Emergency Department Use and Hospital Admissions Among Patients With End-Stage Renal Disease in the United States

Patients with end-stage renal disease (ESRD) have the highest risk for hospitalization among those with chronic medical conditions, including heart failure, pulmonary disease, or cancer.¹ However, to our knowledge, no study has examined use of the emergency department (ED) among the national Medicare population with ESRD. We sought to describe ED visits and hospitalizations through the ED and to determine the sociodemographic and clinical characteristics of patients with ESRD who use ED services in the United States.

We examined all US adult patients 18 years or older with incident ESRD diagnoses from January 1, 2005, to December 31, 2011, and Medicare Part A or B insurance, identified from the US Renal Data System. Data were analyzed from June 1 to September 30, 2015. Use of the ED, hospital admissions, and diagnoses were obtained from the US Renal Data System and Medicare Physician/Supplier and Inpatient databases for Medicare Part A or B claims. Multivariable Poisson regression was conducted to assess the association of relevant patient variables with use of the ED. Patients using the ED more than 50 times in the first year of ESRD were excluded for the possibility of the ED as a primary dialysis provider (n = 502). Sociodemographic and clinical data were obtained from the Centers for Medicare & Medicaid Services 2728 Medical Evidence Report. Emergency department and admission diagnoses were categorized from International Classification of Diseases, Ninth Revision codes using the Clinical Classifications Software from the Healthcare Cost and Utilization Project.² This study was approved by the institutional review board of Emory University, who waived the need for informed consent for this database review.

A total of 769 228 adult patients (433 737 men [56.4%]; 335 484 women [43.6%] women; 7 missing data; mean [SD], age 63.2 [15.3] years) initiated ESRD treatment during the study period. Of these, 535 345 patients (69.6%) had at least 1 ED visit during the study period. More than half (55.0%) of all patients with ESRD visited the ED during their first year of ESRD, and patients with ESRD had a mean of 2.89, 2.48, and 2.54 ED visits per patient-year in the first, second, and third years of ESRD, respectively (Table 1).

In multivariable analysis, factors associated with higher rates of ED use included younger age, female sex, black (vs white) race, comorbid medical conditions, Medicaid insurance (vs Medicare alone), catheter or graft hemodialysis access (vs fistula), tobacco use, institutionalization, and more recent ESRD diagnosis. Factors associated with a lower rate of ED use included pre-ESRD nephrology care, erythropoietin use before ESRD, and private insurance (Table 2).

The most frequently diagnosed comorbid conditions were respiratory abnormalities (727 541 [6.6%]), congestive heart failure (344 244 [3.1%]), and chest pain (322 819 [2.9%]) in the first ESRD year. Hyperkalemia (78 755 [0.7%]) and fluid overload (52 634 [0.5%])—2 potentially preventable ESRD-related complications—accounted for 131 389 of 11 061 866 total ED visits (1.2%).

Nearly half (46.2%) of ED visits by patients with incident ESRD resulted in hospital admission, accounting for 2 108 915 admissions during the study period. The 3 most common admission diagnoses were hemodialysis access complication (107 609 [12.6%]), septicemia (66 554 [7.8%]), and congestive heart failure (64 001 [7.5%]) during the first ESRD year. Hyperkalemia (9345 [1.1%]) and fluid overload (9546 [1.2%]) accounted for 18 891 of 857 245 total hospital admissions (2.2%).

Adults with newly diagnosed ESRD have high rates of ED use and subsequent hospitalization. Patients with ESRD use the ED at 6-fold and 4-fold higher rates than the national mean rates for US adults and Medicare beneficiaries, respectively.^3,4 We identified several potentially preventable causes of ED use, including access to care. Catheter hemodialysis access was the strongest predictor of ED use in multivariable analysis, which may reflect catheter-associated problems, such as infections and inadequate dialysis. The hospital admission rate from the ED was 4 times higher than the mean national rate³ and was influenced by modifiable factors, including infectious and cardiovascular abnormalities. The study data set was limited to Medicare claims; however, nearly all patients with ESRD are eligible for Medicare coverage and thus reflect a national sample. Focusing on modifiable factors associated with ED use, such as ensuring that patients with chronic kidney disease have early access to nephrology care and placement of fistulas, could lead to improved care for patients with ESRD and decreased costs for health systems.

Corresponding Author: Rachel E. Patzer PhD, MPH, Division of Transplantation, Department of Surgery, Emory University School of Medicine, 5101 Woodruff Memorial Research Bldg, 101 Woodruff Cir, Atlanta, GA 30322 (rpatzer@emory.edu).

Published Online: August 22, 2016. doi:10.1001/jamainternmed.2016.4975.

Author Contributions: Ms Zhang had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Concept and design: Lovasik, Zhang, Hockenberry, Schrager, Mohan, Patzer.

Acquisition, analysis, or interpretation of data: Lovasik, Zhang, Schrager, Pastan, Mohan, Patzer.

Drafting of the manuscript: Lovasik, Zhang.

Critical revision of the manuscript for important intellectual content: All authors.

Statistical analysis: Lovasik, Zhang, Hockenberry, Mohan.

Obtaining funding: Patzer.

Administrative, technical, or material support: Pastan, Patzer.

Study supervision: Lovasik, Schrager, Patzer.

Conflict of Interest Disclosures: Dr Pastan reports owning a minority share in Fresnius College Park Dialysis. No other disclosures were reported.

Disclaimer: The data reported herein have been supplied by the US Renal Data System. The interpretation and reporting of these data are the responsibility of the authors and in no way should be seen as official policy or interpretation of the US government.

Additional Contributions: Stephen R. Pitts, MD, MPH, Department of Emergency Medicine, Emory University, and Andrew B. Adams, MD, PhD, Department of Surgery, Emory University, contributed to the design and preparation of this study. Neither was compensated in association with their respective contribution to this article.