. Author manuscript; available in PMC: 2020 Feb 21.

Published in final edited form as: Eur Respir J. 2018 Oct 10;52(4):1801118. doi: 10.1183/13993003.01118-2018

Table 3.

Pairwise correlations of variables associated with lung disease.

	PRAGMA-%Bx	PMNs (%)	MetO (nM)	%OxMet	MPO (μg/ml)
PRAGMA-%Dis	n = 24	n = 23	n = 24	n = 24	n = 23
	ρ = 0.726	ρ = 0.502	ρ = 0.582	ρ = 0.437	ρ = 0.706
	p = 5.8 × 10⁻⁵	p = 0.0146	p = 0.0029	p = 0.0286	p = 1.7 × 10⁻⁴
PRAGMA-%Bx		n = 23	n = 24	n = 24	n = 23
		ρ = 0.6742	ρ = 0.698	ρ = 0.665	ρ = 0.752
		p = 4.2 × 10⁻⁴	p = 1.5 ×10⁻⁴	p = 3.9 × 10⁻⁴	p = 3.5×10⁻⁵
PMNs (%)			n = 23	n = 23	n = 22
			ρ = 0.569	ρ = 0.594	ρ = 0.656
			p = 0.0046	p = 0.0028	p = 0.0092
MetO (nM)				n = 24	n = 23
				ρ = 0.860	ρ = 0.803
				p = 7.2 × 10⁻⁸	p = 3.9 × 10⁻⁶
%OxMet					n = 23
					ρ = 0.837
					p = 6.4 × 10⁻⁷

Pairwise Spearman correlation results are shown for the listed variables. Note that the variable pairs of PRAGMA-%Dis and PRAGMA-%Bx and that of MetO and %OxMet are not independent (PRAGMA-%Bx contributes directly to PRAGMA-%Dis, and MetO contributes directly to %OxMet).