BACKGROUND: The UK national policy promotes expansion of home haemodialysis, but there are no recent data on characteristics and outcomes of a national home haemodialysis population. METHODS: We compared incident home haemodialysis patients in England and Wales (n = 225, 1997-2005) with age- and sex-matched incident peritoneal dialysis, hospital haemodialysis and satellite haemodialysis patients with follow-up until 31 December 2006. Cox regression analyses included time-dependent changes of wait-listing for transplantation (a proxy for health status), start of home haemodialysis and transplantation. RESULTS: There was a median delay of 12 months between starting renal replacement therapy (RRT) and home haemodialysis. During that first year of RRT, > 50% of home haemodialysis patients were wait-listed for kidney transplantation; hospital haemodialysis patients had a lower rate of wait-listing over time [hazard ratio (HR) 0.56, 95% confidence interval (CI) 0.44-0.70; P < 0.001]. In crude analyses, there was a marked survival advantage of home haemodialysis patients compared with other modalities (log-rank P-value < 0.001). In adjusted analyses, being on home haemodialysis yielded a long-term survival benefit compared with peritoneal dialysis (HR 0.61, 95% CI 0.40-0.93), and a borderline advantage compared with hospital haemodialysis (HR 0.68, 95% CI 0.44-1.03). There was no evidence of an advantage compared with satellite haemodialysis (HR 0.94, 95% CI 0.65-1.37). CONCLUSIONS: Home haemodialysis patients have better survival compared with other dialysis modalities. Some of this crude survival advantage is due to selection of a healthier patient cohort as evidenced by higher transplant wait-listing rates. The advantage over peritoneal dialysis persisted after adjustment for wait-listing and transplantation over time.