Opinion
No. CR 05-07-M-DWM.
August 31, 2006.
Kris A. McLean, Office of the U.S. Attorney, Missoula, MT, William W. Mercer, Office of the U.S. Attorney, Billings, MT, Thomas L. Sansonetti, David M. Uhlmann, U.S. Department of Justice, Washington, DC, for Plaintiff.
Angelo J. Calfo, Harold Malkin, Michelle K. Peterson, Yarmuth Wilsdon Calfo, Seattle, WA, Michael F. Bailey, Bailey Antenor, William Adam Duerk, Michael J. Milodragovich, Milodragovich Dale Steinbrenner Binney, Missoula, MT, David S. Krakoff, Gary A. Winters, James T. Parkinson, Mark Holscher, Mayer Brown Rowe Maw LLP, Washington, DC, Ronald F. Waterman, Gough Shanahan Johnson Waterman, Palmer A. Hoovestal, Hoovestal Law Firm, Helena, MT, Jeremy Maltby, O'Melveny Myers, Los Angeles, CA, Elizabeth Van Doren Gray, Sowell Gray Stepp Lafitte, Columbia, SC, William A. Coates, Roe Cassidy Coates Price, Greenville, SC, Stephen R. Spivack, Bradley Arant Rose White, Washington, D.C., David E. Roth, Bradley Arant Rose White LLP, Birmingham, AL, for Defendants.
ORDER
I. Introduction
The Defendants have filed a motion in limine to exclude any evidence or expert testimony relating to the medical screening study conducted in Libby by the Agency for Toxic Substances and Disease Registry (ATSDR), the findings of which were published in a peer-reviewed research journal. The Defendants argue that the study is inadmissible under Fed.R.Evid. 702 because it is irrelevant and not scientifically reliable. The Defendants say the study should also be excluded because its probative value is outweighed by the risk of prejudice and jury confusion, and because the data underlying the study has not been properly disclosed to the defense. The Government argues that the study is admissible to prove that Libby asbestos causes asbestos-related lung abnormalities, which in turn proves endangerment. For the reasons that follow, I find that the ATSDR screening program's findings are inadmissible under Fed.R.Evid. 702 and 403.
Counts II, III and IV of the Superseding Indictment, alleging violations of the Clean Air Act's knowing endangerment provision, 42 U.S.C. § 7413(c)(5)(A), require the Government to prove as one of the elements of the offense that the Defendants's release of a hazardous air pollutant placed another person in "imminent danger of death or serious bodily injury."
The Government filed a second Notice of Appeal in this case on August 23, 2006 (Doc. No. 730), resulting in a partial divestiture of this Court's jurisdiction. The Court retains jurisdiction to decide the issues presented in this motion, however. See City of L.A. v. Santa Monica Baykeeper, 254 F.3d 882, 886 (9th Cir. 2001).
II. Background
The Agency for Toxic Substances and Disease Registry (ATSDR), in conjunction with other state and federal agencies, conducted a program for medical screening of Libby residents during 2000 and 2001. Participants were interviewed in person and tested in Libby in two waves: from July through November of 2000 and from July through September of 2001. A total of 7,307 current and former Libby residents participated in the medical testing program.The study's eligibility rules required that participants have lived, worked, attended school or participated in other activities in Libby for at least six months prior to December 31, 1990. Potential participants were found using the Libby telephone directory, print and media advertising in local and major nationwide markets, and word of mouth and medical referrals. Telephone screening was used as the initial means of determining eligibility. Participants in the study were interviewed using questions from a computer-assisted questionnaire to determine whether they had any historical "pathways to exposure" to asbestos contaminated vermiculite. The interviewers asked questions about residential history, occupational history, household contact history, recreational activities and other potential pathways for vermiculite exposures, cigarette smoking status, medical history, and self-reported symptoms and illnesses. The medical testing consisted of spirometric testing and chest radiographs (x-rays).
Spirometric testing measures lung function and identifies abnormal ventilation in the form of restrictive or obstructive patterns. Patients with asbestos-related lung problems typically exhibit restrictive abnormalities on spirometric testing. ATSDR Preliminary Findings dated February 22, 2001 (ATSDR Report), Ex. 2 to Govt's Resp. Br. at p. 9.
Chest radiographs were not offered to participants under 18 years of age. As a result, only 6,668 of the 7,307 participants received the full range of tests offered by the study.
The x-ray testing included posterior-anterior, right anterior-oblique and left anterior-oblique views. An onsite physician reviewed the x-rays for consistency and quality and completed a referral form if the findings suggested a need for urgent medical attention. After initial evaluation all x-rays were independently read and classified by three B-readers certified by the National Institute for Occupational Safety and Health (NIOSH). Two primary readers reviewed all x-rays and a third reader independently viewed each x-ray for which the primary readers had different interpretations. A subject was classified as having a pleural abnormality if two of the three B-readers identified (a) any unilateral or bilateral pleural calcification on the diaphragm, chest wall, or other site or (b) any unilateral or bilateral pleural thickening or plaque on the chest wall, diaphragm, or costophrenic angle site, consistent with asbestos-related pleural disease, using the posterior-anterior view, the oblique views, or a combination of those views.
The study sought to identify potential exposures to asbestos or vermiculite by asking participants questions about their work history, most notably whether they had ever worked as an employee or contractor for Grace or its predecessors. Other questions sought to determine whether a participant had been a household contact of a Grace worker, a longtime Libby resident, or a user of vermiculite in gardening or insulation. There were also questions about recreational activities, such as playing in vermiculite piles, popping vermiculite, playing on the baseball field near the expansion plant or playing along the road to the Libby mine. Participants were asked to characterize their frequency of these recreational activities or any other contact with vermiculite as "never," "sometimes," or "frequently." All of the questions sought information about exposures prior to December 31, 1990. The study also examined certain demographic variables such as age, sex, smoking status, weight, history of pulmonary disease, lung cancer, and other self-reported conditions.
ATSDR published an initial report of the screening study's findings on February 22, 2001 (ATSDR Report). The ATSDR Report was drawn from the first wave of participants and featured a data set comprised of 1,078 participants. The preliminary findings reported pleural abnormalities among 19% of the participants. The highest rate of abnormality (37%) was observed among participants who identified themselves as former W.R. Grace workers. Household contacts of Grace workers had the next highest rate (20%), followed by those with occupational contact with vermiculite unrelated to Grace (18%). The proportion of pleural abnormalities in other exposure groups were 13% for those with recreational contact to vermiculite, 14% for those who had vermiculite insulation in their homes, and 14% for those with no apparent exposure.
The complete results of the screening study were published in an article entitled "Radiographic Abnormalities and Exposure to Asbestos Contaminated Vermiculite in the Community of Libby, Montana, USA" in the November 2003 volume of the peer-reviewed research journal Environmental Health Perspectives (the Peipins Study)., The authors of the Peipins Study did no independent testing; their findings are based entirely on the data gathered by ATSDR using the methods described above. The first page of the Peipins Study states, "The medical testing program in Libby, Montana, was supported by funds from the Comprehensive Environmental Response, Compensation and Liability Act (CERCLA) trust fund through the Agency for Toxic Substances and Disease Registry, Public Health Service, U.S. Department of Health and Human Services." Peipins Study at 1753.
Attached as Exhibit 1 to the Government's Response (Doc. No. 561).
The Peipins Study does not contain any findings relating to the spirometric testing.
The Peipins Study identifies 29 occupational, recreational, household and other exposure pathways. The study determined that the incidence of pleural abnormality increased with an increasing number of exposure pathways, with abnormalities among 6.7% of those reporting no apparent exposure pathways to 34.6% for those reporting 12 or more pathways. The study found pleural abnormalities among 17.8% of all participants who received x-rays. The factors most strongly related to the presence of pleural abnormalities were being a former Grace worker, being older, being a household contact of a Grace worker, and being male. Other factors associated with pleural abnormalities included having been exposed to asbestos in the military, having played in vermiculite piles, having lived in Libby for a longer period, having been a smoker, and having a high body mass index.
Both the ATSDR Report and the Peipins Study discuss the limitations of the analysis based on data gathered from the medical screening program. The main objectives of the screening program "were to a) identify and quantify possible asbestos-related pleural and interstitial abnormalities among participants and b) examine associations between these outcomes and participants' exposure histories." Peipins Study at 1754. Other important goals of the program were to
(a) provide EPA with information needed to identify and eliminate current exposures to asbestos in the community;
(b) identify the types of illnesses experienced by participants exposed to asbestos in order to better inform local physicians; and
(c) provide the local health care community with an estimate of the additional resources necessary to address health care needs in the Libby area during the next 10-20 years.
ATSDR Report at p. 1.
None of the stated goals of the study relates to establishing a causal relationship between exposure to asbestos in Libby and the development of asbestos-related pleural abnormalities. In response to public comments on ATSDR's draft plan for the screening study, the ATSDR stated, "[T]he purpose of this information is to characterize possible pathways of exposure in the community that EPA needs to evaluate. It is not being collected to attribute causality to asbestos contaminated vermiculite on an individual nor population basis. The medical testing program is not a study." ATSDR Summary of Public Comments, May 19, 2000. The ATSDR Report states, "This testing program was not an epidemiologic study," ATSDR Report at p. 21, and "because this was not an epidemiologic study, no control group was included." Id. "Because no control group was included, direct comparisons of the occurrence of abnormalities above the expected value can not be calculated." Id. While conceding that no control group was included, the authors of the Peipins Study state that "[g]iven the ubiquitous nature of vermiculite contamination in Libby, along with historical evidence of elevated asbestos concentrations in the air," it would be difficult to locate participants to make up a control group, i.e., Libby residents who could be characterized as unexposed to asbestos-contaminated vermiculite. Peipins Study at 1758.
The Defendants' Opening Brief in support of their motion (Doc. No. 500) states that the Summary of Public Comments is attached to the Brief as Exhibit 4. In fact, the record reflects no attachments to Document No. 500. There are no attachments to the entry in the electronic docket, nor are there hard copies on file in the clerk's office. The Missoula filing office has confirmed that no such attachments exist in the record. However, as a courtesy the Defendants supplied attachments to Document No. 500 along with attachments to all other expert-related motions in limine filed by the defense. These courtesy attachments to Document No. 500 are mis-numbered and incomplete, but the Summary of Public Comments is included as Exhibit 4. Despite the Defendants' apparent failure to properly file the attachments to their opening brief, the document in question is a part of the record as Defendant's Exhibit 89 received into evidence during oral argument on July 20, 2006.
Other limitations discussed include the cross-section design of the testing program; the fact that participants self-selected rather than being randomly drawn; the fact that the B-readers were not truly blinded in that they were aware that all of the x-rays were from the Libby screening program; the fact that the x-ray review included no control films that might expose observer bias; and the likelihood that there were some pre-1991 exposure pathways in Libby that were not identified by those performing the study and did not appear in the questionnaire.
III. Analysis
A. Legal Standard
Federal Rule of Evidence 702 provides:
If scientific, technical, or other specialized knowledge will assist the trier of fact to understand the evidence or to determine a fact in issue, a witness qualified as an expert by knowledge, skill, experience, training, or education, may testify thereto in the form of an opinion or otherwise, if (1) the testimony is based upon sufficient facts or data, (2) the testimony is the product of reliable principles and methods, and (3) the witness has applied the principles and methods reliably to the facts of the case.
Rulings on the admissibility of expert testimony under Rule 702 are committed to the sound discretion of the trial court. General Elec. Co. v. Joiner, 522 U.S. 136, 141-42 (1997). InDaubert v. Merrell Dow Pharmaceuticals, Inc., the Supreme Court imposed a gatekeeping obligation on trial judges to engage in objective screening designed to ensure that scientific evidence that is put before the jury "is not only relevant, but reliable." 509 U.S. 579, 589 (1993).
The threshold questions for admissibility of expert scientific testimony are whether the proffered testimony reflects scientific knowledge and whether it will assist the trier of fact. Id. at 592, 113 S.Ct. 2786. The Daubert Court instructed that in exercising the "gatekeeper" function of deciding whether to admit such testimony, a trial court must make "a preliminary assessment of whether the reasoning or methodology underlying the testimony is scientifically valid and of whether that reasoning or methodology properly can be applied to the facts in issue." Id. at 592-593. An expert's testimony must be grounded in the methods and procedures of science, and must be more than unsupported speculation or subjective belief.Id. at 593. An expert's "bald assurance of validity is not enough." Daubert v. Merrell Dow Pharaceuticals, Inc., 43 F.3d 1311, 1316 (9th Cir. 1995) ("Daubert II"). A proponent of the testimony does not have the burden of proving that the testimony is scientifically correct, but rather that the testimony, by a preponderance of the evidence, is reliable. The "`focus, of course, must be solely on principles and methodology, not on the conclusions that they generate.'" Joiner, 522 U.S. at 146 (quoting Daubert, 509 U.S. at 595).
The Supreme Court in Daubert listed four nonexclusive factors that a trial court might, in the exercise of its discretion, consider to assess reliability: 1) whether a theory or technique can be tested; 2) whether it has been subjected to peer review and publication; 3) the known or potential error rate of the theory or technique; and 4) whether the theory or technique enjoys general acceptance within the relevant scientific community. Id. at 593-94.
The Supreme Court revisited admissibility of expert testimony in Kumho Tire Co. Ltd. v. Carmichael, and emphasized that trial judges are entitled to broad discretion when discharging their gatekeeping function, and should not mechanically apply theDaubert factors to testimony, scientific or otherwise, as if the factors were a "definite checklist or test." 526 U.S. 137, 150-51 (1999). Rather, the gatekeeping inquiry must be tailored to the facts of each case, and a trial judge should apply the Daubert factors "where they are reasonable measures of the reliability of expert testimony." Id. at 150, 152. Not only must the trial court be given broad discretion to decide whether an expert's testimony is reliable, it "must have the same kind of latitude in deciding how to test an expert's reliability." Id. at 152. Regardless of the means the court chooses to employ in assessing an expert's testimony, the court must satisfy itself that the expert will adhere to the same standards of "intellectual rigor" in testifying as he would follow when dealing with the same subject matter outside of the courtroom. Id.
The 2000 Amendments to the Rule 702 added the requirement that expert testimony be based on "sufficient facts or data." Fed.R.Evid. 702. The amendment requires an analysis of the sufficiency of underlying facts or data that is quantitative rather than qualitative. See Advisory Committee Notes to the 2000 Amendments to Rule 702. The sufficiency requirement is set forth in a separate subsection but cannot be divorced from the ultimate inquiry into the reliability of the expert's testimony. Id.
The relevance prong of the Daubert test requires a trial court to determine whether the expert testimony is sufficiently tied to the facts of the case to be of use to the trier of fact in resolving a factual dispute. 509 U.S. at 591, 113 S.Ct. 2786. A court must analyze whether it is proper for the jury to apply an expert's reasoning or methodology to the facts at issue. Id. at 591-93, 113 S.Ct. 2786. The evidence must have a valid scientific connection to the disputed material facts in a case. Id. at 591. This ensures a "fit" between the testimony and the issues to be resolved at trial.
Even though an expert may be qualified in one area of expertise, he still may be precluded from offering opinions beyond that area of expertise, or that are not founded on a reliable methodology. Kumho Tire, 526 U.S. at 154-55. "Scientific validity for one purpose is not necessarily scientific validity for other, unrelated purposes." Daubert, 509 U.S. at 591. The trial court's gatekeeper role, however, is not meant to supplant the adversary system or the role of the jury: "[v]igorous cross-examination, presentation of contrary evidence, and careful instruction on the burden of proof are the traditional and appropriate means of attacking shaky but admissible evidence."Daubert, 509 U.S. at 596; see also, Advisory Committee Notes to the 2000 Amendments to Rule 702 ("[T]his amendment is not intended to provide an excuse for an automatic challenge to the testimony of every expert.").
B. Discussion
a. Reliability
The Defendants argue that the ATSDR Report and the Peipins Study must be excluded because the methodology underlying the medical screening program is unreliable. The screening program is unreliable for purposes of determining a causal relationship between exposures to Libby amphibole and the development of asbestos-related disease, the Defendants argue, because the screening program is not a properly constituted epidemiological study. The Defendants contend the screening program is also unreliable for any other purpose because the x-ray readers failed to observe the generally accepted methods for interpreting chest radiographs.
Epidemiology is the field of public health and medicine that seeks to understand disease causation and to prevent disease in groups of individuals. "Epidemiologists employ cohort studies, case-control studies, and ecological studies to determine whether individuals exposed to an agent have a greater risk of developing the disease in question." Siharath v. Sandoz Pharm. Corp., 131 F. Supp.2d 1347, 1356 (N.D.Ga. 2001). Such studies frequently provide the basis for expert opinions relating to causation in legal proceedings. Id. ("epidemiological studies provide the primary generally accepted methodology for demonstrating a causal relation between a chemical compound and a set of symptoms or disease") (internal quotation omitted); 57A Am. Jur. 2d Negligence § 508 (2004) ("Epidemiological evidence is indispensable in carcinogenic tort actions where direct proof of causation is lacking.").
The Defendants argue that the ATSDR Report and the Peipins Study are unreliable because the screening program is not a valid epidemiological study, meaning it fails to identify a causal relationship between an agent (in this case Libby amphibole) and the development of a disease (here, asbestos-related disease). The Defendants point out that the screening program is not a study by the ATSDR's own admission, and that the program lacks two fundamental features of an epidemioloical study: random sampling and a control group.
The Government's expert disclosures do not evince a clear intent on the part of any of the experts to rely upon the screening program to render an opinion on causation. The relevant expert disclosures generally speak of a "relationship" or "association," rather than causation, when discussing the screening program. See Supplemental Disclosure of Dr. James Lockey at p. 16 ("These results are consistent with what would be expected if there was a relationship between environmental exposure to asbestos and chest radiographic changes consistent with this exposure."), p. 17 (commenting on "[t]he findings of an association between both occupational and non-occupational exposure to vermiculite and pleural changes"); Supplemental Disclosure of Paul Peronard at p. 14 (discussing the relationship between exposure to vermiculite and lung abnormalities, but not explicitly discussing causation). Still, given the likelihood that jurors will fail to appreciate the scientific and statistical distinction between association and causation, it is necessary to determine whether the ATSDR Report and Peipins Study are admissible to show causation.
The Government accuses the Defendants of ignoring the distinction "between the ATSDR medical screening program and the ATSDR epidemiological study." Govt.'s Resp. Br. at p. 2. The ATSDR epidemiological study to which the Government refers is the Peipins Study, which is based solely on the ATSDR screening program. The Government has failed to identify any distinction between the program and the Peipins Study, other than that the Peipins Study was published in a peer-reviewed journal. It is noteworthy that despite the Government's characterization, the Peipins Study does not self-identify as an epidemiological study, nor does it discuss its findings in terms of causation.
Participants self-selected for the screening program, and the authors of the Peipins Study acknowledge that "studies involving volunteers are subject to selection bias." Peipins Study at p. 1758. Those who have experienced symptoms or been previously diagnosed may have been more likely to take part, while those who believe they have no exposure or have not experienced symptoms may have chosen not to participate. Id. Selection bias may also have occurred as a result of the requirement that those not then living in Libby travel to the town to participate. Id. A random sampling of the population is necessary to allow the results among participants in a study to be extrapolated to predict results in a broader population. The Government does not counter the Defendants' point about possible selection bias and concedes that the Peipins Study acknowledges the potential problems with the screening program's sample. And while "[t]he fact that a study may suffer from selection bias does not itself invalidate its results," Green, et al., "Reference Guide on Epidemiology," Reference Manual on Scientific Evidence, 365 (2d ed., 2000), the absence of a random sample undermines the reliability of the screening program as an epidemiological study.
Failure to include a control group denies the authors of a study the opportunity to compare the rate of disease among the exposed population to the rate of disease among those not exposed. The Government attempts to minimize the absence of a control group here by claiming that "the medical screening program and the Peipins Study constitute a generally accepted epidemiological `cross sectional study' for which a control group is unnecessary." Govt.'s Resp. Br. at 8. The Government offers no support for this proposition, although the authors of Peipins Study refer to their work as having a "cross-sectional design." Peipins Study at 1758. However, as the Defendants point out, the Government's argument that the cross-sectional nature of the screening program qualifies it as an epidemiological analysis fails to advance the Government's position, because cross-sectional studies are the least desirable epidemiological studies when it comes to reaching reliable conclusions about causation.
The cross-sectional design is identified by the authors as one of the "principal limitations" of the analysis offered in the Peipins Study.
Cross-sectional studies determine the presence (prevalence) of both exposure and disease in the subjects and do not determine the development of disease or risk of disease (incidence). moreover, since both exposure and disease are determined in an individual at the same point in time, it is not possible to establish the temporal relation between exposure and disease — that is, that the exposure preceded the disease, which would be necessary for drawing any casual inference.
Green, et al., Reference Manual on Scientific Evidence at 343.
Although cross-sectional data are often obtained in surveys and can be very useful, they often do not permit the investigator to determine the temporal relationship between exposure and the development of disease. As a result, their value for deriving causal inferences is limited. However, they can provide important directions for further research using cohort, case-control, and nested case-control designs.
Gordis, L., Epidemiology, Third Ed., 202.
The Gordis text is cited in the Defendants' Reply Brief at p. 10 but the Defendants have not provided an excerpt as an attachment. However, the text is part of the record because the excerpt quoted above was presented at oral argument on July 20, 2006 and received into evidence as Defendant's Exhibit 64.
Moreover, the Government's explanation for the absence of a control group is at odds with the reason given by ATSDR as to why no control group was included. While the Government contends that no control group was used because the screening program is a cross-sectional epidemiological study for which no control group is needed, the ATSDR Report states, "[B]ecause this was not an epidemiologic study, no control group was included." ATSDR Report at p. 21.
The Defendants anticipate that the Government's witnesses might attempt to derive meaning from the screening program's findings by comparing the results to control groups from other studies, a tactic the Defendants argue should not be allowed. The Government has not announced an intention to have its experts so testify, but the Peipins Report contains some references to control groups from other studies to determine a normal rate of incidence for an unexposed population. The Defendants have not identified any specific principle of epidemiology that would exclude testimony comparing an exposed group to a control group from another study, but it is clear that the addition of a control group comparison after the screening program is completed is not ideal, and that such a practice is insufficient to transform the screening program into an epidemiological study.
The failure to include a control group and to secure a random sample of the population are shortcomings, but do not compel exclusion of the screening program if its findings appear reliable in spite of those deficiencies. But the real concern with the medical screening program is the fact that those who conceived and executed the program disavowed any goal of establishing a causal relationship between exposure to Libby amphibole and the incidence of asbestos-related disease. ATSDR Report at p. 1, Peipins Study at p. 1754 (listing goals of the screening program and failing to mention goal of identifying causal link); ATSDR Summary of Public Comments, May 19, 2000 ("[Information] is not being collected to attribute causality to asbestos contaminated vermiculite on an individual nor population basis. The medical testing program is not a study."). None of the Government's proposed experts have disclosed an intent to defend the screening program as an epidemiological study of any kind. See Supplemental disclosure of Dr. James Lockey at p. 13 ("[T]he program was not designed to be an analytic epidemiology study, but more of a medical testing or screening program.").
The prosecution notes that the screening program was not prepared for use in litigation, and that this fact should weigh in favor of admissibility, citing Daubert II, 43 F.3d at 1317, in which the court wrote, "One very significant fact to be considered is 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 purposes of testifying." The proposition works against the Government in this case. There are good reasons to lend greater credence to the conclusions of scientists operating outside of a legal proceeding as they are more likely to be candid in their findings and analysis and not tempted to slant the results to serve a legal theory or litigation position. In this case, when the scientists conducted the screening program independent of litigation they were very frank about the limitations of the screening program and the fact that it is not an epidemiological study designed to identify causation. Now that the screening program's findings have been placed at issue in this litigation, the Government would like to explain away those limitations.
The ATSDR medical screening program and the documents reporting the program's findings do not establish a causal link between exposure to Libby amphibole and the development of asbestos-related disease. Fundamental problems with the design of the program were acknowledged by its creators, and it pursued goals that, while useful for other purposes, did not lead to any information about causation. These problems are exacerbated by the failure to secure random sample for study and to include a control group. For these reasons, the Court must conclude that the screening program's methodology is not reliable to reach determinations about causation, and that opinion testimony based on the program's findings is inadmissible as evidence of causal relationship under Fed.R.Evid. 702 and Daubert.
The Defendants make additional arguments about reliability relating to the procedures used by the x-ray readers. Specifically, the Defendants complain that (1) the readers were not truly blinded because they knew all of the x-rays were taken from Libby residents, (2) the set of x-rays reviewed did not include control films, and (3) the readers were allowed to review oblique as well as posterior-anterior views. The procedures identified by the Defendants constitute departures of varying significance from the International Labour Office (ILO) Guidelines for classifying x-rays (2000). The most problematic appears to be the decision to review oblique views as well as posterior-anterior views, as this likely resulted in a higher rate of positives than would have been found using only the posterior-anterior view as recommended by the ILO Guidelines. Peipins Study at 1754. These problems further undermine the reliability of the medical screening program, although they probably would not by themselves require exclusion.
b. Relevance
The Defendants argue that any testimony related to the ATSDR screening program should be excluded as irrelevant because the program's findings do not fit the issues and will not assist the trier of fact in deciding the case. The screening program does not fit, the Defendants argue, because it contains no information about the pre-1991 exposures reported by participants. In the absence of any information about the duration and intensity of the pre-1991 exposures, the Defendants argue that the Government cannot show that those pre-1991 releases and exposures are similar to any exposures occurring after November 3, 1999, and therefore evidence of the pre-1991 exposures is not relevant to whether the Defendants caused releases resulting in endangerment as alleged in Counts II, III and IV of the Superseding Indictment. The Defendants contend the program's findings will not assist the trier of fact because they do not establish causal link between exposure to Libby amphibole and development of asbestos-related disease.
The Court has previously determined that any violations of the Clean Air Act's knowing endangerment provision, 42 U.S.C. § 7413(c)(5)(A), that became complete before November 3, 1999 are time-barred under the statute of limitations. See United States v. Grace, 427 F.Supp.2d 1207, 1239 n. 30 (D.Mont. 2006).
The risk of developing an asbestos-related disease increases with exposures of greater duration and intensity. "In general the longer a person is exposed to asbestos and the greater the intensity of the exposure, the greater the likelihood for asbestos-related health problems." Supplemental Disclosure of Dr. Aubrey Miller at p. 3. This means that the quality of the Government's proof regarding endangerment will depend on the extent to which the Government can introduce evidence of post-November 3, 1999 exposures that are of sufficient duration and intensity to place another person in imminent danger of death or serious bodily injury. The Government may attempt to make this showing by first introducing scientific evidence to establish those levels of duration and intensity beyond which exposure to Libby amphibole causes imminent danger. The ATSDR Report and Peipins Study can not supply that evidence because they contain no measure of the duration and intensity of the pre-1991 exposures reported by the program's participants. There is nothing to link the findings of the screening program to the timely exposures the Government intends to prove at trial, other than perhaps a shared pathway of exposure. But the danger of a particular pathway depends on the concentration of asbestos associated with that pathway and the duration of the exposure, variables that can be assumed to change over time. Evidence that a person was exposed after November 3, 1999 to a pathway identified by the screening program is not evidence that the exposure was of the same intensity and duration as those pre-1991 exposures reported by participants in the program.
The Government argues that the ATSDR Report and Peipins Study are nonetheless relevant to the issues in the case, and helpful to the trier of fact, because of their role in a logical syllogism that underlies the Government's theory. The Government states:
The relevance of the medical screening program, the Peipins Study, and related expert testimony is illustrated clearly in the following logical syllogism:
1. Asbestos exposure causes pleural abnormalities (established by epidemiologic studies during the last 50 years and undisputed by the defendants).
2. The medical screening program and Peipins Study found asbestos related pleural abnormalities in Libby residents.
3. Therefore, exposure to Libby asbestos causes pleural abnormalities.
Govt.'s Resp. Br. at p. 10.
The second proposition of the Government's syllogism is superfluous. If asbestos exposure causes pleural abnormalities, then it follows logically that exposure to Libby asbestos causes pleural abnormalities. The ATSDR screening program and its findings are not necessary to make the Government's logical point about causation. The fact that the medical screening program is irrelevant to the causation question is unsurprising because the screening program yielded no findings relating to causation. The Government concedes as much when it states, "The medical screening program, Peipens [sic] Study, and proffered expert testimony of an association or correlation between asbestos related disease and certain exposure pathways in Libby are pieces of the government's proof of endangerment." Id. (emphasis added). It is not enough for the Government to make a conclusory statement that the evidence goes to endangerment. The Court must ask what fact relating to endangerment is made more or less likely by this proof. An opinion on causation assists the jury, because the jury must decide whether the charged releases placed another person in imminent danger of death or serious bodily injury, i.e., whether the exposures were likely to cause asbestos-related disease. The ATSDR Report and the Peipins Study do not contain opinions as to causation, but instead identify a correlation or association. Moreover, the correlation identified by the screening program is not a correlation between the charged releases and imminent danger; it is a correlation between decade-old exposure pathways and pleural abnormalities that may be asbestos-related disease but have not been diagnosed as such. Evidence of this correlation does not assist the trier of fact.
The Government states that it "intends to offer additional evidence in support of the logical inference that exposure to Libby asbestos occurred and endangered people during the relevant time period." Govt.'s Resp. Br. at p. 10. The Government's intent to present additional unidentified evidence to prove endangerment has no impact on the relevance of the ATSDR Report and the Peipins Study.
The trouble with the screening study is that it tracks exposure pathways without measuring the exposures. Quantified measurements of asbestos exposure are reliable over time. If the Government has evidence from a study showing that asbestos exposure at X intensity for a duration of Y caused asbestos-related disease in 50 percent of those exposed, it can be assumed that the exposure of a different group of people ten years later to the same mineral at the same intensity for the same duration will cause asbestos-related disease in roughly half of those exposed. But if instead the Government has evidence that a certain activity in Libby caused asbestos-related disease among a certain percentage of residents who performed the activity, with no uniformity as to the duration or specific nature of the activity among those tested, the evidence is useless to estimate the risk to individuals performing the same activity ten years later. Exposure pathways don't make people sick; asbestos makes people sick.
The ATSDR Report and the Peipins Study allow for certain tentative conclusions, but they are conclusions about what the screening program measured, i.e., what exposure pathways were most dangerous prior to 1991. It can be presumed that the ATSDR Report and Peipins Study reliably identify the sites in Libby that had the greatest historical concentration of asbestos contamination prior to 1991. But this conclusion is irrelevant to the question of the danger posed by the conditions in Libby after November 3, 1999, and will not assist the jury in deciding whether the charged releases resulted in imminent danger of death or serious bodily injury. Because the evidence is not relevant and does not assist the trier of fact, testimony relating to the ATSDR Report and the Peipins Study is excluded under Rule 702. See Daubert II, 43 F.3d at 1121 n. 17 ("Federal judges must therefore exclude proffered scientific evidence under Rules 702 and 403 unless they are convinced that it speaks clearly and directly to an issue in dispute in the case, and that it will not mislead the jury.").
Even this conclusion is subject to some doubt because areas of relatively low concentration may be associated with an abnormally high rate of disease if the nature of a participant's presence in the area generally required exposures of above-average duration.
Assuming the ATSDR Report and the Peipins Study were otherwise relevant, the Defendants argue they should still be excluded under Rule 403 because of the potential for jury confusion. Rule 403 allows for the exclusion of evidence "if its probative value is substantially outweighed by the danger of unfair prejudice, confusion of the issues, or misleading the jury. . . ." For purposes of the rule, unfair prejudice is "an undue tendency to suggest decision on an improper basis." Advisory Committee Notes to Rule 403. The Supreme Court has held that Rule 403 allows a district court to guard against the introduction of relevant evidence that might "lure the factfinder into declaring guilt on a ground different from proof specific to the offense charged."Old Chief v. United States, 519 U.S. 172, 180 (1997).
The risk in this case if the jury is allowed to hear evidence relating to the findings of the medical screening program is that the jury might fail to appreciate the scientific and statistical distinction between association and causation, leading them to lend undue weight to the findings of the program. The Government argues that there is no risk of confusion or misleading the jury because the Peipins Study describes exposure pathways in detail. This fact does nothing to eliminate the likelihood that jurors will erroneously conclude from the program's findings that "exposure pathways" cause asbestos-related disease.
The ATSDR Report and Peipins Study have low probative value because they do not speak to causation. They are of no use to the jury as valid scientific studies, but they are suggestive of a causal relationship that they cannot support. Lay people do not generally concern themselves with the difference between causation and association; one is no more or less persuasive than the other. It is true that the Defendants could attempt to parse out the distinction on cross-examination, but the chances are good that a juror will ignore such efforts and conclude that intuitively appealing proof of an association is good enough. In that case, the jury will have been misled, and the Defendants prejudiced because of the difficulty of curing the problem. Rule 403 requires the Court to guard against these risks where, as in this case, the likelihood that they will materialize far outweighs the minimal probative value of the evidence.
The Government declares in Footnote 2 of its Response Brief that the ATSDR Report and Peipins Study are relevant to every count in the Superseding Indictment, but the Government has not attempted to explain why the evidence is relevant to any charges other than the Clean Air Act knowing endangerment counts. This bench memo analyzes the probative value of the ATSDR Report and Peipins Study only with regard to the Clean Air Act counts.
3. Fed. Rule Crim. P. 16
The Defendants want the ATSDR Report and the Peipins Study excluded as a sanction for the Government's failure to provide access to x-rays used in the ATSDR screening program and now located in the files of Libby's CARD clinic. The Court has previously required the Government to produce the medical records underlying the medical screening program, which the Government had represented were in the possession of the Department of Health and Human Services (DHHS). See United States v. Grace, 401 F.Supp.2d 1093, 1099-1100 (D.Mont. 2005). The Government now states that the x-rays taken during the screening program were given to the participants, many of whom have in turn given them to the doctors at Libby's CARD clinic.
Upon learning that some ATSDR x-rays were in the CARD clinic's files, the Defendants requested that they all be produced. The Government has made available those ATSDR x-rays contained in the 550 patient database to be relied upon by Government witness Dr. Alan C. Whitehouse in his testimony, the contents of which have been disclosed pursuant to this Court's orders. See Government's Medical Records Certification dated May 15, 2006 (Doc. No. 449). The Government has refused to disclose any ATSDR x-rays located elsewhere in the CARD clinic's files. According to the Government, the requested production would require CARD employees to physically search each file for ATSDR x-rays, something the clinic has refused to do. See June 1, 2006 Letter from Kris McLean to Billy Jacobson, Ex. 16 to Defs' Reply Br., in which Mr. McLean writes, "The CARD clinic advises that it primary obligation to provide medical care makes a hand search of every medical file extremely onerous, time consuming and nearly impossible. Therefore, the CARD clinic advises it will not perform such a search."
The Defendants ask that the ATSDR Report and Peipins Study be excluded as a sanction for failure to disclose all x-rays, presumably under Rule 16(d)(2)(C). Such a sanction is unnecessary in light of the fact that separate grounds exist for the exclusion of the evidence as described above. Moreover, the deadline for discovery motions has long since passed, and the x-rays sought have never been in the Government's possession. It is mere happenstance that some of the x-rays have landed in the files of a Government expert, and there is no indication that Dr. Whitehouse or anyone else possesses a complete set of ATSDR x-rays. Rule 16 does not provide a separate basis for exclusion of the ATSDR Report and the Peipins Study.
IV. Order
The ATSDR screening program was not designed to establish a causal link between exposure to Libby amphibole and incidence of asbestos-related disease. Because it was designed for another purpose, the program's findings, as expressed in the ATSDR Report and the Peipins Study, are not reliable to establish causation (endangerment) and are irrelevant to the issues in the case. Moreover, the limited probative value of the program's findings is outweighed by the risk that they will mislead and confuse the jury.
Accordingly, the Defendants' motion to exclude expert evidence relating to the ATSDR medical screening program (Doc. No. 500) is GRANTED and the ATSDR Report and Peipins Study should are excluded for any purpose relating to the Clean Air Act knowing endangerment counts under Fed.R.Evid. 702 and 403.