Survival in haemodialysis has improved considerably in Okinawa, Japan (1971-2020).

BACKGROUND: Dialysis care has advanced markedly over the past five decades; however, the extent of improvement and whether survival trends differ based on dialysis vintage remain unclear. We examined long-term mortality trends among patients on haemodialysis in Okinawa, Japan. METHODS: We analysed 50-year registry data (1971-2020) from all dialysis facilities in Okinawa. Among the 15 698 adults who had undergone maintenance dialysis, outcomes were ascertained in 96.2%. Hazard ratios (HRs) for all-cause mortality were estimated using a Cox model. Standardized mortality ratios (SMRs) were calculated using age- and sex-specific mortality rates of the general Okinawan population. Patients were stratified by dialysis vintage at risk (<36 vs. ≥36 months), and Joinpoint analysis was used to identify temporal changes in SMR trends. RESULTS: Overall survival improved steadily throughout the decades. Adjusted HRs for mortality declined to 0.43 (95% confidence interval, 0.38-0.49) in 2011-2020 from 1971 to 1980. However, the magnitude differed markedly with dialysis vintage. Among patients with a short vintage (<36 months), mortality declined to HR 0.21 in 2011-2020, whereas among those with a longer vintage (≥36 months; HR 0.67), improvement was attenuated. SMRs declined substantially until 2004, plateaued, then remained elevated in 2020. CONCLUSIONS: In this closed-cohort population, mortality among patients on dialysis declined markedly over five decades; survival gains were concentrated among patients with a short dialysis vintage, whereas improvements among long-vintage patients slowed in the 2000s. The persistent elevation of SMRs despite earlier improvements highlights the need for strategies targeting patients with prolonged dialysis exposure.