Late creation of vascular access increased post-hemodialysis mortality, hospitalization, and health-care expenditure: A population-based cohort study in Taiwan

Abstract

Background/purpose: The optimal timing of vascular access (VA) creation for hemodialysis (HD) and whether this timing affects mortality and health-care utilization after HD initiation remain unclear. Thus, we conducted a population-based study to explore their association. Methods: We used Taiwan's National Health Insurance Research Database to analyze health-care outcomes and utilization in a cohort initiating HD during 2003–2013. We stratified patients by the following VA creation time points: >180, 91–180, 31–90, and ≤30 days before and ≤30 days after HD initiation and examined all-cause mortality, ambulatory care utilization/costs, hospital admission/costs, and total expenditure within 2 years after HD. Cox regression, Poisson regression, and general linear regression were used to analyze mortality, health-care utilization, and costs respectively. Results: We identified 77,205 patients who started HD during 2003–2013. Compared with the patients undergoing VA surgery >180 days before HD initiation, those undergoing VA surgery ≤30 days before HD initiation had the highest mortality—15.92 deaths per 100-person-years, crude hazard ratio (HR) 1.56, and adjusted HR 1.28, the highest hospital admissions rates— 2.72 admission per person-year, crude rate ratio (RR) 1.48 and adjusted RR 1.32, and thus the highest health-care costs— US$31,390 per person-year, 7% increase of costs and 6% increase with adjustment within the 2-year follow-up after HD initiation. Conclusion: Late VA creation for HD can increase all-cause mortality, hospitalization, and health-care costs within 2 years after HD initiation. Early preparation of VA has the potential to reduce post-HD mortality and healthcare expenses for the ESKD patients.

Keywords

• Costs
• Hemodialysis
• Mortality
• Timing
• Vascular access