Opinion
No. MDL 15-02641-PHX-DGC No. CV-16-00474-PHX-DGC
03-01-2018
IN RE: Bard IVC Filters Products Liability Litigation, Sherr-Una Booker, an individual, Plaintiff, v. C. R. Bard, Inc., a New Jersey corporation; and Bard Peripheral Vascular, Inc., an Arizona corporation, Defendants.
ORDER
The parties have filed various motions in limine ("MIL") in advance of the Booker bellwether trial. The Court ruled on some of the motions in an earlier order. Doc. 10075. This order will rule on the remaining motions.
A. Defendants' MIL No. 1 (Recovery filter complications).
Plaintiff Booker was implanted with a Bard G2 filter in June 2007. The filter later tilted, migrated, and fractured. Ms. Booker had surgeries to remove the filter and several fractured struts, but one strut remains embedded in the wall of her inferior vena cava ("IVC"). Defendants seek to exclude evidence of complications with an earlier version of the Bard IVC filter - the Recovery filter - arguing that those complications are not "substantially similar" to the issues with Ms. Booker's G2 filter. Doc. 9862 at 5-7.
The Recovery was Bard's first generation of retrievable IVC filters, and was on the market from the beginning of 2003 to the fall of 2005. The Recovery was followed by the G2. As Defendants explain in their motion: "Based on clinical experience with the Recovery Filter, Bard made several significant changes to the G2 Filter: the filter hook wire diameter was increased, the filter arm tips were curved, the curvature radius of the filter arms at the sleeve were increased, and the spline was modified to accommodate the other dimensional changes." Id. at 4. Because Ms. Booker received a G2 filter, and that filter had been changed significantly from the Recovery filter, Defendants argue that problems with the Recovery have "absolutely no relevance to the issues in this case." Id. The Court does not agree.
Plaintiffs claim that Bard negligently and defectively designed the G2 filter, failed to warn of its risks, and did so with a mindset worthy of punitive damages. In response, Defendants will argue, among other points, that the FDA cleared the G2 filter for market. Defendants vigorously opposed Plaintiffs' motion to exclude evidence of the FDA's 510(k) clearance process, arguing that the steps a manufacturer must take before a product is cleared by the FDA are "highly relevant to a case . . . like this one," and that compliance with the 510(k) process "is certainly probative under Georgia law on the issues of reasonableness of the design, manufacture, and warnings of the G2 Filter, as well as whether Bard's conduct rises to the level justifying punitive damages[.]" Doc. 9690 at 3. The Court agreed with Defendants, noting that the FDA grants 510(k) clearance "only where the device 'is as safe and effective as a [predicate device] and does not raise different questions of safety and efficacy than the predicate device.'" Doc. 9529 at 4 (citation omitted).
Bard identified the Recovery filter as the predicate device for the G2, and avowed to the FDA that the "design, material, components, fundamental technology (mode of device function/operation) and intended use featured with the [G2]" are "substantially equivalent to those featured with the predecessor Recovery Filter System[.]" Doc. 10068-1 at 25. Defendants assert that "one of Bard's goals in developing the G2 Filter was to reduce the number of incidents of filter fracture and migration that Bard had observed with the Recovery Filter." Doc. 9862 at 4.
Given these facts, the Court concludes that Bard's knowledge of problems with the Recovery filter is relevant to central issues in this case - whether Bard properly designed the G2 to correct those problems, whether Bard failed to warn physicians and patients about problems shared by the Recovery and G2, and whether Bard's alleged failure to correct or warn about known problems justifies an award of punitive damages. The Court also concludes that it would be difficult to try this case without evidence of the Recovery filter and the problems it encountered. Defendants' evidence regarding FDA clearance will necessarily include the fact that the Recovery filter was the predicate device for the G2 and was substantially equivalent to the G2. In showing Bard's care in designing the G2, Defendants almost certainly will argue that Bard carefully tracked problems encountered by the Recovery, dutifully reported those problems to the FDA, and took specific steps to correct those problems in the G2. Defendants will use this evidence to show that Bard did not negligently or defectively design the G2 and did not fail to warn of its risks. It is only fair that Plaintiffs be permitted to present evidence to support their claim that the Recovery was known to be more hazardous than Bard admitted publicly or to the FDA, and that it involved problems that were not corrected in the G2 and Bard failed to warn physician and patients of those known problems.
Defendants cite a number of cases for the proposition that other instances of product failure are admissible only if the instances are substantially similar to the product failure at issue in this case. Id. at 5. For example, the Ninth Circuit has held that "[a] showing of substantial similarity is required when a plaintiff attempts to introduce evidence of other accidents as direct proof of negligence, a design defect, or notice of the defect." Cooper v. Firestone Tire & Rubber Co., 945 F.2d 1103, 1105 (9th Cir. 1991). For three separate reasons, the Court does not find this case law controlling.
First, Ms. Booker alleges that her G2 filter tilted, perforated her IVC, fractured, and migrated. Doc. 10068 at 7. These methods of failure include virtually all of the known methods of failure for the Recovery filter.
Second, the facts of this case distinguish it from many of the cases cited by Defendants. This is not simply a product defect case in which Plaintiffs seek to introduce evidence of other potentially unrelated product failures. As already noted, the Recovery was the predicate for the G2 and Bard sought FDA clearance on the basis that the G2 was substantially equivalent to the Recovery. Bard intends to present evidence of that FDA clearance during trial. Defendants surely will note that the clearance required an FDA determination that the G2 was as safe and effective as the Recovery. In this context, evidence regarding the safety and effectiveness of the Recovery filter is plainly relevant, particularly where Plaintiffs claim that the Recovery was not as safe and effective as Bard represented to the FDA. Further, Defendants claim to have designed the G2 to overcome known defects in the Recovery, making the extent and severity of those defects relevant to the adequacy of Bard's design of the G2.
Third, although substantial similarity is required by the Ninth Circuit for direct evidence of negligence, design defect, or notice, it is not required when other examples of product failure are used to impeach an expert's testimony that a product is safe. Cooper, 945 F.2d at 1105. The parties do not address whether the evidence in this case would be used for impeachment or direct evidence.
Defendants also seek to exclude evidence of complications involving cephalad migration (toward the heart), asserting that such evidence is irrelevant because Ms. Booker's filter migrated in a caudal direction (away from the heart). Doc. 9862 at 7-8. Plaintiffs respond that Bard combined caudal and cephalad migration in tracking adverse events internally and in reporting migration rates to the FDA. Doc. 10068 at 9 (citing Doc. 10068-9 at 2-4). On this record, however, the Court cannot determine whether evidence of cephalad migration is relevant to Ms. Booker's claims. The parties should be prepared to address this issue at the final pretrial conference - specifically, what instances of cephalad migration Plaintiffs intend to present at trial, and why those instances are relevant.
Finally, Defendants contend that the probative value of the Recovery filter evidence is substantially outweighed by the risks of unfair prejudice, misleading and confusing the jury, and wasting time. Doc. 9862 at 8-9; Fed. R. Evid. 403. Defendants claim that admission of such evidence will require Bard to introduce its own evidence regarding the design of the Recovery, the technology available at the time, and Bard's investigation of adverse event reports. Id. at 9. This may be true, but that often is the result of admitting relevant evidence. With the possible exception of cephalad migration, the Court concludes that known complications with the Recovery filter are plainly relevant to this case. It will not, in the Court's view, mislead or confuse the jury, waste time, or result in unfair prejudice. The motion in limine (Doc. 9862) is denied, subject to the Court's consideration of cephalad migration at the final pretrial conference.
B. Defendants' MIL No. 2 (development of the Recovery).
Defendants seek to exclude specific evidence relating to the development of the Recovery device: (1) migration resistance tests, (2) testimony from Dr. Asch regarding his clinical trial of the Recovery, and (3) testimony from Bard employee Kay Fuller regarding the 510(k) submission to the FDA. Doc. 9863 at 2-3. Defendants claim that this "bad act" evidence is irrelevant and inadmissible under Rules 404 and 403. Id. at 3-4.
The Court cannot conclude that development of the Recovery filter is irrelevant to the G2 filter claims in this case. Bard based its request for FDA clearance of the G2 on the claim that it was substantially equivalent to, and as safe and effective as, the Recovery. Defendants intend to argue at trial that the resulting FDA approval shows that Bard acted responsibly and produced a safe and effective filter. In this context, evidence that the Recovery filter was itself the result of inadequate and incomplete testing, that a doctor involved in the testing concluded that it was inadequate and incomplete, and that a Bard employee reached the same conclusion, is plainly relevant. After considering the Ninth Circuit's four-part test, see Duran v. City of Maywood, 221 F.3d 1127, 1132-33 (9th Cir. 2000), the Court concludes that such evidence is admissible under Rule 404(b) as probative of Bard's knowledge, intent, and lack of mistake, and that its probative value is not substantially outweighed by the potential dangers set forth in Rule 403. Defendants' motion in limine (Doc. 9863) is denied.
C. Defendants' MIL No. 3 (FDA warning letter).
Defendants seek to exclude a 2015 FDA warning letter, contending that it is irrelevant because it constitutes an informal advisory statement by the FDA issued more than seven years after Ms. Booker received her G2 filter. Doc. 9864 at 2-3. Defendants also contend that specific topics in the warning letter are not related to the G2 filter or any other issues in this case. Plaintiffs counter that the letter is an essential piece of evidence to rebut Bard's suggestion that the FDA took no action against Bard and expressed no concerns about Bard filters. Doc. 9927 at 2. Other than one sentence, however, Plaintiffs do not address Defendants' argument that topics in the letter are not related to issues in this case. That sentence argues only that deficiencies in Bard's handling and reporting of filter failures are relevant to this case. Id.
Many topics in the warning letter lack probative value. Topics 1 and 2 concern the Recovery Cone retrieval system, which is not at issue in this case. See Doc. 9864-1. Topic 4(a) concerns the filter cleaning process, which does not appear to be at issue here. Topics 4(b), 5, and 6 concern the Denali filter, a generation of filter developed after the G2 received by Ms. Booker. Topics 3, 7, and 8 concern Bard's complaint handling and reporting processes, but Plaintiffs have not shown why the specific issues raised in those three topics are relevant to this case.
The Court concludes that topics 1, 2, 4(a), 4(b), 5, and 6 are not relevant to this case, and will grant the motion in limine with respect to them. The Court cannot tell on the present record whether topics 3, 7, and 8 are relevant; that decision must be made at trial. To avoid the potential prejudice that will result if the warning letter is mentioned but irrelevant, the Court directs Plaintiffs to raise the potential admissibility of topics 3, 7, and 8 outside the hearing of the jury before mentioning the warning letter to the jury.
A few other comments are warranted. Plaintiffs argue that Defendants intend to present evidence of the lack of FDA enforcement actions, and to argue that the "FDA's decision not to take any enforcement actions against Bard is probative as to whether Bard acted reasonably in its design and manufacture of the G2 line of filters." Doc. 9690 at 9. In response, Plaintiffs argue, they should be allowed to show that the FDA did issue a warning letter to Bard. But Plaintiffs' statement of Defendants' position is too broad. Defendants argue that they should be permitted to present evidence of "FDA's lack of enforcement action regarding Bard's G2 line of filters." Doc. 9690 at 8 (emphasis added). The warning letter did not concern the G2 line of filters, and was issued in 2015, long after Ms. Booker received her G2 filter in 2007. If Defendants open the door at trial by arguing generally that the FDA has never taken enforcement action of any kind, the Court will entertain a request to admit some or all of the FDA warning letter. But on the present record, the Court concludes that the warning letter does not concern issues in this case, with the possible exception of topics 3, 7, and 8, as noted above.
If the warning letter becomes relevant, the Court concludes that it should not be barred as hearsay. The public records exception applies to documents that describe "a matter observed while under a legal duty to report" or contain "factual findings from a legally authorized investigation[.]" Fed. R. Evid. 803(8)(A)(ii)-(iii). The warning letter reports the FDA's factual findings and matters observed under the agency's investigatory authority, and Defendants have not shown that the letter lacks trustworthiness. Fed. R. Evid. 803(8)(B). The Court will not exclude the letter on hearsay grounds. See Guthrie v. Ball, No. 1:11-cv-333-SKL, 2014 WL 5314576, at *4 (E.D. Tenn. Oct. 17, 2014) ("Courts have held that FDA warnings . . . are admissible under the public records hearsay exception in Rule 803(8).") (citations omitted); Sadler v. Advanced Bionics, Inc., No. 3:11-CV-00450-H, 2013 WL 1311148, at *2 (W.D. Ky. Mar. 26, 2013) (finding FDA warning letter admissible under the public records exception where "FDA officials conducted the investigation themselves as a neutral party with motivations to protect public health and safety"); Sabel v. Mead Johnson & Co., 737 F. Supp. 135, 140-41 (D. Mass. May 14, 1990) (finding FDA letter recommending a warning label admissible as a public record where it was based on an investigation pursuant to the FDA's regulation of the safe marketing of prescription drugs).
Defendants' motion in limine (Doc. 9864) is granted, subject to Plaintiffs' ability to show that topics 3, 7, and 8 are relevant to issues in this case.
D. Plaintiffs' MIL No. 3 (IVC filters as "lifesaving" devices).
Plaintiffs seek to preclude Defendants from putting on a "filters saves lives" defense, or from describing Bard filters as "lifesaving" or "life-extending." Doc. 9867. Plaintiffs state that IVC filters are designed to prevent blood clots from reaching the heart and lungs and any other presumed benefit is "speculative." Id. at 2. But preventing blood clots from reaching the heart and lungs saves lives. Defendants cite statistics showing that some 300,000 people die each year from pulmonary embolisms. Doc. 10059 at 2 n.1. Plaintiffs' own expert has testified that the purpose of IVC filters is to prevent pulmonary embolisms, and in this sense the filters can be lifesaving devices. Doc. 10059-3 at 3-4 (Kinney Dep. Tr. 111:17-112:2).
Georgia law, which applies in this case, includes a risk-utility analysis for design defect claims. This analysis involves balancing "the risks inherent in a product design" against "the utility or benefit derived from the product." Banks v. ICI Ams., Inc., 450 S.E.2d 671, 673 (Ga. 1994). Evidence concerning the benefits of IVC filters is directly relevant to this analysis. Plaintiffs' motion in limine (Doc. 9867) is denied.
E. Plaintiffs' MIL No. 4 (IVC filters are the "gold standard").
Plaintiffs seek to exclude evidence and argument that IVC filters are the "gold standard" or the "standard of care" for the treatment of pulmonary embolisms. Doc. 9869. In response, Defendants state that they will not characterize IVC filters as the "gold standard." Doc. 10054 at 2, n.1. The Court need not address this issue further.
Defendants do intend to present evidence that IVC filters, including the G2, are within the standard of care for the medical treatment of pulmonary embolisms. Id. Defendants assert that such evidence is relevant to the risk-utility analysis for the design defect and negligence claims, and that admission of the evidence will not confuse the jury or result in a mini-trial. Id. at 2-4. The Court agrees. Evidence regarding the use and benefits of IVC filters, and when they are called for, will be relevant to the jury's risk-utility analysis, as well as evaluation of the failure to warn claims and Dr. D'Ayla's decision to implant the G2 in Ms. Booker.
Plaintiffs express concern that reference to the "standard of care" will confuse the jury concerning the standard of care to be applied in this case. But the standard of care for Bard's design and marketing of IVC filters is entirely distinct from the medical standard of care for when filters should be used to treat patients. To avoid confusion, the parties should refer to the "medical standard of care" when referring to the standard for implanting filters. The parties may also seek a clarifying jury instruction if warranted. Plaintiffs' motion in limine (Doc. 9869) is denied.
F. Plaintiffs' MIL No. 6 (nonparties at fault).
Georgia's nonparty at fault statute provides that the "[n]egligence or fault of a nonparty shall be considered if . . . the defending party gives notice not later than 120 days prior to the date of trial that a nonparty was wholly or partially at fault." O.C.G.A. § 51-12-33(d)(1). When such notice is given, evidence of the nonparty's fault may be presented to the jury and the jury may apportion fault and damages between the plaintiff, the defendant or defendants, and the nonparty. Id., § 51-12-33(b)-(c).
Defendants filed a timely notice identifying Dr. Sarwat Amer as a nonparty at fault. Doc. 8844. The notice states that "Dr. Amer was the diagnostic radiologist who read [Ms. Booker's] lumbosacral spine x-ray on March 26, 2009, which showed her G2® Filter had fractured but with all struts adjacent to the filter in the IVC. Dr. Amer reported only: 'IVC Filter is noted.'" Id. at 2. Defendants contend that Dr. Amer failed to properly report the condition of the G2 filter to Ms. Booker's treating physicians, and therefore prevented the physicians from fully evaluating her medical condition and treatment options. Id. Defendants further contend that this alleged failure "constituted the sole proximate cause and/or contributing cause to [Ms. Booker's] injuries." Id.
Plaintiffs seek to exclude evidence and argument related to (1) the fault or negligence of nonparties not included in Defendants' notice, and (2) the "standard of care" of Ms. Booker's healthcare providers not identified in the notice. Doc. 9871 at 1. Plaintiffs argue that the notice identifies only Dr. Amer as a nonparty at fault, and evidence and argument relating to the negligence or fault of any others is improper under § 51-12-33. Id. at 2. For the same reason, Plaintiffs argue that it would be improper for Defendants to present evidence or argument that treatment by other healthcare providers fell below the medical standard of care. Id.
Defendants state that they will not seek to have the jury apportion fault or damages under the Georgia statute for any healthcare provider other than Dr. Amer. Doc. 10055 at 3. Defendants argue, however, that they should be permitted to present evidence of intervening causes of Ms. Booker's injuries. Id. Specifically, Defendants contend that "the jury is entitled to hear evidence that Dr. Brandon Kang tore Ms. Booker's tricuspid valve during his attempt to retrieve a fractured strut in her right ventricle," necessitating open heart surgery to repair the valve. Id. Defendants intend to argue that Dr. Kang's "medical treatment relates to [Ms. Booker's] damages claim and constitutes an intervening act that severed Bard's liability." Id.
The parties' arguments require the Court to examine Georgia law related to nonparties at fault and intervening causation.
The nonparty at fault statute provides that the "[n]egligence or fault of a nonparty" may be considered by the jury when appropriate notice is given. § 51-12-33(d)(1). The meaning of "negligence" is clear, and the Georgia Supreme Court has defined "fault" to mean "any breach of a legal duty that sounds in tort for the protection of the plaintiff, the breach of which is a proximate cause of the injury about which he complains[.]" Zaldivar v. Prickett, 774 S.E.2d 688, 694 (Ga. 2015). Thus, notice is required before a defendant can argue that a nonparty's negligence or other breach of duty proximately caused the plaintiff's injuries, and that some portion of liability and damages should therefore be allocated to the nonparty.
A defendant in Georgia can also escape liability by showing that an intervening cause broke the chain of causation between the defendant and the plaintiff. The Georgia Supreme Court has held that the intervening cause need not be "wrongful or negligent." Jordan v. Everson, 806 S.E.2d 533, 534 (Ga. 2017). That court has also held that "[f]or an intervening act 'to become the sole proximate cause of a plaintiff's injuries, the intervening act must not have been foreseeable by [the] defendant, must not have been triggered by [the] defendant's act, and must have been sufficient by itself to cause the injury.'" Zaldivar, 774 S.E.2d at 698 (quoting Ontario Sewing Machine Co. v. Smith, 572 S.E.2d 533 (Ga. 2002)). "[I]f the character of the intervening act . . . was such that its probable or natural consequences could reasonably have been anticipated, apprehended, or foreseen by the original wrong-doer, the causal connection is not broken." Id. (quotation marks and citation omitted).
From these authorities, the Court reaches four conclusions:
First, with the exception of Dr. Amer, Defendants may not assert at trial that other medical providers, including Dr. Kang, should be apportioned fault under § 51-12-33. For the jury to apportion fault to a nonparty under the statute, Defendants must have given notice regarding the nonparty. See Monitronics Int'l, Inc. v. Veasley, 746 S.E.2d 793, 804 (Ga. Ct. App. 2013) (noting that Georgia's nonparty at fault statute "mandates strict compliance") (emphasis in original). The instructions and verdict form will not permit the jury to apportion fault to any nonparty other than Dr. Amer.
Second, Defendants may assert the separate legal doctrine of intervening cause with respect to Dr. Kang or other nonparties not named in the notice. The Court cannot conclude that failure to identify Dr. Kang in the notice forecloses the defense of intervening cause. The nonparty at fault statute specifically provides that "nothing in in this Code section shall eliminate or diminish any defenses or immunities which currently exist, except as expressly stated in this Code section." O.C.G.A. § 51-12-33(e). The Code section says nothing about the intervening cause defense.
The statute does state that negligence or fault of a nonparty "shall be considered" only if proper notice is given. § 51-12-33(d)(1). Arguably, this phrase could be read broadly to suggest that the jury may consider evidence of a nonparty's negligence or fault - for any purpose - only if notice is given. The Court reads the phrase more narrowly. Section 51-12-33 specifically focuses on apportionment of damages, and the Court concludes that "shall be considered" means "considered" in apportioning damages. This reading is reinforced by the statute's express statement that it does not eliminate other defenses, and seems to comport with Georgia case law. See Vega v. La Movida, Inc., 670 S.E.2d 116, 122 (Ga. Ct. App. 2008) (finding that the defendant was entitled to argue that a nonparty's act was an intervening cause of the plaintiffs' injuries and noting that § 51-12-33 concerns apportionment of damages and therefore was inapposite).
Third, to prevail on their intervening cause defense, Defendants must show that Dr. Kang's intervening act, or the intervening act of any other nonparty, (1) was not foreseeable by Bard, (2) was not triggered by Bard's act, and (3) was sufficient by itself to cause Ms. Booker's injury. See Zaldivar, 774 S.E.2d at 698. The jury will be so instructed.
Fourth, although Defendants will be precluded from arguing that Dr. Kang or others (besides Dr. Amer) were negligent or at fault for purposes of apportioning liability, the Court cannot conclude that they should be precluded from asserting the fault of these nonparties as an intervening cause of Ms. Booker's damages. Georgia law states that an intervening cause need not be negligent or wrongful, but it also recognizes that such a cause can be negligent or wrongful. See Jordan, 806 S.E.2d at 534; Vega, 670 S.E.2d at 122. It appears that Defendants intend to suggest to the jury that Dr. Kang was at fault. In a separate motion in limine, Plaintiffs sought to exclude evidence related to Dr. Kang's social media accounts. Doc. 9877. Defendants opposed the motion, asserting that a central issue is "whether Dr. Kang's retrieval was unduly aggressive in this case." Doc. 10063 at 2. Defendants stated that they "will present expert and other evidence that Dr. Kang should have followed a more conservative approach." Id. Defendants may make this argument as part of their intervening cause defense. The Court will instruct the jury carefully, making clear that the fault or actions of Dr. Kang (or similarly situated nonparties) may be considered only if the three requirements for intervening cause set forth above are satisfied, while the fault of Dr. Amer may be considered by the jury in apportioning fault. Whether the intervening cause defense must apply to all of Ms. Booker's injuries or may be applied to only a discrete part of them is an issue the parties should be prepared to address when jury instructions and the verdict form are finalized.
Defendants argue that they can use the intervening cause defense to assert that Dr. Kang "contributed to" Ms. Booker's injuries. Doc. 10055 at 3. This is not correct. Either his conduct became an intervening cause and broke the causal chain between Defendants and Ms. Booker, eliminating Defendants' liability, or it did not. True, if Defendants are able to show that his actions were the intervening cause for some discrete part of her injury (say, her heart surgery), it may have the same effect as if the jury apportioned fault and assigned 100% of the heart surgery responsibility to Dr. Kang. But the jury could reach this result only through the intervening cause analysis and a finding that the three elements set forth above are satisfied. Defendants cannot use the intervening cause defense to essentially argue for equitable apportionment - that the jury should divide the liability between Defendants and Dr. Kang on the basis of their respective degrees of fault.
Plaintiffs' motion in limine (Doc. 9871) is granted in part and denied in part. The Court will limit Defendants' nonparty-at-fault and apportionment arguments to Dr. Amer, but deny the motion to the extent Plaintiffs seek to preclude Defendants from asserting that the conduct of Dr. Kang or others constituted an intervening cause.
G. Plaintiffs' MIL No. 13 (the alleged fault of Dr. Amer).
In a related motion, Plaintiffs seek to exclude evidence regarding the alleged fault of Dr. Amer on the ground that Defendants have no expert testimony to support a causation finding by the jury. Doc. 9878. In their notice of nonparty at fault, Defendants state that they rely on the expert reports and deposition testimony of Drs. Daniel Cousin and Piotr Sobieszczyk. Doc. 8844 at 2. Defendants argue that, under Georgia law, causation may be established by linking the testimony of different expert witnesses, and that in this case Dr. Cousin opines that Dr. Amer fell below the standard of care and Dr. Sobieszczyk provides the causation opinion. Doc. 10066 at 2-5.
Dr. Sobieszczyk's report references several imaging studies Ms. Booker received, including the radiological exam performed by Dr. Amer in March 2009. Doc. 10067 at 9. He notes that Dr. Amer's exam "showed an IVC filter with one short fragment separated from the filter and pointing up above the filter tip," and another fragment "separated from the filter and moved towards the midline." Id. He refers to the imaging studies as "missed opportunities," opining that they "clearly defined progressive tilting and strut movement" and "[r]ecognition of this process would have allowed timely retrieval of the filter." Id. at 11, ¶ 2.
The Court finds that this opinion, when combined with Dr. Cousin's opinion regarding the standard of care (Doc. 9878-4 at 4), is sufficient to create a jury issue on whether Dr. Amer is at fault for allegedly failing to report the condition of Ms. Booker's filter to her treating physicians. Georgia courts have made clear that "[q]uestions regarding causation are peculiarly questions for the jury except in clear, plain, palpable and undisputed cases." Moore v. Singh, 755 S.E.2d 319, 323-24 (Ga. Ct. App. 2014) (citation omitted). This is not one of those cases. See id. at 324 ("Based on the combined expert testimony, we conclude that Moore presented evidence creating a jury issue as to whether Dr. Singh would have discovered the fracture if she had properly complied with the standard of care [and] . . . whether the failure to diagnose the fracture during that time led to further complications[.]"). Plaintiffs' motion in limine (Doc. 9878) is denied.
H. Plaintiffs' MIL No. 9 (statements from associations and other groups).
Plaintiffs seek to exclude evidence of "statements from various associations, trade groups, organizations, societies of physicians, [and] medical providers[.]" Doc. 9874 at 1. Plaintiffs first contend that such statements constitute inadmissible hearsay, and possibly hearsay within hearsay. Id. at 2. Plaintiffs also contend that admission of such statements through non-experts would evade Daubert scrutiny and would violate Rule 403. Id. at 2-3. Because Plaintiffs identify no specific statement from any particular association, the Court cannot grant Plaintiffs' motion. But the Court will address some of the parties' arguments.
Defendants argue that statement by medical societies may be relevant even if they are not offered for the truth of the matter asserted. Doc. 10058 at 2. Defendants note, for example, that the risk-utility test of Georgia law permits a jury to consider "the gravity and severity of the danger posed by [a product's] design; the likelihood of that danger; the avoidability of the danger, i.e., the user's knowledge of the product, publicity surrounding the danger, or the efficacy of warnings, as well as common knowledge and the expectation of danger[.]" Banks, 450 S.E.2d at 675 n.6. Defendants argue that statements from the Society of Interventional Radiologists ("SIR") about the risks of IVC filters are relevant to several of these factors - the user's knowledge of the product, publicity surrounding the product's danger, and the common knowledge the danger - regardless of whether the SIR's statements are true. Admission for these non-truth purposes, Defendants assert, would not implicate the hearsay rules. Whether statements should be admitted for these purposes must be decided at trial.
Defendants also contend that statements by medical societies are admissible under Rule 803(18) even if offered for the truth of the matter asserted. Doc. 10058 at 3. Rule 803(18) provides that a statement contained in a treatise or periodical is excluded from the hearsay rule if (1) "the statement is called to the attention of an expert witness on cross-examination or relied on by the expert on direct examination," and (2) "the publication is established as a reliable authority by the expert's admission or testimony, by another expert's testimony, or by judicial notice." Fed. R. Evid. 803(18)(A)-(B). If deemed admissible, "the statement may be read into evidence but not received as an exhibit." Id. "This limitation ensures that the jurors will not be unduly impressed by the treatise, and that they will not use the text as a starting point for conclusions untested by expert testimony." 5 Weinstein's Federal Evidence § 803.20[1]. Whether Rule 803(18)'s requirements are satisfied requires a statement-by-statement inquiry, which must be determined at trial. The party proposing to use a statement under Rule 803(18) must be prepared to show that both requirements are met.
On the second requirement, Weinstein's explains: "To establish a proper foundation, the proponent must show that the author of the treatise or article in question is an authority." 5 Jack B. Weinstein & Margaret A. Berger, Weinstein's Federal Evidence § 803.20[2] (2d. ed. 2018).
Plaintiffs further contend that testimony from non-expert witnesses about statements from medical societies such as the SIR would allow Bard to avoid the scrutiny of Daubert. Id. Defendants avow that they will not attempt to offer any expert opinion about such statements through a fact witness. Doc. 10058 at 3. Defendants do state, however, that SIR guidelines were discussed routinely among Bard employees, and that fact witnesses should be permitted to testify about the things they said and the actions they took concerning SIR guidelines. Id. at 3-4. If a discussion at Bard that included reference to SIR statements is relevant to an issue in this case and not otherwise objectionable, then a lay witness's recounting of the discussion may be appropriate. The Court will defer ruling on any objections until trial.
Defendants contend that even if SIR guidelines are themselves inadmissible on hearsay or other grounds, their experts may rely on the guidelines under Rule 703. Defendants assert that some of their experts have relied on the guidelines in forming their opinions and were not challenged by Plaintiffs in any Daubert motion. Doc. 10058 at 3. But even if an expert is allowed to state an opinion formed in part by reliance on the SIR guidelines, that does not mean that the guidelines themselves may be disclosed to the jury. "Rule 703 is not, itself, an exception to or exclusion from the hearsay rule or any other evidence rule that makes the underlying information inadmissible." 4 Weinstein's Federal Evidence § 703.05[2]. Rule 703 generally does not permit an expert to relay inadmissible evidence - as opposed to the expert's own opinion based on such evidence - to the jury. Indeed, there is "a presumption against disclosure to the jury of information used as the basis of an expert's opinion and not admissible for any substantive purpose, when that information is offered by the proponent of the expert." Fed. R. Evid. 703 (2000 Advisory Committee's Note). Thus, if an objection is made, otherwise inadmissible information may be disclosed to the jury under Rule 703 only if the proponent shows that "the probative value of the information in assisting the jury to evaluate the expert's opinion substantially outweighs its prejudicial effect." Id.; Fed. R. Evid. 703.
If the Court sustains an objection to the admission of the SIR guidelines through expert testimony, and Defendants' believe that the probative value of the guidelines in helping the jury evaluate the expert's opinion substantially outweighs any prejudicial effect, Defendants are instructed to raise this issue with the Court outside the hearing of the jury. If the Court concludes that the SIR guideline evidence may be disclosed to the jury under Rule 703, Plaintiffs may request a limiting instruction that the evidence may be used only for evaluating the expert's opinion and not for substantive purposes.
Plaintiffs assert that Bard intends to use the SIR guidelines to establish acceptable failure rates for IVC filter manufacturers, and that use of the SIR guidelines in this way would be misleading and result in unnecessary mini-trials. Doc. 9874 at 3. Plaintiffs further assert that statements from professional organizations with lofty names have limited probative value that is substantially outweighed by the risk of unfair prejudice, confusion, and delay. Id. at 3-4. Efficient management of the evidence and adherence to the trial time limits will avoid the risk of mini-trials and delay. Doc. 10058 at 4. The Court cannot conclude on the present record that any probative value the SIR guidelines may have is substantially outweighed by the danger of unfair prejudice. Plaintiffs may object during trial if they believe specific SIR guideline evidence is inadmissible.
Finally, Plaintiffs seek to exclude evidence regarding the lack of statements from the SIR and similar organizations. Doc. 9874 at 1-2. Plaintiffs cite to the testimony of one of Bard's experts that, to her knowledge, no medical society has taken the position that Bard retrievable filters carry more risks than other IVC filters. Doc. 9874-2 at 3 (Roberts Dep. Tr. 78:9-79:6). Defendants counter that what organizations like the SIR do not say about the risks of IVC filters is relevant to the jury's consideration of the design defect and failure to warn claims. Doc. 10058 at 3. Whether such evidence is relevant and otherwise admissible will depend heavily on context. The Court will make rulings as necessary during trial.
Defendants assert that testimony regarding the absence of a statement is not hearsay. Doc. 10058 at 3 n.1 (citing Llamas v. Seibel, No. 16-CV-05812-WHO, 2017 WL 3782175, at *8 (N.D. Cal. Aug. 31, 2017)). Although this generally is true, silence can be a "statement" for purposes of the hearsay rules where the person "intended it as an assertion." Fed. R. Evid. 801(a); see McGiboney v. CCA W. Props., Inc., No. 1:13-cv-00214-REB, 2016 WL 843253, at *7 (D. Idaho Mar. 1, 2016) (Rule 801 "explicitly requires that the declarant must actually intend his statement (even if it consists of silence) as an assertion"); Weinstein's Federal Evidence § 801.10[2][b] ("Even silence may be intended as an assertion"). If Plaintiffs believe that testimony regarding the absence of statements from medical societies constitutes inadmissible hearsay, they may object. Plaintiffs' motion in limine (Doc. 9874) is denied.
I. Plaintiffs' MIL No. 10 (FDA consent for warnings or recalls).
Plaintiffs seek to exclude evidence or argument that Bard needed the FDA's consent to add warnings to its labels, send warning letters to physicians and consumers, or recall the G2 filter. Doc. 9875. Defendants do not dispute that a medical device manufacturer may send warning letters and voluntarily initiate a recall without FDA consent. Doc. 10062 at 2-4. Defendants will be precluded from presenting evidence or argument to the contrary. Defendants may, however, present evidence and argument explaining the FDA's potential involvement in any recall effort and the reasons why Bard filters were not recalled. Id. at 4.
Defendants dispute that manufacturers may add warnings to device labels without FDA consent. Defendants cite 21 C.F.R. § 807.81, but this regulation says nothing about adding warnings to labels. Rather, the significant changes or modifications that may require FDA premarket notification are those affecting the intended use of the device or the "design, material, chemical composition, energy source, or manufacturing process." 21 C.F.R. § 807.81(a)(3)(ii)-(iii). None of these constitutes adding a warning to a label.
Defendants also cite an FDA guidance document that notes that a labeling change "meant to significantly improve clinical outcomes, to mitigate a known risk, or in response to adverse events . . . likely requires submission of a new 510(k)." Doc. 10062 at 2. As the Court previously noted, this document is meant only to provide guidance and does not bind the FDA or device manufacturers. Doc. 8872 at 24 n.13. Moreover, the document provides that "manufacturers should monitor device usage and promptly revise the warnings and precautions based on user experience," and that "[s]ubmission of new 510(k)s for such labeling changes are generally not required." Guidance Doc. at 22.
See FDA, Deciding When to Submit a 510(k) for a Change to an Existing Device, Guidance for Industry and FDA Staff, at 19 (Oct. 25, 2017), available at https://www.fda.gov/downloads/medicaldevices/deviceregulationandguidance/guidancedocuments/ucm514771.pdf (last visited Feb. 21, 2018).
Defendants cite a regulation that requires FDA approval for label changes or product comparisons, but that regulation applies only to prescription drugs. See 21 C.F.R. § 201.57(c) ("The requirements in this section apply only to prescription drug products described in § 201.56(b)(1)[.]"). Similarly, the regulation Defendants cite regarding clinical trials concerns the FDA's approval of new drugs, not medical devices. See 21 C.F.R. § 314.126.
Plaintiffs' position, according to Defendants, is that "Bard could have added warnings regarding 'the increased risks' of complications 'associated with its products.'" Doc. 10062 at 2. Defendants contend that the concept of "increased risks" necessarily requires a comparison of the G2 and other filters based on medical device reports ("MDRs") and data from the FDA's Manufacture and User Facility Device Experience ("MAUDE") database, and Plaintiffs' proposed warning regarding increased risks runs afoul of FDA regulations governing what content may be used in device instructions for use. Id. (citing 21 C.F.R. § 201.57(c)). As noted above, however, the cited regulation applies only to prescription drugs. Moreover, Plaintiffs' motion states that Bard could have warned about increased risks by voluntarily sending letters to physicians and patients. Doc. 9875. Defendants cite no FDA rule or regulation prohibiting such letters.
Defendants note that MDR and MAUDE data "alone cannot be used to . . . compare event rates between devices" and cannot be "used in isolation to reach conclusions about the existence, severity, or frequency of problems associated with devices." Doc. 10062 at 2. This appears to be correct, but Defendants have not shown that FDA consent would be required to provide a warning about increased risks based on other reliable sources.
See FDA MAUDE Database, available at https://www.accessdata.fda.gov/scripts/cdrh/cfdocs/cfmaude/search.cfm (last updated Jan. 31, 2018; last visited Feb. 27, 2018). --------
Plaintiffs' motion in limine (Doc. 9875) is granted in part. Defendants will be precluded from presenting evidence or argument that Bard needed FDA consent to add any warning to its labels, send warning letters to physicians and patients, or recall its filters. Defendants may, however, present evidence and argument explaining the reasons why Bard filters were not recalled, the FDA's potential involvement in any recall effort, and the fact that warnings about failure rates and increased risks could not be based on MDR and MAUDE data alone.
Dated this 1st day of March, 2018.
/s/_________
David G. Campbell
United States District Judge