Date of Award


Degree Name

Doctor of Philosophy (PhD)



Second Advisor


Third Advisor



Background: The objectives of this study were to first understand whether a dialysis center’s ownership status modified the association between informing end-stage renal disease (ESRD) patients about transplantation options and patients’ transplantation status; and second, to understand whether Medicare’s Prospective Payment System (PPS) implementation affected the cost of care for ESRD patients from Medicare’s perspective.

Method: This study used the United States Renal Data System (USRDS) data containing information on all ESRD patients in the US, linked to Medicare claims. To address the first objective, a multi-level analysis using mixed effect multinomial logistic regression model was conducted with patients aged 18 to 64, having an ESRD onset during 2006 to 2016. To address the second objective, an interrupted time series analysis (ITSA) was used to estimate the effect of Medicare’s PPS implementation on cost of care from Medicare’s perspective. The second aim examined patients 18 years and above with ESRD onset from 2005 to 2015. Total cost was further categorized into total outpatient, outpatient dialysis, outpatient non-dialysis, and inpatient costs, and analyzed using ITSA.

Results: The study showed that informing patients about transplantation options was associated with increased odds of enrolling in kidney transplant wait list (WL) or receiving a live donor kidney transplant (LDKT) at both for-profit and non-profit dialysis centers. However, this effect was less pronounced at for-profit as compared with non-profit dialysis centers (Non-profit dialysis centers: Enrolling in WL OR: 2.23 [95% CI, 2.07-2.40] and Receiving an LDKT OR: 3.35 [95% CI, 2.65-4.25]; For-profit dialysis centers: Enrolling in WL OR: 1.73 [95% CI, 1.66-1.79] and Receiving an LDKT OR: 2.35 [95% CI, 2.08-2.66]). The results also showed that the odds of informing patients was higher for for-profit as compared with non-profit dialysis centers, and characteristics of patients informed were similar between for-profit and non-profit dialysis centers.

With respect to costs, the study showed that PPS was significantly associated with decreased total cost; total cost remained steady before PPS, but cost declined over time after PPS (-0.88% [95% CI: (-0.96, -0.79)]). However, the effect of the PPS was not the same across cost categories. Although, total outpatient and outpatient dialysis costs were decreasing over time before PPS, these cost categories significantly increased immediately after PPS but continued to decline afterward. Outpatient non-dialysis cost was steady before PPS, and after a significant decline after PPS, continued to increase over time. Inpatient costs, which were increasing before PPS, had an immediate decline after PPS and showed a declining pattern over time after PPS. Conclusion: The first part of the study found that the beneficial effect of informing patients on transplantation status differs by the ownership status of the informing facility. Information provided by for-profit dialysis centers was less effective and potentially lower quality than that provided by non-profit dialysis centers. The study highlights the need for

developing guidelines to standardize transplantation information provided, in order to ensure similar informational quality across centers. The study also found that the PPS was associated with decreasing total cost of care, however, the effect was not the same for all cost subcategories. Nevertheless, PPS seems to have achieved the cost containment goal that it set out to achieve.