Mortality rates among prevalent hemodialysis patients in Beijing: a comparison with USRDS data.

BACKGROUND The raw annual mortality rate reported in Chinese patients on maintenance hemodialysis (MHD) was around 10% between 2005 to 2010, and it was around 20% in the US as reported by the United States Renal Data System (USRDS). Our hypothesis was that the large survival difference was caused by differences in race and practice pattern between nations in addition to differences in patient characteristics. METHODS Annual mortality in Beijing prevalent MHD patients per year in 2007, 2008, 2009 and 2010 was reported and relative risks of death were compared with the corresponding mortality of USRDS prevalent MHD patients (in whites, African-Americans and Asian-Americans) after age, gender and primary cause of end-stage renal disease (ESRD) were adjusted. A total of 11 675 MHD patients from 104 dialysis facilities under control of Beijing Blood Purification Quality Control and Improvement Center (BJBPQCIC) from 31 December 2006 to 31 December 2010 were included. A total of 1 937 819 MHD patients (only white, African-American and Asian-American were eligible for inclusion) were subtracted from the USRDS No-60-Day prevalent dataset from the year 2004 to 2009, using the RenDER system. Raw annual mortality for each race was reported as a number per 1000 MHD patients at risk for each year. Age, gender and primary cause of ESRD, adjusted annual mortality and relative risk race of death were reported comparing the Beijing patients and each race of the USRDS. RESULTS The raw annual mortality for the Beijing cohort increased gradually from 47.8 per 1000 patient-years in 2007 to 76.8 in 2010. The raw annual mortality for the white cohort in 2007 was 250.7 per 1000 patient-years, and gradually decreased to 236.3 in 2009. The raw annual mortality for African-Americans (167.8 and 156.7 per 1000 patient-years in 2007 and 2009, respectively) was much lower than that for whites. The annual mortality for Asian-Americans was slightly lower than that for African-Americans. After adjustment, Beijing MHD still had a survival benefit compared with each of the examined USRDS race. The annual mortality rates were 99.4, 80.6 and 94.3 per 1000 patient-years when adjusted to whites, African-Americans and Asian-Americans, respectively, in cohort 2009. CONCLUSIONS The annual mortality for the Beijing MHD patients was lower than that for their USRDS counterparts, and this difference existed after baseline demographics were adjusted. This survival difference between the Beijing and the USRDS MHD cohorts could be attributed to differences in race or practice pattern. More studies are needed to validate our hypothesis.