Photo of Eric Alexander

A long time ago in a law school relatively far away, we took torts as a first year law student.  Many of the cases about which we learned (or were supposed to have learned) were from even longer ago and we had no idea how much some of those old cases would inform our practice.  Three old cases in particular, Palsgraf, Wagon Mound I, and Wagon Mound II, were supposed to be important in defining the duty and causation issues in negligence cases and those following negligence principles.  If you throw in Carroll Towing as a fourth key old case in this space, then you can add Judge Learned Hand to Judge (later Justice) Cardozo in the pantheon of great legal minds to weigh in on these, er, weighty issues.  Beyond the Palsgraf pronouncement that “Proof of negligence in the air, so to speak, will not do,” what have we retained from the discourse in these cases?  Probably not much.  But, we submit, most U.S. lawyers can rattle of the negligence elements of duty, breach, and proximate cause without needing to examine why those are the elements.  The touchstone, at least the one we are addressing here, is foreseeability.

Absent statute or contract, the duties a private citizen (including corporate citizen) owes to other private citizens depend on the foreseeability that its acts will impact them.  In general, how much care is reasonable to exercise in a given situation depends on foreseeability in light of the level of risk and other factors.  In the event of a breach, what will define the boundaries of the harms for which the breacher may be liable?  Why, foreseeability, of course.  Like watching pandemic videos of Rube Goldberg machines that sink baskets or fiction about time travel that results in unintended consequences many years later, we can imagine many paths to remote outcomes that may or may not be consistent with proximate cause (and hence liability).  The trick in all of this is to measure foreseeability at the time of the alleged act or omission without just flipping around hindsight from the bad thing that happened.  Hindsight is easy.  Foresight is hard.  On the books at least, the law does not impose liability based on mere hindsight.

Applying these principles to complicated cases with sympathetic, injured plaintiffs can be a different matter entirely.  If you add in some allegations about off-label use and promotion, then it can be particularly difficult to require foreseeability.  These issues come together in Stiens v. Bausch & Lomb Inc., — S.W.3d –, 2020 WL 7266398 (Ky. Ct. App. Dec. 11, 2020), which addressed an appeal of summary judgment for a prescription drug manufacturer.  The underlying facts and procedural history are complex, but many of the details were ultimately unimportant to the resolution at the trial court and on appeal.  Summarized in light of the dispositive issues, the plaintiff suffered a permanent eye injury from the off-label use of a topical antibiotic during a 2012 surgery to correct her nearsightedness.  (Defendant did not contest medical causation on summary judgment, so we will treat issues of risk and specific causation as proven.)  This antibiotic was approved to treat what is commonly known as pink eye; the off-label use of antibiotics as prophylaxis during this type of eye surgeries and others was standard of care.  There was significant disputed evidence about the promotion of the drug for this off-label use, including direct interaction between one of the manufacturer’s sales representatives and the physician who performed the surgery (whose practice was also a defendant in this case and who seems to have been a defendant in a related proceeding over the same injury).  This physician also reported his experience with plaintiff and later patients receiving the same drug as prophylaxis in connection with eye surgeries, proposing that a component delayed healing and created a risk of injury not previously described with its topical use.  Plaintiff’s negligent failure to warn claim under the Kentucky Product Liability Act initially survived summary judgment based on evidence about off-label promotion influencing the doctor’s choice of the drug and 2010 articles about risks associated with injecting the drug into the eye.  Later, this evidence was excluded and the court allowed a rare second bite at the summary judgment apple.  The trial court held that plaintiff failed to identify “any specific identifiable injury causally connected to Besivance that [the manufacturer] knew was possible or should have known was possible,” rejecting plaintiff’s contentions that knowledge of a risk could be presumed in a negligence case and that Kentucky imposed a duty not to promote off-label.  Id. at *5.

On appeal, the focus was squarely on the issue of foreseeability.  In a sense, though, this could have been a narrower inquiry:  failure to warn claims require proof that the warning at issue is inadequate as to the relevant risk at the relevant time, so one could ask only if plaintiff had admissible proof of that inadequacy to raise a genuine issue of material fact.  The appellate court eventually answered that question in the negative, but the journey there has broader implications.  If, as it appeared, plaintiff experienced the first negative outcome with the drug in ophthalmologic surgery that the manufacturer heard about, then there would need to be some legal and factual basis to say the manufacturer should be liable for not warning about this risk before the plaintiff’s surgery.  Within a year of plaintiff’s surgery, due in part to her surgeon’s efforts, there were publications and a medical society alert about the risk at issue.  That sequence invites substituting hindsight for proof of an inadequate warning, but neither court took that bait.

The start of the appellate court’s analysis was Kentucky negligence law, beginning with duty.  “Every person owes a duty to every other person to exercise ordinary care in his activities to prevent foreseeable injury.”  Id. at *6 (citation omitted).

The Kentucky Supreme Court has held “that so far as foreseeability enters into the question of liability for negligence, it is not required that the particular, precise form of injury be foreseeable—it is sufficient if the probability of injury of some kind to persons within the natural range of effect of the alleged negligent act could be foreseen.” Id. (quoting Miller v. Mills, 257 S.W.2d 520, 522 (Ky. 1953)). Such risk must be foreseeable based on what the tortfeasor knew or should have known at the time of the accident rather than what might be deemed foreseeable in hindsight. Bruck v. Thompson, 131 S.W.3d 764, 767 (Ky. App. 2004).

Id. at *7.  The trial court had mistakenly required plaintiff to prove a risk of “the exact injury actually manifested,” but the plaintiff still had to prove something was foreseeable.  Id.  Knowing she had no evidence of foreseeability, plaintiff offered various arguments against such a requirement.

First, plaintiff contended that the Kentucky Product Liability Act imposed strict liability, so that knowledge of a risk could be presumed and foreseeability was not required.  Rather than delve too much into Kentucky law here (but see here for a survey on this issue), we can say simply the appellate court rejected this argument.  Failure to warn claims in Kentucky require proof that the manufacturer “knew or should have known of the inherent dangerousness of the product and failed to accompany it with the quantum of warning adequate to guard against the inherent danger.”  Id. at *8 (citations omitted).

Second, plaintiff contended that she could establish a failure to warn because other Kentucky cases had found drug manufacturers had failed to warn, but the appellate court noted that these other plaintiffs had “evidence that the drug manufacturers had actual knowledge of reported risks or dangers associated with the use of their pharmaceutical products.”  Id. (citations omitted).  Once the trial court excluded evidence of alleged off-label marketing and articles about the risk of injecting the drug into the eye, the plaintiff had “no evidence suggest [the manufacturer] knew or should have known that there could be risks associated with Besivance use at all.”  Id. at *9.  As such, because foreseeability is required, the appellate court could not hold that the manufacturer “had a duty to warn of or test for risks without any reason to suspect that there may be risks associated with this particular use of Besivance.”  Id.

Plaintiff’s third and last argument was that off-label promotion changes these requirements, that a manufacturer’s “promotion of its product for an off-label use without prior testing is a sufficient basis upon which to predicate foreseeability of harm.”  Id.  We have detailed in various contexts (like here and here) how allegations about off-label promotion are thrown around as a basis for civil liability without a logical connection to the complained-of harm but with intermittent success.  With restraint not seen by enough federal courts sitting in diversity, this state court was clear about its role:

Given that Kentucky does not prohibit the off-label promotion and marketing of drugs, we do not believe that it is appropriate to extend the law of negligence in this regard. This is an issue for the General Assembly to take up, not the courts.


The further exposition for this non-expansion was also helpful:

Off-label use is not unlawful under state or federal law. United States v. Caronia, 703 F.3d 149, 166-67 (2d Cir. 2012). Doctors are permitted and even encouraged to prescribe drugs for both FDCA-approved and -unapproved uses for the benefit of their patients. Id. at 153. Courts have repeatedly noted that off-label promotion is not a private right of action that exists under state law. See Aaron v. Medtronic, Inc., 209 F. Supp. 3d 994, 1010-11 (S.D. Ohio 2016) (citation omitted) (“Off-label promotion” is “not a part of [state] law.”); Thorn v. Medtronic Sofamor Danek, USA, Inc., 81 F. Supp. 3d 619, 628 (W.D. Mich. 2015) (“[T]here is no state law duty to abstain from off-label promotion.”); Caplinger v. Medtronic, Inc., 921 F. Supp. 2d 1206, 1219-20 (W.D. Okla. 2013) (“‘[O]ff-label use’ … is not a part of [state] substantive law.”). Stiens has not established an industry or company standard for off-label promotion outside of the FDCA against which B & L’s promotion of Besivance could be weighed. If Stiens wanted to rely upon FDCA regulation of off-label promotions, Stiens should have brought a claim under the FDCA.

Id.  The only bit with which we take issue is the part about bringing a claim under the FDCA.  In a footnote to the third sentence above, the court rightly observed that “private actions for ‘off-label’ marketing” are preempted, id. at *9 n. 8, but these same cases note that the FDCA does not provide a private right of action based on violations of its provisions.  Under Kentucky law, there is no lower bar for proving a failure to warn when a drug is used off-label.  Until there is a reason to know there is a risk that could be described in the label, there is no failure to warn.  In that context, allegations of off-label promotion cannot cure the lack of proof of foreseeability.