Timing, causes, predictors and prognosis of switching from peritoneal dialysis to hemodialysis: a prospective study

Table 4 Risk of mortality associated with switching from peritoneal dialysis to hemodialysis: relative hazards for mortality for switchers vs. non-switchers

	Relative Hazard (95% CI)
Model	Non-switchers	Switchers
No. of deaths/total no. of patients	62/197	26/65
Incidence rate, per 100 patient-years	18.5	13.5
Unadjusted	1.00 (ref.)	0.94 (0.51–1.73)
Adjusted
Model 1 (Demographics)	1.00 (ref.)	0.87 (0.46–1.66)
Model 2 (Model 1 + Clinical)	1.00 (ref.)	0.68 (0.33–1.40)
Model 3 (Model 2 + Laboratory)	1.00 (ref.)	0.89 (0.41–1.93)

In these analyses, all patients start as non-switchers; if the patient switched modality the patient then became a switcher on the date of switch.
*Demographics: age, race, employment; clinical: ICED, diabetes, CVD, BMI, residual urine output; laboratory: baseline albumin and creatinine. Fully adjusted models included all these variables.

ISSN: 1471-2369