C:\documents and settings\eohcnm.001\local settings\temporary internet files\content.ie5\ps4vpx01\s74[1].pdf

Ten Years of Judicial Gatekeeping Under Daubert and certainty dissolves into probability.”2 In the ten years since Daubert v Merrell Dow Pharmaceuticals, Inc, the standards for admissibility at trial of expert testimony in general and scientific evidence in Legal doctrines developed to guide judges in particular have become more demanding. Reviews of recent cases and empiri- considering such difficult issues of science cal studies of federal judges’ and attorneys’ practices indicate that judges are also are likely to be complex. However, it is more likely to consider the admissibility of expert evidence prior to trial, to inquire worth noting as a preliminary matter that more deeply into the reasoning and methodology that supports the expert opin- much of the testimony offered by testifying ions, and to limit or exclude such evidence from presentation at trial. Studies of experts does not rise to this level of complex- ity. Such expert testimony requires special Recent cases consider more difficult questions arising from the differing meth- expertise, but much of this specialized testi- odologies used in various areas of science. The current legal framework that as- mony appears rather routine when measured sesses admissibility in terms of professional practice outside the courtroom is against the evidence presented in Daubert or poorly suited to cases that require expertise across a wide range of specialties and other cases involving the effects of exposures force judges to choose from among different scientific methodologies. Futureresearch should focus on the pretrial screening of expert testimony and interac- tions between the attorneys and experts in shaping that testimony. (Am J Pub- In one of the earliest empirical studies of lic Health. 2005;95:S74–S80. doi:10.2105/AJPH.2004.044776) expert testimony, Gross analyzed reports froma California jury verdict reporting service todetermine the type and frequency of experts Many federal judges were uncertain how the the tips of the trees that sway in the winds testifying in state civil trials that ended in a 1993 Supreme Court decision in Daubert v while the forest floor remains unstirred? jury verdict.4 Drawing on this data set, Gross Merrell Dow Pharmaceuticals, Inc,1 would af- I review here the limited amount of empiri- and Syverud reported that more than half of fect their work. But Judge Alex Kozinski, au- cal research that has assessed the effect of the the testifying experts were physicians, with thor of the Ninth Circuit appellate court deci- Daubert decision on civil litigation as well as only 3% identified as scientists.5 Champagne, an emerging series of research studies in- Schuman, and Whitaker surveyed attorneys, further consideration by the Supreme Court, tended to strengthen our understanding of judges, jurors, and experts in civil cases in was worried. In reconsidering the case in light the role of expert testimony in courts. I also Texas and confirmed that about half of the of the standards expressed in Daubert, Judge examine a recent case involving conflicting experts were physicians, with scattered repre- evidence from court-appointed scientists, indi- cating the difficulties that arise in applying a More recently, my colleagues and I at the Our responsibility, then, unless we badly mis- standard based on professional practice in read the Supreme Court’s opinion, is to resolvedisputes among respected, well-credentialed sci- cases that require multiple areas of expertise.
judges regarding expert testimony in their entists about matters squarely within their ex- most recent civil trial.7 We then surveyed at- pertise, in areas where there is no scientific torneys who served those same trials. In both consensus as to what is and what is not “goodscience,” and occasionally to reject such expert the judge and attorney surveys, we asked re- testimony because it was not “derived by the spondents to describe the types of experts scientific method.” Mindful of our position in Discussions of admissibility of expert testi- who testified in terms of the occupational cat- the hierarchy of the federal judiciary, we take adeep breath and proceed with this heady task.
mony typically focus on some of the most de- egories used by Gross. Again, medical and manding areas of science. Although judges mental health specialists were the most fre- often consider evidence similar to that used quently presented category of experts, ac- sors and other scholars have filled the law in assessment of public health issues, judges counting for more than 40% of the experts library shelves with articles analyzing pub- must consider such evidence in the context of presented overall. More specifically, medical lished cases following Daubert. But published deciding an individual case rather than the doctors (from all specialties) accounted for cases represent only a portion of the litiga- establishment of a broad social policy. The approximately one third of all experts pre- tion; many more cases are resolved without a Daubert case involved conflicting testimony sented. Specialists from business, law, and published order or decision.3 Are these pub- regarding epidemiology, toxicology, and phar- lished cases indicative of broad shifts in the macology, characterized by Judge Kozinski as orientation of the courts that extend through- concerning “matters at the very cutting edge out litigation? Or are the published cases like of scientific research, where fact meets theory were by far the most common type of expert S74 | Public Health Matters | Peer Reviewed | Cecil American Journal of Public Health | Supplement 1, 2005, Vol 95, No. S1 within this category, as well as the most com- became clear that the interpretation of the de- dence (and, therefore, are likely to follow cision was resulting in a more restrictive ap- Daubert) felt that their gatekeeping role had almost 12% of all experts. (Economists were proach to admissibility of scientific testimony.
changed as a result of Daubert, and only a included in the financial category because A recent analysis by the Rand Corporation of third of the judges believed the intent of they most often present estimates of lost prof- a sample of 399 published and unpublished Daubert was to raise the threshold for admis- its or wages and other financial projections federal district court decisions appearing in sibility. The study also found that state court rather than apply broad economic theories to the Westlaw database over a 20-year period the facts of a case.) Engineers and other indicates the extent to which courts have of the meaning of the scientific criteria sug- safety or process specialists registered close shifted toward excluding proffered scientific gested by Daubert and raised questions behind the business/law/financial sector, ac- and technical evidence.8 That analysis indi- about the ability of the state trial courts to counting for about 22% of all experts. Such cates that challenges to case “elements” in- experts respond to a wide variety of issues volving expert evidence rose in the 3 years thoughtful manner. Few judges, for example, following the Daubert decision.
were able to define the concept of “falsifia- Scientists composed a small portion of the This interpretation is consistent with much bility,” which was one of the factors men- testifying experts. Specialists from scientific of the commentary that followed Daubert.
tioned in Daubert. Other studies have fields such as chemistry, ballistics, toxicology, Many cases decided soon after Daubert in- demonstrated that judges may be insensitive and metallurgy accounted for only 8% of the volved clear instances of evidence that was to methodological problems in social science experts that testified at trials. In this group, not well supported by acceptable scientific research such as experimenter bias or the chemists were the most frequent type, repre- methodology and permitted easy application senting 1.6% of the total number of experts.
of the Daubert standards. For example, the Epidemiologists and toxicologists, who were courts rejected the use of an industrial rather change in the pattern of federal litigation. Be- the focus of the Daubert decision, were rare, than medical methodology in assessing the ginning in 1997, the proportion of cases that together constituting just over 1% of the testi- presence of asbestos fibers in lung tissue,9 the were challenged for unreliable evidence de- fying experts. In comparing these results with substitution of visual inspection for appropri- clined, as did the success rate for such chal- a similar 1991 survey, we concluded that the ate medical testing to conclude that cataracts lenges.8 The authors interpret this as evidence distribution of types of expert testimony in were cause by radiation exposure,10 and sub- that parties responded to the post-Daubert civil trial has changed very little following jective opinions that failed to consider other standards by strengthening the quality of the plausible explanations for harmful effects.11 expert testimony or abandoning those cases Judges reported that the most frequent is- Exclusion of expert testimony in such cases that were unlikely to meet the higher stan- sues addressed by experts at trial were the was generally straightforward and increased dards. Changes in publication practice once existence, nature, or extent of injury or dam- the confidence of judges in reviewing the the case law interpreting Daubert became es- age (68%) and the cause of injury or damage tablished may also have contributed to the (64%), which finding is consistent with the The Federal Judicial Center survey of fed- fact that tort cases represented almost half eral judges and attorneys also confirmed a As the federal courts evolved to consider of all cases reported. Testimony as to the shift toward more demanding standards for more difficult issues of expert testimony, the amount of recovery to which plaintiff was en- distinction between methodology and conclu- titled was offered by experts in 44% of trials; judges and attorneys indicated that judges sions became blurred, and application of the this type of testimony is consistent with the were more likely to scrutinize expert testi- Daubert standards became more difficult. The mony before trial and to limit or exclude Supreme Court decision in General Electric Co ported as having testified. Other issues ad- v Joiner 14 illustrates the increasing difficulty in dressed by expert testimony were the reason- Daubert litigation practice in 1991.7 The considering issues of scientific evidence dur- ableness of a party’s actions (in 34% of trials), survey revealed that motions filed early in ing this period. In such cases, often the issue industry standards/“state of the art” (30%), litigation have become a favored pretrial de- is not the absence of scientific evidence, but vice for challenging the admissibility of ex- rather whether the existing scientific evidence (25%), design or testing of a product (25%), pert testimony and that judges are focusing can be generalized to address the specific and knowledge or intent of a party (16%).
more attention in pretrial proceedings on causal relationships alleged in the case. In Joiner, for example, the plaintiff contended DAUBERT AS A GATEKEEPING
that exposure to polychlorinated biphenyls clear. A 1998 survey of state court judges had promoted the development of his small- by Gatowski et al. found judges split in their cell lung cancer. After establishing an abuse- At the time of the Daubert decision, there assessment of the effect of the Daubert deci- of-discretion standard for appellate review, was considerable uncertainty about its effect sion.12 Only about half of the judges from on admissibility of scientific evidence. It soon states that follow the Federal Rules of Evi- ord in the case and demonstrated the limited Supplement 1, 2005, Vol 95, No. S1 | American Journal of Public Health Cecil | Peer Reviewed | Public Health Matters | S75 extent to which courts should permit experts Daubert (i.e., in 1991 and 1998), the most THE REVEALING CASE OF SOLDO v
to generalize beyond established scientific SANDOZ PHARMACEUTICALS CORP
counter is “experts who abandon objectivity The plaintiff in Joiner had presented a se- Under these Daubert/Jointer/Kumho stan- ries of epidemiology studies that offered hired them.” This perceived lack of objectiv- dards, expert testimony such as clinical med- equivocal findings when taken separately, ity may have a number of different sources.
ical testimony that employs a mix of tradi- but according to the plaintiff, demonstrated a Experts are selected by the parties based on causal relationship when considered together.
the extent to which their testimony will ad- rigorous case study and observational meth- The plaintiff also presented studies with in- vance the parties’ claims, a practice that may odologies is especially problematic. In effect, fant mice showing that injections of large favor the selection of extreme viewpoints.19 the Kumho decision tethered the standard for doses of polychlorinated biphenyls led to admissibility of testimony by physicians to the cancer. The Supreme Court rejected such evi- offer testimony involves a socialization pro- professional standards of medical practice.
dence after focusing on the breadth of gener- cess that is likely to encourage the expert to But many courts have been reluctant to ac- alization implied by the plaintiff’s expert testi- identify with the interest of the party.4 It is cept opinions based on these less rigorous reasonable that judges, who likely were ex- methodologies, even though they are widely posed to such practices prior to their arrival used in clinical medicine.21 More is going on Trained experts commonly extrapolate from on the bench, would be skeptical of testi- here than a straightforward distrust of wit- existing data. But nothing in either Daubert orthe Federal Rules of Evidence requires a dis- nesses. Even when the experts are clearly ob- trict court to admit opinion evidence which is undergone such selection and coaching.
jective, that is, appointed by the court with no connected to existing data only by the ipse The gatekeeping role has evolved beyond a previous association or contact with the par- dixit of the expert. A court may conclude thatthere is simply too great an analytical gap be- device for reviewing only scientific evidence ties, a judge may question the extent to which tween the data and the opinion proffered.
to include all types of expert testimony. In the appointed expert’s testimony can properly Kumho Tire Co v Carmichael,20 which involved inform consideration of the issues in the case.
an allegedly defective automobile tire, the The recent case of Soldo v Sandoz Pharma- this analytical gap between research data and ceuticals Corp22 is revealing of the manner in expert opinion regarding the issues presented approach to all types of expert testimony.
which courts take account of different areas in specific litigation. Because such cases often Prior to Kumho, the courts were divided on of expert testimony. Although this is a single examine opinions based in part on research whether expert testimony based on experi- case rather than a research study, it incorpo- methodologies that are employed in the nor- ence and clinical medical testimony in partic- rates many aspects that would be desirable in mal course of scientific inquiry, they cannot ular should be subject to the Daubert screen- a research study. Reputable scientists from be dismissed out of hand in the manner of ing process. In extending the trial court’s different disciplines with no ties to the parties those early cases decided soon after Daubert.
gatekeeping obligation to all expert testimony, Instead of dismissing the methodology used the Supreme Court noted that “no clear line” prepared independent reports regarding the by the experts, judges who exclude such testi- can be drawn between the different kinds of scientific validity of the proffered testimony, mony typically focus on the reasoning process knowledge, and “no one denies that an expert which was then reconciled by the judge in re- of the experts, questioning whether the meth- might draw a conclusion from a set of obser- solving a motion to exclude the testimony.
ods and findings relied on by the experts can vations based on extensive and specialized Viewed in this light, it might be properly con- reasonably be extended to the facts of the experience.” Although the specific factors sidered a “case study” incorporating critical case at hand. As part of such an analysis, re- mentioned in Daubert may not be relevant to elements that no formal research study has cent courts have dismissed experts’ reliance nonscientific expert testimony, other factors yet achieved and is therefore appropriate to on animal studies15 cautioned against extrapo- may provide a suitable standard for assessing lation of dosage levels16 and objected to gen- such testimony. The Supreme Court indicated In Soldo, a young mother sustained an that all expert witnesses should employ “in intracranial hemorrhage and resulting stroke In questioning the propriety of generaliz- the courtroom the same level of intellectual soon after giving birth. She claimed that ing research findings to the facts of specific rigor that characterizes the practice of an ex- the stroke was a result of her ingestion of litigation, courts often imply that the experts pert in the relevant field.” In effect, this deci- Parlodel, a drug manufactured and marketed have abandoned an objective and impartial sion tethered the standard for admissibility of by the defendant to prevent lactation. Be- role appropriate for scientists and become expert testimony to standards of professional cause a heightened risk of stroke occurs in advocates for the parties.18 Judges have indi- practice. This reliability requirement has also cated persistent concern regarding the possi- the tasks facing the court was to determine if ble lack of objectivity on the part of testify- eral Rule of Evidence 702, strengthening the role of the court in assessing the foundation taken Parlodel following pregnancy exceeds of all expert testimony proffered for litigation.
S76 | Public Health Matters | Peer Reviewed | Cecil American Journal of Public Health | Supplement 1, 2005, Vol 95, No. S1 miologic research, it would require a tremen- nancy. No meaningful clinical trials or epide- The three court-appointed experts, working dous amount of indirect evidence to reach the miology studies of Parlodel exist, so the plain- independently, developed three different in- point that even in the absence of research, the tiff’s experts considered a series of animal terpretations of the scientific reliability of the linkage is ‘obvious’ in the way that the tornado studies, case studies, and other clinical reports testimony by the plaintiff’s experts, thereby in concluding that Parlodel caused the injury.
Regarding the possibility of legitimate and In assessing the scientific reliability of the ployed by various scientific disciplines in as- reasonable disagreement with his views, the testimony by the plaintiff’s experts, the court sessing a similar body of evidence. All of the epidemiologist acknowledged that views vary sought the assistance of three court-appointed experts—a clinical pharmacologist, a neurolo- plaintiff’s experts had made the most of the gist, and an epidemiologist. The three court- data that exist, but they differed in the extent [T]he vast majority of scientists who routinely appointed experts were identified with the consider these sorts of evidence (epidemiolo-gists, researchers in clinical medicine) would assistance of the Registry of Independent Sci- meeting the standards of “scientific knowl- agree with my general conclusions. If forced to entific and Technical Advisors, administered edge.” The epidemiologist and the neurologist guess the proportion, I would estimate 80% of by the Private Adjudication Center of Duke found that the testimony did not meet the my peers would concur. Those who studybasic mechanisms of disease causation (physiol- University School of Law (http://www.law.
standards of scientific knowledge, but for very ogists, pharmacologists, toxicologists) might duke.edu/pac/registry/index.html). Unfortu- different reasons. The clinical pharmacologist well dispute my views in that the plausibility nately, this service has been suspended. A found that the testimony, with one exception, based on those lines of evidence is more sup-portive of the potential for causality. The similar service for identification of court ap- met the standards of scientific medical testi- counterargument to my view is that the di- pointed experts for federal litigation is avail- verse threads based on mechanism of actions able through the American Association for for Parlodel, analogy to other agents in thesame broad category of drug, and temporal linkage of the medication and the illness can information to be sufficiently reliable to be be integrated scientifically into a scientifically (http://www.aaas.org/spp/case/case.htm).
deemed “scientific knowledge” was offered reliable conclusion. However, essentially all sci-entists recognize that when the issue is the All were well-regarded scholars with aca- by the court-appointed epidemiologist. He causation of clinical disease in humans, there is demic appointments at distinguished universi- focused on the “analytical gap” in applying a sizable gap between what is plausible based ties and medical schools. The experts had no the research to the facts of the case, noting on indirect evidence and what is proven basedon clinical and epidemiologic studies. In the previous association with any of the parties chain of reasoning, most scientists would likely and had not previously considered the role share the view that leaping across the huge “[B]ecause the information is so indirectly ap- gulf of critical data moved the person making plicable and hypothetical in nature, the appli- the inference beyond the scope of science.
cation of it to form an opinion is not a “scien- advice of the appointed experts on “whether tifically reliable” process. The linkage between The least demanding standard for assessing those shreds of potentially relevant informa-tion and the opinion that results is so murky a causal relationship was offered by the clini- plaintiff’s medical witnesses . . . in formulat- that it is very difficult to see how the evidence ing their opinions is scientifically reliable leads to the opinions that are offered. Apply- epidemiologic and clinical trial data would be ing any reasonable standards of scientific evi-dence as the basis for drawing a conclusion helpful in assessing the causal relationship, can be properly applied to the facts of this leads to the judgment that we do not know case.” The court recited an expanded ver- enough to offer an opinion on this matter that sion of the Daubert factors and noted that is reasonably well grounded in science.
[M]any important and well-recognized adverse published studies need not be required as a drug reactions are not documented by well basis for testimony, that “differential diagno- The court-appointed epidemiologist was es- conducted epidemiologic studies that have pecially critical of the absence of human studies been specifically conducted to detect them andthat we still believe that the weight of the sci- formed, would generally meet the factors,” and was unwilling to extrapolate findings across entific evidence is sufficient to implicate their and that there must be a proper “fit” be- species, noting that, “Some form of epidemio- involvement to such an extent that we would tween the expert’s opinion and the facts in logic or clinical evidence, even if flawed and in- remove a drug from the market. . . . [M]ostclinical practice is not guided by data from the case. The court further instructed that if complete, is needed for drawing inference [prospective, randomized, placebo-controlled about general causation in making a judgment clinical trials], because they are so difficult and that the testimony of the party’s expert is about Parlodel and intracerebral hemorrhage.” expensive to conduct. In addition, aggregatedata from such trials often do not apply to the not “scientifically reliable,” then the ap- He acknowledged that causal attributions may specifics of individual cases. To assert that any pointed expert should also indicate if opin- be made without such studies if the causal medical practice has no scientific basis because ion by the party’s expert might represent a pathway is clear, such as injuries that arise a randomized, placebo-controlled trial to an-swer the pertinent question was not conducted, “legitimate and reasonable” minority view when a tornado hits a mobile home park, but would be to obviate the vast majority of clini- continued, “In the absence of clinical or epide- cal practice. It follows that other tools must Supplement 1, 2005, Vol 95, No. S1 | American Journal of Public Health Cecil | Peer Reviewed | Public Health Matters | S77 usually be used to provide sufficient evidence ies. In the absence of specific human or ani- though most courts have excluded such testi- mal evidence that Parlodel causes vasocon- mony regarding Parlodel,23,24 at least one striction in cerebral arteries, he found the as- other federal court decided that such differ- sertion of such an association to be without gist found the plaintiff’s experts’ assertion of The Soldo court took no comfort in the a causal link to be scientifically reliable (with diversity of views presented by the experts.
one minor exception). In reaching this con- plaintiff’s experts was not scientifically reli- Faced with such conflicting opinions, the clusion, the clinical pharmacologist used a able, the court-appointed neurologist then as- court attempted to reconcile disputes over sci- “totality of the evidence” approach, bringing sessed the extent to which others in his field entific validity that the scientific community together assessments of animal studies, case might express “a legitimate and responsible” itself had not resolved. In so doing, the court studies, and other available evidence, “none contrary opinion. He acknowledged that oth- of which taken separately may be determi- ers outside his specialty of experimental phar- Kumho in assessing whether the methodology and techniques met “the same level of intel- may be considered convincing.” Although might disagree with his conclusion because lectual rigor that characterizes the practice of such an approach is common in public health they are unfamiliar with the published scien- an expert in the relevant field” and moved on assessments, judges tend to consider sepa- tific evidence regarding the difference in re- to establish legal policy regarding how such rately the validity of each piece of evidence.
sponse of cerebral vasculature and peripheral disputes among scientists were to be recon- Although the epidemiologist dismissed case vasculature to such drugs; he also stated that studies as inherently unreliable, the clinical such knowledge is not typically part of the pharmacologist relied on a series of case training of specialists in the broader fields of published pages, the court dismisses the views studies, including one case study reported in neurology and cerebrovascular disease.
of the court-appointed clinical pharmacologist a peer-reviewed journal that recorded vaso- and the plaintiff’s experts and concludes that constriction changes in carotid arteries with may be found in medical textbooks and peer- the opinions expressed by the plaintiff’s ex- reviewed articles and, in the absence of a perts “failed to use a reliable scientific method- clear cause for the hemorrhage, some mea- ology” to demonstrate general causation and In finding a causal link, the clinical phar- sure of subjective judgment is required in as- specific causation. The court does not explicitly macologist reasoned that evidence of vaso- sessing the evidence. In conclusion, he noted address the use by the court-appointed clinical constriction in peripheral arteries of dogs that other persons generally qualified in this pharmacologist of the “totality of the evidence” and humans caused by Parlodel and related field of expertise might legitimately disagree test, but it notes that while expert opinions substances would permit the inference that may make “appropriate use of all of the avail- Parlodel would cause similar vasoconstriction able information, . . . in the absence of some in cerebral arteries, which may then lead to minimum amount or level of scientific evi- intracranial hemorrhage and stroke. Such rea- dence, the opinions cannot be scientifically de- soning far exceeds the modest stretch across opinions from three distinguished scholars rived because there is too little science from the “analytical gap” that would have been al- from three different disciplines. One can which to derive them. Although it is sometimes lowed by the court-appointed epidemiologist.
necessary in a clinical, regulatory, or business The third court-appointed expert, a neurol- regret, felt by the judge when he reviewed practice to make decisions based on less than ogist, was willing to consider animal studies these three reports that shared little in com- sufficient and/or reliable scientific evidence mon. It is likely that in calling for indepen- due to practical demands that require immedi- more direct evidence of a vasoconstrictive dent assessments from three different special- ate decision-making, such guesses, although effect of Parlodel on cerebral arteries. When ties, he had hoped that the opinions would perhaps reasonable hypotheses based on the the neurologist applied this standard to the converge, thereby strengthening assessment best available evidence, do not constitute a sci- proffered testimony of the plaintiff’s experts, of the scientific methodology underlying the entifically reliable approach when used to as- he too found it lacked necessary scientific experts’ opinions. Instead, he found a dra- sess causality via the scientific method.” matic illustration of the variation in accept- clinical or epidemiologic data are not neces- able methodologies across well-established court-appointed epidemiologist and neurologist: sary to declare a causal relationship, but he objected to reasoning by analogy that evi- At first glance, one might assume that three The body of scientific evidence relating toParlodel and stroke is simply insufficient dence of vasoconstriction in peripheral and different opinions from highly qualified court- to support a scientifically reliable application carotid arteries is indicative of a similar rela- appointed experts would in itself be evidence of plaintiff’s expert methodology. . . . Without tionship in cranial arteries. He pointed out of “a legitimate and responsible disagree- sufficient evidence of general causation,plaintiff’s experts could not reliably apply a that cerebral arteries respond differently to ment” regarding the disputed issues and re- differential diagnosis that comports with the drugs like Parlodel than do peripheral arter- quire that a jury resolve the conflict. Al- scientific method, notwithstanding the fact S78 | Public Health Matters | Peer Reviewed | Cecil American Journal of Public Health | Supplement 1, 2005, Vol 95, No. S1 that physicians in clinical practice may be re- broadly approved professional practices are Through a series of case studies, the Sci- quired to proceed with a differential diagno- scientifically unsound and, therefore, do not ence, Technology & Law Panel plans to exam- sis on the basis of guesses or hypotheses dueto the exigency of the need to treat their meet the standards for admissibility.
ine a number of distinctive features of such lit- igation, including the litigation circumstances that brought about the research, the extent to Having excluded the testimony of the plain- which the participants in litigation participate tiff’s experts, the court then granted summary in the design, analysis, and interpretation of The skepticism of courts toward expert tes- the research, and the extent to which the re- timony in general and scientific testimony in search finds an audience beyond the partici- appointed clinical pharmacologist did not particular seems rooted in a view of science pants in the litigation. This study is intended meet a sufficient standard of scientific reliabil- and litigation that has caused the courts to to identify criteria that courts can use to assess ity, the court made clear what many previous be extremely cautious in acknowledging the the quality of such scientific studies.
decisions have obscured; that is, conclusion value of some scientific methodologies in pro- The Science, Technology & Law Panel of of a casual relationship by a distinguished viding an informed assessment of causal rela- the National Academies also has developed a scholar unaffiliated with the parties and using tionships. Consequently, courts often declare proposal to examine the scientific foundation generally accepted methods of clinical infer- common methods of professional assessment of forensic science testimony submitted in ence may not be sufficient to allow a court to based on animal research or clinical inference criminal cases. This topic has been generally submit such differences of opinion to a jury.
to be so lacking in scientific rigor that they neglected by the broad scientific community, Such a view would appear to conflict with the fail to meet a suitable standard for considera- assurance of the Supreme Court in Kumho Kumho decision with the extension of eviden- that expert opinions are admissible if they Several research projects currently under tiary standards of reliability to all area of ex- employ “in the courtroom the same level of development are likely to shed new light on pert testimony. An empirical analysis of appel- intellectual rigor that characterizes the prac- the use of scientific and clinical medical testi- late decisions in criminal cases by Groscup et tice of an expert in the relevant field.” For this mony in litigation. Carl Cranor and David al. indicates that the Daubert factors have court and an undetermined number of others, been rarely used outside of forensic areas that admissible evidence requires more than meet- National Science Foundation to develop re- are clearly scientific.26 A more explicit role in ing appropriate standards of professional views by independent scientists of synopses screening expert testimony in criminal cases practice; it also requires demonstration of a of expert testimony used in a particularly has emerged following Kumho and the amend- relationship through methodologies that are difficult toxic tort case. The assessments of ments to Federal Rule of Evidence 702. The not an essential part of clinical practice.
the scientists can then be compared with the panel will convene a committee to formulate The court in Soldo indicated as much when court’s interpretation of what constitutes reli- a research agenda for forensic science disci- it noted that even if the plaintiff’s expert opin- able scientific evidence. Such a study is likely plines related to identification (e.g., fingerprints, ions were admissible under Daubert, “such evi- to sharpen the comparison of the interpreta- tool marks, footprints, tire treads, question doc- dence provides but a scintilla of support for tions of scientific methodology inside and out- uments, and handwriting) to strengthen the sci- plaintiff’s position and would not be sufficient entific methodology underlying these areas of to allow a reasonable jury to find that plain- The Science, Technology & Law Panel of forensic science and to promote academic re- tiff’s [intracerebral hemorrhage] had been the National Academies has developed a pro- caused by Parlodel.” One might argue that this posal to examine the characteristics of “litiga- is a more appropriate basis for making such a tion science,” or science that has been devel- which attorneys identify, recruit, and prepare decision rather than for striking the testimony oped in the context of the litigation process.
experts for testimony. Most difficult will be iden- as inadmissible because of some perceived Judge Kozinski recognized that science devel- tifying the manner in which consulting experts flaw in methodology and reasoning. Even if oped in the context of litigation may be sub- who are not designated to testify at trial are the experts use methods and reasoning appro- ject to pressures that may distort the findings used to shape the claims and defenses, because priate to their profession, the courts may set, such activities are protected by attorney work- as a matter of law, a minimum threshold for product privilege. Finding a similar opportunity One very significant fact to be considered is evidence that is sufficient to justify submission whether the experts are proposing to testify to explore these issues with those who have to a jury. Of course, if such a legal standard is about matters growing naturally and directly served as experts will also be difficult.
established in an explicit manner as being in- out of research they have conducted indepen- dent of the litigation, or whether they have de- sufficient as a matter of law, then that decision veloped their opinions expressly for purposes of extent to which courts conduct pretrial in- would be subject to appellate review on a de testifying. . . . [I]n determining whether pro- quiries into the reliability of expert testimony.
novo basis and more vulnerable to reversal on posed expert testimony amounts to good sci- The Federal Judicial Center examined such ence, we may not ignore the fact that a scien- appeal. Specifying such a legal standard, how- tist’s normal workplace is the lab or the field, activities in cases that went to trial, but that ever, would seem preferable to declaring that not the courtroom or the lawyer’s office.
study missed those cases in which such an Supplement 1, 2005, Vol 95, No. S1 | American Journal of Public Health Cecil | Peer Reviewed | Public Health Matters | S79 examination resulted in the case terminating in a post-Daubert world. Law Hum Behav. 2001;25: prior to trial. We do not know the extent to which judges engage in the screening of ex- 13. Kovera M, McAuliff B. The effects of peer reviewand evidence quality on judge evaluations of psycho- pert testimony as part of a routine pretrial logical science: are judges effective gatekeepers? J Appl process and how this screening process varies 14. General Electric Co v Joiner, 522 US 136 (1997).
15. Newman v Motorola, Inc, 218 FSupp2d 769, 780-1(D Md 2002).
16. Amorgianos v National Railroad Passenger Corp, The author is with the Federal Judicial Center’s Program on Scientific and Technical Evidence, Washington, DC. 17. Mitchell v Gencorp Inc, 165 F3d 778, 782 (10th Requests for reprints should be sent to Joe S. Cecil, Fed- eral Judicial Center, One Columbus Circle, NE, Washing-ton, DC 20002 (e-mail: jcecil@fjc.gov). 18. Cacciola v Selco Balers, Inc, 127 FSupp2d 175, This article was accepted August 9, 2004. Note. The views expressed in this article are those of
19. Elliott ED. Toward incentive-based procedure: the author and do not necessarily represent the views of the three approaches for regulating scientific evidence.
Boston Univ Law Rev. 1989;69:487–508.
20. Kumho Tire Co, Ltd. v Carmichael, 526 US 137 Portions of this paper were presented at the Coronado 21. Kassirer JR, Cecil JS. Inconsistency in evidentiary Conference on Scientific Evidence and Public Policy, standards for medical testimony: disorder in the courts.
Coronado, Calif, March 2003. Portions of the discus- sion of the Soldo v Sandoz case also appear in “Constru-ing science in the quest for ‘ipse dixit ’: a comment on 22. Soldo v Sandoz Pharmaceuticals Corp, 244 23. Glastetter v Novartis Pharm Corp, 252 F3d 986(8th Cir 2001).
Daubert v Merrell Dow Pharmaceuticals, Inc, 509 24. Rider v Sandoz Pharmaceuticals Corp, 295 F3d Daubert v Merrell Dow Pharmaceuticals, Inc 25. Globetti v Sandoz Pharmaceuticals Corp, 111 (Daubert II), 43 F3D 1311, 1316 (9th Cir 1995).
Songer DR. Nonpublication in the United States 26. Groscup JL, Penrod SD, Studebaker CA, Huss MT, district courts: official criteria versus inferences from O’Neil KM. The effects of Daubert on the admissibility appellate review. J Politics. 1988;50:206–215.
of expert testimony in state and federal criminal cases.
Psychol Public Policy Law. 2002;8:339–372.
Gross SR. Expert evidence. Wisc Law Rev. 1991; 27. Cecil JS. Construing science in the quest for “ipsedixit ”: a comment on Sanders and Cohen. Seton Hall Gross SR, Syverud KD. Don’t try: civil jury ver- dicts in a system geared to settlement. UCLA Law Rev.
1996;44:1–64.
Champaign A, Schuman D, Whitaker E. Expert witnesses in the courts: an empirical examination. Judi-cature. 1992;76:5–10.
Krafka CL, Dunn MA, Johnson MT, Cecil JS, Miletich D. A survey of judges’ and attorneys’ experi-ences, practices, and concerns regarding expert testi-mony in federal civil trials. Psychol Public Policy Law.
2002;8:309–332.
Dixon L, Gill B. Changes in the standards for ad- mitting expert evidence in federal civil cases since theDaubert decision. Psychol Public Policy Law. 2002;8:251–308.
Braun v Lorillard Inc, 84 F3d 230, 234 (7th Cir 10. O’Conner v Commonwealth Edison Co, 13 F3d1090, 1106-7 (7th Cir 1994).
11. Claar v Burlington Northern R. Co, 29 F3d 499,502 (9th Cir 1994).
12. Gatowski SI, Dobbin SA, Richardson JT, GinsburgGP, Merlino ML, Dahir V. Asking the gatekeepers: anational survey of judges on judging expert evidence S80 | Public Health Matters | Peer Reviewed | Cecil American Journal of Public Health | Supplement 1, 2005, Vol 95, No. S1

Source: http://defendingscience.com/sites/default/files/upload/CecilTENYEARS.pdf