Every time we think about addressing ghostwriting as a recurrent plaintiff-side jury distraction in drug/device product liability litigation, we get earwormed by “Ghost Riders in the Sky.”  Whether one prefers the Johnny Cash or Outlaws version of the song – or one of who knows how many other covers of the song (originally written by Stan Jones in 1948), it’s hard to stop thinking about it once you start.

The most inveterate ghostwriters are, of course, lawyers themselves. Give us a chance (and a fee) and we’ll ghostwrite anything:  opinions for judges, reports for expert witnesses (e.g., McClellan v. I-Flow Corp., 710 F. Supp.2d 1092, 1118 (D. Or. 2010)), and (most annoyingly) pleadings for supposedly “pro se” parties.  But let a drug/device company provide authorship assistance to a busy doctor or a scientist, and the same plaintiffs’ lawyers who routinely massage (if not outright create) their experts’ opinions start screaming and yelling that something terrible is happening.  And yet, there’s no proof (and often not even an allegation) that any of the actual science in the “ghostwritten” article was misstated.

So-called “ghostwriting” is “a fairly common, but little known practice, with a pejorative name would distract the jury and needlessly consume time.”  Okuda v. Wyeth, 2012 WL 12337860, at *1 (D. Utah July 24, 2012).  Plaintiffs regularly attempt to convince juries that routine “ghostwriting” is something nefarious.  Defendants, just as often, try to keep this smoke-and-mirrors type evidence out.  We haven’t blogged about this issue before, so we thought we’d take a look at decisions excluding ghostwriting allegations.

Perhaps the most notorious ghostwriting testimony was the inflammatory rhetoric initially admitted in In re Prempro Products Liability Litigation, 554 F. Supp.2d 871, 885 (E.D. Ark. 2008), to support punitive damages.  There aren’t many judges – especially MDL judges in bellwether cases – willing to admit they were wrong and reverse a verdict, but this was one.  In Prempro “Dr. Parisian testified that the FDA would not be aware of ghostwriting” but “provided no testimony linking FDA regulations and ghostwriting.”  Id. at 885.  Plaintiffs used these (and other) allegations to bamboozle a jury into awarding punitive damages.  Id. at 889, 893, 897 (“Plaintiff asserted that ghostwriting is ‘exactly the type of conduct that necessitates punitive damages.’”) (footnote omitted).  Holding that ghostwriting testimony should never have been admitted, the court granted a new trial:

[T]here is no evidence that this practice is inappropriate or that [defendant] supported articles that it knew were false or misrepresented the science.  Rather, the articles supported [defendant’s] position on the state of the science.  Additionally, there was evidence that ghostwriting was a common practice in the industry.

Id. at 888 (footnotes omitted).  On appeal, the Court of Appeals affirmed.  In re Prempro Products Liability Litigation, 586 F.3d 547, 571 (8th Cir. 2009) (“we cannot say that the district court abused its discretion”).

The same fact pattern was addressed in Cross v. Wyeth Pharmaceuticals, Inc., 2011 WL 2517211 (M.D. Fla. June 23, 2011).  Cross “exclude[d] as irrelevant evidence of ‘ghostwritten’ articles” because “neither [plaintiff] nor her physician relied on a ‘ghostwritten’ article.”  Id. at *4.  Further, “evidence of “ghostwriting’ carries a substantial risk of misleading the jury.”  Id.  See Okuda, 2012 WL 12337860, at *1 (plaintiff could “not produce[] sufficient evidence that she or her prescribing physicians relied on any ghostwritten article in taking or prescribing the . . . drugs at issue or that the information in the articles is false”); Skibniewski v. American Home Products Corp., 2004 WL 5628157, at *1 (W.D. Mo. April 1, 2004) (evidence of ghostwriting also excluded).

Ghostwriting allegations similarly bit the dust in Bailey v. Wyeth Inc., 37 A.3d 549, 574-75 (N.J. Super. L.D. July 11, 2008). That wasn’t really surprising, since plaintiffs’ own expert “admit[ted] the beneficial contribution of the information contained in at least one article” that was allegedly ghostwritten, characterizing the information “provided to the doctor [a]s essential.”  Id. at 574.  Bailey therefore held:

There is no dispute that the articles were subject to a rigorous peer review process and were factually and medically sound.  The identified articles were published after 1994 and would not have “polluted” the information regarding [the drug] already available to the FDA.  There is no proof that these corporate-initiated articles in any way delayed the implementation of what the FDA requested be in the [drug] labeling or diluted the warnings on these drugs.

Id. at 574-75 (granting summary judgment).  Bailey was affirmed on appeal “substantially on the basis of the well-considered and exhaustive opinion . . . in the Bailey matter, which we have determined to be well supported by the evidence and legally unassailable.”  DeBoard v. Wyeth, Inc., 28 A.3d 1245, 1246 (N.J. Super. A.D. 2011).

Allegations that “ghostwriting” was a form of academic impropriety were raised, and rejected, in United States ex rel. King v. Solvay S.A., 2015 WL 8732010 (S.D. Tex. Dec. 14, 2015), a False Claims Act case.  The court determined that “evidence that [defendant] directed progress and revised the final manuscript [of an article] is not probative.”  Id. at *6.  The ghostwriting allegations were simply a prejudicial sideshow:

Relators additionally contend that as part of [defendant’s] publication strategy it commissioned smaller “investigator initiated” studies and then found “thought leaders” willing to lend their names to articles actually written by the writers who worked for [defendant], known as ghostwriters. . . .  This testimony is not an admission that [an author] merely “lent his name” to an article wholly written by [defendant’s] medical writers.  Moreover, even if it were, [plaintiffs] do not link [the] allegedly ghost-written article to any DrugDex [a compendium of off-label research] entries.  While the court understands [plaintiffs’] theory that [defendant] had a strategy to publish articles on small studies with positive outcomes and even had its own staff members write the articles and that these non-authoritative studies ended up supporting off-label use in DrugDex and other compendia, at this stage [plaintiffs] must have evidence specifically linking [defendant’s] conduct to . . . off-label use.  Innuendo related to small articles that may have been partially ghost-written but did not even end up in DrugDex is not sufficient.

Id. at *6-7 (footnote omitted).  Ghostwriting allegations in the air – not relating to anything that influenced an prescriber’s treatment of the plaintiff – also failed in Romero v. Wyeth Pharmaceuticals, Inc., 2012 WL 13036355, at *4 (E.D. Tex. April 25, 2012) (“[f]or these reasons, . . . marketing practices testimony, including . . . ghostwriting, are excluded”).

Other decisions excluding evidence of ghostwriting allegations are:  Hill v. Novartis Pharmaceuticals Corp., 944 F. Supp.2d 943, 952 (E.D. Cal. 2013) (“[Defendant] moves to preclude [plaintiff] from ‘introduc[ing] testimony or evidence that some or many of the articles . . . were actually ghostwritten by drug companies. . . .’  Having reviewed . . . all competent and admissible evidence submitted, the Court agrees such evidence should be excluded.”); Mahaney v. Novartis Pharmaceuticals Corp., 835 F. Supp.2d 299, 318 (W.D. Ky. 2011) (granting in limine motion to exclude “[t]estimony or evidence that articles were ghostwritten by drug companies”), reconsideration granted on other grounds, 2012 WL 12996015 (W.D. Ky. Jan. 4, 2012).

Finally, accusations of ghostwriting have also been a stock-in-trade of notorious plaintiffs’ “expert” Suzanne Parisian, even though she has no relevant expertise in such matters.  Ironically, in at least one deposition, “Dr. Parisian conceded that she had done ghostwriting on behalf of [a major drug company].”  Prempro, 554 F. Supp.2d at 897.  Parisian’s ghostwriting charges were excluded in a lot of Aredia/Zometa cases.  For instance, in Deutsch v. Novartis Pharmaceuticals Corp., 768 F. Supp. 2d 420, 468 (E.D.N.Y. 2011).

[T]he Court grants [defendant’s] motion to exclude Dr. Parisian’s opinions on the use of ghostwriters. . . .  The Plaintiffs argues [sic] that this testimony is relevant because it goes to [defendant’s] “communication of [relevant] risks to health care providers and the public,” which are required to be “fair and balanced” under 21 C.F.R. § 202.1. . . .  Dr. Parisian does not provided [sic] any foundation beyond her personal opinion that the use of ghostwriters . . . does not provide “fair and balanced” information.

Id. at 468.  Accord Kruszka v. Novartis Pharmaceuticals Corp., 28 F. Supp.3d 920, 935 (D. Minn. 2014) (“opinions that [defendant] convinced doctors to write publications favoring [its drugs] under the guise of independent reporting, or ‘ghostwriting,’ are outside the realm of Dr. Parisian’s expertise”); Lemons v. Novartis Pharmaceuticals Corp., 849 F. Supp.2d 608, 615 (W.D.N.C. 2012) (“the Court is not allowing Dr. Parisian to offer testimony regarding . . . ghostwriting”); In re Fosamax Products Liability Litigation, 645 F. Supp.2d 164, 191 (S.D.N.Y. 2009) (Parisian ghostwriting testimony excluded after “she could not name any standard that prohibits such a practice, as long as the information presented is accurate”); Bartoli v. Novartis Pharmaceuticals Corp., 2014 WL 1515870, at *6 (M.D. Pa. April 17, 2014) (“her testimony regarding ghostwriting . . . is inadmissible because she opines, without foundation, that employing such practices does not provide ‘fair and balanced’ information and that it must be disclosed); Earp v. Novartis Pharmaceuticals Corp., 2013 WL 4854488, at *4 (E.D.N.C. Sept. 11, 2013) (“[t]o the extent she also seeks to opine on . . . industry ghostwriting . . . that would unduly prejudicial, irrelevant, or outside the scope of her expertise, [and] the court will not allow her to do so”); Hill v. Novartis Pharmaceuticals Corp., 2012 WL 5451809, at *2 (E.D. Cal. Nov. 7, 2012) (“Defendant’s motion to exclude Dr. Parisian’s testimony regarding ghostwriting . . . is GRANTED”); Georges v. Novartis Pharmaceuticals Corp., 2012 WL 9064768, at *14 (C.D. Cal. Nov. 2, 2012) (“exclud[ing] Dr. Parisian’s testimony regarding ghostwriting”); Zimmerman v. Novartis Pharmaceuticals Corp., 2012 WL 13009101, at *1 (D. Md. Sept. 25, 2012) (“Dr. Parisian may not offer opinion testimony on . . . Ghostwriting”); Winter v. Novartis Pharmaceuticals Corp., 2012 WL 827245, at *3 (W.D. Mo. March 8, 2012) (“Motion to exclude evidence that articles concerning [the class of drugs] in medical journals were actually ‘ghostwritten’ by companies, including [defendant], is granted consistent with prior rulings”); Mahaney v. Novartis Pharmaceuticals Corp., 2011 WL 13209814, at *2 (W.D. Ky. Nov. 15, 2011) (“exclud[ing] Parisian’s testimony on ghostwriting”).

Sharp-eyed readers will note that all of these decisions, except for the King False Claims Act ruling, were issued in the decade between 2004 and 2014 – which we are wont to call the “coprolitic age” of ghostwriting allegations, powered mainly by the aforementioned Suzanne Parisian.  We hope that the other side drew back a nub enough times on this issue that it’s no longer worth the candle to develop.  After all, who knows how many of those expert opinions were ghostwritten by plaintiffs’ counsel?  We’d like the issue to stay dead, so for the good of the order we have compiled all of the favorable precedent here.

Today’s guest post, by Luther Munford of Butler Snow, engages in one of our currently favorite activities, that being informed speculation on what might be the consequences of a favorable Supreme Court resolution of its currently pending preemption appeal in Merck Sharp & Dohme Corp. v. Albrecht.  We hope he’s right.  As always, our guest posters deserve 100% of the credit (and any blame) for their thoughts published here.  We only provide the forum.

**********

In Merck Sharp & Dohme Corp. v. Albrecht, the Solicitor General as amicus curiae argues that judges, not juries, are best suited to evaluate the scope of an FDA determination. Brief of the United States as Amicus Curiae Supporting Petitioner, Merck Sharp & Dohme Corp. v. Albrecht, (No. 17-290), 2018 WL 4562163 (filed Sept. 20, 2018).  Judges, he says, “are trained and experienced in construing legal documents and are far better equipped to understand agency decisions in light of the governing statutory and regulatory context.”  Id. at *15.

To support his argument, he cites the Administrative Procedure Act’s statement that a reviewing court shall “determine the meaning or applicability of the terms of an agency action,” 5 U.S.C. § 706, and precedent that has looked to judges to interpret the meaning of prior adjudications.  See id. at 18-20.

While that case has nothing to do with either medical devices in general or 510(k) clearance in particular, Supreme Court agreement with the Solicitor General on this point could radically alter the way courts view the admissibility of 510(k) clearance, at least where Class II devices are concerned.  To gain 510(k) clearance, unless the FDA decides that more is required, the manufacturer need only establish that a new device is as safe and effective as an existing lawfully marketed device.  No other evidence of safety and effectiveness is required in the absence of FDA action.

At present, courts have treated the meaning of 510(k) clearance as a subject for warring expert testimony as to its meaning, as discussed here on the blog (discussing In re Cook Medical, Inc. IVC Filters Mktg., Sales Practices and Prod. Liab. Litig., 2018 WL 6617375 (S.D. Ind. Dec. 18, 2018)).  In other cases evidence of clearance has been excluded, with one explanation being the mistaken theory that the evidence was of such slight probative value that the battle was best avoided.  In re C.R. Bard, Inc. MDL No. 2187, 810 F.3d 913, 922 (4th Cir. 2016). [Ed. note: Other cases, collected and discussed here, admit evidence of FDA device clearance.]

But if the United States Supreme Court decides that preemption issues are treated as purely legal, the courts will be forced to examine the statutory context that controls Class II clearance using 510(k).  That context shows that such a clearance is almost always an FDA “determination” that the device, with whatever special controls that the FDA ultimately imposes, does not present a potential unreasonable risk of illness or injury.  Nothing could be more relevant to a product liability claim.

To begin at the beginning, in 1976 Congress directed the FDA to engage in a sort of regulatory “triage.”  Triage sorts patients according to the seriousness of their injuries and gives them different levels of care.  The FDA’s statutory scheme sorts devices according to the seriousness of the risks they present and requires different levels of premarket review.

More precisely, Congress directed the FDA to engage medical panels to classify medical devices according to their need for regulation.  Congress specified the qualifications the panel members were to have. 21 USC § 360c(b).  The FDA methodically proceeded over the ensuing decades to classify device types into classes I, II or III according to the risk they present.  It convened the panels, held hearings, published panel recommendations, entertained comments from the public, and fixed the classifications.

Devices that present little risk, such as tongue depressors, were put in Class I and do not need FDA review before they are sold to the public.

Moderate risk devices, such as surgical suture, were put in Class II and, if not exempt, must be “cleared” by the FDA before they are marketed.  To gain Class II clearance, the manufacturer must show in a §510(k) submission that they are as safe and effective as an existing Class II legally marketed device that presents a moderate risk.

Finally, devices that may “present a potential unreasonable risk of illness or injury” or are for sustaining life, such as a pacemaker, were put in Class III and generally must be “approved” by the FDA based on extensive independent evidence of their safety and effectiveness.  21 U.S.C. § 360c(a)(1)(C).

On its face, this is a reasonable way to regulate medical devices.  In fact, Congress stated in the statute that it believed this system of classification and review provided “reasonable assurance” of safety and effectiveness for each class of medical device.  21 U.S.C. § 360c(a)(1)(A), (B), (C).

It is encouraging, but not necessary, to observe that the FDA’s sorting of devices seems to have worked.  Even though they receive less FDA scrutiny, “cleared” devices of moderate risk are less likely to result in a serious recall than “approved” devices that may present an unreasonable risk.  One study showed that while 510(k) cleared devices constitute 98% of all devices, they account for only 71% of serious recalls.  And the 2% of devices that are PMA-approved made up 19% of serious recalls.  Jeffrey Shapiro, Substantial Equivalence Premarket Review: The Right Approach for Medical Devices, 69 Food & Drug L.J. 365, 389-390 (2014).

But that is not all. According to the statute, placement in Class II is itself a determination of safety.  If a device is a Class II device, then, with special controls, it usually does not present “an unreasonable risk of illness or injury” that would require it to be in Class III.  21 U.S.C. §360c(a)(1)(C).  The FDA has made that determination based on the initial work of a medical panel and information in the §510(k) which confirmed that the specific device fell within the panel’s classification of the device type.  Otero v. Zeltiq Aesthetics, Inc., 2018 WL 3012942 *3 (C.D. Cal. June 11, 2018).  This can be confirmed when the decision classifying the device type expresses the opinion that there is no unreasonable risk, or uses words to that effect.

In other words, clearance of a moderate risk device using 510(k) is normally a sign of relative safety, even though the FDA review of more risky Class III devices is more rigorous.  That is what the statute says, and experience seems to bear that out.

If the Supreme Court should agree that the interpretation of an FDA decision is a matter of law for the court, then in any case involving a Class II device, the defendant should be entitled to an instruction on the meaning of that decision.

If the FDA decision classifying the device type rests on a finding of no unreasonable risk then, as a matter of law, the defendant should be entitled to an instruction that “The United States Food and Drug Administration has cleared this device for marketing as a Class II device.  That clearance is a determination that there is reasonable assurance of its safety and effectiveness and that, with whatever special controls may have been imposed, the device does not present a potential unreasonable risk of illness or injury.”

Such an instruction would, of course, be radically different from the treatment courts have recently given §510(k) clearance.  Those courts have not only mistakenly allowed juries to decide the meaning of §510(k) clearance, but they have done so in part because of fundamental legal error in the way they have examined §510(k) clearance.

The fundamental error of those courts has been to overlook the distinction between the normal use of §510(k) to clear devices of a type placed in Class II, and the increasingly rare transitional use of §510(k) to clear devices in Class III based on a pre-1976 predicate.

The use of pre-1976 predicates originated in an interim provision Congress adopted in 1976, when the statutory scheme was new. Congress put all implantable devices in Class III as presenting an unreasonable risk pending medical panel review to reclassify them.  Then, anticipating that it would take medical panels a long time to do their work – and has taken more than 40 years – it allowed these devices to be “cleared” using §510(k) if they could be shown to be equivalent to a device on the market in 1976.  This process, unlike the normal use of §510(k), did not involve any medical panel review and did not require equivalence to a classified device.

This increasingly rare scenario was what the Supreme Court addressed in Medtronic v. Lohr, 518 U.S. 470, 477 n.3 (1996), which, quite incorrectly, has been taken as being representative of all FDA clearance decisions.  But it is not. In fact, now that the medical panel reviews of devices with a potential high risk appear to have been almost completed, the scenario is not representative of practically anything that the FDA is still doing today.  See FDA, FDA Has Taken Steps to Strengthen The 510(k) Program 7 (November 2018).

Where a device has been placed in Class II in what is now the ordinary fashion − based on equivalence in safety and effectiveness to a predicate device placed by a medical panel in Class II because its risks are reasonable − that decision, as a matter of simple statutory interpretation, is a determination of safety.  Even if juries are free to disagree with it, they should not be allowed to ignore it. What the Supreme Court tells us in Merck Sharp & Dohme Corp. may have effects far beyond the resolution of the case at hand.

We’ve been waiting quite a while for the FDA to modernize its positions on the truthful off-label communications by regulated manufacturers.  Under current First Amendment practice, the FDA’s positions are quite likely unconstitutional as both speaker- and topic-based restrictions on the truthful communication of scientific information.  However, the best we’ve seen from the Agency to date was a “statement,” issued last June, that truthful off-label communications to third-party payers (or “payors,” if you’d rather) would henceforth be OK.  Further details were available in an FDA guidance issued at the same time.  We blogged about that statement here.

The June 2018 guidance, however, did not go beyond TPPs.  Id. at 22.  We editorialized at the time:

We fail to see how any constitutionally valid distinction can exist between providing the identical information, with identical disclaimers and limitations, to one “sophisticated” audience (third-party payors) while prohibiting that information’s distribution to another “sophisticated” audience – that being medical doctors that directly prescribe these drugs and devices. . . .  Thus, we believe that, as a practical matter, the OLP Guidance effectively dooms any First Amendment defensibility of an FDA ban on the same truthful information being distributed, in the same fashion, to the rest of the medical community. . . .  After all, just as third-party payers are sophisticated professionals . . ., so are the doctors directly involved in treating [their] patients.

Then, nothing.  Over the last eight months we’ve seen no detectable movement through either FDA guidance documents or, the pending process – a reassessment of which was announced two years ago (see our post here) – to amend the FDA’s archaic “intended use” regulations.

We’re not surprised that a lot of other people are tired of the FDA’s foot-dragging.  Among other things, the FDA stated in that guidance we mentioned above that it would be deferring to “existing current good research practices for substantiation developed by authoritative bodies.”  Guidance at 10.

One of those “authoritative bodies” is the Advanced Medical Technology Association (“AdvaMed,” for short), which describes itself as “a trade association that leads the effort to advance medical technology in order to achieve healthier lives and healthier economies around the world.  AdvaMed’s membership has reached over 400 members and more than 80 employees with a global presence.”  One of the things AdvaMed has done is to create a code of ethics for medical device manufacturers.  The last AdvaMed code was issued about a decade ago and was in need of updating.  So AdvaMed has prepared a new ethics code that’s due to become effective on January, 1, 2020.  The entire code is methodically discussed in a Reed Smith client alert, but for present purposes we’re interested section 10, entitled “Communicating for the Safe & Effective Use of Medical Technology.”

Full disclosure, as stated in the client alert: “Reed Smith was honored to serve as outside counsel to AdvaMed in connection with drafting the original, current, and revised versions of the AdvaMed Code.”

Section X of the new AdvaMed Code starts with a specific description of off-label use, so it’s obvious to all what the code is discussing:

Health Care Professionals may use a product for any use that they determine is in the best medical interests of their patients.  This includes uses that are contained in the Medical Technology’s labeling or otherwise consistent with such labeling, but it could also include uses that are not approved or cleared (i.e. “off-label” uses).  As recognized under U.S. law and by the FDA, off-label use of these Medical Technologies can be an important part of medical practice and may even constitute a medically recognized standard of care.

2020 AdvaMed Code at 29 (emphasis added).  One “key concept” that this section of the new AdvaMed code emphasizes is that “[a]ccess to truthful and non-misleading information relating to Medical Technologies is critical to a Health Care Professional’s ability to exercise his or her medical judgment, to provide high-quality care, and to safely use available Medical Technology.”  Id. (emphasis added).

Thus, AdvaMed has stopped waiting for the FDA and is taking the position that direct off-label communications between medical device manufacturers and the physicians (and other health care professionals) that use their products is 100% ethical if done in a truthful, transparent, and non-misleading fashion:

Industry appropriate communications of such information can include, among other activities:

  • Proper dissemination of peer-reviewed scientific and medical journal articles, reference texts, and clinical practice guidelines;

  • Presentations at educational and medical meetings regarding clinical trial results or research and development data for an investigational use (taking care that no claims are made regarding safety and effectiveness); and

  • Discussions with consultants and Health Care Professionals to obtain advice or feedback relating to topics such as unmet patient needs, product research and development, and the like.

Id.  AdvaMed’s ethical guidelines for such communications (which the FDA would call “promotion”) are about what one would expect a company with good marketing practices to employ generally – authorization, truth, and disclaimers:

  • Company responses that contain information regarding unapproved or uncleared uses should be provided by authorized personnel.

  • Company communications must be truthful and nonmisleading.

  • Information related to unapproved or uncleared uses should be identified as such.

Id. at 30.

Thus AdvaMed – one of the “authoritative bodies” to which the FDA has stated it would be deferring – has stepped up to the plate and declared that truthful manufacturer communication of off-label information to physicians and other health care practitioners is both ethical and desirable.  “Companies are encouraged to develop policies and controls that apply the principles above.”  Id.

FDA, the ball is in your court.  Either take command or lose control.  Industry is not going to wait forever for the Agency to bring its regulation of truthful commercial speech into the Twenty-First Century.

Moreover, from a litigation standpoint, the AdvaMed Code is what we call an “industry standard” (at least when it goes into effect in 2020), and in almost every jurisdiction in the country, compliance with industry standards is admissible evidence in product liability cases.  “[I]ndustry standards promulgated by trade associations” are “acceptable” as evidence in strict product liability cases.  Kim v. Toyota Motor Corp., 424 P.3d 290, 299 (Cal. 2018).  “[T]rade associations consist of manufacturers and other businesses whose conduct comprises the industry custom and practice.”  Id.  Accord, e.g., Adams v. Genie Industries, Inc., 929 N.E.2d 380, 385 (N.Y. 2010) (“weight” of an “industry standard for” the product “was up to the jury”); Mikolajczyk v. Ford Motor Co., 901 N.E.2d 329, 335 (Ill. 2008) (“conform[ance] with . . . guidelines provided by an authoritative voluntary association” held relevant to proving a “unreasonably dangerous” defect); Wash. Rev. Code §7.72.050(1) (“[e]vidence of custom in the product seller’s industry . . . may be considered by the trier of fact”).  So now, when plaintiffs start bleating about “off-label promotion” being such a horrible thing, we have evidence to point to that – provided it’s done correctly and in compliance with a company’s internal guidelines – such activity is recognized as proper and ethical.

Oh, yes, and Happy Valentine’s Day to everyone at the FDA.  Among other things, they keep the nation’s chocolate supply safe.

Today’s guest post is by Reed Smith associate Regina Nelson.  In it she tackles an issue that inevitably arises whenever ediscovery for defendants is successful, that is, what must be done to have the fruits of that discovery be admitted at trial, otherwise known as authentication.  She discusses a recent Pennsylvania appellate case that addressed this issue.  As always our guest bloggers are 100% responsible for their posts, and deserving of all the credit (and any blame) for what they write.

**********

Social media is ubiquitous.  Almost everyone knows someone who is a member of Facebook, Twitter or Instagram, even those of us who do not use these services personally.  As people post about their comings and goings or opinions on social media, it is not surprising that they are becoming an issue in both civil and criminal cases.  The blog has a cheat sheet, “Ediscovery for Defendants,” with over 100 decisions, devoted to discovery of plaintiffs’ social media.

In a recent case of first impression in Pennsylvania, the Superior Court in a criminal case Commonwealth v. Mangel, 181 A.3d 1154 (Pa. Super. Ct. Mar. 15, 2018), discussed what lawyers need to do at trial to authenticate an opponent’s social media pages – in this case, on Facebook.  While Mangel applied Pennsylvania law, it is potentially relevant to federal cases as well because Pennsylvania Rule of Evidence 901 is almost identical to Federal Rule of Evidence 901.

Under Pennsylvania Rule of Evidence 901, authentication is a prerequisite  to admission of most evidence.  Like the federal rule, the standard is adequate proof that the evidence is what it purports to be.  Usually testimony is enough to authenticate evidence.  Evidence can also be authenticated by other methods, including circumstantially.  Id. at 1159.  Admissibility of electronic communications is evaluated on a case-by-case basis.  Merely showing that the electronic communication belonged to someone was not sufficient to authenticate it.  See Commonwealth v. Koch, 39 A.3d 996, 1005 (Pa. Super. Ct. 2011).  Mangel addressed how much circumstantial evidence was sufficient to corroborate the identity of the sender.

In Mangel, the trial court had denied a Motion in Limine to admit Facebook posts and messages allegedly authored by the defendant in an assault and battery case.  Id. at 1155.  The alleged victim could identify the defendant from Facebook pictures.  Id. at 1155-56.  The social media evidence at issue was:  (1) screenshots of certain Facebook account pages in the name of the defendant, and (2) a Facebook screenshot of a photograph of purportedly bloody hands had been posted by another person.  Id. at 1156.

To authenticate the Facebook evidence, the prosecution relied on a qualified expert in computer forensics.  Id.  After researching Facebook and finding a relevant account, the expert determined that the Facebook account she located with a search for the defendant’s name also matched that of the Facebook pages provided to her by the Commonwealth.  Id.  That identification evidence was significantly undermined on cross-examination, however.  First, the prosecution expert conceded that she could not say to a reasonable degree of certainty that the defendant, and not someone else, authored the posts.  Id. at 1157.  Second, the expert failed to obtain an IP address for the account in question.  Id.   Third, the expert was impeached by defense counsel with a Facebook search showing multiple accounts for someone with the same name as the Defendant.  Id.  After hearing this testimony, the trial court denied the Motion in LimineId.

The Superior Court in Mangel expressed concerns about authorship of the Facebook pages and posts similar to the Koch case, which it discussed at some length.  Id. at 1160.  To authenticate social media evidence, the  prosecution had to present direct or circumstantial evidence corroborating the identity of the on-line author.  Id. at 1161.  Critically, the defendant in Mangel did not concede at any time that the Facebook account was his or that he authored the posts.  Id. at 1163.  Establishing authorship was key due to the nature of social media.  “[T]he same authorship concerns . . . in relation to e-mails and instant messages, exist in reference to Facebook and other social media platforms, that can be accessed from any computer or smart phone with the appropriate user identification and password.”  Id. at 1162.  Another failing by the prosecution only compounded its expert’s weaknesses.  It never obtained the username or password for the Facebook account to confirm its authenticity.  Id. at 1163.  That a Facebook page contained the defendant’s name, hometown and school was not sufficient to establish that the defendant actually authored the posts on the page.

The mere fact that the Facebook account in question bore [defendant’s] name, hometown and high school was insufficient to authenticate the online and mobile device chat messages as having been authored by [him].  Moreover, there were no contextual clues in the chat messages that identified [him] as the sender of the messages.

Id. at 1164.

In addition to concerns about identity and authorship, issues also arose with the content of the messages and posts.  Id. at 1163-64.  Nothing established the time the posts and messages were created.  Id. at 1163.  Absence of a timestamp precluded confirmation that the posts were made in connection with the alleged incident.  Id. at 1163-64.  The Facebook posts were ambiguous in that they did not specifically reference the defendant.  Id. Finally, the posts and messages were devoid of any “distinct characteristics” indicative of the author’s personal style that might have corroborated the defendant as their author. Id.

Thus, the Superior Court agreed with the trial court’s decision.  There must be affirmative evidence that the defendant was the actual sender of the messages.  Id. at 1164. Mangel should serve as a reminder that locating seemingly relevant social media evidence is only the first, not the last step.    Social media evidence must also be authenticated before it will be admitted into evidence.  By all means, learn from the mistakes made by the prosecutors in Mangel.

To avoid a similar situation, those seeking to authenticate social media should consider taking certain steps:

  • Establish who, other than the account owner, has access and/or authorship rights.  If such other persons exist, their actions must be excluded.
  • Seek authorship admissions from the owner.
  • If, as in Mangel, such admissions are not forthcoming, conduct additional investigations to exclude other potential authors.
  • If Facebook messenger or some other messaging tool is involved, similarly establish authorship of relevant posts.
  • If necessary, depose the non-authoring recipients of messages to confirm their knowledge of the sender’s identity.
  • Obtain relevant IP addresses of social media evidence, ideally from material provided by the social network operator.
  • Obtain timestamps, or other temporal data, again from the website operator.
  • Obtain available metadata from the operator and have it forensically reviewed.
  • Hire a competent expert, knowledgeable about the applicable certainty standards for opinions, who will not wilt under cross-examination.
  • Include questions about the contents of the social media at issue, such as identification of photographs, in deposition outlines for witnesses other than the account owner.
  • If slang is an obstacle, have slang interpreted by the account owner and (if necessary) others during discovery.

On Wednesday, the Fifth Circuit was finally able speak to what’s been going on in a Dallas courtroom that has racked up over $1.7 billion—that’s billion—in jury verdicts over the last two years in the Pinnacle Hip Implant MDL. And the Fifth Circuit entered the room loudly. It ordered a new trial of the plaintiff’s very first victory, the one that produced a half-billion dollar verdict. The court did not hold back, making it perfectly clear that it vacated the judgment due to “the district court’s evidentiary errors and [plaintiffs’ attorney Mark] Lanier’s deception.” Slip Op. at 6.

The “evidentiary errors” have been a controversial part of these MDL trials since the time of our first post about them two years ago. Even then, we were struck by the “number and nature” of these evidentiary rulings, which in the aggregate suggested “an almost uninterrupted flow of unduly prejudicial and irrelevant information to the jury.”  The Fifth Circuit now agrees, highlighting two of them as the basis to order a new trial.

First, the Saddam Hussein evidence. Slip Op. at 43-46. That’s right. The court allowed evidence concerning Saddam Hussein into a hip implant trial. Its decision was based on a deferred prosecution agreement, one in which J&J took responsibility for the actions of affiliates who had bribed officials in the Iraqi government. These affiliates had nothing to do with the Pinnacle Hip Implant device. Regardless, after defendants elicited testimony on their own positive internal culture and marketing, the district court ruled that the defendants had thereby “opened the door” to Saddam Hussein, the deferred prosecution agreement and all sorts of other stuff. With light now green, the plaintiffs’ attorney thereafter featured Saddam Hussein and bribes and prosecution in his questioning of witnesses and closing argument. Mind you, this was a hip implant medical device trial.

The Fifth Circuit rejected the trial court’s “open door” ruling. It held that “the rules of evidence do not simply evaporate when one party opens the door on an issue.” Id. at 44. Prior bad acts cannot be used to convince a jury that defendants acted as wrongdoers in the case before it. Id. The Fifth Circuit held, however, that the plaintiffs’ attorney asked the jury to do just that. To illustrate this, the Fifth Circuit quoted the paragraph below from plaintiffs’ closing argument. Note that the italicized emphasis in the paragraph was placed there by the Fifth Circuit, as was the single word “Indeed” after that paragraph:

If you go back and look at the DPA, that’s the deferred prosecution agreement where the company paid money one time because of kickbacks to doctors in America, the other time because of the bribes to Saddam Hussein’s government, the bribes in Greece, Romania, Poland and other places where they were bribing people to put in . . . their products. The DPA has [J&J] admitting its responsibility in it. J&J is admitting that they’re responsible. They have already taken this issue out of your hands realistically. That alone is a winner. . . . [J&J] has admitted their responsibility for this. That ought to be enough.

Indeed.

Id. at 45. The Fifth Circuit wrote that this closing argument and the earlier questioning “tainted the result by inviting the jury to infer guilt based on no more than prior bad acts . . . . That alone provides grounds for a new trial.” Id.

The second evidentiary ruling rejected by the Fifth Circuit was the trial court’s decision to allow plaintiffs to use hearsay in a resignation letter from a DePuy employee alleging racism within the company. Slip Op. 46-48. Again, this was a hip implant trial. Calling it a “spectacle,” the Fifth Circuit ruled that reading this letter to the jury “refocused its attention on serious, and seriously distracting, claims of racial discrimination that defendants had no meaningful opportunity to rebut via cross-examination. This spectacle fortifies that a new trial is required.”

These evidentiary rulings alone were enough to upend the judgment. But there were more. As we laid out in our previous posts, the trial court made multiple other questionable evidentiary rulings that allowed questioning and argument on things like suicide, cancer, connections to the tobacco industry, transvaginal mesh suits, and so on. Having already overturned the judgment, the Fifth Circuit declined to address these other evidentiary rulings. But it did warn the trial court to “weigh carefully the application of Rule 403 and 404(b)” when considering these issues at future trials. Slip Op. at n. 71.

Next, the Fifth Circuit found additional grounds to overturn the judgment due to what it called “deception” by the plaintiffs’ attorney regarding plaintiffs’ experts, something we wrote about last year. Plaintiffs classified two of its experts as “non-retained,” meaning not paid. At trial, plaintiffs’ attorney contrasted this with what he called the “bought testimony” of the defendants’ expert. The problem is, however, that plaintiffs’ experts were “bought” too. Before trial, plaintiffs’ attorney donated $10,000 to St. Rita’s Catholic School, the favorite charity of one of the two experts. More blatantly, after trial, plaintiffs paid $65,000 in total to the two experts.

It would be difficult to overemphasize how hard the Fifth Circuit came down on this, labeling it “deception.” Slip Op. at 6. The court was clearly displeased. Its opinion (see pages 49-57) is littered with snide comments. Noting that Mr. Lanier mentioned to the jury that he had shared the “best apple pie in the world” with one of the two experts, the court wrote, “St. Rita’s and the $10,000 check went unmentioned.” Id. at 50. After quoting Mr. Lanier contrasting the supposedly “bought testimony” of the defendants’ experts with the supposed “real life” testimony of his two experts, the Fifth Circuit wrote: “As between ‘real life’ and ‘bought testimony,’ [the jury] chose the former by a margin of $502 million. But that choice was a false one, manufactured entirely by Lanier.” Id. at 52.

With the unmistakable intent to drive its point home, the Fifth Circuit clarified in just about every way possible that this type of maneuvering was improper:

This is the rare case in which counsel’s deceptions were sufficiently obvious, egregious, and impactful to penetrate the layers of deference that would ordinarily shield against reversal.

Lanier’s failure to disclose the donation, and his repeated insistence that Morrey Sr. had absolutely no pecuniary interest in testifying, were unequivocally deceptive.

Lawyers cannot engage with a favorable expert, pay him “for his time,” then invite him to testify as a purportedly “non-retained” neutral party. That is deception, plain and simple.

We find, by the “clear and convincing” evidence of common sense, that Lanier misled the jury in creating the impression that Morrey Jr. had neither pecuniary incentive nor motive in testifying. Neither our double deference nor counsel’s specious reasoning can alter that conclusion.

Calculated or not, falsehoods marred plaintiffs’ victory. The Verdict cannot stand.

Got it. This lambasting by the Fifth Circuit could change the game in the Pinnacle Hip Implant MDL. Proper boundaries may now be back in place, or at least some of them. There are more appeals to come, and they involve serious procedural and jurisdictional issues that were front and center in the two trials that followed this one. In fact, there may be more to come from this opinion, at least from us, as it also granted DePuy judgment on two plaintiffs’ failure to warn claims and J&J judgment on plaintiffs’ aiding and abetting claims—something for us to address another day.

One thing is already clear from this opinion, though. Two years of this MDL may have been wasted on defective jury trials. But, if a reset is needed, it must be done. The goal of an MDL is not to hammer defendants with every possible negative piece of evidence, relevant or not, so as to produce large verdicts. It is to establish a framework under which the parties can properly litigate and value the hundreds or thousands of cases within a complex mass tort. The Fifth Circuit’s opinion is an important step toward that proper construct.