ASN Accountable Care Organization Task Force: Q & A

Dan Weiner, MD, and Amy Williams, MD, were among the 12 ASN members who served on the ASN Accountable Care Organization (ACO) Task Force, which was chaired by Lee Hamm, MD. Here, the task force members discuss the ACO proposed rule and their perspectives on some of the complex issues the rule raised related to care of patients with kidney disease.

What is your overall impression of the ACO concept?

Dan Weiner (DW): In concept, the ACO model has the potential to better integrate medical care across the spectrum of health and disease, particularly when it comes to fairly healthy Medicare beneficiaries and those with less severe chronic illnesses.

Lee Hamm (LH): In practice, it is uncertain how ACOs will deal with high-cost patients with severe chronic diseases as well as those with severe acute illnesses. It is very uncertain whether physicians will have the flexibility to appropriately individualize care for the entire range of patients, including those seen by nephrologists.

Amy Williams (AW): Overall, the proposed ACO structure, governance, and required quality metric monitoring and reporting are too complex and rigid, reducing the flexibility to best manage highly complex patients needing subspecialty care.

The comment letter stated that ACOs may not be well positioned to care for patients on dialysis. Can you elaborate on how the Task Force came to that decision?

DW: The ACO proposed rule focuses on the primary care providers (PCPs) and patients who receive most of their medical care from PCPs. It emphasizes primary and secondary prevention to maintain wellness. Dialysis patients require a very different model of care from the general population. Besides being among the most costly of all Medicare beneficiaries, dialysis patients also have very different disease frameworks.

LH: Many of the care recommendations and proposed ACO quality measures cannot be extrapolated to dialysis recipients. For example, should blood pressure targets for dialysis patients uniformly be below 140 mm Hg systolic? Should dialysis patients with fractures have bone density scans and prescriptions for drugs to treat or prevent osteoporosis? Should chronic disease screening guidelines, such as mammography and colonoscopy, be the same for dialysis patients? These proposed ACO “quality performance measures” often may not be applicable to dialysis patients and, in some cases, could actually prove harmful and unnecessarily expensive to dialysis patients.

AW: In addition, the performance metrics required for the ACOs are very different from those required for the dialysis expanded bundled and quality incentive program (QIP).These differences may lead to confusion and decreased coordination of care. Finally, many dialysis patients receive the majority of their medical care from nephrologists and other dialysis-affiliated professionals; this frequently includes primary care as well as cardiovascular disease and diabetes management. As proposed, ACOs make no allowance for this fact.

What do you see as the potential benefits of ACOs for patients with CKD?

DW: First, most evidence-based medicine recommendations for the general population likely also apply to people with CKD stage 3 and 4. So improving these elements of care for all patients should also lead to similar improvements for patients with CKD.

LH: Second, we have failed to date in timely preparation and education of patients with advanced CKD for their future, be it dialysis, transplantation, or conservative care. The framework for ACOs, by incentivizing preventive care, could improve integration of planning for kidney failure. We would hope that future iterations of ACOs would address this aspect of care.

AW: ASN would gladly partner with CMS to define best practices and expectations for managing advanced CKD in the context of an ACO.

Some in the nephrology community have discussed the possibility of a “renal-specific” ACO. Why did the Task Force believe it was not necessary to discuss a specialty ACO in the comment letter?

LH: There were several reasons. First, the proposed rule from CMS was very clear that the current ACO model was focused on primary care and was not focused on specialist care. Accordingly, while there may ultimately be a role for more renal-specific care models, we felt it was important to deal directly with the issues raised by the current proposed rule.

DW: Second, given the rapidly changing dialysis provider environment, formation of renal-specific ACOs could have further major implications on provider consolidation that need to be considered in greater detail. Finally, dialysis in the United States, under the expanded bundle and Quality Incentive Program (QIP), already incorporates many of the major features of ACOs, with the major difference that hospital care and physician fees, even if related to dialysis, are not included. However, the QIP is tailored to dialysis patients, with dialysis-specific technical expert panels charged with refining dialysis metrics.

AW: Given the recent implementation of the expanded bundle and forthcoming QIP, we felt that it was important to explore the successes and failures of this “limited” ACO model in dialysis before considering substantial expansions. We do support the option to have multiple demonstration projects to further explore the concept of a “renal-specific” ACO.

What did the Task Force think about the 65 proposed quality measures as they might affect patients with kidney disease?

DW: The list of quality measures really reinforced for us that the ACO proposed rule was not meant for dialysis patients. Many of the systems and care coordination measures, if relevant, are already discussed in the Conditions for Coverage, while the vast majority of the patient “evidence-based” measures to promote wellness have no evidence to support their use in dialysis patients and some may actually lead to harm and increased costs.

What were the difficult decisions the group faced?

DW: There were several very difficult decisions. We were fairly certain that dialysis patients did not belong in an ACO as proposed, but remained concerned that excluding them could create a disincentive for ACOs to provide appropriate pre-dialysis care. For example, there would be no financial incentive for an ACO to cover placement of an AV fistula prior to initiation of dialysis if the ACO would not receive the downstream benefit.

LH: We proposed that CMS could solve this dilemma by establishing a quality measure for patients with late stage CKD for timely implementation of a kidney replacement plan. This measure would include creation of hemodialysis access if hemodialysis were the primary planned kidney replacement modality.

AW: Clearly, provision of vascular access remains an important issue that will require collaboration between CMS and CKD providers. We stated in the comment letter that ASN stands ready to work with CMS to develop a standard approach to late stage CKD patient care, and that quality measures based on these recommendations should be included in an ACO’s expectations.

DW: The second difficult decision was what to say about transplant recipients, and we in fact consulted with the ASN Transplant Advisory Group to develop a nuanced position on this issue. Many of the reasons why we felt that dialysis patients were inappropriate for ACOs are also applicable to transplant recipients, particularly those who are in the immediate peri-transplant period. However, ACOs, if successful, could provide substantial benefits to stable transplant recipients. This led the workgroup to call for exclusion of recent transplant recipients from ACOs, and to offer to work with CMS to develop criteria defining a “recent” transplant recipient versus a recipient who has been living stably and could potentially benefit from being in an ACO.

Moving forward, how do you see kidney patients and nephrologists interacting with an ACO?

DW: For the immediate future (if the ACO rules are finalized as proposed), I suspect the program will be very much like the HMO model, with a shift toward primary care doctors providing most medical care for CKD patients until late stage CKD is present. Ultimately I hope that nephrologists, particularly those who provide a lot of primary care to their patients, will be able to participate in an ACO model if they so choose.

AW: The ability of a nephrologist or nephrology group to contract with an ACO is critical to coordinate the care of patients needing subspecialty care. As the relationship between nephrologists and ACO providers evolves, it may become apparent that there are cost savings and improved quality of care when a nephrologist provides primary care to their complicated subspecialty patients. To demonstrate these advantages, and document and report quality metrics, a shared medical record is a necessity.