Economists predict that, in less than 3 years, the average household in the United States will use 50% of its income on health insurance and out-of-pocket medical bills. According to Lee N. Newcomer, MD, MHA, Senior Vice President, Oncology, Genetics and Women’s Health, UnitedHealthcare, Minnetonka, MN, creating payment models that can reimburse physicians for cost-effective care is difficult in itself, but demonstrating that the new models are indeed cost-effective involves another set of hurdles.
“Payers and physicians must continue searching for better payment methods. Today’s methods do not work well enough, and more experiments are necessary to find ways to reduce costs and preserve quality,” said Dr Newcomer in a recent commentary (Newcomer LN. J Oncol Pract. 2014;10:187-189).
One of the prevalent reimbursement models used today is the pay-for-performance (P4P) approach. With this model, it is difficult to accumulate a sufficient number of patients to prove that an outcome is statistically significant. To prove a significant 10% difference in the cost of care, studies must include hundreds of patients; however, aside from Medicare, each payer usually only accounts for 5% or 15% of the patients in the practice. For example, UnitedHealthcare, one of the nation’s largest payers in the country, with more than 29 million members, has only 37 medical oncology groups that see more than 10 patients covered by UnitedHealthcare monthly.
P4P programs provide financial rewards to physicians for meeting prespecified goals, but according to Dr Newcomer, the rewards are often too small to provide an incentive. For example, in a hypothetical P4P program that saves 5% on chemotherapy drug costs or hospitalizations, physicians would receive an annual payment of $2008 for reduced hospitalizations and $7366 for reduced chemotherapy costs. “These sums are quite small when one considers the work effort required to obtain them,” commented Dr Newcomer.
Reimbursing for Value Increased costs for funding a program that improves outcomes must take into account the increase in value. According to Dr Newcomer, an unnamed cancer center recently requested a payment from UnitedHealthcare as a reward for its superior survival rates, but often the center’s rates were only a few percentage points better for a given tumor.
The center’s 5-year survival rates were 19% better than the national average for myeloma and 15% better for lung cancer, but much smaller gains were seen for pancreatic cancer (9%), breast cancer (3%), prostate cancer (2%), colon cancer (1%), and liver cancer (1%). The reimbursement the cancer center requested was not in line with the savings that were actually achieved.
A recent study involving 5 medical groups that treated 810 patients with cancer demonstrated that modifying the fee-for-service payment system for therapy for cancer with feedback data and financial incentives that rewarded outcomes and cost-efficiency reduced the predicted total medical costs by 34% (Newcomer LN, et al. J Oncol Pract. 2014 Jul 8. Epub ahead of print). This study, however, was not designed to determine what expenses caused the difference in the total medical cost.
Dr Newcomer suggests that “it may be time to measure something new. Measuring the effectiveness of specific chemotherapy regimens rather than physician groups may have more value.” This will allow Medicare and some commercial payers to more quickly accumulate large numbers of patients with certain types of tumor in their data sets for comparative effectiveness analyses to compare different regimens in terms of their benefits and total costs. “Regimens could be compared for total costs, toxicities and most importantly outcomes,” he noted.
This will allow payers to give preferential status to superior regimens or nonpreferential status to inferior regimens. This strategy will allow payers to pay for value.
In the search for new payment models, “payers and physicians should remember that the same scientific method which was used to test new therapies and improve processes of care—particularly adherence to statistical rigor when making efficacy claims—should be used to provide equally important insights in the affordability and value of care,” emphasizes Dr Newcomer.
