Table 3. Predictor variables and multivariate model for pneumonia events in patients with diabetic nephropathy.
| Predictors | Hazard ratio | 95% CI | P value |
| Predictor variables | |||
| Age (per 10-year age increase) | 1.27 | 0.87–1.87 | 0.215 |
| Body mass index (per 1 kg/m2 increase) | 1.08 | 0.96–1.22 | 0.185 |
| Hypertension (yes/no) | 1.45 | 0.54–3.90 | 0.459 |
| Cardiovascular disease (yes/no) | 0.60 | 0.18–2.01 | 0.407 |
| Stroke (yes/no) | 0.77 | 0.23–2.60 | 0.674 |
| DBP (per 1 mmHg increase) | 0.98 | 0.95–1.01 | 0.133 |
| MAP (per 1 mmHg increase) | 0.98 | 0.91–1.01 | 0.182 |
| eGFR (per 1mL/min per 1.73m2 increase) | 1.00 | 0.90–1.13 | 0.883 |
| C-reactive protein(per 1 mg/L increase) | 0.98 | 0.91–1.05 | 0.539 |
| Serum albumin (per 1g/L increase) | 0.93 | 0.86–1.01 | 0.066 |
| Multivariate modela | |||
| Body mass index (per 1 kg/m2 increase) | 1.15 | 1.01–1.31 | 0.037 |
| Serum albumin (per 1g/L increase) | 0.87 | 0.78–0.98 | 0.014 |
Abbreviations: DBP, diastolic blood pressure; MAP, mean artery pressure; eGFR, estimated glomerular filtration rate; CI, confidence interval.
a
Adjusted for variables from the above predictor variables using a backward stepwise cox proportional hazards model with a stay criterion of 0.10. P<0.05 represents statistical significant.