. 2020 Mar 25;22(9):1675–1683. doi: 10.1002/ejhf.1787

Table 3.

Analysis of changes in log‐transformed high‐sensitivity C‐reactive protein and serum uric acid levels from baseline to end of treatment with and without additional clinical covariates

	hsCRP changes				SUA changes
	Adjusted for hsCRP baseline level (log‐transformed)		Adjusted for hsCRP baseline level (log‐transformed) and clinical covariates		Adjusted for SUA baseline level (log‐transformed)		Adjusted for SUA baseline level (log‐transformed) and clinical covariates
	Effect ratio^a (95% CI)	P‐value	Effect ratio^a (95% CI)	P‐value	Effect ratio^a (95% CI)	P‐value	Effect ratio^a (95% CI)	P‐value
Treatment groups
Vericiguat 1.25 mg vs. placebo	0.80 (0.56–1.15)	0.235	0.84 (0.59–1.22)	0.367	0.94 (0.88–1.01)	0.083	0.94 (0.88–1.01)	0.081
Vericiguat 2.5 mg vs. placebo	0.76 (0.53–1.08)	0.125	0.71 (0.50–1.03)	0.069	0.94 (0.88–1.01)	0.088	0.95 (0.88–1.01)	0.114
Vericiguat 5.0 mg vs. placebo	0.74 (0.52–1.07)	0.108	0.69 (0.48–1.00)	0.049	0.92 (0.86–0.99)	0.020	0.91 (0.85–0.98)	0.010
Vericiguat 10.0 mg vs. placebo	0.68 (0.47–0.97)	0.035	0.66 (0.46–0.96)	0.024	0.90 (0.84–0.97)	0.004	0.90 (0.84–0.97)	0.004
Linear trend test		0.039		0.013		0.004		0.003

CI, confidence interval; hsCRP, high‐sensitivity C‐reactive protein: SUA, serum uric acid.

As biomarker levels were log‐transformed for the analyses, statistical testing is based on the effect ratio instead of the effect difference. The effect ratio of group 1 vs. group 2 is the ratio of the baseline‐adjusted biomarker levels at end of treatment in groups 1 and 2.