Sometimes a case has one bright shining point floating in a sea of not-so-good ones, or a stand-out result lost among outcomes that should have gone the other way. That describes Isaac v. C.R. Bard, Inc., No. A-19-CV-895, 2021 WL 1177882 (W.D. Tex. Mar. 29, 2021), a case remanded from the Bard IVC filter MDL. As in other filter cases, the plaintiff in Isaac alleged complications following the implantation of a filter following a medical emergency. The defendants’ motion for summary judgment resulted in the dismissal of some claims, but not most, and the order covers a lot of ground. For our part, we found three parts of the order particularly interesting—the parts dealing with warnings-based claims, design defect, and the presumption of non-liability under Texas law for federally approved or licensed products, such as medical devices.
First, the highlight. As we explained just last week, a plaintiff asserting a warnings-based claim bears the burden of proving both that the defendant’s warnings were inadequate and that the inadequate warning proximately caused her alleged injuries. In the drug and medical device world, that means the plaintiff bears the burden of proving that a different or stronger warning would have altered her doctor’s decision to prescribe the drug or device. That is the rule in the Fifth Circuit, too:
The Fifth Circuit has explained that causation in the context of the learned-intermediary doctrine entails two distinct factual predicates: (1) that the doctor would have read or encountered the adequate warning, and (2) that the adequate warning would have altered his treatment decision for, or risk-related disclosures to, the patient.
Id. at *5. We are not so sure about that very last part—that a plaintiff can meet the second factual predicate with evidence that the physician would have altered his or her risk-related disclosures. But for now, we will focus on the first part—that the doctor would have read or encountered the adequate warning. In Isaac, no one took the implanting physician’s deposition, so there was no evidence whether the physician would have seen a different warning, let alone what he or she would have done had the warning been stronger or different. Id. Importantly, it was no substitute to have an expert say that a “reasonable implanting physician” would have changed his or her treating decisions. As the court observed, “the learned-intermediary analysis focuses on the actions of the treating physician, not the opinion of an expert witness.” Id. (emphasis added).
The remainder of the order? Well, not so good. On design defect, the defendants invoked comment k to Restatement (Second) of Torts § 402A. As our faithful readers know, comment k precludes design defect claims for “unavoidably unsafe products,” the rationale being that many products will always carry risks and thus cannot be considered “defective” so long as they come with adequate warnings. The court in Isaac rejected comment k because, according to the court, comment k has never been extended to medical implants, let alone devices cleared under section 510(k). Id at *6.
The court here is drawing two false distinctions to come to the wrong conclusion. First, there is no justifiable basis for applying comment k to drugs, but not prescription medical devices. Both exist to treat injury and illness, and both carry treatment risks that cannot be completely eliminated. That is why federal law allows their use only when prescribed by medical doctors and other qualified medical professionals. Prescription-only products are “unavoidably unsafe” under any measure, and comment k should apply to both prescription drugs and devices. Second, the regulatory pathway to marketing—here the FDA’s 510(k) premarket notification process—is irrelevant to whether a product is worthy of comment k’s protection. The standard is whether a product’s use involves unavoidable dangers or risks, and prescription-only medical devices have known risks, without regard to how the FDA permitted the device’s marketing. This ruling is yet another example of courts incorrectly marginalizing the 510(k) process and wrongly considering the process to be unrelated to product safety.
The court also rejected the defendants’ argument that the plaintiff could not prove that a safer alternate design existed, which was required under the applicable Texas law. Id. at *8-*7. The filter used to treat the plaintiff was a filter designed to be retrievable—that is to say, a physician could opt to remove the filter based on the individual patient’s condition and needs. As a safer alternative, the plaintiff offered only filters designed to be implanted permanently, which are different products. That should have been enough for the court to rule that the plaintiff had not offered an safer alternative design. The court, however, found a triable issue of fact, mainly because the defendant had marketed the filter as both permanent and retrievable and also because the MDL had made a similar ruling in a prior case. Id. at *7.
Which leads us to a third part of the order that piqued our interest. Section 82.008(c) of the Texas Civil Practices and Remedies Code establishes a rebuttable presumption that a manufacturer is not liable for any injury to a plaintiff caused by a product if: (1) the product was subject to pre-market licensing or approval by the federal government; (2) the manufacturer complied with all the applicable procedures and requirements; and (3) the federal government approved or licensed the product for sale after full consideration of the product’s risks and benefits. Id. at *7.
Because the FDA cleared the device—i.e., licensed it for marketing—through the 510(k) process, the defendant was entitled to Section 82.008(c)’s presumption of non-liability. The court, however, rejected the presumption based on the misperception that the 510(k) process does not focus on safety. Id. at *8. Other courts have made this mistake, and nearly all of them cite Medtronic v. Lohr, 518 U.S. 470, 478 (1996), which was outdated virtually since the day it was published. As we have explained in detail, the 510(k) process very much focuses on product safety. The defendant in Isaac should have received the benefit of Texas’ presumption. Given the prevalence of preemption in design defect cases involving Class III medical devices, this holding has the effect of nullifying the Texas compliance presumption in medical device cases. Either the claim is preempted, and the presumption is not necessary, or the presumption itself falls away. Courts, particularly federal courts, should not be interpreting state statutes out of existence.
The court made other rulings. It found that the plaintiff’s expert had raised a triable issue of fact on medical causation and future damages (Id. at *3-*5); it ruled that the plaintiff had presented evidence of negligent manufacturing through an expert stating that the plaintiff’s filter did not meet “performance specifications” (Id. at *7); it declined to apply Texas’ presumption of non-liability for products that comply with federal safety standards or regulations (Id. at *9); and the court applied Texas’ punitive damages law (which allows punitive damages) over Arizona law (which does not), ruling that the state where the plaintiff resided, was treated, and experienced her alleged injury (Texas) had a more significant relationship with the case than the state where the defendant resided and the product was designed and manufactured (Id. at *10).
So there you have it. A beacon of light on the warnings claims, and not much to commend after that. You might even call it a mixed bag (we made it all the way to the end before using that phrase).