Table 3 ORs for the association between cardiovascular events and AI-ECG-based aging effects, separated by sex

From: Explainable AI associates ECG aging effects with increased cardiovascular risk in a longitudinal population study

Cardiovascular event

Overestimation (females)

Overestimation (males)

 

OR

95% CI

p-value

OR

95% CI

p-value

AF diagnosed

4.19

(2.41, 7.28)

< 0.001

2.30

(1.62, 3.27)

< 0.001

Will develop AF

2.06

(1.19, 3.54)

0.009

1.50

(0.96, 2.32)

0.073

HF diagnosed

1.62

(1.32, 1.98)

< 0.001

1.18

(0.95, 1.47)

0.142

Will develop HF

1.49

(1.07, 2.08)

0.018

1.33

(0.96, 1.86)

0.090

MI diagnosed

1.55

(1.00, 2.39)

0.050

1.21

(0.96, 1.53)

0.108

Will develop MI

1.70

(0.99, 2.92)

0.053

1.01

(0.66, 1.55)

0.963

Died within study

1.72

(1.39, 2.14)

< 0.001

1.37

(1.17, 1.59)

< 0.001

Cardiovascular event

Underestimation (females)

Underestimation (males)

 

OR

95% CI

p-value

OR

95% CI

p-value

AF diagnosed

0.24

(0.07, 0.80)

0.021

0.22

(0.08, 0.60)

0.003

Will develop AF

0.25

(0.09, 0.72)

0.010

0.52

(0.22, 1.22)

0.133

HF diagnosed

0.78

(0.61, 0.99)

0.041

0.80

(0.57, 1.12)

0.190

Will develop HF

0.84

(0.58, 1.22)

0.363

1.33

(0.96, 1.98)

0.090

MI diagnosed

0.65

(0.38, 1.14)

0.135

0.57

(0.39, 0.84)

0.005

Will develop MI

0.96

(0.53, 1.73)

0.893

1.25

(0.71, 2.17)

0.438

Died within study

0.61

(0.47, 0.79)

< 0.001

0.50

(0.39, 0.64)

< 0.001

  1. The analysis includes conditions such as the diagnosis and future development (diagnosed during follow-up) of AF, HF, and MI, as well as observed death. Results are presented separately for males and females, with adjustments made for chronological age, hypertension, hyperlipidemia, diabetes, and medications that affect ECG signals, including beta blockers, calcium channel blockers, and antiarrhythmics.
Back to article page