Everyone wants to pay for value (and certainly not just for volume) and the value imperative is nowhere greater than in cancer care. Cancer care cost the US health care system $125 billion in 2010, accounting for 5 percent of total health care spending. Before Sovaldi hit the front pages for offering a breakthrough treatment for Hepatitis C at a cost of $84,000, and Turing Pharmaceuticals pushed Sovaldi out of the headlines for increasing the price of an HIV drug 50 fold, most of the attention to drug pricing was focused on cancer.
With average yearly treatment costs exceeding $100,000, cancer drugs in the US cost nearly twice as much as anywhere else in the world. Drug pricing is driven by state and federal policies, as well as clinical advances that allow the development of specific treatments for targeted populations. It can be difficult to find any correlation between a drug’s price and the social value or efficacy of the drug. We recently published a case study comparing three cancer treatments, one for colorectal cancer and two for multiple myeloma, illustrating how the net health benefits versus costs for new treatments can vary widely.
Medicare has a great deal at stake in the pursuit of value in cancer care because cancer incidence is highly correlated with age. The Center for Medicare and Medicaid Innovation (CMMI) has announced theOncology Care Model (OCM) for physician practices that administer chemotherapy to Medicare patients. Its key features are a monthly care management payment for each patient throughout the six-month episode as well as performance-based payments. If practices are able to spend less than their target amount for the care episode, they will receive a portion of the savings in a performance payment, with the exact amount determined by their scores on quality measures.
Bundles have a natural logic as a means for promoting lower cost for a given episode of care. With a fixed sum of money available, those responsible for organizing and providing care have strong incentives to eliminate unnecessary care and find the most efficient care partners.
The challenges associated with turning this logical case into a practical application are well known. Bundles need a clear beginning and ending point. They need to adjust for underlying patient risk and potentially exclude high-cost/low-frequency elements of treatment where the cost saving imperative of the bundle could yield under-provision of high-value care. Controlling the cost of each bundle does nothing to reduce the number of bundles provided — in fact it creates incentives to increase them.
Even with strong incentives to control the cost of the bundle, providers may lack the information they need to pursue the highest value care for their patients. As policymakers focus on the question of how much bundled payments can save, clinicians see a treasure trove of clinical data on a large number of patients with the same diagnosis. How can we harness those data across the health care system to learn which treatments work best for which patients?
The cancer bundle is, for obvious reasons, focused on oncologists. Yet, care for cancer patients is spread across hospitals, payments for drugs, and a broad range of health professionals. Bundles paid after a diagnosis of cancer fail to engage possibilities for cancer prevention and do little to alter patterns of cancer detection. While the oncologist may lead the care team at certain points, most patients ultimately return to their primary care clinician for ongoing care. How does a shift to bundles for a portion of the continuum of care affect care outside the bundle?
These are great questions to ponder, and I will have the pleasure of posing them to panelists in a session I will moderate at ECRI Institute’s 22nd Annual Conference to be held November 17-18, 2015 at the Omni Shoreham Hotel in Washington, DC. Health Affairs is pleased to once again be a sponsor of this free event. My panelists will be Patrick Conway, Chief Medical Officer of CMS, Lee Newcomer, Senior Vice President at UnitedHealthcare, and Richard Roberts, Professor of Family Medicine at the University of Wisconsin. It is sure to be an interesting conversation.