Outpatient Dialysis for Patients with AKI: A Policy Approach to Improving Care

Abstract

The rate of AKI requiring dialysis has increased significantly over the past decade in the United States. At the same time, survival from AKI seems to be improving, and thus, more patients with AKI are surviving to discharge while still requiring dialysis. Currently, the options for providing outpatient dialysis in patients with AKI are limited, particularly after a 2012 revised interpretation of the Centers for Medicare and Medicaid Services guidelines, which prohibited Medicare reimbursement for acute dialysis at ESRD facilities. This article provides a historical perspective on outpatient dialysis management of patients with AKI, reviews the current clinical landscape of care for these patients, and highlights key areas of knowledge deficit. Lastly, policy changes that have the opportunity to significantly improve the care of this at-risk population are suggested.

Introduction

Between 2000 and 2009, the incidence of AKI requiring dialysis (AKI-D) in the United States rose at a rate of 10% per year, and the absolute number of patients more than doubled (1). The development of AKI-D is associated with hospital mortality rates as high as 50% depending on the clinical setting (2–4). Research has focused primarily on improving the short-term (in-hospital) outcomes of these patients (5,6), and a recent meta-analysis suggests that overall AKI survival may be improving (4). As more patients with AKI-D survive to hospital discharge, some attention has shifted to the long-term management of this population, which remains at significant risk for adverse outcomes (7–9).

Among survivors of AKI-D, between 10% and 30% still require dialysis at hospital discharge (3,6,10,11). The care of these patients is challenging because of both the acute nature of their condition and the significant logistic barriers, such as appropriate outpatient dialysis placement. The latter has emerged as a major issue since the clarification of reimbursement rules by the Centers for Medicare and Medicaid Services (CMS) in 2012, which prohibit Medicare reimbursement for acute dialysis at ESRD facilities.

In an accompanying article (12), we review the clinical implications of renal function recovery in patients with AKI-D and discuss best practices to promote recovery of renal function. Here, we examine the logistic barriers to the care of patients with Medicare and AKI-D who still require RRT at hospital discharge, discuss the implications of current policy in this arena, and present potential policy changes that may improve the care of this at-risk group.

Historical Perspective

In October of 1972, Section 299I of Public Law 92–603 created the US National ESRD Program. This law extended Medicare coverage to individuals with ESRD who require either dialysis or transplantation to maintain life (13). This law was not intended to provide coverage for outpatient dialysis for patients with AKI. However, Medicare pays for acute dialysis under the hospital outpatient prospective payment system (OPPS) when it is furnished to a hospital outpatient with ESRD in nonroutine circumstances (as defined in the Medicare Claims Processing Manual [Publication 100–04, Chapter 4, section 200.2]) or a non-ESRD hospital outpatient (defined as a patient returning to the hospital for ongoing care postdischarge).

Before July of 2012, Medicare recipients with AKI-D had several options for dialysis at the time of hospital discharge. They could return to a hospital-based dialysis unit, go to an outpatient ESRD facility near their home, or be placed at a long-term acute care unit, transition care hospital, or rehabilitation center that provided onsite dialysis. The placement decision was primarily on the basis of patient factors, such as acuity of current illness, comorbid conditions, physical deconditioning, and convenience. In 2009, the CMS stated that ESRD facilities are not restricted from providing dialysis to patients without ESRD but that the payment would be outside of the Medicare ESRD system. At that time, many hospitals entered into contractual agreements with ESRD facilities to provide outpatient dialysis for patients with AKI-D. The hospital billed Medicare and paid the ESRD facility according to the terms of the agreement.

The CMS provided a policy update on outpatient dialysis for AKI-D Medicare recipients with the publication of the 2012 OPPS on November 30, 2011 (14). Although the law had not changed, this document indicated that the CMS was revising their interpretation of Section 410.27 of their regulations broadly to include dialysis services in their definition of outpatient therapeutic services. On July 18, 2012, the CMS further clarified their policy on the provision of dialysis to patients with Medicare and AKI at ESRD facilities:

ESRD facilities cannot furnish acute dialysis to hospital outpatients. The Medicare regulations provide that all therapeutic services furnished to hospital outpatients whether directly or under arrangements, must be furnished in the hospital or in locations that qualify as provider-based departments of the hospital. Our regulations prohibit ESRD facilities from being provider-based departments of hospitals and while an ESRD facility may be located on a hospital’s campus and may share certain overhead costs and administrative functions, CMS does not consider it to be part of the hospital. Therefore, a hospital may not enter into an arrangement with an ESRD facility for the facility to provide, outside of the hospital, outpatient dialysis or any other therapeutic service for which the hospital would bill Medicare (15).

These clarifications effectively reversed the earlier position and resulted in hospitals no longer being able to obtain Medicare reimbursement for AKI-D payments to outpatient ESRD facilities. This has resulted in the restriction of an essential outpatient dialysis option for patients with Medicare and AKI-D, thus impeding the transition to outpatient dialysis for such patients.

Current State: Implications of the CMS Rule Clarification

Health System Perspective

Current options for outpatient dialysis for patients with Medicare and AKI remain limited (Table 1). Until the renal failure can be considered irreversible—a timeframe ranging from weeks to potentially months—these patients must either remain hospitalized or return to a hospital-based unit as outpatients for ongoing dialysis treatments. The first option obviously drives up health care costs; in 2013, the average cost in the United States was $4287 per hospital day according to the price report by the International Federation of Health Plans (16). It also deprives other patients requiring acute care of a much-needed hospital bed, further straining the health care system. Indeed, the National Renal Administrators Association commented to the CMS that “requiring AKI patients to receive dialysis in the hospital is both expensive and inefficient” (17). A third alternative for the hospital would be to discharge patients with AKI-D to a long-term acute care facility that performs dialysis therapy, provided that the patients have other acute medical issues. Most patients will not qualify for this level of service or may face limitations on duration of coverage, and the number of such facilities is limited.

Table 1.

Dialysis options for patients with Medicare and AKI requiring dialysis who are otherwise ready for hospital discharge

Dialysis Setting	Description	Advantages	Disadvantages
Inpatient	Prolong hospitalization while monitoring for renal function recovery	Continuity of care	Risk of nosocomial infections and other hospital–acquired complications
		Patient convenience	High cost and inefficient resource use
Hospital-based unit	Return as outpatient to hospital–based dialysis unit	Care in acute setting with expectations for recovery	Burden of potentially significant travel back to hospital
		Eligible for Medicare payment	Limited availability of such facilities compared with other options
			Higher costs compared with ESRD facility placement
Long–term acute care hospital or rehabilitation center	Onsite provision of dialysis	Convenience for patients who continue to have acute medical or rehabilitation needs	Few patients qualify
			Limited availability of such facilities compared with other options
Community–based ESRD facility	Placement in outpatient ESRD facility	More accessible and convenient for patients	Requires ESRD certification or single-payer agreement between hospital and facility (not billable to Medicare)
			Lack of AKI-specific protocols to promote recovery of renal function

To prevent prolonged costly inpatient stays, some hospitals have adapted by negotiating single-payer agreements with ESRD facilities. Although hospitals can no longer be reimbursed by Medicare for these payments, this approach is associated with cost savings compared with extended hospitalizations. However, even with this option, not all outpatient ESRD facilities will accept patients with AKI-D, and not all hospital systems are willing to enter into these agreements. Furthermore, in some states, the guidelines have been interpreted as prohibiting any acute dialysis in ESRD facilities.

For hospital-based units that also have outpatient ESRD certification, it may be necessary to decertify some previously designated outpatient ESRD chairs and reallocate these specifically for outpatients with AKI. This approach satisfies the CMS ruling that prohibits patients with AKI from dialyzing in an ESRD facility and thus, permits Medicare reimbursement for these patients under the OPPS. This approach is being increasingly adopted and provides a reasonable solution for many patients with AKI-D, particularly when those patients are local and can be managed with non-ESRD protocols. However, use of these designated outpatient AKI stations will be less predictable than with ESRD, and hospitals risk underuse and resource waste. Notably, between 2008 and 2013, the number of freestanding ESRD facilities increased, whereas hospital-based ESRD facilities decreased (18).

Most hospital-based units are not dually licensed as outpatient ESRD facilities and do not have formal clinical or administrative protocols in place for outpatient dialysis. For example, although renal dieticians and social workers may be available for inpatient care, they may not have experience working with outpatients. The increasing need to accommodate outpatients with AKI-D strains a resource that is primarily focused on providing dialysis to an acutely ill inpatient population. One practice is to readmit patients for dialysis in the acute inpatient dialysis unit. This poses significant administrative issues with regards to registration, pharmacy, transport, nursing, and other health personnel having to deal with multiple admissions per week for the same patient with same-day discharges. As the burden of outpatient dialysis for AKI increases, hospitals and health systems are forced to explore implementing new measures to care for this population. These measures may require adding resources, including dedicated personnel with expertise in dialysis for AKI, administrative support for outpatient acute dialysis billing, and designation of a physical space for outpatient dialysis.

Provider Perspective

Clinical judgment remains the sole determination for when renal failure is considered “irreversible and permanent” in signing the ESRD medical evidence report (the CMS Form 2728) (19). An unintended consequence of the CMS rule clarification has been that nephrologists are facing increasing pressure to certify patients with AKI-D as ESRD at the time of hospital discharge (Table 2). Doing so facilitates outpatient placement of these patients in an ESRD facility, which is often the most convenient option for the patient as well as the least costly disposition for the hospital. However, premature designation of patients with AKI-D as ESRD may lead to inadequate follow-up of renal function and constitute a potential lost opportunity to promote recovery of renal function. From a regulatory standpoint, this approach may be considered a fraudulent practice, with potential suspension of medical licensure or even criminal penalties.

Table 2.

Adverse consequences of the current Centers for Medicare and Medicaid Services policy prohibiting ESRD facilities from being reimbursed by Medicare for acute dialysis in patients with AKI

Perspective	Consequences of Current Policy
Hospitals and health care systems	Increased costs associated with either paying for outpatient dialysis at an ESRD facility or providing resources to support outpatient dialysis in a hospital-based unit
Health care providers	Increased pressure to certify patients with AKI requiring dialysis as having ESRD, thereby facilitating placement at outpatient ESRD facilities; knowingly providing ESRD certification when recovery of renal function is reasonably expected may constitute federal fraud
Patients	More limited outpatient dialysis placement options, which may lead to delays in hospital discharge
	Placement in a hospital-based facility may result in prohibitive travel requirements for patients and their families
	Premature certification of ESRD status may significantly decrease likelihood of recovery of renal function

Unfortunately, there are no well validated prediction tools to determine which patients with AKI-D will recover versus progress to ESRD. Several clinical prognostic factors have been identified, but there are no diagnostic tests, and the standard remains an observation period for recovery of renal function. Lack of renal function recovery after 90 days is generally accepted as meeting criteria for ESRD (20), although it is reasonable to declare patients with ESRD earlier on the basis of various risk factors, most notably preexisting CKD, older age, and/or presence of other comorbidities. In our expert opinion, there are insufficient data to recommend one over another approach. Given the complexities and heterogeneity of the population of patients with AKI-D, clinical judgment of the treating nephrologist should remain the standard for determination of ESRD status until evidence-based metrics become available. Additional research, such as development of predictive clinical models and discovery of novel biomarkers for renal function recovery, is clearly required in this area.

In addition to regulatory considerations, the management of outpatient dialysis for patients with AKI-D represents a clinically challenging area. In contrast to the protocol-driven nature of ESRD dialysis care, there are no guidelines or established metrics for patients with AKI-D, but this is a distinct population with specific clinical needs. At a minimum, outpatients with AKI-D should be monitored more closely for evidence of renal recovery, and efforts should be made to facilitate recovery of renal function. For example, in contrast to the standard practice of ultrafiltration challenges to determine dry weight in patients with ESRD, nephrologists might adopt a less aggressive approach to fluid removal in patients with AKI to avoid episodes of hypotension, which may delay or prevent renal function recovery (21,22). At the same time, this must be balanced against the risks associated with excessive fluid overload, such as pulmonary edema. Thus, management of patients with AKI-D can be significantly more time consuming compared with that of patients with ESRD.

Patient Perspective

Recent studies have described the short-term negative effect of AKI-D on patient quality of life (23–25), but the burden of continued outpatient dialysis on patients and their families remains largely unexplored. Unfortunately, each of the currently available options carries significant drawbacks for patients.

The approach of discharging a patient with AKI-D and having them return to the hospital-based unit for dialysis has the advantage of providing continuity of care. In addition, some hospital-based units have clinical protocols in place to more closely promote and monitor for recovery of renal function in this population. Although there are no controlled trials investigating the efficacy of such protocols, it stands to reason that dialyzing in a facility that specializes in the care of those recovering from AKI might improve outcomes. However, in some cases, such as when patients were transferred to a referral medical center for high-acuity care, the travel requirements to return to a hospital-based unit can be prohibitive. Patients with AKI-D who remain dialysis dependent are already pulled in multiple directions, with competing demands from physical rehabilitation needs, follow-up physician appointments, and laboratory testing—all superimposed on the need for three times per week dialysis. The added stress of traveling long distance versus having to find a temporary local living situation adds significantly to the burdens faced by these patients and their families.

Conversely, an early declaration of ESRD may facilitate the placement of a patient with AKI-D at a far more convenient ESRD facility but places the patient under the care of a nephrology team that may not be focused on the potential for recovery of renal function given the diagnosis of ESRD. A diagnosis of ESRD may also have a psychosocial effect on these patients (and families) if they believe that they are unlikely to recover renal function (26).

A last option is to extend an inpatient stay to monitor for recovery of renal function, but this unnecessarily exposes patients to additional complications, including nosocomial infections. Also, the inpatient setting may be less conducive to aggressive rehabilitation efforts and impair speed of overall recovery.

In an ideal world, patients with AKI-D being discharged would be afforded the option of dialysis placement at a facility close to home, where they would be monitored by a dialysis team with AKI expertise that recognizes the ongoing potential for renal function recovery. For many patients with Medicare, this is currently not a possibility.

Distinction between Patients with Medicare and Patients with Commercial Insurance

It is important to note that the CMS rule clarification only affects patients with Medicare, and patients with commercial insurance can receive dialysis therapy at ESRD facilities willing to accept patients with AKI. However, there are several reasons why an overall change to the CMS policy may be beneficial for the broader AKI-D outpatient population. First, the risk for AKI-D increases significantly with age, and the majority of patients are older than the Medicare age threshold of 65 years old (1–3); therefore, it is conceivable that Medicare is already the largest single payer for this group. Second, many dialysis centers do not currently accept patients with AKI-D, regardless of insurance. Reasons include the low frequency of patients with AKI-D per unit, the labor intensiveness required for their management, and the inadequate training for the staff to deal with their complex medical issues. A change in the CMS policy could spur centers to adopt more open admission policies for these patients. Third, a separate CMS reimbursement model would allow a more comprehensive examination of this patient population through mandatory data reporting, similar to the US Renal Disease System for ESRD. This would allow characterization of overall recovery rates from AKI-D, and, potentially, the identification of practices that increase the likelihood of recovery.

Perhaps most importantly, a change in the CMS policy could have a ripple effect, influencing the clinical care of all survivors of AKI requiring outpatient dialysis. Having a defined mechanism for outpatient AKI-D reimbursement would necessitate the establishment of specific clinical protocols for these patients. Currently, the vast majority of outpatient dialysis facilities do not have such protocols, and thus, patients with AKI-D are frequently managed as patients with ESRD. Unfortunately, this practice may lead to patients slipping through the cracks and being maintained on dialysis longer than necessary.

Policy Recommendations

Given the policy clarifications by the CMS, it is understandable how the landscape of dialysis provision for patients with AKI-D has changed in this country over the past couple of years. At best, it has become challenging to discharge patients with AKI-D and Medicare into the community. At worst, we are failing on all three fronts of the “triple aim” approach to health care: improving individual experience of care, improving population health, and reducing per capita health care costs (27). Under the current system, (1) individual patients with AKI may face the hardships of prolonged hospitalization or long travel distances for outpatient dialysis placement, (2) there is a lack of a population-level health approach, and (3) there are increased costs because of the reliance on hospital-based dialysis facilities. With certain changes to current policies, we could improve on cost, safety, and quality of care for this vulnerable population. Below, we make several specific recommendations on the basis of expert opinion to accomplish these goals.

Recognize AKI-D as a Distinct Clinical Population

Fundamentally, patients with AKI-D have a greater chance of kidney function recovery than the typical patient with ESRD starting dialysis. Thus, from a quality of care perspective, patients with AKI must be appropriately labeled as having AKI with the correct diagnosis codes; if patients are labeled as ESRD by Form 2728, then measures to promote kidney function recovery are unlikely to be used. We recommend the use of international classification of diseases, ninth revision code 584.9 to appropriately label patients as AKI. This also prevents patients with AKI from being added to the Medicare ESRD database and biasing analyses regarding complications of ESRD and recovery of renal function.

Methods to clearly distinguish patients with AKI from patients with ESRD must be developed. An area of great concern is that patients with AKI may be inappropriately perceived and managed as having ESRD, potentially delaying kidney function recovery. Approaches to distinguishing patients with AKI from patients with ESRD may include physical measures, such as designating a certain dialysis chair for AKI or identifying the dialysis machine with an AKI label. Patients and staff should have a clear expectation for kidney function recovery and need to be appropriately educated on the signs and symptoms suggestive of kidney function recovery.

Develop Specific Protocols for Patients with AKI-D

In contrast to the myriad recommendations and specific guidelines regarding the care of patients with ESRD, there are no specific guidelines for outpatient dialysis care in patients with AKI. Nonetheless, it is clear that management needs to be different between these two populations, and we recommend the development of standard operating procedures by individual hospital-based dialysis units and ESRD facilities who accept patients with AKI and commercial insurance (Table 3). Unfortunately, as reviewed in the accompanying paper (12), there remains a paucity of evidence to guide these protocols, and we render the following suggestions on the basis of expert opinion.

Table 3.

Recommendations for policy changes to improve the care of patients with AKI requiring outpatient dialysis

Policy Change	Description
Recognize AKI requiring dialysis as a distinct clinical entity and different from ESRD	Clinically and administratively distinguish patients with AKI requiring dialysis from patients with ESRD, with the implicit expectation for recovery of renal function
	Nephrologists must not feel pressured by hospital administrators or dialysis units to declare a patient as having ESRD
	Dialysis providers and patients need to be educated on signs of renal function recovery and approaches to limiting additional renal insults
Develop specific protocols for patients with AKI requiring dialysis	Clinical protocols that promote and monitor for renal function recovery should be used in patients with AKI
	Standard ESRD protocols should not be routinely implemented in patients with AKI
Develop a separate reimbursement system for patients with AKI requiring dialysis	Establish a distinct payment system for outpatient AKI dialysis care in ESRD facilities
	Establish reimbursement rates for AKI dialysis care that reflect a higher cost of care than for patients with ESRD because of increased monitoring needs (both physician assessments and laboratory testing)
	Formal recognition as a separately billed entity will encourage centers to develop AKI-specific protocols and procedures
	Coverage of outpatient dialysis by Medicare will permit characterization of this high–risk AKI population through significantly greater data capture
Prioritize research to improve recovery of renal function	Support research on therapeutics to promote recovery of renal function after AKI
	Support research to develop prediction tools or novel biomarkers for recovery of renal function after AKI requiring dialysis
	Support research aimed at establishing best practices for the promotion of recovery of renal function in patients with AKI requiring dialysis

Approaches to monitoring for kidney function recovery may include measurement of weekly creatinine and BUN before dialysis, asking patients to monitor and record urine output, and measuring 24-hour urea and creatinine clearance (28). We recommend that patients should be asked about increased urine output and that the dialysis staff should document the urine volume before each dialysis session similar to other vital signs. Simply acknowledging the potential for kidney function recovery among physicians, staff, and patients will decrease the odds of inadvertent prolongation of dialysis in patients with recovering kidney function.

The necessary frequency of physician assessment before dialysis in the outpatient setting is unknown. In the inpatient setting, nephrologists routinely perform an assessment for each dialysis treatment—whether this is necessary in the outpatient setting is unclear. Reason dictates, however, that one time per month physician assessment, which often occurs for patients with ESRD, would be inadequate for patients with AKI. Notably, this population is highly vulnerable for complications, including an increased risk for hospital readmissions (29). Even in the ESRD setting, more frequent provider (physician or extender) visits have been associated with improved clinical outcomes (30,31). Importantly, patients are often assessed by physician extenders (physician assistants or nurse practitioners) in ESRD facilities, but most such providers will not have significant experience or training in the care of patients with AKI. Therefore, we strongly recommend that physician assessments should be more frequent for outpatients with AKI requiring hemodialysis and/or that physician extenders caring for these patients should receive AKI-specific training.

Until additional data are available, we recommend that individual centers and dialysis organizations develop policies and standard operating procedures that best fit with local experience and judgment regarding the care of patients with AKI-D. Specific protocols that need to be addressed pertain to (1) frequency of laboratory monitoring, (2) use of erythropoiesis stimulating agents, (3) bone mineral metabolism management, (4) access placement, and (5) others. Raising awareness through the development of standard practices alone will improve care of patients with AKI-D. With such standards in place, better data regarding kidney function recovery in the outpatient setting may be gathered.

Develop a Separate Reimbursement System for Patients with AKI

From both patient and payer perspectives, development of a payment mechanism to allow outpatient dialysis of patients with AKI in ESRD facilities is highly desirable. From a patient perspective, maintaining AKI status distinct from patients with ESRD will promote care tailored to optimize the likelihood of recovery of renal function and discontinuation of dialysis. From a payer perspective, payment for acute dialysis in ESRD facilities will be less costly than acute dialysis in the hospital-based setting or paying for extended hospital stays. Furthermore, if optimizing AKI care results in fewer patients transitioning to ESRD in the long run, then there should be substantial downstream health care savings.

However, making this change is complex and will require thoughtful dialogue with the CMS and lawmakers; instead of another reinterpretation of the current regulations, an actual change in law may be required. Notably, the bipartisan bill Chronic Kidney Disease Improvement in Research and Treatment Act of 2015 (H.R. 1130, S.598) proposed on February 26, 2015, contains a provision for “expanding access for patients with acute kidney injury” by amending the social security act to essentially give patients with AKI the same benefits as patients with ESRD (32). Passage of this bill would certainly increase outpatient dialysis access for patients with AKI and Medicare, but details regarding implementation would still need to be worked out. As outlined above, we believe that a critical aspect is to maintain the distinction between the AKI and ESRD populations.

The reimbursement system must account for increased costs associated with patients with AKI-D compared with patients with ESRD, including need for more frequent testing for residual renal function. At the same time, it may be reasonable to establish a bundled payment system similar to the ESRD program to encourage centers to develop streamlined approaches and avoid overtesting (33). Also analogous to the ESRD program, a future goal may be to develop quality measures for the care of patients with AKI-D, which in turn, can be linked to reimbursement rates. Although the onus will remain on the nephrologist to declare someone ESRD, it is desirable to remove external logistic and financial pressures and allow this decision to be made purely on clinical grounds.

Prioritize Research in Recovery of Renal Function among Patients with AKI-D

An important goal of research should be to improve tools to predict recovery of renal function among patients with AKI-D. In particular, most studies to date have focused on the effect of acute in-hospital interventions on outcomes, and few have explored predictors of renal outcomes beyond acute hospitalization. Validated prediction tools can facilitate appropriate patient placement and monitoring, guide intermediate-term clinical decision-making, and provide patients with more realistic expectations. Prediction tools will also facilitate identification of appropriate risk groups for evaluation in clinical trials of interventions to promote renal function recovery.

A critical limitation to current management of patients with AKI-D in the outpatient setting is the lack of defined best clinical practices, particularly those that may promote recovery of renal function. One reason for this knowledge gap is the lack of large patient registries to describe outcomes among patients with AKI-D beyond hospital discharge. Development of a separate Medicare reimbursement pathway for outpatients with AKI-D (analogous to the CMS Form 2728 for patients with ESRD) would facilitate tracking of these patients at a national level and provide improved characterization of this population, at least among Medicare beneficiaries. The ability to assess the outcomes of these patients beyond hospital survival is a vital step toward identifying which practices may result in the highest likelihood of recovery of renal function. Ultimately, these practices will need to be tested in controlled clinical trials.

Conclusion

After a CMS rule clarification in 2012, Medicare has declined to provide payments for outpatient dialysis at ESRD facilities for patients with AKI-D. As a result, these patients face limited options, each associated with major challenges and pitfalls. Patients with AKI-D who survive to hospital discharge remain a vulnerable population, and there exists significant opportunities to improve their care and overall quality of life. Changes in the CMS policy, particularly relating to dialysis treatment options and reimbursement, have the potential to lead the way forward and guide much-needed improvements to the care of patients with AKI-D.

Disclosures

None.