Study raises concerns about safety-net TEAM participation

Many hospitals required to participate in an upcoming bundled payments model from CMS could bear disproportionate penalties based on their size and patient populations, according to a recent study.

The study, published by researchers from Harvard University and Washington University in JAMA Health Forum, analyzed hospital characteristics of over 700 facilities mandated to participate in the Transforming Episode Accountability Model (TEAM) versus nonparticipants.

Researchers found that hospitals selected to participate in TEAM were larger and served more marginalized patients. They were also often part of the healthcare safety net, raising concerns about financial penalties under the bundled payments model.

Safety-net hospitals required to participate in previous CMS bundled payments models were disproportionately penalized despite reducing spending as much as other hospital participants. Safety-net hospitals generally had lower, and therefore harder-to-meet, spending targets, resulting in skewed penalties, researchers explained.

TEAM calculates spending targets similar to other CMS models, like the Comprehensive Care for Joint Replacement (CJR) model, using historical regional spending data on specific clinical episodes from the previous three years minus a discount. The target includes the costs of the hospital inpatient stay, any readmissions, physician and practitioner services, post-acute care, outpatient services, durable medical equipment and hospice services within the 30 days after discharge.

If the total costs across episodes exceed the spending targets, hospitals must repay CMS the excess amount. However, if total costs are lower, they receive a reconciliation bonus.

The study found a major opportunity to reduce spending on the clinical episodes in TEAM based on mandatory participation. After all, TEAM participants had higher spending compared to nonparticipants.

However, nearly 14% of hospitals required to participate in TEAM were considered part of the safety net, versus about 9% of nonparticipating hospitals. More participating hospitals were also nonprofit (71% versus 65%) and served more patients with dual Medicare and Medicaid enrollment (26% versus 23%).

Patient populations served by safety-net hospitals tend to have more complex medical needs and face socioeconomic barriers to care, affecting total costs of care for most clinical episodes.

CMS has acknowledged these unique challenges for safety-net hospital participants, allowing them to remain in an upside risk-only track for up to three years versus one year for non-safety-net participants. However, after that, safety-net hospitals will have to transition to a track with downside financial risk.

The Alliance of Safety-Net Hospitals (ASH) has called on CMS to make TEAM participation voluntary over concerns about penalties and the safety net's ability to participate.

"We appreciate that the model design permits safety-net hospitals to remain in Track 1 and avoid downside risk, but mandated participants will still be forced to take on new costs and losses while they prepare for the model and strive to increase quality performance for the surgical episodes involved. No downside risk does not mean there is no detriment to the safety-net participants," ASH wrote in a June 2025 letter to CMS.

America's Essential Hospitals, which also represents safety-net hospitals, has also urged CMS to allow safety-net participants to remain in the upside risk-only track for the duration of the five-year model. The group also advised CMS to provide upfront infrastructure payments or targeted technical assistance grants to help safety-net hospitals succeed in TEAM if the model must remain mandatory.

"As CMS has acknowledged in previous rulemaking, safety net hospitals frequently operate with limited access to capital and narrower financial margins due to the high proportion of uninsured and publicly insured patients they serve," America's Essential Hospitals stated. These resource constraints often impede the ability to invest in critical infrastructure required to succeed in a value-based care model with financial risk."

TEAM is slated to launch on Jan. 1, 2026, with approximately 743 hospitals in over 180 core-based statistical areas required to participate.

Jacqueline LaPointe is a graduate of Brandeis University and King's College London. She has been writing about healthcare finance and revenue cycle management since 2016.