When we examined Restatement (Second) of Torts §402A, comment k (1965), in our 2011 research post “Comment K, Some of the Way,” we remarked about how it said “a lot of things,” including: (1) that one can’t design away unavoidable risks, (2) that unavoidably unsafe products aren’t considered “unreasonably dangerous” for purposes of strict liability, and (3) that such products are “especially common” in the prescription medical product area. The first proposition suggests that design defect liability is incompatible with unavoidably unsafe product characteristics. The second indicates that unavoidably unsafe products shouldn’t be subject to strict liability. The third raises the issue of whether all prescription medical products should be considered unavoidably unsafe products – since their inherent risks are why they require a prescription in the first place.
We pointed out that some states, like California and Utah, prohibit design defect liability altogether, following that aspect of comment k. Others, like Pennsylvania, reject strict liability in the context of prescription medical products. Often, as is the case in Massachusetts, the status of negligent design liability is uncertain.
However, the most striking divide has always been between those states following a case-by case approach to determining whether a given product was “unreasonably dangerous” and those states that followed an “across-the-board” approach to comment k in the prescription medical product area. After adding everything up, we had to conclude in our blogpost that, despite its inherent inefficiency and tendency to second-guess the FDA, the case-by-case approach to comment k remained the majority rule.
After that, aside from it being raised in connection with statutory preemption issues in Bruesewitz v. Wyeth LLC, 562 U.S. 223 (2011) − there hasn’t been many developments on comment k beyond occasional cases. Two of the recent cases we’ve discussed, Tersigni v. Wyeth, ___ F.3d ___, 2016 WL 1128256 (1st Cir. March 23, 2016), and Barcal v. EMD Serono, Inc., 2016 WL 1086028 (N.D. Ala. March 21, 2016), are examples. Tersigni, avoided the issue, because negligent design in Massachusetts requires an alternative design, which plaintiff didn’t have. 2016 WL 1128256, at *3. Barcal applied comment k, 2016 WL 1086028, at *3, because Alabama is a well-recognized across the board state. See, e.g., Bruesewitz, 562 U.S. at 234 n.41 (including Alabama in list of across the board states).
What we have seen recently – and it’s unusual – is two significant law review articles on comment k. The first is James A. Henderson, Jr., & Aaron D. Twerski, “Drug Design Liability: Farewell to Comment k,” 67 Baylor L. Rev. 521 (2015). The second is Eric Lindenfeld & Jasper L. Tran, “Prescription Drugs & Design Defect Liability: Blanket Immunity Approach to the Increased Costs & Unavailability of Prescription Medication,” 64 Drake L. Rev. 111 (1st Quarter 2016). Professors Henderson and Twerski, of course, also served as the ALI reporters for the Restatement (Third) of Torts, Products Liability, so it should come as no surprise that their position is that comment k should be replaced by the Third Restatement’s approach to prescription medical product design issues, which is Restatement Third §6(c). The Lindenfeld/Tran article advocates the across-the-board – what it calls “blanket immunity” – approach to comment k, barring design defect claims altogether.
Nobody seems to have much love for the case-by-case approach, even though it’s still probably the majority rule. Henderson & Twerski view it as a plaintiff-side attack on what ALI intended for comment k to accomplish:
Obviously, the plaintiff’s bar and pro-liability academics never have been very happy with [comment k’s] no-liability rule. . . . [T]he issue became whether Comment k could be reinterpreted to allow for judicial review of prescription drug designs without seeming to reject an important part of William Prosser’s legacy. Beginning in the 1980s, a growing number of courts responded to the challenge by requiring drug manufacturers to prove that the drug in question was unavoidably unsafe as a necessary condition to enjoying the Comment k-based no-liability rule. According to these re-interpretations, a drug that is unavoidably unsafe should not bring design-based liability; but unavoidability should be proven, not presumed. . . . Of course, this interpretation of Comment k leaves drug manufacturers with the unenviable task of proving a negative − that no [alternative design] was available at the time of distribution. And it also presents the question conceptually of how a safer alternative drug, which constitutes a different molecule from the one plaintiff alleges to be defective in design, could be said to be a marginal, rather than categorical, variation of the drug under attack. These, and other, puzzlements appear unavoidably to surround adoption of this new interpretation of Comment k.
Farewell to Comment k, at 530-31 (footnotes omitted).
Lindenfeld and Tran (“L&T”) likewise view the case-by-case approach with considerable scorn.
While the case-by-case approach may seem more appealing because it appears to offer a more individualized determination of each drug’s utility versus its dangers, juries are the improper forum to make that determination. Juries are often exposed to only the narrow focus of the trial and are unable to process the larger picture of public health questions, such as whether the benefits of the medication to a broad class of people outweigh its risks.
* * * *
[A] jury sees only the cost of a more dangerous design, and is not concerned with its benefits; the patients who reaped those benefits are not represented in court. Furthermore, the assertion by many courts that tort liability encourages manufacturers to make safer products, is certainly untrue. As evidenced from the extensive FDA review process, the design of a particular drug is already as safe as can possibly be while retaining maximum effectiveness.
* * * *
Like the case-by-case, risk and utility approach employed by the majority of courts, all companies within the pharmaceutical industry undergo a risk versus benefit analysis of their own when deciding to procure a product or to keep it on the market. The possibility of facing multiple claims under a variety of jurisdictions employing different legal standards is factored into the analysis, and as a result, the product will be procured and sold at a higher price to offset costs of potential litigation, or they will chose not to market it at all.
Blanket Immunity Approach, at 127-30 (footnotes and quotation marks omitted).
Henderson and Twerski (“H&T”), as Third Restatement reporters, advocate the alternative approach taken by Restatement Third §6(c). That is – there isn’t a design defect unless the existing design doesn’t benefit any class of patients: “that reasonable health-care providers, knowing of such foreseeable risks and therapeutic benefits, would not prescribe the drug or medical device for any class of patients.” Farewell to Comment k, at 532-33. OK, but why wouldn’t a combination of FDA regulation (the Agency isn’t supposed to approve that kind of product in the first place, is it?) and market regulation be more efficient? According to H&T, patents and information overload justify allowing juries in effect to ban FDA approved products with no alternative designs:
[T]he reason why the market quite often fails in this regard − consists of a combination of excessive patent protection and informational overload. Patent law becomes complicit in encouraging market failure when it extends the period of patent protection to drugs as those drugs approach the end of their original protective time periods. . . . [W]hen patent law discourages the marketing of newer, better drugs, patients are left without any recourse, either from the market or from tort. . . . [The restatement approach] responds to potential market failures − errors by drug prescribers − caused by the inability of such health care providers to cope adequately with overwhelming quantities of data regarding the comparative risks and efficacies of large numbers of more or less substitutable prescription drugs. . . . [Where] a drug is so inefficacious and risky compared with available alternatives that it should not be prescribed for any class for patients, a strong argument can be made that courts are justified in substituting their independent judgments for those of prescribing physicians.
Farewell to Comment k, at 535-36 (footnotes omitted). They also attempt to justify what is a form of generally disapproved categorical liability by equating FDA-approved drugs to “exploding cigars.” Id. at 539-40. While we sometimes stand in awe of intellect of these two, not as to that. That’s just dumb. Drugs are not a gag item. They are evaluated by the FDA for safety and effectiveness before they are ever marketed, and the FDA can withdraw them should they become “obsolete” or are shown to be “lop-sidedly dangerous.” Id. We’d much rather have the FDA, rather than lay juries inflamed by overwrought “reptile” advocacy (designed to facilitate just this type of claim), make that kind of decision.
We’ve always been somewhat AC/DC about §6(c). Generally, we don’t like it when we’re representing manufacturers of prescription drugs, because, like the Supreme Court said in Mutual Pharmaceutical Co. v. Bartlett, 133 S. Ct. 2466, 2475 (2013), you can’t redesign a molecule – you can only make a different molecule. In the FDCA context, that means a different drug that must go through its own FDA evaluation. Id. So we don’t like the exception that §6(c) leaves open to the usual requirement of an alternative design. In the case of medical devices, however, it’s conceptually a lot easier to envision an alternative design. A plaintiff can argue, for example, that what is made out of stainless steel should have been made out of titanium, or vice versa, or that the device’s dimensions should have been different. So in device cases, the §6(c) exception allows a much smaller amount of liability than pre-existing design defect legal theories. So, in device cases, we’re generally in favor of a Third Restatement approach.
H&T don’t dwell on that.
One other thing we don’t see mentioned by H&T, but what will become an increasing barrier to §6(c) liability in drug cases, is pharmacogenomics. Section 6(c) is based on a “class of patients” rationale, and increasingly those classes will be defined by on human genetic variability. Ultimately, genetic variability is probably behind most differences in human response to the same doses of the same molecule, so the increased ability of science to quantify those differences will identify more classes of patients benefitted by even drugs that other people shouldn’t take.
L&T, by contrast, advocate no strict liability at all. In addition to the costs and inefficiency of case by case, they point out something else that H&T overlook – preemption. Maybe Justice Scalia’s death will ultimately result in the Court retracting the “independence principle” that underlies PLIVA v. Mensing, 131 S. Ct. 2567 (2011), and Mutual Pharmaceutical Co. v. Bartlett, 133 S. Ct. 2466 (2013), but until that happens, design defect claims will largely (we would argue, entirely) be preempted. As we’ve discussed at length, the FDA requires agency pre-approval of any design that would significantly affect the safety or effectiveness of a drug or medical device. But unless a design change does just that − significantly alters safety or effectiveness − it won’t make a difference in a product liability case. Since the manufacturer cannot independently comply with an immediate state-law duty to change a design so as to alter product safety or effectiveness, that creates impossibility preemption. That’s true for any kind of drug and any kind of device. So eventually (unless the Supreme Court changes existing law) all design defect claims that could conceivably be causal in a product liability context will wind up preempted.
The growing trend by the Supreme Court and the lower federal courts to preempt state common law doctrines relating to strict liability in design defects implicates a high degree of inconsistency and uncertainty for drug manufacturers within case-by-case jurisdictions. This also holds true with plaintiffs − who do not always work within a contingency agreement with their attorney and who personally invest significant amounts of money in products liability cases. A blanket immunity approach for all jurisdictions would ensure that pharmaceutical companies − along with injured litigants − can to some extent, predict the verdict of a case and better prepare for the costs of litigation or otherwise.
Blanket Immunity Approach, at 135 (footnote omitted). Thus, if drug/medical device design claims aren’t going to be allowed at all due to preemption, why embrace tort theories that enable them?
But then, inexplicably, after relying on preemption, L&T condemn the Third Restatement because it also limits negligence liability:
Like the blanket immunity approach, the Third Restatement effectively prevents most plaintiffs from being successful in strict liability design defect claims, because one would be hard-pressed to find a scenario in which the FDA approves a prescription medicine that does not have some net benefit for a class of patients. . . . [T]he Third Restatement is less of a “restatement” of the current state of the common law as it is an endorsement of the blanket immunity approach employed by a minority of the courts. The Third Restatement’s approach is undoubtedly a step up from the case-by-case approach in that it dissuades much of the duplicative litigation and protects pharmaceutical companies from many of the costs associated with that litigation. However, the Third Restatement’s approach is even more restrictive than the blanket immunity approach in that it excludes negligence as a distinct and available avenue of liability.
Id. at 136 (footnotes omitted). In this respect, however, preemption is like the Third Restatement. Under Mensing/Bartlett a state-law negligent design claim that would require an “impossible” design change (that is, one that would require FDA pre-approval) will be just as preempted as the state-law strict liability design defect claim that was held preempted in Bartlett. Preemption is based on federal supremacy over state law, not differences between state-law tort theories.
The opposite is also true, a duty to recall claim masquerading as a “negligent design” claim that asserts §6(c) categorical liability is almost as bad as a strict liability claim doing the same thing. That’s why, while L&T, suggest that Lance v. Wyeth, 85 A.3d 434 (Pa. 2014), is a “good middle ground,” 64 Drake L. Rev. at 137, our view of Lance has always been, if that’s negligence, then who needs strict liability? In the end, we think that L&T are trying to have it both ways – trying to look reasonable by relying on preemption and on the strength of FDA review when it suits their purpose, see id. at 127-30 (arguing that “the FDA regulatory scheme is an adequate gatekeeper”), and then promptly ditching both to prevent their “blanket immunity” proposal from resembling, well, blanket immunity.
Being defense lawyers, we prefer it when “blanket immunity” means what it says.