Table 7.
Multivariate generalized linear gamma regression explaining the cardiac patients' length of intensive care unit stay (days)
| Parameter | Exponentiated beta coefficient | 95% Wald C. I for Exp(B) | p value | |
|---|---|---|---|---|
| Lower | Upper | |||
| (Intercept) | 0.414 | 0.350 | 0.490 | < 0.001 |
| Sex = male | 1.028 | 0.955 | 1.105 | 0.463 |
| Age (years) | 1.005 | 1.002 | 1.007 | < 0.001 |
| Required inotropic support in the ITU/OR | 1.223 | 1.124 | 1.330 | < 0.001 |
| Body mass type: obese or overweight | 1.032 | 0.967 | 1.101 | 0.347 |
| Aortic cross-clamp time in minutes | 1.013 | 1.010 | 1.015 | < 0.001 |
| First 24 h total blood loss (cc) | 1.001 | 1.000 | 1.001 | < 0.001 |
| EUROSCORE | 1.043 | 1.001 | 1.086 | 0.046 |
Dependent variable: ICU length of stay days. Estimating method = gamma regression with maximum likelihood. Exponentiated beta coefficient is interpreted just like a risk rate or like an odds ratio