The report is a worthwhile resource because it contains a lot of previously unreleased data on specialty drug expenditures in the Medicare Part B program. And while its conclusions are flawed, the report sheds light on how average-price benchmarks work. My take:
- The OIG is completely off their rocker in this report. Their savings estimates presume virtually instant reporting of Average Sales Price (ASP) data, which is both unrealistic and impractical. OIG also seems smugly content to eliminate any positive incentives for generic substitution. The Centers for Medicare & Medicaid Services (CMS) disagrees with the OIG's conclusions, although perhaps CMS just doesn’t want to do the extra work required for more frequent reporting.
- An average price benchmark can guarantee channel profits. As I illustrate below using actual ASP data, the use of an average-price reimbursement benchmark creates guaranteed profits during the period of generic substitution for the channel (wholesaler, provider, or pharmacy). This is a clear counterpoint to more judgmental and variable Maximum Allowable Cost (MAC) methods.
- The OIG skirts the real issue for payers. At what level of provider/wholesaler/pharmacy profits does a payer encourage rapid generic substitution while not “overpaying” for generics? A two-month lag might be too long, but perhaps a shorter lag would lead to slower substitution and therefore be self-defeating. OIG won’t go near that question.
A QUICK PRIMER
As I discuss on pages 66-69 of my pharmacy report, Medicare Part B switched to Average Sales Price (ASP) as the basis of reimbursement for physician-administered injectable drugs as well as some self-administered drugs such as oral anticancer drugs and immunosuppressive drugs. Private payers have been adopting ASP instead of Average Wholesale Price (AWP). See UnitedHealthcare: Cost-Plus for Cancer Drugs.
ASP equals the volume-weighted, per-unit average of manufacturer sales prices for each product that falls within a single Healthcare Common Procedure Coding System (HCPCS) code. ASP is computed using actual sales revenues to a manufacturer, i.e., list price minus all price concessions (volume discounts, prompt pay discounts, cash discounts, free goods, chargebacks, rebates, etc.). Thus, ASP is not a list price like Wholesale Acquisition Cost (WAC).
The ASP computation methodology has important consequences when a brand drug loses exclusivity and generic competitors enter the market. HCPCS codes for single-source drugs—a brand-name drug with no available generic version—typically include products from only one manufacturer. HCPCS codes for multiple-source drugs, i.e., drugs with generic versions, include products from multiple manufacturers.
However, ASPs are published with a lag between when sales occur and when the sales become the basis for payment amounts. For example, sales made in the fourth calendar quarter of 2010 (ending 12/31/10) will be used to compute the ASP that is published and effective in the second quarter of 2011 (starting 4/1/11).
GENERIC LIFE CYCLE PROFITS
The lag matters because physicians and providers will automatically earn higher spread profits during the period of generic substitution. Math does all the work so there is no need to set or manage a Maximum Allowable Cost (MAC) list.
Consider Pfizer’s Camptosar (irinotecan hydrochloride), an injectable drug used to treat patients with colorectal cancer. Nine competing generics launched on February 20, 2008. The generic formulations of Camptosar were priced at about one-third of the brand price and captured about 86 percent of sales in the first month after generic launch (March 2008) per a 2008 OIG report.
For fun (well, my kind of fun), I put together the table below summarizing the ASP story for irinotecan hydrochloride.
As you can see, the average per-unit spread is about $6 both before and long after generics enter the market. But the spread spikes to almost $96 (+1500%) in the first quarter after generic launch.
Why? The Medicare Part B payment amount for the relevant HCPCS code (J9206) does not include the generic versions in price calculations until the third quarter of 2008 (beginning July 1, 2008). Since the generic came out in the middle of Q1:2008, the full price impact does not get reflected until Q4:2008. My example actually understates the profit potential for a provider that immediately substitutes the generic version.
AmerisourceBergen (NYSE:ABC) and McKesson (MYSE:MCK), as the two biggest distributors of specialty drugs, also get a huge financial benefit from this dynamic.
So, does an average price benchmark provide too much profit, as the OIG argues? Or just the right amount of profit, as CMS believes? CMS also argues that the lag discourages manufacturers from rapid price increases to avoid a provider profit squeeze.
The answer will become even more critical because half of the current high-dollar brand-only Part B drugs are biologics.