Trends in Incidence of Kidney Failure and Survival on Dialysis

[摘要] Background and Rationale End-stage kidney disease (ESKD) patients are typically treated through either maintenance dialysis or transplantation. Age-adjusted mortality rates of these patients have fallen over the past twenty years, but the rates remain much higher than in the general population. The incidence of patients starting ESKD treatment has risen over the same period, and both trends will significantly impact future ESKD prevalence. Understanding these trends is important because of the impact of ESKD on morbidity, mortality, quality of life, healthcare utilization, policy, and costs, especially the cost to Medicare. Methods Mediation analyses were conducted using time (1999-2015) as the ;;exposure” and nephrology practice measures as potential mediators that might explain temporal trends in the outcome, patient survival, using data from the international Dialysis Outcomes and Practice Patterns Study (DOPPS).A simulation model to simulate transitions between population prevalence of obesity, diabetes, hypertension, and ESKD incidence for age- and race-specific demographic groups, using smoothed estimates of time-varying parameters for incidence of diabetes, death rate among patients with diabetes, ESKD incidence, and hypertension incidence. Inputs included annual data from 1980-2013 from the Centers for Disease Control National Health and Nutrition Examination Survey, the Centers for Disease Control and Prevention National Health Interview Survey, the United States Renal Data System, and the US Census. The simulation model was used to project future trajectories of incidence rates and prevalence of ESKD, using different patient populations and smoothing methods to test the sensitivity of the model’s predictions to different inputs: increasing obesity versus falling obesity prevalence, and decreasing ESKD death rates versus constant ESKD death rates. These ranges were chosen to be wide enough to include all plausible future trajectories of these important contributors to ESKD incidence and prevalence. Results Based on the DOPPS sample, changes in practice measures explain 12%/decade (95% confidence interval +/-8%) longer survival in Europe, primarily through Kt/V and phosphorous control. Similarly, changes in practice measures explain a 9% (+/-5%) per decade survival improvement in Japan, primarily through the dose of dialysis (Kt/V) and intra-dialytic weight gain (IDWG), and a 26% (+/-9%) per decade survival improvement in the US, primarily through fistula use and phosphorous control. Age- and race-specific ESKD incidence rates and prevalence can be modeled from 1980 through 2013. ESKD incidence and prevalence will continue to rise through 2030 primarily due to demographic trends, such as the aging US population. The projected number of patients diagnosed with incident ESKD in 2030 is 137,000-151,000, a 19-32% increase from 2013, depending on obesity trends. The crude (unadjusted) incidence rate will rise to 381-421 per million/year, a 5-16% increase. The projected number of patients with ESKD in 2030 is 794,000-1,219,000, a 22-88% increase, depending on trends in obesity and ESKD death rates through 2030. Conclusion Evidence-based practices appear to explain a substantial proportion of the decline in lower death rates among hemodialysis patients in the US,Europe, and Japan. Current demographic trends indicate that despite possible decreases in ESKD incidence within age and race groups, possibly due to advances in treatment, we will have an expanding, increasingly older ESKD population in the US. We will need to continue to investigate and apply practices that result in improved outcomes, and then ensure sufficient resources are in place to care for the consequent increase in the number of patients requiring ESKD care.

[发布机构] University of Michigan

[效力级别] dialysis

[关键词] epidemiology;dialysis;kidney disease;simulation modeling;patient survival;mediation analyses;Statistics and Numeric Data;Science;Epidemiological Science