Integrated Care in ESKD

Introduction

ESKD has been a vanguard care environment, involving significant federal regulation, capitated payments, and nationally collected data in advance of other specialty areas of medicine. The care model, however, has remained constant. Patients with ESKD dialyze in center or at home and interact with their primary physician, most often their nephrologist, for appropriate treatment. The dialysis clinic could be a site of care coordination, if not broader care delivery, but it is constrained due to safety, regulatory, and reimbursement factors. However, features of ESKD make it logical to consider an expanded care model. Patients have an interdisciplinary care team inside the dialysis clinic, and the population is overwhelmingly aligned with a single payer: Medicare. Finally, the cost of ESKD (approximately 7% of Medicare expenses while representing only 1% of all Medicare beneficiaries) (1) suggests that improved outcomes could be combined with gains in efficiency and cost-effectiveness if ESKD care can be changed.

Integrated Kidney Care Today

There are two templates for integrated kidney care in the United States today. In certain locations, individuals with ESKD can enroll in a chronic condition special needs plan (C-SNP). These are Medicare-managed care plans with specific benefits and provider networks centered on dialysis care. Patients with ESKD in C-SNPs have lower hospitalization rates and fewer central venous catheters compared with other Medicare beneficiaries with ESKD and annual cost savings that can reach 15% of nondialysis-related expenses (2).

With the advent of the Medicare Shared Savings Program, Accountable Care Organizations (ACOs) enrolled a subset of patients with ESKD. Anecdotal reports indicated that ACOs with patients with ESKD decreased per patient medical spend (3,4). Nonetheless, the patient number was small. A different ESKD-focused shared savings vehicle, the Comprehensive ESKD Care (CEC) Model with ESRD Seamless Care Organizations (ESCOs) (5), borrowed heavily from the ACO construct. Clinicians and providers working in the ESCO are responsible for clinical, quality, and financial outcomes. These are measured not only internally by the ESCO but also, via a common list of quality measures distinct from and in addition to dialysis performance measures as well as Medicare Part A and B spending for the ESCO patients. Patient attribution is through the dialysis facility unlike ACOs, where attribution is through the principal care provider. Four providers implemented 13 ESCOs in the first year of the program. They uniformly delivered reductions in Part A and B Medicare spending totaling >$75 million, with decreased hospitalizations, central venous catheter use, and patient office visits (6).

Assessing Today’s Models—Why It Is Time for a Breakthrough in ESKD Care

Despite expansion of the CEC Model and C-SNPs, both models have shortcomings as does having patients with ESKD in ACOs. The opportunity to scale C-SNPs for patients with ESKD is limited. First, C-SNPs require a payer partner, and thus, they have inherent constraints on the basis of the payer network. Second, C-SNPs engage patients through a specific enrollment process rather than dialysis center attribution, resulting in 5000 patients with ESKD enrolled in C-SNPs or only 1% of the total number of individuals enrolled in all C-SNPs in the United States today (7). Third, C-SNPs often focus on performance measures, such as Healthcare Effectiveness Data and Information Set measures, which are more consistent with managed care entities and may not have the same degree of applicability in the ESKD population. In the CEC Model, nephrologists are required to take on significant financial risk and are not able to participate in shared savings or other key features of the program without being a direct owner in the ESCO. Additionally, geographies with preexisting care efficiency and lower utilization of health care services are penalized in the form of a lower financial benchmark, which is the basis for determining shared savings. Fourth, given the complexity of incorporating the entire disease spectrum for CKD and ESKD, existing models focus on dialysis care.

Significant changes to either program would require governmental action. For C-SNPs, the program is bound by a payer network, limiting growth and obligating patients to change their network of care for services. To the extent the C-SNP can offer a wider network, this might encourage or facilitate maintenance of enrollment. For ESCOs, a significant challenge is the escalating economic hurdles in the financial model as the ESCO progresses over time. This compresses the shared savings opportunity for all participants and may serve as a disincentive. In addition, frequent changes in the financial methodology during performance years pose a significant challenge.

Existing models do not rule out the need to test and innovate with new care models that could more easily include all patients with ESKD and the clinicians who care for them. If there was an opportunity to combine positive features of each model and reduce some of their inherent shortcomings, this could advance integrated kidney care even further and increase the likelihood of achieving the quadruple aim in ESKD: enhanced patient experience, improved population health, reduced costs, and meaningful work for physicians (8). The Dialysis Patients Access to Integrated Care, Empowerment, Nephrologists, Treatment, and Services (PATIENTS) Demonstration Act (H.R. 4143/S. 2065) incorporates elements of existing programs and adds new ones to round out an alternative care model. Patients would be attributed to the model on the basis of their dialysis facility, and they would retain full access to their existing Medicare fee-for-service clinicians and services. Patients would have multiple opportunities to opt out in this demonstration as opposed to the ESCO, where patients can only opt out of data sharing. They also can access supplemental benefits akin to the C-SNPs without increasing taxpayer costs. Payments to the participating organization would occur monthly, smoothing cash flow and allowing providers to identify eligible patients more readily and invest in tailored benefits and timelier interventions. The organization defined by the Act would provide care coordination support to all of its patients with additional education and support for patients transitioning into transplantation, acknowledging the importance of kidney transplantation as a treatment modality for ESKD. Finally, nephrologists are part of the leadership team and governing body for this type of organization, although they do not have to be equity investors in the organization. Their leadership is essential for the organization to achieve safe, effective, equitable, efficient, patient-centered care.

For any of these models to succeed, nephrologists must be involved in their implementation. This requires familiarity with the care model, assessing practice-level capabilities to use and manage patient data for both individualized and population-level care, accessing care coordination resources, and combining the traditional focus on patient treatment with the ability to incorporate outcomes, which include hospitalization reduction, health care utilization into the plan of care, and treatment goals. Policy will also have to change for these models to expand. Whether it be incorporating CKD care or integrating transplant care at some juncture into new care models using the experience gained in C-SNPs today, these efforts to capture the entire care continuum are logistically challenging. Physician-centered proposals have recently emerged, including initiatives from the National Kidney Foundation (NKF; H.R. 3867) and the Renal Physicians Association (RPA). The NKF-sponsored legislation proposes a demonstration project tying Medicare payments to early detection of CKD and coordinated care, whereas the RPA proposal outlines a clinical episode payment model for physicians caring for patients with incident dialysis that incentivizes care coordination, quality of care, preemptive transplantation, and appropriate resource utilization.

Although there have been concerns raised regarding the PATIENTS Demonstration Act model, the opportunity to test an additional model distinct from other integrated kidney care models is essential to advance our community’s knowledge of what works in terms of care delivery and what does not when we start to consider scale, geography, and practice needs. If implemented, the model would have to consider how to incorporate other integrated care proposals, such as the NKF or the RPA proposal. It would also have a set of core quality and performance measures similar to the CEC Model today. Identifying new performance measures that the community agrees would appropriately assess the effectiveness of the model is an important future step and likely a multistakeholder dialogue.

Changes in care models always carry with them concomitant concerns. Even with community input, performance measures can ultimately prove to be weak or invalid, and as such, they do not advance care adequately. Although the PATIENTS Demonstration Act provides for the possibility of supplemental benefits, a potential lack of patient-physician–aligned incentives (9) could reduce the likelihood of success of the model. In addition, this model begins to triangulate patient interactions with the organization helping to oversee the care model and the physicians involved in patient care. Although this is a familiar construct in the hospital setting and even in ACOs, it represents a new patient education challenge. It is different than the traditional relationship of the patient with ESKD with his or her physician and the dialysis facility, and it could potentially alter that relationship in unintended ways.

The calculus of weighing the risks and benefits of a health policy issue, like the PATIENTS Demonstration Act, with existing evidence can be unsettling, but waiting for a level of certainty can be paralytic (10). There are propitious moments when momentum and need align to accelerate a policy idea, even if the data are not as extensive as desired. This confluence is present in ESKD today and argues for broader implementation of integrated kidney care. Gostin (11) recently discussed values to guide health system reform: universal access, equitable access, affordable access or cost, quality, and choice. By adding to the available models of integrated kidney care and expanding those offerings to larger populations of patients with ESKD, it is possible to satisfy these criteria. In so doing, we have an opportunity to transform care for an entire patient population.

Disclosures

B.N.B. and A.R.N. are employees of DaVita, Inc. and own stock and stock options in DaVita, Inc.