Table 2.

Regression analysis of changes in body weight on semaglutide to efficacy outcomes

	Predicted change per 10% decrease in body weight Model 1		Predicted change per 10% decrease in body weight Model 2
Endpoint change at 52 weeks	Slope (95% CI)a	P	Slope (95% CI)a	P
KCCQ-CSS (points)	5.9 (3.6, 8.3)	<0.0001	6.4 (4.1, 8.8)	<0.0001
6MWD (m)	13.2 (4.4, 22.0)	0.0033	14.4 (5.5, 23.3)	0.0016
CRP (ratio)	0.75 (0.65, 0.86)	<0.0001	0.72 (0.63, 0.84)	<0.0001

Results are shown for the intention-to-treat analysis. Data are point estimates and 95% CIs, computed using multivariable regression analyses. P values are two-sided.

Model 1 is adjusted for baseline weight and baseline endpoint (baseline KCCQ-CSS, 6MWD or logarithm to CRP).

Model 2 is adjusted for baseline values of weight, respective endpoint, age, sex, history of atrial fibrillation, history of coronary artery disease, NYHA class, logarithm to CRP and logarithm to NTproBNP.

^aChange per 10% decrease in body weight on treatment with semaglutide.