Table 3.
Sensitivity analysis based on different protein-energy wasting (PEW) conditions.
| Model 3 | Model 4 | Model 5 | Model 6 | |||||
|---|---|---|---|---|---|---|---|---|
| HR (95% CI) | p | HR (95% CI) | p | HR (95% CI) | p | HR (95% CI) | p | |
| DPI tertile (vs. first tertile) | ||||||||
| Second tertile | 0.812 (0.603–1.092) | 0.17 | 0.763 (0.572–1.018) | 0.07 | 0.792 (0.591–1.063) | 0.12 | 0.828 (0.620–1.105) | 0.20 |
| Third tertile | 0.714 (0.511–0.998) | 0.05 | 0.684 (0.491–0.952) | 0.02 | 0.705 (0.506–0.983) | 0.04 | 0.725 (0.523–1.006) | 0.05 |
| No. of PEW components (vs. 0) | ||||||||
| 1 | 1.029 (0.745–1.420) | 0.86 | - | - | - | - | - | - |
| 2 | 0.913 (0.619–1.346) | 0.64 | - | - | - | - | - | - |
| ≥3 | 1.800 (1.181–2.742) | 0.01 | - | - | - | - | - | - |
| PEW components ≥1 (yes vs. no) | - | - | 1.077 (0.799–1.452) | 0.63 | - | - | - | - |
| PEW components ≥2 (yes vs. no) | - | - | - | - | 1.131 (0.860–1.487) | 0.38 | - | - |
| PEW (yes vs. no) | - | - | - | - | - | - | 1.835 (1.297–2.596) | 0.001 |
DPI, dietary protein intake; PEW, protein energy wasting; HR, hazard ratio; CI, confidence interval; No., numbers. HR and 95% CI were analyzed using multivariate Cox proportional hazard regression analysis. The common covariates of models 3–6 were age, sex, current smoking, alcohol drinking, hypertension, diabetes, causes of chronic kidney disease, urine protein to creatinine ratio, blood urea nitrogen, estimated glomerular filtration rate, bilirubin, hemoglobin, and high sensitivity C-reactive protein.