Skip to main content
. 2021 Aug 9;105(3):718–726. doi: 10.4269/ajtmh.21-0234

Table 3.

Association of PAB with mortality and hospital length of stay for COVID-19 patients*

OR (95% CI) P value β (95% CI) P value
Mortality Hospital length of stay, days
PAB1 PAB2 PAB1 PAB2
Model 1 1 (Ref) 5.87 (2.23–15.43) < 0.001 0 (Ref) 2.74 (0.45–5.03) 0.019
Model 2 1 (Ref) 4.59 (1.71–12.29) 0.002 0 (Ref) 2.34 (0.01–4.68) 0.049
Model 3 1 (Ref) 3.48 (1.14–10.64) 0.029 0 (Ref) 3.20 (0.99–5.41) 0.005

CI = confidence interval; COVID-19 = coronavirus disease 2019; OR = odds ratio; PAB = prealbumin.

Model 1: unadjusted.

Model 2: adjusted for age, sex, gastrointestinal disorder.

Model 3: adjusted for model 2 plus hypertension, diabetes, cardiovascular disease, and disease severity.

*

Logistic regression models were used to analyze the association between serum PAB and the prognosis, and linear regression models were used to analyze the association between serum PAB and hospital length of stay. PAB was a binary variable classified as PAB1 (PAB = 150–400 mg/L) or PAB2 (PAB < 150 mg/L).

Information of clinical outcome was added to model 3 additionally to evaluate the association between PAB and hospital length of stay.