Table 2.
Bayesian joint model analysis estimating the association between the change in gut microbiota diversity and mortality at day 60
| Estimate (95% credible interval) | p-value | |
|---|---|---|
| Time-to-event | ||
| SAPS II score at ICU admission | 1.02 (0.99 to 1.05) | 0.26 |
| Charlson comorbidity index | 1.14 (0.97 to 1.35) | 0.11 |
| Reason for ICU admission | ||
| Sepsis | 1.82 (0.22 to 12.25) | 0.52 |
| Cardiac failure | 2.79 (0.70 to 13.07) | 0.15 |
| Respiratory failure | 1.95 (0.39 to 9.40) | 0.40 |
| Neurological disease | 1.65 (0.34 to 9.37) | 0.56 |
| Othera | Reference | Reference |
| Decrease in Shannon index (Hazard ratio) | 1.99 (1.04 to 4.51) | 0.035 |
| Time-varying Shannon index | ||
| (Intercept) | 4.30 (4.00 to 4.60) | – |
| Change per day | −0.10 (−0.20 to −0.10) | <0.01 |
| Change per day * Antibiotics (from admission to S1) | 0.00 (0.00 to 0.10) | 0.08 |
| Change per day * Antibiotics (from S1 to S2) | 0.00 (0.00 to 0.10) | 0.16 |
Sample size = 96; Number of longitudinal observations = 190; Number of events = 24 (25%)
p-value represent the tail probabilities of containing the zero-effect value
1Other: metabolic disorders, kidney failure, abdominal disease, drug abuse, ear-nose-throat surgery