Mensing - Brand Name Wins, Generics Lose

The Eighth Circuit gave brand name drug manufacturers cause to give thanks yesterday, but for generics, is was Black Friday. In Mensing v. Wyeth, Inc., No. 08-3850, slip op. (8th Cir. Nov. 27, 2009), the court answered two questions: (1) can a brand name manufacturer be liable for an alleged inadequate warning on a generic drug, where the plaintiff never used the brand name product, and (2) are warning claims against generic manufacturers preempted by the FDCA requirement that generic drugs have the "same" labeling as their brand name bioequivalents.

The court in Mensing said "no" to the first question and "no" again to the second.

Mensing, like almost all the recent branded/generic cases, involves Wyeth's drug Reglan - also sold generically as metoclopramide. The plaintiff used the drug a long time and suffered tardive dyskinesia. Her lawsuit attacking the warnings for this condition named both the generic and brand name manufacturers. The district court entered summary judgment - for non-use of the product (no duty of care) for the brand name and on preemption because the generic warnings would no longer be "the same" for the generic manufacturerer.

The preemption ruling, frankly, doesn't surprise us. Like the "impossibility" argument in Levine, the "sameness" argument in generic drug preemption depends upon Bush-era FDA statements in the Federal Register. The beating that the FDA's 2006 preemption preamble took in Levine pretty much presaged the same treatment for these other FDA statements about "sameness" and the ability of generic manufacturers to alter their warnings. See Mensing, slip op. at 9 n.4. Significantly, the first case that the Eighth Circuit cites is Bartlett v. Mutual Pharmaceutical Co., ___ F. Supp.2d ___, 2009 WL 3126305 (D.N.H. Sept. 30, 2009). Mensing, slip op. at 4. The moment we saw Bartlett, we thought "the generics better have better preemption arguments than they showed here." We remember feeling that way once before - when we first saw the Vaccine Act-based, anti-preemption arguments in the DTP litigation almost 20 years ago. We didn't have a good answer then, and we lost. We have yet to see a good answer from the generics to the arguments raised in Bartlett.

The Eighth Circuit accepts in Mensing essentially the same simplistic argument that the Supreme Court bought in Levine - that because there's no express preemption in Hatch-Waxman (the FDCA amendments that authorized generic drugs), it presumably was content to leave FDA-regulated generic warnings subject to tort attack. Mensing, slip op. at 7.

As far as unilateral label changes are concerned, the court held that generics may maintain "sameness" by simultaneously seeking changes in both generic and brand name labeling for the same compound - completely separate from the "changes being effected" (CBE) process. Id. at 8-9 ("generic defendants could have at least proposed a label change that the FDA could receive and impose uniformly on all metoclopramide manufacturers if approved"). That's the Bartlett argument for which we haven't seen the generics offer a good answer. In other words, "§201.57(e) does not permit generic manufacturers passively to accept the inadequacy of their drug's label as they market and profit from it." Mensing, slip op. at 9.

Implicit in these comments is the FDA's expectation that generic manufacturers will initiate label changes other than those made to mirror changes to the name brand label and that the agency will attempt to approve such proposals quickly. The availability of one particular procedure (the CBE process, on which the district court expended the majority of its discussion) is immaterial to the preemption analysis in light of this clear directive to generic manufacturers and the availability of the prior approval process.

Mensing, slip op. at 10-11. On more or less the same rationale, the court also held that nothing in federal law prohibits generics from requesting the FDA to send out "Dear Doctor" letters about risks beyond those that appear on a drug's labeling. Id. at 11-12.

The court in Mensing then holds that there's no "obstacle" preemption because strengthening warnings does not necessarily require the sorts of expensive clinical trials that Hatch-Waxman did a way with for generic drugs. In this particular case, the preemption defense was undercut by the FDA, after the fact, requiring a stronger warning about the risk in question:

[T]he FDA did not conduct its own studies when it mandated an enhanced warning for metoclopramide. It simply referenced studies published elsewhere. Requests for label changes must be supported by scientific substantiation, but there is nothing to indicate that the information must be acquired through a manufacturer's own clinical tests.

Mensing, slip op. at 14. Imposing the same requirements to monitor ongoing clinical and research experience that pioneer manufacturers must undertake, the court held, did not present any preemptive "obstacle." Id. at 15.

That's the bad news for generics. The good news for brand name manufacturers is in the second half (third?) of the opinion. The court followed the "ovewhelming majority" of courts (including the Minnesota court of appeals - see our branded/generic scorecard) in rejecting an duty of care owed by a brand name manufacturer to any plaintiff that did not consume the branded product. Mensing, slip op. at 16-17. The court rejected the "determinative" weight given to supposed prescriber reliance in the (grotesque) Conte case from California. Mensing, slip op. at 17. Tort law ain't estoppel - it requires a duty:

"[C]entral" to a fraudulent misrepresentation claim under Minnesota law is "a suppression of facts which one party is under a legal or equitable obligation to communicate to the other, and which the other party is entitled to have communicated to him." In other words, regardless of whether her doctor relied upon the [branded] label, [plaintiff] must show that the name brand manufacturers owed her a duty of care. Duty is a threshold requirement for all of the tort claims [plaintiff] asserts.

Mensing, slip op. at 17 (citation and quotation marks omitted). There was no case plaintiff could cite that imposed liability on a defendant that had not, at least, "intended" to communicate with the plaintiff. Id. at 17 n.9. Unlike Conte bare "foreseeability" isn't enough. To extend a duty to warn about a competing product would "stretch[] the concept of foreseeability too far." Mensing, slip op. at 18. In short, good-bye Conte.

Unless the generics can show us a whole lot better preemption argument than we've seen to date, in the post-Levine era, we think they're unlikely to prevail on this ground. Conversely, with the generics likely to be subject to suit for their own labeling defects, the pressure that motivated the judicial tort activism seen in Conte is unlikely to recur. Thus, Mensing is probably the shape of things to come.

Belated tip of the cyberhat to Jeff Pilkington of Davis Graham for letting us know about Mensing.

Free Case Law On-Line From Google

We're delighted that many of our readers are, like us, Luddites.

We know that some of you folks don't understand the meaning of complicated words like "RSS feed," "gigabyte," and "computer."

So we're betting that a few of you aren't yet aware that Google announced last week that Google Scholar now makes available on-line, and free of charge, a broad range of case law and law review articles. Here's the obligatory link.

This puppy is pretty stripped down. It doesn't have headnotes to cases, has issues with sorting results, and doesn't permit searches of, for example, one word "within X words" of another, which are important features.

Over at the Volokh Conspiracy, the head conspirator ran some searches and complained about what came up.

The Wall Street Journal Law Blog quotes folks from Lexis and Westlaw explaining that the availability of cases on-line and free doesn't worry them at all.

Information Today spent a fair amount of time experimenting with the product and posted a pretty favorable review.

Finally, the Supreme Court of Texas Blog walks you through the entire search process.

So, take a look.

Experiment.

See what you think.

And remember: A "computer" is that thing on your desk that you're staring at as you read this post.

Check That - Let's Look At Something Else (Vioxx and Fosamax)

Yesterday we mentioned that Merck had won summary judgment in a consumer protection-type suit brought by the Texas Attorney General concerning Vioxx. We hoped for an "interesting opinion." Well, check that. We've now seen the order - and that's all there is (at least for now) - an order. Here it is, but it doesn't say much beyond the relief granted.

We couldn't have done much with it any way, given Bexis' involvement in (other) Vioxx litigation, so lets move on to something more interesting.

Yesterday was a good day for Merck (unlike ... umm ... well, just insert an appropriate W.C. Fields reference). The company also won summary judgment in another test case in the Fosamax MDL in New York federal court (If you can make it there, you can make it anywhere). Here's a copy of that opinion - which actually is interesting.

The first thing we have to say, just reading the facts, is where do they get these plaintiffs? If this is a test case, we'd hate to see (actually as defense lawyers, we'd love to see), the cases that even the plaintiffs think are weak. This plaintiff - old, disabled from what the opinion describes as smoking-related lung disease, skin cancer and abcesses in the mouth area, severe osteoporosis, and most importantly for a Fosamax case - a lifetime of miserable dental hygene. Slip op. at 3-4. On top of that, this plaintiff didn't even have a permanent injury, but rather healed up after she finally got dental care (or after stopping the drug, which is one of the issues the opinion addresses). Id.

In short, a wonderful combination of weak liability and low damages. But that's modern mass tort practice, for you.

This plaintiff was treated by two dentists. Neither of the treaters was willing to offer much of an opinion that the injury (which we will assume, for purposes of this post, was osteonecrosis - what the Fosamax litigation is mostly about) was caused by the plaintiff's use of Fosamax. Slip op. at 4-7.

The case is under Mississippi law (you know what I mean! - sorry, that's the best we could do for a Mississippi reference off the top of our heads). Plaintiff gives up on three of her four causes of action (design defect, manufacturing defect, negligence per se) right out of the box - and this is a test case - so we're left with a warning claim. Slip op. at 9.

But it wouldn't really have mattered what theory the plaintiff pursued, becuase she couldn't prove causation. Those treating doctor's "opinions" were all plaintiff had to offer on whether Fosamax caused her an injury. Trouble was, one treater didn't have any opinion at all. See Slip op. at 13 ("Plaintiff cannot establish specific causation through the opinion of [the first treater] because, in short, he does not have an opinion").

Fault.

The other treater admitted that he didn't know squat about Fosamax. Slip op. at 16 (the second treater "instead repeatedly qualified his diagnosis by admitting that 'he knows little' regarding the topic"). His methodology, if it can be called that (the court expressed doubts) "reveals no other reasoning for his diagnosis other than the temporal relationship" and vague rumors ("a little hum") in the medical community. Slip op. at 19.

Double fault. And this is an MDL test case? We have to wonder what exactly is being tested.

So this test plaintiff loses because, with all the resources of a big-deal MDL behind her, she can't come up with an expert witness on causation other than these two treaters, one without any opinion and the other unfamiliar with the drug and incapable of even performing a differential diagnosis:

It is undisputed that [plaintiff] has smoked since childhood and has poor dental hygiene. . . . [A]t his deposition, [the second treater] did not rule out or otherwise address other possible causes for the injury to Plaintiff’s jaw, such as trauma or infection. In fact, he suggested a plausible alternative explanation for [Plaintiff's] injury: “[I]t could have healed up, I guess, either from stopping the Fosamax or it could have healed up on its own if it was just an -- just a horrible looking abscess.”

Slip op. at 20. And the coup de grâce: this supposed "expert" "when pressed at his deposition regarding his opinion, stated that he 'could not say' whether Fosamax was a cause or contributing factor of Plaintiff's injury." Id. at 22.

At the risk of repeating ourselves - this is an MDL test case? Unless the defendant got to pick this case (and defense picks, when permitted, usually end up being voluntarily dismissed by the plaintiffs), we have to doubt (yeah, we're defense counsel, that's what we do) whether there's any "there there" (apologies to Oakland) in the Fosamax MDL.

And congrats to Merck and its lawyers for yesterday's one-two punch.

Merck Wins Vioxx AG Case In Texas

Bexis can't say much anyway, since his firm's involved in Vioxx, and neither of us have seen the opinion, but Merck announced today that it was granted summary judgment today in the Texas state AG consumer protection litigation. Here's a link to Merck's press release. When we get a copy of what promises to be an interesting opinion, we'll pass it along.

If any of our readers have the opinion, we'd appreciate you favoring us with a copy.

Zipursky On Preemption "In The Spirit Of Conversation"

Professor Ben Zipursky (Fordham) just published a guest post over at Torts Prof Blog ruminating about why the Supreme Court found preemption in the context of medical devices that had undergone premarket approval in Riegel v. Medtronic, but did not find preemption in the context of prescription drugs on the facts of Wyeth v. Levine. That's an interesting post, and we commend it to you.


We have a couple of reactions to the post, one general and one specific.


First, the general: As you know, we're big believers in preemption at this blog, and we think the Supreme Court should have found preemption in both Riegel and Levine. And we're still optimistic that courts will find preemption in the context of drugs that have regulatory histories more compelling than the Phenergan involved in Levine.


But, in the spirit of conversation, let's think for a minute about whether Congress could reasonably have wanted to create a world in which preemption existed for certain medical devices, but not for drugs. We think it could have, for three reasons.


The first reason is historical accident. States were not regulating drugs very much in 1962, when Congress revamped the drug laws. But by 1976 that had changed, and states were regulating medical devices at the time Congress revamped the device laws - most notably California, which in 1970 enacted a law requiring its own premarket approval of medical devices. The Medical Device Amendments followed close on the heels of the Dalkon Shield controversy, which prompted states to act. With state regulation in the offing, Congress could reasonably have focused on the need for preemption in 1976 in a way that never occurred to it in 1962. This is particularly true with the most cutting edge, and highest risk/highest benefit devices, which after 1976 has to be premarket approved - the type of device at issue in Riegel. Had this been on Congress' radar screen in 1962, it might have done the same thing with PMA drugs.


The second reason for Congress to have distinguished between drugs and devices is grounded in policy. Historically, drug companies have been large. Bringing a new drug to market can take ten years and cost a billion dollars; drug companies have to be big. But historically, device companies were smaller. An orthopedic surgeon could step out into his garage, gin up a valuable implant with the tools he had on hand, and then start a small device company. These smaller companies have fewer resources to withstand the double jeopardy of first running the gauntlet of the FDA's approval maze, and then being second-guessed under the disparate laws of the 50 states. Don't get us wrong - we think such double jeopardy shouldn't be tolerated in either case, but we're not Congress. Perhaps Congress thought that large drug companies did not need the protection of preemption as much as small device companies did. In fact, the Senate said as much back in 1976, mentioning that its consideration was informed by the importance of the small innovative manufacturer in the invention and development of new medical devices.


The third reason is also grounded in policy. The useful life of a drug can be very long. Penicillin was discovered decades ago, but still serves a purpose, and thus is regularly prescribed. And, back in 1962, before Hatch-Waxman, drugs received patent protection for a long time, so drug companies could sell drugs profitably for many years. The useful lives of medical devices are frequently much shorter. It would be uncommon for a medical device invented today to remain state-of-the-art a decade from now. Perhaps Congress thought that drug companies, which could earn profits on their products for a long time, needed the protection of preemption less than device companies, whose products are not profitable for as long.


That's our general reaction to Zipursky; here's the specific.

We don't think that the difference between Levine and Riegel can be explained by calling Levine a "warning" case and Riegel a "design" case. For one thing, Levine did not say anywhere that design defect claims involving prescription drugs would be preempted. And we're not as confident as Prof. Zipursky that Levine should be considered a true "warning" case. After all, plaintiff never did offer an alternative warning in Levine and argued primarily that the drug should not have been used at all for that particular indication. 129 S. Ct. at 1194. Nor did Riegel hold that warning claims involving PMA medical devices escaped preemption - quite the contrary. Riegel specifically held warning claims were preempted. 128 S. Ct. at 1011 ("the MDA would pre-empt a jury determination that the FDA-approved labeling for a [device] violated a state common-law requirement for additional warnings").

Having litigated a lot of device cases in our lives, we don't agree with Prof. Zipurski that "most" device cases involve design defects. Certainly more do than in the drug area, since it's hard to conceptualize how in most drug cases one would change the "design" of a product and still have the same drug. But let's face it, design defect cases are hard - in most places a plaintiff has to have an alternative design that arguably works better, or at least more safely. That's expensive evidence to obtain, if it can be done at all. Warning claims are much easier to bring, since it's easy and cheap to second-guess a label. Thus, even in the device area, we'd say that there are still substantially more warning than design defect cases.

As litigators, we tend to take what courts say at face value. Since Levine and Riegel don 't purport to draw distinctions between warning and design claims, we're not prepared to read in such a distinction. But if we were inclined to psychoanalyze the Court, we would look more to Riegel as involving an off-label use, and Levine involving a labeled use. We could see the Court (although, of course, neither opinion says so) being less inclined to permit liability where a defendant's product caused injury while being used in a way inconsistent with its FDA-approved labeling, and more inclined to permit liability where the product was being used for an intended purpose, but nevertheless resulted in injury. But again, that's only rank speculation unsupported by anything in either opinion - and we think there should be preemption in either case, as long as the FDA considered the risk in question and made a decision about what the label should say about that risk.

But then again, maybe the most relevant distinction is between law professors, who are paid to think about such things, and us practicing lawyers, who aren't.

Questioning McQuiston

Here's today's question: What do we think of McQuiston v. Boston Scientific Corp., No. 5:07-cv-01723-TS-MLH, slip op. (W.D. La. Nov. 19, 2009) (link here)?

Here's today's answer: We love it!

But not for the reasons you'd think.

Here's the deal: On September 1, 2006, Dr. Thomas Brown performed an angioplasty and stent procedure on James McQuiston. Boston Scientific manufactured the stent. McQuiston said his condition deteriorated after the stent was placed, and he filed a typical product liability complaint against Boston Scientific. (There are actually two plaintiffs, because McQuiston's spouse made a claim for loss of consortium.)

The stent was a Class III medical device that had been approved through the premarket approval process. Judge Stagg thus trots through the holding of Riegel v. Medtronic, 552 U.S. 312 (2008), and routinely applies that holding to find McQuiston's claims of design defect, inadequate testing, failure to warn, breach of express and implied warranties, and manufacturing defect to be preempted. McQuiston, slip op. at 13-16.

What's noteworthy there?

Not much. Although Riegel did not involve warranty claims, an earlier Fifth Circuit case did, so Judge Stagg found the warranty claims preempted. Id. at 15-16. Maybe that's worth mentioning.

McQuiston also pleaded state law negligence and fraud claims. The court dismissed those because they were displaced by the Louisiana Product Liability Act, which provides the exclusive theories of liability against product manufacturers in Louisiana. Id. at 16-17.

But that's still pretty routine. Where's the good stuff?

Our reader who sent us the copy of McQuiston liked the fact that the court refused to allow McQuiston to raise at the eleventh hour a "parallel requirements" claim -- a claim that preemption doesn't apply because Boston Scientific had allegedly violated requirements imposed by the FDA. The court rejected that attempt because there were no corresponding allegations in the complaint and McQuiston had known for over a year that Boston Scientific intended to raise preemption as an affirmative defense. Id. at 18-19.

We like that, too, and we're delighted that you shared the opinion with us. (Thanks!)

But we noticed two other things about this opinion that we liked even more:

At footnote 1 on page 4 of the opinion, the court recites the regulatory status of the Boston Scientific stent -- and it's not purely a Class III device. Rather, it's a "combination device," which involves both drug and device elements. (It was a paclitaxel-eluting stent.) For this stent, the Secretary of Health and Human Services decided that the "primary mode of action" of this combination product was to fulfill a "device function."

We really like that. Cases involving combination devices are hard to come by, and we're delighted to have another precedent to add to that relatively short list. (Riley v. Cordis Corp., ___ F. Supp. 2d ___, 2009 WL 1606650 (D. Minn. June 5, 2009), is another combination device case. It's right there on our (New) Medical Device Preemption Scorecard.)

Also, at footnote 5 on page 17, the court notes that the state law fraud claim may include an imbedded claim of fraud on the FDA -- the implication that Boston Scientific misled the FDA about some aspect of the device. The court properly finds that implied claim to be preempted under Buckman Co. v. Plaintiffs' Legal Comm., 531 U.S. 341 (2001).

So we really like McQuiston, although for reasons that some might think obscure.

Latest Class Action Decision Shows That The Battles Were Important

We aren’t the first to note the latest class action denial in the prescription medical product liability field, In re Panacryl Sutures Products Liability Cases, No. 5:08-MD-1959-BO, slip op. (E.D.N.C. Nov. 13, 2009). 360 (subscription only) beat us to the punch the other day. But just because we aren’t first, doesn’t mean there’s nothing in Panacryl worth blogging about.

We view decisions like Panacryl as vindication of all the fuss we made for more than two years about the ALI’s Principles of Aggregate Litigation project.

We’ve pointed out before – and consider it a significant legal accomplishment – that class certification in personal injury actions involving prescription medical products are routinely denied these days. Earlier this year, we gave the defense bar its highest grade – an “A” – for largely eliminating this sort of class action. We said then:

[I]n the late 1980s, we had to take class actions in product liability litigation very seriously. While there were never a lot of certifications, there were enough of them that – during the Bone Screw litigation, for example – plaintiffs would argue that there was some sort of “modern trend” favoring certification of personal injury class actions. . . .

Then our side prevailed in [Amchem and Ortiz]. After that – with a lot of blood, sweat, and good legal argument from our side – class actions (at least successful ones) largely disappeared from mass torts. . . . The few courts willing to certify class actions in drug and medical device cases have so far gotten shot down on appeal. . . . And with the enactment of CAFA, most class action decisions going forward, and essentially everything in mass torts, will be made by federal courts applying post Amchem/Ortiz law. . . .

As a measure of how far out of the mainstream tort class actions have become over the last couple of decades, the ALI’s Aggregate Litigation principles project, for all its pro-plaintiff leanings in other areas of the law, states quite clearly that personal injury class actions are disfavored for a variety of reasons.

“Taking Stock” post.

More or less removing the threat of class actions in mass tort litigation involving prescription drugs and medical devices has gone a long way to making the risks of this type of litigation manageable – as opposed to the existential threat pharmaceutical mass torts posed back in the days of Bone Screw and Fen-Phen. With the class action threat gone, the likelihood of a mass tort settlement gone wild has become quite remote.

Cases like Panacryl remind us why that is. In Panacryl, the plaintiffs claimed that the defendant’s synthetic surgical stitches (a medical device) were defective and caused or facilitated various types of post-surgical infections. Mass tort litigation followed an FDA Class II recall (It seems like every recall of a drug or medical device turns into a mass tort these days, doesn’t it?), and in the inevitable MDL, plaintiffs sought a nationwide class action that would have joined together more than 2 million users of these stitches – the vast majority of whom suffered no complications from using this product.

Because it was a nationwide class action, the first and foremost question raised was whether the law of all fifty states would apply. If it did, then there’s lots of law holding that the multiplicity of jurisdictions makes renders the class action uncertifiable. What did plaintiffs argue? Why, that the defendant’s principal place of business should apply to all claims, of course. Slip op. at 4. Last refuge of a scoundrel and all that.

Fortunately, the Panacryl court rejected that argument:

First, the court had no trouble finding that the states’ approach to products liability created numerous conflicts. Slip op. at 4-5. Well . . . duh. You've got a court in California, but nobody else, allowing "misrepresentation" liability for a brand-name company in a generic drug case. Then you've got West Virginia, but nobody else, rejecting the learned intermediary rule. Could anybody really deny the conflict question with a straight face?

So what does a court do with that conflict? Before ending up in the MDL, these plaintiffs filed in New Jersey, a state known both for having a lot of drug and device companies headquartered there, and for having relatively (not as much as before, though) pro-plaintiff substantive law. That meant that New Jersey choice of laws principles applied.

That means the court had to decide whether principal place of business could trump the place of injury as a choice of laws principle. Well, that’s one of the main issues that had to be fought out in the ALI’s Principles of Aggregate Litigation project. In fact, one of the three floor motions at the ALI annual meeting last May, was on that precise subject. We – and a lot of other folks – fought long and hard to keep principal place of business as a choice of laws rule out of the final Principles. In that, we were successful. The result, as we reported shortly after the meeting, was:

Principal place of business and choice of law – The reference is still there, but the next sentence now reads: “At the present time, choice-of-law principles that point towards application of the law of the defendant’s principal place of business remain quite rare across the various states.” This position is also now described as an “outlier” in the Reporters' Notes. The Schwartz amendment will add language to the effect that whatever choice of law principles are applicable to litigation generally apply equally to aggregated litigation. We’d rather not see principal place of business mentioned at all, but short of that, it’s essentially a complete fix.

Final report on ALI Principles of the law of Aggregate Litigation, available here.

The defendant in Panacryl succeeded, too. Relying on a recent and directly on point case out of the New Jersey Supreme Court, Rowe v. Hoffman LaRoche, Inc., 917 A.2d 767 (N.J. 2007) (which we reviewed here), the court held that in personal injury actions, the individual plaintiff’s home state/place where the injury occurred should control:

[H]aving considered the contacts relevant to the competing interests of the states in light of Rowe, this Court concludes that the competing interests of the states, the most important factor, weighs in favor of applying the law of each plaintiff’s home jurisdiction.

Panacryl, slip op. at 10.

That’s one – and it’s the big one. Once the court concluded that the plaintiffs’ nationwide class action had to be governed by the law of all 50 states, denial of class certification occurred pretty much as a matter of course.

Except. . . .

Plaintiffs argued that, if the court couldn’t certify the whole class, it should still give them half a loaf and certify an single-issue class action under Fed. R. Civ. P. 23(c)(4). Slip op. at 18-19.

That’s another biggie – the misuse of “single-issue” certifications where the litigation as a whole is too diverse to be certified. This issue, as well, was a major bone of contention in the ALI’s consideration of the Principles of Aggregate litigation. Making sure that single-issue classes stayed rare was the subject of another of the floor amendments offered last May at ALI. Again, we think we got most of what we wanted:

Broad Use of Single Issue Classes – this is the subject of the Beisner amendment that the Reporters largely agreed to. We expect it to end up some neutral language and a statement regarding the law being divided on the point. We expect a substantial, if not complete, fix.

Final report on ALI Principles of the law of Aggregate Litigation, available here.

We pointed out back then that the bulk of precedent rejected the use of Rule 23(c)(4) as a way to break up inherently individualized litigation into bite-sized, certifiable bits. The leading case is Castano v. American Tobacco Co., 84 F.3d 734 (5th Cir. 1996), which flatly held that a “district court cannot manufacture predominance through the nimble use of [bifurcation],” because “a cause of action, as a whole, must satisfy the predominance requirement of (b)(3).” Id. at 745 n.21. Rule 23(c)(4) isn’t a way to make the predominance requirement of Rule 23(b)(3) disappear:

Reading Rule 23(c)(4) as allowing a court to sever issues until the remaining common issue predominates over the remaining individual issues would eviscerate the predominance requirement of Rule 23(b)(3); the result would be automatic certification in every case where there is a common issue, a result that could not have been intended.

84 F.3d at 745. The problem with broad use of single-issue certifications is that it invites courts to ignore what’s left over – which is entirely improper. Cf. McLaughlin v. American Tobacco Co., 522 F.3d 215, 234 (2d Cir. 2008) (rejecting piecemeal certification; “given the number of questions that would remain for individual adjudication, issue certification would not ‘reduce the range of issues in dispute and promote judicial economy”); In re St. Jude Medical, Inc., 522 F.3d 836, 841-42 (8th Cir. 2008) (Rule 23(c)(4) certification improper where “trials will still be required” for remaining individualized issues).

And beyond Castano, a raft of federal district court decisions have reached the same conclusion – a lawsuit that’s individualized and not certifiable as a whole can’t be broken down into smaller, certifiable parts by the use of Rule 23(c)(4). We start with Blain v. Smithkline Beecham Corp., 240 F.R.D. 179 (E.D. Pa. 2007), one of Bexis’ cases that we discussed here. Blain rejected Rule 23(c)(4) single-issue certification, holding that “only after the court has found that the cause of action satisfies the predominance requirements of Rule 23(b)(3) may it certify common issues pursuant to Rule 23(c)(4).” Id. at 190. See also Rowe v. E.I. duPont de Nemours & Co., 2009 WL 2424086, at *2 (D.N.J. July 29, 2009) (“Because the medical monitoring . . . is not applicable to the “class as a whole” . . . [it has] not met the 23(b)(2) requirement and, thus, certification under 23(c)(4) would be improper”); Henry v. St. Croix Alumina, LLC, 2008 WL 2329223, at *4-5 (D.V.I. June 3, 2008) (“a claim for relief taken in its entirety must satisfy the predominance requirement of Rule 23(b)(3) before the court may select certain issues for piecemeal certification”); In re Welding Fume Products Liability Litigation, 245 F.R.D. 279, 312 (N.D. Ohio 2007) (“a court must not manufacture adherence to the requirements of Rule 23 through the nimble use of subdivision (c)(4)”); In re General Motors Corp. Dex-Cool Products Liability Litigation, 241 F.R.D. 305, 314 (S.D. Ill. 2007) (same); Taylor v. CSX Transportation, Inc., 2007 WL 2891085, at *14 (N.D. Ohio Sept. 28, 2007) (“Plaintiffs must still satisfy that the issues to be determined by class adjudication predominate over the claims as a whole, including the claims in the separate individual issue trials.”); In re Katrina Canal Breaches Consolidated Litigation, 2007 WL 2363135, at *1 (E.D. La. Aug. 16, 2007) (“Rule 23(c)(4) issue certification is allowed only if the Rule 23(b) requirements are first met as to the claim”); O’Neill v. The Home Depot U.S.A., Inc., 243 F.R.D. 469, 481 (S.D. Fla. 2006) (“[n]or may the Court certify a single issue when the case as a whole fails to meet the requirements of Rule 23”); Fisher v. Ciba Specialty Chemicals Corp., 238 F.R.D. 273, 316 (S.D. Ala. 2006) (“courts have emphatically rejected attempts to use the (c)(4) process for certifying individual issues as a means for achieving an end run around the (b)(3) predominance requirement”); Hyderi v. Washington Mutual Bank, FA, 235 F.R.D. 390, 398-99 (N.D. Ill. 2006) (quoting and following Castano); Hamilton v. O’Connor Chevrolet, Inc., 2006 WL 1697171, at *6 (N.D. Ill. June 12, 2006) (“a class action movant cannot gerrymander predominance by suggesting that only a single issue be certified for class treatment . . . when other individualized issues will dominate or be meaningfully material to the resolution of the absent class members’ claims”); Snow v. Atofina Chemicals, Inc., 2006 WL 1008002, at *9 (E.D. Mich. March 31, 2006) (“Rule 23(c)(4) may not be used to circumvent the predominance requirement”); Perez v. Metabolife International, Inc., 218 F.R.D. 262, 273 (S.D. Fla. 2003) (“sub-issues cannot be separated out from those that require individualized treatment unless the common issues in the action as a whole predominate”); Rink v. Cheminova, Inc., 203 F.R.D. 648, 651 (M.D. Fla. 2001) (finding Castano analysis of Rule 23(c)(4) “persuasive”); Robertson v. Sikorsky Aircraft Corp., 2000 WL 33381019, at *19 (D. Conn. July 5, 2001) (“[a]n action must be considered as a whole in order to determine whether or not the predominance requirement has been satisfied”); Neely v. Ethicon Inc., 2001 WL 1090204, at *5 (E.D. Tex. Aug. 15, 2001) (Rule 23(c)(4) “does not operate independently from the rule of predominance found in 23(b)(3)”; predominance inquiry cannot be limited to “common issues” alone); In re Jackson National Life Insurance Co. Premium Litigation, 183 F.R.D. 217, 225 (W.D. Mich. 1998) (“certification of the question . . . is inappropriate, for the cause of action as a whole certainly does not satisfy the predominance requirement of Rule 23(b)(3)); Arch v. American Tobacco Co., 175 F.R.D. 469, 496 (E.D. Pa. 1997) (“Plaintiffs cannot read the predominance requirement out of (b)(3) by using (c)(4) to sever issues until the common issues predominate over the individual issues”), aff’d, 161 F.3d 127 (3d Cir. 1998).

To this line of cases can now be added the Panacryl decision. The court held, as to Rule 23(c)(4):

But Rule 23(c)(4) may not be used to manufacture predominance for purposes of Rule 23(b)(3). Plaintiff’s trial plan does not eliminate the necessity of applying the laws of several jurisdictions or the individualized inquiry into whether [the product] caused each plaintiff’s injuries.

Slip op. at 19 (Castano quote omitted).

Panacryl thus demonstrates why engaging on the ALI’s Principles Project for Aggregate Litigation was both a good and necessary thing to do. In Panacryl, as in many class actions involving prescription drugs and medical devices, the key certification questions are those that were most controversial in the ALI – strained choice of law arguments to get around the importance of the plaintiff’s residence in personal injury actions, and slicing, dicing, and pureeing causes of action to avoid the predominance requirement for damages class actions. Of the three main issues that we took to the floor of the ALI, only medical monitoring didn’t raise its ugly head in Panacryl.

There's always good reason to fight the good fight.

There's No Claim for Negligent Misrepresentation on the FDA Either

We had to comment on the most intriguing case of Hughes v. Boston Scientific Corp., 2009 WL 3817586 (S.D. Miss. Nov. 12, 2009). Hughes involved a PMA device, something called a "HydroTherm Ablator," that allegedly malfunctioned and injured the plaintiff. Thereafter, the defendant (1) changed its adverse event reporting algorithm in response to FDA concerns, and (2) still later, instituted a Class II recall of the device.

Faced with the inevitable Riegel-based preemption motion, the plaintiff did two interesting things: First, she tried to turn what was basically a warning-based cause of action into a "manufacturing defect" claim:

[T]his so-called “manufacturing defect” manifests itself through [defentant's] failure to properly label, warn and correct perceived faults in the [device] because [defendant] failed to provide proper injury and malfunction data to the FDA. Under the plaintiff's theory, if [defendant] had properly informed the FDA of the [device's] fault rate and malfunction errors, then the FDA would have required different or supplemental warnings and labeling and perhaps manufacturing changes.

2009 WL 3817586, at *10.

The second thing plaintiff did was to relabel what was essentially a fraud-on-the-FDA claim as "negligent misrepresentation" and "negligence per se" - all based upon the defendant's changing the algorithm upon which its adverse event reporting was based. 2009 WL 3817586, at *10.

Neither of these innovations worked, although the court did call the plaintiff's theories "novel and intriguing." 2009 WL 3817586, at *10.

The court quite correctly held, in light of the policies that led to preemption in Buckman, that scienter - whether the alleged mis-reporting to the FDA was considered "fraudulent" or "negligent" - didn't matter a hill of beans. It was still second guessing the adequacy of the defendant's reports to the FDA combined with speculation that, with "true" information, the FDA might have done something differently that would have prevented the device from being used. As such, that raised all the policy grounds that supported preemption in Buckman:

The Buckman holding did not turn on intentional versus negligent violation of FDA regulations, but on the principle of maintaining a “federal statutory scheme” put in place by Congress. The Supreme Court found that allowing the Buckman plaintiffs' claim would “exert an extraneous pull on the scheme established by Congress, and it [was] therefore preempted by that scheme.” Under this rationale, claims asserting misrepresentations, intentional or otherwise, made to the FDA regarding Class III medical devices are preempted by federal law.

2009 WL 3817586, at *11 (Buckman citations and quotes omitted). In other words "negligent misrepresentation on the FDA" is just as preempted as fraud on the FDA, and for the same reasons.

But there's more. "Novel and intriguing" theories produce novel and intriguing opinions - especially when rejecting such theories. Otherwise they can produce really bad law.

The court then turned both barrels on the negligence per se claim - by both barrels, we mean both state law (failure to state a claim) and federal law (preemption).

The court recognized that FDCA-based negligence per se was widely rejected as a matter of state law - because state negligence per se doctrines do not operate in the face of contrary legislative intent. "[N]umerous courts have rejected negligence per se claims based on alleged violations of FDA regulations as contrary to Congressional intent." 2009 WL 3817586, at *11. We're pleased to note that, of the nine decisions in the ensuing string citation, six of them involved orthopedic bone screws (where we first came up with these state-law-based defenses to FDCA-based negligence per se). As we said yesterday - you scratch our precedent, we'll scratch yours.

Citing another Bone Screw case (if the defendant was Sofamor, Danek, Smith Nephew, or Acromed, and the date is between 1995 and 2001 - chances are its a Bone Screw case), the court goes on to recognize that only "substantive violations" (your product didn't conform to the intended design) rather than "administrative violations" (failure to go through FDA procedures properly) could qualify as negligence per se under state law:

[S]ince such administrative requirements lack independent substantive content, they do not impose a standard of care, the breach of which could form the basis of a negligence per se claim. The defendant contends likewise, i.e., that the reporting regulations contained in the MDA which the plaintiff maintains were not properly met in this case are the type of administrative tools used by the FDA to facilitate the administration of its underlying regulatory scheme and are, thus, administrative, and not substantive, federal regulations. This court agrees.

2009 WL 3817586, at *12. That's two important non-preemption based defenses to FDCA-based negligence per se. They do come in handy.

Switching back to preemption, the court in Hughes observed that negligence per se claims remained preempted where they "would necessarily impose requirements on the device that are ‘different from, or in addition to,’ the requirements of the FDA." 2009 WL 3817586, at *12. That's another defense to remember in this type of case - the purportedly "parallel" claim may well, upon closer examination, not really be "parallel" at all.

[T]here is more to a negligence per se claim (under applicable state law) than a simple prima facie showing that a statute or regulation has not been followed. In other words, the violation in and of itself does not constitute negligence per se. Rather, the defendant may still prevail by showing that he or she acted reasonably under the circumstances. As such, common law tort principles of the reasonableness of the defendant's actions invade the cause of action. Once a state's tort law becomes ingrained with the cause of action, it runs afoul of Riegel.

2009 WL 3817586, at *12 (discussing Bausch v. Stryker Corp., 2009 WL 2827954,(N.D. III. Aug. 31, 2009)). The court concluded this would be the case with plaintiff's negligence per se under Mississippi law.

This court elects to follow the well reasoned and majority rule rejecting negligence per se claims arising out of violations of FDA regulations. Further, the court finds that the attempt by the plaintiff to enlarge or expand the reach of Riegel by claiming that misrepresentation of, or failure to report correct data, to the FDA amounts to a manufacturing defect, while a novel theory, is not warranted under existing law.

2009 WL 3817586, at *12.

Finally, the plaintiff tried to claim that the subsequent recall voided the defendant's device approval and precluded preemption. Harkening back to some of the things we discussed in our "Total Recall" post, the court held that, not only didn't the recall not preclude preemption ("this recall only further serves to prove that regulation of Class III medical devices such as the HTA is the exclusive province of he FDA, 2009 WL 3817586, at *12) - it also amounted to an inadmissible subsequent remedial measure:

[E]vidence of subsequent remedial measures is inadmissible to prove negligence, demonstrate culpable conduct in a breach of warranty claim, or establish product defect. . . . Thus, any evidence of the voluntary HTA recall is irrelevant to the facts of this case and inadmissible under Rule 407, especially in the face of complete preemption of the plaintiff's claims.

2009 WL 3817586, at *12.

Hughes is definately a precedent we're going to be scratching whenever we come up against FDCA-based negligence per se claims.

Doing Good By Doing Well

We just love to see our victories put to good use by other lawyers in later cases.

So we got a big kick out of Pustejovsky v. Wyeth, No. 4:07-CV-103-Y, 2009 U.S. Dist. LEXIS 101513 (N.D. Tex. Sept. 4, 2009).

We didn't get a big kick out of the facts, since we had nothing to do with them: Dr. Collini prescribed generic Reglan to treat Pustejovsky's acid reflux. After using the drug for three years, Pustejovsky developed the movement disorder "tardive dyskinesia." She filed a product liability complaint against assorted manufacturers of generic Reglan.

In earlier decisions, the trial court denied PLIVA's motion for summary judgment on the ground of preemption and granted Wyeth and Schwarz Pharma's motions because they hadn't manufactured the drug that Pustejovsky ingested. Id. at *3-4. In the new decision, PLIVA sought summary judgment based on the learned intermediary doctrine.

The treating physician, however, hadn't relied on anything PLIVA said in deciding to prescribe the drug. Dr. "Collini had not read the package insert . . . and, as a result, did not rely on the warnings included by PLIVA in deciding whether to prescribe" the drug. Id. at *9. "Instead, Collini relied on what she learned in medical school, continuing medical education, her own experience, and the experience of her colleagues in weighing the risks and benefits" of the drug. Id. "Because Collini was aware of the possible risk involved . . . but decided to prescribe it to Plaintiff anyway, the allegedly inadequate warning was not a producing cause of Plaintiff's injury." Id.

Fine, fine, fine.

But we weren't tickled until we read the last chunk of the decision.

Pustejovsky argued that, "based on testimony by other, non-treating doctors, Collini 'would likely have modified her practice as well.'" Id. at *10. The argument is that, even though this particular treating physician denied that the package insert affected her decision to prescribe, other physicians said that a different label changed their prescribing habits, and proof of what others did should suffice to avoid summary judgment.

That issue was front and center back when Herrmann briefed and argued, and Bexis supported with an amicus brief (are we an ugly tag team, or what?) the defense position in Ackermann v. Wyeth, 526 F.3d 203 (5th Cir. 2008). There, the Fifth Circuit went our way, holding that (at least under Texas law) the issue is the subjective conduct of the actual treating physician, not an objective standard of how a supposedly "reasonable" physician would have acted. (That argument is at pages 32 to 36 of Wyeth's opening brief in Ackermann; the brief is available on Westlaw at 2007 WL 5746840.)

In Pustejovsky, the Northern District of Texas naturally followed the Fifth Circuit's lead. The court held that evidence of what non-treating physicians would have done was "too speculative to rebut Collini's unequivocal statements that she did not rely on or even read PLIVA's warning in prescribing" the drug. Pustejovsky, 2009 U.S. Dist. LEXIS, at *10. Additionally, Pustejovsky "must rebut PLIVA's evidence showing that an allegedly proper warning would not have changed the decision of Collini, the treating physician." Id., citing Ackermann (bold in original).

That's proof that defending pharmaceutical product liability cases is a team effort. You scratch our precedent; we'll scratch yours. Keep up the good work.

Rimbert - No Do-Overs

We've discussed the New Mexico case of Rimbert v. Eli Lilly before twice: once concerning the learned intermediary rule, and a second time when the plaintiff's only expert got Daubertized.

Well, the third time's the charm. Rimbert has finally been dismissed. See the order here. Summary judgment was not really surprising after the plaintiff's only expert is excluded. What was is how hard the plaintiff still fought dismissal even after losing her expert.

Plaintiff's latest claim was that she should get a do-over; that is after, standing and falling with one expert, plaintiff should be allowed to take the case back to square zero with a replacement expert. But the plaintiff made it easy for the court to deny the motion. She hid the ball until the end - not even identifying who the "new" expert might be. Slip op. at 6. That, and the fact that the plaintiff had already played out the string until the end with her prior, excluded expert, led to denial of the motion to substitute a new expert. Slip op. at 7-8.

Moral of story: You can't go to the bullpen after the game is over.

Congratulations to Andy See at Shook Hardy for winning the case (and for passing along the news to us). Not asleep at the wheel, that one.

Already Out of Date....

That's us.

At least, that's what blog reader Geoff Klingsporn over at Davis, Graham just told us about our new Conte/generic drug non-liability scorecard. We're not offended, because the reason for our falling short is the best, a new decision - from West Virginia this time - throwing out yet another attempt by a plaintiff to hold a branded drug manufacturer liable for a supposed labeling defect in a generic drug it never made. The new case is Meade v. Parsley, C.A. No. 2:09 -cv-00388, slip op. (S.D.W. Va. Nov. 13, 2009). The best part of Meade? It's explicit rejection of Conte, of course:

So far, Conte, which recognized but declined to follow Foster, is the only decision in several like actions that has allowed the plaintiff to proceed against [a brand-name defendant] when only the generic version of the drug was ingested. . . . Inasmuch as the remaining claims against [the brand-name defendants] require a duty of care to the plaintiffs or proximate cause, summary judgment is proper as to [them].

Slip. op. at 9 (discussion of Foster omitted).

Keep it up guys. We hope Conte gains even less traction than New Jersey's oft-rejected DTC exception to the learned intermediary rule.

Another Trip Through Scholarship

The law reviews just keep on coming.

And one in a thousand articles keeps begging to be read.

We liked Nicholas Pace and William Rubenstein's RAND Working Paper titled, "How Transparent are Class Action Outcomes?: Empirical Research on the Availability of Class Action Claims Data" (on SSRN here). Their thesis is not exactly a news flash: It's very difficult to track what happens in a class action claims process, such as how many class members actually receive payment and, if so, how much. But it's nice that someone looked at this empirically, even if it's only a small sample. In the words of one (of the three) paragraphs of the abstract:

"This paper examines the extent to which claiming data are available and recommends ways to increase transparency in this area. We reviewed the official court files in a sample of 31 class action settlements and we also made direct inquiries to the judges, lawyers, and settlement administrators in another set of 57 cases. Searching through the case files and communicating with the participants, we were able to gain access to data in fewer than one of five closed cases. Despite the significant time and effort we put into the task, the final outcomes of four of five class action cases were beyond our discovery. It is not that the data are non-existent - claims administrators or parties certainly have them - it is, rather, that they are secreted away. The outcomes of publicly approved settlements lie locked in private files."

Pace and Rubenstein were nice for empirics; Jeffrey O'Connell and Patricia Born are nice for thought-provoking. Their article, recently posted to SSRN here, is titled, "The Similar Cost and Other Advantages of an Early Offers Reform for Products Liability Claims for Personal Injury Compared to General Liability Claims Therefor." (Actually, it's nice for thought-provoking, but crappy for "title." We proofread that title a few times to be sure that we'd typed it correctly there.) Anyway, these guys think this system would be a good idea:

"[A] defendant facing a personal injury claim is given the option within 180 days after a claim is filed of offering to guarantee periodic payments for a claimant’s medical expenses and wage loss beyond any other applicable coverage, plus 10 percent for attorneys’ fees. There would be no compensation for pain and suffering. The claimant in return agrees to foreclose further pursuit of a normal tort claim for both economic and non-economic losses.

"Offers could be turned down by claimants, but only in cases where the defendant’s injurious acts were the result of gross misconduct provable beyond a reasonable doubt."

The authors think that such an "early offers plan would reduce the time it takes to pay losses by at least two years, and also greatly reduce the costs of such claims."

Finally, for folks litigating in New York state courts, Patrick Connors has posted at SSRN "Which Party Pays the Costs of Document Disclosure?" Connors wants to maintain flexibility in deciding who will pay the cost incurred in producing electronic discovery. In the words of the second paragraph of the abstract:

"In Lipco Elec. Corp. v. ASG Consulting Corp, the New York Supreme Court concluded that 'the party seeking discovery should incur the costs incurred in the production of discovery material.' However, this rule limits the inherent flexibility of Article 31, and is neither supported by the text of the CPLR, nor by the case law cited in the opinion. This article respectfully submits that the disclosure process will function more efficiently and fairly without a general rule requiring the party seeking 'documents or any things' to bear the costs of production. Parties should be encouraged to discuss disclosure costs as early as possible, and request a protective order from the court if necessary."

That's this week's trek into academia. Now it's back to the real world for a while.