Fracture dialysis patients have a higher risk of 1-year mortality compared with non-fracture dialysis patients. However, it is unclear whether excess mortality persists for more than a year. We conducted a nationwide cohort study in 162,360 hemodialysis patients in Japan. Study outcomes were 5-year all-cause mortality and cause-specific mortality. Cox proportional hazards regression was used to examine the association between hip fracture and mortality in two cohorts: the full cohort, which included potential confounders as covariates in multivariable-adjusted regression models, and the propensity score-matched cohort. Crude mortality rates for fracture patients were double those of non-fracture patients and persisted during the 5-year period. The association between hip fracture and mortality was significant even after adjusting for premorbid conditions (hazard ratio (HR) 1.22, 95% confidence interval (CI) 1.13-1.32). Similar findings were observed in the propensity score-matched cohort of 2410 patients (HR 1.20, 95% CI 1.05-1.36). While cause-specific mortality rates for all categories, with the exception of sudden deaths, were higher for fracture patients relative to non-fracture patients in the full unmatched cohort, only the mortality rate for heart disease was significantly higher for fracture patients relative to non-fracture patients in the propensity score-matched cohort. Excess mortality persisted for many years after hip fracture in hemodialysis patients, and was still present after adjusting for several premorbid conditions and propensity score matching.