Table 1– Participant characteristics in the derivation cohort

	Participants n	TOR L·min⁻¹
		0	1–2	3–6	8–15
Participants n	104	42	20	28	14
Age yrs	104	62±12	62±6	62±7	64±4
Males %	104	79	70	79	93
Body mass index kg·m⁻²	104	28±5	30±6	28±4	30±5
Ever-smokers %	104	71	70	61	64
FVC % pred	104	65±21	45±17	44±16	50±16
Total lung capacity % pred	103	67±18	48±11	49±9	53±10
D_L,CO % pred	101	45±13	37±8	33±11	27±10
Resting supplemental oxygen prescription
L·min⁻¹	104	0 (0–0)	2 (0–2.5)	3 (2–4)	3.5 (0–4)
Room air^#	104	79	40	18	43
P̄_pa^¶ mmHg	83	19±4	22±4	23±7	27±11
PVR Wood units	76	2.3±0.7	2.5±0.7	3.2±1.3	3.0±1.4
6MWD m	104	456±134	372±133	335±101	357±119
End-walk S_p,O₂ %	104	89±7	84±7	83±7	92±6

Data are presented as mean±sd or median (interquartile range), unless otherwise stated. End-walk arterial oxygen saturation measured by pulse oximetry (S_p,O₂) represents oxyhaemoglobin saturation at the end of 6-min walk testing. TOR: titrated oxygen requirement; FVC: forced vital capacity; % pred: % predicted; D_L,CO: diffusing capacity of the lung for carbon monoxide; P̄_pa: mean pulmonary artery pressure; PVR: pulmonary vascular resistance; 6MWD: 6-min walk distance. ^#: 0 L·min⁻¹; ^¶: missing for 14 participants in the 0 L·min⁻¹ group, three in the 1–2 L·min⁻¹ group, two in the 3–6 L·min⁻¹ group and two in the 8–15 L·min⁻¹ group.