CMS Takes Concrete Steps Toward Replacing AWP


Follow SecSovHealth on Twitter

On August 4, 2011 CMS hosted a meeting to update external stakeholders on the status of the implementation and roll-out of the national actual acquisition cost (AAC) survey, the existence of which Secretary Sebelius first revealed in a letter to state governors in February 2011. At the time, she acknowledged only that the survey data would be available to states for use in setting Medicaid drug reimbursement rates this year. CMS still maintains that the new index (now named NADAC instead of AAC) will be available this year.  And, we have confirmed through CMS that the survey results[1] will be posted publicly on the website, thus NADAC will be available for use as a commercial drug pricing benchmark

The survey to establish National Average Drug Acquisition Costs (NADAC) will be conducted by Myers & Stauffer LC – the same firm that handles the Alabama AAC survey, about which we have written extensively[2] – and in many important ways parallels the Alabama survey

AAC (and thus NADAC) is more precise than AWP or MAC, and results in lower pharmacy re-imbursement for ingredient costs. Myers and Stauffer estimate that “90% of generic drug groups will have a margin of error of less than 10% of the mean unit cost at a confidence level of 95% …and an average margin of error as a percent of mean unit cost of 5%.” Clearly this supports a level of reimbursement precision that is orders of magnitude better than current AWP based rates. Using Alabama AAC data, we looked at a sample of 95 of the most prescribed generic molecules and, assuming reimbursement of AWP-50%, MAC’d at the Federal Upper Limit, found less than 2% of generic drugs in our sample had an AWP-based reimbursement that was within 10% of the mean unit cost. And, AAC was not only dramatically more precise than AWP or MAC, but also resulted in significantly lower estimates of acquisition costs, and thus substantially lower dispensing margins to pharmacy.  Specifically, the average difference between acquisition cost (as estimated by AAC) and AWP or MAC-based reimbursement rates was about 200%

To our mind the most important difference between the Alabama survey and the NADAC survey is that participation the national survey is voluntary, whereas participation in the Alabama survey is mandated. While it is difficult to gauge the impact of this difference, the sampling methodology will have accounted for the participation rules; and as the survey manager can increase sample size to offset the effect of non-responses, we have no reason to believe the voluntary nature of the NADAC survey will lead to inaccurate or biased results

Also, we acknowledge that while AAC / NADAC data are far more reflective of ‘true’ acquisition costs than AWP or MAC, AAC and NADAC are imperfect. AAC / NADAC are produced using dispensing pharmacies’ invoice prices, thus off-invoice discounts, rebates, and marketing assistance fees are not captured. However the effect of chargebacks[3] on invoice prices is captured, and at least for the time being chargebacks appear to be (by far) the largest single contributor to the difference between list price and ‘true’ acquisition cost

We continue to believe that the displacement of AWP as a standard pricing benchmark (by indices that more accurately reflect generic acquisition cost – such as NADAC) in commercial contracts will reduce the trades’ (wholesalers, retailers, and PBMs) dispensing margins on generics (we estimate an average of $9.01) to a level on par with dispensing margins for brands (we estimate an average of $5.77). Of the many structural challenges facing PBMs we see a reduction in generic dispensing margins as the most immediately important. Among the trades PBMs are most negatively affected, and among the PBMs greater exposure to mail order means greater exposure to margin compression. Retailers are next most affected, followed by wholesalers

Presentation materials with important background information on NADAC can be found on our website (11 08 16 cms nadac presentation). Our detailed views regarding the PBM industry generally and pricing indices specifically can be found in several of our past notes: “ESRX, MHS, and the PBM Bear Case”, 7/25/2011; “The Thread Holding Generic Dispensing Margins”, 5/5/2011; “Co-Pay Cards and the Stalling of Drug Rebate Growth”, 1/5/2010; “Uncertainty and Motive in Pharmacy Dispensing Mark-Ups”, 11/10/10; and, “Why Generic Dispensing Margins (eventually) Must Fall”, 10/29/10

[1] Obviously these exclude any respondent-identifiable data

[2] See, e.g. July 25, 2011 – “ESRX, MHS, and the PBM Bear Case”; April 6, 2011 – “CMS Says AMP is Coming…”

[3] Chargebacks are used to reduce the price paid by a pharmacy relative to the price paid by the wholesaler.  E.g. a pharmacy may purchase for 0.66 a product purchased by the wholesaler for 1.00, in which case the wholesaler charges the 0.34 difference back to the manufacturer. Importantly, in this case the pharmacy’s invoice price is 0.66, i.e. it reflects the effect of the chargeback, even though the invoice typically would not disclose the specific amount of the chargeback

Print Friendly, PDF & Email