Sunday, July 08, 2007

Comments on AMP Final Rule

PBMs are the big story in CMS’ Final Rule issued on Friday night. See AMP is here! for document links.

Quick Summary: CMS' treatment of mail order will lower AMP, leading to disruptions in the retail pharmacy marketplace. AMP is also going to provide a very unique and new piece of information to the marketplace, although the direct impact on PBMs will be minimal in the short term.


CMS defines Average Manufacturer Price (AMP) quite literally in §447.504(a) as “the average price received by the manufacturer for the drugs in the United States from wholesalers for drugs distributed to the retail pharmacy class of trade.” Manufacturers must include all sales and associated discounts that reduce the amount received by the manufacturer (unless specially excluded). Note that a wholesaler is defined broadly to include any entity purchasing directly from the manufacturer.

AMP had historically only been used to compute brand manufacturer payments to the States under the Medicaid rebate program. But the Deficit Reduction Act (DRA) requires that AMP also be used to limit the reimbursement paid to retail pharmacies for multiple source (generic) drugs but not single-source branded drugs. This limit is called the Federal Upper Limit and will now be computed as 250% of AMP.

  • Since there is only one AMP calculation, this dual role creates conflict between manufacturers and retailers.

  • The Medicaid rebate program means that manufacturers want the lowest AMP possible.
Using AMP in the Medicaid pharmacy reimbursement formula (for generics) means that pharmacies want the highest AMP possible. For instance, DRA specifies that prompt pay discounts can not be deducted from AMP, increasing the average by about 200 basis points. Good for retailers, but bad for manufacturers.

Keep in mind that FULs apply to multiple source drugs – those with a therapeutic equivalent in the Orange Book. Thus, a pharmacy can not dispense a brand name drug with a generic equivalent in order to game the reimbursement system because the brand would be subject to the FUL.

States can top off AMP in whatever way they choose because CMS does not mandate a specific formula or methodology for determining the dispensing fee. The DRA also does not require States to base pharmacy reimbursement for Medicaid scripts on AMP. However, states can not modify the FULs, which was the point of my post from two weeks ago.


Mail order is retail.
CMS clearly states that mail order pharmacies serve the general public and should be included in retail class of trade. Mail order dispensed more than $50 billion of drugs last year, making it the second biggest dispensing channel behind chain pharmacies such as Walgreen (WAG), CVS (retail), and Rite-Aid (RAD).

Chain and independent pharmacy groups vehemently opposed including mail in the retail class of trade because mail’s lower cost of goods brings down the average. PhRMA supported the inclusion of mail, arguing for a function-based definition of dispensing to patients. (See page 57 of CMS-2238-P Paper Comments 125-134.

PBM price concessions are out, but…
Rebates, discounts, or other price concessions to PBMs are excluded from AMP in the final rule, except for purchases through PBM mail order pharmacies. I predicted this general outcome in May based on signals from Congress that CMS was not following “congressional intent.” Sales via a PBM’s retail network involve no physical possession or dispensing on the part of the PBM. These sales transactions get picked up in AMP because they flow through wholesalers to store-based pharmacies, so the exclusion seems logical and internally consistent to me.

... price concessions for PBM mail pharmacies are in.
On the other hand, CMS decided to include PBM discounts in AMP calculations where the PBM is operating a mail order pharmacy. As I understand the regulation, PBM rebates and discounts would lower AMP for the portion of sales dispensed through the mail. This seems like a logical consequence of including mail order in retail.

This decision will significantly reduce AMP because the three largest PBMs – Medco (MHS), Caremark (CVS), and Express Scripts (ESRX) – represent 70% of total mail order dollars. For example, Medco processed $26 billion through its retail network and dispensed an additional $16 billion by mail.

If I am interpreting the Final Rule correctly, then 38% [=$16/($16+$26)] of price concessions to Medco would be included in AMP. Frankly, I did not correctly anticipate how CMS would split the difference by excluding the portion of rebates and discounts from mail. (Please post a comment if I have misinterpreted this provision.)


I’ve summarized the market impact of AMP in two recent posts, both of which are still relevant now that I’ve read the Final Rule: Why AMP will not be Independents' day and An AMP Timeline Appears.

AMP is going to provide a very unique and new piece of information to the marketplace. Here are few ways that AMP will differ from other pricing benchmarks:

  1. AMP will be the first broadly available measure of the price a manufacturer receives for selling a product.
  2. AMP will have a publicly available, defined calculation methodology.
  3. AMP data will be made public on a website at the 11-digit NDC level and updated monthly.
I believe that AMP will shed a lot of light on the economics of the pharmacy supply chain. For better or worse, it will also provide a new data point for benefit managers, insurance companies, policy analysts, and consumer advocates.

That said, I do not see the transparency of AMP as an automatic negative for PBMs, as some are predicting. PBMs are hired by payers to manage pharmacy benefits and moderate the drug spend. New reimbursement models simply provide another basis for contracting and compensating the participants in the pharmacy supply chain.

I’ll post further comments after I bounce some ideas off my clients and other analysts. In the meantime, feel free to send me an email or post a comment. Thoughtful anonymous comments are always welcome!

No comments:

Post a Comment