Table 2.

Logistic regression and competing risk regression assessing association of CHOP ARDS Transcriptomic Subtypes (CATS) clusters and PICU mortality or probability of extubation by day 28 (accounting for the competing risk of death)

	PICU mortality		Probability of extubation
	OR (95% CI)a	p value	SHR (95% CI)b	p value
Unadjusted
CATS-1	Ref	–	Ref	–
CATS-2	0.67 (0.21–2.08)	0.487	1.01 (0.54–1.91)	0.967
CATS-3	0.19 (0.05–0.78)	0.021	2.15 (1.26–3.64)	0.005
Adjusted for PRISM III
CATS-1	Ref	–	Ref	–
CATS-2	0.53 (0.16–1.76)	0.300	1.14 (0.60–2.17)	0.686
CATS-3	0.13 (0.03–0.60)	0.009	2.83 (1.55–5.17)	0.001
Adjusted for immunocompromised
CATS-1	Ref	–	Ref	–
CATS-2	0.44 (0.12–1.58)	0.207	1.20 (0.66–2.18)	0.557
CATS-3	0.24 (0.06–1.07)	0.061	1.74 (1.02–3.00)	0.044
Adjusted for PRISM III + immunocompromised
CATS-1	Ref	–	Ref	–
CATS-2	0.35 (0.09–1.36)	0.130	1.35 (0.74–2.44)	0.329
CATS-3	0.18 (0.04–0.86)	0.031	2.39 (1.32–4.32)	0.004
Adjusted for PRISM + immunocompromised + ANC
CATS-1	Ref	–	Ref	–
CATS-2	0.28 (0.07–1.14)	0.074	1.44 (0.78–2.67)	0.247
CATS-3	0.11 (0.02–0.60)	0.011	2.77 (1.38–5.57)	0.004
Adjusted for PRISM + immunocompromised + ALC
CATS-1	Ref	–	Ref	–
CATS-2	0.42 (0.10–1.67)	0.221	1.25 (0.69–2.27)	0.456
CATS-3	0.19 (0.04–0.90)	0.036	2.37 (1.32–4.25)	0.004

ALC absolute lymphocyte count, ANC absolute neutrophil count, PRISM III Pediatric Risk of Mortality III

^aOdds ratio (OR) < 1: lower odds of mortality

^bSubdistribution hazard ratio (SHR) > 1: greater hazard for extubation alive (i.e., shorter duration of ventilation)