Bundled Payment Reform and Dialysis Facility Closures in ESKD

Abstract

Significance Statement Inclusion of injectable medications into an expanded ESKD payment bundle in 2011 led to concerns that dialysis facilities facing higher costs might close, disrupting the delivery of and access to care. The authors analyzed US Renal Data System registry data to compare facility closures before versus after the payment reform and examined whether patient-, facility-, and geography-related factors influenced closures. Closures affected ≤1.1% of dialysis facilities and ≤2221 patients in a given year. The relative likelihood of experiencing a closure decreased by 37% immediately after enactment of the payment bundle and by an additional 6% per year thereafter, indicating that the payment reform did not increase the risk of closures. However, closure risk increased slightly following payment reform for some high-risk populations, whereas it decreased in others. Background In 2011, inclusion of injectable medications into an expanded ESKD payment bundle prompted concerns that dialysis facilities facing higher costs might close, disrupting care delivery and access to care. Whether this policy change influenced dialysis facility closures is unknown. Methods To examine whether facility closures increased after 2011 and whether factors influencing closures changed, we analyzed US Renal Data System registry data to identify all patients receiving in-center hemodialysis from 2006 through 2015 and to track dialysis facility closures. We used interrupted time series logistic regression models and estimated marginal effects to examine immediate and longer-term changes in the likelihood of being affected by facility closures following payment reform. We also examined whether associations between selected predictors of closures indicating populations at “high risk” of closure (patient characteristics, facility characteristics, and geography-related characteristics) and closures changed after payment reform. Results Dialysis facility closures were uncommon over the study period. In adjusted models, the relative odds of experiencing a closure declined by 37% (odds ratio [OR], 0.63; 95% confidence interval [95% CI], 0.59 to 0.67) immediately after payment reform and declined by an additional 6% (OR, 0.94; 95% CI, 0.91 to 0.97) annually thereafter, corresponding to a 0.3% lower absolute probability of closure in 2015 in association with payment reform. Patients who were black and who dialyzed at small, hospital-based facilities experienced slight increases in closures following payment reform, whereas Hispanic and Medicare/Medicaid dual-eligible patients experienced slight decreases in closures. Conclusions Expansion of the ESKD payment bundle was not associated with increased closure of dialysis facilities, although the likelihood of closures changed slightly for some higher-risk populations.

Topics & Concepts