Provision of Kidney Disease Education Service Is Associated with Improved Vascular Access Outcomes among US Incident Hemodialysis Patients

Abstract

Background: Pre-ESKD Kidney Disease Education (KDE) has been shown to improve multiple CKD outcomes, but its effect on vascular access outcomes is not well studied. In 2010, Medicare launched KDE reimbursements policy for patients with advanced CKD.

Methods: In this retrospective USRDS analysis, we identified all adult patients on incident hemodialysis with ≥6 months of pre-ESKD Medicare coverage during the first 5 years of CMS-KDE policy and divided them into CMS-KDE services recipients (KDE cohort) and nonrecipients (non-KDE cohort). The primary outcome was incident arteriovenous fistula (AVF) and the composite of incident AVF or arteriovenous graft (AVG) utilization. Secondary outcomes were central venous catheter (CVC) with maturing AVF/AVG and pure CVC utilizations. Step-wise multivariate analyses were performed in four progressive models (model 1, KDE alone; model 2, multivariate model encompassing model 1 with sociodemographics; model 3, model 2 with comorbidity and functional status; and model 4, model 3 with pre-ESKD nephrology care).

Results: Of the 211,990 qualifying patients on incident hemodialysis during the study period, 2887 (1%) received KDE services before dialysis initiation. The rates of incident AVF and composite AVF/AVG were more than double (30% and 35%, respectively, compared with 14% and 17%), and pure catheter use about a third lower (40% compared with 65%) in the KDE cohort compared with the non-KDE cohort. The maximally adjusted odds ratios in model 4 for study outcomes were incident AVF use, 1.78, 99% confidence interval, 1.55 to 2.05; incident AVF/AVG use, 1.78, 99% confidence interval, 1.56 to 2.03; incident CVC with maturing AVF/AVG, 1.69, 99% confidence interval, 1.44 to 1.97; and pure CVC without any AVF/AVG, 0.51, 99% confidence interval, 0.45 to 0.58. The benefits of the KDE service were maintained even after accounting for the presence, duration, and facility of ESKD care.

Conclusion: The occurrence of pre-ESRD KDE service is associated with significantly improved incident vascular access outcomes. Targeted studies are needed to examine the effect of KDE on patient engagement and self-efficacy as a cause for improvement in vascular access outcomes.

Keywords: CKD; ESKD; arteriovenous access; arteriovenous fistula; arteriovenous graft; chronic dialysis; chronic hemodialysis; clinical nephrology; dialysis; dialysis access; hemodialysis access.

Graphical abstract

Figure 1.

Study flow chart. USRDS, United States Renal Data System; CMS, Centers for Medicare and Medicaid Services; HCPCS, Healthcare Common Procedure Coding System; KDE, Kidney Disease Education. CMS-2728 report, CMS Medical Evidence report 2728.

Figure 2.

Rates of vascular access outcomes among the KDE and non-KDE cohorts. Any AVF/AVG includes mature-in-use and maturing not-in-use vascular access. CVC, central venous catheter; AVF, arteriovenous fistula; AVG, arteriovenous graft.

Figure 3.

Associations between various patient characteristics and vascular access outcomes in multivariate logistic regression models. Multivariate analyses in model 4 are depicted here and adjusted for age, sex, race, ethnicity, BMI, eGFR, serum albumin level, CHF, DM, CVD, functional limitations, prior nephrology care and duration, and KDE status using logistic regression. Cardiovascular disease in model 3 was defined as the presence of atherosclerotic heart disease, stroke, or peripheral vascular disease. BMI, body mass index; CHF, congestive heart failure; CVD, cardiovascular disease; DM, diabetes mellitus.