Reforming the Payment System for Medical Oncology
Peter B. Bach, MD, MAPP
Infusion of chemotherapy has long been a profit center for medical oncologists. Under the current system, oncologists purchase cancer drugs at one price, then bill insurers at a higher price after they administer the drugs. However, the profits available to oncologists from this practice have been under serious pressure from the pharmaceutical industry and insurers. Pharmaceutical companies have rapidly increased drug prices while insurers, including the Medicare program, have systematically decreased the reimbursement they provide above the cost of the drug. Even though rising drug prices mean, in general, larger absolute profits to oncologists when they administer more expensive drugs (because the amount of profit is proportional to the purchase price of the drug), the decline in profit margins has made running oncology practices more financially challenging. Increases in drug prices mean increasing costs of inventory, and the resulting lower margins mean less financial flexibility when patients or insurers do not pay their bills on time or in total.
The current situation is far different from how it once was. Throughout the 1990s and early 2000s, Medicare reimbursed physicians based on manufacturers’ published wholesale prices even as physicians were acquiring the drugs at a substantial discount. In general, profits were larger when drug prices were higher, which paradoxically meant that the appeal to physicians of administering a chemotherapy drug would increase as the cost of that drug increased. As a result, the price of a typical new cancer drug entering the market increased from a few hundred dollars per month to a few thousand dollars over this period.1 From 1980 to 1989, the median monthly price of a newly approved anticancer drug was $396. By the early 1990s, the median monthly price had increased to just over $1600. The median monthly price of anticancer drugs approved between 2000 and 2005 was more than $4000. (All prices are in 2010 dollars.) However, in 2005, Medicare started to limit profits, reducing the cancer chemotherapy reimbursement level to 6% above the drug purchase price (based on actual prices); this year, the sequester reduced the reimbursement level to 4.2%. President Obama’s fiscal year 2014 budget includes a reduction to 3%.
Lobbyists for oncology physicians are out and Congress is listening. As of June 21, 2013, 91 members of the House of Representatives have sponsored a bill called the Cancer Patient Protection Act of 2013, which would restore the profit margins to 6%.2 One hundred twenty-four members of Congress wrote a letter to the Centers for Medicare & Medicaid Services (CMS) asking them to do the same. The American Society of Clinical Oncology and other associations such as the US Oncology Network have stated that Medicare patients will face access problems if reimbursement levels continue to decline.3 However, Medicare appears to have limited options, and it is difficult to gauge how Congress will act in these restrained fiscal times.
The current perceived crisis should trigger a more comprehensive consideration of how oncology medications should be financed, and what role, if any, profits linked to drug prescribing should play. Three cancer payment reform approaches have been proposed that address the link between cancer drug prices and physician profits.
Episode-based payment creates specific incentives for oncologists to select drugs based on price, but in this case profits increase when physicians choose less expensive rather than more expensive drugs.4 This can be achieved by replacing the current payments to physicians, which are linked to each drug administered, with a single payment to cover the cost of drugs and related services for a period termed an episode. The single payment would be approximately equivalent to the typical cost of caring for a typical patient with the relevant condition but explicitly not linked to the use of one drug vs another. In this scenario, oncologists still purchase cancer drugs before administering them, but they would be expected to purchase less expensive drugs that are appropriate for each patient so they can retain more of the global payment amount as a profit.
For instance, medical oncologists caring for patients with breast cancer might select from one of the less costly rather than more costly preferred regimens for adjuvant treatment. The different drug regimens listed in the National Comprehensive Cancer Network guidelines range from $3007 to $22 570 over the entire prescribed course of adjuvant treatment (based on 2009 reimbursement levels).4 In this payment method, physicians would retain flexibility in treatment choice. How to define similar groups of patients for each unique episode and calculate the size of the global payment are potential challenges of this approach.5
Payment for pathway adherence financially incentivizes oncologists to follow certain treatment approaches (the pathways) in preference to others. These pathways tend to favor lower-priced drugs rather than higher-priced ones when efficacy and toxicity are comparable. One study estimated a 35% cost reduction from following pathways rather than usual prescribing practices.6 Operationally, physicians receive higher overall payment in the form of wider profit margins and add-on payments if they follow the pathways most of the time. Compared with episode-based payment, this approach has physicians both buying and billing for cancer drugs but limits treatment flexibility. Getting physicians to agree on which pathways are appropriate and updating the pathways when new treatment paradigms emerge are challenges to this approach.
Third-party buy-and-bill enables physicians to order the drugs needed to treat a particular patient without having to buy or bill for that drug. Those functions are performed instead by an intermediary who procures the drugs from pharmaceutical companies, ships the medications to the physician’s office on an as-needed basis, and collects reimbursement from the patient and his insurer. A version of this approach, covering nearly all physician-administered drugs, was explored unsuccessfully in Medicare under a provision of the Medicare Modernization Act of 2003 called the Competitive Acquisition Program.7 Private insurers have explored a pared-down version of this approach, in which they contract with a specialty pharmacy to provide certain expensive drugs to physician’s offices when they are prescribed. This approach most closely aligns with the principle that oncologists should be free to prescribe the most appropriate treatment for each patient without incentives to prescribe more- or less-expensive drugs.
These 3 options have different strengths and weaknesses (Table). Episode-based payment appears the most appealing in the long term and is most consistent with the trend in health system reform toward shared accountability and global payments. However, third-party buy-and-bill is the easiest to introduce because it does not require proscriptive decisions by insurers regarding what care is appropriate for which patients and is already enabled in Medicare through existing legislation.
If Medicare were to pursue third-party buy-and-bill again, the program would need several minor modifications. Its first implementation included nearly all physician-administered drugs, including many very inexpensive medications, which caused more of a logistical hassle than an economic boon to physicians. For instance, bags of normal saline had to be obtained from the third-party vendor. A modified program could focus just on highly expensive drugs that account for a meaningful share of Medicare spending—thus removing both the incentives and the risk for practicing oncologists for the drugs that most influence both. A recent Government Accountability Office report and my unpublished analysis (based on the publicly available 2011 CMS Part B National Summary Data File) suggest that nearly three-quarters of all Part B cancer drug spending and cost could be moved to a third party if Medicare transitioned approximately 20 drugs into the program, rather than the nearly 200 drugs and supplies placed in the program last time.8
In addition, Medicare would have to make the program mandatory, add fees for drugs in the program that were linked to the complexity of drug handling and storage, institute systems to ensure on-time delivery, and manage costs associated with wastage and dose reduction. An additional benefit of this approach is that the third-party vendor could assist Medicare in capturing what is probably several billion dollars a year in drug discounts that 340B hospitals and physicians are currently retaining when they obtain discounted drugs through the program and then bill Medicare as if they obtained the drugs at list price.9
The current financing method for physician-administered oncology drugs is unsustainable. Each of the 3 payment approaches outlined are potential improvements.
