Skip to main content
. 2023 Feb 8;8(3):222–230. doi: 10.1001/jamacardio.2022.5309

Table 3. Association Between Long-term Change in N-Terminal Pro–B-Type Natriuretic Peptide (NT-proBNP) With Risk for Incident Heart Failure Events and All-Cause Death.

Factor Hazard ratio (95% CI)a
% Change (continuous variable) Change category
>25% Decrease ≤25% Change >25% Increase
Incident heart failure hospitalization
Model 1b 1.09 (1.06-1.12) 1.24 (1.06-1.46) 1 [Reference] 2.43 (2.00-2.96)
Model 2c 1.06 (1.02-1.10) 1.11 (0.89-1.40) 1 [Reference] 1.84 (1.37-2.48)
Model 3d 1.12 (1.09-1.16) 0.91 (0.72-1.14) 1 [Reference] 4.38 (3.04-6.32)
Model 4e 0.98 (0.93-1.03) 1.27 (1.00-1.60) 1 [Reference] 1.26 (0.91-1.74)
All-cause death
Model 1b 1.08 (1.06-1.10) 1.08 (0.96-1.20) 1 [Reference] 1.82 (1.59-2.09)
Model 2c 1.05 (1.03-1.08) 0.96 (0.82-1.12) 1 [Reference] 1.54 (1.26-1.89)
Model 3d 1.07 (1.05-1.09) 0.87 (0.74-1.03) 1 [Reference] 2.25 (1.75-2.88)
Model 4e 1.01 (0.98-1.04) 1.02 (0.87-1.19) 1 [Reference] 1.25 (1.002-1.55)
a

Percent change modeled continuously as well as a categorical variable.

b

Model 1: age, sex, and race.

c

Model 2: model 1 plus systolic blood pressure, diastolic blood pressure, hypertensive medication use, diabetes, fasting glucose, low-density lipoprotein cholesterol, triglyceride level, cholesterol-lowering medication use, cigarette smoking, estimated glomerular filtration rate, body mass index, and prevalent coronary heart disease.

d

Model 3: model 2 plus visit 2 NT-proBNP (log-transformed).

e

Model 4: model 2 plus visit 4 NT-proBNP (log-transformed).