On March 7, 2019, the Fifth Circuit Court of Appeals affirmed CMS's demand for the return of $8 million in premiums granted to a care management demonstration project that failed to achieve its projected cost savings.

Texas Tech Physician Associates had developed a five-year demonstration program designed to improve care management for high-cost beneficiaries through intensive management, care coordination and intervention services. CMS agreed to pay a monthly per-member management fee that was contingent on achieving a 5 percent reduction in net costs.

Less than a year after inception, the intervention group's costs were 7 percent higher than the costs of the control group. After 16 months, the physician group exercised its early termination option, and the final reconciliation report found a savings shortfall of $16.79 million. CMS then demanded a refund of the $8 million in contingent management fees.

This demonstration project was initiated 13 years ago, and the landscape has changed considerably since that time with respect to alternative payment models and the technologies that may be available to help manage care. Nevertheless, this case illustrates lessons that may be useful to providers in evaluating their own cost containment arrangements.

Appropriate criteria for measuring success

The demonstration project's cost savings were measured against the costs of a control group, which was developed by an independent evaluator. One fundamental problem was that the control group was populated with randomly assigned low-cost, medium-cost and high-cost patients, whereas the intervention group was designed for high-risk patients and contained more nursing home residents and patients who died during the demonstration. The required services and costs for these groups may not have been comparable.

Despite its initial concerns, the physician group assented to the evaluator's proposal for designing the control group. When the first biannual results were issued, the physician group asked CMS to adjust the composition of the control group but CMS declined.

The court of appeals held that the group was not entitled to redesign the criteria and methodology of the project. Providers are therefore advised to consider carefully how success should be defined, whether control groups are defined appropriately, and which specific criteria should be used to determine outcomes.

Access to data

The demonstration agreement required CMS to provide retroactive demographic and claims data for the intervention group at the inception of the project. Any other ongoing data would be provided upon request and approval by CMS.

When the physician group later requested ongoing claim data for the intervention and control groups, CMS required the physician group to execute a Data Use Agreement, which it did not approve until over a year after the demonstration project began. The physician group asserted that it might have been able to better monitor its performance if it had received the data much sooner. The court of appeals held that CMS fulfilled its obligations under the agreement.

As providers consider the appropriate criteria for measuring success, they should also consider the data that will be needed to support those measurements. The demonstration agreement should clearly identify the data that CMS must provide, and require the ongoing delivery of such data from the effective date forward. Moreover, the parties should execute a Data Use Agreement and any other required authorizations before the project is initiated.

No contractual remedies

The court of appeals held that the demonstration agreement was a grant, rather than a governmental procurement contract.  As such, the court held that any disputes must be resolved through the HHS Departmental Appeals Board process, and the board's decisions are presumptively valid.

More importantly, common-law contractual theories do not apply to the analysis and resolution of disputes. Thus, the Departmental Appeals Board is not required to consider any contractual claims or defenses, such as breach, mistake or impracticability.

In a demonstration project, the provider's rights and remedies must be expressed in the grant agreement. The provider will not be allowed to avail itself of extraneous common law contract theories in the event of a later dispute. Providers should make every attempt to predict the contingencies that may affect performance and address those matters in the agreement.

Many of the current alternative payment models and demonstration projects being administered by CMS require the provider to assume some level of financial risk. Providers who are considering such cost containment arrangements should consult with experienced health counsel in developing the parameters and expectations of the parties before proceeding. A few targeted safeguards could result in substantial future benefits.

Find out more about the implications of the new opinion for your healthcare business by contacting the author.