (ref. level)

22477 (62.5%)

–

–

–

–

Active smoker

1300 (3.6%)

0.50 (0.09, 0.92)

0.017

0.41 (−0.01, 0.84)

0.056

Smoked in the past

12212 (33.9%)

0.56 (0.40, 0.72)

2.3e-11

0.34 (0.17, 0.51)

7.9e-05

Alcohol at least 3x per week

16405 (45.2%)

0.24 (0.09, 0.39)

2.3e-03

0.33 (0.17, 0.49)

6.3e-05

Days of moderate PA per week

3.72 (±1.87)

−0.16 (−0.20, −0.11)

8.4e-14

−0.020 (−0.071, 0.030)

0.43

Days of vigorous PA per week

1.93 (±1.58)

−0.25 (−0.30, −0.20)

1.1e-24

−0.16 (−0.22, −0.10)

2.0e-07

  1. The “Info” column lists the number of corresponding participants for categorical features (with the percentage of the total population in parentheses) or the mean value for numerical features (with the standard deviation in parentheses). \(P\)-values and effect sizes in the left double-column are adjusted for age and sex (or only sex for the age-row and vice versa). In the right double-column, the adjustment also includes all other parameters listed in the table. In the “Effect size” columns, values in parentheses denote the lower and upper bounds of the 95% confidence interval. \(P\)-values smaller than the Bonferroni-corrected threshold (\(0.05 / 19 = 0.0026\)) are highlighted in bold. BMI, body mass index; MAP, mean arterial pressure; LDL, low-density lipoprotein; PA, physical activity.
Back to article page