CMS proposes Medicare Part B tests of value-based drug purchasing

It’s perilous times in Washington for the biopharma industry. CMS is issuing a draft proposal (to be published in the Federal Register on March 11) setting up an experimental program to address the high cost of newer, branded pharmaceuticals. And although it is a proposed experiment, and a permanent adoption of its measures is probably years off, it was met with quick negative reaction from biopharma trade associations and others.

At the same time, in a rather transparent attempt to justify the proposal, another part of HHS, the Office of the Assistant Secretary for Planning and Evaluation (ASPE), produced a pair of “issue briefs” measuring the growth in specialty pharmaceutical expenditures (see related news). ASPE estimates that total drug spending in the US was $457 billion in 2015, including $128 billion “non-retail” drugs (i.e., administered in hospitals or clinics under a medical and not pharmaceutical benefit). The total figure puts overall spending at 16.7% of overall personal healthcare expenditures—well above the 10-12% usually cited as drugs’ portion of national healthcare expenditures.

CMS proposes six methods to alter Medicare Part B reimbursements:

• Improving incentives for best clinical care. Physicians often can choose among several drugs to treat a patient, and the current Medicare Part B drug payment methodology can penalize doctors for selecting lower-cost drugs, even when these drugs are as good or better for patients based on the evidence. Today, Medicare Part B generally pays physicians and hospital outpatient departments the average sales price of a drug, plus a 6% add-on. The proposed model would test whether changing the add-on payment to 2.5% plus a flat fee payment of $16.80 per drug per day changes prescribing incentives and leads to improved quality and value. The proposed change to the add-on payment is budget neutral.
• Discounting or eliminating patient cost-sharing. Patients are often required to pay for a portion of their care through cost-sharing. This proposed test would decrease or eliminate cost sharing to improve beneficiaries’ access and appropriate use of effective drugs.
• Feedback on prescribing patterns and online decision support tools. This proposed test would create evidence-based clinical decision support tools as a resource for providers and suppliers focused on safe and appropriate use for selected drugs and indications. Examples could include best practices in prescribing or information on a clinician’s prescribing patterns relative to geographic and national trends.
• Indications-based pricing. This proposed test would vary the payment for a drug based on its clinical effectiveness for different indications. For example, a medication might be used to treat one condition with high levels of success but an unrelated condition with less effectiveness, or for a longer duration of time. The goal is to pay for what works for patients.
• Reference pricing. This proposed model would test the practice of setting a standard payment rate—a benchmark—for a group of therapeutically similar drug products.
• Risk-sharing agreements based on outcomes. This proposed test would allow CMS to enter into voluntary agreements with drug manufacturers to link patient outcomes with price adjustments.

One of the primary problems the changes would address is the negative incentive doctors have today for prescribing lower-cost drugs: because the current reimbursement is tied to ASP+6%, a higher-cost drug will generate more income to the physicians than a lower-cost one. Other proposals would bundle higher-cost and lower-cost drugs of the same therapeutic class, potentially making the higher-cost purchase less attractive. The risk-sharing agreement approach picks up on a practice that is more common in the EU, but which has been tried selectively in the US. And, as CMS notes, “Commercial health plans, pharmacy benefit managers, hospitals, and other entities that manage health benefits and drug utilization successfully employ an array of tools including value-based pricing and feedback on prescribing patterns to improve the value of drug payments;” CMS would like to try some of these reimbursement mechanisms itself.

Reaction was swift in Washington: “[U]pon initial review we are gravely concerned that the Model threatens to disrupt the successful reimbursement framework for drugs covered under the Medicare Part B program,” said BIO. “Proposing sweeping changes to Medicare Part B drug reimbursement without thoughtful consideration and stakeholder input is not the right approach and puts Medicare patients who rely on these medicines at risk,” said PhRMA. A joint statement from Republican leaders of House and Senate committees involved with healthcare oversight asserted that “Yesterday's announcement marks another troubling example of unelected bureaucrats making decisions behind closed doors that impact the American people and their healthcare. This decision was made with a complete lack of transparency and clear disregard for the people and stakeholders who will be impacted the most."

The Community Oncology Alliance (COA), a public interest group representing those practitioners, was flat-out oppositional: “COA notes that it is actively pursuing every legal, legislative, and related option on behalf of members to stop the CMS Medicare Part B experiment,” it asserted. COA also stated directly what other critics have implied: that throttling specialty pharmaceutical prices will harm the practice of community oncology—generally, nonaffiliated oncologists treating individual patients—which it maintains can provide cancer care at a lower cost than oncology centers affiliated with hospital systems. (It’s not clear whether the proposed changes would affect community oncologists better or worse than the oncologists affiliated with health systems, yet that is what COA foresees.)

If CMS goes forward after the 60-day comment period ends, it expects to set up the test programs in the fall, and to have value-based purchasing begin testing no sooner than next January, and for the program to be “fully operational” around late 2018. Gearing all this up in the middle of an election year is sure to be another factor that CMS will have to overcome.