Executive Summary

This policy brief outlines the critical need for a sustained, specialty-focused accountable care model for Medicare beneficiaries with chronic kidney disease (CKD) and End-Stage Renal Disease (ESRD). We present evidence of the current Comprehensive Kidney Care Contracting (CKCC) model’s success, argue for its stabilization and permanency, and propose targeted improvements to ensure long-term viability and savings for the Medicare program.

Beneficiaries with CKD and ESRD represent one of the highest-need and highest-cost populations served by Medicare. Although these beneficiaries comprise a small share of Medicare’s enrollment, they account for a disproportionate share of spending.

Chronic diseases are deeply interconnected, with conditions like diabetes and high blood pressure being the leading drivers of kidney disease. Yet kidney disease often progresses silently, with few or no symptoms, leaving many people unaware until they reach advanced stages and require life-sustaining care. This reality underscores the urgent need for policies that prioritize prevention and access to comprehensive care, early detection, and care management aimed at slowing disease progression. Once a patient reaches CKD Stage 4–5 or ESRD, the nephrologist typically becomes the principal care provider, managing dialysis, transplant preparation, comorbidities, and care coordination. These patients require intensive and specialty-driven care management. To deliver the best possible outcome to patients, nephrology practices must make investments that traditional Medicare Fee-For-Service (FFS) reimbursement does not support and that general primary care–based models are not structured to provide.

Multiple peer-reviewed studies demonstrate that consistent nephrology management slows disease progression, improves outcomes, and reduces avoidable utilization. For example, Shi et al (2018) concluded that multidisciplinary care models can lower the all-cause mortality of patients with CKD, reduce temporary catheterization for patients receiving dialysis, decrease the hospitalization rate, and slow the eGFR decline. Additionally, the authors found the reduction in mortality to be critically linked to the professional care team managing the patient’s care1.

Kidney-specific innovation models can build on this success, allowing for highly tailored and targeted interventions uniquely effective for this patient population, including delaying disease progression, improving preparation for dialysis, increasing transplant and related referrals, increasing home dialysis adoption, and reducing avoidable hospitalizations.

Specialty Models Work for Kidney Care

Traditional Fee-For-Service (FFS) models create misaligned incentives by only recognizing treatment volume. In contrast, VBC models explicitly incentivize care coordination, resulting in better clinical outcomes, which generate cost savings. Several factors make kidney care well-suited to a dedicated accountable care model—more so than other provider specialties and more so than these high-need high-cost patients would receive in general primary care-based models. Kidney care patients have frequent and predictable interactions with the healthcare system, especially nephrologists, creating consistent opportunities for proactive management and coordination that many other chronically ill patient populations lack. Simply stated, the disproportionate spending on this patient population is “modifiable.” Further, unlike many chronic conditions, kidney disease has distinct stages (CKD 3-5, ESRD, Transplant) and objective metrics (e.g., eGFR, catheter rates, home dialysis rates).

The CKCC model built on the lessons from earlier models like the Comprehensive End-Stage Renal Disease (ESRD) Care (CEC) Model (the “ESCO” model), which expanded the focus upstream to late-stage CKD to delay disease progression. The Innovation Center (CMMI) launched CKCC in 2022 as part of the 2019 Advancing American Kidney Health Executive Order to assess value-based program designs aimed at reducing fragmented care, enhancing care coordination, and promoting cost-effective interventions for Medicare beneficiaries living with kidney disease.

In 2023, the second Performance Year (PY), more than 100 kidney care entities (KCEs), 4,331 nephrology professionals, 229 transplant professionals, and 1,594 dialysis facilities participating in the model. CKCC has demonstrated that specialty-focused models can materially improve outcomes including2:

  • 31% improvement in optimal dialysis starts
  • 10% increase in home dialysis
  • 69% increase in pre-emptive transplants

These results help to validate CKCC’s core design, showing meaningful quality gains while maintaining cost discipline. Improvements in quality will only continue to accrue over time, and their impact will compound favorably on health and cost outcomes for this population. Further, the model’s positive performance to date creates strong incentives for participating organizations to continue making investments that will ensure the program’s sustained impact.

Recommendations to Stabilize the Future of Accountable Kidney Care

The current CKCC model has begun delivering measurable improvements in care delivery and outcomes. However, continued transformation in this space requires a multi-year commitment with model refinements to ensure the program’s viability and stability. Model participants have made significant investments in care teams, clinical workflows, technology upgrades, advanced analytics, and patient engagement infrastructure. Maintaining these investments – and incentivizing additional ones – requires a predictable operational environment.

CMMI has shown a willingness to work with the kidney care community in developing and adopting policy changes to advance the mutually shared long-term goal of a viable, permanent VBC delivery model for beneficiaries with kidney disease.

Near term, the most important step in that direction will be to ensure there is no gap between models after the current demonstration ends in 2027. There are multiple paths that CMMI could take to achieve that “no gap” goal, including an extension of the current model or developing a kidney-specific track within LEAD. Whatever form an accountable care model takes for these beneficiaries after 2027, the kidney care community would recommend that CMMI implement the technical changes put forth in the subsequent sections of this paper.

Implementing these technical changes will preserve providers’ engagement and maximize the return on CMS’s initial investment in the CKCC model, laying the foundation for its inclusion as a permanent feature of the Medicare benefit.

The CKCC model has catalyzed transformation in U.S. kidney care. Evaluation results show increased quality and value. The recommendations below are intended to further advance patient access to value-based care, scale CKCC’s positive impact, and ultimately strengthen the Medicare program.

Recommendations

1. Expanding participation of patients and nephrologists in the program

TIN/NPI Methodology

Consistent with the approach established under the LEAD Model, CMMI should manage CKCC model participation at the TIN level. To accommodate Participant Providers who practice within hospital systems, large multispecialty groups, academic institutions, and other organizations whose billing TINs may simultaneously participate in the Shared Savings Program or another Medicare initiative involving Shared Savings, CMS should establish an appeal mechanism through which such providers may seek an exception to the applicable overlap restrictions, enabling their continued participation in the CKCC Model. To ensure clarity. A single TIN should only be able to participate in a single KCE.

Geographic Restrictions

Currently, the CKCC service area methodology is limiting and has unintended consequences. The rationale for using core-based statistical areas (CBSAs) to determine a KCE’s service area is logical. However, the current methodology and rules introduce confusion into the process of partnering with nephrology practices. As KCEs near the maximum number of CBSAs, they face trade-offs when assessing further growth and must choose which CBSAs and practices to add and exclude. Particularly, the five CBSA limit and contiguity rules force out relevant CBSAs not adjacent to major urban centers. It is well known that patients in lower-density metropolitan areas have less access to healthcare; as a result, this unintentionally undermines the availability of models that can address this challenge that contributes to poorer health outcomes.

The CBSA rules can also lead to instability. Healthcare provider groups sometimes restructure in the course of normal practice operations. This restructuring often requires them to obtain new TINs and/or NPIs. The process for obtaining these identifiers is not always fast, clear, or easy to navigate for physicians. Thus, practices may be removed from the model due to minor administrative problems. Further, if these practices anchor connecting Main CBSAs to the Market Area, all participants in the now-disconnected CBSAs would be ejected from the model as well – patients and providers alike.

KCE Market Area and Service Area requirements should be revised to increase overall program participation and better support beneficiary access in rural communities by:

  • Increasing the maximum number of CBSAs permitted in a KCE Market Area from 5 to 10.
  • Increasing the permitted number of contiguous rural counties between CBSAs from 3 to 6.
  • Increasing the permitted number of skipped CBSAs from 1 to 2.
  • Beneficiaries residing in rural counties that are adjacent to a CBSA in the KCE’s Market Area, and who meet alignment criteria, may be aligned to any participating Nephrology Professional with a Primary Address located in a CBSA within the beneficiary’s state.

These changes would address the fact that current Market Area rules disproportionately constrain KCE participation and beneficiary alignment in geographically large, sparsely populated regions, as well as in certain Micropolitan Service Areas. The proposed changes would improve CMS’s ability to ensure access to coordinated kidney care for beneficiaries who live far from metropolitan centers, while preserving the existing safeguards embedded in the Market Area framework.

Further, CKCC rules should apply uniformly to all KCEs regardless of whether a KCE includes on its Participant List one or more Dialysis Facilities owned in whole or in part by a Large Dialysis Organization (LDO). This would ensure equitable treatment of all KCEs under the Model.

Participant Additions

Many interested nephrologists face delayed participation in the CKCC model. Currently, participant nephrologists have only one add window for an 11-day period in June of the year prior to the performance year. The need for defined and limited add windows is reasonable; however, allowing only one window per year has the effect of dampening provider participation, often for reasons beyond their control. Consistent with the LEAD Model, CMS should permit the addition of Participating Providers for alignment purposes within a performance year on no fewer than two occasions per Performance Year to support onboarding and operational flexibility (specifically, an addition window in Q1 for alignment in the start of Q3 and an addition window in Q3 for alignment at the beginning of Q1 for the following year).

Alignment

General Alignment
Eligible beneficiaries should be identified using a 12-month alignment lookback period, applying the same CPT/HCPCS codes as CKCC with the addition of APCM codes G0556-G0558, rendered by participating nephrology professionals, excluding claims with hospital-based places of service to prevent attribution based on one-time encounters during inpatient rounding.

Required beneficiary alignment minimums for a KCE should be 750 beneficiaries with CKD Stages 3B, 4, or 5, and 350 Beneficiaries with ESRD.

Earlier-Stage CKD Eligibility
The current alignment rules allow only patients who have late-stage CKD to participate and have frequently visited a nephrologist engaged in CKCC. Many patients across the country do not have a nephrologist or are not frequent utilizers of nephrology services. Further, patients with CKD 3 cannot align and are thus progressing toward late-stage CKD before benefiting from kidney focused value-based care. This alignment methodology hinders broad access to value-based care. This alignment rule effectively creates populations of underserved patients – i.e., those without nephrologists and those who visit nephrologists infrequently – who are excluded from CKCC. CMMI is missing a valuable opportunity to engage in upstream prevention, education, and care to delay progression. CMMI could accomplish upstream alignment by:

  • Aligning beneficiaries identified as having had at least one Medicare claim during the alignment lookback period with a diagnosis code of CKD Stage 3B.
  • Consistent with the two-touch rule applied to CKD Stage 4, CKD Stage 5, and ESRD beneficiaries, establishing that CKD Stage 3B beneficiaries must have at least two CKD aligning nephrology services claims during the 12-month alignment lookback period, with at least one participating nephrology professional of the KCE.
  • Excluding CKD Stage 3B beneficiaries from the quality measures applicable to CKD Stage 4, CKD Stage 5, and ESRD beneficiaries, or evaluating them under a standalone set of quality measures tailored to the CKD Stage 3B population. Standalone quality measures could possibly include, but are not limited to, a nephrology visit frequency requirement, or clinical outcome measures consistent with those established under CMMI models such as the Long-term Enhanced ACO Design (LEAD) Model or the Advancing Chronic Care with Effective, Scalable Solutions (ACCESS) Model.
  • Establishing a standalone financial benchmark for the qualifying CKD Stage 3B beneficiary population to account for the distinct utilization patterns and expenditure profiles of this population, separate from the benchmarks applicable to CKD Stage 4 and 5, or ESRD.

CARA
Include a CMS-Administered Risk Arrangements (CARA) initiative identical to that featured in the LEAD CMMI model, a voluntary mechanism that facilitates downstream episode-based risk arrangements between participating entities and other specialty care providers to increase specialist accountability for quality and cost.

Departing Providers
If an Aligning Provider is removed from a Participating Entity’s (PE) Participant List during a Performance Year (PY), any beneficiaries aligned to the PE exclusively through that provider’s Qualifying Nephrology Services must be automatically de-aligned, effective on the provider’s termination date. However, there should be an exception for Continuity of Care, meaning, a beneficiary will not be de-aligned if they receive a Qualifying Nephrology Service from another active Aligning Provider on the PE’s Participant List during the remainder of the PY.

2. Improve Financial Stability

CMMI should work to improve the financial stability of the current CKCC model and any successor model. All CKD 4/5 and ESKD beneficiaries eligible for the model should be classified as High Needs beneficiaries, with a concurrent risk adjustment factor (RAF) methodology applied universally. The CKD/ESKD population is inherently high-risk and medically complex, as evidenced by consistently elevated RAF scores. A concurrent risk adjustment model more appropriately captures in-year disease progression and acute conditions, high beneficiary churn, and variability in timing of diagnosis and treatment initiation.

Operationalizing multiple RAF models (e.g., prospective vs. concurrent and ESKD vs. non-ESKD) is discouraged. Such complexity increases administrative burden and creates unnecessary reconciliation risk without meaningfully improving accuracy. A single concurrent model would improve clarity, predictability, and feasibility. Further, risk scores must adequately reflect true beneficiary disease burden. There are two ways risk score growth can be reflected appropriately in the program:

  • Utilize current CKCC approach with a cap based on a moving reference year. This will be simpler to apply across both existing and new ACOs.
  • If using a flat risk score cap for the entirety of the program, set higher risk score caps to reflect the higher acuity of both CKD and ESKD populations. For new ACOs, consider no cap or a materially higher cap than ACOs with membership from CKCC.

This is necessary to ensure risk-adjustment accurately reflects beneficiary risk to both new and existing KCEs given relatively high beneficiary churn year-over-year in the late-stage CKD and ESKD patient population.

Benchmark Reference Year and Trend Methodology

Medical cost trends must represent the applicable population. The LEAD ACO methodology sets a good precedent by including a mix of national and regional trend rates and the current CKCC methodology trues up trend and should be maintained in any successor model.

Further, all retrospective trends should allow for completed runout for the years of trend being developed (e.g., use trend update from April after the program year rather than year before) and not be impacted by corridors or caps. A model that relies only on prospective trend is not an appropriate fit for this patient population. Further, CKD Stage 4/5 trends should be derived from CKD 4/5 beneficiaries, not the broader non-ESKD Medicare population. Absent such adjustments, benchmarks may fail to reflect actual operational environment and clinical realities.

Treatment of Prior Program Savings

Savings achieved through improved care management should not be capped when incorporated into future benchmarks or incorporating a higher cap (5% or greater). An absolute cap penalizes those KCEs with the greatest amount of gross savings already achieved through a more punitive rebasing of those savings. If a cap is needed (e.g., 5%), it should be applied to the combined impact of 50% of historical savings and comparison to regional rate as opposed to the higher of the two. Additionally, prior savings should be calculated before applying program quality withholds or benchmark discounts, especially if only 50% of prior savings are recognized. Finally, in the event of KCE ownership changes or provider composition changes, prior period savings should remain attributable to the organization to preserve continuity and performance incentives.

Benchmark Discounts

CMMI should set the maximum discount for CKD 4/5 beneficiaries at 3%. For ESKD patients the 3% maximum discount should apply only to non-dialysis costs. Dialysis costs for ESKD beneficiaries are largely non-discretionary and non-impactable. Applying discounts to total cost of care, including dialysis, risks creating unattainable benchmarks.

The Advance Shared Savings Payment (ASSP)

The ASSP should be offered and administered in a manner consistent with its implementation in CKCC currently.

3. Ensuring Quality Outcomes

Quality Withhold Structure

CMMI should reduce the quality withhold to 3%, consistent with programs such as LEAD ACOs. Further, CMMI should not apply the withhold to dialysis spend for ESKD patients. Many kidney-specific quality measures are subject to low denominator counts and year-to-year volatility unrelated to actual care quality. A smaller withhold would improve financial predictability while maintaining alignment with CMS quality objectives. A 3% withhold provides a powerful incentive for KCEs/providers caring for CKD 4/5 and ESKD patients with total cost of care much greater than average Medicare patients.

Fixed, Measurable, and Actionable Quality Metrics

It is difficult to have an impact on many of CKCC’s quality metrics. Pairing them with a significant withhold results in financial incentives highly indexed to uncertainty.

CMMI should improve clarity of new metrics or alter to pay-for-reporting only, and metrics should align with the impact on total cost of care. These changes would make the model easier to project financially, simpler to administer, and more attractive for broader participation, which will continue to improve performance that will drive cost savings.

Three to four meaningful and targeted measures are most appropriate for CKCC and any successor model. The group strongly recommends two measures and provides options for an additional one to two measures, along with supporting rationale.

Measure Priority #1: Optimal Starts

CMMI should keep Optimal Starts; however, additional measure transparency is needed.

  • CMMI should collaborate with EQRS to allow KCEs to have real-time, view-only access to their beneficiaries in the EQRS for improved tracking of optimal starts.
  • CMMI should also allow for December claims run out prior to running the semi-final reconciliation report. This will enable December dialysis starts and preemptive kidney transplants to be included in the measure calculation.

Measure Priority #2: Patient Activation Measure (PAM)

CMMI should keep the PAM measure in CKCC.

  • A patient evaluation measure is important for the CKCC program; however, patient friction and administrative burden are important considerations for implementing patient-reported outcome measures (PROMs).
  • We believe alternative measures like CAHPS are not appropriate for the CKCC model.
  • CMMI should engage participants in any discussion of alternative patient-reported outcome measures that would be used instead of PAM.
  • CMMI should work with Phreesia to allow secondary patient identifier options to be added to the Flourish profiles. This will reduce the risk of mismatched patients who continue to receive updated MBIs throughout the measurement year.

Measure Priority # 3: Admit / Readmit Measure

  • The CKCC program should include a hospitalization measure, either the ‘Risk-standardized all conditions readmission’ or the ‘All-cause unplanned admission for patients with multiple chronic conditions’ measures would be appropriate.
  • Patients with advanced kidney disease have a high rate of hospitalization related to their disease and additional complex, chronic conditions. Including a utilization measure addresses appropriate efforts to reduce unnecessary utilization
  • If CMMI selects the ACR readmission measure, the specifications should be adapted to kidney care. For example, benchmarking should compare against an appropriate nephrology or KCE peer group, not a broad general Medicare population. Risk adjustment should incorporate kidney-specific acuity factors, including ESKD status, dialysis status, and CKD severity.
  • Broader Medicare risk adjustment factors, such as age, chronic disease categories, clinical risk factors, and other demographic or stay-level characteristics, are helpful, but may not fully capture the clinical complexity of late-stage CKD and ESRD populations.
  • Because these measures rely on mathematically complex observed-to-expected ratios, they effectively function as a “black box.” CMMI should provide clear visibility into observed events, expected events, risk-adjusted performance, and the patient-level factors driving the expected rate so entities can understand if their performance is driven by true care variation or patient mix.

Additional Measures for Consideration for Future Quality Program

The following five additional quality options could be appropriate choices to round out a total of three to four measures in the future quality program.

Sodium-Glucose Co-Transporter 2 (SGLT2) Use

  • CMMI should not maintain the current Delayed CKD Progression measure in its current form; instead, CMMI should consider SGLT2 use as a claims-based measure.
  • The current delayed progression measure should only be considered if the definition is materially revised to improve actionability, transparency, and fairness. Delay in progression activities require starting upstream from stages 4 and 5 and is not a good fit for measuring the successes of CKCC.
  • SGLT2 inhibitor prescribing is an evidence-based process-of-care measure, and it could serve as an indicator reflecting the adoption of guideline-directed therapy known to slow disease progression and improve CKD, cardiovascular, and mortality outcomes.
  • As an alternative to the current delayed progression measure, we strongly support SGLT2 use as a reporting measure. The SGLT2i measure would be best structured as a reporting measure because it would be a custom measure
  • Prior concerns around access, affordability, and equity remain. However, dapagliflozin generics entered the market recently, which may improve affordability and access.
  • We also recognize there is not currently a SGLT2 measure existing in any CMS programs and we acknowledge that additional due diligence on real-world availability and patient cost before final measure adoption is critical. One potential approach is using claims to measure performance to reduce administrative burden for practices. We welcome methodology discussions as CMMI narrows in on a measure set.

Standardized Mortality Ratio (SMR)

  • SMR is aimed in the right direction, but it is difficult to impact and measure in a single payment year.
  • If SMR is maintained, CMMI should break out CKD and ESRD.
  • Further, additional clarification is needed on how the standardization of the measure works due to limited visibility.
    • Again, these measures rely on mathematically complex observed-to-expected ratios, they effectively function as a “black box.”
    • CMMI should provide clear visibility into observed events, expected events, risk-adjusted performance, and the patient-level factors driving the expected rate so entities can understand if their performance is driven by true care variation or patient mix.
      As an alternative to SMR, CMMI could consider the applicability of an Advance Care Planning measure, which is already part of multiple CMS quality programs.
  • The ACP measure captures meaningful clinician-patient communication and documented discussions, even when patients choose not to complete formal directives. It recognizes that declining life-prolonging treatment or prioritizing comfort can represent high-quality care, not failure.

Glycemic control

  • This HEDIS measure is applicable to a significant portion of the kidney patient population, and uncontrolled diabetes is an important cause of hospitalization.
  • This measure would be useful; however, because diabetes is managed across multiple specialties (Primary care, Endocrinology, and to a lesser degree Nephrology), it can be more difficult to operationalize in a kidney disease-specific program than other measures.

Blood pressure control

  • This HEDIS measure applies to almost all patients with advanced CKD, as well as the majority of those with CKD 3. Hypertension drives significant morbidity, so control is important to cardiovascular health and CKD disease progression.
  • This measure could be useful in a kidney disease-specific program with the typical exclusion for patients with ESKD, because blood pressure control parameters for dialysis patients are not established in the literature.

Kidney Health Evaluation for Patients with Diabetes

  • This kidney-focused HEDIS measure promotes early detection of kidney disease, creating opportunities to delay progression and improve patient outcomes. While PCPs play a critical role in early identification and timely referral to nephrology, ongoing evaluation of kidney function remains essential to guide appropriate medication optimization, risk stratification, and patient education.

Revise Program Evaluation

Without revising the program evaluation framework, the determining the CKCC model’s financial impact from the earlier program years becomes obsolete and does not accurately reflect the program’s potential value moving forward.

CMMI should apply the 2026 program design features retroactively to reassess the early performance years and publicly share the results. This reassessment should incorporate:

  • Changes to the Quality Care Plan (QCP) fee schedule
  • Increased CMS savings guarantees (for both Professional and Global track discounts).
  • Elimination of transplant bonuses
  • Skin substitute fee schedule reduction
  • Removal of providers who left the program

The Future of Value-Based Kidney Care

The success of value-based kidney care depends upon nephrologists’ enthusiasm to participate over the long-term. Nephrologists have invested significant resources and have modified the way they provide care because they believe in value-based kidney care. If a kidney-specific value-based care model does not continue directly following the end of the CKCC model post-2027, nephrologists’ interest will likely diminish. Further, it is imperative that patients do not experience a disruption in their access to value-based care. CMMI should ensure that there is no gap or “desert” period for patients or nephrologists. Our strong preference for this end would be a permanent, certified model. Recognizing that certification may not be possible in the absence of more data demonstrating program savings, we would support other forms of continued, value-based care, including extension of the current program or launching a refined successor model under a rubric similar to LEAD.

All pathways should prioritize stability, simplicity, and sustained provider engagement for the benefit of Medicare beneficiaries. To ensure continued nephrologist participation and avoid disruption to beneficiaries, CMMI should not adopt model elements that would significantly undermine participation in value-based kidney care moving forward.

Conclusion

Medicare beneficiaries with kidney disease represent one of Medicare’s most clinically complex and financially significant populations. Specialty-focused accountable care has demonstrated measurable improvements in patient outcomes and strong provider engagement that would otherwise not be possible under general, primary care-led models. With targeted technical refinements, CMMI can ensure the continued success and sustainability of accountable kidney care, preserve the clinical and quality gains, expand participation, and deliver meaningful results for Medicare beneficiaries and the Medicare program.

Share

Related Posts