Mississippi Medicaid Pharmaceutical Average Wholesale Price Litigation v. State ( 2015 )

IN THE SUPREME COURT OF MISSISSIPPI
NO. 2012-CA-01610-SCT
IN THE MATTER OF MISSISSIPPI MEDICAID
PHARMACEUTICAL AVERAGE WHOLESALE
PRICE LITIGATION: SANDOZ, INC.
v.
STATE OF MISSISSIPPI
DATE OF JUDGMENT: 08/06/2012
TRIAL JUDGE: HON. THOMAS L. ZEBERT
COURT FROM WHICH APPEALED: RANKIN COUNTY CHANCERY COURT
ATTORNEYS FOR APPELLANT: JOSEPH ANGLAND
MICHAEL JOHN GALLAGHER
LUTHER T. MUNFORD
RICHARD T. LAWRENCE
J. COLLINS WOHNER, JR.
ATTORNEYS FOR APPELLEE: CHARLES G. COPELAND
REBECCA SUZANNE BLUNDEN
GEOFFREY C. MORGAN
GEORGE W. NEVILLE
ANDY LOWRY
ELLEN PATTON ROBB
SAMUEL MARTIN MILLETTE
D. RONALD MUSGROVE
BLAKE DAMON SMITH
MICHAEL SHELTON SMITH, II
WILSON DANIEL “DEE” MILES, III
H. CLAY BARNETT, III
NATURE OF THE CASE: CIVIL - OTHER
DISPOSITION: ON DIRECT APPEAL: AFFIRMED.
ON CROSS-APPEAL: AFFIRMED -
10/29/2015
MOTION FOR REHEARING FILED:
MANDATE ISSUED:
EN BANC.
CHANDLER, JUSTICE, FOR THE COURT:
¶1. The State of Mississippi brought a civil action against generic pharmaceutical
provider Sandoz, Inc., (Sandoz) alleging that Sandoz impermissibly exploited Mississippi’s
Medicaid reimbursement program by routinely and exponentially reporting fictitious
“Average Wholesale Prices,” a key data factor in the federally supervised formula used by
the Mississippi Division of Medicaid to reimburse pharmacies serviced by Sandoz. The trial
court, sitting as fact-finder, found Sandoz in violation of the Mississippi Consumer
Protection Act and liable for common-law fraud. Sandoz appeals, and the State cross-
appeals. On our deferential standard of review, we affirm the trial court in full.
FACTS AND PROCEEDINGS BELOW
¶2. Medicaid is a joint state-federal program servicing hundreds of thousands of
Mississippians. Following the passage of the Omnibus Budget Reconciliation Act of 1990
(OBRA ’90), which governs Medicaid, the number of prescription drugs eligible for
reimbursement rose from approximately 1,800 to more than 65,000. More than 900
pharmacies eventually participated, submitting more than one million reimbursement claims
per month. The cost to each pharmacy for each drug from providers like Sandoz changed on
virtually a daily basis and varied depending on the location and nature of the pharmacy.
¶3. For the relevant damages period, Mississippi Medicaid was required by regulation to
reimburse pharmacies the lesser of (1) the Estimated Acquisition Cost (EAC) of the drug plus
a reasonable dispensing fee, or (2) the “usual and customary” price the pharmacy charges to
consumers paying for the drug without government assistance (U&C). 42 C.F.R. §
2
447.512(b).1 Federal Regulations defined EAC as Medicaid’s “best estimate of the price
generally and currently paid by providers for a drug marketed or sold by a particular
manufacturer or labeler in the package size of drugs most frequently purchased by
providers.”
¶4. The federal agency responsible for Medicaid is the Center for Medicare and Medicaid
Services (CMS), formerly known as the Health Care Financing Administration (HCFA).
CMS was required to approve of Mississippi’s “State Plan,” including its reimbursement
formula for prescription drugs. Any changes to Mississippi’s reimbursement rate had to be
approved by CMS. From May 1, 1990, to March 31, 2002, Mississippi Medicaid defined
EAC as the Average Wholesale Price minus ten percent. From April 1, 2002, to June 30,
2005, EAC was defined as AWP minus twelve percent. From July 1, 2005, until October
2005, EAC was defined as (1) the lesser of AWP minus twelve percent or Wholesale
Acquisition Cost (WAC) plus nine-percent for brand named drugs and single-source generic
drugs, and (2) AWP minus twenty-five percent for multiple-source generic drugs.
¶5. To satisfy its regulatory obligations and to efficiently process the large number of
reimbursement claims, Mississippi, similar to more than forty other programs nationwide,
obtained the drugs’ AWPs from national data service First Data Bank. First Data Bank
defined AWP as “the average price paid by the pharmacy to the wholesaler for a particular
drug.” Sandoz generated and submitted the AWPs for its drugs to First Data Bank. Sandoz
1
For the relevant damages period, U&C was employed for only eight percent of
reimbursement claims. Reimbursement was also sometimes based on the “Federal Upper
Limit,” defined at 150 percent of the lowest published number for a generic equivalent.
3
does not dispute that the AWPs it reported to First Data Bank never represented the net
average wholesale prices that Sandoz received for the sale of a particular drug to a pharmacy
or wholesaler, or that Mississippi Medicaid relied on Sandoz’s AWPs as published by First
Data Bank to determine the EAC for purposes of reimbursing pharmacies for the ingredient
cost of Sandoz’s drugs. The discrepancy between the actual cost of the drug and the reported
Average Wholesale Price averaged 886%.
¶6. Mississippi filed suit in 2005, alleging that Sandoz’s conduct of reporting inflated
AWPs caused Mississippi to reimburse pharmacies excessive amounts, and that Sandoz
deliberately created and marketed the discrepancy (or “spread”) between the actual prices of
their drugs and the reported AWPs in order to increase market share and profit.2 The State
asserted that Sandoz violated the Mississippi Consumer Protection Act, the Mississippi
Medicaid Fraud Control Act, and that its conduct constituted commonlaw fraud. Sandoz
argued that the reported AWPs were understood by all parties to be an essentially fictitious
sticker price that had no relation to the actual cost of the drugs.
¶7. In April 2011, the trial court heard more than ten days of testimony, including more
than twenty witnesses. This evidence included the testimony of both parties’ experts, the
testimony of Jack Lee, Mississippi Medicaid’s Director of Pharmacy; the testimony of Helen
Wetherbee, former Director of Mississippi Medicaid; the deposition testimony of Larry Reed,
Technical Director for CMS’s Division of Pharmacy; reports from the Office of Inspector
2
The suit initially was filed against some eighty-six drug companies, most of which
have settled. Sandoz was known as Geneva Pharmaceuticals, Inc., until 2003.
4
General (OIG); and First DataBank’s definitions of AWP from 1991 to 2003. Additional
relevant facts are discussed below.
¶8. The court entered its amended findings of fact and conclusions of law on October 4,
2011, finding that Sandoz committed common-law fraud and violated the Mississippi
Consumer Protection Act. For the damages period of January 1, 1991, through October 20,
2005, the court awarded $23,661,618 to cover the amount Mississippi Medicaid overpaid
in reimbursements to pharmacies, $2,699,000 in civil penalties under the Consumer
Protection Act, and $3,750,000 in punitive damages for Sandoz’s willful and fraudulent
misconduct. The court dismissed the claims against Sandoz under the Mississippi Medicaid
Fraud Control Act and denied the State’s post-trial motion for attorney’s fees, prejudgment
interest, and other relief. Sandoz appealed, and the State cross-appealed.
DISCUSSION
¶9. We will not reverse the trial court’s findings of fact unless manifestly wrong or clearly
erroneous. Puckett v. Stuckey, 633 So. 2d 978, 982 (Miss. 1993). “When a trial judge sits
without a jury, this Court will not disturb his factual determinations where there is substantial
evidence in the record to support those findings.” Transoacean Enter. v. Ingalls
Shipbuilding, Inc., 33 So. 3d 459, 462 (Miss. 2010) (citation omitted). “Fraud is essentially
a question of fact best left for the [fact-finder]. . .” Allen v. Mac Tools, Inc., 671 So. 2d 636,
643 (Miss. 1996). The nine elements of common-law fraud are:
(1) a representation, (2) its falsity, (3) its materiality, (4) the speaker’s
knowledge of its falsity or ignorance of its truth, (5) his intent that it should be
acted on by the hearer and in the manner reasonably contemplated, (6) the
5
hearer’s ignorance of its falsity, (7) his reliance on its truth, (8) his right to rely
thereon, and (9) his consequent and proximate injury.
Franklin v. Lovitt Equip. Co., Inc., 420 So. 2d 1370, 1373 (Miss. 1982). Each element of
fraud must be proven by clear and convincing evidence. Cotton v. McConnell, 435 So. 2d
683, 685 (Miss. 1983).3
¶10. For both the State’s common-law fraud and Consumer Protection Act claims, the
primary issue of fact determined below was the parties’ understanding of the meaning of the
term “Average Wholesale Price” as this phrase was used for purposes of pharmaceutical
reimbursement by the Mississippi Division of Medicaid. The trial court found clear and
convincing evidence supporting that the State understood the Average Wholesale Price data
generated by Sandoz to be related to the pharmacies’ actual drug acquisition cost, that the
State reasonably relied on this data to calculate its best estimate of the actual cost of the drugs
to the pharmacies, that Sandoz intentionally reported fictitious prices that had zero
relationship to the actual cost of the drug while knowing that Medicaid programs used these
prices to estimate how much they should reimburse pharmacies, that Sandoz advertised to
pharmacies the discrepancies between the actual and fictitious prices to attract business, and
that Medicaid reasonably relied on the false data, reimbursing pharmacies $23,661,618 in
3
The dissent, along with reweighing the credibility of the evidence in contravention
of the manifest-error standard of review, adopts an outlier position in its evaluation of the
elements of fraud. Even states that have found that their respective Medicaid programs had
sufficient knowledge of the level of local inflation to defeat claims of fraud still found the
represented AWPs to be false, misleading, and deceptive. Sandoz, Inc., v. Commonwealth
ex rel. Conway, 405 S.W 506, 511 (Ky. Ct. App. 2012); see also AstraZeneca LP v. State,
41 So. 3d 15 (Ala. 2009) (Alabama Medicaid had internal pricing studies on 80% of its
pharmacies, compared here to Mississippi’s sample survey of three pharmacies).
6
excess of the regulatory minimum. We find that the State presented evidence sufficient to
have enabled the trial court to have found the elements of fraud by clear and convincing
evidence.
¶11. The court was confronted with evidence demonstrating the significant changes
Mississippi Medicaid’s pharmacy division faced following the enactment of OBRA 90,
including a shift in the most frequently applied reimbursement formula to calculate the
lowest reimbursement rate reasonably possible for the new open formulary. Rather than most
frequently reimbursing the “usual and customary” price the pharmacy charges to consumers
paying for the drug without government assistance, Medicaid now reimbursed most often by
its best estimate of the acquisition cost plus a reasonable dispensing fee. The reported
Average Wholesale Price was a data factor in 100 percent of reimbursements made under this
method. Sandoz generated and reported its AWPs to First DataBank, to which Medicaid,
along with forty-five other similar programs, subscribed.4
¶12. Helen Wetherbee, the Executive Director of the Division of Medicaid from 1991 to
1999, described in detail the limited resources available to the pharmacy division of
Mississippi Medicaid at the time it was tasked with implementing the new reimbursement
policies under OBRA 90 and after. This description included the necessity of relying on First
DataBank for reported AWPs and the impossibility of surveying pharmacists to get invoice
4
The dissent misleadingly focuses on Medicaid’s “decades” of prior knowledge that
AWPs needed a discount to arrive at a literally accurate number. This is undisputed. The
question for the relevant damages period became the level of knowledge of the degree of
inflation. Moreover, for the first four years of OBRA ’90’s enactment, reimbursement
discounts were frozen by federal law.
7
prices for each drug before reimbursement. The pharmacy division was staffed with two
people in program integrity and three people in program services. She testified that there was
“no practical alternative” to obtaining prices other than subscribing to First DataBank:
We didn’t have the computer resources to do that. We certainly didn’t have the
staff to do anything manually. First DataBank was the national standard, the
gold standard, if you will, and so I felt very comfortable relying on First
DataBank as being the best possible way I knew to serve the Medicaid
program and get prompt, accurate information with respect to pharmacy
product pricing.
The trial court agreed, stating that Medicaid’s decision to rely on Sandoz’s published AWPs
was “the best and most efficient way for [Mississippi Medicaid] to obtain accurate pricing
information for Sandoz drugs.”5
¶13. Mississippi Medicaid was not alone in relying on the numbers reported to First
DataBank, and the evidence supported the finding that this reliance was reasonable. Forty-
five similar programs nationwide used First DataBank to obtain AWPs. The industry did not
expect Medicaid programs independently and continually to investigate the ever-changing
actual price of 65,000 or more prescription drugs purchased by 900 or more pharmacies state-
wide. As the trial court concluded, “because of the constantly changing prices for drugs on
virtually a daily basis, obtaining this accurate information was virtually impossible
considering DOM’s lack of resources.” Other jurisdictions agree. As Wisconsin successfully
argued:
5
This discredits the dissent’s implication that Medicaid somehow copped out by
setting reimbursement rates based on nationally available subscription numbers rather than
by independently investigating and obtaining hundreds of thousands of individual invoices
from pharmacies on a biweekly basis.
8
As with other states throughout the country, the solution that emerged was for
pharmaceutical manufacturers like Pharmacia to report certain figures to
calculate reimbursements. The most important of those figures was AWP.
Pharmacia provided AWPs for its drugs to First DataBank, an independent
company that organized and disseminated information regarding the
pharmaceutical industry to Medicaid, which then plugged them into its
reimbursement formula.
Pharmacia, as with all manufacturers, reported AWPs in agreement with
Medicaid that the AWPs were supposed to reflect what the name suggested:
the average price for which the drug was sold by the wholesaler to the
pharmacy.
State v. Abbott Laboratories, 316 N.W.2d 145, 152-3 (Wis. 2012). See also In re Pharm.
Industry Average Wholesale Price Litigation v. AstraZeneca Pharmaceuticals LP., 582
F.3d 156 (1st Cir. 2009).
¶14. First DataBank defined AWP as “the average price paid by the pharmacy to the
wholesaler for a particular drug.” Wetherbee testified that Medicaid’s understanding of AWP
was consistent with the idea that AWP was developed because there had to be some price
which all parties could agree upon if machine processing of claims was to be possible, with
AWP representing an average price which a wholesaler would charge a pharmacy for a
particular product. First DataBank also stated that “AWP never means that every purchase
of that product will be exactly that price. There are many factors involved in pricing at the
wholesale level which can modify the prices charged even among a group of customers from
the same wholesaler.” The State never denied that these “many factors” informed its
understanding of AWP, including that, to arrive at a true estimated acquisition cost,
reimbursement based on AWPs should be discounted. However, Wetherbee testified, “all of
9
us who were involved in pharmacy reimbursement believed we were getting accurate
information and we were paying a real estimated acquisition cost.”
¶15. Undisputed is that pharmaceutical providers came to exaggerate exponentially their
reported AWPs to the extent that the reported price, even as discounted, had zero relation to
the pharmacies’ actual acquisition costs.6 This “spread” was then used as a marketing tool
to gain pharmacy business in a competitive generic market.7 The trial court concluded that
the AWPs “were not prices at all – otherwise, if they had been, the State’s reimbursement
system would have worked as intended.” Christopher Worrell, Sandoz’s Vice President of
Sales and Marketing from 2002 to 2005 and the individual responsible for assigning AWPs,
denied that Sandoz had engaged in the practice of “marketing the spread” between AWPs
6
The increasingly pervasive pharmaceutical practice of arbitrarily assigning
artificially high drug prices for profit is currently at the forefront of the national conversation
on healthcare costs and corporate ethics. See Andrew Pollack, Sabrina Tavernise, A Drug
Company’s Price Tactics Pinch Insurers and Consumers. N.Y. Times (Oct. 4, 2015),
http://www.nytimes.com/2015/10/05/business/valeants-drug-price-strategy-enriches-it-but
-infuriates-patients-and-lawmakers.html; Andrew Pollack, Once a Neglected Treatment, Now
an Expensive Specialty Drug, N.Y. Times (Sept. 20, 2015),
http://www.nytimes.com/2015/09/21/business/a-huge-overnight-increase-in-a-drugs-price
-raises-protests.html.
7
As explained in Abbott Laboratories:
Over time, however, the manufacturers began reporting inflated AWPs. They
did so to engage in a practice known as “marketing the spread.” When a
manufacturer “marketed the spread,” it reported an inflated AWP to Medicaid
and Medicaid then paid the pharmacy more for the drug than the pharmacist
paid the wholesaler for the same product. . . . Pharmacia “marketed the
spread” along with its competitors, and it reported more and more dramatically
inflated AWPs over time.
Abbott Laboratories, 816 N.W.2d at 153.
10
and actual prices to gain pharmacy business. However, the court was presented with evidence
supporting that Sandoz did “market the spread,” including a 1992 letter to a customer stating
“[w]e offer substantial margins between acquisition costs and [Average Wholesale Price] for
your profit potential.” In 1998, Sandoz bragged to a client that its reported AWP for
Ranitidine was $86, but the acquisition price for the pharmacy would be only $7.16, a spread
of 1,201 percent. Additionally, the court saw examples of confidential8 offers to customers
such as WalMart, listing under a description of the drug prices both the amount of the
contract price and the different AWP amount. Evidence of this practice also was presented
to the court in the form of the OIG investigative reports of the growing problem at a national
level. Worrell also testified that he was aware “there was some other reimbursement formula
that might use AWP” but that from his perspective “[i]t was a very small piece of the
business.” Presented with conflicting evidence, it was the place of the trial court, sitting as
finder of fact, to weigh the credibility of the witnesses and evidence. The evidence supported
the conclusion by the finder of fact that Sandoz advertised two key data points to its potential
pharmaceutical customers: the actual cost of the drug and the fictional cost of the drug, and
that Sandoz then reported the fictional cost of the drug to First DataBank, knowing that
Medicaid programs relied on that data to calculate the actual cost of the drug.
8
The confidential trade-secret nature of the spreads further served to prevent
programs from ascertaining the exact degree of inflation relevant to local programs.
Wetherbee testified that by the late 1990s, “I would say there was a great deal of information
suggesting that AWP was a distorted or inflated figure[,]” but that “[w]hen I was at the
Division of Medicaid, I did not know what the spread was.” This quote explicitly rebuts the
dissent’s pervasive assertion that the State did not take the position below that it was aware
that AWPs were inflated.
11
¶16. The evidence also is undisputed that Medicaid programs became increasingly aware
that the reported AWPs were exaggerated and that they reacted to that awareness by
increasing the discount of AWP. But a qualitative distinction exists between knowledge that
an actual price is exaggerated and knowledge that a “price” is fictitious and assigned solely
for the purpose of creating a fraudulent profit margin. The State persuasively demonstrates
the distinction between knowing that a representation is likely to be only approximately true,
versus knowing that it is a complete fabrication, by pointing to cases involving the sale of
land. A defendant may know that a representation of a building’s square footage is only
approximate, or that a parcel of land is “about” a certain number of acres, but have a viable
fraud claim where the represented numbers were “materially and unreasonably inaccurate”
or “grossly erroneous.” See Kenneally v. Bank of Nova Scotia, 711 F. Supp. 2d 1174, 1185-
1186 (S.D. Cal. 2010).9 This is an apt comparison to Mississippi Medicaid’s understanding
of the price AWP represented. Mississippi knew that AWP “never means that every purchase
of that product will be exactly at that price” and that “many factors” are involved in pricing
that can “modify” the actual price from the reported AWP. What the State was unaware of
was the astronomical discrepancy between the actual cost and the reported AWP, a
discrepancy that averaged 886% and caused Mississippi taxpayers to overreimburse
pharmacies $23,661,618.
9
See also Parker v. Bennett, 231 S.E.2d 10, 12-13 (N.C. Ct. App. 1977) (reversing
summary judgment for defendant who told buyers that parcel was “125 acres, more or less”
but size was ninety-five acres).
12
¶17. To accept Sandoz’s argument that the State knew the reported AWPs had zero
relationship to actual cost would be to say that Medicaid essentially colluded with the generic
drug industry for more than a decade in exploiting the inefficiencies inherent to a government
bureaucracy in violation of Medicaid’s own statutory and regulatory obligations. Wetherbee
testified “[m]y basis [for understanding the meaning of AWP] was the universal awareness
that we were a government program using public funds with a determination to be good
stewards of the money, and I hope a reputation for integrity, basing our reimbursement on
what we thought was accurate information.”10
¶18. The court was presented with evidence supporting that the timing and degree of
adjustments to Mississippi’s reimbursement formula was reasonable given the State’s lack
10
The dissent’s emphasis that Sandoz submitted “Average Manufacturer Prices,” or
AMPs, to First DataBank for the first six years of the damages period overlooks several
relevant points. AMPs were used to calculate federal rebates, not reimbursement, and states
actually were prohibited by CMS from using the AMP figure as a basis for reimbursement.
Also, for the first four years of the implementation of OBRA ’90, state reimbursement rates
were frozen by federal law. We note the coincidence that Sandoz stopped providing that data
to the subscription service after States’ abilities to propose changes to their reimbursement
formulas unfroze and as the confidentially marketed “spread” began to increase exponentially
past anything expected or seen before. The early presence of the AMPs in First Data Bank
does not defeat the trial court’s ultimate determination of credibility upon weighing all of the
evidence. Moreover, the burden was not on the State to ferret out the falseness of Sandoz’s
representation. The court found Wetherbee’s testimony credible that:
Well, the barrier was knowledge. I think if I had had an awareness of all of the
problems associated with AWP that were developing or beginning to develop
at that time, then the AMP would have meant much more to me. . . . I have
since learned that states were prohibited by HCFA from using this AMP figure
as a basis for their reimbursement. . . [reimbursement] is an accounting
function, and the rebates were independent of the reimbursement. This was
brand new to all of us. Rebate was a new game.
13
of knowledge of the degree of AWP inflation as it uniquely affected Medicaid
reimbursement in Mississippi. The 1997 and 2002 OIG reports, while an important impetus
to motivating investigation and (what is, by necessity a slow regulatory process) eventual
change, were not immediately conclusory to Mississippi’s individual concerns. Wetherbee
explained:
Mississippi is at the extreme, both in terms of its poverty level and the rural
nature of its population demography so that what would apply in New York or
California may or may not apply in Mississippi . . . in Mississippi we have
higher level of people living below the poverty level and a greater number of
people living in rural areas, because of which we would also have a higher
proportion of community pharmacies in rural areas. So there were a lot of
reasons why a generalized nationwide determination. . . would certainly be of
interest but . . .I did not conclude that [the OIG report] was directly applicable
to Mississippi. . . . It would be most likely that a chain pharmacy would be the
beneficiary of some sort of purchasing contract or quantity discount that an
independent pharmacy might enjoy.11
A subsequent “look-and-see” investigation by Medicaid Pharmacy Director Jack Lee into the
pricing of a small handful of pharmacies led to a recommendation that “AWP minus 10
percent might no longer be applicable in Mississippi.” The discount subsequently was
raised–following CMS approval– to 12 percent and then again to 25 percent.12 Mississippi
could not act without accountability in setting a plan that would satisfy its regulatory
11
The position of independent, rural pharmacies is born out in the current national
conversation on healthcare costs. See How Generic Drugs Can Cost Small Pharmacies Big
Bucks, National Public Radio, (Oct. 22, 2015,
http://www.npr.org/sections/health-shots/2015/10/22/450600567/how-generic-drugs-can
-cost-small-pharmacies-big-bucks (last visited Oct. 28, 2015).
12
This Court acknowledged the practice of defining EAC as an increasing discount
of Average Wholesale Price in Jones v. Howell, 827 So. 2d 691 (Miss. 2002).
14
obligation to reimburse at the lowest rate reasonably possible. Larry Reed, Technical Director
of CMS’s Division of Pharmacy, explained that CMS would not approve a state
reimbursement plan that did not provide an acceptable method to arrive at the estimated cost
for the drugs. Sandoz’s own expert Robert Helms testified that CMS would not approve a
change in the reimbursement formula that matched the highest OIG estimate of AWP
inflation in other parts of the country. Wetherbee stated that, in a testament to the quality of
work Mississippi invested in its proposed state plans, that Mississippi Medicaid “never had
a federal disallowance with respect to our pharmacy reimbursement.”
¶19. The trial court was presented with evidence to support that Medicaid’s reimbursement
policies consistently were based on an understanding that adjusting the reported AWPs would
arrive at its best estimate of the actual cost, factoring in unpredictable variations in regional
markets. See In re Pharm. Industry Average Wholesale Price Litigation, 582 F.3d 156, 188
(1st Cir. 2009) (affirming finding that “costs of acquiring and acting upon the information
necessary to understand the full extent of the AWP inflation were prohibitive.”).
¶20. In light of the conflicting testimony regarding the parties’ understanding of the term
AWP, the trial court appropriately rejected Sandoz’s argument that all parties understood
AWP to be a fictitious “term of art.” Courts refrain from establishing a technical term of art
when there is conflicting testimony regarding the definition of the term. Commonwealth of
Mass. v. Blackstone Valley Electric Co., 67 F.3d 981, 986 (1st Cir. 1995). A departure from
the plain meaning of a term is only warranted when the term has a “well-defined”
understanding within the relevant field in which it is used. Corning Glass Works v.
15
Brennan, 417 U.S. 188, 201-202, 94 S. Ct. 2223, 41 L. Ed. 2d 1 (1974). Not only this
litigation, but current litigation nationwide supports that AWP was not “well-defined” as a
fictitious term of art having no value to reimbursement calculations. The trial court
exhaustively considered the evidence before it, and we cannot find that the trial court
manifestly erred in determining that Sandoz’s conduct rose to the level of common-law fraud.
Mississippi’s Consumer Protection Act.
¶21. The trial court found that “Sandoz’s submission of [Average Wholesale Prices] to
[First Data Bank] that have no predictable relationship to what customers pay for its drugs
while knowing Mississippi relied on this information in determining [Estimated Actual Cost]
is an unfair and deceptive trade or practice within the meaning of Mississippi’s Consumer
Protection Act. Mississippi’s Consumer Protection Act (CPA) prohibits “unfair methods of
competition affecting commerce and unfair or deceptive trade practices in or affecting
commerce.” The CPA additionally prohibits “misrepresentations of fact concerning the
reason for, existence of, or amounts of price reductions.” See Miss. Code Ann. § 75-24-5
(Rev. 2009). The purpose of the CPA is “to protect the citizens of Mississippi from deceptive
and unfair trade practices.” Holman v. Howard Wilson Chrysler Jeep, Inc., 972 So. 2d 564,
572 (Miss. 2008).
¶22. Mississippi does not require a finding of fraud for an act to be deemed unfair or
deceptive, making our affirmance of the trial court’s judgment on this issue even easier than
affirming its finding of fraud above. Holman v. Howard Wilson Chrysler Jeep, Inc., 972
So. 2d 564, 572 (Miss. 2009); see also AstraZeneca, 582 F.3d at 184 (holding act or practice
16
can be unfair or deceptive even if no violation of common law); Wash. State Physicians Ins.
Exch. & Ass’n v. Fisions Corp., 858 P.2d 1054, 1063 (Wash. 1993) (affirming jury verdict
for violation of similar statute when jury also returned verdict denying fraud claim).
¶23. Rather, when “construing what constitutes unfair or deceptive trade practices . . . the
courts will be guided by the interpretations given by the Federal Trade Commission and the
federal courts to Section 5(a)(1) of the Federal Trade Commission Act (15 USCS 45(a)(1))
as from time to time amended.” Miss. Code Ann. § 75-24-3. A trade practice is unfair if it
(1) causes or is likely to cause a substantial injury; (2) the injury is not “outweighed by any
countervailing benefits to consumers or competition that practice produces;” and (3) the
injury could not have been “reasonably avoided.” 15 U.S.C. § 45(n)(1980). In reviewing
whether a trade practice is unfair, the FTC “may consider established public policies as
evidence to be considered with all other evidence.”
¶24. Evaluating these factors, we find that the trial court did not manifestly err in
determining that Sandoz’s conduct was unfair and deceptive within the meaning of
Mississippi’s Consumer Protection Act. Sandoz’s unfair reporting of fictitious AWPs caused
the substantial injury of $23,661,618 in overpayments to pharmacies participating in
Medicaid. No “countervailing benefits to consumers or competition” resulted from Sandoz’s
conduct. And we already have discussed above that Medicaid’s reliance on the reported
AWPs, which mirrored the practice of similar programs nationwide, was reasonable. The trial
court did not err in finding that Medicaid should not have had pharmacies submit cost
information for every Medicaid reimbursement submission. See AstraZenca, 582 F.3d at 188
17
(affirming finding that “costs of acquiring and acting upon the information necessary to
understand the full extend of the AWP inflation were prohibitive.”); Abbott Labs., 341 Wis.
2d at 524-25 (“Because of the complexity and dynamism of the pharmaceutical industry,
Medicaid required a consistent and broadly applicable formula for determining the
appropriate reimbursements for pharmacies. . . .”).
¶25. The trial court did not err in rejecting Sandoz’s argument that the State is not a
consumer for purposes of the Act. Mississippi is not the first jurisdiction to conclude that the
inflation of Average Wholesale Prices violates standard Consumer Protection Act provisions
and that the State is appropriately viewed as a consumer for purposes of the act. As the trial
court stated “[s]ignificantly, Courts involved in virtually identical AWP litigation in other
jurisdictions have applied similar State Consumer Protection Acts and rejected identical
arguments as Sandoz asserts here. ‘It is apparent that the State is the ultimate purchaser in
the chain of distribution and the one directly affected by the alleged manipulation of the
AWP.’” See Idaho v. Alpharma, USPD, Inc., No. CV-0C-0701847 (Id. Dist. Ct. Aug. 31,
2007); Kentucky v. Alpharma, USPD, Inc., No. 04-CI-1487; Common Wealth of
Pennsylvania v. Bristol-Myers Squibb Co., No. 212 M.D. 2004 (Pa. Commw. Ct. Sept. 10,
2010); Commonwealth v. Johnson & Johnson, No. 212 M.D. 2004 (Pa Commw. Ct. Dec.
7, 2010); see also In re Pharmaceutical Industry Average Wholesale Price Litigation v.
AstraZeneca, 582 F.3d 156 (1st Cir. 2009) (affirming trial court’s finding drug company
liable for unfair and deceptive business practices for publishing inaccurate “average
wholesale price”).
18
¶26. The trial court did not err in finding that Sandoz’s practice of reporting false AWPs
was a deceptive trade practice. The “prices” were exaggerated to an extent that they were not
even prices. Providing false information is deceptive and violates the Act. See Holman, 972
So. 2d at 571-72; S.W. Starving Artists Group, Inc. v. State, 364 So. 2d 1128, 1131 (Miss.
1978). People v. Pharmacia Corp., 895 N.Y.S.2d 682, 693-94 (N.Y. Sup. Ct. 2010) (holding
that where pharmaceutical company knew of reliance on its published prices, it “had an
obligation to refrain from reporting prices in a fraudulent or deceptive manner or causing
fraudulent or deceptive prices to be published.”).
The State’s Cross-Appeal
I. The trial court did not abuse its discretion by denying the State’s
request for attorneys’ fees.
¶27. The State first requested attorneys’ fees in its motion to alter the final judgment. The
trial court denied the State’s motion. The State argues that the trial court erred by declining
to award attorneys’ fees because attorneys’ fees should be awarded whenever punitive
damages are awarded, and because the State was entitled to recover attorneys’ fees under the
CPA. See Miss. Code Ann. § 75-24-19(1)(b) (Rev. 2009).
¶28. “Absent some statutory authority or contractual provision, attorneys’ fees cannot be
awarded unless punitive damages are also proper.” Miss. Power & Light Co. v. Cook, 832
So. 2d 474, 486 (Miss. 2002). When punitive damages are awarded, a post-judgment request
for attorneys’ fees is proper. Fulton v. Miss. Farm Bureau Cas. Ins. Co., 105 So. 3d 284,
285 (Miss. 2012). But even when punitive damages are appropriate, “the allowance and the
amount of a fee is a matter committed to the sound discretion of the trial judge.” Smith v.
19
Dorsey, 599 So. 2d 529, 550 (Miss. 1992). “Attorney fees may be awarded in cases where
punitive damages are proper . . . .” Valley Forge Ins. Co./CNA Ins. Co. v. Strickland, 620
So. 2d 535, 542 (Miss. 1993) (emphasis added). And an award of attorneys’ fees to the
Attorney General is discretionary under the CPA. Miss. Code Ann. § 75-24-19(1)(b) (Rev.
2009) ([t]he Attorney General may also recover . . . a reasonable attorney’s fee) (emphasis
added).
¶29. Whether or not to award attorneys’ fees is within the sound discretion of the trial
court. Smith, 599 So. 2d at 550. Contrary to the State’s arguments, a punitive damages award
does not mandate an award of attorneys’ fees; whether to award attorneys’ fees always
remains within the trial court’s discretion. Smith, 599 So. 2d at 550. Nor does the CPA make
an award of attorneys’ fees mandatory upon the State’s recovery of penalties. Miss. Code
Ann. § 75-24-19(1)(b) (Rev. 2009). Rather, because the CPA states that the Attorney General
may recover attorneys’ fees, the decision to award attorneys’ fees under the CPA is within
the discretion of the trial court. Miss. Code Ann. § 75-24-19(1)(b) (Rev. 2009). We cannot
say that the trial court abused its discretion in denying the State’s request for attorneys’ fees.
II. The trial court did not abuse its discretion by declining to award
prejudgment interest.
¶30. An award of prejudgment interest is within the trial court’s discretion. In re Estate
of Smith, 69 So. 3d 1, 4 (Miss. 2011). The trial court declined to award prejudgment interest
on the State’s successful common-law fraud claim.13 Citing Upchurch Plumbing, Inc. v.
13
The trial court correctly held that no prejudgment interest was available on the CPA
claim because the CPA does not provide for prejudgment interest.
20
Greenwood Utilities Commission, 964 So. 2d 1100, 1117 (Miss. 2007), the trial court found
that prejudgment interest would not be allowed because the fraud claim was unliquidated.
Upchurch held that “prejudgment interest may be allowed in cases where the amount due is
liquidated when the claim is originally made or where the denial of a claim is frivolous or in
bad faith.” Id. (quoting Stockstill v. Gammill, 943 So. 2d 35, 50 (Miss. 2006)).
¶31. The trial court’s denial of prejudgment interest was within its discretion.
“Prejudgment interest ‘is not imposed as a penalty for wrong doing; it is allowed as
compensation for the detention of money overdue.’” Terex Corp. v. Ingalls Shipbuilding,
Inc., 671 So. 2d 1316, 1323, 1324 (Miss. 1996) (Sunburst Bank v. Keith, 648 So. 2d 1147,
1153 (Miss. 1995)). Here, the damages were unliquidated, and the trial court did not find bad
faith. While the State argues that Sandoz’s fraudulent conduct was evidence of bad faith, “the
primary focus of the law in this area concerns bad faith insurance claims,” and this case does
not involve a bad-faith denial of an insurance claim or of money due under a contract. Terex
Corp., 671 So. 2d at 1323. We affirm the trial court’s denial of the State’s prejudgment
interest claim.
III. The trial court appropriately applied the statutory punitive-
damages cap to the State.
¶32. The State argues that, due to its sovereignity, this action is excluded from the punitive-
damages cap in Mississippi Code Section 11-1-65(3)(a). Section 11-1-65(3)(a) states, in
relevant part, that:
(a) In any civil action where an entitlement to punitive damages shall have
been established under applicable laws, no award of punitive damages shall
exceed the following:
21
....
(iv) Three Million Seven Hundred Fifty Thousand Dollars
($3,750,000.00) for a defendant with a net worth of more than One
Hundred Million Dollars ($100,000,000.00) but not more than Five
Hundred Million Dollars ($500,000,000.00). . . .
Miss. Code Ann. § 11-1-65(3)(a) (Rev. 2014).
¶33. The State argues that its ability to recover damages cannot be limited by a statute
unless the statute places an express limitation on the State. The State relies on Jackson v.
Mississippi Building Commission, 350 So. 2d 63, 64 (Miss. 1977). In Jackson, the State
Building Commission filed a complaint arguing that the City unlawfully had required the
Commission to obtain a building permit and pay fees under municipal building codes. Id. at
63-64. A statute authorized the Commission to construct state buildings. Id. at 65-66. This
Court held that the statutes granted the Commission plenary power to construct buildings,
and the municipal building codes, which did not expressly subject the State to the codes, did
not apply to the State. Id. at 66. Jackson cited common-law principles to the effect that
statutes cannot divest the State’s rights in general terms, but must express an intent to limit
the State. Id. at 64-65. Jackson stated:
By resorting merely to well-known principles of statutory construction,
it is evident that the State may not be restricted in its sovereignty except by the
specific provisions of its statutes.
...
[T]he Court said: “It is the settled doctrine that the general words of a statute
do not include the State, or affect her rights, unless she be specially named, or
it be clear and indisputable from the act that it was intended to include the
State.” . . . “It is undoubtedly the general rule that, where the effect of a statute
is to restrict the rights of, or impose liabilities upon, the state or its political
22
subdivisions, it will be held to be inapplicable to them, unless they are
included expressly or by necessary implication.”
Id. (quotations omitted). Jackson also rested its holding upon the principle that a statute
granting plenary power to a state agency supersedes any conflicting local or general
regulations. Id. at 66.
¶34. A plain reading of Section 11-1-65(3)(a) indicates that it applies to actions by the
State, either “expressly or by necessary implication.” Jackson, 350 So. 2d at 65. Section 11-
1-65 states that it applies to “any civil action” where an entitlement to punitive damages is
established. Miss. Code Ann. § 11-1-65(3)(a). “[T]he word ‘any’ has an expansive meaning,
that is, ‘one or some indiscriminately of whatever kind.’” United States v. Gonzales, 520
U.S. 1, 5, 117 S. Ct. 1032, 1035, 137 L. Ed. 2d 132 (1997). “If the words of a statute are clear
and unambiguous, the Court applies the plain meaning of the statute and refrains from using
principles of statutory construction.” Lawson v. Honeywell Int’l, Inc., 75 So. 3d 1024, 1026
(Miss. 2011). We hold that, because the plain language of Section 11-1-65(3)(a) includes
actions by the State, the trial court properly applied the statutory punitive-damages cap to the
State.14 And, although the State argues that it should be exempt from the caps because it
brought the action on behalf of its citizens, an action by individual citizens would be subject
to the caps, and the State cannot recover more for the citizens than the citizens themselves
could recover.
14
We note Sandoz’s argument that one of the large punitive damages awards
motivating tort reform in 2002 involved a $175 million verdict in favor of the Mississippi
Tax Commission. While the statute is unambiguous and is not subject to construction, this
argument supports that the legislative intent was that the punitive-damages caps apply to the
State.
23
¶35. The State devotes a short paragraph in its brief to an argument that the Mississippi
Constitution forbids the Legislature to abrogate the State’s right to collect punitive damages.
Because the State did not raise this argument before the trial court, it is procedurally barred
from appellate consideration. “This Court’s general policy is that ‘errors raised for the first
time on appeal will not be considered, especially where constitutional questions are
concerned.’” Powers v. Tiebauer, 939 So. 2d 749, 752 (Miss. 2005) (quoting Stockstill v.
State, 854 So. 2d 1017, 1023 (Miss. 2003)).
IV. The trial court properly used Sandoz’s net worth from the end of
the prior fiscal year in capping punitive damages.
¶36. As an alternative to its argument that the statutory punitive-damages caps do not apply
to it, the State argues that the trial court did not correctly evaluate Sandoz’s net worth for the
purposes of selecting the appropriate punitive-damages cap. Section 11-1-65(3)(a) caps
punitive damages based on a defendant’s net worth. Miss. Code Ann. § 11-1-65(3)(a) (Rev.
2014). The parties’ discovery efforts yielded quarterly balance sheets beginning in 2010 for
each quarter leading up to the March 2012 punitive-damages hearing, and monthly income
statements from 2010 up to the date of the hearing. At the hearing, Sandoz presented
testimony of chief financial officer James Mastakas that, according to generally accepted
accounting principles, the net worth of Sandoz on December 31, 2011, was a number under
five hundred million dollars.
¶37. Under Section 11-1-65(3)(a)(iv), if a defendant’s net worth is between $100 million
and $500 million, punitive damages are capped at $3.75 million. The State attempted to show
that Sandoz’s net worth in 2012 had risen to more than $500 million, which would have
24
entitled the State to $5 million in punitive damages under the applicable cap. During cross-
examination of Mastakas, the State used 2012 balance sheets and income statements in an
attempt to show that, after January 1, 2012, Sandoz’s net worth had increased to more than
$500 million. But Mastakas stated that the information used by the State was insufficient to
calculate Sandoz’s net worth after January 1, 2012, according to generally accepted
accounting principles. The State presented no expert testimony establishing that Sandoz’s net
worth after January 1, 2012, exceeded $500 million under generally-accepted accounting
principles.
¶38. The trial court assessed punitive damages based on Sandoz’s net worth as of
December 31, 2011. The State argues that the trial court erred by using Sandoz’s net worth
as of December 31, 2011, rather than its net worth as of the date of the judgment on punitive
damages. The State argues that the chancellor’s use of December 31, 2011, as the valuation
date was error because, in Franklin Corp. v. Tedford, 18 So. 3d 215, 241 (Miss. 2009), this
Court held that net worth must be determined at the time of the judgment.
¶39. “The proper assessment of punitive damages under Mississippi Code Section
11-1-65(3) is a question of law, to be reviewed by this Court de novo.” Id. at 241. We find
no error in the trial court’s net-worth determination. Section 11-1-65(3)(b) states that “the
amount of the net worth shall be determined in accordance with Generally Accepted
Accounting Principles.” Miss. Code Ann. § 11-1-65(3)(b) (Rev. 2014). In Franklin, the
Court held that generally accepted accounting principles contemplate that net worth be the
25
current net worth, or the net worth at the time of the judgment on punitive damages.
Franklin, 18 So. 3d at 242. Quoting the trial court’s opinion, the Court stated that:
[T]he language of Miss. Code Ann. Section 11-1-65 which relates to the
imposition of the legislative caps to a punitive damage award provides that the
net worth of the defendant “shall be determined” in accordance with Generally
Accepted Accounting Principles, and that such language implies that the
current net worth of the defendant is to be considered. Further, other portions
of the statute refer to the net worth of the defendant as a factor to be
considered in an effort to determine the defendant’s financial ability to pay the
award, and likewise implies that the current net worth of the defendant is to be
utilized. . . . Further, there is no language in the statute which provides that the
past net worth of the defendant is to be utilized, and without such
distinguishing language, this court must apply the common meaning of the
term “net worth,” as well as the common interpretation as afforded by a
reading of the statute as a whole.
Id. (emphasis in original).
¶40. We observe that the trial court in Franklin used the defendant’s net worth as of
December 31, 2006, but the final judgment was not entered until May 31, 2007. Id. at 230.
So, although Franklin stated that net worth should be calculated at the time of the judgment,
it in fact affirmed a net-worth calculation from the time of the hearing. Id. Due to the
practical limitations of computing net worth before the hearing date, in addition to the fact
that presentation of evidence normally ends at the close of the hearing, we find that net worth
should be calculated as near to the time of the hearing as practicality allows. The evidence
must be sufficient to enable the trial court to determine the defendant’s current net worth,
according to generally accepted accounting principles. Miss. Code Ann. § 11-1-65(3)(b)
(Rev. 2014).
26
¶41. We hold that the trial court’s net-worth determination did not contravene Franklin.
Sandoz proved that its most recent quarterly financial information showed that its net worth
on December 31, 2011, was less than $500 million, according to generally accepted
accounting principles. No evidence was before the trial court that Sandoz’s net worth had
exceeded $500 million by the time of the hearing. Mastakas refuted the State’s suggestions
to that effect on cross-examination, and the State presented no expert testimony by a
qualified accountant that, according to generally accepted accounting principles, Sandoz’s
net worth currently exceeded $500 million. The trial court used the most recent quarterly
financial information to determine Sandoz’s net worth. The sole competent evidence of
Sandoz’s current net worth was presented by Sandoz, and the trial court appropriately
credited the evidence and testimony that, under generally accepted accounting principles,
Sandoz’s net worth was less than $500 million.
V. The trial court correctly defined Sandoz’s violation of the CPA as
each publication of a false AWP, rather than each time Sandoz
caused the State to overpay for a drug.
¶42. The State argues that the chancellor erred in calculating the number of Consumer
Protection Act violations by the number of occasions that Sandoz reported an AWP, rather
than the number of occasions Medicaid reimbursed a claim using a reported AWP. The
Chancellor found 2,699 violations and awarded the State $1,000 per violation.15 While
Mississippi does not have a statute or case on point, United States Supreme Court cases
addressing similar penalties available under the federal False Claims Act and State cases
15
The statute permits a penalty of up to $10,000 per violation.
27
addressing similar AWP litigation support the Chancellor’s decision to apply the “per
violation” language of Section 75-24-19(1)(b) to the number of occasions that Sandoz
improperly set an AWP during the relevant damages period. In United States v. Bornstein,
423 U.S. 303, 96 S. Ct. 523, 46 L. Ed. 2d 514 (1976), the United States Supreme Court held
that, under the federal False Claims Act, which contains a similar per-violation penalty as
Mississippi’s CPA, “the focus in each case [must] be upon the specific conduct of the person
from whom the Government seeks to collect the [penalties].” 423 U.S. at 313. There, the
Court concluded that the number of violations were three fraudulent invoices issued by the
defendant general contractor to a subcontractor, rather than the thirty-five subsequent
invoices billed to the federal government from the subcontractor. The Wisconsin Supreme
Court applied a similar approach in AWP litigation, stating that “[t]he number of times
pharmacies were overpaid is merely a consequence of the alleged fraud, not the fraudulent
conduct itself.” Abbott Labs., 816 N.W.2d at 173-74. We affirm the chancellor’s use of the
reported AWPs to calculate the number of violations under CPA. We also find that the trial
court did not abuse its discretion in failing to award the maximum $10,000 per-violation
penalty for the CPA violations.
VI. The trial court did not err in dismissing the State’s statutory claim
under the Medicaid Fraud Control Act.
¶43. The State argues that the trial court erred in dismissing the State’s statutory claim
under the Medicaid Fraud Control Act on the ground that the Act’s civil liability provision
is inapplicable to Sandoz. The MFCA’s civil liability provision states:
28
A health care provider or vendor committing any act or omission in violation
of this article shall be directly liable to the state and shall forfeit and pay to the
state a civil penalty equal to the full amount received, plus an additional civil
penalty of triple the full amount received.
Miss. Code Ann. § 43-13-225(1). We affirm the trial court’s dismissal of the MFCA claims
on the ground that the civil liability provision does not provide for recovery beyond the
amount the defendant has improperly received. The parties do not dispute that Sandoz never
received any direct payment from the State, and under the language of this statute the parties
are too far removed to permit recovery. As the trial court noted in its opinion, in 2006, 2007,
and 2010, the Mississippi Legislature considered bills that would have replaced the “full
amount received” language with language that permits recovery for the “amount of damages
that the State sustains because of the act of that person.” The fact that the Mississippi
Legislature considered such a change demonstrates its recognition that the existing penalty
provision was limited to the amount a provider or vendor received and thus required an
amendment if it was to also cover the amount of the State’s loss. We therefore affirm the trial
court on this issue.
CONCLUSION
¶44. On our deferential standard of review, we affirm the trial court’s finding that Sandoz
committed common law fraud and violated the Mississippi Consumer Protection Act.
Mississippi Medicaid reasonably relied on Sandoz’s fictitious Average Wholesale Prices to
reimburse pharmacies in excess of the Estimated Acquisition Cost of the drugs. We find the
State’s issues on cross-appeal to be without merit. We affirm the trial court’s judgment in all
respects.
29
¶45. ON DIRECT APPEAL: AFFIRMED. ON CROSS-APPEAL: AFFIRMED.
KITCHENS AND KING, JJ., CONCUR. RANDOLPH, P. J., SPECIALLY
CONCURS WITH SEPARATE WRITTEN OPINION. DICKINSON, P.J., DISSENTS
WITH SEPARATE WRITTEN OPINION JOINED BY LAMAR, PIERCE AND
COLEMAN, JJ. LAMAR, J., DISSENTS WITH SEPARATE WRITTEN OPINION
JOINED BY DICKINSON, P.J., PIERCE AND COLEMAN, JJ. WALLER, C.J., NOT
PARTICIPATING.
RANDOLPH, PRESIDING JUSTICE, SPECIALLY CONCURRING:
¶46. I am constrained to specially concur, while maintaining that the issuance of a
judgment of affirmance is the only appropriate disposition of the case sub judice. Today, four
justices agree to affirm the chancellor’s judgment, and four justices disagree. Thus, the
judgment of the chancery court should be affirmed. See Rockett Steel Works v. McIntyre,
15 So. 2d 624 (Miss. 1943) (“Three of the judges of this Court are of the opinion that the
judgment of the court below should be affirmed, and three [are] of the opinion that it should
be reversed; consequently, that judgment must be, and is affirmed.”).
¶47. Courts of our nation recognized early on that competing opinions of an evenly divided
court should be rejected, for they fail to offer any precedential value. I remain perplexed by
this Court’s recent reluctance16 to apply the principle of law announced by no lesser authority
than Chief Justice John Marshall and applied by numerous justices who followed him in the
United States Supreme Court. In Etting v. Bank of United States, 24 U.S. (11 Wheat.) 59,
78, 6 L. Ed. 419 (1826), Chief Justice Marshall wrote, as follows:
No attempt will be made to analyze [the parties’ arguments and cited cases],
or to decide on their application to the case before us, because the Judges are
16
See Cowart v. State, No. 2012-KA-02051-SCT, 2015 WL 110612, at *17 (Miss.
Jan. 8, 2015); Hentz v. State, 152 So. 3d 1139, 1143 (Miss. 2014).
30
divided respecting it. Consequently, the principles of law which have been
argued cannot be settled; but the judgment is affirmed, the court being divided
in opinion upon it.
Etting, 24 U.S. at 78. In Durant v. Essex Co.,74 U.S. (7 Wall.) 107, 19 L. Ed. 154 (1868),
addressing the effect of affirmance by an equally divided court, Justice Field wrote that:
There is nothing in the fact that the judges of this court were divided in
opinion upon the question whether the decree should be reversed or not, and,
therefore, ordered an affirmance of the decree of the court below. The
judgment of affirmance was the judgment of the entire court. The division of
opinion between the judges was the reason for the entry of that judgment; but
the reason is no part of the judgment itself.
Durant, 74 U.S. at 110. The U.S. Supreme Court has further explained that:
it is obvious that that which has been done must stand unless reversed by the
affirmative action of a majority. It has therefore been the invariable practice
to affirm, without opinion, any judgment or decree which is not decided to be
erroneous by a majority of the court sitting in the cause. . . . [A]n affirmance
by an equally divided court is . . . a conclusive determination and adjudication
of the matter adjudged; but the principles of law involved not having been
agreed upon by a majority of the court sitting prevents the case from becoming
an authority for the determination of other cases, either in this or in inferior
courts.
Hertz v. Woodman, 218 U.S. 205, 212-14, 30 S. Ct. 621, 622-23, 54 L. Ed. 1001 (1910).
¶48. Our Court previously has addressed this issue and cited other state supreme court
decisions which conformed to this principle. In Wise v. Valley Bank, 861 So. 2d 1029 (Miss.
2003), this Court held that:
When this Court is evenly divided, it must affirm the judgment of the court
from which the appeal is taken, even if that judgment is from the Court of
Appeals. There is a long standing history in this regard. Other states have
historically done the same. In Tate v. Christy, 339 N.C. 731, 454 S.E.2d 242
(1995), the North Carolina Supreme Court held that the decision of the Court
of Appeals would be left undisturbed where the participating members of the
Supreme Court were evenly divided as to affirmance or reversal. In Pierce v.
31
Pierce, 244 Kan. 246, 767 P.2d 292 (1989), the Kansas Court of Appeals
affirmed the trial court’s holding. On review by the Kansas Supreme Court, the
justices were evenly divided; therefore, the court affirmed the Court of
Appeals, which had affirmed the district court judgment. In Getschow v.
Commonwealth Edison Co., 99 Ill. 2d 528, 77 Ill. Dec. 83, 459 N.E.2d 1332
(1984), the Illinois Supreme Court ruled that where it was evenly divided on
a portion of the judgment, the Appellate Court’s judgment would stand. In
Christensen v. Epley, 287 Or. 539, 601 P.2d 1216 (1979), the Oregon
Supreme Court was evenly divided on one issue of the case, which had been
heard by the Court of Appeals; therefore, the decision by the Court of Appeals
on that one issue was affirmed. In Benson v. First Trust & Savings Bank, 105
Fla. 135, 145 So. 182 (1932), the justices of the Florida Supreme Court on
second rehearing were equally divided; therefore, the Supreme Court’s
judgment on the first rehearing was sustained.
We hold that when this Court is evenly divided, the order or judgment of the
court from which the appeal is taken must be affirmed.
Wise, 861 So. 2d at 1032-33.
¶49. We should dispose of the case with a short opinion or order, similar to the order issued
by this Court in Beecham v. State, 108 So. 3d 394 (Miss. 2012). In Beecham, this Court
entered an En Banc Order, holding that, because the judgment of the Court of Appeals had
not been decided to be erroneous by a majority of the Court, the Court must affirm, without
opinion, the judgment of the Court of Appeals. Id.
¶50. Not one justice in this case, or our predecessors, has articulated a valid reason or
offered contrary authority why this principle should not be followed. Accordingly, I submit
the chancellor’s judgment should be affirmed without opinion, as his opinion and judgment
have not been determined erroneous by a majority of the justices.
DICKINSON, PRESIDING JUSTICE, DISSENTING:
32
¶51. Any shame brought to light by this case does not lie at the feet of Sandoz, Inc., but
rather at the feet of the State itself. Sandoz submits for publication an index called the AWP
(average wholesale price) for its drugs. This index does not—and for at least the past quarter
century, has not intended to—represent the actual price pharmacies pay Sandoz for
pharmaceuticals. The Mississippi Division of Medicaid knows this and has known it for
decades. So has the Mississippi Legislature.
¶52. The price the Division of Medicaid has reimbursed pharmacies for drugs has been set
by that agency and by the Legislature, and that price has been below AWP for at least the
past twenty-five years. As Justice Lamar points out in her dissent—from evidence in the
record—during the very same period of time
the State is claiming Sandoz’s 57 percent markup is so high it constitutes
fraud, the State was specifically told by federal Medicaid officials that generic
AWPs were between 42 and 65 percent higher than actual prices.
¶53. In fact, as Justice Lamar goes on to point out from evidence in the record, the federal
government in 1975 issued new federal medicare regulations and a statement that fully
informed the states that AWP did not represent the prices pharmacies were actually paying
for drugs. And as Justice Lamar further points out from the evidence in the record, “[t]he
State’s own drug-pricing expert admitted that this statement put Mississippi on notice in
1975 that AWPs were not based on actual prices.” (emphasis added).
¶54. I am not attempting to improve on Justice Lamar’s well-written dissent which, in my
view, is exactly correct and with which I fully join. But I do wish to amplify that, unlike her
excellently researched dissent—that is based on evidence in the record—, much of what the
33
State argues—and which is parroted today by Justice Chandler’s opinion—has no support
in the record. In fact, many of the State’s arguments are flatly contradicted by the record. I
echo the Kentucky Court of Appeals which, after thoughtful analysis, rejected that state’s
AWP fraud claims:
Frankly, it is appalling that the Commonwealth had actual knowledge of this
“shell game” method of pricing employed by the drug companies, the
wholesalers, and the pharmacists. However, even more appalling is the fact
that, in spite of that knowledge, it acquiesced, billed accordingly, and now
seeks reimbursement by way of compensatory and punitive damages.17
¶55. There is no fraud here. At oral argument, the State was unable to articulate evidence
in the record supporting the elements of fraud. Indeed, Justice Chandler’s opinion—which,
incidently, is neither a majority nor plurality opinion—makes no attempt whatsoever to
analyze all the elements of fraud. This case represents nothing resembling an application of
the law to the facts. Rather, it is a stark abuse of power by the State that serves to cover the
State’s own failure to act on information in its possession and to negotiate a better deal for
Mississippi taxpayers. For these reasons, I dissent.
LAMAR, PIERCE AND COLEMAN, JJ., JOIN THIS OPINION.
LAMAR, JUSTICE, DISSENTING:
¶56. Because the State wholly failed to prove fraud or a violation of the Consumer
Protection Act, I dissent. But before I can discuss the merits of this decision, I must provide
necessary evidence omitted from Justice Chandler’s opinion.
17
Sandoz Inc. v. Commonwealth ex rel. Conway, 405 S.W.3d 506, 511 (Ky. Ct. App.
2012) (emphasis added).
34
¶57. The State claims that Mississippi’s Division of Medicaid (Medicaid) over-reimbursed
pharmacies for certain generic drugs because manufacturers intentionally published a
price—called the Average Wholesale Price, or AWP—for its drugs that was not an actual
average and was higher than it should have been. Specifically, the State claims that, because
of Sandoz’s inflated AWPs, pharmacies actually were buying Sandoz’s drugs at a price fifty-
seven percent below Medicaid’s reimbursement rate. According to the State, Sandoz
benefitted from this practice by marketing the profit potential it produced—pharmacies could
make money buying Sandoz’s drugs at one price, then could get reimbursed by the State at
a higher rate that was based on the drug’s AWP.
¶58. The State’s case rests on the premise that Sandoz caused to be published AWPs that
were higher than the prices pharmacies actually were paying and that this resulted in the State
reimbursing pharmacies at an inflated price, because the State was entitled to believe—and
did believe—that AWP represented actual average wholesale prices.
¶59. Sandoz does not deny that its AWPs were not actual averages. And there is no dispute
that pharmacies paid an average of fifty-seven percent less than what Medicaid reimbursed
for Sandoz products. But Sandoz presented evidence that the whole pharmaceutical
industry—including Medicaid—has known for decades that AWPs are not based on actual
transaction prices and are meant to be suggested or benchmark prices. Since the 1970s,
federal Medicaid officials have warned all Medicaid-participating states that AWPs are not
actual averages and are not based on or closely related to the prices pharmacies paid, and
that, as a result, they are an unreliable basis for reimbursement. The State previously has
35
expressly acknowledged these warnings, and its own published definition of AWP calls it a
suggested price and not an actual average. Perhaps most significantly, while the State is
claiming Sandoz’s fifty-seven percent markup is so high it constitutes fraud, the State was
specifically told by federal Medicaid officials that generic AWPs were between forty-two and
sixty-five percent higher than actual prices. And Medicaid’s own pharmacy director did a
study that showed reimbursements for generics might have been as much as forty percent
higher than pharmacies’ actual costs.
¶60. The damages period alleged by the State is January 1, 1991, to October 20, 2005.
Medicaid admittedly did not reimburse any claims during this damages period at 100 percent
of AWP. Rather, they were all reimbursed at a discount off AWP depending on what
formula was in place at the time.18 In 1990, Medicaid set EAC at AWP minus ten percent.
Medicaid considered this rate a “compromise” because, in response to pressure from the
Mississippi Pharmacists Association, Medicaid got permission from federal Medicaid
officials to set that rate in lieu of doing an actual survey of drug costs. In other words,
Medicaid chose to set the reimbursement rate based on negotiations with the pharmacists
rather than on actual acquisition costs. In 2002, the Legislature gave itself, rather than
Medicaid, authority to define EAC and changed it from AWP minus ten percent to AWP
minus twelve percent. See 2002 Miss. Laws Ch. 636B § 1. In 2004, the Legislature once
18
This formula—called the Estimated Acquisition Cost, or EAC—was used only
when it yielded the lowest reimbursement rate for generic drugs, which, during this damages
period, was less than half the time. This was true because most of the drugs were reimbursed
at the mandatory Federal Upper Limit, or FUL.
36
again delegated the authority to define EAC back to Medicaid, and on July 1, 2005, Medicaid
set the rate at AWP minus twenty-five percent, which is where it remains today.
What Federal Medicaid Officials Told Mississippi About Using AWP
¶61. The acronym AWP ceased to represent actual averages at some point prior to 1975
(fifteen years before our damages period here). The federal government publicly
acknowledged this and has warned against using AWP as a basis for
reimbursement—without verification of actual transaction prices—ever since. In 1975
federal officials issued a statement that accompanied some newly issued Medicaid final rules.
40 Fed. Reg. 32284 (July 31, 1975). That statement explained that the federal government
had rejected the idea of basing drug reimbursements on AWP. Id. at 32293. The reason
given was that AWP was “not currently determined by surveying drug marketing transactions
(i.e., by determining the actual price a pharmacist pays to a manufacturer or wholesaler for
a particular drug product), and thus published wholesale prices often are not closely related
to the drug prices actually charged to, and paid by, providers.” Id. (emphasis added). The
State’s own drug-pricing expert admitted that this statement put Mississippi on notice in 1975
that AWPs were not based on actual prices.
¶62. In 1984 (seven years before our damages period), federal Medicaid officials again sent
information to states, including Mississippi, regarding drug prices, urging them to “make a
greater effort to determine more closely the price pharmacists pay for drugs rather than using
AWP.” That was based on an attached report from the Health and Human Services (HHS)
Office of the Inspector General (OIG), which had found that “pharmacies do not purchase
37
drugs at the AWP published in the ‘Bluebook,’ ‘Redbook,’ or similar publications. Thus
AWP cannot be the best—or even an adequate—estimate of the prices providers generally
are paying for drugs.”
¶63. The next year, 1985 (six years before our damages period), HHS sent a letter to
Medicaid warning of impending corrective action if Medicaid continued to use a flat AWP
as the basis for reimbursement. Medicaid Director B.F. Simmons wrote back to HHS,
assuring it that “[t]he content of [its] letter was no surprise to us here in Mississippi.” This
warning came on the heels of a regional workgroup that had come up with a preferred
reimbursement methodology that would result in a rate that was about fourteen percent lower
than AWP. Jim Steele, a Medicaid employee at the time, who would later become
Medicaid’s pharmacy director, participated in that regional workgroup, which was designed
to come up with a better reimbursement basis than AWP, since AWP was not related to
transaction prices.
¶64. But Medicaid did not change the reimbursement methodology, so in January 1989
(two years before our damages period), HHS sent another letter to state Medicaid officials
stating that “nondiscounted or unmodified AWP is not acceptable for State use as the basis
for estimated acquisition cost (EAC), absent any compelling evidence to the contrary.
[Medicaid] policy is that there is a preponderance of the evidence that indicates that AWP
significantly overstates the prices that pharmacists are currently paying for drug products.”
(Emphasis added.)
38
¶65. Later in 1989, HHS again wrote to Medicaid, this time specifically outlining the parts
of Mississippi’s state plan that were out of compliance and the steps needed to bring them
into compliance. Among other problems, the letter pointed out that Mississippi used “[n]on-
discounted average wholesale price . . . to establish the EAC for single-source drugs. There
is a preponderance of the evidence which indicates that AWP does not represent the price
generally and currently paid by providers for a drug.” The letter also warned that “an
arbitrarily assigned discount of the AWP is unacceptable also. The discount must be justified
and substantiated by data.”
¶66. A 1997 study by the Office of the Inspector General (OIG) “estimate[d] that, on
average, actual acquisition cost of generic drugs was 42.5 percent below AWP.” The report
of the study was sent to Mississippi and other states. The OIG again reported on actual costs
versus AWP in March 2002, this time finding that the difference had grown to an average
of 65.93 percent for generic drugs.19
19
Without any record support, or sound reason for doing so, the affirming justices give
credence to the State’s argument that it knew there was some inflation in AWPs, thus all this
evidence of what federal officials told it about AWPs being inflated should not weigh against
it. I see nothing in the record below that supports this claim. This lawsuit was pleaded, tried,
and decided on the question of whether AWPs were supposed to mean the actual average of
prices paid, not an inflated average. This is reflected in the State’s original complaint, its
Sandoz-specific complaint, its arguments and the testimony at trial, its damages calculation,
and the trial court’s amended opinion. The only testimony the State introduced regarding
Medicaid’s understanding of AWPs was that of Medicaid Executive Director Helen
Wetherbee, and she plainly testified that “any difference between AWP and the actual
average transaction prices would be inconsistent with [her] view of AWP.” (Emphasis
added.) The trial court’s understanding of the State’s position was “that the State understood
and relied on the fact that Sandoz’ AWPs meant just what the phrase says “Average
Wholesale Prices,” or the average of those prices that a wholesaler received from the sale of
Sandoz’ drugs to pharmacies . . . .” Conversely, the trial record is devoid of any argument or
evidence on what the “expected exaggeration” was or how “far beyond” such expectations
39
Other Pricing Data Available to Medicaid
¶67. In addition to AWPs, Medicaid had other pricing data for Sandoz’s drugs. Most
importantly, for the first six years of the damages period, Sandoz submitted the actual
average prices that Mississippi retail pharmacies were paying for Sandoz’s generic drugs.
These “Average Manufacturer Prices” (AMPs) were defined in a contract between Sandoz
and Medicaid—and by federal statute—as “the average unit price paid to the Manufacturer
for the drug in the States by wholesalers for drugs distributed to the retail pharmacy class of
the trade . . . .” See 42 U.S.C. § 1396r-8(k)(1)(2000). In other words, AMPs represented the
very number the State claims AWPs were supposed to represent. But AMPs were
significantly lower than AWPs. In fact, one document produced by the State in discovery
showed that, sometime after 1991, a Medicaid employee compared some AMPs to some
AWPs and determined that there was an eighty-three percent difference in the two. Yet,
inexplicably, Medicaid continued to use AWPs to set the reimbursement rate.
¶68. Medicaid also had access to the price Sandoz charges wholesalers for its drugs, not
including any applicable discounts, called a Wholesale Acquisition Cost, or WAC. First
DataBank published these WACs alongside AWPs, and there is no dispute that Medicaid had
Sandoz’s AWPs were. This Court “do[es] not consider issues raised for the first time on
appeal,” Jones v. Fluor Daniel Servs. Corp., 959 So. 2d 1044, 1948 (Miss. 2007). As such,
I think it is improper to consider this argument at all. But this information, admittedly
received by the State, informed Medicaid that AWPs were not based on actual prices and
were not closely related to prices. It also informed Medicaid that pharmacies were paying
anywhere from forty-two to sixty-five percent less than AWP during the damages period.
Medicaid’s reimbursement rates were precisely within that range—at fifty-seven percent over
actual prices. Given all this information in the State’s possession, changing its argument on
appeal does not help it much anyway.
40
access to these numbers. And while AWPs generally stayed the same, over time the WACs
declined as increased competition and other factors drove prices down. One of Sandoz’s
experts testified that if AWPs were meant to be actual averages, this growing discrepancy
would mean that, over time, wholesalers were earning bigger and bigger profits from the sale
of generic drugs, while the manufacturers’ profits were staying about the same. But the
State’s drug-pricing expert testified that this was not the case: wholesalers’ profits were
generally in the two-to-three percent range, regardless of AWP.
The State’s Only Medicaid Witness
¶69. Helen Wetherbee was Executive Director of Medicaid from 1990 to 1999. She was
the State’s only witness from Medicaid. Wetherbee testified that Medicaid did receive the
various OIG reports and information from federal Medicaid officials warning against the use
of AWPs in reimbursement formulae, but that they were “[n]ot of immediate value” to
Medicaid.
¶70. Throughout Wetherbee’s directorship, Medicaid published a pharmacy manual that
defined AWP as “the manufacturer’s suggested wholesale unit price to retailers” with no
mention of it being an actual average. Wetherbee acknowledged this fact but claimed it was
just a “very unfortunate choice of words and [did] not comport with her understanding or the
understanding [she] had with her staff.” Instead, she claimed, “everything we did and
everything we understood was based on the actual Average Wholesale Price, not any
suggestion of a price.” But she admitted that neither Sandoz nor any other drug company
told her that AWPs were averages, and that no federal regulation defined the term. Her only
41
basis for believing that AWPs were an actual average was because “the words are what they
are.” Wetherbee also conceded that she could name no one else who ever had said they
believed AWPs to be an actual average.
¶71. Jack Lee was one of the pharmacy directors under Wetherbee while she was director,
and in 1997 he studied drug prices in Mississippi. Lee conducted the study by reviewing
three years of invoices from two Mississippi pharmacies and determined that a more accurate
discount for generic drugs may be AWP minus thirty to forty percent. Wetherbee testified
that Lee “came to [her] with a recommendation that he thought that AWP minus 10 percent
might no longer be applicable in Mississippi. And based on what he had found in the
pharmacies that he had looked at he was thinking we might need to implement a deeper
percent discount.” Wetherbee testified that Medicaid took no action on these results,
however, because of the small sample size and the fact that it “was not a formal study.”
¶72. Wetherbee left Medicaid in 1999. The State presented no other Medicaid witnesses
or anyone who testified as to Medicaid’s understanding of AWP from 1999 to the end of the
damages period in 2005.
¶73. With this evidentiary background in mind, I now turn to the glaring deficiencies in the
State’s case, beginning with the common-law-fraud claim. While Justice Chandler’s opinion
correctly states the nine elements of fraud, it wastes no ink explaining how even one of the
elements was met; much less all nine, and much less by clear and convincing evidence. And
while I would find the State failed to prove several elements, the failure to prove even one
element by clear and convincing evidence is fatal to its claim.
42
¶74. This Court has explained that “[c]lear and convincing evidence is of such a high order
that [even] the overwhelming weight of the evidence falls short of being clear and
convincing.” Kinney v. Catholic Diocese of Biloxi, Inc., 142 So. 3d 407, 418 (Miss. 2014)
(quoting Brothers v. Winstead, 129 So. 3d 906, 915 (Miss. 2014)). Again, failure to prove
even one element of fraud by clear and convincing evidence is fatal to its case, but I will
focus my analysis on four of the elements: Sandoz’s intent to deceive Medicaid; Medicaid’s
ignorance about what AWPs represented; Medicaid’s reliance on AWPs as actual averages;
and Medicaid’s right to rely on Sandoz’s AWPs as actual averages.
No evidence in the record shows that Sandoz intended to deceive Medicaid.
¶75. The State was required to prove by clear and convincing evidence that Sandoz
possessed “an affirmative intent to deceive.” Russell v. S. Nat’l Foods, Inc., 754 So. 2d
1246, 1256 (Miss. 2000). This element “require[s] that the proof must establish an intent to
deceive and this is indispensable in [a fraud] action.” Anderson Dunham, Inc. v. Aiken,
241 Miss. 756, 133 So. 2d 527, 529 (1961) (emphasis added). The trial court concluded that
“Sandoz submitted its AWPs to [First DataBank] with the intent of inducing Medicaid
agencies such as Medicaid to rely on its AWPs in paying pharmacies for Sandoz’[s] drugs.”
But that conclusion fails to reach the disputed aspect of the intent-to-deceive element:
whether Sandoz intended to deceive the State into believing that AWPs were actual averages
rather than suggested prices. The State was required to prove either that Sandoz tricked the
State into believing AWPs were actual averages, or that Sandoz knew the State believed that
and did nothing to correct that belief. See, e.g., State v. Cummings, 203 Miss. 583, 35 So.
43
2d 636, 638 (1948) (explaining that fraud “consists in deception, intentionally practiced to
induce another to part with property or to surrender some legal right . . . ”). But the State
presented no evidence to support either one of these, instead relying on evidence that
“Sandoz was well aware that MS Medicaid’s reimbursement methodology relied on reported
AWPs.”
¶76. Justice Chandler’s opinion follows the trial court in emphasizing First DataBank’s
definition of AWP as “the average price paid by the pharmacy to the wholesaler for a
particular drug.” See Chandler Op. ¶5. But First DataBank’s definitions have no bearing on
whether Sandoz possessed the specific intent to deceive Medicaid into believing AWPs were
actual averages, because there is no evidence anyone at either Sandoz or Medicaid ever saw
those definitions. Wetherbee testified that she had never seen the definitions, nor could she
name anyone at Medicaid who had seen them. She also could not name any other Medicaid
employee who thought that AWPs were actual averages. Nor could she point to any statute,
regulation, publication, definition, or other outside source for her belief that AWPs were
meant to be actual averages.
¶77. And perhaps most significantly, Sandoz points out that it voluntarily submitted its
actual average prices to the State for six years during the damages period, strong evidence
that it did not intend for the State to interpret AWPs as actual averages. There is no dispute
that Sandoz intended for Medicaid to receive and to rely on its AWPs as a basis for
reimbursement formulae; but there is a complete absence of evidence that Sandoz intended
44
Medicaid to believe those were actual averages rather than a suggested price. In fact, the
record amply supports the opposite conclusion.
¶78. Likewise, Sandoz had no reason to believe that Medicaid understood AWPs to be
something other than a suggested price. Medicaid’s publicly available pharmacy provider’s
manual defined AWP as a “suggested price,” not an actual average. Finally, the only
evidence that Medicaid believed AWPs were actual averages was Wetherbee’s personal
belief, a belief not shared, indeed contradicted, by the two pharmacy directors who worked
under Wetherbee while she was director, and at least one of her predecessors as executive
director. This Court has held that “proving fraud is difficult, as it ought to be. Clear and
convincing evidence is required.” Martin v. Winfield, 455 So. 2d 762, 764 (Miss. 1984). As
such, I would find that the State wholly failed to prove intent to deceive.
The State failed to prove that Medicaid was ignorant about what Sandoz’s AWPs were.
¶79. The State was required to prove by clear and convincing evidence that Medicaid was
ignorant of the fact that Sandoz’s AWPs were not actual averages of prices paid. See Allen
v. Mac Tools, Inc., 671 So. 2d 636, 642 (Miss. 1996). Neither Justice Chandler’s opinion nor
the trial court mentions or analyzes this essential element, but I would find that the record
falls far short of supporting a finding that the State proved it by clear and convincing
evidence. For one, Medicaid’s own published definition of AWP confirms it was not
ignorant at all, referring to AWP as a “suggested price,” not an average. And Medicaid’s own
pharmacy director estimated that pharmacies were paying a thirty to forty percent less than
published AWPs. Additionally, it is undisputed in this record that Medicaid received
45
warnings from the federal government about using AWPs for reimbursement for decades.
The OIG reports foreclose any claim that Medicaid was ignorant that published AWPs were
anything other than a suggested price. It was expressly informed that pharmacies were paying
anywhere from 42 to 65 percent less than these published AWPs. Finally, the statement
accompanying the Medicaid final rules, fifteen years prior to the damages period, forecloses
any claim that Medicaid was ignorant of the fact that published AWPs were not based on
actual transaction prices.
The State failed to prove that Medicaid relied on Sandoz’s AWPs as something other than
a suggested price, or that such a reliance would have been reasonable.
¶80. I examine Medicaid’s reliance and the reasonableness thereof together. Here, the
State was required to prove by clear and convincing evidence (1) that Medicaid relied on the
representation that AWPs were averages of the prices pharmacies paid for generic drugs, and
(2) that such reliance was reasonable. Franklin v. Lovitt Equip. Co., 420 So. 2d 1370, 1372
(Miss. 1982). The trial court did not make specific findings on this issue; rather, it summarily
stated that “Medicaid reasonably relied on the information contained within Sandoz’[s]
published AWPs.” That framing of the issue is unhelpful. The State did rely on Sandoz’s
published AWPs, but the issue in this case is whether the State relied on those as actual
averages, and whether such reliance was reasonable.20
20
The State was required to prove this because that is what it pleaded in its complaint
and what it argued before the trial court. On appeal, as mentioned above, its argument is that
Medicaid “was entitled to rely on Sandoz’s reported AWPs as having a reasonable
relationship to a price.” Notwithstanding the unworkable vagueness in that statement, even
if that is what Medicaid did rely on, it does not change the reasonableness inquiry under
these elements.
46
¶81. One key indication that Medicaid did not actually rely on AWPs as literal averages
is its own pharmacy manual, which defined AWPs not as an actual average, but as the
“manufacturer’s suggested price.” Wetherbee admitted that this contradicted what she
claimed to be Medicaid’s understanding of AWP and lamented that it was a mistake that
should have been corrected but never was. Medicaid’s reliance on Sandoz’s AWPs would
show up primarily in how it set its reimbursement rates. It set those rates three times during
this damages period, but the State failed to produce evidence showing that reliance on
Sandoz’s AWPs as an actual price—or anything other than a suggested price—played any
part in setting those rates. In fact, the State failed to produce any testimony or affidavit of
anyone who actually played a part in those rate changes.
¶82. The trial court’s analysis stopped short in that it determined only that the State relied
on Sandoz’s published AWPs, without requiring the State to prove by clear and convincing
evidence that it relied on those AWPs as actual averages, which was the basis of the entire
lawsuit. I would hold that such an incomplete analysis constitutes error.
¶83. And the trial court’s analysis of the second prong—the reasonableness of Medicaid’s
reliance—is affected by the same error. It ruled that it was reasonable for the State to rely
on published AWPs but made no finding on whether it was reasonable to rely on those AWPs
as actual averages. This analysis relieved the State of its burden of proof and led to an
erroneous finding. The State claimed it was reasonable for it to rely on the published AWPs
as actual averages of prices paid, because the publisher was the “gold standard” for
pharmaceutical pricing. It also pointed out that when the Omnibus Budget Reconciliation
47
Act (OBRA ’90) increased the number of covered drugs from roughly 1,800 to nearly
65,000, the State was forced to automate and computerize the process. These facts do
support the reasonableness of relying on a third-party publisher of drug-pricing information,
but they provide no support for the reasonableness of relying on AWPs as something other
than what Medicaid had for decades been told about AWPs.
¶84. On the other hand, evidence in the record supports Sandoz’s argument that, from the
top of the federal government to Medicaid’s executive director, and from 1975 to 2005, AWP
universally was understood and treated as something other than an actual average of prices
paid. Examples include the following:
In 1985, the federal government warned Mississippi that AWP was not closely related
to actual prices and threatened to withdraw its matching funds, which accounted for
approximately eighty percent of Mississippi’s Medicaid budget, if Medicaid continued
to use a nondiscounted AWP for reimbursement.
During the damages period, both President Clinton and Health and Human Services
Secretary Donna Shalala publicly referred to AWP as a “sticker price.”
For the first six years of the damages period, Sandoz voluntarily submitted to
Mississippi the prices pharmacies actually paid for its generic drugs. Those numbers
showed a significant difference between actual prices and AWP.
The State itself conducted a study in 1989 that indicated pharmacies did not pay
AWP.
The State’s Medicaid witness and former Medicaid director told other states’
Medicaid directors that Medicaid was “paying AWP - 10% and that is where the
[pharmacies] are making their money.”
The State’s drug-pricing expert acknowledged that federal Medicaid officials had, in
1975, “put Mississippi on notice” that AWPs were not closely related to actual prices.
48
¶85. The State did not dispute any of these facts, and in my opinion, the trial court erred
when it dismissed all this as “scattered and sporadic knowledge of government employees”
and as “certain State employees testif[ying] that they believed that Sandoz’[s] published
AWPs were not true average wholesale prices received by Sandoz for its drugs . . . .” In sum,
I echo the Alabama Supreme Court’s conclusion, regarding this same AWP issue: “The idea
of a person knowing a representation to be false and at the same time ‘relying’ thereon is a
contradiction in terms.” Sandoz, Inc. v. State, 100 So. 3d 514, 532 (Ala. 2012).
The State failed to prove that Sandoz’s AWPs proximately caused Medicaid any injury.
¶86. This Court has held that “recovery is not permitted if the proximate cause of the
monetary loss is other than the fraud alleged.” Russell v. S. Nat’l Foods, Inc., 754 So. 2d
1246, 1256 (Miss. 2000). Thus, the State was required to prove that if Sandoz had reported
AWPs that were actual averages, Medicaid would have reimbursed pharmacies at those
averages. The trial court made no specific findings on this element, only ruling that “Sandoz’
[s] conduct caused Mississippi to overpay for its prescription drugs and as a result,
Mississippi sustained proximate injury and damages as a result of Sandoz reporting false and
inflated AWPs.” But the record reveals no testimony from the State’s witnesses, or other
evidence presented at trial, asserting that the State would have reimbursed any differently had
it known that AWPs were not actual averages.
¶87. The State does not even attempt to rebut this argument but simply points to the
damages calculation to show that, because the State used AWP, it paid more than the average
price. But this approach skips the causation inquiry and moves right to damages, bypassing
49
the disputed issue in the case. This Court has held that “[f]raud is never to be presumed or
inferred, but must be proven by clear and convincing evidence.” Boling v. A-1 Detective &
Patrol Serv., Inc., 659 So. 2d 586, 590 (Miss. 1995) (citing Nichols v. Tri-State Brick &
Tile, Co., 608 So. 2d 324, 330 (Miss. 1992)).
¶88. The State neither presented at trial nor argued on appeal any evidence from the record
that indicates Medicaid would have paid a different amount if it had known that AWPs were
something other than actual average prices. The fact is that, during the damages period,
Medicaid never reimbursed at AWP; it always discounted that rate. It received multiple
reports that AWPs were not actual averages and, instead of making any effort to ascertain
what those averages were, Medicaid made a policy decision to find a number that got federal
approval and encouraged pharmacy participation, which is what it was required to do.
Moreover, during the decade since Medicaid allegedly became aware of Sandoz’s
“fraudulent” AWPs, it has not changed the reimbursement formula. I would find that the trial
court erred when it determined that State met its burden to prove causation by clear and
convincing evidence.
The State failed to prove that Sandoz violated the Consumer Protection Act.
¶89. As for the CPA claims, I would find that the State failed to prove that Medicaid was
deceived because Medicaid was repeatedly told that AWPs are not actual average wholesale
prices. Justice Chandler’s opinion ignores this and declares Sandoz’s AWPs to be “false
information,” and therefore deceptive. One cannot help but wonder whether the affirming
justices would find the term “over-the-counter drugs” deceptive if applied to a particular drug
50
that never traveled over an actual counter? What about the “world wide web,” which is
neither world-wide nor an actual web? The rule this Court announces today makes those
terms false and therefore deceptive and a violation of our state’s consumer-protection laws.
Chandler’s Op. ¶ 21. I find that silly.
¶90. Instead, I would follow the Legislature’s admonition and look to federal caselaw for
guidance, and therefore use the test for deceptiveness the Federal Trade Commission has
established: (1) “There must be a representation, omission, or practice that is likely to
mislead the consumer”; (2) “The act or practice must be considered from the perspective of
the reasonable consumer”; and (3) “the representation, omission or practice must be
material.” FTC Policy Statement on Deception (Oct. 14, 1983) (appended to In re Cliffdale
Assocs., Inc., 103 F.T.C. 110, 1984 WL 565319, at *37 (Mar. 23, 1984).
¶91. As to the first element, the Commission in Cliffdale, explained that “[t]he test is
whether the consumer’s interpretation or reaction is reasonable.” Id. at 46. “When
representations or sales practices are targeted to a specific audience, the Commission
determines the effect of the practice on a reasonable member of that group.” Id. For
example, “a practice or representation directed to a well-educated group, such as prescription
drug advertisement to doctors, would be judged in light of the knowledge and sophistication
of that group.” Id.
¶92. Conversely, unreasonable interpretations do not give rise to claims of deception:
Some people, because of ignorance or incomprehension, may be misled by
even a scrupulously honest claim. Perhaps a few misguided souls believe, for
example, that all “Danish pastry” is made in Denmark. Is it therefore an
actionable deception to advertise “Danish pastry” when it is made in this
51
country? Of course not. A representation does not become “false and
deceptive” merely because it will be unreasonably misunderstood by an
insignificant and unrepresentative segment of the class of persons to whom the
representation is addressed.
Id. at 47 (quoting In re Heinz W. Kirchner, 63 F.T.C. 1282, 1290 (1963)) (emphasis added).
¶93. By ignoring the “target-audience” requirement embodied in the first two prongs of the
correct test, I would find that the trial court reached a wrong result when it determined
Sandoz’s published AWPs were deceptive. As far as this record shows, only one member
of the target audience, Wetherbee, claimed to believe AWPs had any definition that would
render them deceptive. There is no record evidence that any other person in her organization,
the pharmaceutical industry, or elsewhere in government shared that understanding. On the
contrary, the record establishes that individuals from all corners of the industry (including
federal and state Medicaid officials, drug-company officials, First DataBank officials, other
publishers of drug-pricing data, and the OIG) had the opposite understanding. As such, I
would find that the chancellor’s finding that Sandoz’s practices were deceptive was
manifestly against the weight of the evidence.
¶94. As for the unfairness claim, I would find that publishing AWPs that were consistent
with everything Medicaid was told, publicly and privately acknowledged, and even published
about AWPs is not likely misleading. Moreover, since Medicaid has known since the 1970s
that AWPs are not based on actual prices, any injury caused by Medicaid relying on AWP
as a number based on actual prices was reasonably avoidable, simply by heeding the federal
government’s thirty years of warnings.
52
¶95. At least as far back as 1975, the Mississippi Division of Medicaid was repeatedly
informed—and warned—that AWPs were not based on transaction prices and were not an
adequate basis for their reimbursement formulae. This is undisputed in the record. The
State’s reimbursement decisions were based on policy considerations, political neogtiations
with the Pharmacists’ Association, and requirements of the federal Medicaid program. In so
doing, Medicaid succeeded in maximizing pharmacy participation and obtaining federal
approval and matching dollars every year of this damages period. Although the State’s
reimbursement formula was based on AWPs provided by Sandoz, the State’s claimed
reliance on those numbers as actual averages or even exaggerated averages was not
reasonable. I would reverse and render this judgment in favor of Sandoz.
DICKINSON, P.J., PIERCE AND COLEMAN, JJ., JOIN THIS OPINION.
53