Skip to main content

Table 3 Multivariate analysis for hospital mortalitya

From: The relationship between serum potassium, potassium variability and in-hospital mortality in critically ill patients and a before-after analysis on the impact of computer-assisted potassium control

 

OR (95% CI)

P -value

Model 1

Sex, female

1.08 (0.97 to 1.20)

0.159

Age

1.018 (1.014 to 1.021)

<0.001

Mean potassium

0.002 (0.000 to 0.008)

<0.001

Mean potassium squared

2.18 (1.85 to 2.57)

<0.001

Potassium variability

9.37 (7.25 to 12.10)

<0.001

Model 2

Sex, female

1.12 (1.01 to 1.25)

0.032

Age

1.017 (1.013 to 1.020)

<0.001

AKI

2.50 (2.25 to 2.79)

<0.001

Mean potassium

0.003 (0.001 to 0.013)

<0.001

Mean potassium squared

2.02 (1.71 to 2.38)

<0.001

Potassium variability

5.83 (4.49 to 7.58)

<0.001

Model 3

Sex, female

1.22 (1.05 to 1.42)

0.012

Age

1.008 (1.003 to 1.013)

0.002

APACHE II score

1.104 (1.091 to 1.116)

<0.001

AKI

1.76 (1.50 to 2.06)

<0.001

Mean potassium

0.008 (0.001 to 0.082)

<0.001

Mean potassium squared

1.84 (1.40 to 2.41)

<0.001

Potassium variability

5.61 (3.64 to 8.66)

<0.001

  1. aCI, Confidence interval; OR, Odds ratio. Data are adjusted for sex, age, acute kidney injury (AKI), severity of illness (Acute Physiology and Chronic Health Evaluation II (APACHE II) score), mean potassium, mean potassium squared and potassium variability observed between 24 hours and 7 days after admission. For all variables except potassium variability (9,228 patients (88%)) and APACHE II score (4,883 patients (51%)), virtually complete data were available, therefore the multivariate analysis was performed with APACHE II score (lower panel) and without APACHE II score. In-hospital mortality was associated with all domains of potassium control. In order to test for a U-shaped relationship of mean potassium with hospital mortality, the mean potassium concentration was both included directly and squared.
Back to article page

Critical Care

ISSN: 1364-8535