Care of Undocumented Individuals With ESRD: A National Survey of US Nephrologists

Results 

Survey Response and Respondent Characteristics 

Of 1,723 eligible nephrologists in 50 states, the District of Columbia, and 1 territory (Puerto Rico), 990 returned completed surveys, yielding an overall response rate of 57%. Three respondents removed identification numbers from their paper surveys and therefore lacked demographic information to include them in the comparison of respondents and nonrespondents. Respondents and nonrespondents did not differ significantly by sex (P = 0.5), age (P = 0.2), region of the country where they practice (P = 0.3), or representation of states known to have a high undocumented population (P = 0.3). Information regarding physician and practice characteristics for respondents is listed in Table 1. Ninety-two percent of respondents practiced clinical nephrology at least 50% of the time.

Table 1. Survey Respondent Characteristics

	Region (%)
	South	33
	Northeast	26
	West	20
	Midwest	21
	Years in practice (%)
	<5	18
	5-9	19
	10-14	16
	15-19	12
	≥20	35
	Race/ethnicity (%)
	White/non-Hispanic	62
	Asian	23
	White/Hispanic	6
	Black	3
	Other/unknown	7
	Location (%)
	Large metropolitan area	57
	Medium-sized metropolitan area	25
	Small/rural town	18
	Practice setting (%)
	Inpatient
	Public sector	34
	Private sector	66
	Not for profit	76
	For profit	24
	Outpatient
	Public sector	18
	Private sector	82
	Not for profit	28
	For profit	72

Nephrologists Who Care for Undocumented Patients With ESRD 

Sixty-five percent (642 of 990) of nephrologists surveyed reported having performed dialysis on undocumented patients with ESRD in the past year. Of these, 64% had cared for fewer than 5 patients, 26% had cared for 5 to 10 patients, 5% had cared for 11 to 19 patients, and 5% had cared for 20 patients or more. Nephrologists represented 44 states and the District of Columbia. In 35 states, the majority of respondents reported having performed dialysis on undocumented patients with ESRD in the past year. As listed in Table 2, providing dialysis therapy to undocumented patients with ESRD was positively associated with practice in a high undocumented population state in multivariate analysis. Characteristics negatively associated with providing dialysis therapy to undocumented patients with ESRD were being from the Northeast or a small town/rural area in multivariate analysis.

Table 2. Characteristics Associated With Providing Dialysis in the Past 12 Months to Undocumented Patients With End-Stage Renal Disease

		Care for Undocumented	Unadjusted Odds Ratio (95% confidence interval)⁎	P	Adjusted Odds Ratio (95% confidence interval)†	P
	Region			0.005		0.1
	South	191(66)	0.49(0.32-0.76)		0.69(0.44-1.10)
	Northeast	151(65)	0.47(0.30-0.74)		0.55(0.34-0.88)
	West	140(80)	Reference		Reference
	Midwest	122(66)	0.49(0.31-0.79)		0.65(0.39-1.07)
	Type of state			<0.001		0.002
	High undocumented prevalence‡	289(76)	1.87(1.40-2.52)		1.67(1.21-2.30)
	All other states	315(62)	Reference		Reference
	Practice setting inpatient
	Public sector	185(66)	Reference		—
	Private sector	395(71)	1.26(0.93-1.72)	0.1	—
	Not for profit	435(69)	Reference		—
	For profit	145(70)	1.09(0.78-1.54)	0.6	—
	Practice setting outpatient
	Public sector	97(61)	Reference		Reference
	Private sector	507(70)	1.50(1.05-2.13)	0.03	1.45(1.00-2.11)	0.05
	Not for profit	156(63)	Reference		—
	For profit	446(70)	1.36(1.00-1.85)	0.05	—
	Practice location§			<0.001		<0.001‖
	Large metropolitan area	376(74)	Reference		Reference
	Medium-sized city	161(73)	0.94(0.65-1.34)		0.99(0.69-1.43)
	Small town/rural	67(41)	0.24(0.17-0.35)		0.27(0.18-0.40)

Note: Values expressed as number (percent) unless noted otherwise.

⁎ Characteristics associated with the outcome at P < 0.25 in bivariate analyses along with variables considered clinically significant were tested in multivariate models. † All variables listed were tested in multivariate models by using a backwards elimination procedure in which the least significant predictor in the model was eliminated sequentially. Only factors with P < 0.05 after adjustment for other significant factors were retained in the model presented. ‡ California, Texas, New York, Florida, Illinois, New Jersey, Arizona, and North Carolina, according to the Pew Hispanic Center. § Large metropolitan area, population greater than 250,000; medium-sized city, population of 100,000 to 250,000; and small town/rural, population less than 100,000. ‖ P value applies to odds ratio for small town/rural.

Many respondents had multiple methods of providing dialysis care to undocumented patients. Of nephrologists caring for undocumented patients with ESRD, 67% reported thrice-weekly outpatient dialysis therapy at a dialysis center, whereas 3% reported dialysis for patients fewer than 3 times/week. Fourteen percent of nephrologists reported providing peritoneal dialysis therapy. Four percent reported that they provide hospital-based dialysis, but it was unclear whether it was inpatient or outpatient. Of nephrologists providing undocumented ESRD care, 59% provided emergent dialysis care, 57% reported providing emergent dialysis in the hospital, and 11% reported that they provide emergent dialysis in the emergency department. Of nephrologists reporting emergent dialysis care, 28% reported undocumented patients receive only emergent care.

Prevalence of Undocumented Patients With ESRD 

Figure 1 shows nephrologists' perceptions regarding the prevalence of undocumented patients with ESRD in their respective states. Sixty percent of respondents had the impression that the prevalence is increasing.

View Large Image Download to PowerPoint

Figure 1. Level of agreement that the number of undocumented patients with end-stage renal disease is increasing in the nephrologist's state.

Access to Dialysis Care for Undocumented Patients With ESRD 

When asked whether undocumented patients with ESRD have adequate access to health care in their state, 44% disagreed, 41% agreed, and 15% did not know. Perceptions regarding access to dialysis care for undocumented patients with ESRD are shown in Fig 2. Seventy-six percent definitely agreed and 15% somewhat agreed that undocumented patients have access to some emergent dialysis therapy in their state; whereas only 24% definitely agreed and 27% somewhat agreed that such patients have access to some maintenance dialysis therapy.

View Large Image Download to PowerPoint

Figure 2. Level of agreement that some maintenance and emergent dialysis therapies are available to undocumented patients with end-stage renal disease.

Table 3 lists characteristics associated with agreeing that undocumented patients with ESRD have access to long-term dialysis therapy according to bivariate and multivariate analyses. Of the respondents, 137 (14%) were unaware whether undocumented patients had access to long-term dialysis therapy. Given the lack of knowledge regarding this main outcome, these respondents were excluded from this analysis. Of note, a separate sensitivity analysis was performed including the “don't know” answers as “no” (data not shown), with similar associations. The characteristic positively associated with agreeing that undocumented patients have access to long-term dialysis therapy was being from a state with a high undocumented population, and being from the southern region of the United States was negatively associated with reporting access to long-term dialysis therapy. When we added state as a random effect, the covariance parameter for state was not significant (P > 0.2), suggesting there was no significant clustering of responses within states and that individual state policies alone are not causing the finding that nephrologists from states with a high undocumented population are more likely to report access to long-term dialysis therapy.

Table 3. Characteristics Associated With Agreeing That Undocumented Patients With ESRD Have Access to Long-term Dialysis Therapy

		Agree Patients Have Access to Long-term Dialysis n (%)	Unadjusted Odds Ratio (95% confidence interval)⁎	P	Adjusted Odds Ratio (95% confidence interval)†	P
	Region			<0.001‡		<0.001§
	South	46(16)	0.32(0.21-0.50)		0.37(0.24-0.57)
	Northeast	72(34)	0.84(0.55-1.26)		0.89(0.58-1.35)
	West	69(38)	Reference		Reference
	Midwest	41(25)	0.55(0.34-0.87)		0.74(0.45-1.20)
	Type of state
	High undocumented prevalence‖	134(35)	2.09(1.54-2.85)	<0.001	1.91(1.37-2.66)	<0.001
	All other states	94(21)	Reference		Reference
	Practice setting inpatient
	Public sector	68(26)	Reference
	Private sector	145(28)	1.09(0.78-1.53)	0.6
	Not for profit	167(28)	Reference
	For profit	47(24)	0.83(0.57-1.20)	0.3
	Practice setting outpatient
	Public sector	41(31)	Reference
	Private sector	178(28)	0.84(0.56-1.26)	0.4
	Not for profit	65(32)	Reference
	For profit	154(27)	0.77(0.55-1.10)	0.2
	Practice location¶			0.1
	Large metropolitan area	132(26)	Reference
	Medium-sized city	68(32)	1.32(0.92-1.87)
	Small town/rural	28(23)	0.82(0.52-1.31)

Note: Values expressed as number (percent) unless noted otherwise. Overall, 137 respondents (14%) did not know whether undocumented patients had access to long-term dialysis therapy and were excluded from further analysis.

⁎ Characteristics associated with the outcome at P < 0.25 in bivariate analyses along with variables considered clinically significant were tested in multivariate models. † All variables listed were tested in multivariate models by using a backwards elimination procedure in which the least significant predictor in the model was eliminated sequentially. Only factors with P < 0.05 after adjustment for other significant factors were retained in the model presented. ‡ P value applies to odds ratios for South and Midwest. § P value applies to odds ratio for South. ‖ California, Texas, New York, Florida, Illinois, New Jersey, Arizona, and North Carolina, according to the Pew Hispanic Center. ¶ Large metropolitan area, population greater than 250,000; medium-sized city, population of 100,000 to 250,000; and small town/rural, population less than 100,000.

Seven percent (72 of 969) of respondents had advised undocumented patients to relocate to another state and 24% (230/967) of respondents had advised patients to relocate to another country within the past 12 months because they could not receive the necessary care. Additionally, 13% (127 of 966) had suggested to referring physicians that their undocumented patients with ESRD might consider relocation to another state or country with the past 12 months.

Qualitative Analysis 

Two hundred physicians (20% of the sample) responded when provided an open-ended comment section at the end of the survey. Respondents to this solicitation were similar to nonresponders except they were slightly older (mean age, 51.9 versus 49.6 years; P ≤ 0.01) and more likely to work at a not-for-profit hospital (85% versus 73%; P < 0.001). Responses were categorized qualitatively into 5 predominant domains regarding care of undocumented patients with ESRD: the impossibility of obtaining outpatient long-term dialysis therapy, lack of reimbursement for dialysis care, certain state Medicaid policies that cover outpatient long-term dialysis care, lack of access to renal transplantation, and a growing prevalence of undocumented patients with ESRD trying to access dialysis care. These domains are listed in Table 4 with representative quotes.

Table 4. Major Conceptual Domains Identified in Invited Comments With Illustrative Quotes

	Domain	No. of Respondents	Illustrative Quotes
	Impossibility of obtaining outpatient long-term dialysis	39	I run a pediatric dialysis unit. We receive no compensation for our undocumented patients (8 currently). When they become adults, no one will accept them..... Eventually, our pediatric dialysis unit in a children's hospital will have a dialysis program filled with adult, undocumented patients. I dialyze these patients in the hospital dialysis units because the medical directors of the outpatient units won't accept them.
	Lack of reimbursement for dialysis care	26	I will provide dialysis to all those who medically require it. However, the absence of resources to fund this can severely tax my practice (and that of my colleagues) and tax the dialysis unit and hospital. Taking care of these individuals is often difficult and frustrating. Definitely related to lack of reimbursement (not just physicians, but hospitals, ancillary services, etc).
	Certain state Medicaid policies that cover outpatient long-term dialysis care	19	Our outpatient units require insurance prior to instituting outpatient dialysis. Undocumented patients get emergency Medicaid for dialysis-related issues only and this allows us to dialyze them. Undocumented patients are usually started on dialysis in the hospital. They often stay in the hospital until they can obtain emergency Medicaid in (state) which covers only dialysis services. Then they are placed to the outpatient center.
	Transplantation is not available to undocumented patients with ESRD	15	My undocumented patients, although some excellent transplant candidates, have to come up with $75,000 before they are considered for transplantation despite this being a much “cheaper” intervention in the long run. About 40%-50% of the patients we dialyze are undocumented. Many are young and would benefit from a transplant but are not eligible due to status.
	There is a growing prevalence of undocumented patients with ESRD trying to access care	11	Huge problem and getting worse. This is a growing problem in our area.

Abbreviation: ESRD, end-stage renal disease.

Discussion 

To our knowledge, this is the first study to evaluate the care of undocumented patients with ESRD from a national perspective. The data suggest that although a greater portion of dialysis care to undocumented patients with ESRD is provided by nephrologists in states with greater undocumented populations, this issue is clearly national in scope, larger than previously appreciated,6 and, from the perspective of physicians performing dialysis, appears to be increasing. Our findings are consistent with changing immigration patterns resulting in greater dispersion of recent immigrants across the United States.12 Whereas ESRD care of undocumented immigrants was previously believed to fall exclusively to public hospitals,8, 15 we found no significant associations between practice setting and report of providing dialysis to undocumented patients with ESRD. A majority of nephrologists agreed that ESRD care is inadequate for this population, and one-third reported that patients with ESRD have access to emergent dialysis only. Despite high reported rates of uncompensated and inadequately reimbursed care, of nephrologists caring for the undocumented population, 67% reported that patients with ESRD have access to long-term dialysis therapy.

The finding that two-thirds of providers are able to provide long-term dialysis therapy seems to contrast with information gleaned from the qualitative data regarding extreme barriers to successful referral of patients to outpatient dialysis units. This may be partially explained by emergency Medicaid policies in certain states allowing for reimbursement for outpatient long-term dialysis.16 Some states, particularly those more experienced with immigrant populations, have determined that eliminating public funding for care of the undocumented in certain circumstances is more costly than funding such care. For example, in 2000, California determined that elimination of public funding for prenatal care of the undocumented would prove far more costly to taxpayers by substantially increasing low-birth-weight, prematurity, and postnatal costs.17 The Renal Physicians Association (RPA) has advocated that all undocumented individuals be eligible for emergency Medicaid services if they do not have insurance or resources to pay for dialysis.16 Still, a greater than expected number of nephrologists who reported caring for undocumented patients with ESRD reported that undocumented patients have access to long-term dialysis therapy. One explanation might be that other entities (such as counties, charities, and hospitals) are subsidizing outpatient care to avoid expensive inpatient care. A small portion of the undocumented population may have sufficient medical insurance and therefore more readily access options for long-term dialysis therapy.

Access to long-term dialysis therapy for undocumented patients with ESRD may be changing. A prior study9 suggested that the limited number of undocumented patients with ESRD in certain areas may have permitted local and state funds to cover the cost of outpatient maintenance dialysis care (∼$69,000/person/y).18 However, such programs are terminating because of inadequate resources to cover the increasing undocumented ESRD patient population,9, 16 resulting in greater rates of emergent care only (28% observed in this study). Greater rates of emergent dialysis care are concerning given the greater risk of associated complications, hospitalization mortality,19 and costs. Comparing undocumented patients with ESRD who received emergent versus long-term care, Sheik-Hamad9 found that the emergent care group required more transfusions (24.9 versus 2.2) and had more emergency department visits (26.3 versus 1.4), more inpatient days (162 versus 10.1), more intensive care unit days (6.1 versus 1.5), and more hospitalizations (12.8 versus 1.1) compared with a maintenance peritoneal dialysis group. Emergent dialysis care is more often (or exclusively) administered through catheters instead of arteriovenous fistulas or grafts, and prior studies have shown that catheter dialysis access is associated with greater rates of infections and mortality compared with the other forms of vascular access.20, 21, 22, 23 As a result of high associated morbidity and location of care in emergency departments and hospitals, recurrent emergent dialysis therapy is likely to be associated with greater morbidity and cost than long-term dialysis therapy. However, this financial trade-off has not been studied.

This study is particularly timely given the ongoing debate about immigration issues in the United States.24, 25, 26, 27, 28, 29, 30, 31, 32, 33 Some investigators and some of our respondents would favor sending all undocumented patients back to their country of origin and argue that immigrants come here to reap the benefits of our strong economy, particularly our health care system.34 This notion is refuted by the literature showing that foreign-born, especially undocumented, persons use fewer medical services and contribute less to health care costs in relation to their population share.35, 36, 37 Coritsidis et al8 found that most patients with ESRD who had been in the United States for more than 5 years were unaware of having a kidney problem before presentation and were employed. Referring patients with ESRD back to their home country, Mexico in most instances, is problematic for several reasons. First, although renal replacement therapy in Mexico has grown from 140 in 1992 to greater than 450 recipients/million in 2006, dialysis accessibility in Mexico still pales in comparison to the situation in the United States.38 Also, when dialysis is accessible in Mexico, mortality may be greater as a result of late presentation and lack of primary care infrastructure. Garcia-Garcia et al39 showed that adjusted mortality was 3-fold greater in peritoneal dialysis patients in Jalisco, Mexico, compared with uninsured Hispanic peritoneal dialysis patients in the United States. Leaving the United States would often mean leaving established support networks of friends, families, and employers. Last, some believe our federal government has an “ethical and fiscal responsibility to provide care for patients within our borders.”16 For these reasons, some argue that US solutions should be sought for caring for undocumented patients with ESRD.7, 8

Enhanced pre-ESRD care might delay the need for renal replacement therapy and thus curb costs and would be one means of addressing the growing problem of providing long-term dialysis therapy to undocumented patients with ESRD. Almost a decade ago, the RPA suggested that “optimal health maintenance for non-citizens with renal failure will invariably reduce the need for costlier hospitalizations and invasive procedures.”16 Proactive screening for chronic kidney disease and early preventive health care may decrease the incidence of ESRD and reduce the need for dialysis coverage. Current studies have shown that most undocumented patients with ESRD are young (40s) when they present,8, 9 meaning they might require many years of renal replacement therapy. Our qualitative analysis suggests that some nephrologists believe providing renal transplants to such patients would be not only better for the patients, but a less costly intervention longitudinally. Another avenue to investigate would be whether a national policy to provide compensation for long-term dialysis therapy for undocumented patients with ESRD would not only enable improved care, but also decrease costs of providing dialysis care by decreasing the frequency of emergent dialysis. The RPA also recognized the variability of nephrological care for undocumented individuals across the states and called for a national approach to the problem.16

This study has strengths and several limitations. This is the first national survey to assess the issue of providing dialysis to undocumented patients with ESRD, with a superior response rate (57%) to most national surveys of similar size.40 ASN members were chosen to survey because greater than 90% of nephrologists in the United States are ASN members. Nonetheless, this study was subject to selection and information biases inherent to survey methods. Nephrologists who are passionate about this topic may have been the predominate responders and thus responses may not be entirely generalizable. Of those treating undocumented patients with ESRD, the rate of outpatient dialysis therapy (67%) may be an overestimate of actual practice. Also, undocumented patients might have been difficult to identify for some respondents given that nephrologists may be less focused on this specific demographic attribute. Last, some respondents work for different types of institutions (for profit and not for profit or public and private), and although they were instructed to answer with regard to their primary hospital and/or outpatient dialysis unit, their answers may represent a blend of their experience in different institutions.

The high rates of emergent care and uncompensated or inadequately reimbursed dialysis care for undocumented patients with ESRD found in this study are unlikely to be sustainable. The burden of care is falling to individuals, local governments, and states, entities that have little say in immigration policies.16 In this context, it seems reasonable to establish federal policy regarding delivery of dialysis services to undocumented individuals in the United States. Such a policy could be informed by the RPA position statement on dialysis therapy for noncitizens.16 Health care providers and policymakers need to better communicate and assess this changing landscape and explore methods to mitigate this increasing health care concern.