Tuesday, February 02, 2010

On Judicial Estoppel

Once upon a time, quite a few years ago, one of us had the pleasure of a having a case before the (late?) Hon. John Dowling of the Dauphin County (Harrisburg, Pa.) Court of Common Pleas. We still remember winning the case on summary judgment – after finishing our argument, the first words out of the judge’s mouth to the other side were, “How are you going to get out of that?” They didn’t.


Anyway, Judge Dowling had quite a literary bent, which he put to work in his opinions. You could always tell when you were reading a Dowling opinion. Once he had a case involving a party with a habit of taking contradictory positions. Ultimately he was forced to invoke the doctrine of judicial estoppel to put an end to that. His estoppel discussion began:

Not since Joan de Pucelle in Shakespeare’s Henry VI, Part I attempted to defend herself from a capital charge by proclaiming herself a virgin and then, seeing that that particular defense was unlikely to prevail, informed the judge that she was with child, has anyone argued a judicial point with a more breathtaking lack of concern for consistency.
Ligon v. Middletown Area School District, 110 Dauph. Co. R. 9, 11 (Pa. C.P. Dauphin Co. 1989) (quoted in Ligon v. Middletown Area School District, 584 A.2d 376, 379 (Pa. Commw. 1990).

Not exactly one of Shakespeare’s better known works – but that was Judge Dowling – a renaissance man.

We got thinking about Judge Dowling and Joan de Pucelle because of some antics by the other side in litigation that shall remain nameless. The other side first argued, after Wyeth v. Levine, 129 S. Ct. 1187 (2009), of course, that no, there was no preemption because the drug manufacturer was free to ignore and supplement the FDA approved labeling language due to (you guessed it) the FDA’s “Changes Being Effected” (CBE) regulation. In short, the plaintiff argued that there’s nothing mandatory, and therefore preemptive, about FDA-approved drug labeling.

Unfortunately, that’s a position that – after Levine – a lot of courts buy. We don’t think that’s how the FDA really works, but as we’ve said before, strange things happen when preemption meets product liability.

Be that as it may.

After the plaintiff took that position, and prevailed in the trial court, the case moves closer to trial. We move to exclude labeling changes and other FDA regulatory activity that took place after the plaintiff last took the drug. They’re subsequent remedial measures, we argue. The plaintiff’s response was no they’re not subsequent remedial measures at all because only voluntary actions qualify – and the FDA’s later actions are (you guessed it, again) mandatory.

Mandatory?

But didn’t the same plaintiff just win a preemption motion with that argument that FDA mandated labeling wasn't mandatory?

Our own words temporarily failing us, we thought of Judge Dowling and Joan de Pucelle. Then we thought of judicial estoppel, which is what Ligon was all about. The basic rule is quite simple – no party can take a position in open court, prevail on that position, and then assert the opposite position. There’s no judicial estoppel without winning the first go round (a party can lose and change its position), but after prevailing, well, a party that made its bed with respect to its winning legal position has to sleep in it. Thus as a legal matter a party can claim to be either a virgin or pregnant, but not both. There’s no exception for immaculate conception.

Don’t get us wrong, though, we don’t think much of the underlying argument either. We just might not have been moved to post about it without the blatant flip flop. But since we have….

We don’t claim to know about all states, but recalls, and subsequent drug label changes have been considered subsequent remedial measures in Pennsylvania – with no distinction between “voluntary” and “mandatory.”  See Incollingo v. Ewing, 282 A.2d 206, 222-23 (Pa. 1971) (subsequent drug label change a subsequent remedial measures; admitted under exception to rule); Matsko v. Harley Davidson Motor Co., 473 A.2d 155, 158 (Pa. Super. 1984) (“statutory” motorcycle recall “mandated by federal statute” is a subsequent remedial measure; subsequent remedial measures held admissible in strict liability), overruled on other grounds, Duchess v. Langston Corp., 769 A.2d 1131, (Pa. 2001) (subsequent remedial measures equally excludable in negligence and strict liability); Hardin v. Upjohn Co., 47 Pa. D. & C.3d 66, 70 (Pa. C.P. Philadelphia Co. 1986) (FDA recall an excludable subsequent remedial measures). If Pennsylvania is any indicator, there are probably bunches of cases that consider government-ordered conduct to be subsequent remedial measures. If our plaintiff keeps this up, that’s probably going to force us to look for those cases at some point.

But since we’re about drugs and devices, we turn to Bexis’ book, where we find loads of precedent holding that subsequent changes in drug labeling, and other FDA actions (such as recalls) are considered subsequent remedial measures: Chlopek v. Federal Insurance Co., 499 F.3d 692, 700 (7th Cir. 2007) (label change); Stahl v. Novartis Pharmaceuticals Corp., 283 F.3d 254, 270 n.10 (5th Cir. 2002) (label change); Kehm v. Procter & Gamble Manufacturing Co., 724 F.2d 613, 622 & n.7 (8th Cir. 1983) (recall; allowed in strict liability under pre-1997 version of Fed. R. Civ. P. 407); DeLuryea v. Winthrop Laboratories, 697 F.2d 222, 227-29 (8th Cir. 1983) (label change excludable under pre-1997 version of Fed. R. Civ. P. 407); Lindsay v. Ortho Pharmaceutical Corp., 637 F.2d 87, 94 (2d Cir. 1980) (label change excludable under pre-1997 version of Fed. R. Civ. P. 407); Werner v. Upjohn Co., 628 F.2d 848, 853 (4th Cir. 1980) (label change excludable under pre-1997 version of Fed. R. Civ. P. 407); Smyth v. Upjohn Co., 529 F.2d 803, 804-05 (2d Cir. 1975) (label change) (label change excludable under pre-1997 version of Fed. R. Civ. P. 407); Hughes v. Boston Scientific Corp., 2009 WL 3817586, at *12 (S.D. Miss. Nov. 12, 2009) (recall); Gerber v. Hoffmann-La Roche Inc., 392 F.Supp.2d 907, 919 (S.D. Tex. 2005) (label change); Appleby v. Glaxo Wellcome, Inc., 2005 WL 3440440, at *6 n.7 (D.N.J. Dec. 13, 2005) (label change); Swank v. Zimmer, Inc., 2004 WL 5254312, at *1 (D. Wyo. April 20, 2004) (Dear Doctor letter and recall); Figueroa v. Boston Scientific Corp., 2003 WL 21488012, at *4-5 (S.D.N.Y. June 27, 2003) (recall); In re Propulsid Products Liability Litigation, 2003 WL 1090235, at *2 (E.D. La. March 11, 2003) (recall); [why the big 1989-2002 snooze, we don't know]Kociemba v. G.D. Searle & Co., 683 F. Supp. 1579, 1581-82 (D. Minn. 1988) (recall); Martinkovic v. Bangash, 1987 WL 28400, at *1 (N.D. Ill. Dec. 18, 1987) (label change); Richardson-Merrell, Inc. “Bendectin” Products Liability Litigation, 624 F. Supp. 1212, 1239 (S.D. Ohio 1985) (recall), aff’d, 857 F.2d 290 (6th Cir. 1988); Wetherill v. University of Chicago, 565 F. Supp. 1553, 1557-58 (N.D. Ill. 1983) (FDA submission seeking label change); Wolf v. Procter & Gamble Co., 555 F. Supp. 613, 623 (D.N.J. 1982) (recall); Skill v. Martinez, 91 F.R.D. 498, 509 (D. N.J. 1981) (label change); Janssen Pharmaceutica, Inc. v. Armond, 866 So. 2d 1092, 1100-01 (Miss. 2004) (label change); Wagner v. Roche Laboratories, 671 N.E.2d 252, 258 (Ohio 1996) (label change); Cosgrove v. Merrell Dow Pharmaceuticals, Inc., 788 P.2d 1293, 1296-97 (Idaho 1989) (recall); Northern Trust Co. v. Upjohn Co., 572 N.E.2d 1030, 1038 (Ill. App. 1991) (label change); Witherspoon v. Ciba-Geigy Corp., 1986 WL 2138, at *2 (Tenn. App. Feb. 12, 1986) (label change); Ortho Pharmaceutical Corp. v. Chapman, 388 N.E.2d 541, 560-61 (Ind. App. 1979) (label change); Smith v. E.R. Squibb & Sons, Inc., 245 N.W.2d 52, 58 (Mich. App. 1976) (label change), aff’d, 273 N.W.2d 476 (Mich. 1979).

Whatever comes, we’re ready. And thank you Judge Dowling for the Shakespearean inspiration.

5 comments:

Anonymous said...

Well, to be fair, it might be possible to be pregnant and a virgin without divine intervention. . .

http://scienceblogs.com/effectmeasure/2010/02/oral_conception.php#more

Anonymous said...

Antics? I think you misunderstand Wyeth v. Levine.

The fact that your side argued that you can't change a label unless the FDA says it's okay doesn't mean the other side argued "Oh yes you can" (i.e., label changes are voluntary). Actually, the other side argued, and the Wyeth Court agreed, that not only can manufacturers change their label without FDA approval, they must do so, at least in some circumstances. If it were just "voluntary," it wouldn't result in tort liability!

So for the other side to take the position that label changes are not "voluntary" is in no way contradictory to the position that FDA approval of a label ought not excuse a manufacturer from fulfilling its duty to provide adequate warnings.

And since you brought it up, just how does the FDA really work?

Beck, et al. said...

I think we understand exactly how Levine operates. It was not a negligence per se case. There was no claim that under federal law a label change was "mandatory." Levine involved only state law claims. The court used the CBE requirement to avoid the defense argument that compliance with FDA labeling approval was mandatory. Thus our analysis is correct, and the anonymous comment is wrong because it is predicated on a claim based on federal law that was neither before the Court in Levine nor proper under 21 U.S.C. 337(a). For the argument that compliance with FDA-approved labeling was mandatory (and therefore preemptive), see the dissent in Levine.

As far as how the FDA "really works" there's an excellent FDLJ article by Cooper (don't have cite handy) that demonstrates that, in practice, manufacturer doesn't dare try to change a label significantly without first telling the FDA and getting its blessing to proceed.

David said...

Would the Joan de Purcelle example fit into judicial estoppel since she did not prevail under her former argument (she was a virgin)? You mention that it only comes into play if you win your first argument and then seek to argue the opposite, but it's ok to argue the opposite if the first argument failed to persuade.

My follow-up question is whether estoppel is triggered if the arguments are made concurrently. Can you say, OK if we don't win under X originalism standard, then we should win for these reasons under Y evolving document standard (I realize those might not be precise opposites)?

Beck, et al. said...

Come to think of it, David is absolutely right. She didn't win with the virgin argument, therefore she shouldn't have been estopped (at least judicially). It's just such a great quote, we couldn't resist - but he makes a very valid point.