Opinion
Civil Action No. 2:96-0163.
7-24-1998
Harry K. BLACK, et al., Plaintiffs, v. RHONE-POULENC, INC., Defendant.
Roger D. Hunter, Neely & Hunter, Charleston, WV, Wilson H. Barnes, Henry Dart, Metairie, LA, for Plaintiffs. J. Rudy Martin, Lynn Oliver Frye, James Eric Whytsell, A. L. Emch, Jackson & Kelly, Charleston, WV, Wilson H. Barnes, Metairie, LA, for Defendant.
Pending is the remainder of Defendant's motion for summary judgment as to Plaintiffs' claims for intentional and negligent infliction of emotional distress. The Court conducted extensive hearings pursuant to Daubert v. Merrell Dow Pharmaceuticals, Inc., 509 U.S. 579, 113 S.Ct. 2786, 125 L.Ed.2d 469 (1993) on March 13, May 27-28 and July 6-7, 1998 relating to Plaintiffs' experts. As a result of the hearings, the Court GRANTS the remainder of Defendant's motion and DISMISSES Plaintiffs' claims for negligent and intentional infliction of emotional distress.
I. FACTUAL BACKGROUND
Plaintiffs are putative class representatives of persons seeking redress for events that occurred on February 15, 1996. A fire broke out at Rhone-Poulenc's Institute, West Virginia plant on that date. The fire occurred in the Toluene Drying System, a system that removes water from toluene, located in the Sevin Unit at the plant. It was extinguished within one hour. The apparent cause of the mishap was the failure of a pump.
Pre-established emergency response procedures were implemented by Defendant and public officials. As a result, some area residents were ordered to "shelter-in-place" at approximately 7:00 a.m. and certain thoroughfares were closed temporarily. The shelter in place was lifted at approximately 8:20 a.m.
The pump involved in the fire had been visually checked by Sevin Unit operator Chris Myers no more than twenty minutes before the fire. Myers claims to have not seen or heard anything unusual. Myers also testified another co-worker passed the pump just prior to the fire:
He walked right by that pump and then walked probably 150 feet into the control room. And when he walked by that pump, it wasn't on fire, but when he came in the control room the alarm came in. Then I walked immediately out the door, and it was on fire.
Dep. of Chris Myers, Def.'s mot. for summ. jgt. ex. 7 at 110.
Following the fire, Rhone-Poulenc determined that 9,186 pounds of toluene was lost from the system. The toluene was over 99% pure and contained trace amounts of chemicals used in the Sevin manufacturing process including water, Sevin, methyl isocyanate ("MIC"), naphthol, ethyl benzene, dimethyl urea and chloroform.
Plaintiffs retained Dr. Joseph Scotti, a clinical psychologist, in support of their emotional distress claims. Dr. Scotti filed a "Preliminary Report on Class Action Plaintiffs and General Population Survey" on September 23, 1996 and was deposed on September 25, 1996. Although discovery did not close until December 31, 1997, the Scotti report was never supplemented.
The Court continued in a Memorandum Opinion denying Defendant's motion to dismiss in May 1996: "Plaintiffs have promised reliable expert testimony in support of the emotional distress alleged.... [T]he Court will expect [such] testimony." McClenathan v. Rhone-Poulenc, 926 F.Supp. 1272, 1279 and n. 10 (1996).
The report is the result of a study Dr. Scotti performed for Plaintiffs concerning the emotional effects of the release on area residents. The study was designed by Dr. Scotti and an associate. Dr. Scotti commented as follows on the design of the study:
[We were to] [d]ecid[e] what would be the most appropriate instruments to use for an evaluation of psychological behavioral effects, to determine the course of those symptoms over time, that is, what were people like before the event, right after and then currently, currently being July and early August. And what would be reliable valid measures, what measures are common to this area of research.
Dep. of Dr. Scotti at 82.
The population of the study was 162 people, 36 of whom were claimants intending to sue Rhone-Poulenc. This was done ostensibly for purposes of comparing Plaintiffs with non-Plaintiffs. The balance of the study group, contrary to what is stated in Dr. Scotti's report, was taken from several different groups, which the Court details below.
The participants first were read an introduction and then given a "self-report packet" to determine the effects of the fire on them emotionally. Pls.' resp., ex. 11. The tests contained in the packet were, inter alia, (1) the State-Trait Anxiety Inventory (STAI); (2) the Beck Depression Inventory (BDI); (3) the Impact of Event Scale (IES); (4) the Symptom Checklist-90-Revised (SCL-90-R); (5) the Modified Fear Survey (MFS-III); (6) the Keane Post-Traumatic Stress Disorder Scale of the MMPI (PTSD Scale); and (7) the History of Psychosocial Stressors (HPS), a measure apparently created by Dr. Scotti. Some of the tests for some of the periods were modified by Dr. Scotti from their original form as received from the author.
The STAI is a psychometric device for determining levels of arousal, distress, and anxiety.
The BDI is a psychometric device for determining levels of depression.
The IES is a psychometric device utilized in the field of traumatology to assess the frequency of avoidant and intrusive symptoms. This scale was completed for the post-fire and summer 1996 periods only.
The SCL-90-R is a psychometric device for assessing a range of psychological symptoms through ten subscales, such as somatization, depression and anxiety, and a global severity index. It was completed for the summer 1996 period only.
The MFS-III is a measure of fears and phobias with subscales for, e.g., sexual assault, medical-treatment fears and classical fears. It was completed for the summer 1996 period only.
The PTSD Scale is a subscale of the Minnesota Multiphasic Personality Inventory (MMPI). It was designed to assess the level of distress related to traumatic events and was completed for the summer 1996 period only.
The HPS is a checklist of traumatic events. The participant indicates which events he or she has experienced and rates the level of distress associated. It was used to rank the relative severity of events a person has experienced over their lifetime compared with that experienced during the fire.
As a result of the testing, Dr. Scotti and his associate concluded, in part, "that two factors were overwhelmingly related to the psychological and behavioral impact experienced by this sample of individuals ... Exposure to the toxic cloud ... and being forced to Shelter-in-place. " Pls.' ex. 11 at 9 (emphasis in original). For each test used, Dr. Scotti and his associate provided two sources of comparative information: (a) established clinical norms for the test; and (b) the scores obtained with other trauma populations such as combat veterans and auto accident survivors.
In addition to Dr. Scotti, Plaintiffs retained Dr. Thomas Schrager to testify on certain toxicology issues. Dr. Schrager's proposed testimony, however, is much more circumscribed than that of Dr. Scotti. By example, Dr. Schrager has indicated he will not be offering any opinions on causation.
Plaintiffs filed this action on February 26, 1996. Previous rulings by the Court have left Plaintiffs with claims for (1) strict liability pursuant to Restatement (Second) of Torts § 519; (2) intentional infliction of emotional distress; (3) negligent infliction of emotional distress; and (4) negligence. In its ruling denying in part Defendant's motion for summary judgment, however, the Court stated as follows:
Pending is Defendant Rhone-Poulenc, Incorporated's motion for summary judgment. The Court DENIES in part Defendant's motion and sets the remaining issues of the motion for a hearing pursuant to Daubert v. Merrell Dow Pharmaceuticals, Inc., 509 U.S. 579, 113 S.Ct. 2786, 125 L.Ed.2d 469 (1993).
The Court concludes genuine issues of material fact remain extant on Plaintiffs' strict liability, negligence and punitive damages claims. Regarding the strict liability claim, the Court further lacks sufficient technical and other information on the toluene drying system at issue to make a fully informed decision as to whether the practice constitutes an abnormally dangerous activity for purposes of Restatement (Second) Torts § 519.
The viability of Plaintiffs' claims for intentional and negligent infliction of emotional distress hinge largely on the admissibility of the findings and conclusions of Dr. Joseph R. Scotti. The affidavits of Dr. Paul Lees-Haley and Dr. Bruce K. Bernard signal serious shortcomings in Dr. Scotti's methodology and analysis, which could result in the exclusion of his opinions under Daubert. Accordingly, the Court sets a Daubert hearing for evidentiary development of this issue ....
Summ. jgt. op. at 1-2 (footnote omitted).
Given the United States Supreme Court's recent decision in Bragdon v. Abbott, ___ U.S. ___, 118 S.Ct. 2196, 141 L.Ed.2d 540 (1998), delaying a final decision on the summary judgment motions appears to have been an approved course. The Supreme Court stated in Bragdon "Scientific evidence and expert testimony must have a traceable, analytical basis in objective fact before it may be considered on summary judgment." Id. at ___, 118 S.Ct. at 2212.
The Court was further presented with Defendant's attacks on the methodology forming the basis of the opinions of Dr. Schrager and stated as follows:
The Court also is concerned with the basis of the opinions of Plaintiffs' [toxicology] expert Dr. Thomas Schrager as discussed in footnote 23 of Defendant's memorandum in support of summary judgment. Nonetheless, it appears Plaintiffs may now not be seeking to use Dr. Schrager's expertise to prove causation and exposure, relying instead on individual Plaintiff testimony of adverse health effects they experienced in close temporal proximity to the fire that are consistent with exposure to the chemicals at issue. Accordingly, this issue may be moot. To the extent Plaintiffs intend to rely on Dr. Schrager's testimony [on these issues], however, and to the extent Defendant objects to that testimony under Daubert, the issue also should be addressed at the [Daubert] hearing.
Id. at 2 n. 2.
The Court conducted a five-day hearing spaced out over five months. After hearing the testimony, reviewing the transcripts and exhibits and receiving further submissions from the parties, the admissibility question is ripe for determination.
The hearings were spread out over this period of time to accommodate the scheduling conflicts of counsel and the expert witnesses.
II. DISCUSSION
A. Controlling Legal Principles
At the outset, the Court notes there are many cases applying the Daubert considerations to equally as many differing factual scenarios. Accordingly, the Court believes it more appropriate, given the standard of review to examine the testimony here with reference to the factual and general legal principles contained in decisions emanating only from the United States Supreme Court and our Court of Appeals. Cf. Calderon v. Thompson, ___ U.S. ___, ___, 118 S.Ct. 1489, 1506, 140 L.Ed.2d 728 (1998)("[T]he variety of subjects left to discretionary decision requires caution in synthesizing abuse of discretion cases.")(Souter, J., dissenting).
The district court's determination on the question of admissibility is subject to an abuse of discretion standard. See, e.g., U.S. v. Scheffer, ___ U.S. ___, ___, 118 S.Ct. 1261, 1271, 140 L.Ed.2d 413 (1998)("Well reasoned opinions are concluding, consistently with this Court's decisions in [Daubert and General Elec. Co. v. Joiner, ___ U.S. ___, ___, 118 S.Ct. 512, 519, 139 L.Ed.2d 508 (1997)] ... that the federal rules wisely allow district judges to exercise broad discretion when evaluating the admissibility of scientific evidence.")(Stevens, J., dissenting); General Elec. Co. v. Joiner, ___ U.S. ___, ___, 118 S.Ct. 512, 519, 139 L.Ed.2d 508 (1997); Talkington v. Atria Reclamelucifers Fabrieken BV (Cricket BV), No. 97-1386, 1998 WL 394593, at *9 (4th Cir. Jul.15, 1998).
Our Court of Appeals commented recently at length concerning the applicable standard of review in Maryland Cas. Co. v. Therm-O-Disc, Inc., 137 F.3d 780, 784 (1998):
Daubert clearly contemplates the vesting of significant discretion in the district court with regard to the decision to admit expert scientific testimony. Beyond establishing the two criteria of reliability and helpfulness, the Court has left the means by which these criteria are evaluated to the sound discretion of the district judge. This is apparent not only from Daubert itself, but from subsequent Supreme Court precedent, holding that abuse of discretion is the proper standard by which evaluations of proffered evidence should be reviewed. In addition, this circuit has taken the position that the Daubert court "was not formulating a rigid test or checklist," and was "relying instead on the ability of federal judges to properly determine admissibility."
Id. (footnotes omitted)(quoted authority omitted); see also Cavallo v. Star Enter., 100 F.3d 1150, 1153-54 (4th Cir.1996)(rejecting a stricter standard of review from a sister circuit); Benedi v. McNeil-P.P.C., Inc., 66 F.3d 1378, 1383-85 (4th Cir.1995); United States v. Dorsey, 45 F.3d 809, 814 (4th Cir. 1995); United States v. Powers, 59 F.3d 1460, 1471 (4th Cir.1995); United States v. Bynum, 3 F.3d 769, 773 (4th Cir.1993). Of particular note here is the Court of Appeals' observation that a district court does not abuse its discretion in excluding challenged scientific testimony where "extensive, unanswered evidence weigh[s] against the scientific validity of the" testing. Powers, 59 F.3d at 1471.
Our Court of Appeals' adherence to the abuse of discretion standard proved prescient. The Supreme Court rejected the plenary-review approach of the Third Circuit evidenced in In re Paoli R.R. Yard PCB Litig., 35 F.3d 717, 744 (3d Cir.1994) in favor of the less restrictive standard employed by the Fourth Circuit. General Elec. Co. v. Joiner, ___ U.S. ___, ___, 118 S.Ct. 512, 519, 139 L.Ed.2d 508 (1997).
Daubert worked a significant shift in the jurisprudence controlling the admissibility of expert testimony. In replacing the "general acceptance" test announced in Frye v. United States, 293 F. 1013 (D.C.Cir.1923), Daubert emphasized a "flexible" approach to the determination, consistent with the "liberal thrust" of the Federal Rules of Evidence on the question of admissibility. Daubert, 509 U.S. at 588, 594, 113 S.Ct. 2786; Joiner, ___ U.S. at ___, 118 S.Ct. at 517; Ruffin v. Shaw Industries, Inc., No. 94-1882, 1998 WL 394992, at * 2 (4th Cir. Jul.16, 1998); Therm-O-Disc, 137 F.3d at 783; Cavallo, 100 F.3d at 1158 ("[T]he Supreme Court itself viewed Daubert as a liberalization, not a tightening, of the rules controlling admission of expert testimony"); Dorsey, 45 F.3d at 813.
Turning to the mechanics of the admissibility determination, "Daubert instructs district courts to make a `preliminary assessment of whether the reasoning or methodology' underlying expert testimony `is scientifically valid.'" Freeman v. Case Corp., 118 F.3d 1011, 1016 n. 6 (4th Cir.1997). Specifically, the Court in Daubert set forth a two-pronged inquiry to be used by the district court in exercising its gatekeeper role:
Faced with a proffer of expert scientific testimony, then, the trial judge must determine at the outset, pursuant to Rule 104(a), whether the expert is proposing to testify to (1) scientific knowledge that (2) will assist the trier of fact to understand or determine a fact in issue. This entails a preliminary assessment of whether the reasoning or methodology underlying the testimony is scientifically valid and of whether the reasoning or methodology properly can be applied to the facts in issue.
Daubert, 509 U.S. at 592-93, 113 S.Ct. 2786; Therm-O-Disc, 137 F.3d at 783; Powers, 59 F.3d at 1470; Dorsey, 45 F.3d at 813.
To aid the district court in making the determination under the scientific knowledge prong of the inquiry, the Supreme Court set forth a few considerations that have since been employed by our Court of Appeals: (1) whether the testimony, theory or technique can be and has been tested, (2) whether it has been published or exposed to peer review, (3) its rate of error, (4) whether there are standards and controls over its implementation, and (5) whether it is generally accepted. See generally Ruffin, 149 F.3d 294, 1998 WL 394992, at *2; Cavallo, 100 F.3d at 1158, 1159 ("[T]he five factors [Daubert] established ... require that the methodology and reasoning used by a witness have a significant place in the discourse of experts in the field."); Powers, 59 F.3d at 1471; Dorsey, 45 F.3d at 814.
The five considerations are a non-exclusive listing. Daubert, 509 U.S. at 593, 113 S.Ct. 2786; Ruffin, 149 F.3d 294, 1998 WL 394992, at *2; Therm-O-Disc, 137 F.3d at 784; Benedi v. McNeil-P.P.C., Inc., 66 F.3d at 1384 (noting the Daubert factors are not "a rigid test or checklist ...."). Rather, in keeping with the flexible nature of the inquiry, the district court may consider other items it deems appropriate under the circumstances. In that regard, the Court believes a few other considerations are appropriate here:
1. Whether the experts are proposing to testify about matters growing naturally and directly out of research they have conducted independent of the litigation or whether they have developed their opinions expressly for the purpose of testifying. 4 Jack B. Weinstein & Margaret A. Berger, Weinstein's Federal Evidence § 702.05[2] (2d ed.1998).
2. The independence of the investigator and his staff.
3. The presence of methods and practices contrary to "good science."
Because of the close fit, the Court treats this consideration in tandem with the fifth consideration outlined by our Court of Appeals in Cavallo. For the same reason, the Court also combines the rate of error consideration with the considerations of litigation motivation and the independence of the investigator.
B. Analysis of the Applicable Considerations in Relation to Dr. Scotti
Before discussing the applicable considerations, a brief synopsis of the testifying experts' qualifications is in order. Dr. Scotti is a Licensed Clinical Psychologist and an Associate Professor of Psychology at West Virginia University. He has authored peer reviewed articles and has a special interest in the area of post-traumatic stress. Among other things, Dr. Scotti is a member of the Board of the Journal of Traumatic Stress.
Dr. Bernard is a Research Professor at Campbell University and works in the field of hypothesis generation, testing and evaluation in the fields of both toxicology and psychology. He has performed hundreds of scientific studies and has worked extensively with the Food and Drug Administration on acceptability of data. He has performed over 260 peer reviews.
Dr. Lees-Haley is a Licensed Clinical Psychologist. He is board certified with a Diplomate from the American Board of Professional Psychology and the American Board of Psychological Specialties. He has performed peer reviews for the Journal of the American Medical Association; Psychological Reports; Journal of Clinical Psychology; Archives of Clinical Neuropsychology; Perceptual and Motor Skills; Behavioral Sciences and the Law; Applied Neuropsychology; and many others. He is a recognized expert in psychological testing methodologies.
1. Whether the Theory or Technique Can Be and Has Been Tested
Dr. Bernard testified Dr. Scotti's study violated the "prime directive" for scientific research by not proceeding from the null hypothesis, the very starting point for testability. In short, instead of following the scientific method by developing a proper hypothesis and working to disprove it, Dr. Scotti determined there was at least some effect and sought to determine the level of the effect. Both Dr. Scotti's report and his letter to the law firm of Neely & Hunter, Plaintiffs' counsel in this action and the sponsor of the study, are illustrative. Quoting from the report, Dr. Scotti stated as follows: "The purpose of the present investigation was to determine the psychological, behavioral, and physical impact of a technological disaster on the community surrounding the Rhone-Poulenc chemical plant ...." This sentence, along with others in the report and the letter to Neely & Hunter, indicates an "impact" was presumed and that the level of the presumed "impact" was the actual object of the study.
Dr. Bernard's statements concerning the importance of developing an appropriate hypothesis and properly testing it cannot be underestimated: "`Scientific' knowledge is generated through the scientific method — subjecting testable hypotheses to the crucible of experiment in an effort to disprove them. An opinion that defies testing, however defensible or deeply held, is not scientific.'" United States v. Bynum, 3 F.3d at 773 (emphasis added)(quoted authority omitted). The Supreme Court in Daubert likewise observed as follows:
"Scientific methodology today is based on generating hypotheses and testing them to see if they can be falsified: indeed, this methodology is what distinguishes science from other fields of human inquiry."
Daubert, 509 U.S. at 593, 113 S.Ct. 2786 (quoted authority omitted)(emphasis added).
Dr. Bernard also testified, without disagreement from Dr. Scotti, that good science requires the investigator to keep good documentation so the study can be reconstructed wholly independent of the investigator. Independent reconstruction, and thus re-test verification, are largely impossible given Dr. Scotti's poor record keeping. By example, Dr. Scotti did not make a list of the 2000 people he phoned to participate in the study, thus leaving lingering questions concerning bias, more fully discussed below.
Dr. Scotti stated recently this problem has now been solved. He originally agreed the call lists could not be reconstructed. He testified at the final set of hearings, however, that personnel from Neely & Hunter approached him since that time about discarded materials found in the "Class Action Office" at the Days Inn Hotel, the location where potential class members executed claim forms and spoke with Neely & Hunter about the case. The materials were the missing call sheets. Even if this portion of the study could now be reconstructed, however, the safeguarding of the records, i.e., the fact they ended up in the custody of the sponsoring law firm, raises serious questions about the independence of the investigator, as discussed below.
Dr. Scotti also kept no records of those who (1) responded to a flyer he posted at a college located near the site of the fire; and (2) ultimately participated in the study as a result.
The flyer reads, in part, as follows: NOTE: OPINION CONTAINING TABLE OR OTHER DATA THAT IS NOT VIEWABLE
Next, of the 74 claimants on the list provided to Dr. Scotti at the outset of his research, half chose not to participate in the study. No records were kept as to the severity of the injuries of those who participated and did not participate nor their reasons for participating or not participating.
Finally, some participants brought others to the class action office with them when they came to perform the study. There is no documentation of who these individuals were and which ones were permitted to participate.
Dr. Bernard commented that "There is no way to reconstruct this study. There is not documentation. In the absence of it, it is total speculation." Elec. trans. at 270. He suggested this alone was enough to "throw out" the research. Id. Dr. Scotti all but conceded his woeful record keeping but defended by saying his sketchy report did a better job at allowing independent reconstruction than similar studies. Despite his protestations and attempted justifications, however, the Court finds independent reconstruction would be exceedingly difficult if not impossible. This fact, taken in concert with Dr. Scotti's irregular hypothesis generation and other relevant items, weighs against admissibility.
In an effort to expedite its Opinion, the Court's citations of testimony are from the electronic transcript of the hearing. The electronic version's pagination will vary from the hard copy of the transcript.
On the subject of reconstruction, and relating to other attacks on his testimony, Dr. Scotti fell back on, and extrapolated to, studies from other experts he asserted supported his methodology and ultimately his conclusions. More often than not, however, his citation to other studies was sketchy and often tenuous on close inspection. See Joiner, ___ U.S. at ___, 118 S.Ct. at 519 ("Trained experts commonly extrapolate from existing data. But nothing in either Daubert or the Federal Rules of Evidence requires a district court to admit opinion evidence which is connected to existing data only by the ipse dixit of the expert.").
2. Peer Review and Publication
Dr. Scotti noted his participation in two symposiums, one in Florida and one in New York, where he discussed his research in this case. Dr. Lees-Haley, however, obtained a copy of what Dr. Scotti delivered at the Florida conference. He testified Dr. Scotti did not present the methodology he used to his peers at the conference. He said the presentation was quite sketchy and was certainly not an accurate depiction of the Rhone-Poulenc study. The New York presentation appears to have a similar problem. The presentation was but 14 minutes and, again, Dr. Scotti did not reveal several significant issues relating to the study, such as (1) the content of the flyer; (2) the poor record keeping which occurred; and (3) his disclosure to participants of Neely & Hunter's sponsorship of the study. In sum, there is little to support a finding that appropriate peer review occurred. See Ruffin, 149 F.3d 294, 1998 WL 394992, at *2 ("Plaintiffs, on the other hand, failed to submit any evidence of peer reviews supportive of Dr. Anderson's methodology or any proof that her studies had been published.")
Again, Dr. Scotti made reference to other similar studies found in peer-reviewed literature, and suggested his methods and results comport with the peer review requirement. Rarely, however, did he provide sufficient details to permit the Court to perform the necessary comparative analysis between his work here and that of his colleagues elsewhere.
His reference to other works further assumes the publication of an article necessarily validates what is contained therein. As noted by Dr. Bernard, however, "It depends upon how the peer review was performed and who the participants were. There are peer reviews ranging from those which are worthless to those which are of high quality." Elec. trans. at 210. The superficial showing attempted by counsel and Dr. Scotti often did not provide the Court the necessary information to make a reasoned determination on this point, often lacking reference to the author's reputation, much less that of the journal.
The difficulty of simply assuming validity as the result of publication is illustrated in Effie J. Chan, The "Brave New World" of Daubert: True Peer Review, Editorial Peer Review, and Scientific Validity, 70 N.Y.U. L.Rev. 100, 113 (1995):
In addition to being wholly voluntary, editorial peer review also is completely unregulated. It is not a monolith but a variegated collection of different practices. Members of the scientific community are well aware that the types of editorial peer review at different scientific publications vary greatly. And although there is no official "pecking order" of publications based on quality of editorial peer review, there is a general correlation between a journal's prestige and the quality of its editorial peer review.
Id. at 117. The balance of the article highlights significant problems with the entire process of editorial peer review.
The importance of adequate, true peer review cannot be overstated. Equally true is that mere publication of an article is not the end of the peer review process; it is but the beginning. See generally Chan, supra at 113 ("The body of published scientific literature is the most visible and prevalent forum through which modern-day scientific claims are communicated to the global audience of scientists. Journal articles are `temporal statements in an ongoing debate' and their publication is a starting point for the open discourse of true peer review.") (emphasis added).
Considering the hearing testimony, the Court finds Plaintiffs failed to make the necessary showing on the peer review consideration. Accordingly, this factor weighs in favor of exclusion.
3. The Rate of Error
The Court harbored concerns about the potential for error throughout the hearings. Most troubling was the potential rate of error in Dr. Scotti's research emanating from the bias that appears to have infected his study, bias that could be attributed to both the study participants, based on a litigation motive and certain ways they were handled and included in the study, and with Dr. Scotti and his staff, based on a lack of independence and training.
Bias is a serious source of error in scientific studies, especially in a study where one attempts to extrapolate effects found in a small group to a large population. As noted by Dr. Lees-Haley, and not contested by Dr. Scotti, bias is multiplicative and not additive; in common terms, it grows at a healthy, perhaps astounding, rate once introduced. There are many potential sources of bias.
First, while Dr. Scotti insists no one at Neely & Hunter gave him any instructions, the firm provided him a list of 74 Plaintiffs/claimants out of which 36 persons chose to participate. This non-random cluster amounts to nearly 25% of the total number of survey participants.
Second, Dr. Scotti's company, Survey Associates, came into being only as a result of his work on a related chemical leak case for Neely & Hunter several months before the fire. It has apparently performed work only for that law firm. The reason for creating Survey Associates is that Dr. Scotti's institutional employer, West Virginia University, according to his own admission, "wanted to distance itself from ... involvement in this project." Elec. trans. at 35. Dr. Bernard testified that universities rarely distance themselves from researchers and their research.
Third, persons hired by Neely & Hunter were the ones who called study participants and scheduled them to come in for interviews with Dr. Scotti's team. Dr. Scotti said he does not believe those hired by Neely & Hunter would have excluded people called from the reverse directory for an improper purpose, but he did not know for sure. Proper record keeping and control would have permitted reconstruction of the study to demonstrate a lack of bias in participant selection. As noted above, however, that did not occur.
Fourth, the phone number posted on the flyer seeking study participants was the number for the Neely & Hunter Class Action Office at the Days Inn, the same place where individuals also completed claim forms to participate in the litigation.
Fifth, study interviews were conducted in the Class Action Office. This, along with the participation of Neely & Hunter paid personnel in the study process, was particularly troublesome to Dr. Bernard, who stated "at some level the participation in the study can be a reflection of whether they liked or didn't like or agree with or don't agree with the positions of Mr. Neely. That — I mean, that's not science. You can't deal with that." Elec. trans. at 154.
Counsel on both sides concede that this and similar litigation pursued by Neely & Hunter against another chemical company has been the subject of much debate, favorable and unfavorable, in the Kanawha Valley.
Sixth, there was a $5,000.00 witness fee to be paid to Dr. Scotti after the completion of the study.
Seventh, the address on the consent form to participate in the study is: "Survey Associates, c/o Neely and Hunter, Lawyers." This reference is both unnecessary and has a strong potential to introduce bias.
Eighth, there was an indemnification agreement between Neely & Hunter and Dr. Scotti. In commenting on the presence of such an agreement in the field, Dr. Bernard observed "I've been in business 20 years and I've never asked a client to indemnify me against me doing a scientific study ...." Elec. trans. at 153.
Ninth, another potential source of bias emanates from the level of participation in the study that was obtained by utilizing the reverse directory to call over 2,000 individuals. Of the 2,000 individuals called, only 10 to 15% agreed to participate. This level of participation, according to Dr. Bernard, "leaves you wide open for anything to have skewed it. And you must show that [bias] did not occur. We can't show that here." Elec. trans. at 156. The potential for bias with such a low response rate was characterized as "astronomical." Elec. trans. at 249. Dr. Lees-Haley said a 10-15 percent response rate is unacceptable for purposes of a random survey. It shows a strong possibility of bias on the part of the 10-15% because so many others declined to participate.
Dr. Bernard's and Dr. Lees-Haley's concerns regarding the random nature of the selection process and the level of nonresponse is consistent with the position of the Reference Manual on Scientific Evidence:
One suggested formula for quantifying a tolerable level of nonresponse in a probability sample is based on the guidelines for statistical surveys issued by the former U.S. Office of Statistical Standards. According to these guidelines, response rates of 90% or more are reliable and generally can be treated as random samples of the overall population. Response rates between 75% and 90% usually yield reliable results, but the researcher should conduct some check on the representativeness of the sample. Potential bias should receive greater scrutiny when the response rate drops below 75%. If the response rate drops below 50%, the survey should be regarded with significant caution as a basis for precise quantitative statements about the population from which the sample was drawn.
Shari Seidman Diamond, Reference Guide on Survey Research in Federal Judicial Center, Reference Manual on Scientific Evidence 221, 239 (1994)(emphasis added).
The Court also notes there are several significant areas where participant bias could have occurred. First, the flyer discussed above talks about a "Toxic Release" and persons "concerns about [their] Health and Safety." (Emphasis in original). These terms were unnecessary and have a strong propensity to bias. As noted in the hearings, the scientific method seeks an unbiased group to begin with and then takes all measures to keep the group unbiased throughout the study.
In response to concerns about bias, Dr. Scotti suggested conclusorily his use of a random, representative sample knocked out the sources of bias that were present in this case. As demonstrated throughout this Opinion, however, it is apparent Dr. Scotti's sample was anything but random and representative. Even in relation to the variables Dr. Scotti claimed were important in post-traumatic stress research (age, education, intellect, etc.), he conceded (1) these critical variables did not appear in his report nor any charts provided to Defendant; and (2) no comparison was made between the study participants and the community at large on the variables.
There were several examples in the testimony where Dr. Scotti made assumptions, like this one, that were not supported by science or the record. This alone could constitute a basis for excluding his opinions and testimony. Smith v. Virginia Commonwealth Univ., 84 F.3d 672 (4th Cir.1996) (en banc)("[A]n expert's opinion is inadmissible when it is based on assumptions that are speculative and are not supported by the record.")
Second, Neely & Hunter was mentioned several times to the potential participants in the phone script and in the consent form which study participants signed in order to participate in the research.
Plaintiffs pointed out guidelines for research which require the identification in the consent form of sources of external support for the research. As noted by Dr. Bernard, however, deception may be used if disclosure will alter the results of the study by introducing bias. Dr. Scotti agreed he could have used deception, but chose "the welfare of the participant" over the significant potential for bias. Elec. trans. at 473.
Third, there is a statement in the consent form the investigator is looking for effects on the community. This alone could introduce bias, as it informs participants exposure may have occurred and consequently affected the community.
Another opportunity for bias was engendered by less than adequate training of Dr. Scotti's team. First, interviewers were permitted to "paraphrase in [their] own words" the "Introduction for All Participants" that was read to study participants at the outset of their participation. As noted in the testimony, uniformity is critical in psychological testing, lest different stimuli be presented to different persons. No one challenged Dr. Bernard's citation to studies that have found interviewers simply responding "uh huh" to a participant as opposed to "really?" can make a statistically significant difference in the data collected. Elec. trans. at 252.
Second, Dr. Bernard stated 5 hours or more of training is the norm for studies similar to this and surmised there was only 1/2 to 1 hour of training done here. Scotti defended on rebuttal by saying most of the interviews were conducted by his graduate students who had taken courses on validity and reliability, as well as the administration of some of the tests in question. He said the students had more than 1 hour of training and Scotti himself supervised them for the first two weeks. Nonetheless, the Court has serious concerns about training, especially considering events such as those described further below.
Third, responses to oral interview questions were not taken down verbatim. This may seem to be a minor point in isolation. The Reference Manual on Scientific Evidence, however, states "Interviewers also should be instructed to record verbatim the respondent's answers, to indicate explicitly whenever they repeat a question to the respondent, and to record any statements they make to or supplementary questions they ask to the respondent." Seidman, supra, at 257. Once again, the meticulous record keeping that is the hallmark of science played little role in this study.
Fourth, Dr. Bernard testified it would have been best to do a "double-blinded" study, one where neither the interviewer nor the interviewee knew what the investigator was looking for in the study. This study, however, was not even single-blinded. As noted in the Reference Manual on Scientific Evidence, "the interviews themselves are not directly visible, and any potential bias is minimized by having interviewers and respondents blind to the purpose and sponsorship of the survey and by excluding attorneys from any part in conducting interviews and tabulating results." Seidman, supra, at 232; see also Ruffin, 149 F.3d 294, 1998 WL 394992, at *6 ("The most common criticisms of Anderson Labs' methodologies and procedures related to its failure to insure a `blinded' study, its failure to perform necropsies or autopsies on the deceased mice in the studies and irregularities in recording indications of sensory and pulmonary irritation.").
Fifth, certain examples of poor training were adduced at the hearings. Dr. Bernard noted one interviewer made a comment in a file to the effect the participant being interviewed "barely remembers [the fire] or it didn't affect him at all. Who scheduled this person?" Elec. trans. at 184 (emphasis added). The Court agrees this statement demonstrates the interviewer is looking for an effect. Further, no other data was collected on this participant, yet he was still inexplicably included in the study. Dr. Bernard noted, and the Court is inclined to agree, the interviewer was biased and anything he recorded should be suspect.
Other problems with lack of documentation and errors or inconsistencies in the data reporting were evident. For instance, some changes were made to the raw data when participants apparently changed their minds about certain answers. The scientific method for accounting for these changes, however, was not followed.
Sixth, in assessing the participant's believability, Dr. Scotti admitted he was simply asking his interviewers for their "gut response." Elec. trans. at 501. The scientific basis, if any, for this request is undisclosed.
Further examples concerning bias and potential bias, and thus error, could be offered. The instances discussed, however, make the point.
In relation to the law firm's involvement in the study, the Court stresses this case did not merely consist of lawyer sponsorship of litigation research. Perhaps one, or even several, of the issues set out above would, in isolation, be a matter of weight rather than admissibility. One can readily see the distinction here. The depth and breadth of litigation taint is so substantial the very validity of the study is compromised. Given this and other matters of record, the Court deems the error consideration to weigh in favor of exclusion.
By example, a post traumatic stress study of survivors of the Buffalo Creek mining disaster was sponsored by a plaintiffs' law firm and performed by the University of Cincinnati.
4. The Maintenance of Appropriate Standards and Controls Over the Technique's Operation
The Court's greatest concern under this consideration is Dr. Scotti's failure to abide by his own protocol and plans for conducting the study. As stated by Dr. Lees-Haley, "[Y]ou normally set out in your plan your research and follow your ... steps through very carefully, and if you do modify them, you document what you did and why and when." Elec. trans. at 356. There were many deviations from protocol here.
First, Dr. Scotti stated in his report he used a random two mile radius/reverse directory selection process to garner many of the study participants. In fact, however, no such method was used. He simply picked streets around the plant and called persons using the reverse directory. In explaining the error, Dr. Scotti simply said that paragraph of the report was mistakenly included in the rush to complete the study.
Second, Dr. Scotti stated he was going to get 300 participants for the study but stopped at 160. The primary reason proffered for halting participation is everyone had to get back to school to attend to other things.
One of the Court's primary concerns with regard to this deviation was explained by Dr. Bernard. When one identifies a target number for a study population but then stops well short, an inference of "data dredging" can be made. Elec. trans. at 174. Data dredging occurs where an investigator stops a study when a run of data supporting a conclusion appears and a decision is made to not risk it by going further with more data collection. The Court is not stating dredging occurred here, but the otherwise unnecessary deviation from protocol raises the issue.
Third, the mini-SCID instrument employed by Dr. Scotti was not finished. According to Dr. Lees-Haley, Dr. Scotti "didn't analyze it, didn't interpret it." Elec. trans. at 309.
Fourth, Dr. Scotti promised in his report to analyze data by direction and distance. He surmised these "analyses are very likely to have significant impact on the findings presented here regarding being sheltered-in-place." (emphasis in original). Nonetheless, the analyses were never performed.
On these issues, Plaintiffs did very little to address the Court's concerns. Strict controls are important in a scientific study. Those controls, as shown above and elsewhere in the testimony, were either missing or seriously lacking. Accordingly, this factor weighs in favor of exclusion.
5. Whether the Method is Generally Accepted and the Presence of Methods Contrary to Good Science
Dr. Scotti relied heavily on a meta-analysis by Anthony V. Rubonis and Leonard Bickman. Without going into detail, however, it was amply demonstrated throughout the hearing that the article both helped and hurt Dr. Scotti's methodology in a number of areas. In any event, the Court recalls its previous discussion of the concept of peer review and notes Dr. Lees-Haley's comment "A meta analysis does not normally validate the individual studies it looks at." Elec. trans. at 316.
In short, a meta-analysis simply pools all of the data from many studies and treats them as one mega-study.
Anthony V. Rubonis & Leonard Bickman, Psychological Impairment in the Wake of Disaster: The Disaster-Psychopathology Relationship, 109 Psychol. Bull. 384 (1991).
The Court considers a host of varying testimony here, including the appropriate use of the chosen survey measures. As noted above, there are serious concerns about the motives of some of the participants in this study and the unconscious or conscious biases that may have been operating. Dr. Lees-Haley stated where such are present, it is important to use tests that have validity scales in order to judge truthfulness. None of the tests given by Dr. Scotti had such scales.
Second, Dr. Scotti all but conceded his retrospective use of the BDI, the STAI and the IES was not generally accepted. He went so far as to state that retrospective data was some of the "shakier" data one could use. Elec. trans. at 493. Dr. Lees-Haley stated Scotti's use of certain tests in this fashion led to implausible and speculative results. Further, Dr. Lees-Haley stated the BDI in particular is extremely sensitive to response bias. There are further substantial questions in the record concerning Dr. Scotti's use and administration of the measures the Court does not detail here.
As for good scientific practices, several things Dr. Scotti did in his analysis simply did not satisfy "good science." First, Dr. Lees-Haley noted Dr. Scotti never defined what he meant by the term "exposed" as used in his categorization of participants. Awaiting an explanation on rebuttal, the Court was told by Dr. Scotti that (1) the exposure categorization was based on the event interview; and (2) it did not matter to him if the participant was 50 miles away at the time of the fire if they thought they were exposed he treated them as such.
Second, some people who did not answer certain categorization questions were lumped into the "shelter-in-place" category, a category for which Dr. Scotti provides no definition in his report. Third, some of the categories employed by Dr. Scotti overlapped and participants also were placed in more than one category. Dr. Lees-Haley stated it is impossible to do group comparisons when this occurs.
Fourth, there is a sizeable non-random cluster of participants in the study that includes persons (1) from the claimant list; (2) from the Charleston Job Corps Center; (3) responding to the flyer; and (4) friends and family who Dr. Scotti allowed to participate in the study out of "fair[ness]." Elec. trans. at 171.
Fifth, in regard to the friends included in the study, Dr. Scotti conceded the friends who showed up could very well have been friends or family members of the claimants. He does not know how many showed up. Also, he does not know whether these people already had other non-litigation claims with different attorneys pending against Rhone-Poulenc.
Sixth, twenty-two participants hailed from the Job Corps Center in downtown Charleston, approximately 10 or more miles from the site of the fire. Some of the Job Corps Center participants were categorized as exposed and some as not exposed. This is quite perplexing, as both the "exposed" and "unexposed" Job Corps Center participants were in the same place at the same time. The Job Corps office was located about 100 feet away from the offices of Neely & Hunter.
Seventh, one participant was asked whether he was exposed or not and responded on a form by stating "Who Knows." When questioned about how he would categorize this person, Dr. Scotti said he would be conservative and put him in the unexposed group. In fact, however, the person was put in the exposed group.
Given these and other items of record, the Court deems this consideration to weigh in favor of exclusion.
Moving to the "helpfulness" prong of the Daubert inquiry, and perhaps meshing it a bit with the first prong of the analysis, the Court observes "Rule 702's `helpfulness' standard requires a valid scientific connection to the pertinent inquiry ...." Daubert, 509 U.S. at 591-92, 113 S.Ct. 2786. That connection simply is not provided here. Admissibility also runs counter to Rule 403 because the testimony could be unfairly confusing and prejudicial.
Although Daubert liberalized the admissibility equation, there is still an evidentiary line in the sand:
Conjectures that are probably wrong are of little use ... reaching a quick, final, and binding legal judgment — often of great consequence — about a particular set of events in the past. We recognize that in practice, a gatekeeping role for the judge, no matter how flexible, inevitably on occasion will prevent the jury from learning of authentic insights and innovations. That, nevertheless, is the balance that is struck by the Rules of Evidence designed not for the exhaustive search for cosmic understanding but for the particularized resolution of legal disputes.
Daubert, 509 U.S. at 597, 113 S.Ct. 2786; Powers, 59 F.3d at 1472; Dorsey, 45 F.3d at 813-14.
Based on the foregoing, the Court deems each and every consideration to weigh strongly in favor of exclusion. Accordingly, the Court EXCLUDES the research, opinions and testimony of Dr. Scotti from the trial or trials to be conducted in this class action.
In so ruling, again, the Court has considered the justifications offered by Dr. Scotti in support of his methodology. Some of those justifications have been discussed infra and, by example, he further stated at the hearing he was doing "quasi-experimental" research and consequently he should be accorded some "leeway." Elec. trans. at 444. His explanations, however, are wholly insufficient to resurrect the study and opinions.
In making its determination, the Court wishes to stress it is not attacking Dr. Scotti as a scientist generally. No one disputes Dr. Scotti has an impressive vitae. Further, the Court emphasizes that in excluding Dr. Scotti, it is not doing so on the basis of how compelling his conclusions were factually or any other consideration which is properly reserved for the trier of fact. See Daubert, 509 U.S. at 595, 113 S.Ct. 2786; Therm-O-Disc, 137 F.3d at 783 (The burden on Plaintiffs "is not ... `to prove their case twice — they do not have to demonstrate to the judge by a preponderance of the evidence that the assessments of their experts are correct, they only have to demonstrate by a preponderance of evidence that their opinions are reliable.'")(quoting In re Paoli R.R. Yard PCB Litig., 35 F.3d 717, 744 (3d Cir.1994)); Cavallo, 100 F.3d at 1158 ("Daubert governs whether evidence is admitted, not how persuasive it must be to the factfinder."). Rather, the Court is concerned solely with the methodology employed by Dr. Scotti, judged by the considerations set forth in Daubert.
Further, the Court is cognizant that (1) testing the expert on cross examination before the trier of fact; (2) submitting contrary evidence; and (3) careful instructions on the burden of proof "are the traditional and appropriate means of attacking shaky but admissible evidence." Daubert, 509 U.S. at 596, 113 S.Ct. 2786; Cavallo, 100 F.3d at 1158 ("[E]xclusion is the least favored means of rendering questionable scientific evidence ineffective"). Nonetheless, the Court cannot abdicate its role as "gatekeeper" and subject the jury unfairly to confusing and misleading "pseudoscientific" research. Cavallo, 100 F.3d at 1159 ("Although Daubert eliminated the requirement of general acceptance, the five factors it established still require that the methodology and reasoning used by a witness have a significant place in the discourse of experts in the field."). Exclusion is the only appropriate remedy in this case.
C. Proposed Testimony of Dr. Schrager
The Court confesses it remains perplexed as to what Dr. Schrager will testify to or about. Near the end of the hearings, Plaintiffs' counsel acknowledged:
"We have stipulated that Dr. Schrager is not going to testify with regard to any individual causation.... not either individual or group causation."
Elec. trans. at 263. Then, in apparent contradiction:
Dr. Schrager's job is to first talk about the chemical. Now, there is no question about what the chemicals were because Rhone-Poulenc has given a list of them and provided them in discovery, preeminently it is toluene and the other one is MIC, and then there are a number of other chemicals that we have not made any big thing about and there are not very many of them, not very many pounds of them. Toluene and MIC are the major ones. Dr. Schrager is going to talk about what those ... chemicals are. And there is also no debate in this case with regard to how much was released. Rhone-Poulenc has told us that eight tons of toluene, more or less, four and a half pounds of MIC were released, and we are not challenging that. We have no way of challenging it for one thing. So we know that the chemicals were released. We also have photographs of the fire. We have roughly 20 minutes of video tape taken by three television stations from all kinds of different angles from helicopters, from the ground, from across the river, from the other side of the river, from Route 35. There is all of this video tape material that Mr. Emch, I'm sure, has, because I think he provided it to us, but I may confusing that with the FMC case where they had have a video tape from Charlie Ryan before the corpse was dead and actually made us the tape — so there is no question that these tapes exist and they have been seen by everybody. Dr. Schrager has seen the tapes several times. What he is going to explain to the jury is what happens when 8,000 tons — 8 tons or 8,000 pounds of toluene is burned. And then we are going to show the jury the tape and the jury can see the dispersion of this stuff. The jury can see with their own eyes that this stuff was going everywhere but, indeed, there was a predominant wind direction. Then Dr. Schrager is going to talk about the general effects of these chemicals on persons who are exposed to them. Now, the dose response effect comes in because one thing that Dr. Schrager is going to talk about is that different people, as he did today, different people respond to the same dose differently. That's his beer analysis. I mean, you know, a 14 year old girl drinks one beer and finds herself pretty well high as a kite. An experienced Phi Si at the university may ... down ... two six packs ....
Id. at 263-77 (emphasis added). Finally, to obfuscate further, the representation is:
[H]e's not going to opine about causation at all.
Id. at 277.
The Court sought clarification of the substance of Dr. Schrager's testimony via supplemental briefing. The supplemental filing from Plaintiffs was contradictory at best.
On this record, a definitive ruling is impossible. By example, if Dr. Schrager limits his testimony as promised by counsel during argument, a Daubert analysis would likely not be warranted. Most, if not all, of the areas mentioned by counsel involve matters arguably not within the scope of a Daubert inquiry. See, e.g., Talkington, 1998 WL 394593, at *9; Freeman, 118 F.3d at 1016 n .6 (In cases like this one, where an expert relies on his experience and training and not a particular methodology to reach his conclusions, "application of the Daubert [analysis] is unwarranted."). Were Dr. Schrager to venture into some other areas mentioned in the supplemental brief, however, Daubert would be implicated.
In an attempt to narrow this issue for trial, the Court will conditionally permit testimony by Dr. Schrager on the nature and quantity of chemicals released during the fire. Further, Plaintiffs may present the videotape discussed at the Daubert hearing. Dr. Schrager will not, as conceded by counsel, be permitted to testify on any causation-related matters. The Court may permit inquiry into other areas if they become relevant, and otherwise admissible, at trial.
D. Effect of the Court's Rulings
Taking all facts and reasonable inferences in Plaintiffs' favor, the Court over two years ago observed Plaintiffs would be required to submit expert testimony in support of their claims for intentional infliction of emotional distress. McClenathan, 926 F.Supp. at 1279 and n. 10. This ruling was in compliance with Tanner v. Rite Aid of West Virginia, Inc., 194 W.Va. 643, 461 S.E.2d 149 (1995), which permits "A determination by the trial court as to whether a plaintiff has presented sufficient evidence, absent expert testimony, such that the jury from its own experience can evaluate the claim, its causal connection to the defendant's conduct and the damages flowing therefrom." Id.; Travis v. Alcon Labs., Inc., No. 24207, 1998 WL 261007, at *9 (May 21, 1998).
Given (1) the relatively brief nature of the incident; (2) the number of individuals opting out of the class and not filing individual lawsuits even when advised they could recover if they suffered "mental or emotional distress;" and (3) other considerations, the Court deems its original ruling remains appropriate today.
As amply demonstrated by Defendant's supplemental memorandum on the issue filed July 15, 1998 the Supreme Court of Appeals of West Virginia requires the introduction of expert testimony in order to prove a claim for negligent infliction of emotional distress. See, e.g., Stump v. Ashland, Inc., 201 W.Va. 541, 552, 499 S.E.2d 41, 52 (1997). Given the exclusion of Dr. Scotti, the Court is obliged to dismiss this claim as well.
Accordingly, the Court GRANTS Defendant's motion for summary judgment to eliminate Plaintiffs' claims for negligent and intentional infliction of emotional distress.
The Clerk is directed to send a copy of this Memorandum Opinion and Order to counsel of record.