This is the third consecutive week a case from Indiana has been in our sights. By and large, we have been impressed with the quality of the Indiana judiciary in both the state and federal courts. One hallmark of exemplary judging is the capacity to revisit rulings and change course when necessary. We clerked for a judge (William Norris on the Ninth Circuit) who was brilliant and hard-working and hardly ever made mistakes, but when he concluded he or his panel got something wrong, he stepped up immediately and fixed what needed fixing. As recent events have made crystal clear, judicial integrity is a bedrock of the republic.

Today’s case, In re Cook Medical, Inc., IVC Filters Marketing, Sales Practices and Product Liability Litigation, 2020 WL 241097 (S.D. Indiana Jan. 6, 2020), shows a court realizing that a piece of evidence admitted after a close call was so abused and so consequential that a new trial was warranted.

First, have a few bites of background. (But not too much; you don’t want to spoil your appetite for tomorrow’s festivities.) The plaintiff claimed that an IVC filter’s defective design caused it to fracture within her, leading to follow-up operations, after which some pieces of the filter remained inside her. She brought the action in the Northern District of Georgia, but the case was scooped up by the Multi-District Litigation, and then was remanded to the Southern District of Indiana, where it went to trial. The jury returned a verdict for the plaintiff in the amount of $3 million. The defendant then filed post-trial motions for judgment as a matter of law (Fed. R. Civ. P. 50) and for a new trial (Fed. R. Civ. P. 59).

The Rule 50 motion for judgment as a matter of law did not succeed. Such motions usually do not. They face a tough standard. A court may enter judgment against a party who has been fully heard on an issue during a jury trial if a reasonable jury would not have a legally sufficient evidentiary basis to find for the party on that issue. The defendant here had spotted some evidentiary gaps at trial on the issues of product defect and whether such defect actually caused the injury. Nevertheless, the court concluded that the plaintiff had presented sufficient evidence for a reasonable jury to find in her favor on her strict liability and negligence design defect claims.

The Rule 59 motion for a new trial did succeed, and that is a big deal. Getting a new trial is not quite as tough as getting judgment as a matter of law, but the standard is still onerous. In this case, the questions were whether there was an error in the admission of evidence, whether the improperly admitted evidence likely had a substantial influence on the jury, and whether the result reached was inconsistent with substantial justice. Remarkably, the court answered these questions in the affirmative, threw out the verdict, and ordered a new trial.

What was the improperly admitted evidence? The plaintiff introduced a defendant employee email chain that attached a table listing patients who died after implantation of the defendant’s IVC filters — some, but not all, of those filters being the same one at issue in the case. Mind you, the plaintiff in the case most definitely did not die, so it is easy to understand why this exhibit could be confusing and devastating. The attachment to the embank chain was a complaint summary regarding adverse events (AEs) that resulted in deaths. There is nothing nefarious about collecting or reviewing such information. There are specific FDA reporting obligations regarding death AEs.

Why would anyone dream of admitting the death email chain in a case not involving death? The court admitted the exhibit at trial, finding “it’s prejudicial” but “it’s probative on the issue of risk utility” and that the defendants objections went to weight and not admissibility.

The court came to regret that decision.

One category of evidence that plaintiffs often seize upon is other similar incidents. The burden is on the proponent of the evidence to show substantial similarity. Having the benefit of the trial record, the court concluded that the plaintiff did not satisfy that burden. The plaintiff was happy to use the emails to suggest both that the company was on notice of the defect and that the defect could cause very serious injuries. Death, to be sure, is serious. But there was not really substantial similarity between the deaths described in the email that were associated with the particular type of IVC filter, which involved fatal retroperineal hemorrhages, and the very much alive plaintiff’s injury, which involved a fracture resulting in strut pieces migrating to her thigh, psoas muscle, and spine.

Moreover, again exercising hindsight, the court deemed the death reports to be inadmissible hearsay. The emails were written between the defendant’s employees, but they were “merely recording information provided by third parties.” The plaintiff argued that the emails and attachment were statements against interest, but the material statements were not statements by the defendant or its employees. There was also nothing in the record to show that the company was adopting the information as true.

The court also included that the email chain, as utilized by the plaintiff at trial, ran afoul of Fed. R. Evid. 403’s proscription against evidence that is unfairly prejudicial. The lack of similarity with the plaintiff’s injury and the hearsay factor meant that the email’s probative value was minimal. Meanwhile, the plaintiff lawyer milked the death issue for all it was worth, and then some, in direct exams and cross-exams. This milking culminated in the closing argument (both initial closing and rebuttal), when the plaintiff lawyer “referred to death and dying a multitude of times.”

Consequently, the court held that the erroneous admission of the death email “successfully crossed the hurdle from harmless error to prejudicial error.” (It is hard to argue that something was harmless error if you lingered on the point in closing argument. When we prosecuted cases in Los Angeles back in the 1990s, if we ever got something into evidence on a close call, such as a prior conviction, we usually avoided emphasizing it in closing argument. We could count on the jurors discussing it during deliberations. We took care to give the Ninth Circuit one less reason to reverse.)

As mentioned above, getting a new trial is no easy thing. The court in the IVC filters case did “not come to this conclusion lightly. It has thoroughly reviewed the 4,000-page trial transcript and the parties’ memorandums of law.” The court showed admirable diligence. And then it showed admirable integrity by deciding that the email had been erroneously admitted and that it likely “had a substantial effect on the jury’s verdict.”