Oncology Care Model
July 1, 2016
Centers for Medicare & Medicaid Services
On June 29, 2016, the Center for Medicare and Medicaid Innovation (CMMI or the Innovation Center) formally launched the Oncology Care Model (OCM) at the Cancer Moonshot Summit. The OCM is a 5-year voluntary pilot project intended to improve the quality of cancer treatment, and to lower Medicare spending, by coordinating care. The OCM is a multi-payer initiative, with 17 private payers participating along with Medicare. In total, 196 physician practices in 31 states are participating in the OCM. HHS projected that the OCM will include more than 3,200 oncologists and will cover approximately 155,000 Medicare beneficiaries.
Payment for OCM practices will be based upon costs for 6-month episodes of care triggered by a Part B or Part D chemotherapy claim. During an episode, all Part A and B services received by the beneficiary will be included in the OCM calculation, as well as certain Part D expenditures. Participating physicians will continue to bill traditional Medicare fee-for-service claims throughout OCM episodes. Additionally, the practices will receive a $160 Monthly Enhanced Oncology Services (MEOS) payment. Practices will also have the opportunity to receive retroactive semi-annual performance-based payments. Practices’ performance-based payments will be determined by the amount (adjusted for quality measures) that total Medicare expenditures for beneficiary episodes over the course of a six month performance period are less than a target amount, which will include an adjustment for “novel therapies” to the extent that practices use such treatments to a greater extent than non-OCM practices.
For the first two years of the program (through Performance Period 3), practices will not be liable for payments if Medicare expenditures exceed target price for a performance period. Starting in year three (Performance Period 4), practices will have the option to switch to a symmetrical model that requires a participating practice to pay the difference when Medicare expenditures for a performance period exceed the target amount, but that will also include greater opportunities for shared savings.
The OCM will begin with episodes of care for cancer triggered by a new Part B or Part D chemotherapy claim on or after July 1, 2016.
Background and Timeline
The OCM is an initiative designed to improve the quality and effectiveness of cancer treatment, and to lower Medicare costs. More than 1.6 million people are diagnosed with cancer annually in the United States. Cancer patients are a high cost population, and a significant proportion of them are Medicare beneficiaries.
The OCM seeks to use multi-payer financial incentives to improve coordination of, appropriateness of, and access to care for beneficiaries undergoing chemotherapy. It seeks to accomplish this by aligning Medicare payments and those of other payers. The OCM is implemented under the authority of the CMMI. CMMI’s mission is to “test innovative payment and service delivery models to reduce program expenditures under the applicable titles while preserving or enhancing the quality of care furnished to individuals under such titles.” Section 1115A of the Social Security Act grants the Secretary of Health and Human Services the authority to waive all of the title XVIII requirements of the Medicare program.
CMMI has set out the timeline below for the OCM:
- July 1, 2016: Model begins
- Private payers encouraged to begin their performance periods within 90 days
- June 30, 2018: End of Performance Period 3 (episodes that end January 1 - June 30, 2018)
- Last performance period in which only the one-sided risk arrangement operates
- July 1, 2018: Beginning of Performance Period 4 (Episodes that end July 1 - December 31, 2018)
- Two-sided risk arrangement added
- May switch between arrangements semi-annually
- Practices failing to achieve performance based payments in the one-sided risk model at time of initial reconciliation for this Performance Period (expected to be August 2019) must leave the model or enter two-sided risk arrangement
- June 30, 2021: Model scheduled to end
HHS has selected 196 practices for the OCM. A list of the selected practices can be found here. All Medicare beneficiaries receiving included chemotherapy for cancer under the management of an OCM-participating practice and who meet the beneficiary eligibility criteria for episode definition are included. HHS has also selected 17 private payers to participate in the OCM.
Participation in Other CMS Initiatives
Practices and practitioners in the OCM may also participate in (and beneficiaries may be attributed to) Medicare Accountable Care Organizations (ACOs), the Bundled Payments for Care Improvement Initiative (BPCI), Comprehensive Care for Joint Replacement (CJR), and the Medicare Care Choices Model (MCCM). In addition, CMS indicated in the March 11, 2016 Proposed Rule for the Medicare Part B Drug Payment Model program that it would also apply to OCM practices. CMMI has stated that it may make adjustments to the payment methodology to account for changes in Federal regulation such as a Final Rule for the Part B Drug Payment Model.
Practices enrolled in the OCM will continue to submit claims to Medicare and be paid on a fee-for-service basis. In addition to those reimbursements for services, practices participating in OCM receive from Medicare a $160 per-month Monthly Enhanced Oncology Service (MEOS) payment for all Medicare cancer patients, and they also potentially will receive performance based bonus payments.
Monthly Enhanced Oncology Services (MEOS) Payment
OCM practices will be eligible to receive the MEOS payment monthly for each beneficiary during an episode attributed to them regardless of cancer type, unless the beneficiary enters hospice or dies. This fee will be $160 per beneficiary per month for the 6-month episode for a total of $960 per beneficiary per episode. The amount of the monthly fee will stay the same for the duration of the program.
OCM practices, or pools of practices,1 will have the opportunity to receive semi-annual retroactive performance payments based on their savings relative to a target amount. Only episodes assigned to a high volume cancer type, which CMS has designated as “reconciliation eligible,”2 will be included in these calculations.
CMMI has delineated a series of steps for determining performance-based payments based upon historical national and practice-specific spending adjusted for a variety of factors described below. In essence, benchmark prices are established for each episode in a performance period by comparing the episode to similar episodes during a baseline period and trending the resulting price forward and adjusting for the use of novel therapies on a practice-wide level. These benchmark prices are then discounted to come to a target price, which is summed with target prices for all other episodes at the practice during the performance period to create a target amount that is then compared with actual practice expenditures during the performance period. The difference between the two amounts is modified by the practice’s quality metrics, as well as geographic price variations and the effects of sequestration. The difference between the target amount and the practice’s actual expenditures (as modified) equals the practice’s Performance Based Payment for each 6-month performance period.3
In establishing the payment methodology, CMS has defined a variety of terms that can be somewhat confusing. The following descriptions may be useful for understanding their methodology and calculations:
- Baseline expenditures—Expenditures calculated for baseline period episodes.
- Baseline price—Modeled expenditures for a performance period episode calculated by comparing the episode to similar episodes during the baseline period.
- Benchmark price—Baseline price trended forward and adjusted for use of novel therapies on a practice level.
- Target price—Benchmark price with the application of the OCM discount.
- Benchmark amount—Sum of benchmark prices for all episodes at a practice during a performance period.
- Target amount—Sum of target prices for all episodes at a practice during a performance period.
- Actual expenditures—Sum of episode expenditures at a practice during a performance period.
The following is a more detailed description of the methodology for determining Performance Based Payments:
Step 1: Determining Episodes for the Practice’s Baseline
CMMI will define a baseline set of historical episodes that will be used to develop baseline episode expenditures on which target amounts will be based. The period used to determine this set of episodes is January 2012 through June 2015.4
An episode begins on the date associated with a trigger event. A Part B claim qualifies as a trigger event if it contains both an initiating cancer therapy and diagnosis of a cancer included in the model (that is, listed on the spreadsheet cited above). The Part B claim cannot have a place of service code indicating an inpatient hospital. When the trigger event is a Part B claim, the episode begins on the date of service on the Part B chemotherapy drug claim.
A Part D claim qualifies as a trigger event if it contains an initiating cancer therapy designated by CMS along with a Part B claim for a cancer diagnosis included in the model has occurred on the prescription fill date or in the preceding 59 days. Where the trigger event is a Part D claim, the episode begins on the fill date for the Part D chemotherapy drug claim.
An OCM episode lasts for 6 months,5 regardless of whether the beneficiary receives chemotherapy treatment throughout the period. Re-initiating chemotherapy within 6 months of the initial chemotherapy does not trigger a new episode—the re-initiated chemotherapy is considered part of the first episode. A triggering event after the end of the 6-month period triggers a new 6-month episode. There is no set limit on the number of episodes that a beneficiary can trigger during the OCM program period.
For an episode to be included in the OCM, the Medicare beneficiary must meet the following criteria throughout the 6-month period: 1) Enrollment in Part A and Part B; 2) Medicare as primary payer; 3) does not receive the Medicare End Stage Renal Disease benefit; and 4) not covered under Medicare Advantage or any other group health program; 5) has received chemotherapy treatment for cancer; and 6) has at least one Evaluation & Management (E&M) visit with an OCM-included cancer diagnosis during the 6 months of the episode.
Each episode will be assigned a cancer type based on which type is attributed to the most E&M visits. Additionally, each episode will be assigned to a practice based on which practice received the most E&M visits associated with an included cancer type diagnosis code over the course of the episode. These plurality determinations will be made on the individual practice level even where practices are pooled together for performance-based payment calculation purposes.
Step 2: Calculating Baseline Episode Expenditures
Next, Medicare FFS expenditures are totaled for all baseline period episodes included in calculations of performance-based payment (that is, those involving cancer types that are reconciliation-eligible). Baseline expenditures include reimbursement for all claims for which the service date is within the episode. Baseline episode expenditures include all Medicare Part A and Part B FFS expenditures. From Part D, they include only the Low Income Cost Sharing Subsidy (LICS) amount and 80% of the Gross Drug Cost above the Catastrophic (GDCA) threshold. Before finalizing baseline episode expenditures, CMMI will make adjustments 1) to account for overlap of OCM episodes and other CMS models; 2) to remove the effects of sequestration; 3) to trend all episode expenditures for the baseline period to the same level as expenditures for episodes ending in the most recent 6-month period included in the baseline (January – June 2015), and 4) to truncate episode expenditures at the 5th and 95th percentiles of per-episode expenditures by cancer type.
Step 3: Determining Baseline Price for Each Performance Period Episode
For each performance period episode included in calculations of performance-based payment, CMMI will predict baseline price associated with the specific characteristics of that episode, using a prediction model that will be calibrated using the national set of baseline episodes created in Step 1 and the baseline episode expenditures calculated in Step 2. The model will make adjustments by regressing baseline episode expenditures using a list of covariates determined to influence episode expenditures, including cancer type, age, sex, and selected comorbidities.
CMMI will then apply an adjuster to the baseline expenditures predicted by the model to reflect the experience of each practice. The adjuster will represent the ratio of average actual baseline expenditures for the practice’s baseline period episodes to the average predicted expenditures (using the same predictor model described above) for the practice during the baseline period. This adjuster will be weighted at 50%. The baseline price for each episode is the predicted baseline expenditures for the episode (calculated using the prediction model) multiplied by the weighted experience adjuster specific to that practice.
Step 4: Determining Benchmark Price for Each Episode for a Practice
The benchmark price for each episode is the baseline price for the episode multiplied by a trend factor and an adjustment for the use of novel cancer therapies, both specific to each practice. The trend factor will be based on episode expenditures for practices not participating in the OCM during the performance period adjusted using regression analysis to reflect the case mix of the OCM practice during the performance period.
Novel Therapy Adjustment
In the benchmark methodology materials CMS announced a “novel therapy adjustment” to account for a practice’s greater spending if it had a higher proportion of expenditures for newly-FDA approved oncology drugs compared to non-OCM practices. To be considered for inclusion in this adjustment, oncology drugs must have received FDA approval (for a new drug or a new indication) after December 31, 2014, and the use of the therapy must be consistent with FDA approved indications. Specifically, CMS states “Oncology drugs will be considered “new” for 2 years from FDA approval for that specific indication for the purpose of the adjustment.” [emphasis added] In addition, for purposes of calculating the adjustment for each performance period, CMS notes that the “new” designation may extend longer to align with OCM reconciliations.
CMMI will calculate the percentage of actual episode expenditures at the practice for the performance period associated with new oncology drugs. For Part B drugs the cost of new oncology drugs includes the full Medicare expenditure amount; for Part D drugs it includes the LICS amount and 80% of GDCA. CMMI will then compare that percentage to the percentage of actual episode expenditures associated with new oncology drugs among all episodes attributed to practices not participating in the OCM.
If the percentage for the OCM practice exceeds that for the non-participating practices, the difference between the percentages will be multiplied by the OCM practice’s actual expenditure amounts for the performance period. This product will then be multiplied by 0.8 (the policy factor). That number will then be divided by the sum of the practice’s trended baseline prices for the performance period’s episodes (i.e., the practice’s baseline amount). The resulting percentage is the adjuster by which the practice’s trended baseline prices are increased to arrive at per-episode benchmark price. (CMMI’s example calculation of this adjustment is at pp. 42 – 43 of its Methodology document).
CMS may also reduce the novel therapies adjustment in the future for medicines that have “lower clinical effectiveness,” but that would not occur any earlier than Performance Period 3 (i.e., for episodes beginning July 2, 2017 through January 1, 2018).
Step 5: Determining Target Price for Each Episode
The target price for each episode is calculated by applying the OCM discount to the benchmark price. Under the one-sided risk sharing arrangement the OCM discount is 4%; under the two-sided arrangement it is 2.75%.
Step 6: Determining Benchmark and Target Amounts for Each Practice
The benchmark and target amounts for each practice are the sum of the benchmark and target prices for all episodes during the performance period that are included in performance-based payment calculations. Episodes are assigned to a performance period based on the date the episode terminates.
Step 7: Determining Actual Expenditures for Episodes in a Performance Period
Episodes identified and attributed for each performance period will be defined in the same way as those for the baseline period as set out in Step 1.
Actual episode expenditures include expenditures for all claims where the service date is during the 6 month period (for episodes that are included in performance-based payment calculations). They include all Part A and Part B expenditures, some Part D expenditures (LICS + 80% of GDCA), and all OCM MEOS payments. Those amounts are adjusted, as with baseline episode expenditures, to account for model overlap, to remove the effects of sequestration, and to Winsorize outliers.
Step 8: Calculating Performance-Based Payments
To receive a performance-based payment, the practice’s episode expenditures for the performance period must be less than the target amount. This comparison is called reconciliation. The performance-based payment is equal to the difference between actual episode expenditures and the target amount,6 multiplied by the performance multiplier (which is discussed further below), adjusted for geographic variation, and reduced for sequestration. Performance-based payments for pools are calculated in substantially the same manner as for individual practices.
To be eligible for performance-based payment, the practice must also receive an Aggregate Quality Score (AQS) of at least 30%.7 The AQS is equal to total quality points earned divided by maximum quality points possible for the performance period. The performance multiplier (50%, 75%, or 100%) will also be based on AQS. AQS is calculated based on twelve quality measures in four domains—communication and care coordination, person- and caregiver-centered experience and outcomes, clinical quality of care, and patient safety.8 Some measures are calculated from claims data, others are practice-reported, and patient-reported experience of care will be determined by a survey of patients deployed by a CMS contractor. Through Performance Period 2, practice-reported metrics are “pay for reporting” while claims-based metrics are “pay for performance.”9 Beginning in Performance Period 3, all measures become pay for performance.10 Reporting and performance rates are converted into quality points, which are then compared against the maximum number of points to calculate AQS. Quality measures for pooled practices are scored in much the same way as for individual practices.
CMMI will calculate performance based payments for each performance period three times. The first reconciliation will include 2 months of claims run-out (claims submitted after the end of the performance period), the second reconciliation will include 8 months, and the third will include 14 months. These additional reconciliations may result in additional payments to a practice or in recoupments by CMMI. Calculations for each reconciliation will generally begin within 6 weeks of the last month of run-out for that reconciliation. Generally, results of the first reconciliation will be communicated to the OCM practices by the 8th month after the end of each performance period, and the results of the second and third reconciliations will be communicated 6 and 12 months later, respectively. That is, practices will get initial information about their performance 8 months after a performance period ends, or 14 months after the start of a performance period, e.g., for episodes beginning between January 2, 2017 and July 1, 2017 (Performance Period 2), practices will receive their first reconciliation information around August 2018.
Two risk arrangements will be available to participants in OCM:
- One-Sided Risk
- Under this model, the participating practice will be eligible to receive a performance-based payment if expenditures are below the target price. The target amount for the one-sided model is set at a 4% discount from benchmark amount. Participants in this one-sided model do not receive any financial penalty for failing to reach the target price. All participants are on this risk model for the first three performance periods, and participants may remain in this risk model for the entire five years, subject to the qualifications discussed below.
- Two-Sided Risk
- After the second year (Performance Period 3), participants can elect to change to a symmetric two-sided risk model. The target amount for this risk model is set at a 2.75% discount from benchmark amount. Under this model, the participant still receives a performance-based payment if expenditures are below the target amount, but is also financially responsible for expenditures that exceed the target amount. This recoupment is the difference between the actual expenditures and target amount, adjusted for geographic variation and reduced for sequestration.11 Under the newly proposed Quality Payment Program, which implements provisions of MACRA, this track will be considered an “Advanced Alternative Payment Model.”
Beginning with Performance Period 4 (episodes beginning 1/2/18 through 7/1/18), participants may switch between the risk models on a semi-annual basis. Participants that fail to achieve a performance-based payment by the time of the initial reconciliation of Performance Period 4 (i.e., August 2019) must exit the model or enter the two-sided risk arrangement until they achieve a performance-based payment.
As noted, 17 private payers participating along with Medicare including national payers like Aetna, Cigna, and regional payers. Private payers methodologies will be aligned with - but may be different from - Medicare’s for the OCM practices, but specific payment incentives may differ. The OCM payment methodology document released on June 29, 2016 explicitly states that it “reflects only the methodologies that will be used for Medicare FFS beneficiaries.” As to how private payers will be participating in the OCM, it says only that “there may be differences in certain model design aspects between the subset of OCM for Medicare FFS beneficiaries and the subset for other payer beneficiaries, such as specific payment incentives. However, the approach to practice transformation is consistent across the OCM.”
In June 2015, CMS indicated that payers have significant flexibility to design their own financial incentives and to define episodes, as long as the incentives align with those of the OCM and certain basic structural similarities are maintained. For example, CMS stated that it wanted other payers to align with its general two-pronged (enhanced services and actual performance) payment approach and that it would collaborate with payers to arrive at a basic set of quality measures applicable across all payers.
CMS also stated in June 2015 that private payers were encouraged, but not required, to begin their performance period within 90 days of the OCM performance period.
1. Practices may join together for the OCM for purposes of calculating their performance, and hence any bonus payments. For the purposes of this paper, the term “practice” will include both individual practices and pools of practices.
2. All episodes will get MEOS payments, but only “reconciliation eligible” cancer types will be eligible for inclusion in the calculation of Performance Based Payments.
3. Calculations for each 6-month performance period include spending for episodes of care (which are 6 months long) that end during that performance period. Thus, episodes that start in mid-July 2016 will end in the first performance period of 12/31/16 to 6/30/17.
4. That is, it includes episodes beginning on January 1, 2012 and after, and ending June 30, 2015 and before.
5. Unless the beneficiary dies.
6. The maximum expenditure reduction to which the performance multiplier can be applied is 20% of the practice’s benchmark. CMS calls this the stop-gain limit.
7.The practice must also report to the OCM Data Registry on all practice-reported quality measures and implement all Practice Redesign Activities.
8. Quality measures are excluded if there are insufficient episodes to calculate the measure reliably. For instance, three of the quality measures are unique to breast cancer and therefore will not be universally applicable.
9. Patient-reported experience of care is not included in Performance Period 1 and is pay for performance in Performance Period 2.
10. The performance rate for a pay for performance quality measure will be the average of the rates from the current performance period and the prior one.
11. This amount is subject to a maximum repayment of 20% of the benchmark amount.