Comparison of outcomes between emergent-start and planned-start peritoneal dialysis in incident ESRD patients: a prospective observational study

Table 2 Multivariate logistic regression analysis showing predictors for emergent initiation of dialysis in incident ESRD patients

Emergent-start dialysis	PD		HD
Emergent-start dialysis	Odds ratio (95% CI)	p value	Odds ratio (95% CI)	p value
Predialysis education program (no vs. yes)	3.17 (1.19–8.42)	0.021	24.77 (10.18–60.25)	< 0.001
eGFR (per ml/min/1.73 m²)	0.68 (0.51–0.92)	0.012	0.63 (0.53–0.75)	< 0.001
Albumin (per g/dl)	0.14 (0.04–0.43)	0.001	0.26 (0.15–0.46)	< 0.001
Comorbidities (yes vs. no)
Coronary artery disease	–	–	2.87 (1.05–7.86)	0.040
Valvular heart disease	–	–	5.89 (1.17–29.78)	0.032

Variable selection for multivariate logistic regression model was performed using stepwise multiple regression (stepwise forward and backward selection method as p < 0.05 criterion). For PD patients, it showed a percentage of concordant pairs = 94.6%, adjusted generalized R² = 0.261, estimated area under the receiver operating characteristic curve = 0.793, and Hosmer and Lemeshow goodness of fit test p = 0.853 > 0.05 (d.f = 8). The predicted probability of emergent-start PD = 1/(1 + exp.[− (8.93 + 1.153 x (predialysis education program) - 0.38 x (eGFR) - 1.99 x (albumin)]). For HD patients, it showed a percentage of concordant pairs = 91.2%, adjusted generalized R² = 0.414, estimated area under the receiver operating characteristic curve = 0.888, and Hosmer and Lemeshow goodness of fit test p = 0.115 > 0.05 (d.f = 8). The predicted probability of emergent-start HD = 1/(1 + exp.[− (10.06 + 3.21 x (predialysis education program) - 0.46 x (eGFR) - 1.35 x (albumin) + 1.06 x (coronary artery disease) + 1.77 x (valvular heart disease)])

ISSN: 1471-2369