Troutmansanders.com

“JAWS”
Attacks On The Daubert Trilogy
A Case Study: The Parlodel® Litigation
Stephen D. Otero is a senior associate in the Richmond office of Troutman Sanders LLP He is a
graduate of the University of Virginia (B.A 1989) and William & Mary Law School (J.D. 1995), where
he was a member of the William & Mary Law Review and the Moot Court Board. Since completing a
clerkship in the Eastern District of Virginia in 1995-96, Mr. Otero has had extensive experience in state
and federal courts in a broad range of products liability matters, including mass tort experience in diet pill
and hypertension pill multi-district litigation panels and products liability cases involving printing
presses, industrial equipment and controls, office equipment, ladders and recreational equipment. Mr.
Otero has written several articles on issues relevant to products liability law, including “Challenging Non-
Scientific Experts After Kumho Tire,” 41 For the Defense 12 (Defense Research Institute June 1999) (co-
author) and “The Case Against FIFRA Preemption: Reconciling Cipollone’s Preemption Approach with
Both the Supremacy Clause and Basic Notions of Federalism,” 36 William & Mary Law Review 783
(1995). He is a co-author of Virginia Law of Products Liability (Harrison 2002). Mr. Otero is Chair-
Elect of the Virginia Bar Association Young Lawyers Division and a member of the DRI Drug & Medical
Device, Products Liability and Young Lawyers Committees.
Melissa Roberts Levin is a senior associate in the Richmond office of Troutman Sanders LLP
She is a Phi Beta Kappa graduate of the University of Virginia (B.A. 1994) and the University of VirginiaLaw School (J.D. 1997), where she was a finalist in the William Minor Lile Moot Court competition anda member of the editorial board of the Virginia Environmental Law Journal. Ms. Levin’s practice focuseson products liability and tort litigation. She has assisted in the defense of pharmaceutical and medicaldevice manufacturers in the diet pill, hip implant and PPA mass tort litigation in federal and state courts.
She also has defended various national and international manufacturers and distributors of recreationaland other equipment. Ms. Levin is a co-author of Virginia Law of Products Liability (Harrison 2002).
She is on the Board of Directors of the Young Lawyers Section of the Richmond Bar Association and is amember of the DRI Drug and Medical Device, Products Liability and Young Lawyer Committees.
Mr. Otero and Ms. Levin wish to thank Michael H. Gladstone, a partner in the Richmond, Virginia officeof Troutman Sanders, L.L.P., for suggesting this topic.
INTRODUCTION
In March 1999, the Supreme Court issued its third opinion in six years regarding the admissibility of expert evidence. See Daubert v. Merrell Dow Pharmaceuticals, Inc., 509U.S. 579 (1993); General Electric Co. v. Joiner, 522 U.S. 136 (1997); Kumho Tire Co. v.
Carmichael, 526 U.S. 137 (1999) (hereinafter “the Daubert Trilogy”). The opinion in KumhoTire concluded one battle of the experts by making it clear that a trial court’s gatekeepingobligation applies to “non-scientific” experts as well as scientific experts. See Kumho Tire Co.,526 U.S. at 149. Kumho Tire did not, however, bring an end to the expert wars. To the contrary,it has had the practical effect of extending the front lines of the expert wars to all kinds of experttestimony. See, e.g., Carr, Dabney J. et al. “After Kumho Tire: Challenging Non-ScientificExperts,” 41 For the Defense 12, 13 (Defense Research Institute 1999).
This article examines the litigation over the pharmaceutical Parlodel® in general, and an appeal argued this week in the Eleventh Circuit in particular, as a representative battle inthe continuing expert wars. The purpose of the case study is to learn: How courts are applying the Daubert Trilogy in pharmaceutical cases; and The latest tactics employed to wage the expert wars in pharmaceutical At a minimum, the Parlodel® litigation should be of interest to pharmaceutical Recent proceedings reflect that it is a determined and organized attempt by the plaintiffs’ bar to eliminate, or at least minimize the importance of, epidemiological evidencein toxic exposure cases.
Plaintiffs are pursuing appeals of two district court opinions excluding contested scientific evidence in both the Tenth Circuit (Hollander v. SandozPharmaceuticals Corp., 95 F. Supp.2d 1230 (W.D. Okla. 2000)) and the Eleventh Circuit(Siharath v. Sandoz Pharmaceuticals Corp., 131 F. Supp.2d 1347 (N.D. Ga. 2001)),notwithstanding an unsuccessful appeal of the exclusion of the same evidence in the EighthCircuit (Glastetter v. Novartis Pharmaceuticals Corp., 107 F. Supp.2d 1015 (E.D. Mo. 2000),aff’d 252 F.3d 986 (8th Cir.) (per curiam), pet. for reh’g and reh’g en banc denied (8th Cir.
2001)).
The Siharath appeal currently pending before the Eleventh Circuit has captured the attention of distinguished legal and medical scholars, as evidenced by the threeAmicus Curiae briefs filed in support of the Plaintiffs-Appellants. The Amicus briefs were filedby two law professors and a professor of medicine who have authored parts of the FederalJudicial Center’s Reference Manual on Scientific Evidence, and three other distinguishedprofessors of medicine, one of whom served as Editor-in-Chief of the New England Journal ofMedicine from 1991 through 1999. See infra § IV.
In the Eleventh Circuit appeal of Siharath, the Plaintiffs-Appellants argue that the standard of review is de novo because the district court misinterpreted the DaubertTrilogy. See infra § IV.A. This argument appears to be an invitation to retreat from Joiner’sholding that “abuse of discretion is the proper standard by which to review a district court’sdecision to admit or exclude scientific evidence.” See Joiner, 522 U.S. at 146.
This article begins with some background on Parlodel® and an overview of the scientific evidence at issue in the litigation. It then recounts lower court opinions determiningthe admissibility of the contested scientific evidence and summarizes the briefs of the parties andthe Amici Curiae in the Siharath appeal argued in the Eleventh Circuit this week. Lastly, thearticle offers some explanations of the significance of the battle over the experts in the Parlodel®litigation.
BACKGROUND
Parlodel®:
Parlodel® is manufactured by Sandoz Pharmaceuticals Corporation (“SPC”), which is now Novartis Pharmaceuticals Corporation. See Siharath v. SandozPharmaceuticals Corp., 131 F. Supp.2d 1347, 1349 (N.D. Ga. 2001).
Parlodel® blocks secretion of the hormone prolactin, which induces the breasts to secrete milk. See Rider v. Sandoz Pharmaceuticals, Corp. and Siharath v. SandozPharmaceuticals Corp., Corrected Brief of Defendants-Appellees, Case Nos. 01-11965-BB, 01-11966-BB (consolidated) (Appeal to the 11th Circuit, filed Aug. 30, 2001) (hereafter “Brief ofDefendants-Appellees”), at 7. Significantly, Parlodel®’s active ingredient, bromocriptinemesylate, has been associated with vasodilatory properties and hypotensive effects. See Brief ofDefendants-Appellees at 7-8.
SPC began selling Parlodel® in 1980 as a treatment for amenorrhea (the absence of menstruation) and galactorrhea (milk flow not related to childbirth or nursing). SeeRider v. Sandoz Pharmaceuticals, Corp. and Siharath v. Sandoz Pharmaceuticals Corp., JointBrief and Appendix of Plaintiffs-Appellants, Case Nos. 01-11965-BB, 01-11966-BB(consolidated) (Appeal to the 11th Circuit, filed July 18, 2001, 2001) (hereafter “Brief ofPlaintiffs-Appellants”), at 5 and n.3. SPC continued to market Parlodel® for the prevention ofphysiologic lactation (“PPL”) from 1980 through 1994. See Brief of Defendants-Appellees at 8;see also Brief of Plaintiffs-Appellants at 6.
In 1983, the FDA issued an “ADR Highlights” about Parlodel® and reports of hypertension, seizure or stroke. See Brief of Plaintiffs-Appellants at 6. In 1983 and1985, the FDA asked SPC to include a warning about hypertension, seizure and stroke in thedrug’s labeling. See id. at 7.
In 1987, at the FDA’s request, SPC revised the Parlodel® package insert to include a statement regarding adverse effects such as seizure, stroke and myocardial infarction and issued a “Dear Doctor” letter to advise of the labeling changes. See Eve v. SandozPharmaceutical Corp., 2001 U.S. Dist. LEXIS 4531 at *18 (S.D. Ind. 2001).
“[I]n June 1989, the FDA reported to its Maternal Health Drugs Advisory Committee ‘that it had received a total of 85 serious ADEs, including 10 deaths, since approvalof [Parlodel® use for PPL] in 1980.’” Brief of Plaintiffs-Appellants at 7. Although itrecognized that the ADE’s did not prove that Parlodel® caused hypertensive crises, seizures orcerebral vascular accidents, the FDA concluded that “‘in the aggregate [they] suggested thatbromocriptine may be the cause of these serious adverse experiences.’” Id. In 1989, the FDA twice requested that SPC withdraw Parlodel®’s indication for PPL, and in 1994, it initiated withdrawal proceedings. In January 1995, SPCvoluntarily withdrew the drug as indicated for PPL. See id. at 9.
Parlodel® remains FDA-approved for the treatment of other conditions, including Parkinson’s disease, amenorrhea, and galactorrhea. Eve, 2001 U.S. Dist. LEXIS 4531at *11; Brief of Defendants-Appellees at 7-8.
Overview of scientific evidence regarding a possible association between
Parlodel® and strokes or seizures
Epidemiological studies
Kenneth Rothman, An Epidemiological Evaluation of the Possible Relation Between Bromocriptine, Puerperal Seizures and Strokes, (Epidemiologic Resources,Inc. Sept. 30, 1998) (referred to as “ERI” or the “ERI Study”): This study concluded that there is no statistically reliable association between Parlodel® and stroke. “The ERI study, commissioned by [SPC], is the onlyepidemiologic study using case controls and cohorts that has sought to determine whether acausal relationship exists between Parlodel® and stroke. This study reviewed hospital records of280,096 postpartum women. Out of a total of ten postpartum strokes in this population, only oneoccurred in a woman who had taken Parlodel®.” Siharath, 131 F. Supp.2d at 1356.
This study also concluded that “although there is a positive association between bromocriptine and seizures among those [patients in the study] who alsoreceived [another drug not at issue in the Siharath or Rider cases], there is a weak negativeassociation among those who did not receive [the other drug].” Id. at 1357 (emphasis inoriginal).
Andrea D. Witlin, et al., Postpartum Stroke: A Twenty-Year Experience: This study concluded that postpartum women who take bromocriptine are eighttimes less likely to experience stroke than other patients who are exposed to the drug. Id. at1358; see also Brief of Defendants-Appellees at 9 n. 6 (noting that Witlin studied approximately130,000 women).
HCIA, Inc., Postpartum Complications and Parlodel® (October 1995): This study was commissioned by SPC and “analyzed 533,816 delivery records from 128hospitals” to track postpartum complications to correlate them with Parlodel® use. Siharath,131 F. Supp.2d at 1357. “For both preexisting and non-preexisting hypertensive women, thestudy concluded that there existed a negative association between bromocriptine … andhypertension.” Id. (emphasis in original).
R.M.C. Herings, et al, Bromocriptine and Suppression of Postpartum Lactation, Pharmacy World and Science 17:133-37 (1995): “In this study,investigators compared hospital admission and drug use of 2,130 women to identify the existenceof ischemic heart disease, hypertension, and cerebrovascular events such as stroke before, duringand after use of Parlodel® for postpartum lactation. The study found that no women whatsoeverwere admitted to hospitals for any of these conditions during the presumed exposure period or inthe following two months.” Siharath, 131 F. Supp.2d at 1357. “The authors noted thatsignificant adverse events may ‘wrongly be associated with bromocriptine use’ because of theconfounding fact that pregnancy itself causes such events.” Brief of Defendants-Appellees at 9n.8.
Case reports and dechallenge / rechallenge tests
SPC’s Drug Monitoring Center concluded that bromocriptine “probably” caused an ischemic stroke (i.e., strokes caused by the lack of blood flow to the brain)in one case report. The patient in this case, however, was 62 years old, was not postpartum, hadlongstanding hypertension, and suffered from a life-threatening pituitary disease that admittedlycan lead to stroke. Siharath, 131 F. Supp.2d at 1360.
In another case, a 23 year-old German woman who took Parlodel® for three months suffered from hypertension and cerebellar incoordination. The patient in thiscase, however, was not postpartum and was taking Parlodel® to treat a pituitary condition thatcan itself lead to hypertension and incoordination. Moreover, the patient had multiple sclerosis,the classic symptom of which is cerebellar incoordination. Id. A 22 year old French woman who took Parlodel® to suppress PPL later developed hypertension and convulsions. The case report reveals, however, that she washypertensive before delivery, that her hypertension decreased after taking Parlodel®, and thatshe suffered from postpartum eclampsia, which can lead to seizures and stroke. Id. In another case, a 20 year old Arkansas woman took Parlodel® to suppress PPL and later developed hypertension. Some plaintiffs’ experts have emphasized thedechallenge aspect of this case (i.e., the discontinuation of the drug treatment), because thepatient’s hypertension improved after she was taken off Parlodel®. Id. at 1361. Defense expertshave minimized this aspect of the case, however, since the patient’s hypertension continued forfour or five days after she was taken off the drug. Id. Animal Studies
Bertholet and Sutter study of the “hind limb” of a dog: “This study attempted to determine, by injecting bromocriptine into the hind leg of a dog, whetherbromocriptine acts as a vasoconstrictor and, if so, at what point vasoconstriction takes place.”Id. at 1367-68. Plaintiffs maintain that this study demonstrates that Parlodel® is avasoconstrictor, but admit that it does not show that Parlodel® causes stroke. Id. at 1368.
Defendants question the relevance and reliability of this study because vasoconstriction occurredat 1,250 times the human dosage and because the interspecies variation between dog arteries andhuman arteries is unknown. Id. Carotid artery study: This study “attempted to determine the effects of bromocriptine on the carotid artery of a dog.” Siharath, 131 F. Supp.2d at 1368.
Plaintiffs contend that this study demonstrated that bromocriptine is a vasoconstrictor, but admitthat it does not demonstrate that the drug causes stroke. Id. Defendants contend that the studyonly demonstrated an increase in vascular resistance, which could have any number of causesother than vasoconstriction. Id. Defendants also contend that the study used a flawed method torecord blood flow. Id. Pithed animal studies: Plaintiffs contend that a number of studies conducted on pithed rats, mice, dogs, cats and rabbits show vasoconstriction caused bybromocriptine so severe that the tails of rats and mice and the ear margins of dogs becamedeprived of blood and fell off. Id. at 1369. Defendants criticize the method of these studies onthe grounds that pithed animals are particularly susceptible to changes in blood pressure becausethe area of the brain controlling the cardiovascular system has been obliterated. Id. FDA Action
By 1983, SPC and the FDA began receiving reports relating Parlodel® to “hypertension and related effects.” Kuhn v. Sandoz Pharmaceuticals Corp., 14P.3d 1170, 1174 (Kan. 2000). In 1989, the FDA asked SPC to voluntarily withdraw Parlodel®from the market. Id. at 1175. SPC declined to do so, and on August 17, 1994, the FDA issued apaper stating that it had initiated procedures for withdrawing approval of Parlodel® to preventPPL. Id. The next day, SPC withdrew the Parlodel® indication for the prevention of lactation inthe United States. Id. On August 24, 1994, the FDA issued the following statement: Since approval of bromocriptine for use in preventingphysiological lactation, FDA has received a number of reports ofserious and life-threatening adverse experiences (hypertension,seizures, and CVA’s [cardiovascular accidents]) associated withthe use of bromocriptine for this indication. FDA believes thatthe number of women experiencing such adverse experiencesmay well be greater than those reported to the FDA. The aboveevidence, in aggregate, calls into question bromocriptine’s safetyfor use in postpartum women given that bromocriptine may beresponsible for hypertension, seizures and CVA’s in a small but significant number of patients. … Accordingly, the Directorconcludes that the potential risks associated with the use ofbromocriptine for the prevention of physiological lactationoutweigh its limited benefits and bromocriptine is no longershown to be safe for use in preventing physiological lactation.
Siharath, 131 F. Supp.2d at 1365-66 (quoting 59 Fed.Reg. 43347, 43351 (Aug. 24, 1994)).
Effects of Other Ergot Alkaloids
“Plaintiffs’ argument in this regard is as follows: Parlodel®’s active ingredient is bromocriptine. Bromocriptine is a semi-synthetic ergot alkaloid. Ergotalkaloids are a class of drugs that can cause vasoconstriction. Vasoconstriction can lead tohypertension, seizures and ischemic strokes. Hemorrhages are another type of stroke, so it ispossible that they also are caused by Parlodel®.” Siharath, 131 F. Supp.2d at 1364.
Defendants argue that there is no evidence that ergot alkaloids cause hemorrhagic (i.e., bleeding) strokes, only ischemic (i.e., lack of blood to brain) strokes,and that no epidemiological studies or learned treatises link ergot alkaloids to hemorrhagicstrokes. Id. at 1365. Defendants also argue that bromocriptine differs significantly from otherergot alkaloids in its chemical structure and in that it can act as either a vasoconstrictor or avasoldilator, depending upon vascular tone. Id. at 1364.
Learned Treatises
“In the Physician’s Desk Reference … there is well- documented evidence of strokes in women receiving bromocriptine for postpartum breast milksuppression…” Id. at 1369 (quoting M.D.B. Stephens, ed., Detection of New Adverse DrugReactions 383).
“Drug interactions and use after pregnancy can induce life- threatening responses,” and “[s]evere HT [hypertension] with stroke has been reported after usefor suppression of lactation.” Siharath, 131 F. Supp.2d at 1369 (quoting Williams & Wilkins,Ellenhorn’s Medical Toxicology: Diagnosis and Treatment of Human Poisoning 26 tbl. 1-34,867 & 868).
“Adverse effects [for bromocriptine] which occur more rarely, but which are serious … include unusual and continuing headache, vision changes,seizures or strokes.” Siharath, 131 F. Supp.2d at 1369 (quoting USP, Material Safety Data Sheet(1995)).
“Many postpartum patients who developed stroke and/or seizures in association with bromocriptine therapy complained of constant and often progressively severe headaches hours prior to the acute event.” Siharath, 131 F. Supp.2d at 1370(quoting American Hospital Formulary Service Drug Information 2560 (1995)).
Dukes, Meyler’s Side Effects of Drugs (13th ed. 1996): Discusses bromocriptine but fails to state that Parlodel® causes stroke. Siharath, 131 F. Supp.2dat 1370 (noting that this treatise is edited by one of plaintiffs’ experts, Dr. Dukes); see also Briefof Defendants-Appellees at 9-10.
Kittner, et al., “Pregnancy and the Risk of Stroke,” 33 New Eng. J. Med. 768-74 (1996): Pregnancy is a risk factor for stroke. Specifically, Kittner found anincreased relative risk of 28.3 for intracerebral hemorrhagic stroke in postpartum women. SeeBrief of Defendants-Appellees at 7, 37.
Ellenhorn, Ellenhorn’s Medical Toxicology: Diagnosis and Treatment of Human Poisoning (2d ed. 1997): This text includes a table summarizing thedisparate properties of different ergots. Brief of Defendants-Appellees at 40. It also reports thevasoconstrictive property of bromocriptine as zero. Id. HISTORY OF PARLODEL® LITIGATION
Federal district courts excluding plaintiff’s expert causation testimony
Brumbaugh v. Sandoz Pharmaceuticals Corp., 77 F. Supp.2d 1153,
1157 (D. Mont. 1999). After being brutally attacked by her boyfriend, plaintiff delivered a baby
by cesarean section. 77 F. Supp.2d at 1155. Plaintiff’s treating physician prescribed Parlodel®
for plaintiff to reduce her breast engorgement. Id. Plaintiff suffered a seizure shortly thereafter.
Id. In support of plaintiff’s case against SPC, plaintiff’s expert relied on anecdotal case reports
and his theory that drugs similar to Parlodel® are vasoconstrictors, but he did not rely on any
epidemiological studies. Id. SPC, on the other hand, relied on five studies, two of which were
epidemiological, that showed no statistically significant relationship between Parlodel® and
strokes. Id. at 1155. The court was persuaded that Parlodel® caused vasodilation rather than
vasoconstriction. Id. The court held that plaintiff’s expert’s opinions were too unreliable and
speculative to be admissible. Id. at 1157. Accordingly, the court granted defendant’s Daubert
motion and defendant’s motion for summary judgment. Id.
Caraker v. Sandoz Pharmaceuticals Corp., 172 F. Supp.2d 1046 (S.D.
Ill. 2001); final order entered 2001 U.S. Dist. LEXIS 22397 (Nov. 21, 2001). Plaintiff
suffered an intracerebral hemorrhage after taking Parlodel® to suppress lactation after child
birth. 172 F. Supp.2d at 1047. Plaintiff’s experts attempted to link the two events. Id. Noting
that the court “is not required to simply ‘take the expert’s word for it,’” the court excluded the
testimony of plaintiffs’ experts as scientifically unreliable. Id. (citation omitted) (stating that the
court must “rigorously scrutinize” the sufficiency of the data relied upon by the expert, the
reliability of the principles and methods employed by the expert and the reliability of the expert’sapplication of the principles and methods to the facts of the case).
The court noted that plaintiff admitted that most of her data, including case reports, animal studies, chemical inferences, FDA actions and flawedepidemiological studies, individually would not show that Parlodel® causes intracerebralhemorrhage, but argued that cumulatively, the data shows such a causal link. Id. at 1048. Afteranalyzing the evidence relied upon by plaintiff’s experts piece by piece and in the aggregate, thecourt concluded that plaintiff’s experts’ methodology was questionable and based on insufficientdata. Id. at 1053.
The court was suspect of plaintiff’s experts’ use of the differential diagnosis method. For such a method to be effective in the realm of science “as opposed to itsuse by treating physicians in the practice of medicine out of necessity,” the experts must rely onsufficient and reliable data to rule in and rule out potential causes. Id. The court determined that the epidemiological studies relied upon by both parties were flawed. 2001 U.S. Dist. LEXIS 22397 at *15-17. The court, however,“imposes no absolute epidemiology requirement or any other requirement, except reliability andrelevance.” Id. at *18.
Glastetter v. Novartis Pharmaceuticals Corp., 107 F. Supp.2d 1015,
1044-45 (E.D. Mo. 2000), aff’d 252 F.3d 986, 989 (8th Cir. (per curiam), petition for reh’g
and reh’g en banc denied (8th Cir. 2001).
Plaintiff and her husband sued Novartis Pharmaceuticals Corp.
after plaintiff suffered an intracerebral hemorrhage following 13 days of Parlodel® drug therapyafter the birth of her second child. 107 F. Supp.2d at 1017. Plaintiffs’ experts, Drs. Kulig andPetro, employed the differential diagnosis method, which involved ruling out other possiblecausative factors, and concluded that Parlodel® caused plaintiff’s intracerebral hemorrhage. Id.
at 1019-28.
Defendant moved in limine to exclude plaintiffs’ causation experts and for summary judgment. Id. at 1016. After a Daubert hearing, the court concluded thatplaintiffs’ causation evidence failed the test set forth in Daubert for scientific reliability andawarded summary judgment to defendant. Id. Plaintiffs’ experts conceded that their differential diagnosis method “is not helpful in assessing general causation.” Id. at 1027. Plaintiffs then came forward withthe following evidence to support their claim that Parlodel® can cause intracerebralhemorrhages: “(1) peer reviewed articles, texts, and treatises; (2) multiple human dechallengeand rechallenge studies and reports; (3) an epidemiology study on stroke and Parlodel, whichplaintiffs admit is ‘underpowered and thus partially flawed’ …; (4) determinations by the FDAthat Parlodel is unsafe because of stroke and other vasospastic risks; and (5) allegedly hiddeninternal company admissions by Sandoz concluding that Parlodel can cause such conditions.”Id. at 1028.
The court found this evidence insufficient to establish general First, as SPC and plaintiff’s own expert noted, case reports do not establish causation. Id. at 1029-30. The court did not believe that the case reports,including dechallenge / rechallenge information, were sufficient “to establish the requisitecausation, as they fail to take into account the postpartum incidence of stroke and other factors.”Id. at 1031 (citation omitted).
Second, the court rejected findings in several published texts and articles relied upon by plaintiffs’ experts because they were “simply case reports.” Id.
at 1033-34. The court also was not persuaded by other texts and articles regarding the tendenciesof other ergot alkaloids in the same family as bromocriptine to cause hypertension, vasospasmand stroke. Id. at 1034. The court found that plaintiffs’ evidence does not establish that“bromocriptine and the other ergots have sufficiently similar physiological effects to warrantcomparison.” Id. (citation omitted). Moreover, the court noted that vasospasm is a differentinjury than intracerebral hemorrhage. Id. Third, the court found plaintiffs’ reliance on the FDA’s withdrawal of approval for bromocriptine for use in preventing lactation “is misplaced.” Id. at1036. The FDA did not affirmatively state that there is a connection between bromocriptine andintracerebral hemorrhage, only that bromocriptine “‘may be an additional risk factor in patientswho are already at risk for seizures and strokes.’” Id. (emphasis in original). Thus, the FDA’sstatement does not establish the reliability of plaintiffs’ experts’ causation testimony. Id. Fourth, the court determined that none of the SPC documents relied upon by plaintiffs established that Parlodel causes intracerebral hemorrhages.
Id. at 1036-37 (noting that some of the documents merely reflected case reports).
Fifth, the court found a lack of similarity between the animal studies relied upon by plaintiffs’ experts and the facts of this case. Id. at 1044. Finally, the court emphasized that although plaintiffs’ experts admitted that epidemiological evidence is the “best evidence supporting a connectionbetween a drug and an adverse effect,” they did not rely on epidemiological evidence to supporttheir position that Parlodel® causes stroke. Id. at 1042. Plaintiffs’ experts admitted that they didnot know of a statistically significant epidemiological study demonstrating an associationbetween bromocriptine or Parlodel® and stroke or intracerebral hemorrhage. Id. at 1042-43. “Inthe absence of their own epidemiological evidence supporting the conclusions of their expertsthat Parlodel can cause [intracerebral hemorrhage], the best plaintiffs can do is attack defendant’sstudies. However, … such attacks do not ‘meet the law’s requirements,’ because plaintiffs ‘mustcome forward with reliable scientific evidence of [their] own to defeat a summary judgmentmotion when [the] case is based on the expert’s proof.’” Id. at 1044 (citation omitted).
Thus, the court held that “in the absence of reliable scientific evidence supporting their causation testimony, plaintiffs’ experts must be excluded” anddefendant is entitled to summary judgment Id. at 1045.
Hollander v. Sandoz Pharmaceuticals Corp., 95 F. Supp.2d 1230, 1238-
39 (W.D. Okla. 2000) (appeal pending). Plaintiff suffered a hemorrhagic stroke after ingesting
Parlodel®. 95 F. Supp.2d at 1232. In support of their theory that Parlodel® can cause strokes,
plaintiff’s experts relied on determinations by the FDA that the safety of Parlodel® had not been
demonstrated, a finding of causation by a Kentucky jury, incidents of challenge/de-challenge/re-
challenge data, case reports, the effects of other ergots, animal studies and epidemiological
studies. Id. at 1234-35. After a Daubert hearing, the court excluded testimony of Dr. Kulig and
three other experts as unreliable and granted defendant’s motion for summary judgment. Id. at
1234.
Defendant challenged plaintiff’s experts causation evidence on the grounds that plaintiff’s experts: “(1) cannot explain how Parlodel causes vasoconstriction, whichthen causes hypertension and strokes; (2) improperly rely on anecdotal case reports and temporalproximity, which do not constitute scientifically reliable bases for their opinions; (3) improperlyreason that because some ergot alkaloids, which are in the same class as bromocriptine may alsocause hypertension; and (4) improperly rely on animal studies that are too different from thefacts presented by this case to be reliable.” Id. at 1235. The court agreed with defendant. Id. The court rejected case reports as a scientific basis for establishing causation. Id. at 1237. Likewise, the court rejected plaintiff’s expert’s reliance on FDAdeterminations because the FDA’s standard of proof is lower than that required by courts. Id. at1234, n.9, 1237. As for the epidemiological studies relied upon by plaintiff’s experts, the courtfound that none showed a statistically significant link between Parlodel® and strokes. Id. at1236.
The court did not conclude that expert testimony must be based on epidemiology to be admissible. Id. at 1237, n.22. Instead, the court concluded that “due to theabsence of supportive epidemiological evidence, the differences between bromocriptine and theother ergot alkaloids, the dissimilarity of the animal studies, and the unreliability of case reports,the data and methods relied on by the plaintiffs’ experts do not furnish a scientifically valid basisfor their conclusion that Parlodel causes strokes.” Id. at 1238-39.
Siharath v. Sandoz Pharmaceuticals Corp., 131 F. Supp.2d 1347 (N.D.
Ga. 2001). With this decision, the United States District Court for the Northern District of
Georgia, Atlanta Division, addressed defendants’ motions to exclude and for summary judgment
in two cases: Siharath v. Sandoz Pharmaceuticals Corp. and Rider v. Sandoz Pharmaceuticals
Corp.
Facts of Siharath: In 1989, Ms. Siharath, age 17, delivered her second child, decided not to breast feed and started taking Parlodel® for the prevention oflactation. Siharath, 131 F. Supp.2d at 1349. Shortly thereafter, Ms. Siharath suffered threeseizures and a subarachnoid hemorrhagic stroke. Id. Plaintiff’s treating physicians were unable to diagnose the cause of her stroke and did not find evidence of cerebral vasoconstriction. Id.
None of Ms. Siharath’s treating physicians concluded that Parlodel® was a cause of her stroke,and none of them testified as experts on her behalf in the litigation. See Brief of Defendants-Appellees at 4-5. Ms. Siharath filed suit in 1995 claiming that her 1989 ingestion of Parlodel®caused her stroke. Siharath, 131 F. Supp.2d at 1349.
Facts of Rider: In 1993, Mrs. Rider, age 39, had a child, decided not to breast-feed, and began taking Parlodel® to suppress lactation. Id. at 1350. Shortlythereafter, she suffered an intracranial hemorrhagic stroke. Id. After testing, Mrs. Rider’sdoctors determined that vasospasm caused her stroke and that her significant history of smoking,her hemorrhagic AVM, abnormal vascularity or left-sided structural lesion were all possibleetiologies of her stroke. Id.; Brief of Defendants-Appellees at 5. None of Mrs. Rider’s treatingphysicians concluded that Parlodel® was a cause of her stroke, and none of them testified asexperts on her behalf in the litigation. Brief of Defendants-Appellees at 5.
Plaintiffs’ experts opined that Parlodel®’s active ingredient bromocriptine, as an ergot alkaloid, causes hemorrhagic stroke by vasoconstriction in somepostpartum women. Siharath, 131 F. Supp.2d at 1355. Defendant moved in limine to excludeplaintiffs’ expert witnesses and for summary judgment on the ground that plaintiffs could notprove causation. Id. at 1350. Defendant argued that plaintiffs’ experts’ testimony fails to meetDaubert’s requirements because plaintiffs’ experts: “(1) have failed to provide any evidence,either published or unpublished, that Parlodel® increases one’s risk of stroke; (2) rely onuncontrolled and unreliable spontaneous reports and anecdotal case reports as the basis for theiropinions; and (3) cannot show that their opinions have an acceptable error rate or are otherwisegenerally accepted.” Id. at 1352.
After a three day evidentiary hearing, the district court excluded plaintiffs’ experts’ opinions as neither scientifically reliable nor relevant and granted summaryjudgment to defendants. Id. at 1350, 1355-56.
The court pointed out that epidemiological studies are the primary method to prove causation in toxic tort cases. Id. at 1356. The epidemiological studiesinvestigating a possible link between Parlodel® and strokes, however, did not show astatistically significant relationship between the two and therefore did not support plaintiffs’causation theory in this case. Id. at 1356-58. The court concluded that “the lack ofepidemiological studies supporting Plaintiffs’ claims creates a high bar for Plaintiffs to surmountwith respect to the reliability requirement, but it is not automatically fatal to Plaintiffs’ case.” Id.
at 1358 (stating that if other reliable scientific knowledge exists, plaintiffs can overcome thishurdle).
The court held that the anecdotal case reports relied upon by plaintiffs’ experts did not involve cases where Parlodel® caused hemorrhagic stroke inpostpartum women, nor do they satisfy scientific method requirements sufficient to establishcausation. Id. at 1359. In addition, the court concluded that the medical treatises relied upon byplaintiffs’ experts were nothing more than case reports and similarly were unreliable. Id. at1370.
The court rejected plaintiffs’ experts’ comparisons of bromocriptine to similar drugs because they raised “serious questions of ‘fit.’” Id. at 1363.
The court also would not permit plaintiffs to rely on the FDA’s determinations that the risks of Parlodel® outweighed its benefits in suppressinglactation. Id. at 1366. The FDA’s risk-utility standard is lower than the standard of proofrequired in tort actions. Id. The court determined that none of the animal studies relied upon by plaintiffs’ experts establish that Parlodel® causes stroke in animals or humans. Id. at1367. The animal studies also had flaws that “prevent any conclusion that they ‘fit’ withPlaintiffs’ causation theory.” Id. The court concluded that none of plaintiffs’ evidence, individually or collectively, establishes a prima facie case that Parlodel® causes stroke. Id. at1370. The court held that as one of plaintiffs’ experts previously wrote, “one cannot lumptogether lots of hollow evidence in an attempt to determine what caused a medical harm.” Id. at1371.
The court rejected plaintiffs’ arguments that they “have employed the same methodology as is applied by doctors throughout the world in their clinicalpractices” and that plaintiffs’ experts used “the best methodology available.” Id. at 1372. Thecourt found that their opinions were not supported by reliable scientific evidence and insteadseemed based more on their personal opinions. Id. The court noted that defendant “continually researched” whether there was an association between Parlodel® and strokes and had not establishedcausation. Id. at 1373. The court did not want to create “an unintended disincentive forpharmaceutical companies to engage in ongoing research as to their products’ safety” bylowering the Daubert standard based on anecdotal evidence. Id. Federal courts of appeals excluding plaintiff’s expert causation testimony
Glastetter v. Novartis Pharmaceuticals Corp., 107 F. Supp.2d 1015,
1044-45 (E.D. Mo. 2000), aff’d 252 F.3d 986, 989 (8th Cir. (per curiam), petition for reh’g
and reh’g en banc denied (8th Cir. 2001). Plaintiff and her husband filed suit after Mrs.
Glastetter suffered an intracerebral hemorrhage while taking Parlodel® to suppress lactation after
child birth. 252 F.3d at 987-88. The district court excluded plaintiff’s experts’ proposed
causation testimony as not scientifically valid and granted defendant summary judgment. Id. at
988. On appeal, the Eighth Circuit affirmed the decision of the district court.
The Eighth Circuit held that plaintiffs’ experts “failed to produce scientifically convincing evidence that Parlodel causes vasoconstriction.” Id. at 989. Plaintiffs’experts relied on published case reports, human re-challenge/de-challenge data, medical treatises,animal studies, analysis of the vasoconstrictive properties of other ergot alkaloids, defendant’s internal documents, and the FDA’s decision that the possibility of harm from Parlodel®,including the possibility of stroke and seizure in certain women, outweighed its benefits as alactation suppressant. Id. at 988-89. The Eighth Circuit agreed with the district court, however,that all of the data the experts relied on in support of their opinions were unreliable. Id. at 990-91. Specifically, the Eighth Circuit found that: none of the animal studies concluded thatintracerebral hemorrhage was associated with bromocriptine; the leading toxicology treatiseconcluded that bromocriptine does not have any vasoconstrictive properties; case reports “are notscientifically valid proof of causation”; defendant’s documents did not “admit” that Parlodel®causes intracerebral hemorrhage; and the FDA’s standard for evaluating pharmaceuticals isdifferent than the court’s causation standard. Id. The Eighth Circuit concluded that “[v]iewed in isolation, [plaintiffs’] different pieces of scientific evidence do not substantiate her experts’ conclusion thatParlodel can cause [intracerebral hemorrhages]. Likewise, we do not believe that the aggregateof this evidence presents a stronger scientific basis for [plaintiffs’] supposition that Parlodel cancause [intracerebral hemorrhages].” Id. at 992.
The Eighth Circuit stated that contrary to plaintiffs’ assertion, the district court did not require that she introduce epidemiological evidence of causation to satisfyDaubert’s requirements. Id. The parties agreed that the epidemiological evidence wasinconclusive regarding links between Parlodel® and intracerebral hemorrhages. Id. The EighthCircuit noted that the “absence of epidemiological evidence did not doom [plaintiffs’] case,” but“epidemiological evidence might have assisted [plaintiffs] in establishing causation, and thus itsabsence limited the available tools with which she could prove causation.” Id. Eleventh Circuit: Siharath v. Sandoz Pharmaceuticals Corp. and Rider
v. Sandoz Pharmaceuticals Corp. (appeal of exclusion of plaintiff’s causation evidence pending)(oral argument on May 2, 2002).
Tenth Circuit: Hollander v. Sandoz Pharmaceuticals Corp. (appeal of
exclusion of plaintiff’s causation evidence pending).
State court Parlodel® cases excluding plaintiff’s expert causation testimony
Revels v. Novartis Pharmaceuticals Corp., No. 03-98-00231-CV, 1999
WL 644732 (Tex. App. Aug. 26, 1999) (unpublished op.), reh’g denied (Tex. App.), petition
for review denied (Tex. 2000). In this case, plaintiff suffered heart failure and sudden death
after taking Parlodel® to suppress lactation after the birth of her third child. 1999 WL 644732 at
*1. Plaintiffs offered seven experts to testify that Parlodel® could cause coronary artery
vasospasm in the general population and caused plaintiff’s coronary artery vasospasm in this
case. Id. Plaintiffs’ experts relied on case reports, adverse event reports submitted to the FDA,
one report of a bromocriptine challenge/re-challenge test, FDA findings, and analysis of
structurally similar compounds to support their causation opinion. Id. at *2. The trial court
excluded plaintiffs’ experts’ testimony as unreliable under Texas’s Daubert analog, and the
Court of Appeals of Texas affirmed. Id. at *6. The court held that “[w]hile the case reports
illustrate an association between adverse drug experiences and Parlodel, the [Texas] supremecourt has clearly warned that such an association does not equate to causation.” Id. at *5.
Federal district courts admitting plaintiff’s expert causation testimony
Brasher v. Sandoz Pharmaceuticals Corp., 160 F. Supp.2d 1291 (N.D.
Ala. 2001). Plaintiffs suffered strokes shortly after delivering children and taking Parlodel® to
stop postpartum lactation. The court concluded that the proffered experts opinions that
plaintiffs’ strokes were caused by their ingestion of Parlodel® were sufficiently reliable and
admissible. 160 F. Supp.2d at 1299 (denying defendants’ motion for summary judgment).
Defendants argued that “absent a scientifically appropriate epidemiological study showing an increased risk of stroke associated with Parlodel use,plaintiffs’ experts’ opinions are nothing more than unscientific speculation.” Id. The court held that the proffered opinions that plaintiffs’ strokes were caused by “cerebral arterial spasms arising from the vasoconstrictive effects of the Parlodelboth women were taking” were “based on ‘good grounds’ tied to the scientific method” and“possess sufficient evidentiary reliability (that is, trustworthiness) that a jury should be allowedto consider the opinions in the determination of the facts of this case.” Id. at 1296.
The court stated that plaintiffs’ experts relied on: (1) animal studies that showed that Parlodel® has vasoconstrictive properties; (2) the fact that otherergot alkaloids are known to cause vasoconstriction; (3) case reports suggesting that womentaking Parlodel® have suffered hypertension, stroke and myocardial infarction, including thosereported to the FDA; and (4) several treatises and medical textbooks, which identifybromocriptine as a risk factor for stroke. Id. at 1296. The court also noted that scientistsroutinely use such evidence to draw conclusions. Id. “Although it is true that none of these bits of evidence establish conclusively that Parlodel can cause vasoconstriction and vasospasm, taken togetherthey present a compelling picture, one which can support a reasonable scientific inference.” Id. The court further held that “[u]nquestionably, epidemiological studies provide the best proof of the general association of a particular substancewith particular effects, but it is not the only scientific basis on which those effects can bepredicted. In science, as in life, where there is smoke, fire can be inferred, subject to debate andfurther testing.” Id. The court found that the fact that the lack of epidemiological studies is dueto the difficulty and danger in structuring such a study and the rarity of strokes in women ofchild-bearing years. Id. at 1297.
The court decided to leave for the jury to determine whether defendants’ alternative explanations for plaintiffs’ strokes, such as history of smoking,obesity, and family history of stroke, have merit. Id. at 1299.
The court in Brasher did not feel obligated to follow the Eighth Circuit Court of Appeals’ recent decision affirming the grant of summary judgment infavor of SPC in Glastetter v. Novartis Pharmaceuticals Corp., 252 F.3d 986 (8th Cir. 2001). Id.
at 1299 n.17. The Brasher court noted that the Glastetter court relied on an outdated treatise andthat the more recent edition of the treatise indicated a relation between bromocriptine and seizureand stroke. Id. Eve v. Sandoz Pharmaceutical Corp., 2001 U.S. Dist. LEXIS 4531 (S.D.
Ind. 2001). The court concluded, based on the written submissions of the parties, that the
testimony of plaintiffs’ experts Drs. Kulig and Petro are “scientifically reliable and therefore
admissible.” 2001 U.S. Dist. LEXIS at *55.
In this case, Drs. Kulig and Petro relied upon case reports, treatises, texts and journals, human re-challenge studies, inferences based on similarities of otherchemical compounds, adverse drug reports submitted to the FDA, FDA regulatory findings,defendants’ internal documents and animal studies to support their causation opinions. Id. at*60-61, *66-73. The court held that the “cumulation of this evidence satisfies the Daubertrequirements of scientific reliability.” Id. at *61.
The court stated that although plaintiffs’ experts admitted that they do not have any epidemiological evidence to support their opinions, “Daubert simply requiresreliable evidence.” Id. at *57.
The court noted that the “Seventh Circuit chastises trial court judges who act as overly aggressive gatekeepers.” Id. Globetti v. Sandoz Pharmaceuticals, Corp., 111 F. Supp.2d 1174 (N.D.
Ala. 2000). Plaintiff’s experts opined that plaintiff’s myocardial infarction was caused by her
ingestion of Parlodel® after delivery of her sixth child. 111 F. Supp.2d at 1176. Plaintiff’s
experts relied on animal studies, case reports, adverse drug reaction reports submitted to the
FDA, medical textbooks, the results of a 1993 de-challenge/re-challenge experiment, and the
“generally accepted notion in the medical community” that Parlodel is a risk factor for
myocardial infarction because of its vasoconstrictive effects. Id. Defendant argued in support of
its Daubert challenge that absent an epidemiological study showing an increased risk of
myocardial infarction associated with the use of Parlodel®, plaintiff’s experts’ opinion is sheer
speculation. Id. The court held that plaintiff’s experts’ causation opinion “is based on ‘good
grounds’ tied to the scientific method” and “possesses sufficient evidentiary reliability… that a
jury should be allowed to consider it.” Id. at 1177.
“Although defendant is correct that there is no epidemiological study showing an increased risk of [myocardial infarction] associated with bromocriptine, thereis more than adequate evidence of a scientific nature from which a reliable conclusion can bedrawn about the association.” Id. at 1179. Thus, the court “inferred” that plaintiff’s myocardialinfarction was caused by her ingestion of Parlodel®. Id. The court agreed with plaintiffs that an epidemiological study of the association between Parlodel® and myocardial infarction is not practical because of the rarityof myocardial infarctions among postpartum women. Id. Whether there are alternative causes of plaintiff’s myocardial infarction is a question that goes to the weight to be given plaintiff’s experts, not theadmissibility of their opinions. Id. The court explained why it disagreed with other courts that had The court believes that in those cases the Daubert standard was appliedincorrectly, creating much too high a standard of admissibility. Both ofthese cases seem to equate Daubert’s reliability standard with scientificcertainty, which is far from what the Supreme Court intended in Daubert.
[The other courts] failed to recognize that Daubert does not require, oreven allow, the trial court to determine the scientific ‘correctness’ orcertainty of the evidence, but only that the facts from which the opinion isinferred are themselves sufficiently reliable.
Id. at 1180 (finding plaintiff’s experts’ opinions scientifically reliable and denying defendant’smotion for summary judgment on medical causation).
Kittleson v. Sandoz Pharmaceuticals Corp., Civil No. 98-2277 (N.D.
Minn. March 3, 2000) (unpublished). The court summarily held that plaintiffs presented
sufficient causation evidence to allow the case to withstand defendants’ Daubert challenge and
proceed to the jury.
Federal courts of appeals admitting plaintiff’s expert causation testimony
State court Parlodel® cases admitting plaintiff’s expert causation testimony
Kuhn v. Sandoz Pharmaceuticals Corp., 14 P.3d 1170 (Kan. 2000).
After the birth of her child, plaintiff received one Parlodel® tablet, and shortly thereafter becameill, went into a coma and died. 14 P.3d at 1173-74. Plaintiffs’ experts’ conclusions were basedon the medical methodology of differential diagnosis. Id. at 1177. At trial, the court foundplaintiffs’ experts’ causation opinions unreliable, excluded them and granted SPC summaryjudgment. Id. at 1173. The Supreme Court of Kansas reversed the trial court’s award ofsummary judgment in favor of SPC, finding that the trial court erred in applying the Frye test toexclude plaintiffs’ experts’ testimony on causation. Id. Applying a de novo standard of review, the Supreme Court of Kansas determined that Frye was not applicable in this case because it requires a showing thatthe basis for an expert’s opinion is generally accepted as reliable within the expert’s field before it may be admissible. Id. at 1178-79. The court found that the experts in this case were offering“pure opinion,” which is distinguishable from testimony based on scientific method or technique,and therefore the Frye test did not apply. Id. at 1179-81. “Pure opinion” is “an expert opiniondeveloped from inductive reasoning based on the expert’s own experience, observation orresearch.” Id. at 1179. The court held that “pure opinion is tested by cross-examination of thewitness.” Id. The Supreme Court of Kansas further found that the trial court erred when it required plaintiffs to prove both general and specific causation. Id. at 1184. TheSupreme Court of Kansas held that a finding of general causation is not necessary. Id. The courtreasoned: ‘Cases that have not imposed this requirement [general causation] typicallyinvolve injuries that may be placed in the ‘sporadic accident model of tort law.’In [these] cases, where only a single plaintiff or a few plaintiffs have allegedlysuffered an injury due to some exposure, a medical doctor will be permitted torender an opinion as to whether the exposure caused the plaintiff’s injury solelyon an examination of the plaintiff and a differential diagnosis of the source of theplaintiff’s injury, sometimes supplemented with toxicological evidence. …. Inmany of these cases there is relatively little epidemiological data available andthe courts are reluctant to burden ‘first plaintiffs’ with the task of usingepidemiology to prove general causation.
Id. at 1184-85 (citations omitted). Thus, the Supreme Court of Kansas reversed the decision ofthe trial court excluding the experts and remanded the case with directions. Id. at 1185.
Sandoz Pharmaceuticals Corp. v. Roberts, No. 89-CI-653 (Ky. Ct. App.
Aug. 9, 1996). Plaintiff was hospitalized for pregnancy-induced hypertension. Id. at 2. After
she delivered her baby, her blood pressure returned to normal. Id. Although Parlodel® was not
indicated for plaintiff because of her previous hypertension, it was administered pursuant to her
doctor’s standing orders at the hospital. Id. In addition, plaintiff had suffered adverse effects
from the drug ten months earlier when she had taken it for infertility. Id. Shortly after taking
Parlodel® postpartum, plaintiff suffered a stroke and ended up severely disabled. Id. at 3. After
the trial of this case, the jury returned verdicts against SPC and awarded plaintiff over
$1,000,000 in compensatory and $1,000,000 in punitive damages. Id. at 2, 4. On appeal, SPC
contends that appellees failed to establish legal causation. Id. at 2. The Kentucky Court of
Appeals affirmed the judgment of the trial court. Id.
Plaintiff’s experts relied on their review of plaintiff’s medical records, the depositions of other witnesses, epidemiological studies, peer-reviewed medicalarticles and their own professional experiences with Parlodel®. Id. at 8-9. The Kentucky Courtof Appeals held that “the information which they relied upon is of a type reasonably relied uponby experts in the medical field when forming opinions.” Id. at 9 (citing Glaser v. ThompsonMedical Co., Inc., 32 F.3d 969 (6th Cir. 1994) (where the article and studies relied upon by theexperts were “not definitive or conclusive,” but they were data upon which medical experts reasonably rely on a daily basis). Thus, the court held that the trial court did not abuse itsdiscretion by permitting them to form opinions based on this information. Id. The Kentucky Court of Appeals held that “legal causation may be established by such circumstantial evidence as would permit a jury to reasonably infer that thedrug was a legal cause of the plaintiff’s injury.” Id. at 4.
The Kentucky Court of Appeals determined that Daubert was not controlling in this case and applied Kentucky Rule of Evidence 702. Id. at 6. In applying thatrule, the court concluded that the experts’ scientific testimony assisted the jury in determiningthe cause of plaintiff’s stroke, the experts were qualified, and their opinions were scientificallyreliable. Id. at 6-7.
11th CIRCUIT APPEAL OF SIHARATH / RIDER:
Brief of Plaintiffs-Appellants
Plaintiffs-Appellants contend that the standard of review for the district court’s exclusion of evidence is de novo because the district court misapplied Daubert. See Briefof Plaintiffs-Appellants at 15. Compare General Electric v. Joiner, 522 U.S. 136, 141 (1997)(“abuse of discretion is the proper standard of review of a district court’s evidentiary rulings.”).
With respect to the merits, Plaintiffs-Appellants first argue that the district court established a “statistics plus a checklist” standard for the admissibility of expert testimonyregarding medical causation, which standard is beyond what science requires and contrary to thelaw. See Brief of Plaintiffs-Appellants at 17-18, 28-32.
According to Plaintiffs-Appellants, the Daubert Trilogy requires only that an expert provide “an empirically supported rational explanation within the bounds ofnormal scientific discourse.” Id. at 21.
Plaintiffs-Appellants allege that their experts satisfied the Daubert standard by relying on empirical evidence in the form of: i) dechallenge / rechallenge tests onten patients; ii) the ERI epidemiological study showing a relative risk of 8.4; iii) animal studiesshowing that bromocriptine can cause either vasodilation or vasoconstriction; iv) case reports;and v) the chemical structure of ergot alkaloids. See id. at 25-28 and n.26.
In addition, Plaintiffs-Appellants maintain that the district court misapplied the Daubert Trilogy when it imposed a “statistics plus a checklist” standard whichwould require all of the following for admissibility: i) peer-reviewed epidemiological literature;ii) a predictable chemical mechanism; iii) general acceptance of a causal relationship in thescientific literature; iv) a plausible animal model; and v) dozens of well-documented casereports. See id. at 17-18, 28-32 (citing Siharath, 131 F. Supp.2d at 1370). They emphasize thatthe scientific method is an untidy and loosely defined process of testing and refinement akin tothe piecemeal completion of a crossword puzzle by various scientists. They argue that the adoption of such a stringent definition of scientific knowledge would have resulted in therejection of the work of Galileo, Newton, Einstein and Benjamin Franklin. See Brief ofPlaintiffs-Appellants at 29-32. It also would require greater intellectual rigor from scientists inthe courtroom than they apply in their daily practices, contrary to the requirements of KumhoTire.
Plaintiffs-Appellants next argue that reversal is appropriate even under a less stringent standard of review because the district court abused its discretion by makingfundamental mistakes about the facts. See Brief of Plaintiffs-Appellants at 33-42.
Plaintiffs-Appellants take issue with the district court’s conclusion that “Plaintiffs have not pointed to a single case report involving a postpartum woman whosuffered a hemorrhagic stroke.” Siharath, 131 F. Supp.2d at 1361. They cite five publishedreports of such events, some of which covered more than one case. They also point out that atleast one FDA scientist concluded in 1985 that Parlodel® causes stroke. See Brief of Plaintiffs-Appellants at 34.
According to Plaintiffs-Appellants, the district court also abused its discretion by concluding that the epidemiological evidence either shows no relationship or anegative relationship between Parlodel® and stroke. Although they acknowledge that it is notstatistically significant, they emphasize that the relative risk yielded in the ERI Study was 8.4and that the other studies were not conducted under appropriate controls. Thus, they argue thatthe epidemiological evidence “does not point away from causation,” as the district court found,but rather is either “indeterminate or perhaps a weak indication of causation.” Id. at 38.
Plaintiffs-Appellants also contend that the district court ignored seven reports of rechallenge / dechallenge testing in a total of ten patients, five of whom werepostpartum. They maintain that these data demonstrate that bromocriptine causesvasoconstriction. See id. at 39-42.
Lastly, Plaintiffs-Appellants imply that the district court abused its discretion by “preferentially relying on exclusionary precedent” from other Parlodel® casesexcluding expert testimony. See id. at 43. They argue that the court’s reliance on earlierParlodel® cases failed to recognize that “lawyers working on a new kind of toxic tort litigationcan take several cases and several years to learn about the science and how to present it to ajudge or jury.” Id. They also emphasize that “the reasons for avoiding a rush to judgment areespecially compelling in the Parlodel[®] litigation because epidemiology . . . simply cannotprovide an answer.” Id. at 48.
Amicus Brief of Margaret A. Berger and Jerome P. Kassirer
Margaret A. Berger teaches evidence, civil procedure, science and the law, and toxic torts at Brooklyn Law School. From 1993 to 1996, she was the Reporter for theAdvisory Committee on the Rules of Evidence. She has authored chapters in both editions of theFederal Judicial Center’s Reference Manual on Scientific Evidence. See Amicus Brief of Prof.
Margaret Berger and Dr. Jerome Kassirer at 1-2.
Dr. Jerome P. Kassirer is a Distinguished Professor at Tufts University School of Medicine and a Professor Adjunct of Medicine at the Yale University School ofMedicine. He teaches clinical reasoning at both institutions and co-authored a leading textbookon that subject. From 1991 to 1999, he was the Editor-in-Chief of the New England Journal ofMedicine. See id. at 2.
In describing their interest in the case, Professor Berger and Dr. Kassirer state that they “venture no opinion on whether Parlodel[®] does cause strokes,” but believe thatthe trial court “misconceived its role” by deciding causation in a Daubert hearing, instead ofleaving the causation question for the jury. See id. at 5-6. They argue that the test the trial courtapplied for admissibility in a toxic tort case “would have the effect of making it virtuallyimpossible for any plaintiff to overcome the causation hurdle.” Id. at 6. The brief does notindicate whether Professor Berger and Dr. Kassirer are being compensated.
Professor Berger and Dr. Kassirer first argue that the district court applied a formula to decide the question of admissibility, and in so doing, both misinterpreted theDaubert Trilogy and failed to grasp “how the medical community would undertake a causalityassessment under the circumstances in these cases.” See id. at 5.
The legal premise of this syllogism is the uncontroversial notion that the Daubert Trilogy makes clear there is no legal formula for determining the admissibilityof expert evidence. The overriding principle is merely that the proffered expert address theparticular issue “with the same intellectual rigor that an expert in the relevant field would use . . .
.” Id. at 11-12.
The scientific premise of this syllogism is that there is no scientific formula within the medical community for determining whether a potentially toxic substancecauses an observed medical condition. According to Professor Berger and Dr. Kassirer, thedetermination of medical causation can depend upon a multitude of factors, such as biology,chemistry, temporal connection, the gravity of the medical condition, dechallenge andrechallenge tests, differential diagnoses, animal studies and statistical data. See id. at 13-15.
Although they recognize the strength of epidemiological studies and the fact that doctors relyheavily on such studies in toxicity cases, they emphasize that the availability and usefulness ofhuman epidemiological studies is limited. Accordingly, they argue that “[p]hysicians try toincorporate any reliable data that seem to bear on a possible causal relationship” into theirclinical decision-making. See id. at 16 (emphasis in original).
The conclusion of Professor Berger’s and Dr. Kassirer’s syllogism is that the trial court erred by deciding the admissibility of the proffered evidence using aformula, instead of determining how the medical community would make such a determination.
According to Professor Berger and Dr. Kassirer, the district court erroneously “equate[d] proofof general causation with epidemiology,” id. at 18, and in so doing, ignored Kumho Tire’sadmonition about matching “‘certain kinds of questions to certain kinds of experts.’” Id. at 18(quoting Kumho Tire). The testimony of Plaintiffs’-Appellants’ experts is admissible, theyargue, because the experts considered some kinds of evidence medical experts would use in a clinical setting to determine causation: epidemiological and toxicological evidence, case reports,and dechallenge / rechallenge tests. Id. at 20.
Professor Berger and Dr. Kassirer also argue that the district court confused the issues of admissibility and sufficiency of the evidence and imposed a stringentscientific standard of proof that is higher than the preponderance of the evidence.
Professor Berger and Dr. Kassirer contend that the district court conflated the sufficiency of the evidence standard with the admissibility standard when itrequired the Plaintiffs-Appellants to produce evidence that would allow a reasonable jury to findgeneral and specific causation to a reasonable degree of medical certainty. See id. at 22-25; seealso Siharath, 131 F. Supp.2d at 1352. They argue that the court should have first determinedwhether the proffered evidence was admissible and then determined if it was sufficient toestablish general causation, not vice versa. In proceeding otherwise, they argue, the district courtevaded the summary judgment standard and acted as fact finder instead of gatekeeper, therebyviolating the Seventh Amendment right to a jury trial. See id. at 23-24.
In addition, Professor Berger and Dr. Kassirer argue that the December 2000 amendments to Rule 702 do not impose a sufficiency standard for admissibilitybecause those amendments were adopted by a rule-making process and therefore must beconstrued narrowly. See Fed. R. Evid. 702 (allowing an expert to testify “if (1) the testimony isbased upon sufficient facts or data, (2) the testimony is the product of reliable principles andmethods, and (3) the witness has applied the principles and methods reliably to the facts of thecase”). To construe the amendments otherwise would work “a shift in the existing balance ofpower between plaintiffs and defendants, and a curtailment of the right to trial by jury.” SeeAmicus Brief of Berger-Kassirer at 27.
Lastly, Professor Berger and Dr. Kassirer maintain the district court’s requirement that scientific evidence of general causation be more than inconclusive ismuch higher than the preponderance of the evidence standard. In other words, they argue that byrequiring the plaintiffs to satisfy the scientific standard for conclusive proof, the district courtimposed a standard that is “far more demanding than the preponderance of the evidence standardgenerally applicable in civil litigation.” Id. at 28.
Amicus Brief of Michael D. Green and Sander Greenland
Michael D. Green is a professor of law at Wake Forest University. He is a co-author of the epidemiology chapter in the Federal Judicial Center’s Reference Manual onScientific Evidence (2d ed. 2000), and a co-Reporter for the American Law Institute’sRestatement (Third) of Torts: Liability for Personal Harm (Basic Principles). See Amicus Briefof Green-Greenland at 1.
Sander Greenland is a professor of epidemiology at UCLA School of Public Health, a professor of statistics at UCLA College of Letters and Science, and researchprofessor of preventive medicine at UCLA School of Medicine. He is the co-author of one of the leading texts on epidemiology and author or co-author of 230 peer-reviewed epidemiologic andstatistical articles. See id. In describing their interest in the case, Professors Green and Greenland point out that they are not being compensated for preparing their brief. Their interest is in“having courts understand the appropriate role of epidemiology in lawsuits in which the causalrole of an agent is at issue.” See id. at 1-2. They “only seek to urge certain scientific and legalprinciples be employed by this Court in deciding this appeal.” Id. Professors Green and Greenland criticize the district court for adopting a standard that either requires epidemiological evidence in toxic tort cases as a threshold matter or,at a minimum, raises the bar for proof of general causation in the absence of epidemiologicalevidence. They argue that because toxic agents differ, the evidence required to prove causationby various toxic agents must be flexible. See id. at 6-15.
According to Professors Green and Greenland, an understanding of how the biological mechanisms of an agent operate on the human body is sufficient evidence ofmedical causation (e.g., a bloody nose may be caused by a punch). In the case of toxic agents,however, the biological mechanisms by which varying diseases occur are rarely well enoughunderstood to play a substantial role in proof of causation. See id. at 6-7.
Professors Green and Greenland recognize that where a substantial body of epidemiological evidence exists, it is often the best evidence of general causation (e.g.,Benedictin, breast implant, asbestos and DES litigation). See id. at 7-9. Nevertheless, they alsorecognize potentially enormous impediments in the development of epidemiological evidence,such as the rarity of certain conditions or diseases, the ethical issues that can arise in exposingsubjects to potentially toxic agents, and the time and expense required to conduct a validepidemiological study. See id. at 5, 10-11.
Accordingly, they contend that where epidemiological evidence is understandably absent, courts should consider other evidence of causation without skepticismbased upon the absence of epidemiological evidence. See id. at 14-15.
Professors Green and Greenland next argue that because epidemiological evidence is understandably absent in the case of Parlodel®, courts should consider otherevidence of causation, including case reports, dechallenge / rechallenge tests, biologicalmechanism evidence, chemistry and animal toxicology studies. See id. at 15-21.
Professors Green and Greenland acknowledge the limited probity of animal toxicology studies where a substantial body of epidemiology exists. They argue,however, that because of the absence of epidemiological evidence, the district court should havemore carefully assessed the probative value of the bromocriptine animal studies, rather thansummarily dismissing them on the basis of case law from the Benedictin and breast implantlitigation, where substantial epidemiological evidence existed. See id. at 16-18.
Similarly, Professors Green and Greenland recognize that as a general matter, case reports are not sufficient to establish general causation. See id. at 18. Theymaintain, however, that in unusual circumstances, case reports can have “considerable saliencyin proving causation.” Id. at 19. The unusual circumstances that might make case reportsprobative include cases where there is an acute reaction to a toxic agent as opposed to a latentreaction, where there are few competing causes (e.g., because the disease is rare) and where theproportion of case reports to the population exposed is substantial. See id. Several of thesecircumstances are present in the Parlodel® litigation.
Amicus Brief of Howard Kipen and Daniel Wartenberg
Howard Kipen, M.D., M.P.H., is Professor and Acting Chair of the Department of Environmental & Community Medicine at the University of Medicine andDentistry of New Jersey – Robert Wood Johnson Medical School. He is a co-author of thechapter “Reference Guide on Medical Testimony” in the current edition of the Federal JudicialCenter’s Reference Manual on Scientific Evidence. He has also authored over 100 scientificarticles on topics including epidemiological and clinical exposures including asbestos, benzeneand the Gulf War syndrome. See Amicus Brief of Kipen-Wartenberg at 1.
Daniel Wartenberg, Ph.D., is a Professor in the Department of Environmental & Community Medicine at the University of Medicine and Dentistry of NewJersey – Robert Wood Johnson Medical School. He has authored over 100 scientific articles ontopics including nuclear workers, Gulf War syndrome and various other toxic exposures. See id.
at 1-2.
In describing their interest in this case, Drs. Kipen and Wartenberg state that they “are concerned about this case because they believe it represents a departure from howmedicine and science actually determine causation in toxic exposure situations.” Id. at 2.
Neither they nor their counsel are being compensated for their work in this case. See id. at 2-3.
Drs. Kipen and Wartenberg first argue that the district court’s decision is premised upon a mistaken view of the certainty involved in causation decisions in the scientificand medical fields. They emphasize that it is extremely rare to have definitive scientific proof ofcausation, that there is no formula for determining causation, and that doctors and scientistsroutinely make critical judgments regarding causation based upon a variety of data sources andfields of expertise. Thus, while the scientific evidence could lead to varying conclusionsregarding the alleged causal relationship between Parlodel® and postpartum stroke, theconclusions offered by Plaintiffs’-Appellants’ experts are based on exactly the type ofinformation and offered to the level of certainty that is typical in the medical and scientificcommunities. See id. at 6-9.
Drs. Kipen and Wartenberg also argue that the district court made several specific errors in its evaluation of the scientific evidence.
First, they contend that the district court ignored powerful evidence in the form of dechallenge / rechallenge tests. They maintain that these tests are analogous to controlled laboratory experiments and are in many ways superior to epidemiological studiesbecause they eliminate many types of bias and confounding. See id. at 9-11.
Second, they argue that the district court gave the case reports short shrift. Although they acknowledge that individual case reports have limited value indetermining causation, they maintain that a significant number of case reports, as in this instance,should be given substantial weight until the causal associations they depict are called intoquestion by epidemiological or other evidence. See id. at 11.
Lastly, Drs. Kipen and Wartenberg assert that the district court misinterpreted the epidemiologic evidence. Like the Plaintiffs-Appellants, they emphasize thatthat the only true epidemiological study yielded a relative risk of 8.4. Thus, they argue that theepidemiology points toward rather than away from causation. See id. at 12.
Brief of Defendants-Appellees
As an initial matter, SPC maintains that the trial court’s decision may only be reversed for an abuse of discretion. See Brief of Defendants-Appellees at 10. It emphasizesthat the deferential abuse of discretion standard “‘applies as much to the trial court’s decisionsabout how to determine reliability as to its ultimate conclusion.’” Id. at 10 (quoting Joiner, 522U.S. at 141-43).
SPC argues as a general matter that the district court properly exercised its discretion as gatekeeper in finding that the Plaintiffs-Appellants failed to carry their burden ofproving the reliability and relevance of their experts’ testimony.
First, SPC argues that the district court did not abuse its discretion because it provided ample support for its determination that the contested evidence was notreliable. See Brief of Defendants-Appellees at 17-18.
The court considered the reliability factors set forth in Daubert and concluded that “plaintiffs’ theory is largely untested and without peer review, ‘[t]herate of error is unknown,’ and ‘[t]he theory has not attained general acceptance within thescientific community.’” Id. at 17 (quoting Siharath, 131 F. Supp.2d at 1355-56; see alsoGlastetter, 107 F. Supp.2d at 1045 n.28.
The court also found the experts’ methodology unreliable due to unjustified reliance on anecdotal case reports, unexplained disregard for chemicaldissimilarities and improper extrapolation of animal studies. See id. at 17-18.
Second, SPC argues that the district court properly found that the data, such as the results of animal studies, cited by plaintiffs’ experts was not relevant because itdid not “fit” the plaintiffs’ causation theory. See id. at 18; Siharath, 131 F. Supp.2d at 1363-65,1367. Specifically, the district court determined that [t]hree scientifically unwarranted ‘leaps of faith’ exist in [plaintiffs’]causal chain. First, a serious question exists whether bromocriptine is likeother ergot alkaloids since it generally causes hypotension rather thanhypertension. Second, even if Parlodel® can occasionally causehypertension, Plaintiffs have not established that it can cause hypertensionso severe as to cause seizures and stroke in humans. Third, even ifParlodel® can cause hypertension severe enough to cause stroke inhumans, Plaintiffs have not shown that it causes hemorrhagic stroke. … Inthis case, ‘there is simply too great an analytical gap between the data andthe opinion proffered.’ Siharath, 131 F. Supp.2d at 1371 (quoting Joiner, 522 U.S. at 146).
SPC also argues that the district court did not abuse its discretion given the consistency of its opinion with other Parlodel® cases. Noting the Eighth Circuit’s decision andthe majority of district court opinions that have excluded similar evidence in Parlodel® cases,SPC argues that “the fact ‘that other courts, after thoroughly sifting through the scientific data,have come to the same decision, and indeed have even excluded some of the same experts as thedistrict court did here in the exercise of its gatekeeping role [is] an indication that the districtcourt was not operating on the outer fringe of its discretion.’” Brief of Defendants-Appellees at22 (quoting Allison v. McGhan Med. Corp., 184 F.3d 1300, 1311 (11th Cir. 1999) (breast implantlitigation)).
On a more detailed level, SPC argues that the district court was well within its discretion in finding that the contested experts’ reliance on case reports, chemicalanalogies and animal studies did not satisfy Daubert. See Brief of Defendants-Appellees at 23-46.
Case Reports: SPC provides several reasons supporting its
contention that the district court correctly concluded that anecdotal case reports were “ofdoubtful validity” in this case. See id. at 24 (citing Siharath, 131 F. Supp.2d at 1361).
First, the Eighth Circuit in the Parlodel® litigation has found that case reports do not constitute scientifically valid proof of causation: Much of the evidence relied upon by Drs. Kulig and Petro has beenculled from case reports in which doctors reported patient strokesfollowing their ingestion of Parlodel. A case report is simply a doctor’saccount of a particular patient’s reaction to a drug or other stimulus,accompanied by a description of the relevant surrounding circumstances.
Case reports make little attempt to screen out alternative causes for apatient’s condition. They frequently lack analysis. And they often omitrelevant facts about the patient’s condition. Hence, ‘[c]ausal attributionbased on case studies must be regarded with caution.’ (citation omitted).
Though case reports demonstrate a temporal association between Parlodel and stroke, or stroke-precursors, that association is notscientifically valid proof of causation.
Id. at 25 (quoting Glastetter, 252 F.3d at 989-90 and citing Brumbaugh, 77 F. Supp.2d at 1156(Parlodel® case reports “do not contain a testable and systematic inquiry into the mechanism ofcausation…”)).
Second, plaintiffs own experts had previously acknowledged that case reports do not establish general causation. See Brief of Defendants-Appellees at 26-28.
Third, given that “[f]rom 1980 to 1994, millions of women took Parlodel®[,] [t]he modest number of case reports associating the drug with stroke or evenpostpartum hypertension is not what would be expected.” Brief of Defendants-Appellees at 27(quoting Siharath, 131 F. Supp.2d at 1361).
Fourth, many of the case reports relied upon were not reports of patients with hemorrhagic stroke. See Brief of Defendants-Appellees at 29 (citingSiharath, 131 F. Supp.2d at 1361, 1363 (discussing heart attack and noting absence of scientificsupport in the record for its relationship to stroke)).
Fifth, contrary to Plaintiffs’ assertion on appeal, the district court did not ignore plaintiffs’ “dechallenge” and “rechallenge” case report data. In fact, thecourt specifically held that defendants “effectively discredited these additional case reports asevidence of a relationship between Parlodel® and postpartum stroke.” Brief of Defendants-Appellees at 30 (quoting Siharath, 131 F. Supp.2d at 1361). This conclusion was not an abuse ofdiscretion. Indeed, the Eighth Circuit also rejected the dechallenge / rechallenge reports on thegrounds that they are statistically insignificant and involve artery spasms and heart attacks, asdistinct from hemorrhagic strokes. See Brief of Defendants-Appellees at 32-33.
Sixth, SPC takes issue with the plaintiffs assertion that the trial court ignored at least five published case reports of postpartum hemorrhagic strokes. SPCemphasizes that four of these five reports were authored by a former Parlodel® expert whoseopinions were excluded in another case and that the fifth report did not involve a hemorrhagicstroke. See id. at 33-34.
Seventh, the FDA advises against relying on case reports to determine causation. See id. at 34. Moreover, the FDA never concluded that bromocriptinecauses seizure and stroke. See id. at 35 (citing Siharath, 131 F. Supp.2d at 1366). Indeed, thecurrent FDA-approved Parlodel® package insert states that a cause-and-effect relationshipbetween Parlodel® and stroke has not been established. See Brief of Defendants-Appellees at 36(citations omitted).
Lastly, SPC argues that the “strong epidemiological evidence that pregnancy itself is a strong risk factor for stroke” also justifies the district court’srejection of case reports. See id. at 36-37 (quoting Siharath, 131 F. Supp.2d at 1363).
Chemical Analogies: SPC also provides several reasons
supporting the conclusion that the expert testimony “extending general conclusions about similardrugs does not meet Daubert’s requirements of reliability.” Brief of Defendants-Appellees at 38(quoting Siharath, 131 F. Supp.2d at 1363-64).
For instance, “bromocriptine differs physically from other ergot alkaloids,” and “Plaintiffs’ experts in this case cannot show that bromocriptine, the activeingredient in Parlodel®, affects the body in a manner similar to other ergot alkaloids.” Brief ofDefendants-Appellees at 38-39 (quoting Siharath, 131 F. Supp.2d at 1363-64) (other citationsomitted). Even plaintiffs’ experts recognize that just because bromocriptine is an ergot alkaloiddoes not mean that it causes vasoconstriction. See Brief of Defendants-Appellees at 39-40.
In addition, a leading treatise on medical toxicology recognized by plaintiffs’ experts indicates that the vasoconstrictive property of bromocriptine iszero. See id. at 40.
Furthermore, in rejecting the chemical analogies, the district court followed ample post-Daubert precedent for the proposition that “‘[s]malldifferences in molecular structure often have significant consequences.’” See id. at. 41 (quotingSiharath, 131 F. Supp.2d at 1364 (citing Schudel v. General Electric Co., 120 F.3d 991, 996-97(9th Cir. 1997), cert. denied, 523 U.S. 1094 (1998))).
In addition, plaintiffs’ own expert admitted that “no epidemiology or even learned treatises link [any] ergot alkaloids to hemorrhagic strokes.” Briefof Defendants-Appellees at 42 (quoting Siharath, 131 F. Supp.2d at 1365).
Animal Studies: SPC argues that the district court properly
found that “the animal studies upon which plaintiffs’ experts relied were so dissimilar to the factsof this case, and the experts’ proposed extrapolations were so speculative, that the studies couldnot offer relevant or reliable support for their opinions.” Brief of Defendants-Appellees at 43(citing Siharath, 131 F. Supp.2d at 1366-69).
Plaintiffs’ experts did not cite any studies showing that bromocriptine causes stroke, and they could not state that Parlodel® caused hypertension inanimals. See Brief of Defendants-Appellees at 44 (citations omitted).
As the Eighth Circuit has held and plaintiffs’ experts admitted, it is difficult to extrapolate reliably from the results of studies on small animals toeffects on humans. See id. at 45-46 (citing Glastetter, 252 F.3d at 991).
SPC counters the argument of Plaintiffs-Appellants that the district court should have viewed the totality of the evidence instead of dissecting each piece with quotes fromone of the contested experts indicating that each element of proof must have individual validityand warning against “‘lumping together several pieces of evidence which are equally hollow so that together they hide one another’s fundamental deficiencies.’” See Brief of Defendants-Appellants at 46-47.
Lastly, SPC asserts that there is no merit to the claim of Plaintiffs- Appellants that the district court imposed an absolute epidemiology requirement, a “statisticsplus a checklist” requirement, or any other requirement other than reliability and relevance. Seeid. at 50. Accordingly, SPC argues, the district court did not abuse its discretion. See id. at 55.
SPC emphasizes that the plaintiff’s experts did not and cannot rely on epidemiology to support their opinions and failed to cite any study showing an increase ordecrease in the incidence of postpartum stroke relative to Parlodel®. While it does not contendthat such evidence is a requirement to prove causation, SPC argues that the non-existence of suchscientific studies is surely “a factor that a court should consider in assessing the reliability ofexpert opinions on causation, as Daubert makes clear.” Id. at 51-52.
SPC also takes issue with the argument of Plaintiffs-Appellants that the district court misinterpreted the existing epidemiological evidence. It maintains that theepidemiological evidence consisted of three studies where no positive causal relationship wasfound and one study that tended to show a negative relationship between Parlodel® and stroke.
See id. at 53 n.40.
SPC also notes that the district court explicitly held that “‘[e]pidemiological evidence is not the only legally sufficient proof for establishing a primafacie case of medical causation.’” Id. at 53-54 (quoting Siharath, 131 F. Supp.2d at 1358); seealso Siharath, 131 F. Supp.2d at 1358 (lack of epidemiological support “is not automaticallyfatal to Plaintiffs[’] case”). It then examined the other types of causation evidence proffered bythe plaintiffs and found that none of them were sufficiently reliable or relevant to be admissible.
See Brief of Defendants-Appellees at 54-55 (citing Siharath, 131 F. Supp.2d at 1370).
Reply Brief of Plaintiffs-Appellants
In their Reply Brief, Plaintiffs-Appellants first attempt to distinguish the present cases from other toxic tort cases and other Parlodel® cases relied upon by SPC.
Specifically, they argue that these cases are distinguishable from Joiner because the contestedexperts did not attempt to infer causation directly from animal and epidemiological studies, butrather explained their inference of causation from a multitude of data sources. Similarly, theydistinguish the breast implant litigation, in which there existed a substantial body of negativeepidemiological evidence. They also highlight differences between the proffered evidence inthese cases and the earlier Parlodel® cases relied upon by SPC. See Reply Brief of Plaintiffs-Appellants at 4-8.
Plaintiffs-Appellants next argue that the present cases are distinguishable “because the court was so explicit in its errors.” See id. at 8. In particular, they maintain that thedistrict court erred by: Saying it was not requiring epidemiology and then making it clear that it was really requiring epidemiology plus chemical evidence, general acceptance, a plausibleanimal model and dozens of well-documented case reports; Finding negative epidemiology in the absence of a valid study Ignoring the difference between dechallenge / rechallenge tests and Completely failing to mention important pieces of evidence. See Plaintiffs-Appellants next argue that the central scientific issue is whether Parlodel® causes vasoconstriction, not whether vasoconstriction causes hemorrhagic stroke, asSPC argues. They maintain that the Eighth Circuit’s opinion in Glastetter, 252 F.3d at 989,recognizes that vasoconstriction can cause either ischemic stroke or hemorrhagic stroke. SeeReply Brief of Plaintiffs-Appellants at 16. Thus, according to Plaintiffs-Appellants, all they needto prove in these cases is that Parlodel® causes vasoconstriction. They maintain that this burdenis satisfied because the drug does cause vasoconstriction where there is low vascular resistance,as is typical in postpartum women. See id. at 14.
Finally, Plaintiffs-Appellants contend that the district court either ignored or misinterpreted the proffered causation evidence. See id. at 17-33.
The few citations to a small part of the record are evidence that the district court failed to properly consider at least four specific pieces of dechallenge / rechallengetests, as well as objective medical tests measuring vasoconstriction. See id. at 17-20 Plaintiffs-Appellants also contend that their experts did not cite animal studies as direct evidence of causation, as the district court thought, but rather as evidencethat bromocriptine can cause either vasodilation or vasoconstriction. See id. at 22-23.
Similarly, Plaintiffs-Appellants maintain that their experts did not rely upon chemical analogies to directly infer causation, as the district court believed, but only todemonstrate that bromocriptine was suspect for vasoconstriction because it is an ergot alkaloid.
They also cite a later edition of the medical toxicology treatise cited by SPC that listsbromocriptine as a hypertensive agent and notes postpartum stroke symptoms possiblyassociated with the drug. See id. at 23-24.
Plaintiffs-Appellants explain that the paucity of case reports is due to the relative rarity of postpartum stroke, the same reason that explains the lack ofepidemiological evidence. See id. at 25-27.
Lastly, Plaintiffs-Appellants reiterate that the epidemiological evidence does not point away from causation, as SPC contends and the district court found.
They highlight the fact that the one study that found a negative relationship between Parlodel®and stroke was withdrawn from publication. Of the remaining three studies, one showed astatistically insignificant positive relationship between the drug and strokes. See id. at 28-30.
Thus, Plaintiffs-Appellants contend, the epidemiological evidence “points weakly towardcausation, not away from it.” Id. at 30.
SIGNIFICANCE OF PARLODEL® LITIGATION TO PHARMACEUTICAL
DEFENSE COUNSEL
Demonstrates That Most Lower Courts Are Taking Their Gatekeeping
Responsibilities Seriously
In Joiner and Kumho Tire, the Court engaged in detailed analyses of the proffered expert evidence. See Joiner, 522 U.S. at 143-147; Kumho Tire, 526 U.S. at 153-58.
The level of detail in Joiner and Kumho Tire illustrates the importance getting specific whenassessing the reliability and relevance of shaky expert testimony. See, e.g., Carr, Dabney J. etal., “After Kumho Tire: Challenging Non-Scientific Experts,” 41 For the Defense 12, 50-51(Defense Research Institute 1999) (emphasizing the need to sell the importance of thegatekeeping responsibility, to identify the specific opinion proffered and to engage in a detailedDaubert analysis of the expert’s method in arriving at the contested opinion).
The Parlodel® litigation demonstrates that for the most part, lower courts are taking the Supreme Court’s examples in Joiner and Kumho Tire to heart and carefullyexamining proffered expert testimony, in keeping with their important gatekeepingresponsibilities. See supra § III.
Accordingly, counsel owe it to their clients to vigilantly and proactively assess the reliability and relevance of proffered expert testimony.
Reveals A Determined And Organized Attempt By Plaintiffs’ Bar To
Minimize/Eliminate Significance Of Epidemiological Evidence In Toxic
Exposure Cases
Undeterred by an unsuccessful appeal of the exclusion of the contested evidence in the Eighth Circuit in Glastetter, 252 F.3d 986, plaintiffs are pursuing appeals of twodistrict court opinions excluding the same evidence in both the Tenth and Eleventh Circuits. SeeHollander, 95 F. Supp.2d 1230 and Siharath, 131 F. Supp.2d 1347.
The thrust of plaintiffs’ arguments on appeal is that the district court applied too stringent a standard because it emphasized the lack of epidemiological evidence, andwas skeptical of the other proffered evidence of general causation in light of the lack ofepidemiological evidence. See supra § IV; see also, e.g., Reply Brief of Plaintiffs-Appellants at8-9 (district court said it was not requiring epidemiology but really was requiringepidemiological evidence plus additional scientific evidence). In short, plaintiffs argue againstany standard that emphasizes epidemiological and other objective evidence and advocate instead a standard that allows general causation to be determined on the basis of more subjectiveevidence frequently relied upon within the medical community. See supra § IV; see also, e.g.,Brief of Plaintiffs-Appellants at 25-28 and n.26. Needless to say, the resolution of this issuecould have far-reaching effects in toxic exposure cases.
The Siharath appeal currently pending in the Eleventh Circuit has captured the attention of distinguished legal and medical scholars, as evidenced by the threeAmicus Curiae briefs filed in support of the Plaintiffs-Appellants. The Amicus briefs were filedby two law professors and a professor of medicine who have authored parts of the FederalJudicial Center’s Reference Manual on Scientific Evidence, and three other distinguishedprofessors of medicine, one of whom served as Editor-in-Chief of the New England Journal ofMedicine from 1991 through 1999. See supra § IV.
Furthermore, the Amicus Curiae Briefs, like the Briefs of Plaintiffs- Appellants, decry the use of any formula in determining the admissibility of expert testimonyand advocate instead the use of a more subjective standard. See Amicus Brief of Berger-Kassirerat 15-18 (advocating a standard that would admit any reliable data that seem to bear on a causalrelationship); Amicus Brief of Green-Greenland at 6-15 (evidence required to prove causation byvarious toxic agents must be flexible); Amicus Brief of Kipen-Wartenberg at 6-9 (no formula fordetermining causation in science).
Includes An Invitation By Plaintiffs’ Bar To Retreat From The Abuse Of
Discretion Standard Established In Joiner
In the Eleventh Circuit appeal of Siharath, the Plaintiffs-Appellants argue that the standard of review is de novo because the district court misinterpreted the DaubertTrilogy. See infra § IV.A.
If accepted, this argument would represent a definite retreat from Joiner’s holding that “abuse of discretion is the proper standard by which to review a district court’sdecision to admit or exclude scientific evidence.” See Joiner, 522 U.S. at 146.
This argument is also significant because virtually any appeal of a district court decision regarding the admissibility of expert evidence could be couched as amisinterpretation of the Daubert Trilogy. Thus, if accepted, this argument could result in a seachange in the degree of deference accorded trial court decisions in this area of the law.
Demonstrates The High-Stakes Nature Of The Expert Wars In
Pharmaceutical Litigation
The Parlodel® litigation also provides a compelling example of the high-

Source: http://www.troutmansanders.com/files/Publication/675254c9-12b4-4f53-8c86-344a4dac7fb5/Presentation/PublicationAttachment/a33fcb79-edc6-413e-bff1-2610298376a2/art-otero-levin2.pdf