Table 2 Risk of death.

From: Deep neural network-estimated electrocardiographic age as a mortality predictor

 

CODE-15% (n = 218,169)

ELSA-Brasil (n = 14,263)

SaMi-Trop (n = 1,631)

 

HR (CI 95%)

p value

HR (CI 95%)

p value

HR (CI 95%)

p value

Adjusted by age and sex

 ECG-age < age−8 years

0.78 (0.74–0.83)

<0.001

0.74 (0.63–0.88)

<0.001

0.89 (0.52–1.54)

0.681

 ECG-age > age+8 years

1.79 (1.69–1.90)

<0.001

1.75 (1.35–2.27)

<0.001

2.42 (1.53–3.83)

<0.001

Adjusted by age, sex, hypertension, diabetes mellitus, and smoking

 ECG-age < age−8 years

0.78 (0.74–0.83)

<0.001

0.82 (0.69–0.98)

0.030

0.90 (0.52–1.55)

0.702

 ECG-age > age+8 years

1.79 (1.68–1.89)

<0.001

1.56 (1.20–2.03)

<0.001

2.48 (1.56–3.94)

<0.001

Adjusted by age, sex, hypertension, diabetes mellitus, smoking, and dyslipidemia

 ECG-age < age−8 years

0.78 (0.74–0.83)

<0.001

0.82 (0.69–0.98)

0.030

Not available

 ECG-age > age+8 years

1.78 (1.68–1.89)

<0.001

1.56 (1.20–2.03)

<0.001

Adjusted by age, sex, hypertension, diabetes mellitus, smoking, dyslipidemia and obesity

 ECG-age < age−8 years

Not available

0.82 (0.69–0.98)

0.030

Not available

 ECG-age > age+8 years

1.57 (1.21–2.04)

<0.001

  1. The table displays the hazard ratios (HR) when the difference between ECG-age and chronological age are larger than 8 years (either positive or negative). The HR summarizes the Cox regression models obtained for overall mortality. The models were adjusted by different selection of variables (including age, sex, and cardiovascular risk factors).
  2. The number of death events was n = 8,118 for CODE-15%, n = 617 for ELSA-Brasil, and n = 104 for SaMi-Trop. When the ECG-age is more than 8 years smaller than the chronological age n = 1,861, n = 239, and n = 19, respectively, for the CODE-15%, ELSA-Brasil, and SaMi-Trop cohorts. When the ECG-age is more than 8 years greater than the chronological age n = 1,675, n = 69, and n = 41, respectively.
Back to article page