Oncology Rounds

What can we learn from United’s medical oncology episode-based payment pilot?

by Lindsay Conway

Last week United Healthcare announced that their medical oncology episode-based payment pilot reduced cancer care costs by 34%, or approximately $40,000 per chemotherapy patient. Those results would be remarkable for any value-based reimbursement model, but they are especially exciting in cancer care. To date, other oncology payment pilots have resulted in cost savings of only a few hundred dollars per cancer patient, if any.

Surprisingly, the pilot participants achieved those cost savings despite spending 179% more on chemotherapy drugs than the comparator group. The results raise new questions about how to realign incentives to promote value and whether episode-based payments are feasible in medical oncology.

Goals to reduce cancer costs, decouple drug choice from physician income

The primary goal of the pilot was to reduce the overall medical costs of cancer care. United hypothesized that by removing the economic incentive for oncologists to prescribe higher-cost chemotherapy drugs, they would reduce spending.

To that end, United reimbursed participating practices for chemotherapy drugs at average sales price (ASP). They then added an episode-payment to replace the drug margin and fee for service payments for physician hospital care, hospice management, and case management. All other services, including physician office visits, chemotherapy administration, and diagnostics, were paid on a fee-for-service basis.

Takeaway #1: United plans to scale up the pilot – a little

In the past, both United Healthcare and the participating oncology practices had reported that the administration of the episode-based payments was labor intensive, requiring separate manual billing processes. As a result, neither party was interested in expanding the pilot to include more patients.

However, the magnitude of the cost savings generated by the pilot may well justify investment in automated patient enrollment and claims systems, which would make it possible to expand the model more broadly. As a next step, United has indicated that it plans to continue the pilot and triple the number of participating practices.

Takeaway #2: Data on oncology performance improvement opportunities remain elusive

Unfortunately, like other oncology payment pilots, United’s study did not generate enough data to yield statistically significant conclusions about what led to the cost reduction. Nor did it yield much data on clinical quality, although quality outcomes appear to be similar between the pilot participants and comparator group.

A subset analysis revealed that the pilot participants had lower hospitalization rates and lower use of radiation therapy. These changes may have been caused by the bundling of physician payment for hospital care and hospice management.

Anecdotally, there were variations across the pilot practices in service utilization. For example, one practice had a significantly higher hospitalization rate. The practice’s leadership determined that after being discharged from the hospital, patients often waited several weeks for follow up appointments in the clinic. As a result, patients were frequently readmitted for the same problems.

To address the problem, the practice changed their processes to ensure post-discharge patients could be seen in the clinic within 48 hours of discharge. As a result, their hospitalization rates now match the other practices.

Another notable variation across the pilot participants was a four-fold difference in the frequency of imaging for patients with metastatic disease. Although the physician leaders discussed the issue, they were unable to reach consensus about the appropriate use of imaging for metastatic patients.

Takeaway #3: Care delivery innovation—not drug choice—may present the best opportunity for cost savings

Given the high price of chemotherapy drugs, and the fact that traditional fee-for-service payment creates an economic incentive for physicians to prescribe higher cost therapies, many industry observers have assumed that controlling drug spending would be key to reducing cancer care costs. However, the results of this pilot suggest that’s not necessarily the case.

As noted above, the pilot participants decreased overall costs by 34%, despite spending 179% more on chemotherapy. Although it’s not clear what accounts for this seeming contradiction, there are several possible explanations:

  • The pilot participants prescribed therapies that, although they were most costly, resulted in fewer complications, ED visits, and hospitalizations, thus reducing overall spending.
  • The pilot participants invested in other care delivery innovations (e.g. pre-chemo education, advanced directives, remote symptom management) that reduced overall costs more than enough to offset the additional spending on chemotherapy.
  • The episode payment replaced physician payment for hospital care, creating a small economic incentive for physicians to reduce hospitalizations.
  • The episode payment included modest reimbursement for case management, which is not typically reimbursed under traditional fee-for-service. The additional revenue may have made it more feasible for the practices to dedicate resources to non-reimbursed activities that affect patient outcomes, e.g. care coordination and psychosocial support.

Takeaway #4: Not clear that these results are replicable

An important question is whether these results can be replicated at other oncology practices. The five original practices participating in the pilot are by no means representative of oncology practices nationally. Most, if not all, of the pilot participants have risk-based contracts with other payers and have invested in care delivery innovations designed to increase quality and reduce costs.

Consequently, it’s possible that United’s pilot had little to no impact on these practices. Rather, they were implementing – or had implemented - care innovations for other reasons which contributed to the cost savings.

More study is needed

The pilot results generate more questions than answers. Even the study authors admit, “the source of the cost savings is enigmatic.”

We’ll continue to track new developments as the United pilot proceeds, and also report on the outcomes of other risk-based contracts in oncology.


Join us at Wednesday's Oncology Medical Homes Update webconference, where we'll discuss the applicability and feasibility of the medical home model for different organizations.