Skip to main content
. 2017 Dec 27;2017:8374215. doi: 10.1155/2017/8374215

Table 5.

Uni- and multivariable logistic regression analyses of risk factors for POF.

Univariable analysis Multivariable analysis
Odd ratio (95% CI) P value Odd ratio (95% CI) P value
Sex, male 0.94 (0.42, 2.13) 0.89
Age, ≥60 years 1.12 (0.46, 2.74) 0.8
Hemoglobin, >150 g/L 2.98 (1.30, 6.86) 0.01
#Glucose ≥ 11.1 mmol/L 7.85 (3.17, 19.41) <0.001
Albumin < 32 g/L 4.72 (1.86, 11.99) 0.001 7.18 (1.73, 29.79) 0.007
#Calcium < 2 mmol/L 26.00 (8.18, 82.62) <0.001
PDW ≥ 16.45% 21.94 (6.97, 69.07) <0.001 39.42 (8.64, 179.77) <0.001
Ranson score ≥ 4 4.75 (1.79, 12.57) 0.002 2.38 (0.58, 9.76) 0.227
SIRS score ≥ 3 5.743 (2.32,14.22) <0.001 3.78 (0.99,14.36) 0.051

#As serum glucose and serum calcium were not independent of Ranson score, they were excluded from multivariable analysis. Due to the small number of the study population and because the OR of albumin is higher than that of hemoglobin in the univariable model, we decide to include albumin, PDW, and Ranson score in multivariable analysis. CI: confidence interval; PDW: platelet distribution width.