Thursday, December 09, 2010

How to Stop Medicaid from Overpaying for Drugs

A fascinating new study from the Lewin Group estimates how much Medicaid overpays for prescription drugs. You can download this lengthily-titled report for free here: Potential Federal and State-by-State Savings if Medicaid Pharmacy Programs were Optimally Managed.

Lewin estimates that Medicaid spending would drop by 14.8% ($2.5 billion) in 2011 if fee-for-service Medicaid prescriptions were “optimally managed,” which means at levels comparable to private plans. A ten-year extrapolation estimates more than $30 billion in savings.

Shocking? Not really. I’ve long been critical of the overly political nature of pharmacy reimbursement under Medicaid. Just look at the windfall to South Carolina pharmacies versus their neighbors (in A Victory for Pharmacy Profits in South Carolina) or the excessive payments to pharmacies under the old Federal Upper Limits (in Won’t get FULed again). Lewin found that higher payments to pharmacies are not even associated with higher generic dispensing rates.

The data seem clear to me. But as always, I encourage you to read the study for yourself and make up your own mind.


The Lewin study computes the savings figure using a straightforward analytic approach:
  • Compute the savings from moving Medicaid fee-for-service (FFS) plans to levels typically found in Medicaid and commercial managed care or Part D plans in four areas: dispensing fees, ingredient cost, generic fill rates, and utilization

  • Add the increase in administrative costs associated with more active benefit management activities, i.e., higher costs by an internal state Medicaid program or the costs of an external Pharmacy Benefit Manager (PBM)
Page 8 of the report shows savings by state. Three states—New York ($402 million), California ($372 million), and Texas ($266 million)—account for 42% of the total savings. The savings figures for Alabama and Oregon are probably a bit high because both states are moving to cost-plus models to save money.

Here’s where the estimated savings come from:

One surprise is the fact that lower dispensing fees account for almost one-third of the savings. As Lewin notes: “On average, Medicaid FFS programs pay pharmacies a dispensing fee of $4.60 for brand drugs and $4.90 for generic drugs, more than twice the amount paid by private sector health plans.”

True. Compare Lewin’s computed Medicaid average dispensing fee of $4.81 with the average $1.62 fee from the PBMI's most recent survey of 372 employers (source).

Keep in mind that the dispensing fee typically accounts for only 15% to 20% of the total gross profit per prescription earned by a pharmacy from a typical third-party payer. “Spread pricing” provides retail pharmacies with most of their profits from prescriptions because pharmacies consistently acquire drugs for less than the ingredient cost reimbursement amount. Lewin found that several states are just very munificent—they are "high-end payers" to retail pharmacies for both dispensing fees and ingredient costs.

For more details on pharmacy profits, see the “Pharmacy Reimbursement by Third-Party Payers” section of The 2010-11 Economic Report on Retail and Specialty Pharmacies.


I’m sure some pharmacy owners will carp about the fact that the study was commissioned by the Pharmaceutical Care Management Association (PCMA), the association that represents PBMs. There is clearly some enlightened self-interest here because PBMs would like to manage the Medicaid pharmacy benefit on behalf of states—and earn fees for doing so.

But having PBMs manage Medicaid drug trend makes sense to me as both an industry analyst and as a taxpayer, especially given historically dismal oversight of pharmacy reimbursement by state Medicaid programs. Factor in the projected growth in Medicaid enrollment (illustrated in Health Reform: Impact on Drug Channels) and there is a very compelling story here.


What do you think? Are the study's conclusions reasonable or not?


  1. Adam-

    Pharmacy carve outs are a state by state decision. Yep-some states do a very poor job and have no business managing pharmacy spend. However, many states that have a carve out manage the drug spend very effectively and have brand and generic reimbursement rates below that of the MCOs. The state generic dispensing rates in the Lewin report are not accurate. They are off in some cases by 8-11%. This casts doubts on the entire analysis. How does the 10 year extrapolation incorporate the waive of upcoming patent expirations and the obvious savings that will occur without regard to who manages the benefit? What "optimally managed" means is that physicians, nurses, patients and pharmacists will be subject to prior auth requirements that are not clinically sound. Very simply put, MCO's institute ridiculous PA requirements and restrictive formularies in order to maximize their profit at the administrative expense of providers. What vested interest does the MCO have in the long term health of the patient? Virtually none as they won't be around or assume the risk when the relatively healthy women and children eventually get sick. Why aren't the sickest Medicaid patients in managed care...because they won't accept those patients because of the financial risk of caring for this high cost population. The report does a very poor job of explaining how they adjusted for population mix differences. Not subject to lobbying..... Just take a the recent KY Passport issues:

    This is rampant across the country. "Not for Profit???"--just go take a look at the publicly available 990 forms and see where the taxpayer dollars are going.

    Thanks for the opportunity to comment.

  2. You may want to look closely at Lewin’s averages and total expenditures for each state; various states’ numbers are grossly over- and/or underestimated. For example, Lewin lists generic %s as well as total claims that do not seem up to date. When you are talking avg brand price at around $180-200 per claim, and avg gen/OTC price of $25 per claim, that % generic fluctuation can (and does) result in millions skewed dollars on Lewin’s part. Just something to think about.

  3. Adam,

    You can't just compare the dispensing fees. I wonder why they did not compare reimbursement product cost on generics Medicaid versus PBM mail order. I will take a lower dispensing fee if they will reimburse at the level PBMs pay themselves. I am not going to "carp" about the report but who paid the bill for the study? If I ask for a review of whether my car was "cleaned and washed" by the business that "cleaned and washed"it what do you think that review would be? Rocket science !!!!

  4. Adam , your audience has found to date your analysis on pharmaceutical supply chain dynamics very insightful. Alas , on this occasion a deeper dive is warranted and expected by your readers in relation to Medicaid Pharmacy Programs.
    The report is commissioned by PCMA , Lewin Group is a subsidiary of United HealthCare Group who owns Prescription Solutions i.e. a PBM.The executive board of PCMA includes the CEO of Prescription solutions.
    Previous two commenters have touched on some of the critical issues that the Lewin report neglects to address or demonstrates a lack of understanding of the Fee For Service (FFS) Medicaid Pharmacy Programs.
    Federal and State Statutes provide the regulatory framework for the administration of Fee For Service pharmacy programs relating to eligibility requirements , provider enrollment , reimbursement methodologies including ingredient costs and reasonable dispensing fees, prior authorizations requirements , mandatory drug coverage requirements , formulary design criteria ….etc the PBM industry focuses on contractual relationships that are crafted to maximize profits , PBMs have the luxury to exclude providers from network , increase premiums and cost sharing , exclude high risk high utilization patients , deploy utilization control methodologies that impede access to care , prior to Affordable Care Act PBM transparency was non existent.
    Under 42 U.S.C. 1396a(a)(30)(A), a State’s plan for medical assistance under the Medicaid Act must “provide such methods and procedures relating to the utilization
    of, and the payment for, care and services available under the plan * * * as may be necessary * * * to assure that payments are consistent with efficiency, economy, and quality of care and are sufficient to enlist enough providers so that care and services are available under the plan at least to the extent that such care and services are available to the general population in the geographic area.”
    The shift of a pharmacy reimbursement model to market based acquisition cost pricing is a threat to the survival of PBM industry .The Industry , as you point out , is trying to create a business portfolio offering to cash strapped State Legislatures.
    Over to you!

  5. I'm not claiming that this study is perfect. However, none of these comments address the real issue: Why are Medicaid prescriptions so much more lucrative for pharmacies in some states?

    Just look at my brand-name computations in A Victory for Pharmacy Profits in South Carolina or the MACs shown in Morgan Stanley's study in New Study Finds Small AMP Impact, But Trouble in Six States.

    Are the real savings 14.8%? or 10%? or 20%? I'm not sure. But it's hard to deny that there is room for improvement.

    And I agree that the sponsorship is important to the interpretation, which is why I highlight it above.


  6. ADAM :However, none of these comments address the real issue: Why are Medicaid prescriptions so much more lucrative for pharmacies in some states?
    Answer :Rebate Focused programs , AWP (AINT WHATS PAID)based reimbuserment methodologies and Federal regulations that establish criteria Rebate programs and reimbursment methodologies including Dispensing Fees.Why did CMS recently approve a dispensing fee in Alabama for more that $10 ?

  7. Most state's lack the political will to lower reimbursement to competitive levels. At the end of the day, that's the answer to your question.

  8. Yes, statutory and supplemental manufacturer rebates do influence the net economics for a state.

    But as Lewin correctly notes: "Rebates do not diminish or otherwise impact the savings that are achieved from dispensing fee savings and ingredient cost reductions." (page 15)


  9. Rebate Revenue is signficant from Brand name products , ingredient costs for these products to date have been inflated by Manufacturers to improve the price spread to providers ie increased reimbursement.IF you move to generic only programs , rebates will diminish.
    Dispensing Fees in Medicaid Programs are defined to be reasonable ; consideration of over head costs , labor and profit are statutory obligations for States.That is probably why Alabama is paying a dispensing fee of more than $ 10 per prescription.PBMs NEGOTIATE Contracts with select providers for dispensing fees in the range $1- 2.50 per prescription.What is the true cost of dispensing a prescription $1 or $10 ? THE LEWIN group analysis is not considering the dynamics and the inter-relationships between ingrdeient costs , dispensing fees , rebates , product mix on formularies.

  10. Anytime Lewin does a study the results are immediately suspect. Why-not only are they in bed with the PBMs...THEY also set capitation rates for some states Medicaid MCO programs. Lewin has conflicts of interest all over the place.

    Secondly, take the top ten Medicaid MAC programs and compare them to the commercial MAC programs (if you can get your hands on the data). The real numbers will surprise you. The Medicaid programs will blow them out of the water making the dispensing fee matter a non-issue. Why you say--the commercial MACs are kept artificially high because they pay their own mail order pharmacies at that rate. Lewin doesn't have a clue with respect to rebates. On paper ACCA seems to make Pharma look like they are paying more in base rebates. Truth of the matter is that there is virtually no new money on the table from the baseline rebate perspective. Best price rebates are way above the increased minimum rebate percentage. Two products/classes would have made some difference--hemophilia factor and Synagis...but interested parties were able to get legislative relief so that they were not subject to any sort of meaningful increase.

  11. Blog #1 is absolutely on target regarding dispensing rates and the Kentucky Passport Issue which is an ongoing Kentucky problem.

    Blog #2 Also on target, again observing what appears to be fuzzy math regarding GSR, who determined what the generic definition was in this report?

    Blog #3 Bulls eye….the PBM industry focuses on contractual relationships that are crafted to maximize profits , PBMs have the luxury to exclude providers from network , increase premiums and cost sharing , exclude high risk high utilization patients , deploy utilization control methodologies that impede access to care.
    The shift of a pharmacy reimbursement model to market based acquisition cost pricing is a threat to the survival of PBM industry.

    Question: Why do PBM always do “forecast” regarding Medicaid accounts? There is more than an amble amount of collected data to compare numbers head up. What we have found is that when you plug existing PBM numbers with existing Medicaid numbers based on past history, here in Ohio the PBM model is always more expensive

    Bog #4 another ringer….Take the top ten Medicaid MAC programs and compares them to the commercial MAC programs (if you can get your hands on the data). The real numbers will surprise you. The Medicaid programs will blow them out of the water making the dispensing fee matter a non-issue.
    Here total agreement; having audited Ohio Medicaid with full access to data I have found that the total dollar expenditure per Rx to be less when comparing Medicaid to the MCO/PBM model.

    Adam, The enormous amount of conflict in this report is undeniable.

    True Acquisition + Professional Dispensing fee = PBM failure

  12. Hmm, many Medicaid fans on the blog today.

    Would you believe the OIG? Here's the conclusion from an OIG study comparing Medicare Part D to Medicaid using claims data. Guess which one pays more to pharmacy for generic drugs?

    "Nationally, the average Medicaid pharmacy reimbursement amounts typically exceeded the average Part D reimbursement amounts for selected multiple-source drugs. Based on data from all States, Medicaid pharmacy reimbursement amounts exceeded Part D pharmacy reimbursement amounts by at least 10 percent for 28 of the 39 multiple-source drugs under review. Medicaid reimbursed less than Part D, on average, for just three of these drugs. At the median, the Medicaid reimbursement amount was 17 percent greater than the Part D amount for the 39 selected multiple-source drugs."

    Source: Comparing Pharmacy Reimbursement: Medicare Part D To Medicaid


  13. Adam - When are you going to learn that your facts don't matter to pharmacy owners? This is their pocketbook issue. They have already made up their mind!

  14. ADAM ,the question that the responses are addressing here (and very well articulating) , are not whether Medicaid Reimbursement (aggregate of Estimated Acquistion Cost and reasonable dispensing fee) are generally higher than private payors or Medicare Part D plans.
    You stated :"The data seem clear to me. But as always, I encourage you to read the study for yourself and make up your own mind."
    The Lewin report is deeply flawed in its analysis --- that is what is being called out.The issues you raise and point to from previous blog posts are well known to this group.
    If you believe the Lewin report is balanced , fair , independent , and accurately reflects the issues discussed by the commentry above , please elaborate with respect to the details in the report.

  15. Michael HowardDecember 09, 2010

    How about some focused data from the past 3 to 4 years from states such as NY where they have established mandatory generic dispensing programs. Using data from the past 10 years skews the numbers in favor of PBM's.

  16. ADAM ;In light of the significant discussion on the validity of the Lewin report ,do you continue to believe the Lewin study makes the "data clear "?
    I have certainly appreciated the insights provided on your website postings in relation to this report and would conclude the report as no more than a marketing gimmick from PBM lobbyists.Further it is interesting that Lewin group or United Healthcare Group have not stepped up to clarify/answer/defend their marketing piece.Thank you for attracting a great expert audience.

  17. Yes, I've also been impressed by the many thoughtful comments.

    As I note above, the report is not perfect and the sponsorship may or may not have influenced the results. I do still believe that the report is directionally correct, but the Drug Channels audience has raised legitimate issues about the magnitude of the effect.

    I'll contact Lewin and see if they want to respond.


  18. RE: Referenced OIG Study

    A few comments for consideration-

    1) This study was from 2006 data. Data is no longer relevant as Medicaid states have drastically increased their cost-containment activities
    2) This study only included 5 states out of a potential 51 Medicaid programs. One didn't even have a MAC program. In the world of averages, this greatly skews the results
    3) It will be interesting to see current single source drug reimbursement metrics. Nearly every state kept their current AWP discount when the compendia rolled back AWP. Most commercial insurers adjusted their AWP discount or converted to an equivalent WAC so that interested parties were not economically harmed
    4) How were copays taken into account if they were only examining ingredient cost data? Med D copays are greater than Medicaid copays thereby reducing Med D ingredient cost payments to pharmacies. This would need to be normalized for a true comparison.
    5) Multi-source drugs-Only 39 drugs with several where multi-source status had recently taken place. Why is this important--FFS Medicaid programs paid for the brands because the net cost after rebates was below that of the generics.
    6) OIG and CMS relationship-while these two bodies are both federal entities--they don't get along, period. OIG knows which states to select to make their point to CMS. Will the OIG ever use comparison states that manage a tight program? How about Massachusetts, Indiana, Maine or Minnesota?

    This is exactly why the CMS Retail Price Survey is so important. It will eliminate all of these poorly done studies and be updated monthly to provide real benchmarks for comparisons.

  19. Check reimbursement in Rhode Island, WAC + 3.40 dispensing fee, mac is lowest in the country, all pharmacies still take Medicaid, or many patients are in managed care plans administered by UHC, or Neighborhood Health....Both these managed care plans pay better than straight medicaid...

  20. Adam,

    The best way to learn any industry (or anything for that matter) is NOT simply studying the numbers and trends. One needs to to get the "real feel" of the operation by working in the trenches, with the small business owners, viewing how claims are processed, and visiting with employees & customers.

    There has been storm clouds hanging over the retail pharmacy world for the past five years, and most folks feel it's getting you can tell from your readership.

    I'd suggest you spend a couple half-days at an indy pharmacy and shadow the owner. Ask for specific drugs which prove margin erosion. See where the dollar (and pct) margin have gone over the past five years. Get the gut read from the druggist.

    Kind of like a field trip for grade school kids. Hands on is the best way to learn.

    Take care. b

  21. Adam,

    Has Lewin responded to the request for comment you said you'd send them on 12/9/2010?

    A Medicaid Fan

  22. Lewin Group would not comment on the blog. Frankly, I don't blame them since most of the comments (including yours) are anonymous.

    I re-read the paper and think many of the "critiques" above are addressed in the methodology of Lewin's paper. For instance, some people above have criticized Lewin's data. However, I see that they used the 2009 Medicaid data from the CMS website. (See their footnote 4 on page 9.)

    As I state above, I do not claim that the study is perfect. However, Lewin appears to have analyzed the available public data in a relatively transparent manner. I presume that "medicaid fans" out there will publish an alternative (and more favorable) analysis at some point.

    Have a happy new year!


  23. Adam,

    You make a valid point, and I suppose I don’t blame them for not responding to anonymous comments (including mine). I wonder if they would also decline to respond to a more official inquiry. Perhaps we’ll find out.

    I also agree that one would think Lewin’s source of states’ information, CMS (link provided on pg 24 of Lewin’s report: Source: CY2009 CMS website data. Available Online:, would result in fairly accurate projections. Turns out that’s not the case though. I’m disappointed that a group like Lewin apparently made no effort to validate what they were coming up with. For example, Lewin uses TN data to estimate pharmacy cost for each expansion enrollee (see pg 13), but did not validate the accuracy of what they apparently pulled down for TN from the CMS website. Page 24 of their report shows pharmacy expenditures of only $61 million for CY2010, but the actual expenditures are closer to $730 milion… yes, that’s right – they report 10% of the true value. With a quick Google search, I found a publically available document (that was posted well before this report, so the Lewin authors would have been able to find it if they had looked) reporting their actual expenditures (

    I don’t dispute that they are being transparent – and their transparency reveals that they really didn’t do their homework on this one.

    Thanks for answering my question, and a Merry Christmas to you!

    Medicaid Fan

  24. Your citation is to a document posted on the web site of a private public relations firm that represents health care providers. (

    I don't think it's appropriate to fault Lewin for pulling data from the official government web site. If CMS is publishing inaccurate data, then blame CMS, not Lewin.

    Enjoy the holiday!


  25. I don't fault them for pulling it from an official government website, I fault them for not validating it.

    If the New England Journal of Medicine decided to publish a study without peer-reviewing it just because they'd previosly gotten good information from that author, and it was subsequently determined to be a very flawed study, it would be NEJM taking the heat for that.

    Lewin certainly isn't the NEJM, but they surely have an obligation to make sure the information they're publishing isn't completely flawed before they run with it, in my opion.

    -Medicaid Fan

  26. I disagree. It's not a peer-reviewed academic paper. Neither Lewin nor PCMA have claimed otherwise. The conclusions of a privately published report are held to a different standard than the NEJM. I presume that my readers understand the difference.

    IMO, Lewin made sufficient disclosure of their methods and data. In fact, they disclosed enough that you've been able to raise some very specific criticisms.

    No study is perfect. Lewin made a useful contribution to the debate. Further research can now address any real or perceived deficiencies in their approach. That's the nature of the knowledge-generation process.


  27. Well, guess we'll have to agree to disagree about that one. Call me a perfectionist, but if I was the author of this study I'd sure feel like an idiot for being that far off on the basics.

    And I'd say that you're right - your readers do understand the difference between the NEJM and Lewin... unfortunately, a lot of the politicians that PCMA will be handing this out to don't. Believe me, that's one that I know from personal experience.

  28. Perhaps pharmacists should wish for socialized medecine like they have in Canada. Over there they reimburse $7 per Rx. No wonder their pharmacies are thriving and have no lines. I reimburse my pizza delivery guy more than these PBMs reimburse pharmacists.