Ranbaxy Laboratories Inc. v. First Databank, Inc.

PARKER, Circuit Judge: Plaintiff-Appellant, the pharmaceutical company Ranbaxy Laboratories Inc., seeks money damages and injunctive relief for alleged misrepresentations made by Defendant-Appellee First Databank, Inc. (“FDB”), a company that publishes a drug information database for use by pharmacies across the United States. Ranbaxy alleges that FDB’s database, MedKnow-ledge, falsely represents that Ranbaxy’s acne drug Absorica is non-unique. After expedited discovery on the issue of falsity, the district court granted summary judgment in favor of FDB, concluding that FDB did not publish any false statements about Absorica. Because we agree that Ranbaxy has not raised a genuine issue of material fact with regard to falsity, we affirm the order and judgment of the district court. I. BACKGROUND A Absorica Ranbaxy is the manufacturer of Absori-ca, an Isotretinoin-based product used to treat serious acne and other skin diseases. Although Absorica shares many features with other generic acne treatments, it is unique in that it is effective even if taken without meals (in a “fasted state”). The FDA issues a publication called the “Orange Book,” which is used by pharmacists in many states to help identify which drugs are interchangeable with other drugs. The Orange Book provides a wide range of information about drugs approved by the FDA, but only two metrics are relevant here: pharmaceutical equivalence and therapeutic equivalence. Two drugs are pharmaceutical equivalents if “they contain the same active ingredient(s), are of the same dosage form, route of administration and are identical in strength or concentration.” App. 74-8 at vi-vii. The Orange Book designates Absorica as phar-maceutically equivalent to several other *1336 Isotretinoin-based acne medications. By-contrast, two drugs are therapeutic equivalents if “they are pharmaceutical equivalents and if they can be expected to have the same clinical effect and safety profile when administered to patients under the conditions specified in the labeling.” App. 74-8 at vii. Because of Absoriea’s unique effectiveness when taken in a fasted state, the Orange Book has given Absorica a “BX” rating, which indicates that no drugs are therapeutically equivalent to Absorica. The Orange Book is used in many states as the authoritative source for determining whether a pharmacist may substitute a prescribed drug with a cheaper generic version. In “Orange Book states,” pharmacists may only substitute a drug for another if the two drugs are designated by the Orange Book as being therapeutic equivalents. However, in “non-Orange Book states,” pharmacists are not required to consult the Orange Book (though they may choose to do so), and instead make substitution decisions by relying on their own professional judgment and the information provided by their companies’ software programs. B. MedKnowledge Database FDB publishes the MedKnowledge database, which is a collection of information about various drugs for use by pharmacies when they fill prescriptions. MedKnow-ledge is á raw data file — it is not organized in a way that is meaningful or useful to pharmacists at local drug stores. Instead, FDB sells subscriptions to the MedKnow-ledge database to customers who then develop software that sorts and organizes the raw data into a display format usable by pharmacists. MedKnowledge provides thousands of fields for each drug, with each field populated with a coded piece of information. For example, in one field that relates to the Orange Book’s therapeutic equivalence designation, Absorica is marked as “BX,” indicating it has no therapeutic equivalent. FDB’s customers choose which data to display to pharmacists and how to display it. FDB has no control over how the information is displayed to the pharmacists issuing prescriptions. Because MedKnowledge is merely a collection of thousands of coded data fields, FDB provides its customers with access to the MedKnowledge user documentation. Reference to the documentation is necessary to understand the various fields of coded data, many of whose meaning is not self-evident. FDB customers can retrieve the documentation, which is nearly 4,000 pages long, by either requesting a CD of the documentation or downloading it from FDB’s website. Ranbaxy’s complaint concerns two pieces of data published in the MedKnow-ledge database. First, each drug is assigned a 5-digit Clinical Formulation ID. Several drugs may be assigned the same Clinical Formulation ID if they have the same active ingredients, route, dosage form, and strength (the same factors considered in determining pharmaceutical equivalence). The MedKnowledge documentation indicates that “[although the Clinical Formulation ID ... can be used to develop a list of candidates for substitution, these candidates are only pharmaceu-tically equivalent; it is not sufficient to determine therapeutic substitutability.” App. 74-2 at 5. Elsewhere, however, the documentation indicates that FDB may assign a unique Clinical Formulation ID to a drug with pharmaceutical equivalents if the drug has a clinically unique dosage form. that is not accounted for in the Orange Book. For example, while the Orange Book groups all drugs taken as a tablet under a single dosage form, FDB has twenty-five different dosage forms for tablets. Thus, two drugs identified as phar-maceutically equivalent by the Orange *1337 Book might still have different Clinical Formulation IDs if their dosage forms differ in a way recognized by FDB, but not the FDA. Although FDB employs substantially more dosage forms than the FDA, it does not have a dosage form relating to whether a drug may be taken in a fasted state. Absoriea thus shares its Clinical Formulation ID with several other Isotre-tinoin-based acne medications. Ranbaxy protests that because the Med-Knowledge documentation indicates that “new dosage forms are added when the clinical uniqueness of a novel dosage form has been established,” App. 74-2 at 3, the assignment of a non-unique Clinical Formulation ID to Absoriea falsely represents that Absoriea does not have a clinically unique dosage form, thereby misleading pharmacists into believing that Absoriea is substitutable with other acne medications. Ranbaxy specifically points to testimony from FDB’s corporate representative that “differences in absorption when a product is taken in the fed or fasted state” may be a sign of clinical uniqueness. App. 74-11 at 27. The second piece of data challenged by Ranbaxy is Absorica’s designation in the Multi-Source/Single Source Indicator (NDCGI1) field as a “multi-source” drug. The MedKnowledge documentation explains that this field “specifies whether a product’s clinical formulation (i.e., its particular active ingredient, dosage form, route of administration and strength) is only available from a single labeler [ (single source) ] or from multiple labelers [(multi-source)].” App. 74-2 at 6. The documentation goes on to state that “[pjroducts that have the same clinical formulation are not necessarily therapeutically equivalent.” App. 74-2 at 6. But Ranbaxy’s expert testified that, contrary to the definition offered in the M'edKnow-ledge documentation, “multi-source” is a term of art in the pharmaceutical industry used to indicate that a drug has a therapeutic equivalent. Thus, a pharmacist who sees Absoriea designated as “multi-source” may erroneously conclude that Absoriea has therapeutic equivalents that may be offered as cheaper substitutes. Ranbaxy admits that, in Orange Book states, there is no risk of confusion because pharmacists there are required to consult the Orange Book code before offering a substitute for a drug, and Absorica’s BX rating, accurately notated in the Med-Knowledge database, indicates that it has no therapeutic equivalent. Ranbaxy contends, however, that in non-Orange Book states, a pharmacist might not consult the Orange Book code for Absoriea,, and instead will see its non-unique Clinical Formulation ID or its multi-source designation and wrongly conclude that other generic acne drugs may be safely substituted for Absoriea. C. Procedural History After Ranbaxy brought this action, FDB moved to dismiss the complaint, transfer venue, and strike the complaint pursuant to California’s anti-SLAPP statute, which allows courts to dismiss actions seeking to restrain speech unless the plaintiff demonstrates a probability of success on the merits. The district court denied all three motions. The parties then agreed on an expedited discovery schedule limited to the issue of falsity. Following completion of limited discovery, FDB moved for summary judgment on the ground that there was no genuine issue of material fact as to falsity. The district court granted the motion, reasoning that “[n]o reasonable reader of MedKnowledge’s data would consider these statements ‘false’ or ‘reasonably capable of a defamatory interpretation.’ ” Ranbaxy Labs. Inc. v. First Databank, Inc., No. 3:13-cv-859, 2015 WL 3618429, at *12 (M.D. Fla. June 9, 2015). The court *1338 agreed with FDB that “this is just a disagreement between Ranbaxy and FDB about the proper characterization and placement of Absorica in the MedKnow-ledge database. But such a disagreement is not the stuff of a trade libel claim.” Id. Ranbaxy appealed. II. STANDARD OF REVIEW We review a district court’s ruling on summary judgment de novo. Skrtich v. Thornton, 280 F.3d 1295, 1299 (11th Cir. 2002). A court may grant summary judgment only if it determines that “there is no genuine dispute as to any material fact and the movant is entitled to judgment as a matter of law.” Fed. R. Civ. P. 56(a). In reviewing the record, we must “construe the facts and draw all inferences in the light most favorable to the nonmoving party.” Mathews v. Crosby, 480 F.3d 1265, 1269 (11th Cir. 2007). III. DISCUSSION A. Jurisdiction Before we examine the merits of this case, we must first consider whether we have subject-matter jurisdiction. This case comes before us as an appeal from a final decision of the District Court for the Middle District of Florida. See 28 U.S.C. § 1291. The district court exercised diversity jurisdiction pursuant to 28 U.S.C. § 1332(a)(1). A district court may exercise diversity jurisdiction only if there is complete diversity between the parties, that is, no two adverse parties are citizens of the same state. See Owen Equip. & Erection Co. v. Kroger, 437 U.S. 365, 373, 98 S.Ct. 2396, 57 L.Ed.2d 274 (1978). A corporation is a citizen of any state in which it is incorporated and of the state “where it has its principal place of business.” 28 U.S.C. § 1332(c)(1). The parties agree that FDB is a Missouri corporation with its principal place of business in California, but they disagree as to whether Ranbaxy, a Delaware corporation, has its principal place of business in Florida or New Jersey. However, “a party need not always prove the exact location of a corporation’s principal place of business.... If it can be shown, for example, that the company’s principal place of business is one of two states, and the opposing party is not a citizen of either of them, subject matter jurisdiction will be upheld.” 13F Charles Alan Wright et al., Federal Practice & Procedure § 3625 (3d ed. April 2016 update). Though we have riot explicitly adopted this principle, we have acknowledged its logic in other circumstances, see Cabalceta v. Standard Fruit Co., 883 F.2d 1553, 1560 (11th Cir. 1989) (discussing, without criticizing, decision of district court to exercise jurisdiction without determining corporation’s place of business), and we agree with the parties that its application is appropriate here. We are therefore satisfied that the district court had jurisdiction over this case. B. Falsity We now turn to the question of whether Ranbaxy has raised a genuine issue of material fact with regard to the alleged falsity of FDB’s statements in the Med-Knowledge database. We conclude that it has not. The parties disagree as to whether Florida, California, or New Jersey law controls this dispute. The district court concluded that Florida or New Jersey law likely applies, but that there was no conflict between the two. We agree with the district court that the question of which state’s law applies is immaterial. Ranbaxy has brought claims for trade libel and tortious interference with business relations. Under both Florida and New Jersey law, FDB is only liable for those claims if it published false information. Compare Border Collie Rescue, Inc. *1339 v. Ryan, 418 F.Supp.2d 1330, 1348 (M.D. Fla. 2006) (trade libel), and Cherestal v. Sears Roebuck & Co., No. 6:12-cv-1681, 2014 WL 644727, at *4 (M.D. Fla. Feb. 19, 2014) (tortious interference), with Arista Records, Inc. v. Flea World, Inc., 356 F.Supp.2d 411, 427-28 (D.N.J. 2005) (trade libel), and E. Penn Sanitation, Inc. v. Grinnell Haulers, Inc., 294 N.J.Super. 158, 682 A.2d 1207, 1218 (N.J. Super. Ct. App. Div. 1996) (tortious interference). The operative question is thus whether FDB’s published statements regarding Absorica are false. In determining falsity, we analyze how a “reasonable reader” would understand the disputed material. Dunn v. Air Line Pilots Ass’n, 193 F.3d 1185, 1193 (11th Cir. 1999) (citing Masson v. New Yorker Magazine, Inc., 501 U.S. 496, 111 S.Ct. 2419, 115 L.Ed.2d 447 (1991); Nat’l Ass’n of Letter Carriers v. Austin, 418 U.S. 264, 94 S.Ct. 2770, 41 L.Ed.2d 745 (1974)). We consider the context of the statements and the commonly understood meaning of terms. Id. at 1193. The plaintiff in a trade libel case bears the burden of proving that the statements in question are false. See Bothmann v. Harrington, 458 So.2d 1163, 1168 (Fla. Dist. Ct. App. 1984). We now turn to the two claims made by Ranbaxy: (1) that FDB’s assignment of a non-unique Clinical Formulation ID to Absorica is false, and (2) that FDB’s designation of Absorica as “multi-source” is false. 1 1. Clinical Formulation ID Ranbaxy argues on appeal that the assignment of a non-unique Clinical Formulation ID to Absorica falsely represents that Absorica is not clinically unique, thereby misleading pharmacists into believing they may safely substitute generic acne medications for Absorica. At the outset, FDB correctly notes that this is not the theory Ranbaxy pursued in its complaint. There, Ranbaxy alleged that “[b]y assigning the same [Clinical Formulation ID] to Absorica and all other Isotretinoin-based products ..., FDB falsely and incorrectly indicates to FDB Subscribers that Absorica is therapeutically equivalent to, and may be safely substituted for, other branded or generic Isotretinoin-based products.” App. 1 at 7. It was only after discovery that Ranbaxy advanced the theory that FDB falsely represented the clinical uniqueness of Absorica. In any event, however, the district court was correct in concluding that there is no genuine issue of material fact because no reasonable reader would understand Abso-'rica’s Clinical Formulation ID to mean that Absorica had therapeutic equivalents or that it could be substituted for other drugs. *1340 Ranbaxy first argues that because the MedKnowledge documentation indicates that unique Clinical Formulation IDs may be assigned when a drug’s dosage form is “clinically unique,” Absorica should have its own Clinical Formulation ID. Ranbaxy points to testimony from MedKnowledge’s corporate representative explaining that clinical uniqueness may include “differences in absorption when a product is taken in the fed or fasted state in clinical trials,” App. 74-11 at 27..But there is no evidence that FDB has ever assigned a unique Clinical Formulation ID merely because a drug may be taken in a fasted state. 2 And despite an extensive list of dosage forms, nothing in the MedKnow-ledge documentation, suggests that such a metric is relevant to FDB’s determination of clinical uniqueness. That an FDB corporate representative admitted that a drug’s ability to be taken in a fasted state may be a sign of clinical uniqueness has no impact on what a reasonable reader would glean from the database and the accompanying documentation. The MedKnowledge documentation belies any claim that the assignment of a non-unique Clinical Formulation ID to Absorica is misleading to a reasonable reader: the documentation makes clear that the Clinical Formulation ID “is not sufficient to determine therapeutic substitutability,” and nothing in the documentation suggests that a drug’s effectiveness in a fasted state is a metric that warrants a new dosage form. App. 74-2 at 5. Ranbaxy protests that the MedKnow-ledge documentation is so lengthy and cumbersome that no representative from FDB could even say they read the entire manual, and there is no evidence that any customers have actually done so. Ran-baxy’s insistent reference to the size of the documentation is of no help. The user documentation is not a novel to be read cover-to-cover; it is a reference manual, designed to be consulted and searched as needed. The very passages cited by the parties in this litigation can be found simply by referencing the Table of Contents, Index, or other search function. And as explained above, the MedKnowledge documentation, which is made available to all ■MedKnowledge subscribers, is necessary to understand the data provided in the coded fields. It strains credulity to suggest that FDB’s customers, responsible for designing software to make prescription decisions for ailing patients, would simply neglect to consult the authoritative guide explaining the various data fields. Tellingly, Ranbaxy’s argument regarding clinical uniqueness makes sense in the first place only if we assume that customers read the portion of the documentation explaining that clinical uniqueness may be a basis for developing new dosage forms. Nor is it appropriate to call the detailed explanations in the MedKnowledge documentation “disclaimers.” Unlike in the cases cited by- Ranbaxy, this is not a situation in which FDB has made false statements and has attempted to insulate itself from liability by disclaiming responsibility for their accuracy. See Off Lease Only, Inc. v. Carfax, Inc., No. 12-80193-cv, 2012 WL 1966372, at *3 (S.D. Fla. May 31, 2012) (defendant provided disclaimer that “in no event [are] the [reports] warranted as being error free” (alterations' in original)); Harcrow v. Struhar, 236 Ga.App. 403,511 S.E.2d 545, 546 (1999) (defendant implied that plaintiff was guilty of a crime, but added that “I’m not saying that they are responsible for this atrocious act”). 3 *1341 FDB does not disclaim responsibility for the accuracy of its data in the documentation; it provides explanations for each of the coded fields so that its customers can translate those fields into usable data for pharmacists. Ranbaxy next points to two publications outside the MedKnowledge database that it claims give rise to an issue of fact. Like the district court, we are uncertain whether these separate publications are germane to our analysis of the falsity of the Med-Knowledge database. Nevertheless, we conclude that these publications do not create a genuine issue of material fact. Ranbaxy cites language in a series of “MEDITE CH Customer Connection” newsletters, sent to a subset of FDB’s customers, explaining that “because FDB’s Clinical Form ID classification groups identical products under a shared numerical value (ID), users are able to easily identify a replacement [National Drug Code] for a pharmaceutically substitutable product.” App. 74-24 at 6. Though Ran-baxy is correct that these newsletters describe Clinical Formulation IDs as identifying “identical products,” it ignores the rest of the sentence, which plainly states that the codes merely identify “pharma-ceutically substitutable produces].” And a table produced just one page earlier-in the same document makes clear that Clinical Formulation IDs group together products with the same “Ingredients, Strengths, Dosage Forms, and Routes.” App. 74-24 at 5. Ranbaxy is not permitted to make its case by taking terms out of context and ignoring the plain meaning of the immediate context. A reasonable reader does not read terms in isolation, but puts them in the context in which they were published. See, e.g., Fid. Warranty Servs., Inc. v. Firstate Ins. Holdings, Inc., 74 So.3d 506, 515 (Fla. Dist. Ct. App. 2011); Ward v. Zelikovsky, 136 N.J. 516, 643 A.2d 972, 980 (1994). Ranbaxy also cites to a “Monthly Interest” newsletter in which FDB discussed, as an example of clinical uniqueness, two drugs, one that needed to be taken in the evening with. dinner and one that needed to be taken in the morning without regard to meals. Accepting, as we must at this stage, that a reasonable reader would interpret this newsletter to mean that FDB assigns a unique Clinical Formulation ID to drugs that can be taken in a fasted state, there is no basis upon which to assume that a reasonable reader would ignore the numerous explanations that “[t]he purpose'of the [Clinical Formulation ID] is to allow grouping of pharmaceutically equivalent products,” and that “[a] good rule of thumb to follow is to use the [Clinical Formulation ID] plus the Orange Book code to identify possible generic equivalents.” App. 74-25 at 4-5. These passages inform readers that the fact that two drugs share a Clinical Formulation ID is only sufficient to show pharmaceutical equivalence, and that substitution decisions should be made by consulting the Orange Book. No reasonable reader would understand that assignment of a non-unique Clinical Formulation ID indicated therapeutic equivalence or substitutability. Again, Ranbaxy seeks to remove small portions of text from their plain context; such a strategy does not create a genuine issue of material fact. The last piece of evidence relied upon by Ranbaxy is a set of FDB customer inquiries. These inquiries questioned why Abso-rica shares a generic code (its Clinical Formulation ID) with other Isotretinoin-based medications, even though Absorica is not generally substitutable with other *1342 drugs. Although we have acknowledged that in the context of a Lanham Act claim, “evidence of actual confusion” is the “best evidence of likelihood of confusion,” Amstar Corp. v. Domino’s Pizza, Inc., 615 F.2d 252, 263 (5th Cir. 1980), we are assessing falsity, not likelihood of confusion. And the inquiries are not evidence of actual confusion, but are instead targeted questions about how FDB organizes its data. Moreover, deferring to the judgment of actual customers would require us to ignore the Supreme Court’s direction that falsity is viewed from a reasonable reader’s perspective — the inquiry is an objective, not a subjective, one. In the face of the plain language of the MedKnowledge documentation, which provides clear and ample explanation of how Clinical Formulation IDs are generated and used, five isolated instances of customers inquiring about the Clinical Formulation ID are insufficient to raise a genuine issue of material fact. The MedKnowledge documentation, which is necessary to understanding the vast fields of data provided by FDB, dispels any notion that FDB has published false information about Absorica by assigning it a non-unique Clinical Formulation ID. There is nothing in the Med-Knowledge database or the accompanying documentation that would lead a reasonable reader to believe that every drug that may be taken in a fasted state is assigned a unique Clinical Formulation ID. 2. Multi-Source Designation Ranbaxy’s second ground for liability is that the MedKnowledge database falsely represents that Absorica is a multi-source drug despite the fact that, according to Ranbaxy’s expert, “multi-source” is a pharmaceutical term of art understood in the industry to mean that a drug has therapeutic equivalents. Although Ran-baxy’s expert is unable to cite any treatise, journal, or other authority for this contention, we must assume at summary judgment that his characterization of the term, is accurate. However, we need not ignore the plain reality that “multi-source” is susceptible to multiple meanings, as evidenced by the Orange Book’s use of the term, which is consistent with the definition employed by FDB in the MedKnowledge documentation. 4 Ranbaxy argues that where a statement is susceptible to multiple meanings, a jury or other finder of fact must decide whether the statement was understood in a defamatory sense. See Smith v. Cuban Am. Nat’l Found., 731 So.2d 702, 705 (Fla. Dist. Ct. App. 1999). But that is not the case where, as here, the publisher has offered clear and unambiguous guidance as to how the term should be understood. FDB did not leave interpretation of the term “multi-source” to the discretion of the reader, but rather provided a detailed explanation of how the term is employed in the MedKnowledge database. Ranbaxy protests again that mere “disclaimers” do not absolve FDB from liability. As set forth above, the explanations in the MedKnowledge documentation are not disclaimers; they are guides for understanding the fields of data that FDB publishes and are a necessary reference for all of FDB’s customers. In particular, the data field indicating whether a drug is multi-source or single source contains only a “1” or a “2”; users have to reference the documentation to understand that “1” corresponds to multi-source and “2” corresponds to single source. 5 *1343 Moreover, not only does FDB indicate in a separate field that Absorica has no therapeutic equivalents, it also indicates in another field that Absorica is single source pursuant to the definition suggested by Ranbaxy’s expert. Any reasonable reader viewing the database in context would understand that multi-source and single source are susceptible to different interpretations, and that reference to the user documentation was necessary to understand the meaning employed by each field. IV. CONCLUSION There is little dispute that the Med-Knowledge documentation directly undercuts each of Ranbaxy’s claims. We are not persuaded that the sheer volume of the documentation undermines FDB’s reliance on it. Because FDB provides ample explanation of the information and terms in its database, no reasonable reader would conclude that Absorica was therapeutically equivalent to or substitutable for other drugs. Accordingly, we AFFIRM the order and judgment of the district court. Footnotes [1] Both parties reference a case in the Northern District of California, Schering Corp. v. First Databank Inc., wherein the plaintiff made similar allegations against FDB to those made here. FDB points out that the plaintiff in that case moved for a preliminary injunction, but the court denied the motion, finding that the plaintiff had not established a likelihood of success on the merits because "[n]othing in First DataBank's ... database actually states that the ... products are therapeutically equivalent.” Schering Corp. v. First Databank Inc., No. C07-01142, 2007 WL 1068206, at *5 (N.D. Cal. Apr. 10, 2007). But Ranbaxy points out that just days later, the same court denied FDB’s motion to strike the complaint under California's anti-SLAPP statute because, among other reasons, the plaintiff had made "a prima facie showing of facts to sustain a favorable judgment." Schering Corp. v. First Databank Inc., No. C 07-01142, 2007 WL 1176627, at *9 (N.D. Cal. Apr. 20, 2007). Whatever the meaning of these apparently contradictory statements, we note only that these unpublished decisions, rendered by a district court in another circuit, offer limited persuasive value, as the court there was operating under a different set of facts and the parties there had not yet engaged in discovery. [2] Even Ranbaxy's expert seemed unconvinced by this argument: “Q: And is it your testimony that FDB is required to create a new dosage form for Absorica? ... A: I — I can’t— I can’t — that’s up to them.” App. 74-13 at 38 (deposition testimony). [3] A third case cited by Ranbaxy, Machleder v. Diaz, 538 F.Supp. 1364 (S.D.N.Y. 1982), is *1341 inapposite. There, the “disclaimer” ignored by the court was a statement made by the plaintiff, explaining that he was not responsible for the acts improperly attributed to him by the defendant. 538 F.Supp. at 1372-73. [4] Ranbaxy's expert admitted that "the FDA did not consider Absorica a single-source product, at least in how they used the terms.” App: 74-13 at 31. [5] Ranbaxy suggests that users could learn the meaning of the MedKnowledge codes orally from more experienced users without referencing the user documentation. Again, in the absence of any evidence to the contrary, we *1343 find it implausible that FDB’s customers, who are responsible for developing software for dispensing medication to patients, would eschew the clear explanations provided in the MedKnowledge documentation in favor of word-of-mouth explanations by other users.