Part D reform has faded from the policy debate. This rerun explains why it is still needed. FYI, this is my favorite article from 2020 (so far).
Click here to see the original post and comments from January 2020.
Our high-list-price/high-rebate system remains a fundamental source of warped incentives and cascading problems within the Medicare Part D program.
For proof, check out the previously unpublished data below on market share for products that treat hepatitis C. Despite manufacturers offering products with lower list prices, Medicare Part D plans have rejected the therapeutically identical but lower-priced versions of these drugs.
List prices significantly affect seniors’ out of-pocket costs, so Part D plans are needlessly costing many of them thousands of dollars. The federal government's Medicare spending is also unnecessarily higher.
Anyone concerned about drug prices should pay close attention to this situation. Part D plans and seniors who don’t need specialty medications are benefiting, while seniors who need treatment with specialty medications are ripped off. Just another day inside the gross-to-net bubble!
Understanding the Part D pricing issues requires a brief detour into the treatment of chronic hepatitis C virus (HCV):
- There are six genetically distinct groups of the virus. About 75% of patients infected with HCV have genotype 1 (subtypes 1a or 1b), 20% to 25% have genotypes 2 or 3, and a small percentage of patients has genotypes 4, 5, or 6.
- Drug treatment depends on the patient’s genotype. Two popular therapies—Mavyret and Epclusa—can be used with all six genotypes. Harvoni can be used only with genotypes 1, 4, 5, and 6. Zepatier can be used only with genotypes 1 and 4.
- Harvoni (brand): $94,500
- Harvoni (authorized generic; AG): $24,000
- Epclusa (brand): $74,760
- Epclusa (authorized generic; AG): $24,000
- Mavyret: $26,400
These list prices do not meaningfully reflect the net cost to a payer, because rebates can be substantial. . In fact, I suspect that the net, post-rebate prices of these drugs are fairly comparable.
For instance, Gilead started a wholly-owned subsidiary called Asegua Therapeutics to launch the authorized generics of Harvoni and Epclusa. At the time of launch, the list prices for the authorized generic versions reflected the average, post-rebate net prices of the brand therapies.
MARKET SHARE MUDDLE
Students of economics might assume that the market will choose the products with the lowest prices.
For example, the branded and authorized generic versions of Harvoni and Epclusa are identical products. So, you might think that the plans would naturally prefer the lower-priced version because the patient is getting the exact same product.
Alas, that’s not how the U.S. drug channel works.
Below are data on new prescriptions for the five HCV products we’ve listed above. These data include only those patients who had not initiated therapy on any HCV product in the prior year. The results cover the three major payers: Medicare Part D, commercial health plans, and Medicaid. (Special thanks to IQVIA for sharing these previously unpublished data.)
[Click to Enlarge]
As you can see, there are some startling differences in Medicare Part D compared with the other payers.
- Medicare Part D favors high list price drugs. In January 2019, the high list price versions of Harvoni and Epclusa accounted for 47% of all new prescriptions in the HCV class. In August 2019, these two products were still receiving 46% of all new prescriptions, though the share of Harvoni dipped at the expense of Epclusa.
The share of these products’ respective authorized generics reached an underwhelming 8% by August of last year. Meanwhile, Mavyret’s share of new prescriptions ranged from 40% to 44% throughout the first eight months of the year.
- Commercial plans are adopting the lower list price products. The two high list price products accounted for more than half (51%) of new prescriptions in January 2019. However, their share of new prescriptions shrank to 29% by August 2019. Mavyret and the Epclusa authorized generic had 62% of new prescriptions by that time.
- Medicaid programs have also adopted the low list price drugs. Mavyret accounted for three-quarters of all new Medicaid prescriptions at the start of 2019. It has lost share to the low list price version of Epclusa.
Note that the last few months of these data also reflect new state Medicaid arrangements. In July 2019, the Washington and Louisiana state Medicaid programs entered into separate “Netflix subscription” agreements with manufacturers. Washington contracted with Abbvie for Mavyret and Louisiana contracted with Asegua for Epclusa AG. See this Health Affairs article by Mark Trusheim and Peter Bach for background on these programs.
Here’s why the structure of the Medicare Part D benefit makes the market share of high list price drugs so troubling.
Medicare beneficiaries, unlike those in most private insurance plans, can face unlimited out-of-pocket prescription drug costs if they reach the catastrophic coverage limit. Medicare covers 80% of the cost in the catastrophic phase, while plans pay 15% and the beneficiary pays 5% coinsurance.
Progression through the Part D benefit tiers is based on the prescription price negotiated between the plan and the pharmacy. That negotiated price excludes rebates, so beneficiaries reach catastrophic coverage—and its unbounded 5% out-of-pocket expense—sooner when using products with higher list prices.
Manufacturers are not permitted to provide copayment support to beneficiaries of federal healthcare programs, so the burden of using a higher-price product falls entirely on the patient.
Consider the following data from The Out-of-Pocket Cost Burden for Specialty Drugs in Medicare Part D in 2019:
[Click to Enlarge]
A Part D beneficiary’s median out-of-pocket costs for Mavyret were $3,520 in 2019, while median out-of-pocket costs were $5,283 for brand Epclusa and $6,338 for brand Harvoni. This study didn’t include the authorized generics, but I estimate that their out-of-pocket costs would be about $3,300.
Despite the out-of-pocket advantages, the market share data above demonstrate that half of seniors in Part D plans are still initiating therapy with branded Harvoni and Epclusa instead of the lower list price alternatives. That's crazy.
Part D plans and their PBMs appear to prefer drugs with higher list prices and higher rebates. Such products push more beneficiaries into the catastrophic reinsurance phase, where the plan liability is low.
Here’s the twist: Plans can use the additional rebates earned from the high-list price products to reduce monthly premiums and grab more Part D market share. The weird math of Part D bidding and its benefit structure encourage this gaming. For an explanation, see Will CMS Pop the Gross-to-Net Bubble in Medicare Part D With Point-of-Sale Rebates?.
Part D plans’ continued use of brand Harvoni and Epclusa exemplifies how patients don’t benefit directly from rebates on the drugs they use.
- Plans are using rebates to reduce overall Part D premiums, rather than offset the prices paid by patients whose prescriptions are generating the rebates.
- Part D plans' preference for the high list price versions shifts costs and risk to the federal government. The Medicare Payment Advisory Commission (MedPAC), among others, has highlighted these problems. I highly recommend its recent status report and proposal for restructuring. Over the past ten years, the government's share of Part D costs has almost quintupled, from 5% in 2007 to 23% in 2017. Meanwhile, plans' share of costs has plummeted , from 53% in 2007 to 29% in 2017.
Part D plan’s failure to adopt lower list price versions of high list/high-rebate drugs perfectly illustrates the warped incentives baked into the drug channel and Part D benefit design. Your friendly neighborhood blogger remains frustrated that most conversations and policies aimed at drug prices ignore the very real consequences of benefit design for patients.