First, a minor correction. We lost by 6-3, not 6-2 (when you’re typing fast you’re all thumbs). We lost Kennedy, Thomas, and Breyer. We needed to win at least two of them.
The court relied on two facts established by the trial: (1) that a stronger warning would have made a factual difference (eliminating consideration of possible medical malpractice), and (2) that this was only a failure to warn and not a duty to contraindicate case. Slip op. at 7-8. Thus it avoided having to deal with a lot of the facts we thought in our prior posts (click on the “Levine” label) were pretty good for our side.
It’s now clear that (1) congressional intent is the “touchstone” of implied preemption cases as well as express preemption cases, and (2) the presumption against preemption applies to implied preemption cases. Slip op. at 8. The latter ruling was presaged earlier in the term by Philip Morris v. Good.
Looking at congressional intent, the majority found no intent to preempt state-law duty to warn cases. It found the opposite, viewing the 1962 Amendments as adding a “saving clause” and noting that there was no express preemption provision similar to that enacted for the Medical Device Amendments in 1976. Slip op. at 10. The 2007 “rule of construction” “ma[d]e it clear that manufacturers remain responsible for updating their labels.” Slip op. at 11.
The CBE regulation, allowing strengthening of certain types of warnings, defeats preemption by reason of impossibility. Slip op. at 11-12. Even under the 2008 amendment a unilateral label change is allowed for “new analyses of previously submitted data.” The defendant could have re-analyzed its accumulated adverse drug reaction data under even the 2008 version of the CBE regulation. Slip op. at 13. This is an important ruling because it implies (but does not hold) that the result won’t be any more favorable to preemption in post-2008 injury cases.
“The FDA’s belief that a drug is misbranded is not conclusive” in civil trials because juries have to resolve misbranding prosecutions. Slip op. at 13. Thus the majority concludes that a threat of FDA enforcement action does not result in preemption by reason of impossibility. The majority views such actions as unlikely in any event. Slip op. at 13-14.
The majority rejects the argument that labeling is primarily the FDA’s responsibility, rather than the manufacturer:

[I]t has remained a central premise of federal drug regulation that the manufacturer bears responsibility for the content of its label at all times. It is charged both with crafting an adequate label and with ensuring that its warnings remain adequate as long as the drug is on the market.

Slip op. at 14.
If the FDA actually rejected a CBE label change (which did not happen in Levine) there might be preemption. Slip op. at 15 (“absent clear evidence that the FDA would not have approved a change”). The majority found no evidence of that in this case:

And while [defendant] does suggest that the FDA intended to prohibit it from strengthening the warning about IV-push administration because the agency deemed such a warning inappropriate in reviewing Phenergan’s drug applications, both the trial court and the Vermont Supreme Court rejected this account as a matter of fact. In its decision on [defendant’s] motion for judgment as a matter of law, the trial court found “no evidence in this record that either the FDA or the manufacturer gave more than passing attention to the issue of” IV-push versus IV drip administration.”

Slip op. at 15. The implication here is that there’s room for a different result if the FDA had made “an affirmative decision” and had gave a particular issue “more than passing attention.”
That’s (or more accurately, is not) impossibility preemption.
But the majority rejects implied conflict/obstacle preemption as well, holding that “relies on an untenable interpretation of congressional intent and an overbroad view of an agency’s power to pre-empt state law.” Slip op. at 17. Ouch.
There follows a discussion of congressional intent. The majority’s major point is that, “[i]f Congress thought state-law suits posed an obstacle to its objectives, it surely would have enacted an express preemption provision at some point.” Slip op. at 18. Thus the majority allows implied preemption to be driven by express preemption (or, more properly, the lack of it) more than it has in previous cases.
The majority then clobbers the FDA’s 2006 preemption preamble: (1) it’s not tied to any preemption-related regulation, slip op. at 19-20; (2) the Court doesn’t defer to agency conclusions specifically about preemption, slip op. at 20; (3) the FDA didn’t give proper notice and comment with respect to the preamble, slip op. at 21; (4) the preamble is “at odds” with congressional intent as interpreted by the majority, slip op. at 21; (5) the preamble “reverses” the FDA’s own position on preemption, slip op. at 21-23. This is a significant rebuke to the positions taken by the Bush administration FDA.
Finally the majority distinguishes Geier v. American Honda Motor Co., 529 U.S. 861 (2000), as involving “formal rulemaking and then. . .a plan to phase in a mix of [options].” Slip op. at 24. In Levine, by contrast, there was no FDA “regulation bearing the force of law.” Slip op. at 24. Unlike Geier, the court read the relevant regulatory history as a “longstanding coexistence of state and federal law” at “each” time the FDA took regulatory action concerning the drug Phenergan. Slip op. at 24. Significantly the majority refrains, in its statements about Geier, from making formal rulemaking an absolute prerequisite to preemption as the plaintiffs had argued.
So what’s left? This: “we recognize that some state-law claims might well frustrate the achievement of congressional objectives, this is not the case.” Slip op. at 25.
So what might those claims be? Two things jump out. As discussed, there’s the situation actually involving “formal rulemaking.” The second would be a situation where, unlike the way the majority read the FDA regulatory history in Levine, the FDA did, in fact, consider a particular risk in considerable depth and made an affirmative decision to reject a warning of the sort that a particular plaintiff demands. SSRIs come to mind, and some other situations as well. But defendants are going to have to litigate more selectively – and take great pains to develop a solid regulatory history.
We’ve been on the wrong side of Supreme Court preemption decisions before. In the middle of Bone Screw Lohr came down and knocked us for a loop. Levine isn’t good news for our side by any means, but it’s no Lohr. The Court did not hold that there’s no preemption at all, or no preemption without an express regulation, either of which it could have done. It didn’t impair either Riegel (which, being an express preemption case, is completely different) or Buckman. In short, Levine doesn’t make preemption impossible, just a lot more difficult. After Levine it’s pretty clear though that implied preemption is going to depend quite a bit on what the FDA has to say about specific drug risks, and not very much on what the FDA has to say about preemption.
That’s it for now. We’ll look through the concurrences and dissent later.