Exploring the Association between Anemia and Negative Treatment Outcomes in Dialysis Patients at Medicare Facilities in the United States by County Poverty Levels Open Access

Fawcett, Emily (2016)

Permanent URL: https://etd.library.emory.edu/concern/etds/nk322d716?locale=en


End-stage kidney disease (ESKD), the final stage of chronic kidney disease (CKD) is a condition that affects thousands of adults annually. ESKD requires dialysis to ensure survival and has many comorbidities. Anemia is a common complication and is present in over half of all ESKD patients in the US. Anemia is also associated with an increased risk of negative health outcomes, most notably mortality and cardiovascular disease. Community-level poverty has also been shown to be associated with increased incidence and prevalence of ESKD and poorer medical care received.

This study investigated the association between anemia and the standardized ratios for mortality, hospitalization, readmission, and transfusion (SMR, SHR, SRR, and STrR) among Medicare dialysis facilities. In addition, poverty was included as an effect modifier of this relationship.

Data are from the Medicare Dialysis Facility Compare database, which collects patient data at dialysis facilities across the US. Poverty prevalence was obtained from 2010-2014 US Census data and matched to facilities based on county. Anemia was divided into high and low prevalence (facilities with less than 20% anemic patients were considered low prevalence), and poverty was divided into quintiles.

One-way ANOVAs were used to determine the relationship between anemia level and each of the four outcome variables individually. Next, multi-factor ANOVA incorporated poverty quintile as a predictor and an interaction term between anemia and poverty. Finally, two-way ANOVAs with only anemia and poverty as predictors were used for the outcome variables for which the interaction term was insignificant.

The means for SMR, SHR, SRR, and STrR all differed across anemia levels. In the multi-factor ANOVA, the interaction term was significant for the SMR and SRR models. Two-way ANOVA was run for the SHR and STrR models. Means for these two variables differed across both anemia levels and poverty quintiles.

Increased anemia prevalence had a negative impact upon SMR, SHR, SRR, and STrR. The effect of poverty was less well-defined, but means for the outcome variables varied across poverty quintiles. Trends were less coherent across poverty quintiles.

Future studies involving patient-level data rather than facility-level may improve the reliability of the results found here.

Table of Contents

Table of Contents







About this Master's Thesis

Rights statement
  • Permission granted by the author to include this thesis or dissertation in this repository. All rights reserved by the author. Please contact the author for information regarding the reproduction and use of this thesis or dissertation.
  • English
Research field
Committee Chair / Thesis Advisor
Partnering Agencies
Last modified

Primary PDF

Supplemental Files