The $253 Million Vioxx Verdict: What Does It Mean?

September 7, 2005 

8:45 a.m.	Registration
9:00	Panelists:	Dan Troy, Sidley, Austin, Brown & Wood
		Evan Schaeffer, Schaeffer & Lamere, P.C.
		John E. Calfee, AEI
		Ted Frank, AEI
	Moderator:	Peter Wallison, AEI
11:00	Adjournment

Proceedings:

            PETER WALLISON: Good morning. I am Peter Wallison. I’m a resident fellow here at the American Enterprise Institute and I’ll be moderating the panel this morning on the $253 million Vioxx verdict, its meaning, its significance. We have a terrific panel with very interesting backgrounds, and all of whom are students of mass tort litigation and health issues for you to listen to this morning.

            I think we’ll go in alphabetical order. Seems fair. I’ll introduce each person as his chance to speak comes up. So we’ll start with Jack Calfee, who’s just to my right. Jack’s a resident scholar here at AEI. From 1980 to 1986 he served in the bureau of economics at the Federal Trade Commission. He’s taught marketing and consumer behavior at the University of Maryland in College Park and at Boston University, and he was a visiting senior fellow at – what was that? Brookings. Something like that. Brookings Institution, I think it’s called.

            His research is focused on regulation, especially the FDA regulation on health care, advertising and information, tort liability and related areas. He’s the author of several books on this subject. After Jack, we’ll proceed down the panel. So Jack, why don’t you start.

            Oh, let me add one thing, and that is, after each of the panelists has spoken, we’ll then have a discussion among the members of the panel and then we’ll proceed to questions from the audience. So make notes of the questions you want to ask so you’re ready when the question period begins. Thanks very much. Jack?

            JOHN CALFEE: Thank you, Peter. I’ll go through several points, starting with what is the main issue in most of this litigation, which is the cardiovascular effects of Vioxx, where cardiovascular effects usually refers to heart attacks and strokes, usually heart attacks. I think when it comes to information about Vioxx and cardiovascular risk, we’re really dealing with two eras or time periods: before February 2005 and February 2005 and later.

Before February Vioxx often, perhaps usually, appeared to be unusually dangerous compared to the drugs with which it mainly competed. I think the research that supported that conclusion was actually a good deal weaker than it seemed to be to a lot of people at the time. It mainly relied upon two large-scale clinical trials, VIGOR and APPROVe. VIGOR was run against naproxen, which is an older over-the-counter NSAID (non-steroidal anti-inflammatory drug) usually sold as Aleve. And essentially what was discovered after the trial was that if you more or less cherry-picked the data that you discovered that you get more cardiovascular events for Vioxx. Mortality was not higher but the heart attacks and stroke rates were higher, although that was only true for patients for whom aspirin was indicated but was not actually taken. And then later on there was the approved trial, the one that caused Vioxx to be pulled from the market, which was run against a placebo, in which a significant elevation of risk emerged after 18 months of use.

            But of course that left open the question of whether or not other NSAID’s would have given similar results if they had also been exposed in long-term trials against a placebo.

            Then in February of 2005 the FDA convened a meeting of two advisory committees and made, as they always do, their own staff presentations. And the outcome of that episode essentially was a finding that, first of all, Vioxx may impose extra risk compared to placebos, but this is probably true of the entire class of Cox-2 inhibitors, which included at the time Celebrex and Bextra, plus one or two other drugs in the rest of the world. But also, although the Cox-2 inhibitors appeared to involve an elevated risk, probably that’s true of the NSAID’s generally except for aspirin. In other words, there was not good evidence allowing one to say that Vioxx was more dangerous than Cox-2’s, or Cox-2’s were more dangerous than NSAID’s generally, with the possible exception of naproxen, or Aleve, which happened to be the drug that Vioxx had been tested against in the bigger trial, the results of which were published in November of 2000.

            So what came out of that meeting was what had looked at first like a Vioxx safety episode was shaping up as an NSAID safety episode. And, in fact, the committee voted overwhelmingly, 31-1, to keep Celebrex, the leading Cox-2 at the time, on the market, and by a close vote they were in favor of returning Vioxx to the market.

            In April of 2005 the FDA took more action. It released a staff memo by Jenkins and Seligman which reinforced what had emerged in the February of 2005 meeting through greater analysis of the data. The FDA also announced that they were seeking vigorous cardiovascular risk warnings for all NSAID’s, not just the Cox-2’s and certainly not just Vioxx, if Vioxx were to return to the market.

            In July of 2005 an expert panel in Canada was convened. They had access to a new and comprehensive meta-analysis, which as far as I know has not yet been published. And that meta-analysis and the committee’s deliberations again reinforced what had come out of the February of 2005 and April 2005 FDA deliberations, and by a vote of 11-1 the expert committee in Canada voted in favor of bringing Vioxx back to the market, emphasizing that the Cox-2 inhibitors were useful drugs, acknowledging that they did pose risks but noting that those risks were comparable to those posed by the older NSAID’s with which those drugs normally compete.

            This brings me to the two essential issues in the litigation – causation and failure to warn. Let’s look first at causation. I think that it’s fair to say that the plaintiffs would have a difficult time mounting a purely scientific case for what lawyers refer to as general causation—that is, the idea that in general Vioxx tends to cause undue risk to patients. To the extent that they could demonstrate that, their demonstration would apply to NSAID’s generally rather than just to Vioxx by itself.

            I also think it’s fair to say that the plaintiffs have no purely scientific way to demonstrate what lawyers call specific causation – that is, the question of whether or not Vioxx caused a particular heart attack or stroke that is the subject of litigation. As far as I know, there is almost never a diagnostic test that can tell you whether a heart attack or stroke was specifically caused by Vioxx as opposed to the many other causes that can come into play.

            So we have a situation in which you would normally have dueling experts on both sides. You will have evidentiary rules. Often in state courts a preponderance of the evidence rule in which, as a lawyer emphasized in the Ernst case, a 49 percent doubt is not enough. You have to go with a 51 percent to find liability, which means the plaintiffs will often win these cases. They may not win anywhere near half of them but they will often win cases in this situation, at least on the causation issue.

            Which brings us to failure to warn, a point that also raises some fairly serious questions. One of them is, what should Merck have warned patients about in connection with Vioxx? Should the warning have been about Vioxx versus the NSAID’s with which it competed, in which case there may not have been a whole lot to warn patients about? Should it have been about NSAID’s generally rather than just Vioxx by itself? Should the warnings have cherry-picked the data, such as in the VIGOR trial, and said that one large trial has shown that Vioxx appears to pose higher risk for some kinds of patients but not for other kinds of patients, et cetera?

            And then there’s the question, if you have some idea what the substance of a warning would be, how would that warning be presented to patients? As anyone who’s ever delved into this issue knows, drug warnings are messy and complicated things. There’s a lot of things to warn about in drugs as powerful as NSAID’s, including the gastrointestinal or ulcer complications that all NSAID’s involve, and which in some cases can be very, very serious.

            Another question is, once you know you’re going to have this substance, once you have some idea about how that information should be provided, then what justification might there be for holding Merck liable for doing it in some way other that what the FDA has required? And of course Merck was always in compliance with FDA requirements.

            Then beyond the substance of the warning and the presentation, there’s the question of what the effects of a warning would have been. Would it have been that patients would simply stop using NSAID’s altogether? Would they have switched to other NSAID’s besides Vioxx, which seems like a far more likely result, different in new warnings? If so, would that have made patients safer or would have left them more or less in a situation they would have been if they’d been using Vioxx, because essentially the result would have been to induce patients to switch from Vioxx, about which a fair amount was known, to traditional NSAID’s, about which much less was known.

            When you think about those core issues in this litigation, you have to ask yourself how well the liability system can handle these kinds of things. In doing that I would consider the following questions. First of all, we know that on the essential issues of causation and failure to warn, I think the plaintiffs’ cases are inherently weak in the sense of having a thin and quite ambiguous scientific base upon which to rely. But there is a situation of dueling experts from both sides, and a preponderance of the evidence rule. And then you add to that the typical plaintiff’s strategy, at least so far, of demonizing Merck, demonizing the pharmaceutical industry in general, and demonizing the FDA because, after all, this is a situation in which Merck was doing essentially what the FDA had required it to do.

            I think that if you combine these uncertainties, the dueling experts, preponderance of evidence rules with the strategy of demonization, you end up with a situation in which the plaintiffs may well win a minority of cases, maybe a small minority of cases, but when they do win they’re going to get very, very large awards. In Texas it was a quarter of a billion. The punitive damages will be reduced under Texas law. There are some states in which such a verdict would not be reduced. According to Texas law they might be reduced on other grounds.

            The cumulative burden of such victories, even if it’s a small minority of cases, could be very, very large. It could easily be tens of billions of dollars. It could even threaten Merck as an institution. And all of this could happen even if the scientific truth is that Vioxx is no more dangerous than the typical Cox-2 inhibitor, no more dangerous in terms of cardiovascular risk than the typical NSAID. And in fact it’s less dangerous than the typical NSAID in terms of gastrointestinal risk – that is, ulcers and bleeding.

            This suggests at least to me that there are some dangers to watch out for as we look forward in Vioxx litigation. After all, the fact pattern that characterizes Vioxx litigation may turn out to be fairly common. That is, you have a popular drug class. It treats a chronic condition, in this case chronic pain from arthritis and other conditions. And therefore, like other popular drugs, it is in long-term use. There is an unknown risk. That is, there’s much that we don’t know about the drugs used for chronic conditions. Probably, as we are learning in the case of Vioxx, probably these unknown risks, when they are revealed do not tip the balance. The drugs still remain useful as they have been in the past.

            But given all these fact situations, if you add the fact that the FDA approval, FDA regulation provides no defense to the manufacturers, then through this sort of random process that we have in litigation with dueling experts, et cetera, you can get occasional awards that are very, very large. And since this could happen not just with Vioxx but with other drugs or even other classes of drugs, that raises two questions which are quite alarming.

            One of them is, if you’re the manufacturer of drugs that fit this pattern – that is, that your drug is a member of a larger class of drugs which is very popular and used for long-term use—does it really pay off to learn more about your drug? Are you going to make yourself better off by conducting further research on that drug, even if the research you’re doing (as with Vioxx) is exploring new uses for the drugs that may be extremely valuable, and that may in fact have been available for older drugs in the class but had never been researched until you came along. That was the case with Cox-2’s in cancer prevention.

            It also raises another question, which is, if you’re in this market, if you’re active in such a drug class, why should you try to develop a better drug for that particular drug class? A drug that may be safer, that may have some applications that previously had not been revealed, if the results can be that in the process of doing more thorough research you’re going to reveal risks about your drug that in fact may be typical of the entire class but have not been revealed before. The company may, in such a case, suffer much more than it benefits from its research. I think those are the kinds of things that could play out in the Vioxx litigation. I think they’re things to worry about.

            Peter, those are my remarks.

            MR. WALLISON: Thank you, Jack. Jack, if Vioxx is held responsible, as it was in this case, for heart problems, and the NSAID’s are all deemed to be of the same character, are ibuprofen and similar kinds of over-the-counter drugs also possibly in jeopardy? Could not someone construct a case that taking ibuprofen resulted in a heart attack?

            MR. CALFEE: I can imagine it in theory. I think it would be more difficult to pursue that kind of litigation because most of those drugs have not been actively researched recently (because they are all drugs that have been off patent for 10 or 20 years). What drove this litigation is the fact that Merck was actively researching these drugs. It was conducting trials far larger than had ever been done with NSAID’s before, except for the other Cox-2 inhibitors. And when they do that, they were coming up with results that you have to look at very carefully, possibly new safety results. So you’re going to get this kind of e-mail exchange that has been revealed from Merck, in which people are saying, we may have this problem, we may have that problem, we may want to think about this and that, et cetera.

            And the result is, if you have an active research program, you’re going to generate the kind of exchanges that can be used to alarm juries about this particular drug and about this particular company. I doubt, although I don’t know for sure, that there’s that kind of record for these older drugs which, after all, have not been researched very much. Or if they’ve been researched in the last 5 or 10 years, that research has been done by public agencies rather than by manufacturers.

            MR. WALLISON: Okay, well, we’ll turn now to the litigation itself, and Ted Frank, who is a resident fellow at AEI and director of the AEI liability project. He manages the institute’s research about liability reform proposals, tort law, class actions, and civil procedure. Before joining AEI Ted was at a law firms in Los Angeles and Washington, D.C., and clerked for Judge Frank Easterbrook. His litigation work included Vioxx and automobile liability cases, class action defense, and antitrust and patent cases. Since 2003, he has been a regular contributor to the tort reform weblog Overlawyered.com.

            Ted, under-lawyer us.

            TED FRANK: Thank you. We’ll start off with a disclaimer. I did use to work for a law firm where I did work for Merck in several Vioxx cases, but not the Ernst case, but everything I say here today is my opinion and not that of Merck, my former clients, or any of my former employers.

            The Ernst case presents several problems, some of which are complementary problems that magnify each other. And I could talk for hours about it but fortunately for you I only have 15 minutes. I invite you to read the much longer piece that’s been distributed with the binders.

            The first issue I’d like to discuss today is the question of the damages award, the $253 million. I think this case illustrates the problems that can arise when juries are asked to assign monetary values to immeasurable intangibles like non-economic damages, like punitive damages. This jury came up with a figure of $450,000 in economic damages – that’s lost wages – $24 million in non-economic damages, lost companionship. (In other circumstances pain and suffering would be included, but this fellow died in his sleep so that wasn’t an issue.) And $229 million in punitive damages.

            Texas has had some liability reform. They have caps on the punitive damages that are tied to a formula relating them to the amount of economic damages if no crime has been committed. The punitive damages will thus be capped at $1.6 million, which got recorded in the 12^th paragraph of some stories, but the resulting total of $26 million is still too high. I think it creates some distorting effects in litigation.

            Let’s step back for a minute. Why do we have damages? On the one hand it is to compensate the injured who suffered from alleged torts. On the other it is to deter actors from wrongdoing. But when you have a figure like $26 million, I think it fails on both accounts.

            Let’s look on the plaintiff’s side. Rob Ernst was a 59-year-old produce manager at Wal-Mart. Even setting aside the question of whether the present value of his lost wages over the rest of his life was really $450,000 (and if it was, you’d think Wal-Mart would have fewer labor problems) the problem comes with $24 million in non-economic damages. It creates a sort of paradox where Mrs. Ernst is better off because her husband died from sudden arrhythmia rather than from a brain tumor or a tornado or any other form of sudden death. All tragic events, all creating loss. In one instance she gets her life insurance policy, in the other if she can persuade a jury, she can get $24 million. I think the odds are overwhelming that she did not have a $24 million life insurance policy to compensate her in the event of his death.

            That’s the issue here. Compensatory damages are supposed to compensate the victim, leaving the plaintiff in the same position as if the tort never happened. When you have something like loss of companionship, there’s no rational way to compute that. It becomes solely a matter of the whim of the jury. So if Mrs. Ernst is less grandmotherly looking, if she’s a different race than the jurors, if she has a hideously grating speaking voice, if she’s otherwise unsympathetic, the award becomes completely different. There are fundamental issues of fairness here. And that’s the theory behind a lot of legislation to cap non-economic damages, to constrain jurors in a sense.

            Had there been a $250,000 cap on non-economic damages in place with this award, the jury’s award still would have added up to $2.3 million, which again is surely more than whatever insurance policy the Ernst’s held, and can hardly be considered an injustice, at least on the plaintiff’s side. Certainly Mrs. Ernst may well prefer to have her husband of one year alive rather than having the money, and that’s I think the fundamental absurdity of compensatory damages that even large amounts don’t seem to compensate.

            But as a society we also have to look at the other side of the equation, the deterrence side. And when damages are too large, they over-deter. Now over-deterrence might not be an issue when you’re talking about Russell Crowe throwing a telephone at a hotel clerk. If that happens less often than it should economically, we don’t care that much. But when you’re talking about things like producing life-saving drugs, it makes a big difference.

            Even with high drug prices, the manufacturer of a life-saving drug cannot collect $2 million for each life their product saved. There’s no way to extract that, even if there were a way to determine whose life was saved by the fact that they took one Cox-2 inhibitor instead of a drug that had gastrointestinal side effects. And you can imagine a case where a drug saves 100 lives for every life that the drug’s side effects cost. We’re better off as a society if that drug is being sold and if those lives are being saved. We don’t want the manufacturer to be deterred from producing that drug, from selling that drug. But if the damages are too high, and I think this is Jack’s point, that is precisely what will happen.

            Even if juries act perfectly in deciding liability, if they’re allowed to range wildly in damages, this cost-benefit calculus can be upset because the defendant will be paying a multiple of the true costs that are imposed on society by whatever wrongdoing they’ve done. And we don’t have that best-of-all-possible-worlds situation where the juries act perfectly. You have to ascribe mystical wisdom to the jury system to believe that a randomly selected panel of a dozen lay people without scientific training, who are unanswerable to their peers, are going to act omnisciently in weighing costs and benefits. And the more freedom to wildly fluctuate in deciding award amounts, the more distortion gets introduced into the system when juries make mistakes.

            You can imagine a world where 99 juries correctly exonerate an innocent defendant, but the hundredth awards $253 million. In that case the defendant will be facing an average expense of $2.5 million, plus their attorney’s fees, per plaintiff. Now even if you believe (as I do) that juries get it right the majority of the time, even the vast majority of the time, the fact that an outlier jury making a mistake can essentially undo the work of the juries that are getting it right gives that jury a veto over the other juries and infects the system as a whole. This shows the irrelevance of some arguments out there that juries get it right most of the time, or defendants win most of their cases.

            There’s another distorting effect because the more likely the jury is to make a mistake, the greater the damages that a mistaken jury can award, the greater incentive there is for plaintiffs and attorneys to treat litigation like a lottery ticket and bring cases that are marginal, or even worse than marginal. If a plaintiff’s attorney can bring 1,000 cases and recover in the tens of millions for each victorious case, he or she doesn’t have to win very many cases to make a very healthy living. It might be possible to grow rich simply on the mistaken cases.

            We saw this problem in the breast implant litigation, where there was nothing wrong with the products but plaintiffs’ lawyers succeeded in driving the manufacturers into bankruptcy. The prospect of jackpot justice also creates incentives for outright fraudulent claims, and we’ve seen this in the asbestos litigation, we’ve seen it in the silicosis litigation, and we’ve seen it in the Fen-phen litigation, which got settled. Perhaps a majority of these mass tort claims involve phony diagnoses from doctors-for-hire, and attorneys that are committing outright fraud and extracting settlements from defendants who are overwhelmed with more cases than they can defend.

            This might even be happening in some cases in the Vioxx context. In Rogers v. Merck, which was originally scheduled to go to trial in Alabama before Ernst, Merck argues that Rogers didn’t even take Vioxx because the widow took a box of Vioxx to the deposition and said this is the Vioxx my husband took. And it was fascinating because the manufacturing label indicated that it left the factory six months after her husband died. And that case is still pending. The judge refused to grant summary judgment. She’s not getting prosecuted for perjury—she might collect millions of dollars.

            Now the absence of caps to moor a jury’s decision-making can compound the effect of errors on the system, distort the incentives the corporations face, and distort the types of cases that end up getting litigated. So even if juries were capable of deciding cases almost omnisciently, with 95 to 99 percent accuracy, the tort system still can be creating distortions that deter pharmaceuticals from producing life-saving drugs that we want them to be producing because individual mistakes by juries and courts can have an impact that far outweighs the correct decisions that the juries make. And we have good reason to believe that the Ernst jury did make a mistake.

            There is a fascinating Wall Street Journal article published the Monday after the verdict interviewing the jurors. Quote, “Jurors who voted against Merck said much of the science went right over the heads. `Whenever Merck was up there, it was like wah, wah, wah,’ said juror John Ostrom, imitating the sounds Charley Brown’s teacher makes in the television cartoon. `We didn’t know what the heck they were talking about.’” It goes on. “One juror, Lorraine Blas, had written in her questionnaire that she loves the Oprah Winfrey show and tapes it.” (This raises the side note, why are we letting lawyers ask jurors about their favorite television shows?) The jury consultant told Mr. Lanier, “`This jury believes they’re going to get on Oprah. They only get on Oprah if they vote for the plaintiff.’” So Mark Lanier, the plaintiff’s attorney, went on to suggest to the jurors that they could be famous. They might appear on Oprah if they voted for Ernst. And again from the Wall Street Journal, “As he made the Oprah reference, Mr. Lanier looked Ms. Blas in the eye. She said she broke out into laughter and liked the lawyer’s attention to her. `That told me he read these profiles and tried to assess each and every one of us,’ Ms. Blas said.”

            This is how decisions are getting made. Now, “wah, wah, wah” might not mean that the jury didn’t care about the science. They might just have felt that Lanier was more persuasive. I’ve seen other second-guessing of what happened at the trial. On the internet I’ve seen claims Merck should have had a more dynamic lead attorney. Merck shouldn’t have had a bald attorney. Merck’s attorney should have done more investigation of the coroner. Merck’s attorney shouldn’t have cross-examined the widow. Merck executives performed poorly on the stand. The argument goes, the jury heard the argument and decided for Ernst, therefore, if Merck did not deserve to win it was because Ernst’s attorneys did a better job and Merck’s attorneys failed to persuade the jury of the soundness of their arguments. I think this attitude toward the jury system comes straight of Voltaire’s Dr. Pangloss. Any result that has happened is the best of all possible worlds.

            I think this is a curious rationalization. And this leads me to my second point today. Is it really the case that we want billions of dollars of the economy, and that’s how much Merck’s share price declined over the weekend after the verdict, that those billions of dollars of our economy should be redistributed purely on the basis of oratorical guile? It turns litigation into a game show. And perhaps game show contests are appropriate for entertainment on Thursday nights at 9:00, 8:00 Central, but do we as a democratic society want to resolve issues that have such huge public policy implications this way? Why do we prefer to have these decisions made on the aesthetics rather than on the science?

            Thousands of years ago the great ancient Greek playwright Aristophanes satirized this problem in his play “The Clouds.” Power in Nephelococcygia, or Cloud-Cuckoo-Land, which was a thinly veiled version of Athens, resided in those who could persuade the citizenry in their votes through the most sophistical reasoning. The skills needed to proceed in the litigious Athenian social life were so valued that the best sophists became rich and the citizenry spent excessive amounts of time and money learning from them. It’s thousands of years old, but the only thing missing from this play is the LSAT.

            And I think that those who defend the Ernst decision by arguing that it was the consequence of the differing skill of the attorneys are doing more to indict the system as a whole than any critique of individual legal or procedural rules could.

            I would like to talk about one of those procedural rules today, though. And I think Mark Lanier won this case with the help of a loophole in federal law. This is a nationally important case, as you can see by the fact we’re all here discussing it today, and it affected citizens of each of the 50 states. This is a case that belongs in a national court, but it was tried in a state court in Brazoria County. There were some major differences between how this court acted and how the vast majority of federal courts would act. Because Ernst was in a state rather than a federal court, Lanier was able to obtain a victory from a non-unanimous jury. The jury voted 10-2 rather than unanimously.

            As another example, I think Lanier was allowed to mislead the jury by having his experts get on the stand and give evidence that was based purely on speculation. As the argument goes, Vioxx could have caused the death because it could have caused a thrombosis which could have caused a heart attack, which could have caused the arrhythmia that actually killed Ernst. I therefore find that it did cause the death because I say so.

            Well, you can’t do that in federal court, and if you believe the Texas Supreme Court, you can’t do it in Texas either, which is why I think that this decision will be reversed eventually. As the Texas supreme court has said, “Reasonable probability cannot be created by the mere utterance of magic words by somebody designated as an expert.” There are other significant differences, other significant failures of the trial court that I’m not going to go into today, but suffice it to say that if Ernst had been in a federal court rather than a local Brazoria County courthouse, the result might very well have been different. I think it’s worth exploring the tactical maneuver that allowed Lanier to pick his forum.

            Under the Constitution federal courts can have jurisdiction in what are called diversity cases. That means that if a plaintiff sues an out-of-state defendant, the defendant has the ability to seek the federal forum. The idea behind this diversity jurisdiction, expressed by Alexander Hamilton in Federalist number 80, is to create a neutral forum to resolve all those cases in which state tribunals cannot be supposed to be impartial and unbiased. Hamilton argued that in order to achieve the inviolable maintenance of that equality of privileges and immunities to which the citizens of the union will be entitled, the national judiciary ought to preside in all cases in which one state or its citizens are opposed to another state or its citizens.

            However, in 1806 an obscure Supreme Court decision interpreted the statute to be limited to those cases where each plaintiff was of a different state from each defendant, and that with the limited exception of the Class Action Fairness Act from earlier this year, Congress has never corrected this 200-year-old loophole that was created by the judiciary, and plaintiffs have learned to take advantage of it in forum-shopping.

            If Texas plaintiff Ernst sues New Jersey defendant Merck and nobody else, the case ends up in federal court. But if, as actually happened, Texas plaintiff Ernst sues New Jersey defendant Merck and a Texas defendant doctor, the case stays in state court. The partiality or bias that justified bringing the case to federal court in the first place hasn’t gone away because of the presence of the doctor. But Mark Lanier, by suing the doctor, succeeds in divesting the federal court of jurisdiction. And then a year later he dismisses the doctor from the case without penalty, the case stays in state court, and we’re still now Texas plaintiff, New Jersey defendant. The bias that Hamilton warned against is more than theoretical. Alex Tabarrok and Eric Helland have done a study that showed the average verdict against out of state defendants is substantially larger than those against in-state defendants, and the difference is even more pronounced in those states with partisan judicial elections.

Plaintiffs’ attorneys often sue in-state defendants whom they have no intention of bringing to trial in order to avoid the federal diversity jurisdiction and the more stringent federal rules for evidence and jury selection. This is called fraudulent joinder and it’s not allowed, but most courts permit the tactic without penalty except in most obviously egregious cases. Just a minor reform of the removal laws to prevent plaintiffs’ lawyers from benefiting by gaming the system would have tremendous benefits for the country as a whole by preventing this sort of forum-shopping. Thank you.

MR. WALLISON: Thanks, Ted. Let me ask you a question about fraudulent joinder because I don’t quite understand how we could adjust the removal laws in order to deal with this problem. It’s already prohibited.

EVAN SCHAEFFER: Can I ask you a question about that too, Peter?

MR. WALLISON: Of course.

MR. SCHAEFFER: You call it fraudulent joinder. Did Merck actually remove the case to federal court in the beginning of the litigation? Because either Merck removed it and then it went to federal court, where you want these decisions being reached, and a federal judge said no, there was no fraudulent joinder so it went back down to Texas state court; or Merck didn’t raise it all and it didn’t remove. I’m not sure what the answer is.

MR. FRANK: Well, the fraudulent joinder standard is so hard to prove, with some courts even requiring you to prove intent to fraudulently join parties, that it’s often futile to remove, though I know from personal experience that Merck has removed several cases where there has been fraudulent joinder and had the federal court decide there’s enough there under the standard for proving fraudulent joinder to keep the case in state court, and keep the case in state court. It’s very, very hard to prove fraudulent joinder. It’s only, again, the most egregious cases that you can divest and remove the case when there’s the plausible state defendant.

And you sue the doctor, it creates a plausible defendant, even though Lanier had no intention of keeping the doctor in the case because the jury might award the damages against the doctor instead of Merck, and then he doesn’t get the $26 million or the $253 million. And he doesn’t get the national publicity and he doesn’t get Oprah.

But to answer Peter’s question, I think we can just change the diversity jurisdictions. What the Supreme Court did in 1806 was interpretation of a statute, and it’s very simple for a majority of Congress, affirmed by the president, to say no, you interpreted the statute wrong, and from here on out you shall interpret the diversity statute to apply to all cases with an out-of-state defendant. I don’t think that’s that hard. We saw it done in the Class Action Fairness Act for class actions over certain jurisdictional amounts, and I think we can see it in products liability cases or other cases that we think are nationally important.

MR. WALLISON: Well, it would seem to me that there is a policy issue here in that an in-state defendant might feel, for example, that he would have a better chance to succeed in a state court than in a federal court. So I mean, we’re not just dealing with the plaintiff and a big out of state drug company. We could be dealing with a case in which the local defendant (the doctor, for example), may believe that he would be better off in a state court.

MR. FRANK: I think that’s a problem to the extent you believe that plaintiffs’ lawyers really want to be recovering from doctors here rather than just naming the doctors to keep the cases in state court.

MR. WALLISON: It sounds to me like a pretty tough case to make to Congress because we’re not just talking about drug companies and doctors but we’re talking about any kind of out of state defendant and an in-state defendant in any kind of tort suit. The in-state defendant may well claim that he wants to keep his jurisdiction in his state court, where he thinks he might have a better chance to succeed than in federal court in defending himself.

I just think this whole idea has to be thought through a little bit more because it’s a tough change in the removal laws to get something like that done by Congress.

MR. FRANK: Well, historically we haven’t had problems creating federal jurisdiction for such things as the securities laws, and saying that securities cases need to be tried in a federal court, even though there might be defendants and plaintiffs of the same state. So I think it comes down to, do we think these are cases that are local cases or national cases?

MR. WALLISON: Thanks, Ted, very much.

Our next speaker is Evan Schaeffer. He’s on the other side of the issue from Ted. Evan is principal of Schaeffer and Lamere. He’s a plaintiff’s lawyer, licensed in Missouri and Illinois, who practices primarily in the area of mass torts and class actions. His firm, along with two other St. Louis firms, has filed against Merck on approximately 1,000 users Vioxx users. He also publishes two popular legal weblogs. One is called Evan Schaeffer’s Legal Underground and the other is The Illinois Trial Practice Weblog. Evan?

MR. SCHAEFFER: Do any of you actually read weblogs? They may not be that popular.

Now that Ted has joined the AEI full-time, I’m the only trial lawyer on the panel, which means in general that I’m more fun-loving. That being the case, I had a joke to get started with today but I decided on the way over from the hotel that it wasn’t very funny, so I’m going to dispense with the hilarity and get right to my presentation.

I want to say first, thank you very much for inviting me, Ted, and thanks to the AEI for having me. I think this is a great event. It’s very interesting to hear you all speak, although I don’t agree with a lot of what you say. I’m glad that you invited me.

Also before I get started I should say that although much has been made about my Madison County roots by the organizers of this event, very few of the Vioxx cases that I filed along with the two other firms that I’m working with have been filed in Madison County. So as a result I won’t be addressing Madison County very much in my talk today

What I hope to do is to correct some misimpressions about the science that was presented at the Ernst trial. The magazine The Economist had an article that’s on the newsstands right now with a line about the science. In this case The Economist got it wrong. It’s there on the slide, except the slide’s not showing up. I’ll keep going. The Economist said that Mark Lanier, the plaintiff’s lawyer, won the case despite the lack of any scientific evidence that Vioxx actually killed Mr. Ernst. It’s no wonder that some people were angry about the result of the verdict.

One example was my brother-in-law. He’s a very nice guy, although he’d feel right at home here in Washington under the current administration, and he demanded an explanation from me after reading about the Ernst verdict. The plaintiff won without presenting any evidence at all, he told me. The other panelists have echoed much the same claim here today. If what The Economist said were true, which it’s not, the first person who observers should be outraged at is the trial judge for letting the case get to a jury in the first place. The trial judge is the gatekeeper of the scientific evidence. He gets to decide what evidence is too speculative or too unscientific for the jury to hear, even before the jury hears it. The judge in Texas determined, even before the jury saw the evidence, that the plaintiff’s evidence was scientifically valid and wasn’t speculative. And that’s something that’s often overlooked about our jury system. It’s one of the things that makes it so resilient. The jury decides the facts but the judge decides the law.

A judge decides what evidence the jury gets to hear. There are many other safeguards built in along the way. The trial judge can throw out a case based on the defendant’s motion to dismiss, or based on a defendant’s motion for summary judgment, or after the trial is half-way through on a defendant’s motion for a directed verdict. The trial judge can even throw out the case after the trial is over, after the verdict has been reached and a judgment, notwithstanding the verdict.

So why aren’t the very vocal Ernst critics complaining much about Judge Hardin, the gatekeeper of the scientific evidence down in the Texas state court? Back in Madison County, Illinois, where I practice quite a bit, I’ve been hearing plenty about state court judges ever since President Bush came to town last year to make a speech about “tort reform.” When I do this with my fingers, making a quotation mark, it’s because when people say they want to “reform” the tort system, often what they want to do is tilt the playing field in favor of big business.

I have heard since President Bush came to town that the judges in Madison County are biased, that the judges are unfair, all of which is code, in my opinion, for another message – namely, that what we need in Madison County is more Republican judges. And that’s what the advocates of tort reform have been saying down in Texas too, only in Texas they got the Republican judges, including the one, Judge Ben Hardin, that presided over the Ernst trial. That makes it hard for the Ernst critics to blame Judge Hardin for the result.

What’s to be done when you can’t blame the trial judge, the gatekeeper of the scientific evidence? You blame the jury. You say that the jury was too stupid to understand the science. That’s the claim that’s been made a lot about this case and a claim that I don’t happen to agree with.

We’re back to the science. Do you have to take my word for it as to whether the science at the Ernst trial was any good? No, you don’t. That’s because the plaintiff, just like the defendant, presented its evidence through medical experts who know a whole lot more about the science than I do. Let me tell you about the plaintiff’s medical experts, and in particular the plaintiff’s expert cardiologist. His name was Dr. Isaac Wiener. He graduated from Harvard Medical School and now teaches at UCLA. He’s board certified not only in internal medicine and cardiology but also in cardiac electrophysiology, which is the study of irregular heartbeats, including arrhythmias, probably the key scientific issue that was at issue at the trial.

Mr. Ernst, you remember, suffered an arrhythmia. The issue at trial was what caused that arrhythmia. Dr. Wiener is also the co-director of the cardiac arrhythmia center at the UCLA medical center in Los Angeles. Dr. Wiener saw all the evidence and all the medical reports and read all the studies, and he testified about those things to the jury. Here’s the bottom line, in my opinion. No one can claim that Dr. Wiener is a junk scientist.

What about the specific evidence that the plaintiff presented? Vioxx has been linked to blood clots and medical problems caused by blood clots: heart problems, strokes, embolism. The theory presented by Mark Lanier was that it was a blood clot-induced heart attack that led to Ernst’s arrhythmia and sudden death. The blood clot that caused the heart attack, Lanier said in his opening statement, was caused by Vioxx.

Dr. Wiener’s medical opinion was a little different. As one of the medical experts, he doesn’t have to agree with everything the lawyers say. In fact, he doesn’t have to agree with any of it, and when questioned by Mark Lanier, Dr. Wiener didn’t agree that Ernst died of a “heart attack.” He said people who die suddenly of a cardiac event often don’t show evidence of a heart attack. He preferred to use the term “sudden cardiac death.” But whatever the terminology used, Dr. Wiener also testified based on all the medical records in the case, and after ruling out all the other possible causes, that Vioxx had been a significant contributing factor in Ernst’s sudden cardiac death, as did two other medical experts who testified at the trial, both of them doctors. I know this is a little technical, but anyone who claims the jury avoided the science, as many have done, has to first have some understanding of the science.

What about some of the other questions that were raised about the evidence after the trial was over? At the trial, as I said, the plaintiff’s experts testified that a Vioxx-induced blood clot led to Ernst’s death. Even so, it was impossible after the fact to see the clot with their own eye.

MR. WALLISON: Evan, your material is up on the screen now.

MR. SCHAEFFER: Thank you. I’m not ready to switch yet.

Couldn’t find the blood clot, but that’s a non-issue. It’s well established that when a blood clot leads to sudden cardiac death, the clot usually dissolves. Blood clots are not something you can go hunting for afterward, like an Easter egg. Usually they vanish and leave no trace, and that’s going to be true not only in Ernst’s case but in many other upcoming trials.

What about Merck’s evidence of causation? Merck argued that one of Ernst’s arteries was narrowed due to plaque, something called arteriosclerosis, and that this caused his arrhythmia and sudden death. According to Merck, Ernst’ use of Vioxx had nothing to do with his death. Both sides agreed about the arteriosclerosis, that one of Ernst’s heart arteries was narrowed due to plaque, yet there was evidence in the trial that this narrowing by itself wouldn’t cause an arrhythmia and sudden death. Merck’s experts in my opinion didn’t really get around this problem, but presented their theory anyway that Ernst’ arteriosclerosis might have caused his sudden death. The biggest problem with the theory was that the narrowing in Ernst’s artery wasn’t severe enough to cause an arrhythmia and sudden death without something more, and that something more according to the plaintiff’s evidence was the Vioxx-induced blood clot, the only thing that could explain why the flow of blood was cut off to Ernst’s heart, leading to the arrhythmia and sudden death.

Next, what about the question of speculation that some have raised since the trial ended? It’s not speculation when you have well-credentialed medical experts testifying about causation based on their review of all the medical records, which happened in this case. Here again you don’t have to take my word for it because of those safeguards that I said were built into the jury system. If Merck thinks the evidence was too speculative or too unscientific, it has a chance to raise this again in front of the trial judge after the trial was over, as well as on appeal, not once but twice, being that in Texas there are two levels of appeal. That’s another reason why I can say that the Ernst trial is evidence that the jury system is working.

One final point about the science at the Ernst trial. Some are asking, hey, I thought that Vioxx was okay as long as you didn’t take it 18 months, but Ernst took it for only eight. The 18-month number comes from the VIGOR and APPROVe studies. Neither – and this is important – VIGOR nor APPROVe was designed to assess the cardiovascular risk of Vioxx. The VIGOR study specifically excluded from the study group most people thought to be at risk for cardiovascular events. Had people at risk for cardiovascular events been included in the study group, the results might have been different.

The VIGOR and APPROVe studies were not designed to assess cardiovascular risk. That’s why the VIGOR and APPROVe studies are of limited use in saying Vioxx definitely didn’t cause a bad event. What they do show is an association between the use of Vioxx and development of blood clots in general, even though not designed specifically to assess the amount of risk.

Causation in specific cases like the Ernst trial will have to be decided on a case-by-case basis by having qualified doctors rule out other alternative causes, which is exactly what went on in Ernst, and, I should also point out, is also exactly the way that real doctors treat real patients and make real-life life-or-death decisions, by ruling out other alternative causes and reaching a clinical diagnosis.

What about the other studies that were cited by Merck supporters? Every study that’s out there is of limited utility on the question of cardiovascular risk because none were designed to test that risk. The tests instead were designed to test new markets and new uses for Vioxx, and that’s a problem for Merck because it suggests that Merck was ignoring the risk, despite evidence that its scientists knew about and appreciated the dangers of Vioxx before it came on the market.

Mr. Calfee talked about an active research program. My question is, if the research program is so active, why didn’t Merck study the one thing that was thought to be the biggest problem, which is cardiovascular risk?

It’s also recently come to light that Merck did design a study to assess cardiovascular risk. It was the VALOR study, a large-scale cardiovascular risk study that Merck had in place to start in 2002. But just days before Merck was to submit that study to the FDA for approval, Merck halted the study. The reason why Merck halted the study has never been fully explained.

A final important Merck study is the VICTOR study, which made an appearance in the Texas trial and will probably come up again in the upcoming trial in New Jersey. The VICTOR study, even though it also wasn’t designed to assess cardiovascular risk, showed an immediate difference in cardiovascular events between subjects on Vioxx as compared to those taking a placebo. Merck hasn’t yet published the results of the VICTOR study. I don’t know if they plan to, but there will probably be more information about the study results in upcoming months. The study wasn’t completed. It was halted when Vioxx was taken off the market, but nonetheless there exists a lot of data.

So that’s the science behind the Ernst verdict. Does the jury’s original $253 million verdict tell us anything about problems with the jury system? First of all, I think it’s a mistake to confuse the verdict with the final result. But no matter what happens with the Ernst verdict, and already as has been pointed out the $229 million punitive award has been reduced to just $.16 million, the verdict does demonstrate in my mind that the system works.

Just looking at the science alone, here’s what happened in the Ernst case. Merck’s explanation for Ernst’s death, presented through its experts, didn’t hold up in comparison to the plaintiff’s explanation in the collective opinion of 10 of the 12 jurors who heard all the evidence. The Wall Street Journal—and it was a fascinating article—said that the jurors only took an hour to decide about causation, but what that fails to take into account is that they listened to the evidence for five weeks. They had five weeks plus an hour to decide about causation.

It’s easy to second-guess a jury, but I’d still be defending the Ernst decision even if the result had gone the other way. You can ask any trial lawyer, plaintiff’s lawyer, defense lawyer—Ted said it himself just a minute ago—you can ask any judge, state or federal, as to whether they respect the ability of lay juries to sift through the evidence and reach a just result. They’ll almost universally say yes. I have some data about that on a slide. So if Merck wins the next trial and the next trial, I might have some questions about what evidence was presented. I might have some questions about the way that evidence was presented at the trial, exactly the same sorts of questions that people have been raising about the Ernst trial. But if juries do decide for Merck from here on out, I won’t be questioning the jury’s competence or the jury’s intelligence to reach a just result based on the evidence that was presented to it.

Now despite my views, some are willing to draw conclusions about the Ernst verdict, so let me address a few of those, then I’ll wrap it up. First the question of punitive damages. Even though the punitive damage award was reduced, aren’t punitive damages a bad idea in the first place. It’s important to keep in mind that punitive damages are not awarded very often. When they are awarded, they’re meant to punish, to act as a reminder that outrageously reckless conduct shouldn’t happen again. In the Ernst case the conduct being punished included failing to appropriately warn about a likely clotting risk of Vioxx, that such a risk was known to Merck, and marketing Vioxx in a misleading way to the very people who would most likely be affected by such a risk.

The specifics of these failures have been well documented by the New York Times, the Wall Street Journal, other newspapers, as well as a congressional study that was released on May 5^th, 2005. If you want further details about the specifics of Merck’s conduct, those are good places to start.

What about the effect of the Ernst verdict on the development of new drugs? Despite other pharmaceutical mass torts, drug companies continue to do very well financially. Drug companies continue to develop new drugs. Take Wyeth, for example, a company that has recently paid $21 billion in settlements for the diet drug litigation. It still has a healthy pipeline of new drugs nonetheless. But we can set examples like that aside.

Why would anyone even consider granting favors to the pharmaceutical industry by tort immunity when the fix is so simple? First, warn appropriately about the known risks, and second, market the drugs to the most appropriate users, those for whom the benefits of the drugs are most likely to exceed the risks. That’s all consumers are asking. Drug companies should fully disclose the risks about which they know and study the ones they suspect so that doctors and consumers can make informed choices.

No one wants to keep drug companies from developing new drugs. No one wants to keep drug companies from making money. No one expects prescription drugs to be free of all risk. All drug companies have to do is fully inform consumers of the known risks. In my opinion that’s something that Merck didn’t do with respect to Vioxx, and that’s unfortunate because had Merck been forthright about the clotting problems and done a clinical study to assess the risk, Vioxx might still be available to the people who would benefit from it. Who are those people? Not a group of people large enough to create a blockbuster drug. I mean instead a smaller group of people without heart problems who are in serious pain and need a painkiller that won’t give them the gastrointestinal problems other painkillers might. That was the promise of Vioxx but it wasn’t to be.

Now as a result, a very small minority of commentators are calling for tort immunity for drug companies and discussing reforms to the jury system that would benefit drug companies. I think these ideas are radical and unnecessary. The tort system and the jury system may not be perfect, but they’re certainly better than any of the alternatives, especially those involving granting favors to drug companies that might lead to even greater health risks.

People talk about jury system reform, people talk about tort reform. If anything needs to be reformed, it’s the business practices of a very few drug companies that are willing to put consumers’ lives at risk in the search for greater profits. Again, thank you very much for having me.

MR. WALLISON: Thank you very much, Evan. I guess I have one question, and that is that you have 1,000 plaintiffs. If they were successful in anything like the plaintiff in the Ernst case was successful, would you agree that a Merck, for example, is gone? Is out of business, is bankrupt? And do you have any sense of how that issue should be resolved? Should we allow drug companies to be destroyed by asbestos-like issues that involve questionable—that is, debatable—science, simply because of what might be characterized as a mistake? Is this a good public policy from your point of view?

MR. SCHAEFFER: I think you want me to address what a case might be worth, and that’s not something that –

MR. WALLISON: No, I’m just saying if you have 1,000 cases and you’re successful and each of them is worth $5 million, just to pick up a number.

MR. SCHAEFFER: My point is, it doesn’t make sense. There’s only been one data point in a graph that has yet to be developed. It doesn’t make sense to assign, to speculate like that and assign a value to a case and then multiply it times the number of cases and then say, aha, you’re going to put Merck out of business, because I think that that’s highly unlikely to happen. Merck can keep it from happening.

MR. WALLISON: I’m not asking you to speculate, but I’m really asking about the role of a plaintiff’s trial lawyer in a case like this. It isn’t your job of course, but we in Washington think a lot about the policy results of things like this, and we have examples like asbestos, where many companies have been put out business by mass tort litigation of this kind, with again, science that is not entirely clear but based on abuses, alleged abuses by the asbestos companies. Do you think about that as part of your role, or is it simply a question of getting as much as you can for your clients?

MR. SCHAEFFER: I think these are important questions to raise, but my role as plaintiff’s attorney is to seek compensation for people who are injured. I think it’s unlikely Merck is going to go out of business or declare bankruptcy and have to deal with the claims in bankruptcy. I don’t think that’s going to happen. But let’s say it happened. You put that public policy problem, as you stated, against the fact that Merck didn’t warn about a known risk, at least in my opinion, and thousands and thousands of people have been affected. If you examine it like that, I’m in favor of the right side of the scale: making sure the drug companies warn.

MR. WALLISON: Okay, thank you.

Our next speaker is Dan Troy. He’s a partner now at Sidley Austin Brown & Wood’s Life Sciences Practice as well as its Appellate Litigation practice. Before joining that firm he was the chief counsel for the Food and Drug Administration. In addition to providing strategic counseling on FDA-related matters, Dan practices administrative and constitutional law and litigation, with particular focus on the pharmaceutical, biotechnology, food, medical device, cosmetic and media industries. They are all related, of course. From 1987 to 1990 he was in the Justice Department’s office of legal counsel, and before that he clerked for D.C. Circuit judge Bob Bork. From 1997 to 2000, indeed, he was an adjunct scholar here at AEI. Dan?

DANIEL TROY: Thank you very much. It’s always good to be back at AEI. Some disclaimers – my firm does some limited work for Merck. We represent Pfizer in the Cox-2 litigations. Everything I’m going to say are not the views of my clients or my firm.

I’m not here to litigate the Vioxx cases. I’m not qualified to do so. I will note that there was a warning about cardiovascular risk. The question is the adequacy of the warning. But what I really want to do is take a step back and take a look at what the public policy implications are, not really just of the verdict but of the entire Vioxx situation, the Vioxx withdrawal and the attendant public brouhaha about it. And I’m going to be somewhat personal about this because I’m going to talk about what the situation at FDA was before 2004 as I observed it.

I found when I came to the agency a healthy appreciation for the fact that all drugs do have risks, a healthy appreciation for the fact that it was a risk-benefit calculus. FDA viewed itself as a risk management agency. I think that the agency very much had learned the lessons of the AIDS experience from the 80s, pressure from the cancer patients, and we have some friends here from CEI who did a fabulous job at highlighting the effects of deadly over-caution and of highlighting the fact that not approving therapies has risks as well, and not being able to treat things can kill as many or more people than all of the drug safety problems associated with drugs that FDA had put on the market.

Dr. McClellan was trying to lead a discussion about Bayesian statistics. I won’t go into it, but in the debate between the clinicians and the statisticians at FDA and at the Center for Drugs, there’s always this, you know, pressure or desire on the part of some to ask for more highly powered trials before the drug gets approved, more highly powered trials, longer trials, and what Bayesian statistics does is it takes account of what we do know as opposed to traditional statistics, which assumes that you don’t know anything. We were really trying to come up with the right risk-benefit calculus.

FDA understood that the real way to address drug safety problems is, to a certain extent, pre-market, that you’re not going to know everything about a drug pre-market. You’re always putting these drugs on the market with their studies in 1,000, 2,000, 5,000 patients. Once a drug is taken by many more people, you’re going to find out much more about it. So post-marketing surveillance needs to be the answer, if you will, to drug safety.

I also think that the role of the Prescription Drug User Fee Act, which is assailed by many as putting FDA in the pocket of drug companies, worked enormously well not so much because it gave the agency additional resources, although that was important, but also because PDUFA has associated with it deadlines by which FDA needs to get back to the company, and they keep the dynamic scientific negotiation that is the drug approval process moving forward.

So what happened in 2004? We won’t spend a lot of time on it, but I think there were at least three things that contributed to the current regulatory situation. One is flu vaccine, which caused people to attack FDA. Next is the whole brouhaha about SSRI’s, the anti-depressants. One of the things that’s really interesting about anti-depressants is if you just do a content analysis of the media stories about SSRI’s before FDA put the black box warning on it, it was all, “Oh, people are committing suicide from these drugs.” And then almost to the day that the black box warnings were put on, all the media stories are, “Depression is being under-treated.” In fact the AMA has now said that depression is being under-treated and has asked FDA to take a look at the warnings on SSRI’s, and that is a perfect example of the risks of over-warning, which is something that plaintiffs’ lawyers completely ignore. And then of course came Vioxx.

And so in short, what are the consequences, at least at FDA? The pendulum, which has always swung a little bit, has swung dramatically back in the direction of drug safety. It hasn’t shown up yet in the statistics but there is a slow-down in approvals. The reason why it hasn’t shown up in the statistics is because we’re really only talking about things since September, October, November. I can tell you experientially there is an increased request for pre-market studies once again. The New York Times did a front page story on the now-routine use of black box warnings. Kip Viscusi has done work showing that at a certain point if you keep warning people, they just turn off and there is a dramatic risk of over-warning in that respect. And if every drug has a black box warning, what does a black box mean? Routine risk, what are called risk management plans which limit the access to the drug and which has FDA doing what it has traditionally avoided doing, which is regulating the practice of medicine, deciding which doctors can prescribe, putting enormous amounts of paperwork, which they have no legal authority to do. I’ll talk a little bit about that in a moment. But it leads to diminished access.

I’ll give you just one example. Doctors won’t prescribe Acutane. If you want Acutane, which is for some people a very useful and appropriate and important drug, you have to go to a hospital-based practice. Private practitioners, private dermatologists will not deal with it. It’s just too burdensome. And you see that with a lot of these drugs that are under these risk-management plans.

FDA did a pause in the Tsabri marketing. Tsabri is a product that is two-thirds more effective than any other product for multiple sclerosis. There are 100,000 patients in this country who are not being treated for MS at all, and the pause in marketing was after one or two adverse events were found. And then an opioid called Palladone (a sort of follow-on to Oxycontin) was withdrawn from the market after it was found that if you take an opioid with alcohol, bad things can happen. Well, there’s a shocker. And so instead of saying don’t take alcohol, FDA pulled the drug from the market.

Then of course we had the Vioxx verdict. Mr. Schaeffer tried to avoid the question, but there was a Sanford-Bernstein analysis that came out before the most recent verdict, which basically estimated the net present value, potential for liability as $25 billion up to $36 billion, at net present value of $25. The reports I’ve seen in the Wall Street Journal range everywhere from a potential of $4 billion to $50 billion of liability to Merck. Apparently Merck tried what’s called the Baycol approach to these product liability cases, which is, you know, we’re going to try every one of them, but in Baycol Bayer has won every case that they have tried and in order to have the Baycol approach succeed, you need to win, otherwise the plaintiffs’ lawyers smell blood in the water.

As we’ve talked about, it was a very high verdict. It’s going to be reduced, but as Peter Wallison pointed out, if it’s times 4,000, 40,000, 140,000, whatever the multiple is, it’s a very, very difficult and dramatic situation. And what this reflects is that the new regulators really are plaintiffs’ lawyers, state juries, state attorneys general and US attorneys, none of whom really are, you know, primarily motivated by the science. FDA is not perfect—I’ve been a big critic of the agency before, when and after I was there —but it really does try to do things based on science. And even the New York Times and USA Today noted the non-science-based nature of the verdict.

You’re always going to have doctors who are well credentialed who are on the other side. They get well paid for this, and if the fact that a few doctors say that the science says this is outcome-determinative, then you’re never going to be able to have an accurate assessment of the science.

So what are the consequences of over-caution due to a product’s liability situation? They are dramatic. There are reams of studies that show that current product liability situation leads to much higher prices in America. Richard Manning did a study in which he compared a generic drug that’s been around for many years, a thyroid therapy, and found that a huge amount of the price differential between the US and Canada is accounted for by product liability. It is also one of the reasons why there’s been exit from the vaccine market and one of the problems with developing new vaccines. (I will note today that Merck, which Mr. Schaeffer seems to not care about if it goes away, did today come out with a new childhood vaccine that is measles, mumps, rubella, and chicken pox, and that is exactly the kind of drug or vaccine that is derided by critics of the drug companies as a so-called me-too drug. But if you’re a parent and you only have to go once instead of four times, or even twice, to get your child vaccinated, let me tell you as a parent of three young children, that’s a big benefit worth paying for. And more important than that, it leads to greater compliance, and we have a problem in this country with not enough people being vaccinated, and of course a vaccine only works if enough people take it.)

So there are consequences to price and consequences to availability. What do I mean? There’s a product called Bendectin. It was the only therapy ever approved for morning sickness. Morning sickness for those of you who have wives who’ve had it, is not just some mild nausea. It can lead to dehydration, it can lead to hospitalization. The product was on the market, approved by FDA safe and effective, and was withdrawn only because of product liability concerns. At one point the product liability costs came very close to the total amount the company was earning from the product. So the product is not available here. It’s available in Europe. FDA made a formal determination recently that it was not withdrawn for reasons of safety and effectiveness. Not available here in the US. Is this a good thing? I would say no. Norplant, a contraceptive that was implanted, is not available in the US because of product liability concerns.

In general companies steer away from a lot of research in women’s health issues. Mr. Schaeffer said, well, you know, Wyeth has a healthy pipeline. But the question is what drugs are not being developed and what’s happening is there’s a focus on therapies for cancer and other serious diseases rather than on therapies for the broader population and for chronic diseases. And what’s happening is this product liability and drug safety situation is driving companies in exactly the opposite direction of where we want them to be.

We have learned that the best bang for your public health buck is of course for people to eat right and exercise, but after that to address chronic conditions. The best thing that happened in the public health probably over the last 10 or 15 years is the advent of the statin drugs. If you don’t believe me, ask Bob Temple, who’s the dean of drug approvals at FDA. And again, Mr. Schaeffer said, well, all Vioxx was trying to do was expand markets.

Well, let’s be clear about what expanding a market is. Expanding a market is finding a new use for a drug. And so if you have a drug that is an AIDS drug and you find that it is enormously effective for hepatitis B and it’s the best therapy for hepatitis B, is this a bad thing? Is that just expanding markets trying to make money? No. This is a good thing. We want to find FDA-approved, science-based uses for new drugs. I certainly do agree with Jack that there are going to be disincentives to study products further.

And so what has FDA done in its response? First of all, there is this increased over-caution. I think the pink sheet yesterday reported that David Graham said something I think is remarkable. He said, we shouldn’t wait for statistical significance or scientific evidence before we act on a drug safety issue. Well, okay! Why don’t we just pull them all off the market? Sid Wolfe thinks no one should use a new drug for the first five years after it’s approved. Well, if nobody uses the new drug for the first five years after it’s approved, there won’t be any new drugs. And so the suggestion is that companies should just warn. Just tell the people what they need to know. But it’s not that easy to come up with a warning.

Again, Vioxx did have warnings with respect to cardiovascular risk. The question is the adequacy, and it’s always going to be possible when we discover new things, and science is dynamic and knowledge marches forward and we’re always going to learn new things about drugs once they get put into the broader population, it’s always going to be possible for plaintiff’s lawyer to say to 12 jurors, probably most of whom do not have a background in science, they should have said this instead of that.

Let’s be clear. Drugs are more closely regulated than perhaps any other product in the economy (except perhaps nuclear power plants). You can’t test it, market it, say a word about it, or manufacture it without FDA approving every jot and tittle of what’s on the label, what’s in the drug. People hear about a new drug application and they think it’s an application like you fill out for college. A new drug application has enough boxes of documents that it would probably at least fill this room. When an NDA is given over to the agency, literally trucks come by and they give over the data, and FDA reviews all of the data. There is no product that is or can be more comprehensively regulated.

The real public policy question here is, what additional utility in terms of the public health and safety does the tort system provide? Does it make drugs safer and more effective, more available, lower cost? And I would say the question to all of those questions is no.

I’m not going to go into great detail about risk management plans. The hour is late and we do want to leave people time to questions and have some discussion. But FDA has increasingly been using these risk management plans. They are highly problematic from a legal perspective. As I often say, they’re extra-legal. FDA does not have the explicit authority to use them. Of course a company is free to agree to conditions as a condition of getting its drug on the market, but the case law is very clear that FDA has no right to put manufacturers to the Hobson’s choice of either accepting recommendations for actions which the agency couldn’t require, or facing rejection of their NDA’s, even though that’s what’s happening all the time.

FDA is also doing this drug safety oversight board and there are many in Congress who want it to go much further and have an independent board. The problem is this disassociates the benefits from the risks and people do what you pay them to do. If all the incentives for these drug oversight boards are just to