[Skip to Content]
Access to paid content on this site is currently suspended due to excessive activity being detected from your IP address 54.205.176.107. Please contact the publisher to request reinstatement.
Sign In
Individual Sign In
Create an Account
Institutional Sign In
OpenAthens Shibboleth
[Skip to Content Landing]
Download PDF
Figure 1.
Arteriovenous Fistula (AVF) and Arteriovenous Graft (AVG) Utilization Within Medical Insurance Status Categories Comparing Black Patients and Hispanic Patients With White Patients
Arteriovenous Fistula (AVF) and Arteriovenous Graft (AVG) Utilization Within Medical Insurance Status Categories Comparing Black Patients and Hispanic Patients With White Patients

A, Black patients and Hispanic patients initiated hemodialysis with an AVF less frequently than white patients irrespective of medical insurance status. B, Black patients initiated hemodialysis with an AVG more often than white patients and Hispanic patients irrespective of medical insurance status.

aSignificant at P < .05.

Figure 2.
Arteriovenous Fistula (AVF) and Arteriovenous Graft (AVG) Utilization Within Categories of Receipt of Nephrology Care Comparing Black Patients and Hispanic Patients With White Patients
Arteriovenous Fistula (AVF) and Arteriovenous Graft (AVG) Utilization Within Categories of Receipt of Nephrology Care Comparing Black Patients and Hispanic Patients With White Patients

A, With increasing duration of receipt of nephrology care, fewer black patients and Hispanic patients initiated hemodialysis with an AVF compared with white patients. B, Black patients initiated hemodialysis with an AVG more often than white patients and Hispanic patients irrespective of receipt of nephrology care.

aSignificant at P < .05.

Table 1.  
Characteristics of Patients by Race/Ethnicitya
Characteristics of Patients by Race/Ethnicitya
Table 2.  
Characteristics of Patients Stratified by Hemodialysis Access Typea
Characteristics of Patients Stratified by Hemodialysis Access Typea
Table 3.  
Propensity Model–Adjusted Odds Ratios of Initiating Hemodialysis With an AVF
Propensity Model–Adjusted Odds Ratios of Initiating Hemodialysis With an AVF
1.
U.S. Renal Data System. USRDS 2013 Annual Data Report: Atlas of Chronic Kidney Disease and End-Stage Renal Disease in the United States. Bethesda, MD: National Institute of Diabetes and Digestive and Kidney Diseases, National Institutes of Health; 2013.
2.
Foley  RN, Chen  SC, Collins  AJ.  Hemodialysis access at initiation in the United States, 2005 to 2007: still “catheter first.” Hemodial Int. 2009;13(4):533-542.
PubMedArticle
3.
DeSilva  RN, Sandhu  GS, Garg  J, Goldfarb-Rumyantzev  AS.  Association between initial type of hemodialysis access used in the elderly and mortality. Hemodial Int. 2012;16(2):233-241.
PubMedArticle
4.
Wasse  H, Speckman  RA, McClellan  WM.  Arteriovenous fistula use is associated with lower cardiovascular mortality compared with catheter use among ESRD patients. Semin Dial. 2008;21(5):483-489.
PubMedArticle
5.
Ravani  P, Palmer  SC, Oliver  MJ,  et al.  Associations between hemodialysis access type and clinical outcomes: a systematic review. J Am Soc Nephrol. 2013;24(3):465-473.
PubMedArticle
6.
Stehman-Breen  CO, Sherrard  DJ, Gillen  D, Caps  M.  Determinants of type and timing of initial permanent hemodialysis vascular access. Kidney Int. 2000;57(2):639-645.
PubMedArticle
7.
Reddan  DN, Szczech  LA, Klassen  PS, Owen  WF  Jr.  Racial inequity in America’s ESRD program. Semin Dial. 2000;13(6):399-403.
PubMedArticle
8.
Frankenfield  DL, Rocco  MV, Roman  SH, McClellan  WM.  Survival advantage for adult Hispanic hemodialysis patients? findings from the End-Stage Renal Disease Clinical Performance Measures project. J Am Soc Nephrol. 2003;14(1):180-186.
PubMedArticle
9.
Arce  CM, Mitani  AA, Goldstein  BA, Winkelmayer  WC.  Hispanic ethnicity and vascular access use in patients initiating hemodialysis in the United States. Clin J Am Soc Nephrol. 2012;7(2):289-296.
PubMedArticle
10.
Allon  M, Ornt  DB, Schwab  SJ,  et al; Hemodialysis (HEMO) Study Group.  Factors associated with the prevalence of arteriovenous fistulas in hemodialysis patients in the HEMO Study. Kidney Int. 2000;58(5):2178-2185.
PubMedArticle
11.
Malas  MB, Canner  JK, Hicks  CW,  et al.  Trends in incident hemodialysis access and mortality [published online March 4]. JAMA Surg. 2015.doi:10.1001/jamasurg.2014.3484.
12.
Rhee  CM, Lertdumrongluk  P, Streja  E,  et al.  Impact of age, race and ethnicity on dialysis patient survival and kidney transplantation disparities. Am J Nephrol. 2014;39(3):183-194.
PubMedArticle
13.
Kucirka  LM, Grams  ME, Lessler  J,  et al.  Association of race and age with survival among patients undergoing dialysis. JAMA. 2011;306(6):620-626.
PubMedArticle
14.
Arce  CM, Goldstein  BA, Mitani  AA, Winkelmayer  WC.  Trends in relative mortality between Hispanic and non-Hispanic whites initiating dialysis: a retrospective study of the US Renal Data System. Am J Kidney Dis. 2013;62(2):312-321.
PubMedArticle
15.
Derose  SF, Rutkowski  MP, Crooks  PW,  et al.  Racial differences in estimated GFR decline, ESRD, and mortality in an integrated health system. Am J Kidney Dis. 2013;62(2):236-244.
PubMedArticle
16.
Feldman  HI, Joffe  M, Rosas  SE, Burns  JE, Knauss  J, Brayman  K.  Predictors of successful arteriovenous fistula maturation. Am J Kidney Dis. 2003;42(5):1000-1012.
PubMedArticle
17.
Lok  CE, Allon  M, Moist  L, Oliver  MJ, Shah  H, Zimmerman  D.  Risk equation determining unsuccessful cannulation events and failure to maturation in arteriovenous fistulas (REDUCE FTM I). J Am Soc Nephrol. 2006;17(11):3204-3212.
PubMedArticle
18.
Ocak  G, Rotmans  JI, Vossen  CY,  et al.  Type of arteriovenous vascular access and association with patency and mortality. BMC Nephrol. 2013;14:79.
PubMedArticle
19.
Propper  B, Black  JH  III, Schneider  EB,  et al.  Hispanic ethnicity is associated with increased costs after carotid endarterectomy and carotid stenting in the United States. J Surg Res. 2013;184(1):644-650.
PubMedArticle
20.
Sonel  AF, Good  CB, Mulgund  J,  et al; CRUSADE Investigators.  Racial variations in treatment and outcomes of black and white patients with high-risk non–ST-elevation acute coronary syndromes: insights from CRUSADE (Can Rapid Risk Stratification of Unstable Angina Patients Suppress Adverse Outcomes With Early Implementation of the ACC/AHA Guidelines?). Circulation. 2005;111(10):1225-1232.
PubMedArticle
21.
Epstein  AJ, Gray  BH, Schlesinger  M.  Racial and ethnic differences in the use of high-volume hospitals and surgeons. Arch Surg. 2010;145(2):179-186.
PubMedArticle
22.
Centers for Medicare & Medicaid Services. Medicare coverage of kidney dialysis & kidney transplant services. 2012. http://www.medicare.gov/pubs/pdf/10128.pdf. Accessed June 1, 2013.
23.
Powe  NR, Melamed  ML.  Racial disparities in the optimal delivery of chronic kidney disease care. Med Clin North Am. 2005;89(3):475-488.
PubMedArticle
24.
Weech-Maldonado  R, Morales  LS, Elliott  M, Spritzer  K, Marshall  G, Hays  RD.  Race/ethnicity, language, and patients’ assessments of care in Medicaid managed care. Health Serv Res. 2003;38(3):789-808.
PubMedArticle
25.
Ishaque  B, Zayed  MA, Miller  J,  et al.  Ethnic differences in arm vein diameter and arteriovenous fistula creation rates in men undergoing hemodialysis access. J Vasc Surg. 2012;56(2):424-431.
PubMedArticle
26.
Sales  CM, Schmidt  RJ; FFBI Clinical Practice Workgroup.  Fistula First clarifies its message. J Vasc Surg. 2010;52(6):1699-e1. doi:10.1016/j.jvs.2010.10.038.
PubMedArticle
27.
Dember  LM, Imrey  PB, Beck  GJ,  et al.  Objectives and design of the Hemodialysis Fistula Maturation Study. Am J Kidney Dis. 2014;63(1):104-112.
PubMed
28.
Wasse  H, Hopson  SD, McClellan  W.  Racial and gender differences in arteriovenous fistula use among incident hemodialysis patients. Am J Nephrol. 2010;32(3):234-241.
PubMedArticle
29.
Kazmi  WH, Obrador  GT, Khan  SS, Pereira  BJ, Kausz  AT.  Late nephrology referral and mortality among patients with end-stage renal disease: a propensity score analysis. Nephrol Dial Transplant. 2004;19(7):1808-1814.
PubMedArticle
30.
Kessler  M, Frimat  L, Panescu  V, Briançon  S.  Impact of nephrology referral on early and midterm outcomes in ESRD: Epidémiologie de l’Insuffisance Renale Chronique Terminale en Lorraine (EPIREL): results of a 2-year, prospective, community-based study. Am J Kidney Dis. 2003;42(3):474-485.
PubMedArticle
31.
Longenecker  JC, Coresh  J, Klag  MJ,  et al.  Validation of comorbid conditions on the end-stage renal disease Medical Evidence Report: the CHOICE study: Choices for Healthy Outcomes in Caring for ESRD. J Am Soc Nephrol. 2000;11(3):520-529.
PubMed
32.
Hopson  S, Frankenfield  D, Rocco  M, McClellan  W.  Variability in reasons for hemodialysis catheter use by race, sex, and geography: findings from the ESRD Clinical Performance Measures Project. Am J Kidney Dis. 2008;52(4):753-760.
PubMedArticle
Original Investigation
June 2015

Racial/Ethnic Disparities Associated With Initial Hemodialysis Access

Author Affiliations
  • 1Department of Surgery, Dartmouth-Hitchcock Medical Center, Lebanon, New Hampshire
  • 2Division of Vascular Surgery, Department of Surgery, Johns Hopkins Medical Institutions, Baltimore, Maryland
  • 3UC Davis Health System, University of California at Davis, Sacramento
  • 4Former Editor, JAMA Surgery
JAMA Surg. 2015;150(6):529-536. doi:10.1001/jamasurg.2015.0287
Abstract

Importance  Superior outcomes have been established with the use of an arteriovenous fistula (AVF) at first hemodialysis. However, considering the influence of comorbidities, medical insurance, and specialist care, racial/ethnic differences in the patterns of utilization of AVFs are unknown and deserve evaluation.

Objective  To assess national trends in initial hemodialysis access with respect to race/ethnicity stratified by comorbid disease, nephrology care, and medical insurance status within the US Renal Data System.

Design, Setting, and Participants  A retrospective analysis of all patients with end-stage renal disease in the US Renal Data System who initiated hemodialysis between January 1, 2006, and December 31, 2010. Univariable statistics (χ2 test and analysis of variance) and logistic regression were used to compare racial/ethnic groups (white vs black vs Hispanic). Multivariable logistic regression and propensity score–matching techniques were used to evaluate hemodialysis access rates between patients of different races/ethnicities with comparable characteristics.

Main Outcomes and Measures  Utilization rates of AVF, arteriovenous graft, and intravascular hemodialysis catheter.

Results  In this cohort of 396 075 patients, more white patients initiated hemodialysis with an AVF than black patients or Hispanic patients (18.3% vs 15.5% and 14.6%, respectively; P < .001). Black patients and Hispanic patients initiated hemodialysis with an AVF less frequently despite being younger and having less coronary artery disease, chronic obstructive pulmonary disease, and cancer than white patients with an AVF. When stratified by medical insurance status, black patients (odds ratios, 0.90 [95% CI, 0.82-0.98] for uninsured and 0.85 [95% CI, 0.84-0.87] for insured) and Hispanic patients (odds ratios, 0.72 [95% CI, 0.65-0.81] for uninsured and 0.81 [95% CI, 0.79-0.84] for insured) persistently initiated hemodialysis with an AVF less frequently than white patients (P < .05 for all). Arteriovenous fistula utilization at initial hemodialysis was lower among black patients (odds ratio, 0.81 [95% CI, 0.78-0.84]) and Hispanic patients (odds ratio, 0.86 [95% CI, 0.82-0.90]) compared with white patients within the category of patients who had nephrology care for longer than 1 year (P < .001 for all).

Conclusions and Relevance  Black patients and Hispanic patients tend to initiate hemodialysis with an AVF less frequently than white patients despite being younger and having fewer comorbidities. These disparities persisted independent of factors that drive health access for fistula placement, such as medical insurance status and nephrology care. The sociocultural underpinnings of these disparities deserve investigation and redress to maximize the benefits of initiating hemodialysis via fistula in patients with end-stage renal disease irrespective of race/ethnicity.

Introduction

In 2010, end-stage renal disease (ESRD) affected 593 086 persons in the United States, 383 992 of whom were treated with hemodialysis.1 Arteriovenous fistulas (AVFs) confer a survival benefit relative to intravascular hemodialysis catheters (IHCs) in the incident hemodialysis population, a finding reported in multiplicity before and during the era of the fistula first–based initiative.25 The results of racially/ethnically stratified studies69 suggest that minority patients tend to receive inferior incident hemodialysis access, either with arteriovenous grafts (AVGs) or IHCs. Minority patients have more comorbid diseases or socioeconomic challenges that limit their access to the health care system.9 The concern about increased surgical risk results in higher incident hemodialysis with an IHC. A lack of medical insurance coverage delays permanent hemodialysis access placement until the final stages of ESRD, leaving too short of an interval to mature a fistula before starting hemodialysis.7,10

However, the actual demographic differences between white vs black vs Hispanic ESRD populations, as well as the effect of these differences on incident hemodialysis access, are unknown. Within the US Renal Data System (USRDS) database, this study evaluates national trends in incident hemodialysis access with respect to race/ethnicity (both overall and stratified by comorbid disease), receipt of nephrology care, and medical insurance status.

Methods
Study Design

A retrospective analysis of a prospectively maintained database was performed that examined all patients initiating renal replacement therapy between January 1, 2006, and December 31, 2010, in the USRDS. The Johns Hopkins Medicine institutional review boards and the USRDS approved this study and waived the need for informed consent. All patients diagnosed as having ESRD in the United States are registered in the USRDS. Data entry required a physician, nurse, or social worker to complete the Centers for Medicare & Medicaid Services Form 2728 within 45 days after patients receive a kidney transplant or establish chronic dialysis, even if the patient died within this period. Forms were not submitted for patients with acute renal failure. General and medical demographics were collected, as well as medical insurance status, hemodialysis access type, nephrology care, and ESRD network. Patients were categorized as white, black, or Hispanic race/ethnicity as delineated on the Centers for Medicare & Medicaid Services Form 2728. Patients in the category of other race/ethnicity were excluded from the analysis. Also excluded were patients with hemodialysis access established before 2006, as well as patients who received a kidney transplant at any time (as determined from cross-referenced data from the United Network for Organ Sharing).

Statistical Analysis

The primary outcome of the study was incident hemodialysis access with an AVF vs AVG vs IHC. Statistical tests, including χ2 test and analysis of variance, were performed, and logistic regression was used to calculate the odds of each racial/ethnic group’s receiving a particular hemodialysis access type. Propensity score matching was performed to evaluate hemodialysis access rates between patients of different races/ethnicities who had comparable demographic and medical characteristics. First, the independent predictors of initiating hemodialysis with an AVF were determined. Thereafter, propensity scores were generated based on the predictive variables. Logistic regression of matched subgroups was performed, and the probabilities of initiating hemodialysis with the reference access type were calculated. All analyses were performed with statistical software (STATA, version 12.1; StataCorp LP). P < .05 was considered significant.

Results
Overall Patient Characteristics

Our cohort consisted of 396 075 patients initiating dialysis between January 1, 2006, and December 31, 2010, as identified in the USRDS. Most patients (55.4%) were of white race/ethnicity, followed by 30.3% black patients and 14.3% Hispanic patients (Table 1). More white patients initiated hemodialysis with an AVF than black patients or Hispanic patients (18.3% vs 15.5% and 14.6%, respectively; P < .001). An IHC was the dominant hemodialysis access mode, used in 312 351 patients (78.9%) among the cohort. Diabetic nephropathy (43.2%) and hypertension (29.4%) were the most common causes of ESRD.

Overall, black patients and Hispanic patients were younger than white patients (mean age, 58.8 vs 58.9 vs 67.0 years, respectively; P < .001) and were more frequently female (48.1% and 43.3% vs 41.4%, respectively; P < .001) (Table 1). White patients had significantly more comorbidities, including congestive heart failure, coronary artery disease, peripheral arterial disease, chronic obstructive pulmonary disease, and cancer (P < .001 for all). Black patients and Hispanic patients also had receipt of nephrology care less frequently than white patients (57.9% vs 54.0% vs 63.2%, respectively) and were uninsured more often than white patients (11.7% vs 14.1% vs 4.4%, respectively) (P < .001 for both).

Race/Ethnicity and Incident Hemodialysis Access Type

When categorized by incident hemodialysis access type, white patients had a 22% greater chance of initiating hemodialysis with an AVF than black patients (odds ratio [OR], 1.22; 95% CI, 1.20-1.24) and a 32% greater chance of initiating hemodialysis with an AVF than Hispanic patients (OR, 1.32; 95% CI, 1.28-1.35). In contrast, black patients had 80% higher odds of having a functional AVG at first hemodialysis than white patients (OR, 1.80; 95% CI, 1.74-1.86) and 97% higher odds of having an AVG than Hispanic patients (OR, 1.97; 95% CI, 1.87-2.08). Hemodialysis initiation with an IHC was highest among Hispanic patients compared with white patients (OR, 1.29; 95% CI, 1.26-1.33).

A complete breakdown of patient demographics stratified by race/ethnicity and incident hemodialysis access type is summarized in Table 2. Women initiated hemodialysis with an AVG more frequently across all studied groups. White patients initiating hemodialysis with an AVF were older than black patients or Hispanic patients initiating hemodialysis with both an IHC and AVG (P < .001 for all). Black patients and Hispanic patients demonstrated significantly lower rates of congestive heart failure, coronary artery disease, chronic obstructive pulmonary disease, and cancer than white patients within each of the 3 hemodialysis access groupings (P < .001 for all). Black patients and Hispanic patients initiating hemodialysis with an IHC or AVG exhibited higher frequencies of and more severe diabetes mellitus, hypertension, and drug abuse and alcohol abuse (P < .001 for all).

Propensity Score Analysis

The propensity score analyses summarized in Table 3 indicate that older age, higher body mass index, presence of hypertension, receipt of nephrology care, and medical insurance coverage were all significantly associated with the presence and use of a functioning AVF at incident hemodialysis (P < .001 for all). Female sex, congestive heart failure, peripheral arterial disease, chronic obstructive pulmonary disease, cancer, diabetes mellitus, alcohol abuse, and immobility were all significantly associated with lower odds of initiating hemodialysis with an AVF (P < .001 for all). Coronary artery disease, cerebrovascular disease, tobacco use, and drug abuse were not independent predictors of initiating hemodialysis with an AVF.

The propensity score generated from the group of white patients initiating hemodialysis with an AVF was then applied to the combined AVG and IHC groups for white, black, and Hispanic patients to obtain the probabilities of initiating hemodialysis with an AVF within each racial/ethnic category. The probability that white patients from the combined AVG and IHC groups would initiate hemodialysis with an AVF was 0.163 compared with 0.151 for black patients and 0.143 for Hispanic patients (P < .001 for both).

Race/Ethnicity and Hemodialysis Access Type Stratified by Medical Insurance Status and Nephrology Care

Medicare was the most common medical insurance status across all hemodialysis access groupings regardless of race/ethnicity (Table 2). However, black patients and Hispanic patients initiating hemodialysis with any access modality were significantly more likely to have no medical insurance or Medicaid compared with white patients (P < .001). To compare similar groups, the racial/ethnic groups were stratified by medical insurance status (no insurance vs Medicaid vs Medicare vs private insurance). Regardless of medical insurance status, both black patients and Hispanic patients initiated hemodialysis with an AVF less frequently than white patients (Figure 1A). When stratified by medical insurance status, black patients (ORs, 0.90 [95% CI, 0.82-0.98] for uninsured and 0.85 [95% CI, 0.84-0.87] for insured) and Hispanic patients (ORs, 0.72 [95% CI, 0.65-0.81] for uninsured and 0.81 [95% CI, 0.79-0.84] for insured) persistently initiated hemodialysis with an AVF less frequently than white patients (P < .05 for all). Within the uninsured population, Hispanic patients also initiated hemodialysis with an AVF significantly less often than black patients (P < .001). Incident AVF use was similar between black patients and Hispanic patients who had Medicare (P = .92). Black patients initiated hemodialysis with an AVG more often than either white patients or Hispanic patients regardless of medical insurance status (Figure 1B).

The weighted impact of nephrology care on initiating hemodialysis with an AVF was at least 7 times higher than that of other variables. Within the stratum of patients who had no nephrology care, AVF use was higher among black patients (OR, 1.29; 95% CI, 1.18-1.40; P < .001) and Hispanic patients (OR, 1.13; 95% CI, 1.02-1.26; P = .02) compared with white patients (Figure 2A). Arteriovenous fistula use was similar among the races/ethnicities within the category of patients who had nephrology care for less than 6 months, with ORs of 0.99 (95% CI, 0.95-1.06; P = .99) for black patients and 0.99 (95% CI, 0.92-1.07; P = .88) for Hispanic patients. Among the patients who received nephrology care for 6 to 12 months, AVF use was less for black patients and Hispanic patients (OR, 0.87; 95% CI, 0.84-0.91; P < .001) compared with white patients (OR, 0.87; 95% CI, 0.83-0.92; P < .001). Within the category of patients who had nephrology care for longer than 1 year, AVF use was even lower for black patients (OR, 0.81; 95% CI, 0.78-0.84) and Hispanic patients (OR, 0.86; 95% CI, 0.82-0.90) (P < .001 for both) compared with white patients. Black patients consistently initiated hemodialysis with an AVG compared with Hispanic patients and white patients irrespective of nephrology care (Figure 2B).

Discussion

The use of an AVF for incident hemodialysis is well known to result in superior outcomes compared with an AVG or IHC.11 Previous findings suggest that minority populations may initiate hemodialysis with an AVF less frequently than white patients, presumably due to worsening disease status, comorbidities, or reduced access to medical care.9 In this study, we demonstrated that race/ethnicity is significantly associated with patients’ incident hemodialysis access modality in several ways. Overall, minority patients initiate dialysis with an AVG or IHC significantly more often than white patients. This observation is true despite the fact that black patients and Hispanic patients tend to be significantly younger and have fewer comorbidities (including coronary artery disease, chronic obstructive pulmonary disease, and cancer) than white patients. Current ESRD treatment strategies result in fewer functional AVFs for racial/ethnic minority patients receiving their first hemodialysis treatment compared white patients even when black patients and Hispanic patients receiving an AVG or IHC appear to be better surgical candidates. These differences persist despite statistical adjustment for demographic and comorbid characteristics.

The racial/ethnic disparities in incident AVF access that we describe deserve elucidation. The high rates of catheter use despite national programs to reverse this trend is unacceptable. A recent publication by our group identifies a policy gap among other contributors to the persistent use of an AVG and IHC at initial hemodialysis.11 However, inadequate policy, comorbidities, and medical insurance status do not account for the racial/ethnic disparities that we have shown. It is likely that clinical judgment informs surgeons when treating patients with ESRD because no unified algorithm exists to date to aid in selecting between hemodialysis access types. Black and Hispanic hemodialysis patients older than 40 years demonstrate better survival than white hemodialysis patients.12,13 Therefore, one argument for the more frequent use of an AVG or IHC in these groups may be that survival is better regardless of hemodialysis access type. However, minority populations are also less likely to receive kidney transplants than white patients,12,14 which makes the use of a permanent form of hemodialysis access that much more pertinent in these groups. It is possible that black patients and Hispanic patients with chronic kidney disease are simply progressing too quickly to ESRD15 to make AVFs a viable initial hemodialysis access option. Fistulas generally take up to 6 to 12 weeks to mature.16 Consistent with this notion, both black patients and Hispanic patients appear to be younger than white patients at the time of dialysis initiation. Certain comorbidities may also exert an influence on perceived operative risk or fistula maturation. Congestive heart failure, coronary artery disease, and peripheral arterial disease are well-known operative risk factors, with the latter 2 also impacting fistula maturation.17 However, the prevalence of these conditions in our study was actually lower among minority patients than among white patients. Diabetes mellitus has been shown to be associated with higher rates of primary fistula patency loss,18 and the prevalence of diabetes mellitus (particularly severe disease) was significantly higher in the black and Hispanic groups in our study.

Medical insurance status has been consistently shown to contribute to disparities in hemodialysis access type.9 Age effects associated with Medicare enrollment, the dominant medical insurance mode, likely contribute to some of the observed racial/ethnic variations in AVF use. The mean age in all subgroups for black patients and Hispanic patients within our study was younger than 65 years, whereas white patients tended to have a mean age older than 65 years, rendering them eligible for Medicare. Furthermore, black patients and Hispanic patients were more likely to have no medical insurance or Medicaid, a finding echoed by other large database studies.1921 Unfortunately, the use of Medicaid limits reimbursement for predialysis care, disincentivizing surgeons from establishing AVF access before a patient completes the prescribed 3-month period of dialysis required to establish an ESRD diagnosis and trigger related physician payments.22 This protocol likely limits the feasibility of establishing mature AVFs in many cases. Although we similarly describe some variation in hemodialysis access type stratified by medical insurance status, minority patients within our cohort consistently initiated dialysis using non-AVF access regardless of medical insurance status. In addition, disparities exist within specific medical insurance status subgroups. Within the uninsured population, Hispanic patients initiated hemodialysis with an IHC more frequently than both black patients and white patients. The persistence of disparities within groups with similar types of medical insurance suggests that there are social, cultural, or behavioral determinants of access to health care within these minority populations as stated by others.23 Certainly, language barriers within the Hispanic population probably have a role because barriers to health care are consistently reported among racial/ethnic and linguistic minority patients.24 Identification and understanding of these sociocultural and behavioral determinants provide room for further study.

We also demonstrated that black patients initiated hemodialysis with an AVG much more frequently than either white patients or Hispanic patients. Black patients have been shown to have smaller veins available for use as fistulas by ultrasonography. This finding might inform a surgeon’s decision to use a prosthetic rather than an autogenous conduit.25 An AVF is the preferred form of hemodialysis access over an AVG, as emphasized by a 2010 update to the Fistula First Breakthrough Initiative (change concept 3).26 Furthermore, the use of AVFs in the black population is feasible. For example, Ishaque et al25 successfully established functional AVFs in 74% of black patients in whom they were attempted. These data suggest that adequate intraoperative vein assessment, use of venous dilation and a proximal location, and careful follow-up can lead to long-term AVF patency even in patients in whom preoperative imaging findings are discouraging. The ongoing Hemodialysis Fistula Maturation Study27 may provide further insights into parameters associated with successfully maturing AVFs rather than resorting to AVGs, as will developments in the identification and triaging of high-risk patients with chronic kidney disease into a multidisciplinary ESRD care model.

Our propensity score analysis showed that nephrology care is perhaps the most significant factor influencing hemodialysis initiation by an AVF. Management by a nephrologist increased the odds of a patients’ beginning dialysis with an AVF more than 11-fold. Prior investigations have demonstrated a similar finding.28 We found that black and Hispanic populations without nephrology care initiate hemodialysis with an AVF more often than white patients (Figure 2A). The reverse occurred on receipt of nephrology care. It is alarming that AVF utilization decreases among black and Hispanic patients compared with white patients with increasing length of nephrology care. It can be argued that there is less receipt of nephrology care among minority patients.9 However, the disparity persisted after stratified analyses, propensity score matching, and adjustment for demographic and medical characteristics. The situation raises critical questions about the possible social and cultural contributors to this occurrence and deserves an urgent system review, considering the vital role nephrologists have in improving hemodialysis initiation with an AVF.29,30 As a result, it is imperative that the American Medical Association, American College of Surgeons, National Kidney Foundation, and Centers for Medicare & Medicaid Services must identify and address the reasons why minority populations do not receive the well-known benefits of this multidisciplinary approach to the care of patients with ESRD with regard to AVF placement.

The USRDS is a large database derived from the Centers for Medicare & Medicaid Services Form 2728, which is completed by a myriad of practitioners. Therefore, variability in diagnoses, patient demographics, and data completeness is expected. However, prior investigations have demonstrated reliability in diagnoses with the USRDS database.31 We acknowledge a potential for racial/ethnic heterogeneity, particularly because these are self-reported characteristics and because overlap between the 3 study groups is possible. However, our sample size is large enough such that small inaccuracies should not significantly affect the overall results. In addition, regional variation in hemodialysis access creation has been reported,28,32 which we have not considered in the present study. Furthermore, nephrologists and surgeons may preferentially select a particular hemodialysis access type for reasons that cannot be accounted for by a retrospective data set. It is not possible to adjust for all individual patient–based, system-based, and practice-based selection effects that may bias the USRDS data set. However, this preliminary description of racial/ethnic variations in incident hemodialysis access type can serve to initiate important conversations surrounding the presence of and potential reasons for racial/ethnic disparities in ESRD care and should raise interest in addressing areas that may be amenable to intervention with minority-based and similar initiatives in the future.

Conclusions

Black patients and Hispanic patients initiate dialysis using AVFs significantly less frequently than white patients despite their being younger and having fewer comorbidities. These disparities persisted independent of factors that drive health access for fistula placement, such as medical insurance status and nephology care. The sociocultural underpinnings of these disparities deserve investigation and redress to maximize the benefits of initiating hemodialysis via fistula in patients with ESRD irrespective of race/ethnicity.

Back to top
Article Information

Accepted for Publication: October 29, 2014.

Corresponding Author: Mahmoud B. Malas, MD, MHS, Division of Vascular Surgery, Department of Surgery, Johns Hopkins Medical Institutions, 4940 Eastern Ave, Ste A547, Baltimore, MD 21401 (bmalas1@jhmi.edu).

Published Online: April 29, 2015. doi:10.1001/jamasurg.2015.0287.

Author Contributions: Drs Zarkowsky and Malas had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.

Study concept and design: Zarkowsky, Canner, Freischlag, Malas.

Acquisition, analysis, or interpretation of data: Zarkowsky, Arhuidese, Canner, Obeid, Schneider.

Drafting of the manuscript: Zarkowsky, Arhuidese, Hicks, Canner, Obeid.

Critical revision of the manuscript for important intellectual content: Hicks, Canner, Qazi, Abularrage, Freischlag, Malas.

Statistical analysis: Arhuidese, Canner, Schneider.

Study supervision: Freischlag.

Conflict of Interest Disclosures: None reported.

Disclaimer: Dr Freischlag is the former editor of JAMA Surgery but was not involved in the editorial review or the decision to accept the manuscript for publication.

Previous Presentation: This study was presented at the 40th Annual Meeting of the New England Society for Vascular Surgery; September 28, 2013; Stowe, Vermont.

References
1.
U.S. Renal Data System. USRDS 2013 Annual Data Report: Atlas of Chronic Kidney Disease and End-Stage Renal Disease in the United States. Bethesda, MD: National Institute of Diabetes and Digestive and Kidney Diseases, National Institutes of Health; 2013.
2.
Foley  RN, Chen  SC, Collins  AJ.  Hemodialysis access at initiation in the United States, 2005 to 2007: still “catheter first.” Hemodial Int. 2009;13(4):533-542.
PubMedArticle
3.
DeSilva  RN, Sandhu  GS, Garg  J, Goldfarb-Rumyantzev  AS.  Association between initial type of hemodialysis access used in the elderly and mortality. Hemodial Int. 2012;16(2):233-241.
PubMedArticle
4.
Wasse  H, Speckman  RA, McClellan  WM.  Arteriovenous fistula use is associated with lower cardiovascular mortality compared with catheter use among ESRD patients. Semin Dial. 2008;21(5):483-489.
PubMedArticle
5.
Ravani  P, Palmer  SC, Oliver  MJ,  et al.  Associations between hemodialysis access type and clinical outcomes: a systematic review. J Am Soc Nephrol. 2013;24(3):465-473.
PubMedArticle
6.
Stehman-Breen  CO, Sherrard  DJ, Gillen  D, Caps  M.  Determinants of type and timing of initial permanent hemodialysis vascular access. Kidney Int. 2000;57(2):639-645.
PubMedArticle
7.
Reddan  DN, Szczech  LA, Klassen  PS, Owen  WF  Jr.  Racial inequity in America’s ESRD program. Semin Dial. 2000;13(6):399-403.
PubMedArticle
8.
Frankenfield  DL, Rocco  MV, Roman  SH, McClellan  WM.  Survival advantage for adult Hispanic hemodialysis patients? findings from the End-Stage Renal Disease Clinical Performance Measures project. J Am Soc Nephrol. 2003;14(1):180-186.
PubMedArticle
9.
Arce  CM, Mitani  AA, Goldstein  BA, Winkelmayer  WC.  Hispanic ethnicity and vascular access use in patients initiating hemodialysis in the United States. Clin J Am Soc Nephrol. 2012;7(2):289-296.
PubMedArticle
10.
Allon  M, Ornt  DB, Schwab  SJ,  et al; Hemodialysis (HEMO) Study Group.  Factors associated with the prevalence of arteriovenous fistulas in hemodialysis patients in the HEMO Study. Kidney Int. 2000;58(5):2178-2185.
PubMedArticle
11.
Malas  MB, Canner  JK, Hicks  CW,  et al.  Trends in incident hemodialysis access and mortality [published online March 4]. JAMA Surg. 2015.doi:10.1001/jamasurg.2014.3484.
12.
Rhee  CM, Lertdumrongluk  P, Streja  E,  et al.  Impact of age, race and ethnicity on dialysis patient survival and kidney transplantation disparities. Am J Nephrol. 2014;39(3):183-194.
PubMedArticle
13.
Kucirka  LM, Grams  ME, Lessler  J,  et al.  Association of race and age with survival among patients undergoing dialysis. JAMA. 2011;306(6):620-626.
PubMedArticle
14.
Arce  CM, Goldstein  BA, Mitani  AA, Winkelmayer  WC.  Trends in relative mortality between Hispanic and non-Hispanic whites initiating dialysis: a retrospective study of the US Renal Data System. Am J Kidney Dis. 2013;62(2):312-321.
PubMedArticle
15.
Derose  SF, Rutkowski  MP, Crooks  PW,  et al.  Racial differences in estimated GFR decline, ESRD, and mortality in an integrated health system. Am J Kidney Dis. 2013;62(2):236-244.
PubMedArticle
16.
Feldman  HI, Joffe  M, Rosas  SE, Burns  JE, Knauss  J, Brayman  K.  Predictors of successful arteriovenous fistula maturation. Am J Kidney Dis. 2003;42(5):1000-1012.
PubMedArticle
17.
Lok  CE, Allon  M, Moist  L, Oliver  MJ, Shah  H, Zimmerman  D.  Risk equation determining unsuccessful cannulation events and failure to maturation in arteriovenous fistulas (REDUCE FTM I). J Am Soc Nephrol. 2006;17(11):3204-3212.
PubMedArticle
18.
Ocak  G, Rotmans  JI, Vossen  CY,  et al.  Type of arteriovenous vascular access and association with patency and mortality. BMC Nephrol. 2013;14:79.
PubMedArticle
19.
Propper  B, Black  JH  III, Schneider  EB,  et al.  Hispanic ethnicity is associated with increased costs after carotid endarterectomy and carotid stenting in the United States. J Surg Res. 2013;184(1):644-650.
PubMedArticle
20.
Sonel  AF, Good  CB, Mulgund  J,  et al; CRUSADE Investigators.  Racial variations in treatment and outcomes of black and white patients with high-risk non–ST-elevation acute coronary syndromes: insights from CRUSADE (Can Rapid Risk Stratification of Unstable Angina Patients Suppress Adverse Outcomes With Early Implementation of the ACC/AHA Guidelines?). Circulation. 2005;111(10):1225-1232.
PubMedArticle
21.
Epstein  AJ, Gray  BH, Schlesinger  M.  Racial and ethnic differences in the use of high-volume hospitals and surgeons. Arch Surg. 2010;145(2):179-186.
PubMedArticle
22.
Centers for Medicare & Medicaid Services. Medicare coverage of kidney dialysis & kidney transplant services. 2012. http://www.medicare.gov/pubs/pdf/10128.pdf. Accessed June 1, 2013.
23.
Powe  NR, Melamed  ML.  Racial disparities in the optimal delivery of chronic kidney disease care. Med Clin North Am. 2005;89(3):475-488.
PubMedArticle
24.
Weech-Maldonado  R, Morales  LS, Elliott  M, Spritzer  K, Marshall  G, Hays  RD.  Race/ethnicity, language, and patients’ assessments of care in Medicaid managed care. Health Serv Res. 2003;38(3):789-808.
PubMedArticle
25.
Ishaque  B, Zayed  MA, Miller  J,  et al.  Ethnic differences in arm vein diameter and arteriovenous fistula creation rates in men undergoing hemodialysis access. J Vasc Surg. 2012;56(2):424-431.
PubMedArticle
26.
Sales  CM, Schmidt  RJ; FFBI Clinical Practice Workgroup.  Fistula First clarifies its message. J Vasc Surg. 2010;52(6):1699-e1. doi:10.1016/j.jvs.2010.10.038.
PubMedArticle
27.
Dember  LM, Imrey  PB, Beck  GJ,  et al.  Objectives and design of the Hemodialysis Fistula Maturation Study. Am J Kidney Dis. 2014;63(1):104-112.
PubMed
28.
Wasse  H, Hopson  SD, McClellan  W.  Racial and gender differences in arteriovenous fistula use among incident hemodialysis patients. Am J Nephrol. 2010;32(3):234-241.
PubMedArticle
29.
Kazmi  WH, Obrador  GT, Khan  SS, Pereira  BJ, Kausz  AT.  Late nephrology referral and mortality among patients with end-stage renal disease: a propensity score analysis. Nephrol Dial Transplant. 2004;19(7):1808-1814.
PubMedArticle
30.
Kessler  M, Frimat  L, Panescu  V, Briançon  S.  Impact of nephrology referral on early and midterm outcomes in ESRD: Epidémiologie de l’Insuffisance Renale Chronique Terminale en Lorraine (EPIREL): results of a 2-year, prospective, community-based study. Am J Kidney Dis. 2003;42(3):474-485.
PubMedArticle
31.
Longenecker  JC, Coresh  J, Klag  MJ,  et al.  Validation of comorbid conditions on the end-stage renal disease Medical Evidence Report: the CHOICE study: Choices for Healthy Outcomes in Caring for ESRD. J Am Soc Nephrol. 2000;11(3):520-529.
PubMed
32.
Hopson  S, Frankenfield  D, Rocco  M, McClellan  W.  Variability in reasons for hemodialysis catheter use by race, sex, and geography: findings from the ESRD Clinical Performance Measures Project. Am J Kidney Dis. 2008;52(4):753-760.
PubMedArticle
×