Wednesday, August 31, 2005
7th Circuit Reverses Verdict Against Orthopedic Implant Manufacturer Based on Unreliable Testimony re Causation and Defect
Sensing, perhaps, that their expert's opinion might profit from some bolstering, plaintiff's counsel in Fuesting opened their appellate brief with a five-page discussion of published literature from medical journals to support their expert's theory -- viz., that implant sterilization by gamma irradiation in air causes ultra high molecular weight polyethylene components to oxidize. The Seventh Circuit's opinion concedes that the cited literature shows "some modicum of acceptance" for the expert's theory, but strikes this portion of plaintiffs' appellate brief nevertheless, because the cited articles (it says) were not part of the trial record. Plaintiffs' counsel did quote them in cross-examining defendant's expert witness. But the quoted portions "almost invariably met with [the defense expert's] disapproval." According to the panel, the conditions for the hearsay exception in Fed. R. Evid. 803(18) were therefore unsatisfied. (Our own understanding is that Rule 803(18) depends only on the expert's agreement that the learned treatise or periodical is authoritative, and not on the expert's agreement with the particular content on which the cross-examiner seeks to rely. But that's an issue for another day.)
The remainder of the Seventh Circuit's opinion is decidedly on the interventionist end of the appellate spectrum. It faults the district court for a desultory reliability analysis that merely referred to the reasoning offered by plaintiff's expert at deposition, rather than reciting his reasoning anew or addressing defendants' objections specifically. But it does not remand with instructions for the district court to conduct a more thorough reliability analysis, nor show any explicit deference to the reliability findings that the district court actually made. It faults plaintiff's expert for not citing or personally conducting tests or experiments, complaining that the expert seeks "insulation of his theory from the dispassionate crucible" (whatever "the dispassionate crucible" may be), without however adverting to the panel's earlier acknowledgment that plaintiff's theory does enjoy some measure of support in the literature. It criticizes plaintiff's expert for developing his opinion for litigation purposes -- without discussing whether it was his specific opinion, or the underlying theory on which it was based, that was crafted for the occasion. And it conducts an unusually searching appellate analysis of "analytical gaps":
Rather than relying upon tests, experiments or studies, [plaintiff's expert] attributed his opinion to "basic polymer science," which, even [defendant] concedes, reveals that gamma irradiation of polyethylene can create free radicals that bond with oxygen, thereby decreasing its molecular weight by keeping molecular chains from reforming, increasing its density, and making the polyethylene susceptible to delamination. The problem here is that [plaintiff's expert] did not bridge the analytical gap between these basic principles and his complex conclusions. For example, he did not specify, with respect to [plaintiff's] implant in particular, what quantum of each variable is required to set this agreed upon chain reaction in motion. How much radiation does it take to cause oxidation, and to what degree? How much oxidation must occur to render polyethylene more susceptible to delamination? And once polyethylene becomes more susceptible to delamination, how then does oxidation affect delamination? Are all forms of polyethylene, including that used by [defendant] (which the company claims to be oxidation-resistant), susceptible to delamination? What effect, if any, does implantation into the human body have on the rate of oxidation? Basic polymer science alone cannot answer these questions. Some greater methodology is required to bridge the analytical gap between general principles and particular conclusions, and to vest thereby the opinion with requisite reliability.We're not saying the outcome is wrong (though it does seem questionable to undo a verdict partly on the ground that the district court's reliability analysis was inadequate, without remanding for a new analysis). We do suspect, however, that the panel's outlook may have been affected by its perception that the plaintiff/appellant's citations to the record (in the stricken portion of the briefing) were misleading. Readers can take a look at plaintiff's (corrected) appellate brief and judge for themselves. Maybe it would have been better, in retrospect, to have acknowledged forthrightly, in the brief, that the quoted material was from counsel's questions, and that the quoted articles were not themselves in the record. But does the brief really evince an intent to mislead? It does source the quotations to identified locations in the appellate record.
The moral may be that in matters of candor to the tribunal, the appearance of propriety can be as important as propriety itself.
New Study Finds No Link Between Cell Phones, Brain Cancer
Friday, August 26, 2005
ACSH Asks EPA to Limit Reliance on Animal Testing in Identifying Carcinogens
Possible Chelation Fatality
Now comes a tragic report that one autistic five-year-old has died after a course of chelation. The mother reportedly doesn't blame the therapy. An official pronouncement on the cause of death awaits further investigation.
Toxic French Fries and the Lone Star State
It is a standard that would doom the claims of many toxic tort claimants. But according to a new study, breast cancer victims who consumed more than four weekly helpings of french fries during their childhoods might have a shot.
Update 8/28/05: Is it the acrylamides? California is on the case.
More on the Medico-Scientific Evidence About Fetal Pain
Courtesy of Health Law Prof Blog, we now learn of a new study, published in JAMA, concluding that fetuses lack the necessary nerve connections in the brain to experience pain prior to 29 weeks, and therefore should not need anesthesia during abortions. An August 24 NYT story has more.
Rule 702 Meets Rule 23(f)
If we understand this opinion correctly, then a representative plaintiff whose expert testimony in support of class certification is excluded by the trial court is without appellate recourse in the Fifth Circuit, except upon eventual entry of final judgment, even though the appellate court does possess interlocutory jurisdiction over the class certification decision under Fed. R. Civ. P. 23(f).
That outcome seems plainly wrong.
Here's what happened in Bell:
(1) Ascendant Solutions, Inc. ("ASI"), committed alleged securities fraud.The Fifth Circuit's opinion first holds that plaintiffs may not simply rely on the pleadings in support of class certification, but must rather come forward with evidence to show that Rule 23's criteria are satisfied -- even where that evidence may overlap with merits issues. On the specific issue of market efficiency, the opinion says:
(2) Plaintiffs sued and moved for class certification under Fed. R. Civ. P. 23(b)(3).
(3) ASI opposed class certification, relying in part on its expert's opinion that ASI's common stock did not trade in an efficient market -- on the basis of which ASI argued that the "fraud-on-the-market" theory was unavailable to plaintiffs, and that plaintiffs therefore could not show reliance on a class-wide basis.
(4) Plaintiffs countered with an expert of their own, who opined that ASI's stock did in fact trade in an efficient market.
(5) The district court: (a) struck plaintiffs' expert testimony as unreliable under Daubert; (b) found that plaintiffs lacked sufficient alternative evidence to show an efficient market; and (c) denied class certification.
(6) Plaintiffs took an interlocutory appeal under Fed R. Civ. P. 23(f).
[A]lthough "there is no requirement for expert testimony on the issue of market efficiency ... many courts have considered it when addressing this determination, which may often benefit from statistical, economic, and mathematical analysis." [Unger v. Amedisys, Inc., 401 F.3d 316, 323 n.6 (5th Cir. 2005).] Indeed, though Unger admonishes district courts "not to insist upon a 'battle of the experts' at the class certification stage," id., we quoted with approval a statement from the district court's opinion in this case in defense of considering at least the reliability of expert testimony on market efficiency at the class certification stage. Id. at 323-24 n.6 ....
Under Unger, in other words, market efficiency may not simply be presumed from the bare allegations of the complaint. Under Unger, the "market efficiency" issue is sufficiently integral to the class certification analysis that district courts should consider devoting special scrutiny to the parties' expert evidence on that subject.
So far, so good. Reasonable people could perhaps differ over just how deeply the district courts should scrutinize expert evidence at the certification stage. Cf. In re Visa Check / Mastermoney Antitrust Litig., 280 F.3d 124 (2d Cir. 2001) (district courts deciding class certification should merely ensure that basis of expert opinion is not so fatally flawed as to render it inadmissible as a matter of law). But the level of scrutiny contemplated by the Fifth Circuit does seem rationally defensible.
The Bell opinion goes on, however, in the very next breath, to say:
This second thought seems in significant tension with the first. Does the panel's refusal to review the district court's Daubert ruling rest entirely on the claimed concession, by plaintiffs in Bell, that such review is impermissible? If that were so, there would be little reason to cite Bertulli. In the event, there's little reason to cite it anyway. Nothing in Bertulli, it transpires, deals with the question whether the admissibility of evidence offered in support of class certification may be reviewed in connection with a Rule 23(f) appeal of the district court's class certification ruling.
Because we hear this appeal on an interlocutory basis, ... our review is bridled by rule 23(f). Under that rule, "a party may appeal only the issue of class certification; no other issues may be raised." Bertulli v. Indep. Ass'n of Cont'l Pilots, 242 F.3d 290, 294 (5th Cir. 2001). Consequently, as plaintiffs concede, we may not review the exclusion of their expert report, so we must look to the remainder of their market efficiency showing and determine whether the district court abused its discretion in finding it wanting.
Indeed, the principal reason given by the Bertulli panel for limiting the scope of Rule 23(f) appeals (and its principal guidance defining what those limits are) is that merits issues should not be addressed at the class certification stage under Eisen v. Carlisle & Jacquelin, 417 U.S. 156 (1974). (Take a look at footnote 7 in Bertulli.) Of course, if Eisen meant that the courts must be agnostic on every Rule 23 issue that happens to overlap with a merits question, then it would be error for the courts even to consider the "market efficiency" issue at the class certification stage (because "market efficiency" is indeed a merits issue). Plaintiffs in Bell, that is, would be right, and the Fifth Circuit wrong. But if Eisen's prescription of agnosticism on the merits requires (as the Bell opinion insists) only that the certifying court rest its ruling on the evidence relating to the elements of Rule 23, and not on the perceived strength of the underlying claims, then little reason appears for reading Bertulli as the Bell panel does. In other words, if Eisen's mandated agnosticism on merits questions does not preclude the district court's consideration of the reliability of expert evidence on market efficiency, it doesn't preclude appellate consideration of that issue either.
Once in a great while, logic harmonizes with practicality. The idea behind interlocutory appeals under Rule 23(f) is to permit early appellate resolution of class certification issues, without requiring either party to wait until final judgment. Defendants want early appellate review, it has been said, because they must meanwhile live under the in terrorem effect of any class certification order. Plaintiffs want prompt appellate review of adverse certification rulings too, because their claims may not be economically viable for trial as individual actions. To permit interlocutory review of certification orders, while delaying until final judgment all appellate review of the evidentiary rulings on which the class certification orders depend, pointlessly escalates transaction costs and maximizes uncertainty.
Sunday, August 21, 2005
Prof. Heinzerling's Doubts About Daubert
Saturday, August 20, 2005
Minnesota Supreme Court Holds Frye Inapplicable to Expert Evidence on Battered Child Syndrome
The Frye standard is particularly unsuitable, says the MacLennan opinion, for "syndrome" evidence, in which the expert is not permitted (in Minnesota, anyway) to opine on whether the individual actually suffers from the relevant syndrome. A "syndrome" expert may testify (in Minnesota, anyway) only in general terms, by way of describing the syndrome's typical features. The admissibility of such testimony, under MacLennan, should depend on whether it satisfies Minn. R. Evid. 702, which focuses inquiry primarily on whether the evidence will assist the jury.
Monday, August 15, 2005
Does Peer Review Work?
Saturday, August 13, 2005
Mississippi Supreme Court Upholds Defendant's Medmal Causation Testimony Under Daubert
Daubert and Cross-Racial Identifications
Maryland's highest court has just issued a puzzling 4-3 opinion on this subject, holding that it was error to disallow closing argument on cross-racial identification where the identifying witness claimed to be "extremely good with faces." See Smith v. State, No. 116 (Md. Aug. 12, 2005). The witness's identification was sole significant evidence against the defendants, and no expert evidence on cross-racial identifications was offered.
The Court of Appeals opinion begins by canvassing the social scientific literature on the issue, concluding that although most research seems to support the existence of some differential between the accuracy of same-race and cross-racial identifications (especially as regards identifications of blacks by whites), there is no clear consensus on the strength of the effect. For that reason, says the opinion, "we cannot state with certainty that difficulty in cross-racial identification is an established matter of common knowledge."
So far, so good. If indeed it is not a matter of jurors' common knowledge that cross-racial i.d.'s are less reliable, it should follow, seemingly, that the inference cannot permissibly be argued in closing, absent record evidence to support it. But the Court of Appeals reaches the opposite conclusion in the opinion's very next sentence: "Here, however, the victim's identification of the defendants was anchored in her enhanced ability to identify faces."
The question is, how does the witness's claim to be an above-average identifier of faces in general open the door to any inference of a cross-racial differential? We're at a loss. (We can see how it would open the door to cross-examination on the issue, but that's another story.)
We're not saying the result is incorrect. Maybe criminal defense counsel should have wide latitude in closing arguments, and the theory may enjoy enough empirical support to warrant a per se rule permitting such arguments without reliance on expert testimony. Maybe expert testimony would just foster uncertainty and inconsistency from case to case.
But we don't follow the opinion's reasoning at all.
The Basis and Reasons Therefor: Rule 703 and Expert Discomfiture
We don't mean to pick on anyone who has managed, in the great American tradition, to secure publication, in an industry journal, of an "article" in which his new web service is hawked -- except to mention that it all reminds us of an old skit on Saturday Night Live, where Steve Martin played a medieval barber and would-be healer known as Theodoric of York. You remember:
Joan: You charlatan! You killed my daughter, just like you killed most of my other children! Why don't you admit it! You don't know what you're doing!But Dr. Feinman's article does evince a more general confusion over the rules of evidence that it might be helpful to address, before it becomes an urban legend in the "tort reform" wars.
Theodoric of York: [steps toward the camera ] Wait a minute. Perhaps she's right. Perhaps I've been wrong to blindly follow the medical traditions and superstitions of past centuries. Maybe we barbers should test these assumptions analytically, through experimentation and a "scientific method." Maybe this scientific method could be extended to other fields of learning: the natural sciences, art, architecture, navigation. Perhaps I could lead the way to a new age, an age of rebirth, a Renaissance! [thinks for a minute ] Naaaaaahhh!
As Dr. Feinman recounts the saga of his brush with malpractice liability, the tort system can have at least one beneficent effect. It can send doctors scurrying to the medical library, even if the project of designing effective treatments for their patients had not originally supplied sufficient occasion to do so. The lawsuit in which Dr. Feinman was initially named involved a woman who died after treatment for an infection and chronic renal failure secondary to multiple myeloma. Once served with process, Dr. Feinman tasked a medical librarian with a literature search about the range of appropriate treatments in such a situation. The outcome? "We found no evidence," he says, "that any treatment was more effective than what I had ordered, or that my treatment could have hurt the patient's prognosis."
Dr. Feinman took his research to defense counsel and said he wanted to invoke it in his trial testimony. (Although the claims against Dr. Feinman had been dismissed by now, his testimony was still required on the claims against other defendants.) The following colloquy reportedly ensued:
As events actually unfolded at trial, Dr. Feinman says, "my attorney managed to give me the opportunity to explain to the jury how valid and relevant medical evidence supported my treatment decisions." This, perhaps, might be seen as one clue that the system is not so badly broken, or in need of "reform," as Dr. Feinman had initially supposed. And indeed, as a trip to the law library would have revealed, there is, in fact, at least one entire Federal Rule of Evidence devoted to the very problem that Dr. Feinman found so troubling. Under Fed. R. Evid. 703:
Defense counsel: Medical evidence from clinical trials isn't easy to get admitted in court. It's considered hearsay. An expert or defendant can bring his own experience, knowledge, and training to support his testimony, but you can't bring actual copies of medical or scientific evidence, or quote from it.
Dr. Feinman: You mean I can rely on that evidence to form my opinion, but I can't cite the data from the clinical trials, or bring copies of the trial results to court? You mean a $500-an-hour expert can say whatever he wants in court, based on his individual experience, but you can't admit medical data that proves that a specific intervention may have caused or prevented a bad outcome?
Defense counsel: Yes, that's how the medicolegal system works.
Dr. Feinman: In that case, the system is broken, and it's time for reform.
The facts or data in the particular case upon which an expert bases an opinion or inference may be those perceived by or made known to the expert at or before the hearing. If of a type reasonably relied upon by experts in the particular field in forming opinions or inferences upon the subject, the facts or data need not be admissible in evidence in order for the opinion or inference to be admitted. Facts or data that are otherwise inadmissible shall not be disclosed to the jury by the proponent of the opinion or inference unless the court determines that their probative value in assisting the jury to evaluate the expert's opinion substantially outweighs their prejudicial effect.
A review of the Advisory Committee Notes to Rule 703 can help to place the rule within the larger context of evidentiary procedures under the modern adversary system. Enacted in 1975, Rule 703 was actually designed to broaden the permissible sources on which expert opinion might be founded beyond the limits then prevalent in many jurisdictions. Those traditional limitations may now seem quaint in Daubert's aftermath. We now live in an era where experts can scarcely sneeze without first citing peer-reviewed literature in support. But in olden times, before disco stalked the earth, the permissible bases for expert opinion were theoretically confined, in many courts, to facts established either through the expert's personal observations or via testimony from other witnesses in court. Rule 703 expanded the range of material available for expert consideration in recognition of the plain truth that experts commonly and necessarily consult a wide range of data and background material in their normal professional practice, when not in court.
What nonlawyers may find puzzling is that the evidentiary issues don't stop there. If the expert can rely on learned literature in forming his opinions, surely he should be permitted to invoke that literature on the stand, during direct examination, to defend and explain his opinions, no?
Well, not necessarily.
For the benefit of the uninitiated: If a learned treatise or a piece of medical literature is going to come into evidence, it is important, for various reasons, to keep track of what proposition the evidence is offered to prove. That need arises, in part, from firmly entrenched rules about hearsay. The adversary system relies very heavily on cross-examination to test whatever evidence the parties may offer. Lawyers can't cross-examine a journal article (as opposed to the article's authors), and so some of our system's key procedures for enabling the jury to weigh and assess the evidence in rational fashion would be unavailable, if journal articles could routinely be offered directly, without further ado, to prove the truth of their contents. To be sure, many scientists and physicians might be unbothered by that outcome. They might believe, for example, that the discursive crucible of peer review is already a far more reliable engine of truth than jury trials. (Certainly no practicing physician would be temperamentally inclined to make treatment decisions by assembling twelve random laypersons in a room and asking for their collective judgment on how to proceed.) But neverteheless, it is a bedrock principle of our existing system that evidence, in general, must be sponsored by witnesses who claim to know that the evidence is true, and who are prepared to answer the parties' searching questions about such claims under oath.
Consider Dr. Feinman's own malpractice case (or more precisely, a hypothetical case modeled on Dr. Feinman's account). Substantive medical malpractice law requires the jury to decide whether the treatment afforded by the physicians satisfied the prevalent standards of medical care. And according to Dr. Feinman himself, he was unacquainted with the entire concept of a "standard of care" before he was sued. It isn't taught in medical school, he says. And in the actual practice of medicine, doctors don't pause during CPR to have earnest conversations with their colleagues about what other reasonable practitioners of their specialty might do under comparable circumstances.
Fair enough. It follows, perhaps, that Dr. Feinman could not himself testify to the applicable standard of care -- not, at least, unless he first went and educated himself at some such venue as a medical library. In that library, he might discover a dusty tome entitled The Mayo Clinic Guide to the Standard of Care for Treating Infections and Renal Failure Secondary to Multiple Myeloma. And if, by training and experience, Dr. Feinman is the sort of fellow who possesses specialized knowledge enabling him to interpret the Mayo Clinic Guide in now forming a helpful testimonial opinion on the standard of care, the rules allow him to do so.
But suppose now that Dr. Feinman is the only defense witness to testify about the standard of care. And suppose further, and quite hypothetically, that Dr. Feinman breaks down in tears on cross-examination and says: "You know what? The defense lawyers are holding my family hostage, and they have threatened to torture and kill my wife and children them unless I give this opinion. I don't really believe a single word of the testimony I've just offered." May the jury permissibly find that the defendants' version of the standard of care is nevertheless correct, because after all, the Mayo Clinic Guide vouches for it?
Well, no, you see, because the Mayo Clinic Guide is hearsay. (Here we are leaving recondite peripheral issues, such as Fed. R. Evid. 803(18), to one side.) And thus if Dr. Feinman is going to mention the Mayo Clinic Guide to the jury during direct examination, we have to be mindful of the risk that the jury will be confused about whether that dusty tome counts as direct evidence on the standard of care, or merely as support for any opinion that Dr. Feinman himself may offer. So at a minimum, we have to weigh the pros and cons, and then maybe tell the jury, if we let the reference in, what the Mayo Clinic Guide is and is not in evidence to prove. By the same token, of course, there is also some risk of confusion if Dr. Feinman is forced to take the stand and testify to his ultimate opinion without being able to defend it by explaining how it's something more than a mere pontification. And so we have to give poor Dr. Feinman some chance to defend his views.
The solution afforded by Rule 703 is to vest discretion in the trial judge about how much of the underlying bases for his opinions an expert may disclose on direct. In actual practice, courts appear to be relatively lenient on this point, in the Daubert era, unless some meaningful risk of confusion or prejudice is presented. Few trial judges, we believe, will currently confine direct examination to an unsupported statement of the expert's conclusions (much as many juries might actually prefer that). But even when and if they do, the structure of the rules creates a natural tendency for those bases to emerge as the natural product of the adversary system. If an expert offers only an unexplained conclusion on direct examination, the other side will be loathe to forgo attack, and so there will be cross-examination about how on earth the expert can hold such an opinion, and what he bases it on. And once that cross is conducted, the door is opened for further exploration of the foundations for the expert's opinions on redirect examination. Meanwhile, we were really able to get to Kansas all along, whenever we wanted, just by clicking our heels together three times and cross-examining the adversary's expert about the Mayo Clinic Guide.
Experts are often uncomfortable with these procedural mechanisms, because they are often understandably uncomfortable with the entire adversary process. In the normal practice of their disciplines, experts commonly have the opportunity to state and defend their opinions in sustained expository fashion, without interruption, in whatever manner they please. Their views won't be elicited through a tiresome process of Q&A, and they won't be confined to answering whatever questions somebody else sees fit to ask -- somebody, perhaps, whose level of understanding the expert holds in fairly low regard. Serving as an expert witness at trial is an inherently irritating process, and experts frequently bemoan their unaccustomed lack of control over the direction of courtroom inquiry. We're not saying such complaints are unfounded. But we do think there's an important distinction, of which both experts and lawyers often lose sight, between: (1) "The legal system won't allow witness w to state proposition x at time y under circumstances z," and, e.g., (2) "The legal system 'can't admit medical data that proves that a specific intervention may have caused or prevented a bad outcome.'"
Meanwhile, there are tensions and paradoxes lurking within Rule 703's approach. In part, the rule reflects the same ambivalence manifest in Daubert about how far juries may be trusted to grasp the underlying theories and data on which expert opinions are based. No doubt experts could find it more than a little ironic to be severely limited in the explanations they can provide to the jury, when every tiny little step in their reasoning had to be explained and rigorously defended, often at painful length, during pretrial proceedings on admissibility. We'll say this much in favor of Dr. Feinman's ideas for reform: His instinct, apparently, is to give more information to the trier of fact, rather than suppressing more.