Table 5.
Logistic regression analysis showing the association of SIRS score at admission with the secondary study outcome (pooled mortality three months after discharge) in elderly patients with suspected sepsis who completed the study follow-up (n = 247).
| Parameters | OR (95% CI) | p * |
|---|---|---|
| Model 1—Univariable | ||
| SIRS score | 1.369 (1.032–1.816) | 0.03 |
| Model 2—Bivariable | ||
| SIRS score | 1.260 (0.876–1.812) | 0.21 |
| Frailty | 4.855 (2.632–8.954) | <0.001 |
| Model 3—Multivariable | ||
| SIRS score | 1.149 (0.696–1.897) | 0.57 |
| Frailty | 2.505 (1.038–6.044) | 0.04 |
| Age | 0.995 (0.920–1.077) | 0.91 |
| Sex (female vs. male) | 0.414 (0.149–1.148) | 0.09 |
| Institutionalization | 0.693 (0.180–2.665) | 0.59 |
| Dementia | 1.105 (0.319–3.829) | 0.88 |
| CIRS-SI | 1.543 (0.930–2.562) | 0.09 |
| CRP | 0.999 (0.992–1.005) | 0.68 |
| PaCO2 | 1.011 (0.969–1.033) | 0.22 |
* p values < 0.05 are indicated in bold. The multivariable model was built considering the parameters to be significantly different between survivors and non-survivors at descriptive analysis. Frailty was considered as a categorical variable based on the Rockwood Clinical Frailty Scale score (1–3 = fit; 4–5 = frail; 6–9 = disabled). OR = Odds Ratio; SIRS = Systemic Inflammatory Response Syndrome; CIRS-SI = Cumulative Illness Rating Scale Severity Index; CRP = C-Reactive Protein.