The effect of home versus in-centre dialysis on patient-reported and clinical outcomes: a target trial emulation study

Abstract Background To make an informed decision between dialysis modalities, reliable information is needed on how each option may affect outcomes that matter to patients. Therefore, we aimed to 1) assess the effect of home dialysis (i.e. peritoneal dialysis [PD] and home haemodialysis) versus in-centre haemodialysis (ICHD, daytime and nocturnal), and 2) assess the effect of PD versus conventional daytime ICHD on patient-reported and clinical outcomes. Methods We conducted a multicentre cohort study in Dutch and Belgian incident dialysis patients. Primary outcomes were health-related quality of life and dialysis-related symptoms. Secondary outcomes were risk of first hospitalization and all-cause mortality. The target trial emulation framework was used and marginal structural models were fitted. Results In this study, including 2110 incident dialysis patients, no differences between home dialysis and ICHD were found in patient-reported outcomes and risk of first hospitalisation over three years. No difference in all-cause mortality was observed in the first two years. The 3-year hazard ratio (HR) showed a difference in favour of ICHD (HR = 1.77, 95% CI 1.08; 2.92), but sensitivity analyses did not confirm this result. These findings were consistent when comparing PD to conventional ICHD. Conclusion We found no differences in patient-reported outcomes, risk of first hospitalisation, and mortality between home dialysis and ICHD, nor between PD and conventional ICHD, in the first three years after dialysis initiation. The absence of evidence for a difference in these outcomes underlines the importance of patient preference in dialysis modality selection to ensure the therapy fits optimally into the patient’s life.