On April 10, 2024, the Centers for Medicare & Medicaid Services (CMS) issued the fiscal year (FY) 2025 Hospital Inpatient Prospective Payment Systems (IPPS) for Acute Care Hospitals and the Long-Term Care Hospital (LTCH) Prospective Payment System proposed rule. See the press release here. The rule included a proposal for a mandatory payment model called the Transforming Episode Accountability Model (TEAM). A fact sheet for the model can be found here. CMS also released answers to frequently asked questions on the model, which can be found here.
OVERVIEW
The (TEAM) model is a mandatory, episode-based payment model whose performance period would start January 1, 2026, and last for five years, ending on December 31, 2030. TEAM participants would be limited to acute care hospitals paid under the Hospital Inpatient Prospective Payment System (IPPS). An episode would be initiated by one of the following surgical procedures – lower extremity joint replacement, surgical hip femur fracture treatment, spinal fusion, coronary artery bypass graft, and major bowel procedure – and would last until 30 days after the patient’s discharge from the hospital. During this period, the hospital would be responsible for cost and quality of care. Participants would be presented with three tracks representing different levels of financial risk. Track 1, which would be available to all participants for the first year of the model, would only have upside risk, to allow participants to become familiar with the model before taking on risk.
Participants would continue to bill Medicare Fee for Service (FFS) during the model. To incentivize hospitals to control costs, CMS would set price targets for episodes based on predicted Parts A & B spending. Depending on their track, hospitals could earn a payment from CMS if they spend below this target or be subject to a recoupment if they exceed the target. This payment/recoupment would be subject to a pay-for-performance quality adjustment, the magnitude of which would depend on a participant’s risk track. Similar to past episode-based models like the Comprehensive Care for Joint Replacement (CJR) Model and the Bundled Payments for Care Improvement Advanced (BPCI Advanced) Model, participants would also be subject to stop-gain and stop-loss limits, applied after their quality adjustment.
CMS will utilize stratified random sampling methodology to select Core Based Statistical Areas (CBSAs) for participation in the TEAM model. Every hospital within a selected CBSA would be required to participate. For a full list of eligible CBSAs see page 1119 of the unpublished rule. CMS plans to select roughly a quarter of eligible CBSAs. Due to the stratified random sampling methodology, a hospital’s chance of being selected would depend on the stratum their CBSA is assigned to and would range from 20 to 50 percent. The stratum each CBSA is assigned to can be found starting on page 1119 of the unpublished rule and the probability of a CBSA within each stratum being selected for the model can be found on page 1135.
To encourage collaboration with physicians, TEAM participants would be permitted to share reconciliation payments with certain types of providers and suppliers[1], deemed TEAM Collaborators. Agreements between participants and collaborators would need to documented in writing and made available to CMS upon request.
The Center for Medicare and Medicaid Innovation (CMMI) intends to use TEAM to build on lessons learned from past episode-based models like the CJR Model and the BPCI Advanced Model. If TEAM is successful, CMS hopes the model can “establish the framework for managing episodes as a standard practice in Traditional Medicare.”
EPISODES
There would be five episode categories under the model, the Lower Extremity Joint Replacement (LEJR) episode category, the Surgical Hip and Femur Fracture Treatment (SHFFT) episode category, the Coronary Artery Bypass Graft (CABG) episode category, the Spinal Fusion episode category, and the Major Bowel Procedure episode category.
An episode would begin on the day of the outpatient procedure or date of admission on the IPPS claim associated with the inpatient hospitalization and would end 30 days following discharge from this procedure or hospitalization. The MS-DRG and HCPCS codes that would initiate an episode can be seen in Table 1. below:
Table 1. Proposed Categories and Billing Codes[2]
| Episode Category | Billing Codes (MS-DRG/HCPCS) |
| LEJR | MS-DRG 469, 470, 521, 522 HCPCS 27447, 27130, 27702 |
| SHFFT | MS-DRG 480, 481, 482 |
| CABG | MS-DRG 231, 232, 233, 234, 235, 236 |
| Spinal fusion | MS-DRG 453, 454, 455, 459, 460, 471, 472, 473 HCPCS 22551, 22554, 22612, 22630, 22633 |
| Major bowel procedure | MS-DRG 329, 330, 331 |
The decision to use 30 days was based largely on trying to keep the model focused on costs the hospital could control, with CMS worrying that longer episode lengths would increase the risk of hospitals being impacted by medical events outside of the intended scope of the TEAM model. Similar to CJR, if a patient was transferred from a hospital after initiating an episode with that hospital, the transfer would not initiate a new episode but would be included in the existing episode.
All items and services paid under Parts A & B would be included in an episode, outside of an exclusion list designed to ensure that the model only captures services related to the anchor hospitalization or procedure. For a full list of exclusions, see pages 1150-1153 of the unpublished proposed rule.
PAYMENT
On an annual basis, CMS would compare each participant’s spending on episodes for that performance year to their target prices. Consistent with the CJR model, reconciliation would occur 6 months after the end of the performance year to allow for half a year of claim runout. At the time of reconciliation, CMS would apply a series of adjustments, such as beneficiary-level risk adjustment, to target prices. The difference between the adjusted targets and participant spend would be the reconciliation amount. This reconciliation amount would then be adjusted based on participant CQS scores. Stop-loss and stop-gain limits would then be applied to this quality adjusted reconciliation amount to create the Net Payment Reconciliation Amount (NPRA). Finally, the NPRA would be combined with a post-episode payment calculation, designed to ensure that hospitals were not deferring procedures until after the episode had concluded, to create the final reconciliation amount for each participant.
Stop-Loss & Stop-Gain Limits
Stop-loss/gain limits reduce the financial risk for model participants. Under these limits, a participant’s gains or losses are capped at a percent of their target for a given performance year. For example, if a participant’s target was $100, and the stop-gain/stop-loss limits were 10 percent, then the maximum the participant could earn would be $110, and the minimum the participant could earn would be $90.
Post-Episode Payment Calculation
CMS is concerned that, because episodes only include 30 days following a patient’s inpatient discharge or outpatient procedure, there would be an incentive for hospitals to delay necessary care past this 30-day cut off. To reduce this incentive, CMS would calculate all Parts A & B spend in the 30-day period following each episode. If post-episode spending was greater than three standard deviations above the regional average, then the difference between this three standard deviation threshold and the post episode spend would be either added to the repayment amount the hospital owed CMS or subtracted from the reconciliation amount CMS owed the hospital.
High-Cost Outlier Cap
To avoid penalizing hospitals for outlier cases where episode costs were extremely high for reasons outside of the hospital’s control, the model would include a high-cost outlier cap. Under this cap, CMS would calculate the 99thpercentile episode cost for each of the 216 MS-DRG/HCPCS episode type and region combinations. This 99thpercentile cost would function as the cap, with episode costs above this value being set to the cap.
TRACKS
The TEAM model would have three tracks. Track 1 would have no downside risk and would be available to all participants in the first year of the model. Track 2 would only be available in years 2-5 of the model, and only for the following types of participants: safety net hospitals, rural hospitals, Medicare dependent hospitals (MDHs), sole community hospitals (SCHs), and essential access community hospitals. To reduce the risk taken on by these participants, Track 2 would have lower stop-loss and stop-gain limits, and allow for larger reductions to repayment amounts owed to CMS via CQS adjustments. Finally, Track 3 would be used by all participants who did not qualify for Track 2 after the first year of the model. The various stop-gain/stop-loss limits and CQS adjustment limits associated with each track can be seen in the table below:
Table 2. Financial Risk by TEAM Track[3]
| Track | Financial Risk |
| Track 1 | • Upside risk only (10% stop-gain limit) • CQS adjustment percentage of up to 10% for positive reconciliation amounts |
| Track 2 | • Upside and downside risk (10% stop-gain/stop-loss limits) • CQS adjustment percentage of up to 10% for positive reconciliation amounts and CQS adjustment percentage of up to 15% for negative reconciliation amounts |
| Track 3 | • Upside and downside risk (20% stop-gain/stop-loss limits) • CQS adjustment percentage of up to 10% for positive and negative reconciliation amounts |
Beginning in the second year of the model, Track 2 participants would have the option to switch between Track 2 and Track 3 on an annual basis.
BENCHMARK PRICES
The benchmark price for each episode would be calculated separately for each MS-DRG/HCPCS episode type and region[4]. This price would be based off the average standardized spending for each episode type/region combination over three years of data. For performance year one of the model (beginning January 1, 2026), episodes initiated between January 1, 2022 and December 31, 2024 would be used to calculate benchmarks. This data would then be rolled forward one year for each performance year, so episodes initiated between January 1, 2023 and December 31, 2025 would be used to set benchmarks for performance year 2, and so on and so forth. More recent years would be weighed more heavily in calculating benchmarks. Under this methodology, TEAM participants would be rewarded for being more efficient than other hospitals within their region.
Preliminary target prices would be provided to participants by the end of November prior to the performance year beginning in January.
CMS is concerned that using regional targets, as opposed to hospital specific targets, could hurt safety net hospitals, who were disproportionately likely to owe repayments under CJR after the model moved to regional targets. Accordingly, the agency seeks comments on “adjustments to regional target prices for Track 1 or Track 2 TEAM participants, that would decrease the likelihood of safety net hospitals being disproportionately penalized by regional target prices.”[5]
COMPOSITE QUALITY SCORE
CMS believes that payments under the TEAM model should incentivize not only lower costs but also better patient outcomes. Accordingly, the agency will link payments to quality of care via payment adjustments based off performance on quality measures. The combined score across these measures is referred to as the Composite Quality Score (CQS).
There would be three quality measures initially used in the TEAM model, all of which are already reported under the Hospital IQR and/or HAC Reduction Programs. This means that there would be no additional quality data collection required for TEAM participants, as CMS would simply analyze data already submitted through those programs. Quality measures for TEAM participants would be publicly available online starting in 2027.
The three measures are:
- Hybrid Hospital-Wide All-Cause Readmission Measure with Claims and Electronic Health Record Data (CMIT ID #356), which would apply to all episode types,
- CMS Patient Safety and Adverse Events Composite (CMS PSI 90) (CMIT ID #135), which would apply to all episode types, and
- Hospital-Level Total Hip and/or Total Knee Arthroplasty (THA/TKA) Patient-Reported Outcome-Based Performance Measure (PRO-PM) (CMIT ID #1618), which would only apply to LEJR episodes.
CMS is also considering adding the Thirty-day Risk-Standardized Death Rate among Surgical Inpatients with Complications (Failure-to-Rescue) (MUC2023-049) measure, which calculates the percentage of “surgical inpatients who experienced a complication and then died within 30-days from the date of their first “operating room” procedure,” to all episode categories.
PAYMENT CALCULATION
The CQS score for a given measure would be the percentile of the participant’s score among a national cohort of hospitals. This cohort score distribution would be calculated for CY 2025, and this CY 2025 data would be used to determine percentiles for the duration of the model. So, despite the use of percentiles, it would be theoretically possible for all TEAM participants to score 100 on CQS if each participant scored higher on all quality measures than any hospital in CY 2025.
For example, if a participant’s score for a measure landed in the 55th percentile, their CQS score for that measure would be 55. CMS would weigh the measure scores for each participant based on volume, so if 50 of a hospital’s 100 episodes were scored using the Hybrid Hospital-Wide Readmission (CMIT ID 356) measure, that measure would be assigned a weight of .5. The final CQS score would simply be the sum of the weighted measure scores:
Table 3. CQS Score Calculation
| Measure | Score | Weight | Weighted Score |
| CMS Patient Safety and Adverse Events Composite (CMIT ID 135) | 55 | 0.4 | 22 |
| Hybrid Hospital-Wide Readmission (CMIT ID 356) | 43 | 0.4 | 17.2 |
| Hospital-Level, Risk-Standardized Patient-Reported Outcomes Following Elective Primary Total Hip and/or Total Knee Arthroplasty (CMIT ID 1618) | 62 | 0.2 | 12.4 |
| CQS | — | — | 51.6 |
The amount to which this CQS score would influence reconciliation amounts would vary based on the participant’s track.
Table 4. CQS Adjustment Percentage Calculation[6]
| Track | Reconciliation Value | CQS Adjustment Percentage Formula |
| Track 1 | Positive | = (10%-10% * (CQS/100)) |
| Track 2 | Positive | = (10%-10% * (CQS/100)) |
| Track 2 | Negative | = (15% * (CQS/100)) |
| Track 3 | Positive | = (10%-10% * (CQS/100)) |
| Track 3 | Negative | = (10% * (CQS/100)) |
For example, if the CQS score of 51.6 in table 3 were awarded to a TEAM participant in Track 3 with a positive score, the CQS adjustment percentage would be calculated as follows:
CQS Adjustment Percentage = .1 – (.1 * (51.6/100)) = .1 – (.1* 0.516) = .1 – 0.0516 =
.0484 or 4.84%.
The CQS adjustment percentage would then be multiplied by a participant’s reconciliation amount to determine the CQS adjustment amount, which would then be subtracted from the participant’s reconciliation amount to create their quality-adjusted reconciliation amount. This means that, for participants receiving a positive reconciliation amount, the quality-adjusted reconciliation amount will always be lower their original reconciliation amount unless they receive a score of 100. Similarly, for participants who owe a repayment to CMS, quality-adjusted repayment amount will always be lower than their original repayment amount unless they receive a score of 0.
Table 5. CQS Adjustment Applied to Track 3 Participant
| Reconciliation or repayment | Value | CQS | CQS Adjustment Percentage | CQS Adjustment Amount | Quality-Adjusted Reconciliation Amount |
| $38,000 | Positive | 51.1 | 4.89% | $1,858 | $36,142 |
| $38,000 | Negative | 51.1 | 4.89% | $1,858 | $36,142 |
********
This Applied Policy® Summary was prepared by Will Henkes with support from the Applied Policy team of health policy experts. If you have any questions or need more information, please contact him at whenkes@appliedpolicy.com or at (202) 558-5272.
[1] The following provider and supplier types could qualify as TEAM collaborators: Skilled Nursing Facilities, Home Health Agencies, Long-Term Care Hospitals, Inpatient Rehabilitation Facilities, Physicians, Nonphysician practitioners, Therapists in a private practice, Comprehensive Outpatient Rehabilitation Facilities, Providers or suppliers of outpatient therapy services, Physician Group Practices, Hospitals, Critical Access Hospitals, Non-physician provider group practices, Therapy group practices, and Medicare ACOs
[2] TABLE X.A.-04 on page 1147 of the unpublished rule
[3] TABLE X.A.-01 on page 1114 of the unpublished rule
[4] Per CMS, this would result in 216 different MS-DRG/HCPCS episode type/region-level benchmark prices
[5] Page 1190 of the unpublished rule
[6] TABLE X.A.-08 on page 1221 of the unpublished rule