Wednesday, May 11, 2016

Why the CMS Part B Drug Payment Demo Could Hurt Cancer Patients and What Should Replace It

Cancer treatment is becoming less and less affordable every year.  The primary reason for the rapid increase in cancer treatment costs is the increasingly high prices manufacturers charge for the drugs cancer patients need.  However, instead of finding ways to reduce the high prices of drugs, the Centers for Medicare and Medicaid Services (CMS) has proposed to cut the Medicare payments that enable oncology practices to buy and use the drugs cancer patients need.

CMS believes that the current method it uses to pay physicians to administer drugs in their offices creates a “financial incentive to prescribe high cost drugs over lower cost ones when comparable low cost drugs are available.”  To address this, it proposed the “Part B Drug Payment Model” in March 2016.  If it were implemented, this mandatory demonstration would cut Medicare payments to physicians for all of the drugs they administer (regardless of whether there are lower-cost drugs available) in order to “test” whether this would lead to a reduction in Medicare spending on drugs.  Although the change is represented as being budget neutral for physician practices overall, data presented by CMS indicate that the change would actually cut payments to oncology practices by over $32 million.

Many individuals and organizations have expressed strong opposition to the proposal, while others have supported it.  However, it is difficult to determine whether to support or oppose the proposal without understanding the complex way that Medicare pays for physician-administered drugs.  As explained in detail below, once you understand how payment works today, it becomes clear that the most likely effect of the change proposed by CMS will be to make it more difficult for cancer patients to obtain the drugs they need.  Moreover, it also becomes clear that more comprehensive reforms are needed to the way oncology practices are paid that would support improved care for patients and reduce truly avoidable spending.

Do Medicare Payments for Cancer Drugs Create Incentives to Use More Expensive Drugs?

Today, when a Medicare patient with cancer comes to an oncology practice for a chemotherapy infusion treatment, the drug the patient receives had to first be purchased by the practice from a drug wholesaler and then stored in the practice’s pharmacy until it was used to treat the patient.  After the patient receives the drug, Medicare pays the practice a predetermined amount for the drug, and by law, that amount is calculated by taking the “Average Sales Price” for the drug (ASP) six months earlier and adding 6%.  Currently, the actual payment is only ASP + 4.3% because of the across-the-board 2% cut in Medicare payments due to sequestration.

Many people have been led to believe that the 6% add-on is “profit” for the oncologist or the oncology practice.  In reality, the 6% is used by the oncology practice to cover at least five types of costs that are not otherwise reimbursed by Medicare:

  • the cost of operating the pharmacy in the oncology practice that enables drugs to be safely stored, mixed, and administered to patients;
  • the cost of wastage and breakage for drugs (although the practice must pay a drug company for the full price of a vial of chemotherapy, it can only bill Medicare for the portion of the vial that is used);
  • the difference between the “Average Sales Price” (ASP) and the actual acquisition cost of drugs (a more detailed explanation of this point is provided below);
  • the time the practice spends in trying to get financial assistance to help patients who don’t have supplemental insurance to pay the high cost-sharing on expensive drugs; and
  • the losses practices incur by not being able to collect the full cost-sharing amount from patients who cannot afford it.

Many of the above costs are higher for more expensive cancer drugs, which is why it makes sense to base the payment at least in part on a percentage of the drug price.  For example, if an oncology practice cannot use 10% of what is in a drug vial, the practice will not be reimbursed for 10% of what it paid for that drug, and a drug that was five times as expensive as another will cause the practice to lose five times as much.  Practices need to spend more time trying to help patients obtain financial assistance in paying their cost-sharing on expensive drugs than on lower-cost drugs, and practices incur more bad debt for patients receiving expensive drugs than low-cost drugs.

Contrary to what CMS and others believe, covering all of these costs and covering the higher costs associated with more expensive drugs does not create an “incentive” for a practice to use an expensive drug.  If anything, the percentage payment avoids creating a disincentive for the practice to use the expensive drug, so the oncologist can choose the drug that is best for the patient without worrying (as much) about whether the practice will lose money by using the drug.

No one knows what the “right” payment amount is to cover all of these costs.  The current 6% statutory amount is not based on an analysis showing that amount would cover the costs physician practices in general or oncology practices in particular incur, and CMS has not presented any new analysis indicating that 6% is too much.  In the Part B Drug Payment Model, CMS has proposed replacing the ASP+ 6% formula with a three part formula: ASP + 2.5% + a $16.80 flat payment per drug.  (With the sequestration adjustment, the actual payment would only be ASP + 0.86% + $16.53.)  However, CMS presented no analysis justifying that 2.5% would better match costs than 6%.  It used that amount because it was used by the Medicare Payment Advisory Commission (MedPAC) in an analysis MedPAC did.  MedPAC used the 2.5% amount in its analysis because it felt that this “should be sufficient to cover markups from wholesalers.”  CMS indicates that this was based on “anecdotal evidence” that such markups are between 1% and 2%, but that MedPAC “was not aware of data that could verify this estimate.” CMS states in its regulation that it is “not aware of data that could verify this assessment.”

Rather than seeking to obtain better data to determine what the right percentage should be, CMS is proposing to just cut the amount from 6% to 2.5% and see what happens.  The $16.80 flat payment was selected by CMS in order to offset the loss in revenue caused by the cut from 6% to 2.5%.  However, the analysis presented by CMS only shows that adding a $16.80 flat payment would offset the cut from 6% to 2.5% on average in 2014; the analysis shows that some types of physician practices – particularly oncologists, rheumatologists, and ophthalmologists – would experience large cuts in revenue, while other physician practices, such as primary care physicians, orthopedic surgeons, and cardiologists, would experience large increases in revenue.  No analysis is presented to suggest that these increases and cuts in revenue for different specialties would better match the costs the physicians in those specialties are incurring to deliver medications to their patients.

Some combination of a flat fee and a percentage markup would probably be a better match for a practice’s costs than a pure percentage-based markup, because a pure percentage markup under-reimburses pharmacy operations costs when lower-priced drugs are used (since there are fixed costs to operate the pharmacy that have to be covered regardless of the price of the drugs used), but a flat fee alone wouldn’t cover the higher costs the practice incurs when it uses more expensive drugs (because, as explained earlier, the practice’s costs are higher for higher-priced drugs).  However, the specific combination of a flat fee and percentage that CMS is proposing would clearly not be a better match for an oncology practice’s costs than the current percentage payment, because CMS’s own calculations show the proposed formula would result in a more than $30 million cut in the payments that oncology practices currently use to cover the costs of operating their pharmacies and purchasing the drugs their patients need.  Moreover, measuring whether the total payments are higher or lower than they are currently begs the question of whether the payment amounts were correct to begin with.

Available data indicate that commercial health plans pay a higher percentage markup on drugs than Medicare does, not a lower markup as CMS is proposing.  Some of these higher percentage markups likely do exceed the costs cited earlier that are associated with purchasing, storing, and administering drugs to patients.  However, data show that these higher payments do not represent “profits” to oncologists or to their practices; rather, these higher payments cover the costs of services that oncology practices deliver to their patients that are not paid for, or are inadequately paid for, by Medicare and the commercial health plans, such as the costs of patient education and counseling services, the time spent in coordinating care, etc.  Data from the National Practice Benchmark for Oncology indicate that current fee-for-service payments from Medicare and other payers only cover 2/3 of the costs of the services that oncology practices provide to their patients.  Many oncology practices are forced to rely on higher commercial payments for drugs to subsidize the other services they offer.

However, focusing only on the small percentage markup ignores the serious problems with the ASP portion of the formula, which represents 96% of Medicare’s spending on drugs and 96% of what the practice receives to cover the costs of acquiring drugs.  Most people do not realize that Medicare does not reimburse a practice for its actual acquisition cost associated with an expensive chemotherapy drug.  Rather, Medicare pays the practice based on the “average sales price” (ASP) of that drug two calendar quarters earlier.  As everyone knows, the prices of most cancer drugs are increasing rapidly.  This means that in most cases, the ASP payment from Medicare will be less than what the oncology practice will have to pay to purchase the drug, because the ASP amount was based on the price of the drug six months earlier, not the price when the practice actually bought the drug.  There are also concerns that the formula calculating ASP incorporates discounts received by wholesalers that are not actually passed on to physician practices, which means that the ASP amount is less than the average amount that physician practices actually paid for a drug.  Moreover, because larger practices often can obtain discounts that smaller practices cannot, smaller practices will generally have to pay more for a drug than larger practices, and that means they will pay more for a drug than what is calculated as the “average sales price.”  Consequently, when Medicare bases payments on ASP, it is often paying less than a practice’s actual acquisition cost for drugs, particularly for small practices.

The result of this very complex system is that many oncology practices, particularly small practices, lose money on many of the chemotherapy drugs they purchase.  Medicare’s 6% add-on payment (which has been only 4.3% under sequestration) on top of ASP helps to offset this loss in some cases, but not all.  The CMS proposal to significantly cut that add-on payment will mean that practices will lose money on even more drugs than they do today.  CMS is not proposing any improvements to the ASP system – which represents 96% of the payment for drugs – to make it more accurate or to remove any undesirable incentives it might create, it is only proposing to change the remaining 4% and to do so without any solid analysis indicating that the new amount better matches costs than does the current amount.

If CMS does not pay adequately to cover the losses and costs oncology practices incur in buying and administering chemotherapy, oncology practices will not be able to afford to administer the drugs patients need.  This will affect their ability to administer the lower-priced drugs CMS wants to encourage practices to use as well as their ability to use expensive drugs.

As noted earlier, paying adequately for the costs of administering chemotherapy does not give an oncologist an “incentive” to use one drug over another.  Rather, it ensures the oncology practice is not penalized financially for choosing the most appropriate drugs for their patients.  Moreover, for many patients, there may be only one drug that is appropriate to treat their disease at a particular point in time, in which case there is no choice that could possibly be “incentivized” by how much CMS pays for drugs.  In these cases, the proposal by CMS to cut drug payments would simply create a financial penalty for oncology practices when their patients need a high-cost drug or a drug whose price has been increasing rapidly.

The bottom line is there are two primary types of impacts that would be likely to occur if CMS implements the Part B Drug Payment Model.  One is that cancer patients will be unable to receive chemotherapy treatments that they need because their oncologists can no longer afford to purchase and administer them.  The second is that community oncology practices that try to purchase the drugs their patients need with inadequate reimbursement from Medicare will lose money and potentially be forced to close, and that in turn will mean that patients will have to travel farther and pay more to obtain cancer treatments.  It would be inappropriate for CMS to use the authority provided under the Affordable Care Act to “test” which of these impacts will occur or how big the impacts would be.

How to Control Cancer Spending Without Harming Patients

Instead of this problematic proposal, CMS should pursue implementation of comprehensive oncology payment reforms that will actually improve care for patients while making that care more affordable.  Last year, after many months of work, the American Society of Clinical Oncology (ASCO) announced a proposal for comprehensive payment reform called Patient-Centered Oncology Payment (PCOP).  PCOP is designed to provide adequate payment to oncology practices for many essential services that Medicare and health plans don’t pay for today, such as patient education, counseling, care coordination, etc.  Under PCOP, instead of oncology practices being forced to try and pay for essential patient services using revenues generated from drugs, the practices would be paid directly for those services.  PCOP is also specifically designed to reduce the overall cost of cancer care by (a) identifying the kinds of drugs, tests, and treatments that patients do not need and reduce the use of those services and (b) reducing the rates of complications and hospitalizations that patients experience while undergoing cancer treatment.  Under PCOP, oncology practices would take responsibility for implementing evidence-based guidelines developed by ASCO for prescribing tests and drugs, and the practices would receive payments enabling them to determine an accurate diagnosis and to select the most appropriate treatment based on the guidelines.  This would improve care for patients and make care more affordable, rather than achieving savings at the expense of quality.

Although PCOP has many advantages over the current payment system for oncology, it has specific advantages compared to the CMS Part B Drug Payment Model.  Instead of a policy that tries to discourage oncology practices from using expensive drugs that patients need, as the CMS Part B Drug Payment Model would do, Patient-Centered Oncology Payment would discourage practices from using drugs when patients don’t really need them.  Since many of those drugs are very expensive, this would save money without harming patients.

For example, the data that CMS released in conjunction with its Part B Drug Payment Model proposal showed that Medicare spent over $1 billion in 2014 on a very expensive drug called pegfilgrastim, the fourth highest amount Medicare spent on any drug in the Part B program.  Pegfilgrastim is not used to treat cancer, but rather to help patients who are receiving chemotherapy to avoid developing infections.  ASCO has developed guidelines for when the drug should and should not be used, but studies have shown that pegfilgrastim is being used for many patients who do not really need it.  The drug isn’t being overused because oncology practices have a financial incentive to do so, it’s being overused because oncologists want to help as many patients as possible avoid complications that can lead to hospitalizations.  However, not all types of chemotherapy are equally likely to cause the kinds of complications that pegfilgrastim can prevent, and the drug causes serious side effects that can outweigh its benefits for many patients.  Rather than simply cutting all payments for pegfilgrastim whether patients would benefit from it or not (as the proposed Part B Drug Payment Model would do), PCOP would enable an oncology practice to have adequate time to determine which patients really need the drug and also pay for the staff resources needed to create more cost-effective approaches for preventing hospitalizations for the others.

Reducing unnecessary spending on a few frequently used, expensive drugs could result in far greater savings for Medicare than anything the proposed Part B Drug Payment Model could achieve.  The PCOP payment model would provide oncology practices the time and resources they need in order to implement and follow complex, evidence-based guidelines that can control spending while protecting patients.

Reducing avoidable spending on drugs and tests is just one way that PCOP would achieve savings on cancer care in ways that benefit patients.  Another major focus of PCOP would be to enable cancer physicians to turn their practices into “oncology medical homes,” including providing rapid response to complications of chemotherapy in order to avoid patients being taken to the emergency room or being hospitalized.  A national demonstration project called COME HOME ( that was funded by CMS showed that dramatic reductions in the frequency of ED visits and hospitalizations and overall savings for Medicare could be achieved by giving oncology practices the time and resources they need to deliver more services to patients.  Unfortunately, these services are not paid for under the current Medicare fee schedule, and the practices that have implemented these services will have to discontinue them if a better payment system like PCOP isn’t implemented soon.

CMS has developed another oncology payment reform demonstration called the Oncology Care Model that could provide some of the resources oncology practices need to help their patients avoid ED visits and hospitalizations.  Unfortunately, the Oncology Care Model would also create significant financial incentives for oncology practices to give patients low-cost drugs regardless of whether those drugs would be effective for their patients, and it would penalize practices by dropping them from the program if they don’t find ways to reduce spending on drugs and other services.  A report prepared by the Center for Healthcare Quality and Payment Reform titled A Better Way to Pay for Cancer Care explains why the PCOP payment model is superior to the Oncology Care Model and to other alternative payment models that have been proposed or used by CMS.  It is available at

The current payment system for oncology care in America is badly broken.  Cancer patients deserve much better.  There are significant opportunities to reduce the cost of cancer care in ways that help patients rather than hurt them, but these can only be implemented with an appropriately designed payment system that provides adequate funding for good cancer care grounded in evidence-based guidelines for treatment.  Rather than testing problematic and piecemeal payment “incentives” that could seriously harm both patients and oncology practices, CMS should implement a truly comprehensive payment reform that strengthens oncology practices and enables them to deliver the best possible care to patients at the most affordable cost.



No comments yet.

RSS feed for comments on this post.

TrackBack URL

Leave a Comment

©2008-2011 Center for Healthcare Quality and Payment Reform. All rights reserved.
320 Ft. Duquesne Blvd., Suite 20-J - Pittsburgh, PA 15222 - (412) 803-3650 -