Photo of Steven Boranian

We have closely followed the development (or perhaps more accurately, stagnation) of so-called “innovator liability.” It started with the wrongly reasoned and wrongly decided opinion from the California Court of Appeal in Conte v. Wyeth, 168 Cal. App. 4th 89 (2008), where the court held that an innovator drug company could be liable for injuries allegedly caused by generic products it did not make and did not sell. We rolled our collective eyes at this unmooring of tort law, where lines of cases dating back to cars with wooden wheels placed responsibility with parties who made and sold the allegedly offending product.

In the eight years since Conte, courts throughout the country in more than 100 different cases have roundly rejected innovator liability, with only a few exceptions. You can view our innovator liability scorecard here, and you can click on the “Conte” link on the side of the page to see all of our commentary on the subject. There is a lot of it. Our 50-state survey of innovator liability is here, which is as good a place to start as any. The upshot is that innovator liability should become another Sindell v. Abbott Labs, which introduced market share liability in 1980, only to become nothing more than a curious historical footnote in product liability law.

We revisit the issue now because the California Court of Appeal has gone off the rails again. In T.H. v. Novartis Pharmaceuticals Corp., No. D067839, 2016 Cal. App. LEXIS 179 (Cal Ct. App. Mar. 9, 2016), the court held that a listed drug manufacturer can be liable for injuries alleged caused by a generic version of the drug terbutaline sulfate, even though the listed manufacturer divested itself of the product six years before the plaintiff even used it. At least in Conte the plaintiff alleged that Wyeth still held the NDA. Here, the listed manufacturer had been completely out of the terbutaline business for several years.

How did this happen? The FDA approved terbutaline as a bronchodilator for treating asthma in 1974. Id. at **3-4. The defendant manufactured and sold the product for that purpose until December 2001, when it divested its interest in the product and sold the NDA to another pharmaceutical company. Id. at *4, *10. The plaintiffs’ mother, while she was pregnant, was prescribed terbutaline six years later, in 2007, to inhibit uterine contractions and to prevent premature labor, which was off-label use. The treatment allegedly contributed to autism in her twins, who sued not the company that made and sold the drug that their mother ingested (knowing full well that those claims against a generic product would have been preempted), but instead sued the company that formerly held the NDA. Id. at *11. We will emphasize again, because it’s really important, that the defendant was not selling any terbutaline at the time, and it had divested its interest in the product six years earlier.

The trial court therefore dismissed the case on the pleadings and ruled that the defendant owed the plaintiffs no duty—the correct and obvious result. Id. at *13. The Court of Appeal, however, saw it differently and held that the defendant innovator manufacturer could be liable for injuries alleged caused by another company’s product. According to the Court, the issue was negligent failure to warn, and the plaintiffs argued on appeal that the defendant had information about the risk of adverse neurological effects before it sold the terbutaline NDA in 2001. Thus, the defendant had a duty to strengthen the drug’s warnings in 2001, and if it had done so, it would have prevented their injuries, which were diagnosed in 2012. Id. at **16-17. Relying on Conte, the Court of Appeal stated the general rule that “all persons have a duty to use ordinary care to prevent other from being injured” and found that it was foreseeable that a patient could be injured by relying on a brand-name manufacturer’s product information, even though the patient used a generic form of the drug. Id. at **18-20.

We have written so much about Conte and innovator liability, we will cut to the chase. This opinion is wrong for four reasons that we immediately see. There may be others.

First, it contradicts the defining principle of product liability law. Product liability flows from the manufacture and sale of a product, and California is no exception to this general rule. Justice Roger Traynor and the California Supreme Court essentially created the modern law of product liability in Greenman v. Yuba Power Products, Inc., 59 Cal. 3d 57 (1963), where Justice Traynor wrote that “a manufacturer is strictly liable in tort when an article he places on the market, knowing that it is to be used without inspection for defects, proves to have a defect that causes injury to a human being.” (Emphasis added). And then there is the more recent opinion from the California Supreme Court in O’Neil v. Crane Co., 53 Cal. 4th 335, 360 (2012), where the Court rejected a duty to warn against hazards in another manufacturer’s product because “[a]n interpretation of [the law] that would require a manufacturer to warn about all potentially hazardous conditions surrounding the use of a product, even when those hazards arise entirely from the product of another manufacturer, reaches too far.” O’Neil v. Crane Co., 53 Cal. 4th 335, 360 (2012). By holding a product manufacturer potentially liable for injuries allegedly caused by a product that it did not make and did not sell, the T.H. court departed from this premise and decades of controlling case law.

Second, the opinion is against the great weight of authority. Innovator liability is itself no longer innovative. It has been rejected with near unanimity, and rightly so. Although the Court of Appeal in T.H. rejected the argument that the California Supreme Court overruled Conte in O’Neill v. Crane, we continue to believe that the two opinions cannot be reconciled and that Conte must give way.

Third, make no mistake—the issue is duty, and the Court of Appeal in both Conte and T.H. both engaged in unwarranted expansion of tort duties. Both courts cited the general rule that “all persons owe a duty to use ordinary care,” but that is a gross oversimplification. Legal “duties” are “merely conclusory expressions that, in cases of a particular type, liability should be imposed for damage done.” Christensen v. Superior Court, 54 Cal. 3d 868, 885 (1991) (citation and quotation marks omitted). Duty therefore is a policy determination, and ever since the scales toppled over on that train platform on Long Island and injured poor Mrs. Palsgraf, we have known that duties have limits. Here, the Court in T.H. missed the mark even worse than Conte by focusing so heavily on foreseeability. As the California Supreme Court has observed, “there are clear judicial days on which a court can foresee forever.” Thing v. La Chusa, 48 Cal.3d 644, 668 (1989). There are other considerations, known in California as the Rowland factors. There is, for example, no direct connection between a patient and a company that neither made nor sold the drug she used, and any alleged connection is further strained where the company abandoned the business years earlier. There also is no apparent moral blame to attach to product manufacturer that has not sold the product for several years, had no intentional role in delivering the product to the patient, and who did not promote the product for the alleged off-label use. It also does not help to prevent future harm when the company has already stopped selling the product and can neither stop others from selling nor control how they market. The relative costs weigh heavily against creating a new duty in this context because the organizing principle for product liability law is that a product’s manufacturer/seller is in a position to spread the cost of resulting injuries or to procure insurance. That dynamic becomes moot when imposing a duty on a party that neither manufactured nor sold the product, and all it could possibly do is raise the price of other product for which there is no alleged defect. Even worse, the Court has arguably converted the former listed manufacturer into an insurer for current sellers of terbutaline without compensation and against its will. We could riff like this for a while, but we think you get the point.

Fourth, the opinion impermissibly allowed the plaintiffs to recast their product liability claims as alleged misrepresentation claims. The Conte court did this too, but the plaintiffs’ characterization of their claims should not have changed the analysis. T.H. is a product liability case, and there is no allegation that the defendant wrote or said anything that the plaintiffs ever saw or heard.

The T.H. opinion came to the wrong result for the wrong reasons. We still believe it is a weak case, even accepting its current posture. Even if the defendant owed a duty in 2001, it is long haul to show that the defendant breached that duty or that a breach proximately resulted in injuries to the plaintiffs. It is disappointing, however, that the defendant has to engage in those disputes.