Accountable Care Organizations: Who Can Join and What Will They Mean for Nephrology?

Can my nephrology practice join an ACO?

Amy Williams: Nephrologists and nephrology practices are eligible to join an ACO and have patients assigned to them according to the two-step process described in the first article of this series (March Kidney News). If the patient received any primary care services from a primary care provider in an ACO, but has received most of these services from a specialist (e.g., a nephrologist) eligible to have patients assigned to them as part of an ACO, the patient will be assigned to the specialist and his or her ACO.

ACO providers that are not eligible to have ACO patients attributed to them (medical and surgical specialists or acute-care hospitals) will be able to participate in more than one ACO. Thus, if a nephrologist associated with an ACO is not identified to be the provider of primary care services attributed to any patient in the ACO, that nephrologist is eligible to participate in more than one ACO. However, if the primary care services of the nephrologist are used to assign the patient to the ACO, the nephrologist must be exclusive to that ACO for the purposes of the Shared Savings Program ACO. The ACO will report the Tax Identification Numbers (TINs) and National Provider Identifiers for each practice and individual provider associated with them to the CMS.

Can dialysis organizations join an ACO? Would they want to?

Dan Weiner: Dialysis organizations cannot form an ACO themselves, but they can be a part of a larger ACO structure. Similarly, individual nephrologists can and will be part of ACOs, primarily as specialists and, in rare occasions, as designated primary care providers. In fact, if nephrologists bill under more than one TIN, they can be included in more than one ACO. Given that nephrologists do, in some circumstances, act as primary care physicians and that their group of primary care patients will be far smaller than that of most primary care providers, this policy may enable nephrologists to maintain the primary care relationships with some patients, albeit with some administrative uncertainty.

In one sense, dialysis providers will definitely want to participate in ACOs, and ACOs should want the input of dialysis providers. The dialysis team has far more contact with a dialysis patient than any other medical provider the patient is likely to encounter, and the dialysis provider is uniquely positioned to monitor health and health interventions that are most relevant to a dialysis patient. Additionally, given that the major health care issue for a dialysis patient is almost always the sequelae of kidney failure and kidney failure itself, the expertise of the dialysis provider in managing these issues is critical for optimizing patient success.

Finally, even though the ACO final rule specifically mentions antitrust concerns, dialysis providers will not want to lose possible patients to other nearby facilities, suggesting that, if possible, the providers will want the opportunity to collaborate with local ACOs. However, there are inherent problems.

First, dialysis patients make up a small proportion (only about 1 percent) of Medicare beneficiaries, suggesting that ACOs will not develop efficient practices for caring for these patients. Second, given the varying catchment areas of dialysis facilities and of ACOs, it is likely that a single dialysis facility will end up working with multiple ACOs. With each ACO having its own administrative and information technology infrastructure—not to mention the CMS infrastructure and reporting requirements for dialysis units (e.g., Consolidated Renal Operations in a Web-enabled Network [CROWNWeb]), their chain affiliations (if present), and local departments of public health—the administrative burden could be substantial. Third, the quality metrics for dialysis units under the Quality Incentive Program (QIP) and for ACOs within the final rule are inherently different, and quality indicators for the general population often are not applicable to dialysis patients.

What will it mean for me as a nephrologist if my dialysis patients are attributed to an ACO?

Emily Robinson: As with patients, there will be pros and cons for nephrologists if their patients are attributed to an ACO. It is hoped that improvements in communication and coordination of care plans will help the nephrologist as well as the patient.

However, nephrologists will need to be increasingly diligent in ensuring that medication changes and screening tests ordered by primary care physicians to meet quality guidelines are actually appropriate for each individual patient, and they may have to spend more time talking with patients and other physicians about the appropriateness of these interventions.

Dialysis centers have already set up reporting systems for documentation in dialysis patients, but these are different from the systems used for ACO reporting. Thus, a nephrologist/dialysis unit that joins an ACO may be required to use multiple reporting systems for documentation, increasing work, confusion, and financial burden to put the systems in place. Although there are no specific provisions of the ACO that counter the Prospective Payment System bundle and QIP, the amount of effort to satisfy both systems is quite large. It also would remain to be seen whether referral patterns to nephrologists would change. Although patients in a specific ACO do not need to see all of their specialists in that ACO, primary care physicians might urge them in that direction.

How does CMS plan to coordinate the ACO reporting and quality measurements with the QIP program reporting and quality measurements?

Dan Weiner: Unfortunately, no coordination is planned for data reporting or quality metrics between these two CMS programs. This is somewhat ironic, given the time and expense that CMS has devoted to developing dialysis-specific reporting (in the form of CROWNWeb) and dialysis-specific quality measures. The most notable item here is the lack of applicability of the ACO performance measures to a dialysis population.

For example, colorectal cancer screening and breast cancer screening likely are neither cost effective nor beneficial for dialysis patients aged 50 to 75 years who are not eligible for transplantation and therefore have life expectancies of less than 5 years. Similarly, there are no evidence-supported blood pressure (BP) targets, hemoglobin A1C targets, or low-density lipoprotein cholesterol targets for dialysis patients, and no data to support the supposition that any intervention to address BP levels, diabetes control, or hypercholesterolemia has a benefit to the dialysis population. In theory, one could be concerned that ACOs will aggressively pursue these performance metrics at increased cost and increased burden both to the health care system and to individual patients.

Is CMS going to allow formation of renal ACOs?

Emily Robinson: At this time, the answer is no. In the final rule the CMS did not allow for the formation of any specialty ACOs, including a renal/ESRD ACO, stating “although we do not see the need to design distinct ESRD or cancer specific ACOs, neither of these provider types are in any manner excluded from participation in an ACO.” So, for the time being, the only types of ACOs that can form are “general” ACOs.

What other kinds of new care delivery models exist?

Amy Williams: The ACO model is not the only one being considered to improve the value of care provided. The Federally Qualified Health Center (FQHC) Advanced Primary Care Practice Demonstration’s Patient Centered Medical Home (PCMH) program is another primary care–based comprehensive, coordinated care model sponsored by the CMS in collaboration with the Health Resources Services Administration (HRSA).

As with the ACO model, responsibility for chronic care is centered on the primary care team with the goal of providing patient-centered care by improving care coordination and promoting health while decreasing the overall cost of care. During the demonstration period, individuals with ESRD are excluded. The PCMH involves the subspecialist by designating the subspecialist a PCMH Neighbor with a well-defined graduated role in the care of PCMH patients with subspecialty illnesses. This model would designate the nephrologist as having primary responsibility for patients with acute complicated renal disease as well as chronic complicated subspecialty care needs, such as those receiving dialysis or having undergone renal transplantation.

Emily Robinson: In addition, the Centers for Medicare and Medicaid Innovation (CMMI) has statutory authority to test new innovative models of care and could potentially conduct a demonstration or pilot project on a renal-specific care delivery model.

The ASN ACO Task Force and members of the ASN leadership are in an ongoing dialogue with the Innovation Center, as it is known, about potential opportunities and challenges that a nephrology integrated care model could yield. The ASN has developed a series of principles about the formation, structure, and scope of nephrology integrated care models that the society has discussed with the Innovation Center and made available on the ASN home page (

Stay tuned: The next Q & A will focus on nephrology integrated care delivery models and ASN’s principles related to a potential pilot project or demonstration project. If you have questions you would like the ACO Task Force to address in this series, please email ASN Manager of Policy and Government Affairs Rachel Shaffer at