Table 3

Adjusted ORs using osmolarity_max as the design variable in patients with or without respiratory admission diagnosis

Variable	OR	95% CI	p	Variable	OR	95% CI	p
Model1				Model2
osmolarity_max (<290)	2.22	1.37 to 3.60	<0.001	osmolarity_max (<290)	1.64	0.68 to 3.94	0.26
osmolarity_max (290–309)	Ref.			osmolarity_max (290–309)	Ref.
osmolarity_max (310–319)	1.98	1.69 to 2.33	<0.001	osmolarity_max (310–319)	1.26	0.89 to 1.78	0.18
osmolarity_max (320–329)	3.20	2.69 to 3.82	<0.001	osmolarity_max (320–329)	1.17	0.79 to 1.74	0.4
osmolarity_max (330–339)	3.93	3.20 to 4.82	<0.001	osmolarity_max (330–339)	1.51	0.92 to 2.46	0.10
osmolarity_max (≥340)	4.45	3.58 to 5.53	<0.001	osmolarity_max (≥340)	2.03	1.20 to 3.42	0.01
Vasopressin use	1.89	1.54 to 2.31	<0.001	Vasopressin use	2.05	1.17 to 3.57	0.01
Fluid electrolyte disorder	1.42	1.25 to 1.61	<0.001	Congestive heart failure	1.40	1.06 to 1.86	0.02
Lymphoma disease	2.03	1.36 to 3.04	0.001	Lymphoma disease	3.43	1.57 to 7.49	0.01
Metastatic cancer	3.16	2.48 to 4.01	<0.001	Metastatic cancer	3.38	2.34 to 4.89	<0.001
Maximum SOFA score	1.23	1.21 to 1.25	<0.001	Maximum SOFA score	1.27	1.22 to 1.32	<0.001
Gender	1.58	1.40 to 1.78	<0.001
Cardiac arrhythmia	1.52	1.33 to 1.74	<0.001
Paralysis	1.52	1.02 to 2.29	0.04

Notes: Model1 contains information from 14 736 patients with non–respiratory admission diagnosis and the mean VIF was 1.43. Model2 contains information from 1750 patients with respiratory admission diagnosis and the mean VIF was 1.40. osmolarity_max represents the maximum serum osmolarity during intensive care unit stay and was categorised into six levels.
SOFA, sequential organ failure assessment; VIF, variance inflation factor.