Magistrini v. One Hour Martinizing Dry Cleaning, 180 F. Supp. 2d 584 (D.N.J. 2002)

U.S. District Court for the District of New Jersey - 180 F. Supp. 2d 584 (D.N.J. 2002)
January 4, 2002

180 F. Supp. 2d 584 (2002)

Kathy MAGISTRINI, Plaintiff,

No. CIV.A.96-4991.

United States District Court, D. New Jersey.

January 4, 2002.

*585 *586 *587 *588 Esther Berezofsky, Williams, Cuker & Berezofsky, Cherry Hill, NJ, for Plaintiff Kathy Magistrini.

Richard J. Shackleton, Shackleton, Hazeltine & Bishop, Ship Bottom, NJ, for Defendant Dow Chemical Co.

Helen Witt, Kirkland & Ellis, Chicago, IL, Joseph F. Lagrotteria, St. John & Wayne, Newark, NJ, for Defendant R.R. Street & Co., Inc.

Larry S. Reich, Tenzer & Greenblatt, LLP, New York, NY, for Defendants One Hour Martinizing Dry Cleaning and Martin Franchises, Inc.


HOCHBERG, District Judge.

This matter comes before the Court following a Daubert[1] hearing to determine the admissibility of the expert testimony on both sides of this complex products liability action. Defendants move to exclude the testimony of Plaintiff's experts: Dr. David Ozonoff is proffered to opine on the issue of medical causation and Mr. George Stanton, an industrial hygienist, seeks to testify about chemical exposure *589 and the adequacy of Defendants' warnings. Plaintiff seeks to exclude the testimony of Defendants' medical causation expert, Dr. James Jandl.[2]


Kathy Magistrini (hereinafter "Magistrini" or "Plaintiff") worked for Defendant One Hour Martinizing Dry-Cleaning for approximately a two-year period, from September, 1977, through October, 1979. While working there, Plaintiff was exposed to the dry-cleaning fluid identified as perchloroethylene ("PCE").[3] Plaintiff's exposure ceased when she left One Hour's employ. In January, 1981, Plaintiff was diagnosed as having acute myelomonocytic leukemia ("AMML"), a subtype of acute myelogenous leukemia ("AML"). As a result of this diagnosis, Plaintiff underwent a bone marrow transplant and total body radiation treatment and has suffered from infections, infertility, loss of body hair and nails, mouth sores, throat damage and severe weight loss.

In August, 1994, Plaintiff's husband drew her attention to a television broadcast which discussed an alleged relationship between PCE and leukemia. On August 8, 1996, Plaintiff filed suit against Defendants One Hour Martinizing Dry Cleaning, Martin Franchises, Inc., Dow Chemical Company and R.R. Street & Co., Inc. The Complaint sets forth four causes of action sounding in: (1) negligence; (2) strict liability; (3) gross negligence; and (4) breach of warranty. The gravamen of each of Plaintiff's claims is that Defendants placed PCE into the stream of commerce in an allegedly defective, unsafe and inherently dangerous way that is, allegedly without adequate warnings, precautions and instructions. Plaintiff alleges that the PCE used by her employer was supplied by Dow and R.R. Street.

Following extensive discovery, the parties now challenge their opponents' respective expert witnesses. Anticipating a complex Daubert hearing, Defendants moved to amend the final pretrial order and add expert witnesses Drs. Laura Green and David Garabrant to testify as experts solely at the pretrial hearing of Defendants' motion to exclude the testimony of Plaintiff's medical causation expert, Dr. David Ozonoff. Magistrate Judge Stanley R. Chesler granted that motion. Plaintiff has moved to exclude the pretrial hearing testimony of Drs. Garabrant and Green[4] and submitted the affidavits of two additional experts, Drs. Philip Landigran and Anthony Robbins, to buttress Dr. Ozonoff's expert methodology.

On March 26, 2001, in light of the highly complex and contentious scientific, technical, and medical issues in this case, the Court designated Mark Weiss, M.D. to act as the Court's technical advisor solely for purposes of the Daubert hearing in this case. As the Court's technical advisor, Dr. *590 Weiss' duties were to assist and advise the Court with regard to the non-legal, technical, scientific and medical data, literature, and testimony supplied on behalf of the parties and to assist the Court in formulating questions to assess the reliability and relevancy of the methodology used by the parties' experts. Dr. Weiss did not opine on the ultimate issues of causation or warnings and offered no evidence or testimony in the case.

On May 21, May 23, and May 30, 2001, this Court held a Daubert hearing, to ascertain the reliability and relevancy of the medical causation experts in this case. At the hearing, the Court heard testimony of Drs. Ozonoff, Jandl, Greenland, Garabrant and Green regarding the reliability and relevance of the scientific methodology underlying the respective opinions offered on the medical cause of Plaintiff's AML. On June 11, 2001, the Court conducted a Daubert hearing with respect to the industrial hygienist expert, offered by Plaintiff to quantify Plaintiff's exposure to the chemical agent in question and to opine on the adequacy of Defendants' warnings[5]


The various experts' testimony is replete with scientific concepts, which are summarized herein:


"Epidemiology is the field of public health and medicine that studies the incidence, distribution and etiology of disease in human populations." Michael D. Green, et al., Reference Guide on Epidemiology, in Federal Judicial Center, Reference Manual on Scientific Evidence at 335 (2d ed.2000). The purpose of conducting epidemiological studies is to gain a better understanding of disease causation and to prevent disease in groups of individuals. "Epidemiological evidence identifies agents that are associated with an increased risk of disease in groups of individuals, quantifies the amount of excess disease that is associated with an agent, and provides a profile of the type of individual who is likely to contract a disease after being exposed to an agent." Id. at 336. The focus of epidemiology is on general causation (i.e., is the agent in question capable of causing disease?) and not specific causation (i.e., did the agent cause a disease in a particular individual?).

There are two categories of epidemiological studies: experimental studies and observational studies. Experimental studies, in the form of randomized trials, clinical trials, or true experiments, generally involve two groups, one of which is exposed to the agent in question while the other is not. In observational studies, individuals who have been exposed to the agent at issue are observed and compared to a group of individuals who have not been so exposed. When an agent is suspected to be harmful, most epidemiological studies concerning that agent are observational.

*591 The two primary types of observational studies are cohort studies and case control studies.[6] Cohort studies compare the incidence of disease among individuals exposed to an agent with an unexposed group, while case control studies look at the frequency of exposure in individuals who have the disease as compared to a group of individuals who do not have the disease.

Epidemiological studies attempt to identify agents that are associated with an increased risk of disease. Thus, the first question an epidemiologist asks is whether an association exists between exposure to an agent and a particular disease. An association between exposure to an agent and a disease exists when the two occur together more frequently than they would by mere chance. Once an association is observed, the scientist undertaking the study must assess the strength of the association as well as whether the reason for the observed association is due to bias, chance or real effect.

A measure of the strength of an association in an epidemiological study can be expressed in terms of its "relative risk." Relative risk indicates the difference in risk of contracting a disease in people exposed to a risk factor, as compared to those not exposed (but otherwise similar). Determining the relative risk is important in understanding the results of a study because virtually every disease associated with a risk factor also occurs, at some rate, in the general population not exposed to the risk factor.

Relative risk is commonly calculated by dividing the risk of developing a disease observed in an exposed group by the risk observed in an unexposed, but otherwise similar group.[7] If the risks of the unexposed and exposed are the same, then the relative risk estimate (which mathematically is simply the former divided by the latter) is 1.0. This value is also called the null value, and indicates that exposure is not associated with the disease in that study. Thus, a relative risk of 1.0 means that the agent has no effect on the incidence of disease. Similarly, if the relative risk estimate is 1.3, then risk appears to be 30% higher among the exposed compared to the non-exposed. When the relative risk reaches 2.0, the risk has doubled, indicating that the risk is twice as high among the exposed group as compared to the non-exposed group. Thus, "the threshold for concluding that an agent was more likely than not the cause of an individual's disease is a relative risk greater than 2.0." Green, supra, at 384.

In evaluating epidemiological studies, it is important to note that:

"[A]n association is not equivalent to causation. An association identified in an epidemiological study may or may not be causal. Assessing whether an association is causal requires an understanding of the strengths and weaknesses of the study's design and implementation, as well as a judgment about how the study findings fit with other scientific knowledge."

Id. at 336-337. Generally, there are three reasons that a positive association may be observed: (1) bias (including confounding factors[8]), (2) chance, and (3) real effect. *592 Each must be evaluated to extract a valid message from the study. Evaluation of these factors measures the "internal validity" of an epidemiology study, that is, the extent to which the study's findings are viable and sound.

"Bias" in epidemiology is systematic error, which includes "confounding bias." "The underlying mechanisms for these biases is to make the two groups being compared different in more ways than just the variable being studied." (October 1, 1997 Report of Dr. David Ozonoff at 13 (hereinafter "Initial Report").) Sources of bias must be considered in interpreting an epidemiological study because bias can produce an erroneous association. Green, supra, at 363.

The possible role of chance must always be considered in evaluating the results of an epidemiology study. To evaluate the role of chance in an epidemiology study, epidemiologists use statistical methods. Statistical methods provide information about how likely it is that chance or the random distribution of unidentified differences between the two groups being comparedis the reason for an observed association. The two primary techniques used to assess random error are statistical significance and confidence intervals. Green, supra, at 354.

A confidence interval is an indication of the range of associations between a factor and an outcome that is supported by the data in an epidemiology study. The confidence interval provides a measure of the precision, or statistical stability, of an association. In this case the factor under evaluation is exposure to PCE and the outcome is acute myelogenous leukemia. The wider the confidence interval, the less precise and less stable the association observed in a study.

Statistical significance, calculated as a P-value, is another statistical method of evaluating the possible role of chance. The P-value measures the likelihood that one would see the observed association even in the absence of a true association. The lower the P-value, the less likely the observed result is due to chance alone. In other words, as the P-value gets smaller and smaller, the data is less and less consistent with the "null hypothesis": the hypothesis that there is no association between the factor and the outcome.

Traditionally, P-values are termed "statistically significant" when they are less than 5 percent (P

In determining whether an observed association between a chemical and a disease is causal (i.e., general causation), scientists are guided by various factors, which are often referred to as the Hill criteria.[9] These factors include: (1) strength of association (i.e., whether the association is strong and statistically significant); (2) temporal relationship (i.e., whether the timing of the exposure and the onset of disease is consistent with the latency period *593 of the disease); (3) consistency of the relationship (i.e., whether the results of multiple scientific studies are consistent); (4) biological plausibility (i.e., whether there exists a biologically plausible mechanism by which the agent could cause the disease); (5) consideration of alternative explanations (i.e., whether the association could be accounted for by other factors); (6) specificity (i.e., whether the specific chemical is associated with the specific disease at issue); and (7) dose-response relationship (i.e., whether an increase in exposure yields an increase in risk).

Animal Studies

Another gauge by which to measure the toxicity of an agent in humans is through animal toxicology studies, or animal bioassays. The purpose of animal studies is not to predict what specific types of cancer a particular carcinogen might cause in humans, but rather to identify whether it can cause cancer at all. However, animal bioassays are of limited use in determining whether a particular chemical causes a particular disease, or type of cancer, in humans. "Where both animal toxicology and epidemiologic studies are available, no universal rules exist for how to interpret or reconcile them. Careful assessment of the methodological validity and power of the epidemiologic evidence must be undertaken and the quality of the toxicologic studies and the question of interspecies extrapolation and dose-response relationship must be considered." Green, et al., supra, at 346-47.

Standard of Admissibility for Expert Testimony

Federal Rule of Evidence 702, which governs the admission of expert testimony in federal court, provides that:

"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 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."

A Rule 702 determination is a question of law for the district court. Thus, when a party seeks to admit expert testimony, the district court must make an initial determination, in a preliminary hearing under Fed.R.Evid. 104(a), that the requirements of Rule 702 have been met. Daubert v. Merrell Dow Pharmaceuticals, Inc., 509 U.S. 579, 592, 113 S. Ct. 2786, 125 L. Ed. 2d 469 (1993). In Daubert the Supreme Court held that Fed.R.Evid. 702 imposes a special obligation on the trial judge to "ensure that any and all scientific testimony or evidence admitted is not only relevant, but reliable." Id. at 589, 113 S. Ct. 2786.

"[T]o qualify as `scientific knowledge,' an inference or assertion must be derived by the scientific method. Proposed testimony must be supported by appropriate validation i.e., `good grounds,' based on what is known. In short, the requirement that an expert's testimony pertain to `scientific knowledge' establishes a standard of evidentiary reliability." Id. at 590, 113 S. Ct. 2786. Additionally, Rule 702 requires that the evidence or testimony offered must "assist the trier of fact to understand the evidence or to determine a fact in issue...." Thus, the Daubert Court articulated the role of the district court as that of a gatekeeper:

"Faced with a proffer of expert scientific testimony, then, the trial judge must determine at the outset, pursuant to *594 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 that reasoning or methodology properly can be applied to the facts in issue."

Daubert, 509 U.S. at 592-593, 113 S. Ct. 2786.

The United States Court of Appeals for the Third Circuit has further clarified the trial court's gatekeeping role. In Paoli R.R. Yard PCB Litigation, 35 F.3d 717, 741-42 (3d Cir. 1994), the Third Circuit articulated the requirements of Rule 702:(1) the proffered witness must be a qualified expert; (2) the expert must testify about matters requiring scientific, technical, or specialized knowledge; and (3) the expert's testimony must "fit" the facts of the case. See also Kannankeril v. Terminix Int'l Inc., 128 F.3d 802, 806 (3d Cir.1997). Additionally, in response to the Supreme Court's decision in Daubert, Rule 702 has been amended. "The amendment affirms the trial court's role as gatekeeper and provides some general standards that the trial court must use to assess the reliability and helpfulness of proffered expert testimony." Fed.R.Evid. 702 advisory committee's note.

The first step of the Rule 702 inquiry is to determine whether the expert is qualified to testify. Paoli, 35 F.3d at 741. This requirement is liberally construed by the Third Circuit, which has "eschewed imposing overly rigorous requirements of expertise and [has] been satisfied with more general qualifications." Id. A broad range of knowledge, skills and training qualify a witness as an expert. Id. Second, the Rule 702 inquiry requires the expert testimony to be reliably based upon scientific methods. Paoli, 35 F.3d at 742. "Daubert explains that the language of Rule 702 requiring the expert to testify to scientific knowledge means that the expert's opinion must be based on the `methods and procedures of science' rather than on `subjective belief or unsupported speculation'; the expert must have `good grounds' for his or her belief." Id. (citing Daubert, 509 U.S. at 590, 113 S.Ct. 2786). These concepts have been articulated as factors to guide the Court's assessment of the reliability of proffered scientific expert testimony: (1) whether the theory or technique can be tested, (2) whether the theory or technique has been subjected to peer review, (3) whether there is a high rate of known or potential error, (4) whether there are standards controlling the technique's operation, (5) whether the theory enjoys "general acceptance," (6) whether there is a sufficient relationship between the technique and methods which have been established to be reliable, (7) whether the expert witness' qualifications are sufficient, and (8) whether the method has been put to non-judicial uses. Daubert, 509 U.S. at 593-94, 113 S. Ct. 2786; Paoli, 35 F.3d at 742 n. 8. Some courts also consider additional factors relevant in determining reliability, including: (i) whether the expert's proposed testimony grows naturally and directly out of research the expert has conducted independent of the litigation (see Daubert v. Merrell Dow Pharmaceuticals, Inc., 43 F.3d 1311, 1317 (9th Cir.1995)); (ii) whether the expert has unjustifiably extrapolated from an accepted premise to an unfounded conclusion (see General Elec. Co. v. Joiner, 522 U.S. 136, 146, 118 S. Ct. 512, 139 L. Ed. 2d 508 (1997)); (iii) whether the expert has adequately accounted for alternative explanations (see Claar v. Burlington, N.R.R., 29 F.3d 499 (9th Cir.1994)); (iv) whether the expert is being as careful as he would be in his professional work outside of the litigation *595 context (see Sheehan v. Daily Racing Form, Inc., 104 F.3d 940, 942 (7th Cir. 1997)); and (v) whether the field of expertise asserted by the expert is known to reach reliable results for the type of opinion proffered by the expert (see Kumho Tire Co. v. Carmichael, 526 U.S. 137, 149-50, 119 S. Ct. 1167, 143 L. Ed. 2d 238 (1999)). See Fed.R.Evid. 702 advisory committee's notes.

The Supreme Court in Daubert emphasized that the Rule 702 inquiry was "a flexible one" and emphasized that the individual factors are neither exclusive nor dispositive. Daubert, 509 U.S. at 594-595, 113 S. Ct. 2786. As such, this Court considers the relevant factors set forth by the Supreme Court and the Third Circuit, as well as those articulated by sister courts in their gatekeeping decisions.

In assessing whether the proffered scientific expert testimony is reliable, the Third Circuit has explained that the trial court should admit expert testimony "if there are `good grounds' for the expert's conclusion" notwithstanding the judge's belief that there are better grounds for some alternative conclusion. Heller v. Shaw, 167 F.3d 146, 152-53 (3d Cir.1999) (citations omitted). "Novel" conclusions should not be excluded where the methodology and its application are reliable. Id. at 153. "[A]n expert opinion must be based on reliable methodology and must reliably flow from that methodology and the facts at issue but it need not be so persuasive as to meet the party's burden of proof or even necessarily its burden of production." Id. at 152. "The ultimate touchstone of admissibility is helpfulness to the trier of fact." U.S. v. Velasquez, 64 F.3d 844, 850 (3d Cir.1995) (citations omitted).

The Court's inquiry "must be solely on principles and methodology, not on the conclusions that they generate." Daubert, 509 U.S. at 595, 113 S. Ct. 2786. However, "conclusions and methodology are not entirely distinct from one another." Joiner, 522 U.S. at 146, 118 S. Ct. 512. The Court must also "examine the expert's conclusions in order to determine whether they could reliably flow from the facts known to the expert and the methodology used." Heller, 167 F.3d at 153. "But nothing in either Daubert or the Federal Rules of Evidence requires a district court to admit opinion evidence that is connected to existing data only by the ipse dixit of the expert. A court may conclude that there is simply too great an analytical gap between the data and the opinion proffered." Joiner, 522 U.S. at 145-46, 118 S. Ct. 512. The recent amendments to Rule 702 make clear that the trial court must evaluate not only the principles and the methodology employed by the expert, but also whether the expert has properly applied those principles and methods to the facts of the case. See Fed.R.Evid. 702.

The third requirement of the Rule 702 inquiry is that the proffered testimony must "fit" the facts of the case. See Velasquez, 64 F.3d at 850. "Fit" requires that the proffered testimony must in fact assist the jury, by providing it with relevant information, necessary to a reasoned decision of the case. Paoli, 35 F.3d at 743.

As a general matter, the Rules of Evidence "embody a strong and undeniable preference for admitting any evidence" that could potentially assist the trier of fact and Rule 702 is liberally interpreted by the district courts. Holbrook v. Lykes Bros. Steamship Co., Inc., 80 F.3d 777, 780 (3d Cir.1996) (citations omitted). Acknowledging the "liberal thrust" of the Federal Rules of Evidence, the Supreme Court in Daubert instructed that "[v]igorous cross-examination, presentation of contrary evidence, and careful instruction on *596 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. The 2000 Amendments to Rule 702 have not changed this basic premise. With these principals as a guide, the Court assesses the admissibility of each challenged expert's testimony, following the Fed.R.Evid. 104 hearing.

A. Plaintiff's Daubert Challenge to the Joint Affidavit of Drs. Garabrant, Green and Jandl

Plaintiff has filed a motion in limine to bar this Court's consideration of the Joint Affidavit of Drs. Garabrant, Green and Jandl, submitted pursuant to Judge Chelser's July 31, 2000 Order permitting the designation of additional experts solely for the purposes of the Daubert hearing. In essence, Plaintiff asserts that this testimony fails under Daubert to meet the standard for consideration in a Daubert hearing, i.e., Daubert within Daubert.

Plaintiff's bases for seeking exclusion of the Joint Affidavit are as follows: (1) the Joint Affidavit disputes Dr. Ozonoff's conclusions under the guise of attacking his methodology; (2) the Joint Affidavit is itself not reliable under Daubert; (3) none of the three affiants is qualified to opine on medical causation because none is an epidemiologist; and (4) none of the three affiants is qualified to opine on scientific method and/or statistics.

Plaintiff's challenge asks this Court to act as gatekeeper to itself, to undertake a hearing within a hearing and exclude the Joint Affidavit from the Court's own consideration. Plaintiff admits that "no real preceden[t] exists concerning the standards for experts offering opinions solely for the purposes of determining the competence of another expert's [opinion]" and implores the Court to apply the Daubert standard to the experts designated by the Defendants solely for purposes of the hearing on the trial experts. (Pl. Post-Hearing Br. at 7.) This Court has found no authority requiring it to apply Rule 702 to experts testifying solely for the purpose of assisting the Court with its gatekeeping function at a Daubert hearing.[10]

Nonetheless, this Court finds that Drs. Garabrant,[11] Green[12] and *597 Jandl[13] are qualified to offer their opinions for the purpose of the Daubert hearing in this case and their proffered opinions stem from reliable scientific methodology and are relevant.[14] Thus, for the purposes of this Daubert hearing, the Joint Affidavit and the respective experts' testimony are admitted for the Court's determination as to what weight it should be accorded in reaching its conclusion in the Daubert hearing. The Court relies on Plaintiff's thorough and knowledgeable counsel for her "[v]igorous cross-examination [and] presentation of contrary evidence," Daubert, 509 U.S. at 596, 113 S. Ct. 2786, and assesses the probative value of the evidence accordingly. To the extent that the Joint Affidavit attacks Dr. Ozonoff's conclusions, as opposed to his methodology, it, and the attendant testimony, has been disregarded.

B. Defendants' Daubert Challenge to the Admissibility of Dr. David Ozonoff's Testimony at Trial

Defendants have moved to exclude the testimony of Plaintiff's medical causation expert Dr. David Ozonoff, asserting that: (1) His opinions are based upon studies that do not support his conclusions; (2) His assumptions about disease causation are inconsistent with known science; and (3) He failed to rule out other possible causes. Specifically, Defendants assert that Dr. Ozonoff's opinion is unreliable both as to the general cause of AML and to the cause of Plaintiff's AML (specific causation) because he: (i) improperly relies on inconclusive epidemiological data that does not show that PCE doubles the risk of developing cancer; (ii) fails to take into account negative epidemiological studies; (iii) relies on/extrapolates from studies involving cancer and blood diseases other than AML; (iv) relies on/extrapolates from studies involving chemicals other than PCE; (v) relies on high dose animal studies that do not "fit" with the facts in this case; (vi) fails to quantitatively assess the dose of PCE received by Plaintiff; (vii) fails to account for background risks (including history of cancer in Plaintiff's family); (viii) fails to rule out alternative causes; and (ix) fails to properly consider the latency period of Plaintiff's illness.

1. Qualifications

Dr. David Ozonoff is a graduate of Cornell Medical College and is licensed to practice medicine in Massachusetts. As part of his medical training, Dr. Ozonoff studied oncology, hematology and pharmacology, which incorporate toxicology; he is well trained in epidemiology and chronic disease epidemiology. As an undergraduate *598 at the University of Wisconsin, Dr. Ozonoff studied mathematics, probability theory and statistical processes. Dr. Ozonoff obtained a Masters in Public Health from Johns Hopkins University and was employed by Massachusetts Institute of Technology for ten years. Dr. Ozonoff has worked on various environmental matters for the World Health Organization, has acted in an advisory capacity to local state and federal agencies and has held various teaching positions, including appointments at Harvard Medical School and Boston University's School of Public Health.

Dr. Ozonoff has been researching perchloroethylene since 1980. Specifically, he has examined and researched the toxicity of PCE as it relates to ground water contamination and was commissioned by the Commonwealth of Massachusetts and the National Institute of Health ("NIH") to research and study the effects of PCE contamination of groundwater. Dr. Ozonoff was selected by NIH to direct a research center, funded by NIH, which focuses on PCE toxicology, PCE epidemiology and the effects of exposure to PCE. Additionally, Dr. Ozonoff is the principal and/or co-investigator on several PCE related projects, including one that involves PCE and cancer. Dr. Ozonoff has published extensively on epidemiological and toxicological issues, including publications involving exposure to PCE, and has served as a peer reviewer for federal agencies. Dr. Ozonoff has also taught epidemiology and related subjects to masters and doctoral students since 1977. In light of his impressive credentials and experience, Dr. Ozonoff clearly is qualified to testify as an expert in this case.

2. Dr. Ozonoff's Opinion  
a. Dr. Ozonoff's October 1, 1997 Report

Dr. Ozonoff's testimony is proffered to establish the medical cause of Ms. Magistrini's acute myelomonocytic leukemia. On October 1, 1997, Dr. Ozonoff issued his Initial Report (hereinafter the "Initial Report"). The substantive portions of the Initial Report cover three broad areas: (1) an overview of the methods used by scientists to inquire into the causes of occupational diseases; (2) Dr. Ozonoff's opinion on general causation and the relationship between PCE and leukemia; and (3) Dr. Ozonoff's opinion on specific causation, that is, the cause of Ms. Magistrini's acute myelomonocytic leukemia.

(i) General Causation

Dr. Ozonoff opined that: "On the basis of the cited scientific evidence and reasoning given above, I have concluded, to within a reasonable degree of medical probability, that PCE can be a substantial factor in causing human leukemia (all kinds)." (Initial Report at 36.) In support of his opinion, Dr. Ozonoff relied on animal bioassays, 14 human epidemiological studies and toxicology studies, which he referred to as "the left bookend," "the right bookend" and "between the bookends," respectively.

Animal Bioassays

According to Dr. Ozonoff, animal bioassays are the "gold standard" for identifying chemical carcinogens. (Initial Report at 24.) Animal bioassays are classic experiments in which one group of animals is exposed to a chemical and another, the control group, is not. (Initial Report at 25.) Rodents are chosen for these types of experiments because they are mammals with biological processes similar to humans. (Initial Report at 24-25.) "Results of such bioassays are an important part of the basis for regulatory decisions about the carcinogenicity of chemicals ...." (Initial Report at 25.) "[T]he importance of the bioassay findings is that PCE can cause this biological effect [cancer] at all, *599 immediately placing it in a different category than most chemicals." (Initial Report at 25.)

Three animal bioassays conducted by or under the auspices of the National Toxicology Program (the "NTP") and the National Cancer Institute (the "NCI") are cited as having been considered by Dr. Ozonoff in rendering his opinions in this case. In the first, cancer of the liver was found in the NCI's 1977 study of mice that orally ingested PCE. In the second, cancer of the liver was found in the NTP's 1986 study of mice that inhaled PCE. In the third, leukemia and kidney cancer were found in the NTP's 1986 study of rats that inhaled PCE. An elevated level of leukemia was not observed in the two studies involving mice.

In the Initial Report, Dr. Ozonoff raises and attempts to refute two common criticisms of animal bioassays generally: (1) certain animal strains are predisposed to develop cancer and (2) such high doses are used in bioassays that the carcinogenic process is fundamentally altered. Dr. Ozonoff states that, "Differences in response by site of action is most likely a function of genetic differences between the strains" and that the importance of bioassays is that they indicate that a chemical is capable of causing cancer at all. (Initial Report at 25.) To refute the second criticism, Dr. Ozonoff implies that PCE is genotoxic[15] and that scientists generally accept the proposition that low dose extrapolations are valid for chemicals that are genotoxic.

Human Epidemiological Studies

Dr. Ozonoff also considered fourteen human epidemiological studies in rendering his opinions in this case. In deciding what epidemiological studies to include, Dr. Ozonoff made two basic assumptions: (1) the chemicals known as the chlorinated ethylenes (specifically TCE and PCE) are sufficiently similar such that they can be treated together when determining their toxicological effects and (2) lymphohematopoietic cancers[16] can be treated together for etiological purposes. On the basis of these assumptions, Dr. Ozonoff included in his assessment studies that involved exposure to chemicals other than PCE and studies that involved cancers of the blood other than leukemia.[17]

The Initial Report listed, in table format, the fourteen studies together with each study's population, design and results. In most instances, the chemical being tested was not listed on the table and in many cases the results of the study did not reference leukemia or any other lymphohematopoietic cancer. Additionally, exposure data was not included in the Initial Report. It was unclear from the Initial Report whether the "results" were those of the studies' authors or of Dr. Ozonoff. Dr. Ozonoff concluded that, "Taken together, these studies have showed associations of dry cleaning work *600 with leukemia, cancer of the urinary tract (bladder and kidney) cervix, lung, colon, pancreas and liver." (Initial Report at 27-28.)

Toxicological Studies

Dr. Ozonoff acknowledges that "[t]he basic studies of the toxicological mechanisms of PCE supplement but are not the basis for the conclusion that PCE is a human carcinogen." (Initial Report at 31.) Dr. Ozonoff instead relies on toxicology data to support his reliance on the animal bioassays. Specifically, he asserts and explains that "evidence suggests" that humans metabolize PCE in the same way that it is metabolized by some animals. (Initial Report at 32.)

Additionally, Dr. Ozonoff cites to a review of the carcinogenicity of PCE by the International Agency for Research on Cancer ("IARC"), which concluded, inter alia, that: (1) PCE is a probable[18] carcinogen; (2) PCE induced leukemia in rats and (3) several epidemiological studies showed elevated risks from esophageal cancer, non-Hodgkin's lymphoma, and cervical cancer. (Initial Report at 34.)

Method of Weighting Studies Included in "Weight-of-the-Evidence" Opinion in Initial Report

While Dr. Ozonoff's Initial Report accumulated and listed fourteen epidemiological studies and three animal bioassays, the Initial Report did not explain the methodology he used in weighing them against contra scientific evidence to reach his conclusions on general causation. The data provided only a framework from which to begin to approach the critical questions in this case.

Dr. Ozonoff also discussed scientific concepts such as internal and external validity of studies, statistical significance and statistical methods. However, he never addressed his method for assessing the internal and/or external validity of each of the studies that he allegedly "weighed,"[19] nor did he methodologically evaluate statistical significance or confidence intervals ascribed to those studies in deciding how much weight to accord each.

(ii) Specific Causation

Dr. Ozonoff opined that: "It is my opinion, to within a reasonable degree of *601 medical probability, that Ms. Magistrini's leukemia was caused by her exposure to perchloroethylene (PCE) in the course of her employment in the drycleaning industry." (Initial Report at 2.) Using the method known as a differential diagnosis (discussed below), Dr. Ozonoff evaluated the history of Ms. Magistrini's illness, her prior and family medical histories, her exposure history (work, recreational and environmental), her social history and her use of medications prior to her illness. He also evaluated her medical records.

Dr. Ozonoff discussed the process of differential diagnosis, listed Plaintiff's history and concluded:

"On the basis of my education, training, research and experience, taking into account the existing scientific literature, and after ruling out any other plausible causes for Ms. Katherine Magistrini's acute myelomonocytic leukemia, I have reached the opinion, within a reasonable degree of medical certainty, that the cause of Ms. Magistrini's disease was her exposure by inhalation, skin absorption and ingestion to tetrachloroethylene (PCE), a solvent used to dryclean clothes and abundantly present in her workplace prior [to] the onset of her disease."

(Initial Report at 40-41.) Dr. Ozonoff did not specifically address the history of cancer in Plaintiff's family nor did he further discuss Plaintiff's history of smoking.

b. Dr. Ozonoff's Supplemental Affidavit

In response to the Joint Affidavit submitted by Defendants, Dr. Ozonoff prepared and submitted a Reply to the Affidavit of Drs. Garabrant, Green and Jandl, in further support of his Initial Report (hereinafter the "Reply Affidavit"). In the Reply Affidavit, Dr. Ozonoff identified the methodology he used in rendering his opinions as "the weight-of-the-evidence methodology." According to Dr. Ozonoff, "Although there is no accepted definition [of the] methodology, the essence of the `weight-of-the-evidence' approach requires that different types of data be evaluated together. This may include toxicology and chemical/structural studies, epidemiological studies, animal studies and comparison with toxicity benchmarks." (Reply Affidavit at 10.) Factors to be included in an evaluation of disparate data using the weight-of-the-evidence approach are: (1) tumor data from human studies; (2) tumor data from animal studies; (3) structure-activity relationships; (4) short-term test findings; (5) results of appropriate physiological, biochemical and toxicological observations; and (6) comparative metabolism and pharmacokinetic studies. (Reply Affidavit at 10.) However, Dr. Ozonoff still did not offer any scientific method to guide what weight he had accorded each piece of evidence nor did he explain why he weighted the evidence as he did nor how, given the body of evidence before him, he arrived at his ultimate conclusions.

c. Testimony During the Daubert Hearing

Read together, the Initial Report and the Reply Affidavit were insufficient for this Court to determine and assess the reliability and relevance of Dr. Ozonoff's methodology. This paucity of a "weighting methodology" could have been supplied at the extensive Daubert hearings held during the course of several days of live testimony. See Oddi v. Ford Motor Co., 234 F.3d 136, 152 (3d Cir.2000) (citing Padillas v. Stork-Gamco, Inc., 186 F.3d 412, 417 (3d Cir.1999)). "The opportunity to be heard is important because it allows a plaintiff a chance to have his or her expert demonstrate and explain the good grounds upon which the expert evidence rests." Id. (citations omitted). However, after three days of testimony no reliable scientific *602 method for conducting the weighing of the evidence was offered by Dr. Ozonoff. (See below).

3. The Weight-of-the-Evidence Methodology

The "weight-of-the-evidence" methodology has been recognized by the United States Environmental Protection Agency as well as the International Agency for Research on Cancer ("IARC"). In its Guidelines for Carcinogen Risk Assessment, the EPA described the weight-of-the-evidence methodology as follows:

"Evidence of possible carcinogenicity in humans comes primarily from two sources: long-term animal tests and epidemiologic investigations. Results from these studies are supplemented with available information from short-term tests, pharmacokinetic studies, comparative metabolism studies, structure-activity relationships, and other relevant toxicologic studies. The question of how likely an agent is to be a human carcinogen should be answered in the framework of a weight-of-evidence judgment. Judgments about the weight of evidence involve considerations of the quality and adequacy of the data and the kinds and consistency of responses induced by a suspect carcinogen. There are three major steps to characterizing the weight of evidence for carcinogenicity in humans: (1) Characterization of the evidence from human studies and from animal studies individually, (2) combination of the characterizations of these two types of data into an indication of the overall weight of evidence for human carcinogenicity, and (3) evaluation of all supporting information to determine if the overall weight of evidence should be modified."

Guidelines for Carcinogen Risk Assessment, 51 Fed.Reg. 33992, 33996 (September 24, 1986).

This Court draws on the Third Circuit's discussion of the reliability of the differential diagnosis methodology in Paoli as instructive in this context. Importantly, because the weight-of-the-evidence methodology involves substantial judgment on the part of the expert, it is crucial that the expert supply his method for weighting the studies he has chosen to include in order to prevent a mere listing of studies and jumping to a conclusion. How else can one expert's choice of "weight" be helpful to a jury which may be called on to assess a "battle of weighers"? The particular combination of evidence considered and weighed here has not been subjected to peer review. However, the weight-of-the-evidence methodology has been used, in a non-judicial context, to assess the potentially carcinogenic risk of agents for regulatory purposes. The existence and maintenance of standards controlling the technique's operation when used for regulatory purposes is informative here. The EPA's Guideline for Carcinogenic Risk Assessment specifically set forth the types of information that should be considered and the standards that information must meet before it should be considered at all. 51 Fed.Reg. at 33996. When a weight-of-the-evidence evaluation is conducted, all of the relevant evidence must be gathered, and the assessment or weighing of that evidence must not be arbitrary, but must itself be based on methods of science.

While flexible application of the Daubert factors permits this Court to find that, properly applied, the weight-of-the-evidence methodology is not an unreliable methodology, in order for Dr. Ozonoff's opinion to go to a jury, the application of that methodology also must be reliable.

*603 4. Dr. Ozonoff's Application of the Weight-of-the-Evidence Methodology

In order for Dr. Ozonoff's expert opinion to be admitted, his weighing process must be based on methods and procedures of science, rather than on subjective belief or unsupported speculation. Paoli, 35 F.3d at 742. The question here is not the reasonableness, in general, of undertaking a "weight-of-the-evidence" analysis. See generally Kumho, 526 U.S. at 153-54, 119 S. Ct. 1167. "Rather, it [is] the reasonableness of using such approach, along with [the expert's] particular method of analyzing the data thereby obtained, to draw a conclusion regarding the particular matter to which the expert testimony was directly relevant." Id. The Court finds that Dr. Ozonoff's application of the weight-of-the-evidence methodology is flawed for several reasons stated below.

a. Inclusion of the Fourteen Epidemiological Studies in Dr. Ozonoff's Assessment

Dr. Ozonoff has not presented "good grounds" for the two assumptions upon which he based his decision to accord weight to studies involving chemicals other than PCE and to studies involving cancer other than leukemias.[20] As a result, the body of evidence that Dr. Ozonoff included in his assessment was not shown to be reliably composed.[21]See Paoli, 35 F.3d at 745.

First, Dr. Ozonoff contends that all lymphohematopoietic cancers can be treated together for etiological purposes. Dr. Ozonoff simply does not provide sufficient scientific support for this theory. When pressed, he could not identify any literature that supported this proposition. Moreover, Defendants proffered the testimony of a renowned hematologist, Dr. James Jandl, whose decades of medical practice and research stood behind his refutation of the methodological premise that all lymphohematopoietic cancers can be treated together for etiological purposes. Even Plaintiff's own statistical epidemiologist testified that, in conducting a metaanalysis, he would not include studies of cancers other than AML without consulting an oncologist first to determine whether it was biologically appropriate to do so. Moreover, Dr. Ozonoff provides absolutely no support for including cancers other than those of the blood in his assessment. Thus, it is the opinion of this Court that Dr. Ozonoff did not demonstrate a scientifically reliable basis for the inclusion of studies showing an association between PCE and cancers other than leukemia.[22]

Second, Dr. Ozonoff contends that the chemicals known as the chlorinated ethylenes, specifically TCE and PCE, are sufficiently structurally and functionally similar that they can be treated together when determining their toxicological effects. He likewise states that the two chemicals are *604 metabolized similarly. Again, Dr. Ozonoff does not sufficiently support this premise with scientific methods. It is not contested that certain chemicals are structurally and functionally similar to one another such as benzene and toluene where the former is a known carcinogen and the latter is known to pose no carcinogenic risk whatsoever. Plaintiff's other epidemiology expert testified that he would not include studies implicating chemicals other than PCE without consulting with scientists to determine that it was biologically sound and that, even if included, the weighting of such studies should distinguish between the types of chemicals. Yet Dr. Ozonoff pointed to no scientific authority that supported his position to include these "other chemical" studies other than his own opinion, and supplied no weighting adjustment for them. Thus, no good grounds for treating these two chemicals similarly was presented and Dr. Ozonoff did not properly include epidemiological studies involving chemicals other than PCE in his assessment.

b. Validity of the Studies Relied Upon by Dr. Ozonoff

In his Initial Report, Dr. Ozonoff explains that "Internal validity refers to the extent to which the experiment or study produces valid information on its own terms, i.e., the extent to which it is internally valid." (Initial Report at 12-13.) When evaluating the internal validity of a study, the researcher or scientist must account for the roles of bias, confounding factors, and the likelihood that the observed association is due to chance. Guidelines for Carcinogenic Risk Assessment, 51 Fed.Reg. at 33999. This evaluation must be undertaken before a causal association can be inferred between exposure to an agent and cancer in humans.

Even if all fourteen epidemiological studies were properly included in his assessment, Dr. Ozonoff did not adequately explain his methods for assessing the internal validity of the studies that he relied upon in rendering his opinion. As stated earlier, with respect to the human epidemiological studies, Dr. Ozonoff stated only that, "Evaluating these studies for internal validity, they present a mixed picture, not unexpectedly. Some are very well done, others have limitations. Most suffer from exposure misclassifications that would be expected to underestimate the risk."[23] (Initial Report at 31.) Neither his Initial Report nor the Reply Affidavit address the internal validity of the animal studies.[24] Moreover, Dr. Ozonoff's testimony sheds little light on which studies he considered "well done" and which he thought "had limitations." There is little to no discussion *605 of possible biases or confounding factors found in the studies.

Most importantly, Dr. Ozonoff does not explain which studies were more or less reliable based upon statistical methods that determine the probability that an observed association is due to chance. Rather, he states blanketly that, "Chance has been dealt with by the usual methods and is evaluable in each study." (Initial Report at 31.) Since epidemiological studies look at health and risk factors in only a sample and not the entire population, statistical methods are required to determine the precision with which risks have been estimated.[25] Both Drs. Ozonoff and Garabrant contend that confidence intervals better gauge the import of a particular study and provide more information than does the showing of statistical significance.[26] The Court accepts their contention for the purposes of this motion.[27]

As stated, a confidence interval measures the range of associations between a factor and an outcome that is supported by the data in an epidemiological study, providing an important measure of the precision, or statistical stability, of an association. Dr. Ozonoff cites confidence intervals for only two of the fourteen studies he cites in his Initial Report and does not otherwise discuss the confidence intervals of the studies. His testimony at the Daubert hearing shed no new light on the strength of the associations, if any, found in the studies he considered or on the precision or stability of those associations.

The testimony of Dr. David Garabrant underscored the flaws in using a "weight-of-the-evidence" methodology without identifying or factoring in the confidence interval of each study that is either given weight or excluded from the weight scale.[28]*606 Defendants' Exhibit 17, a summary of studies of PCE and leukemia, graphically sets forth the relative risks calculated in the underlying studies relied on by Dr. Ozonoff. In all but one study, the calculated confidence intervals include a relative risk of 1.0, or the null hypothesis. The only study that does not include the null hypothesis is the Ashengrau study, coauthored by Dr. Ozonoff. Yet the confidence interval of that study was so wide that even Plaintiff's statistical expert acknowledged that the results were unstable and imprecise. (Testimony of Dr. Sander Greenland, 5/23/01 Tr. at 69.)

c. Dr. Ozonoff Did Not Explain How He Applied His Methodology

Even if the Court were to accept the premises upon which Dr. Ozonoff bases his general causation opinion and even if the Court were to determine that Dr. Ozonoff had adequately assessed the internal validity of the studies upon which he relies, the single most serious flaw is the most basic: he simply has not set forth the methodology he used to weigh the evidence. Dr. Ozonoff describes the "weight-of-the-evidence" methodology as follows:

"[T]he weight-of-the-evidence is sort of what is sounds like. And it's what happens in this room, you know in front of a jury. You have evidence and you need to weigh that evidence. Or give different weights to the various pieces of evidence. So, you know, you may count the animal studies for something, you may count this epi study for something else, and this one for something else, and essentially what you're doing is you're putting together, you're trying to put together a picture of what you think is happening. It's like a jigsaw puzzle.... [b]ecause you've got these pieces, there are gaps between the pieces, there can be uncertainty as to whether this piece goes in the middle of the picture, or it's a piece that's really at the edge, or one that's a piece of blue sky that doesn't really add information, but it does go somewhere. Or maybe it's a piece from another puzzle, doesn't belong there at all. You got to fit that together in some kind of way. And depending upon what you think of these pieces, and how you weight them, you may, people construct different pictures, and that is real life.... In order to do it though, you need to have as many of the pieces around as are relevant to figuring out where all of the pieces go. So you need the animal studies and you need the toxicology and the exposure information as best you can get it. And the epi information and the case studies, all of those things tell you something. People select different things in different ways, they put more weight on some, and they don't pay attention to other things and they come up with different pictures."

(Testimony of Dr. David Ozonoff, 5/21/01 Tr. at 85-86 (emphasis added).)

Absent from the Initial Report, the Reply Affidavit and his subsequent testimony was an estimate of the relative risk found in each of the epidemiological studies Dr. Ozonoff relied upon. See discussion supra at 6-7. "[A] relative risk of 2.0 is not so much a password to a finding of causation as one piece of evidence, among others for the court to consider in determining whether an expert has employed a sound methodology in reaching his or her conclusion." Landrigan v. Celotex Corp., 127 N.J. 404, 419, 605 A.2d 1079, 1087 (1992). However, Dr. Ozonoff's failure to adequately address the relative risks found in the studies that he relied on weighs heavily in this Court's ultimate conclusion that his methodology is not sufficiently reliable to pass through the "gate" to the jury.

*607 In order to ensure that the "weight-of-the-evidence" methodology is truly a methodology, rather than a mere conclusion-oriented selection process that weighs more heavily those studies that supported an outcome, there must be a scientific method of weighting that is used and explained. In this case it appears that Dr. Ozonoff relied most heavily on his own study, which itself looked at only seven (7) cases of leukemia and had a huge confidence interval, indicating that the results of the study are unstable and imprecise. He neither explained why the confidence interval in that study was not of concern to him, nor did he sufficiently discredit other studies that found no association or a negative association with much more precise confidence intervals, nor sufficiently explain why he did not accord weight to those studies.

For example, an IARC working group evaluated PCE and the studies concerning the agent in 1995. International Agency for Research on Cancer, Dry Cleaning, Some Chlorinated Solvents and Other Chemicals, 63 IARC Monographs on the Evaluation of Carcinogenic Risks to Humans (1995). IARC concluded that there is limited evidence of carcinogenicity in humans based upon on "consistently positive associations between exposure to tetrachloroethylene and the risks for oesophageal and cervical cancer and non-Hodgkin's lymphoma," and not on data concerning leukemia. Id. at 202. Regarding the Ashengrau study, IARC stated that "although the increase in the relative risk for leukemia was significant, the result was based on only two cases," and concluded that "[n]o consistent evidence for an elevated risk for leukemia was seen in the cohort studies." Id. Additionally, no international or federal agency has classified PCE as a known, or established human carcinogen or leukemogen and the National Toxicology Program (NTP) considers the evidence pertaining to PCE's overall carcinogenicity in general to be inconclusive. The U.S. EPA has no official position on PCE's carcinogenicity to humans. While the 9th Report on Carcinogens of the U.S. Department of Health and Human Services classified PCE as "reasonably anticipated to be a human carcinogen based on sufficient evidence of carcinogenicity in experimental animals," it further states that there is no data available to evaluate the carcinogenicity of PCE in humans and considers evidence of it being a leukemogen as inconclusive. (Def.Ex. 6A.) Dr. Ozonoff does not provide any methodology for weighing and discounting this and other information.

Plaintiff asks this Court to make the following finding of fact:

"Dr. Ozonoff, in assessing the cause of Ms. Magistrini's leukemia, utilized the scientifically accepted `weight-of-the-evidence' methodology which is a manner in which an epidemiologist would place a value on each piece of evidence and then in a methodical systematic process attempt to put the pieces together like a puzzle to determine a picture of causation." (Pl.'s Proposed Findings of Fact and Conclusions of Law at 19.)

What this Court was never told, after three days of hearing and hundreds of pages of pre-hearing and post-hearing submissions, is what "methodical systematic process" was used by Dr. Ozonoff. And, it is that very process that is the heart of this Court's inquiry. Rule 702 requires the Court to determine whether "the process or technique ... used in formulating the opinion is reliable." Paoli, 35 F.3d at 742. Here, by contrast, Dr. Ozonoff has failed to explain how he valued each piece of evidence, his methodological basis for relying heavily on certain pieces of evidence and his methodological basis for completely discounting others. It may be *608 that Dr. Ozonoff determined that the only studies he found to be internally and externally valid were ones that suggested an association between PCE and leukemia. And, it may be that he found evidence to the contrary to be weak because those studies were somehow compromised. However, given multiple opportunities to aver the same in written submissions and in his testimony at the hearing, Dr. Ozonoff simply failed to provide this Court with good grounds for how he applied his methodology. Plaintiff's own statistical expert, Dr. Sander Greenland, testified that he could not identify what, if any, methodology was used by Dr. Ozonoff in weighing the evidence. (Testimony of Dr. Sander Greenland, 5/23/01 Tr. at 78-79.)

Dr. Ozonoff opines that causation is a "judgment" made about experimental data. (Reply Affidavit at 2.) "Judgment" does not substitute for scientific method; without a reliable method, result-oriented "judgment" cannot be distinguished from scientifically or methodologically-based judgment. Where, as here, elements of judgment pervade the methodology, it is essential that the expert set forth the method for weighing the evidence upon which his opinion is based. Absent that, this Court's role as gatekeeper to assess the reliability of the methodology applied in this case is nullified. This is precisely what the Supreme Court addressed in Joiner, when it stated that "[N]othing in either Daubert or the Federal Rules of Evidence requires a district court to admit opinion evidence that is connected to existing data only by the ipse dixit of the expert. A court may conclude that there is simply too great an analytical gap between the data and the opinion proffered." Joiner, 522 U.S. at 145-46, 118 S. Ct. 512. It is the opinion of this Court that, on many levels, "there is simply too great an analytical gap" between the data purportedly relied on by Dr. Ozonoff and the opinion he has proffered. This Court cannot admit Dr. Ozonoff's testimony as reliably based on a scientific method.

5. "Fit" or Relevance

The third prong of the Rule 702 inquiry, "fit," asks whether the proffered testimony can be applied to the facts at issue. See Paoli, 35 F.3d at 743. Because the Court finds Dr. Ozonoff's application of the "weight-of-the-evidence" methodology unreliable, the Court does not reach the issue of "fit."

6. Specific Causation

Dr. Ozonoff also opined that Ms. Magistrini's acute myelomonocytic leukemia was caused by her exposure to PCE while working at Defendant One Hour Martinizing Franchises. As stated earlier, Dr. Ozonoff performed a differential diagnosis to evaluate Ms. Magistrini and assess the specific cause of Ms. Magistrini's AML. This evaluation included taking a medical history, family history (which included a family medical history and residential history), and an occupational history. Dr. Ozonoff evaluated: (1) the circumstances of the onset of her illness; (2) past medical history so as to rule out pre-existing disease and other possible exposures; (3) her family history so as to ascertain propensities and vulnerabilities to disease; (4) information as to her present illness and associated symptomology; (5) her environmental and occupational history. (Initial Report at 36-41.) The environmental and occupational history considered (1) Plaintiff's exposure history to ascertain activities which may bear on relevance to the development of her leukemia; (2) information related to possible recreational and environmental exposures; (3) her social history including information as to alcohol consumption and nicotine usage; (4) her residential information in order *609 to determine other possible causes of her leukemia; (5) her past and current jobs, exposure to PCE, health problems in relation to work; (6) medication usage history; and (7) her past medical records regarding her development and diagnosis of AML.

In this Circuit, the technique of differential diagnosis has been found to be a reliable technique when properly performed. See Paoli, 35 F.3d at 758-759. As the Third Circuit pointed out in Paoli, the element of judgment present in the evaluation renders "most of the Daubert factors testability, general acceptance, peer review, and degree of production of errors ... of only limited help" in assessing whether a particular differential diagnosis was reliably conducted. Paoli, 35 F.3d at 758. The existence of standards controlling the technique's operation is informative to the reliability inquiry. Id.

Differential diagnosis, or differential etiology, is a standard scientific technique which identifies the cause of a medical problem by eliminating the likely causes until the most probable one is isolated. See Baker v. Dalkon Shield Claimants Trust, 156 F.3d 248, 252-53 (1st Cir. 1998). A reliable differential diagnosis typically is performed after "physical examinations, the taking of medical histories, and the review of clinical tests, including laboratory tests," and generally is accomplished by determining the possible causes for the patient's symptoms and then eliminating each of these potential causes until reaching one that cannot be ruled out, or determining which of those that cannot be excluded is the most likely. Kannankeril, 128 F.3d at 807 (explaining that "[d]ifferential diagnosis is defined for physicians as `the determination of which of two or more diseases with similar symptoms is the one from which the patient is suffering, by a systematic comparison and contrasting of the clinical findings.'") (quoting Seedman's Medical Dictionary 428 (25th ed.1990)).

A reliably performed differential diagnosis includes considering plausible alternative causes. See Paoli, 35 F.3d at 758 ("[D]ifferential diagnosis can be considered to involve the testing of a falsifiable hypothesis ... through an attempt to rule out alternative causes...."). "The task ... is to decide `is there any other way of explaining the set of facts before us, is there any other answer equally, or more, likely than cause and effect?'" (Initial Report at 36.) It is within the Court's discretion to exclude an opinion on specific causation where either: (1) The expert engaged in very few standard diagnostic techniques by which doctors normally rule out alternative causes and the expert offered no good explanation as to why his or her conclusion remained reliable; or (2) The defendants pointed to some likely alternative cause and the expert offered no reasonable explanation as to why he or she still believed that defendant's actions were a substantial factor in bringing about plaintiff's illness. Paoli, 35 F.3d at 760. While an expert is not required to rule out all alternative possible causes of a plaintiff's disease, "where a defendant points to a plausible alternative cause and the doctor offers no reasonable explanation" for why he still concludes that the chemical was a substantial factor in bringing about the plaintiff's disease, "that doctor's methodology is unreliable." Paoli, 35 F.3d at 759 n. 27, 760.

It is undisputed that 90-95% of leukemias are of unknown origin, occurring without any known causal factor. That leaves 5-10% of leukemias that are caused by exposure to some carcinogen or whose cause is otherwise known. In this case, Plaintiff was exposed to PCE as well as to the chemicals in primary and secondary cigarette smoke, which contain benzene, a *610 known human carcinogen. Plaintiff smoked for several years prior to the onset of AML, and she was subjected to secondhand smoke from her father's heavy smoking habit. Thus Dr. Ozonoff's differential diagnosis needed to reliably determine that PCE exposure more likely caused Plaintiff's disease than did exposure to smoke.

In order to properly assess the specific cause of an individual's illness, a quantitative risk assessment must be undertaken, which assesses the risk that a particular exposure caused a particular illness. When an individual is exposed to two or more risk factors, the same method for both risk factors would have to be applied in order to assess the risk attributable to each of those risk factors.

While the health evaluation performed by Dr. Ozonoff generally was quite thorough, Dr. Ozonoff did not use a reliable scientific methodology in determining that PCE was more likely than cigarette smoke to be a cause of Plaintiff's AML. As is stated in the history taken by Dr. Ozonoff, Plaintiff was a smoker who "[b]egan to smoke in 1973 (1-2 cigarettes/week), which by 1979 had progressed to 1-2 packs/day, filtered cigarettes."[29] (Initial Report at 40.) Additionally, prior to contracting leukemia, Plaintiff was exposed to secondhand smoke from her father's smoking for most of her life. At the Daubert hearing, both Drs. Green and Jandl opined that, while smoking was not a great risk factor for leukemia, it was a greater risk factor than PCE exposure. While Dr. Ozonoff is entitled to reach a different expert conclusion as to the relative risks of smoking and PCE in causing AML, he must evaluate their respective risks using the same scientific method. This he did not do in ruling out smoke as a causal factor.

In his Initial Report, Dr. Ozonoff discounted entirely smoking as a risk factor, without discussing why smoking was ruled out as a cause of Plaintiff's illness. At the Daubert hearing, Dr. Ozonoff conceded that smoking is a risk factor, but distinguished between its effects in men and women. He testified: "It's my opinion that there is pretty good evidence that [smoking] does cause [leukemia] in males," later adding that, "I'm willing actually to believe that [smoking] does cause leukemia in females." Dr. Ozonoff also testified that Plaintiff's smoking exposure was significant. Yet, it is clear that Dr. Ozonoff conducted neither a weight-of-the-evidence analysis nor any other similar analysis in order to compare the respective risks to Plaintiff of her exposures to smoking and PCE. Thus, his ultimate expert decision to discount smoking as an alternative cause of Plaintiff's AML is not based on any reliable method. Paoli, 35 F.3d at 759-760.

7. Rule 703 Inquiry

Defendants have also moved to exclude the testimony of Dr. Ozonoff pursuant to Fed.R.Evid. 703, which provides that expert opinions based upon otherwise inadmissible hearsay evidence shall be admissible only if facts or data are "of a type reasonably relied upon by experts in the particular field in forming opinions or inferences upon the subject ...." Fed. R.Evid. 703. Although the Court does not expressly reach this issue, Dr. Ozonoff's consideration of studies involving cancers other than leukemia and chemicals other than PCE also render his opinion vulnerable *611 to exclusion pursuant to Fed.R.Evid. 703.

C. Plaintiff's Daubert Challenge to the Testimony of Dr. James Jandl

Plaintiff has moved in limine to exclude the proffered opinion of Defendants' expert witness on medical causation, Dr. James Jandl. The basis of Plaintiff's motion is twofold: (1) Dr. Jandl, a hematologist, is not qualified to opine on the issue of medical causation and (2) Dr. Jandl's opinion is an insufficient "net" opinion, which should be excluded due to its brevity and lack of foundation.

1. Qualifications

It is undisputed that Dr. James Jandl, an experienced hematologist and renowned professor, is both a researcher and clinician, expert in teaching and treating hematopoietic disorders. Dr. Jandl is a Professor of Medicine Emeritus at Harvard Medical School and a clinician at Beth Israel Hospital and Brigham and Women's Hospital. He has authored, in whole or in part, 126 peer reviewed publications and is the sole author of the treatise Blood: Textbook of Hematology, which is widely used as a medical text.

Plaintiff first challenges Dr. Jandl's qualifications to be an expert on the medical causation of leukemia in this case because he is neither a toxicologist nor an epidemiologist. Expert qualifications, however, can derive both from expertise in how outside substances affect the human organism and how the human organism becomes ill. Dr. Jandl is a renowned expert in hematology. As a hematologist he often tackles the issue of medical causation because of its crucial importance in treatment and medical management decisions. Determining the cause of a patient's leukemia is critical to his or her course of clinical treatment. Additionally, in his text Blood, Dr. Jandl wrote two chapters devoted to hematopoietic malignancies and their causes and AML and its causes, with a specific section on "Chemical Leukemogenesis." Both chapters cite to hundreds of sources upon which Dr. Jandl relied in drafting these chapters of his treatise.

Dr. Jandl testified that he is knowledgeable about leukemogenesis and hematologic toxicities. "[A] broad range of knowledge, skills and training qualify an expert as such." Paoli, 35 F.3d at 741. This Court finds that Dr. Jandl is qualified to opine on the issue of medical causation in this case.

2. The Reliability of Dr. Jandl's December 30, 1998 Report

In his December 30, 1998 Report, Dr. Jandl opined that:

"Based upon my thorough knowledge of the literature on chemical leukemogens, including scrutiny of the data on tetrachloroethylene exposure, it is my professional opinion that Ms. Magistrini's leukemia was not caused by exposure to any dry cleaning agent used in the course of her employment by the One Hour Martinizing franchise."

Plaintiff asserts that Dr. Jandl's opinion lacks adequate factual foundation and is therefore an impermissible "net opinion." Plaintiff further argues that Dr. Jandl conducted no critical investigation of the relevant medical and scientific literature on the issue of causation and did not accurately analyze the studies upon which he relied in forming his opinions on causation. Plaintiff also argues that Dr. Jandl's failure to examine Ms. Magistrini or conduct an exposure assessment renders his opinion unreliable.

In Dawson v. Bunker Hill Plaza Associates, 289 N.J.Super. 309, 323, 637 A.2d 847, 853-54 (N.J.Super.1996), the New Jersey *612 Appellate Division explained the "net opinion rule":

"Under New Jersey law, an expert's opinion must be based on a proper factual foundation. In other words, `expert testimony should not be received if it appears the witness is not in possession of such facts as will enable him [or her] to express a reasonably accurate conclusion as distinguished from a mere guess or conjecture.' This prohibition against speculative expert testimony has been labeled by modern courts as the `net opinion rule'. Under this doctrine, expert testimony is excluded if it is based merely on unfounded speculation and unquantified possibilities."

Id. (citations omitted.)

At the Daubert hearing, the Court afforded both Dr. Ozonoff and Dr. Jandl the opportunity to testify about their respective methodologies, in order to allow both parties "a chance to have his or her expert demonstrate and explain the good grounds upon which the expert evidence rests." Oddi, 234 F.3d at 152 (citations omitted). At the hearing, Dr. Jandl explained the breadth of literature and research knowledge that he relied on in formulating his opinion and clarified that his December 30, 1998 Report cited to only a few of the large number of sources he drew upon in rendering his opinion. In sum, Dr. Jandl relied on the body of medical research and experience comprising his tome Blood, which he alone wrote and which is a classic work on blood-borne diseases. Dr. Jandl's opinion is neither conclusory nor speculative and it is not a "net opinion" under New Jersey law.

At the hearing, Dr. Jandl set forth the method upon which his opinion is based. Dr. Jandl's opinion concerning the cause of Ms. Magistrini's illness is based upon his own extensive knowledge of the peer-reviewed medical literature on chemical leukemogens, the causes of leukemia, his review of the data on PCE and leukemia, his decades of clinical experience and his authoritative treatise Blood.

A review of the relevant Daubert factors demonstrates that Dr. Jandl's reasoning and methodology are reliable. It is testable, as the literature that he relied on and its underlying data could be reviewed to assure that it supported his conclusion. Dr. Jandl's own work and the medical literature on which he relies has been subject to extensive peer review and his evaluation of the literature was broad enough to encompass all of the relevant literature. Dr. Jandl's method is both generally accepted and widely used in a non-judicial setting by scientists and medical doctors. Additionally, the widespread use of his treatise demonstrates Dr. Jandl's prominence in the field of hematology. Dr. Jandl's opinion about the etiology of AML in humans "fits" the facts of this case and will assist the jury, by providing it with relevant information, necessary to a reasoned decision of the case. Paoli, 35 F.3d at 743. Plaintiff's Motion to Bar the Testimony of Dr. James Jandl is denied.[30]

*613 D. Defendants' Challenge of Mr. Stanton

Defendants Martin Franchises Inc. and Dow Chemical Company have each moved to exclude the opinion of Plaintiff's industrial hygienist, George Stanton. In support of that motion and pursuant to Judge Chesler's July 31, 2001 Order, Defendant Martin Franchises submitted the affidavit of engineer Steven Landon, solely for the purposes of the Daubert motion. That affidavit was struck after Mr. Landon failed to appear for the Daubert hearing, at which he would have been cross-examined. Defendant Martin Franchise's Motion for Reconsideration of the Court's bench ruling has been denied this day in a separate opinion.

In his report, Mr. Stanton offers two opinions. He opines on the concentrations of PCE to which Plaintiff was exposed and he opines that the equipment in question was defectively designed and lacked adequate warnings.

1. Qualifications

George Stanton is a certified industrial hygienist and engineer who has taught and practiced industrial hygiene in the State of New Jersey for almost thirty years. He received Bachelors and Masters Degrees in chemical engineering and also holds Masters Degrees in business administration and occupational health and safety. He has studied and specializes in, inter alia, industrial ventilation, and chemical labels and warnings. Mr. Stanton teaches fundamentals of industrial hygiene, including issues of chemical exposures in the workplace and air monitoring. Mr. Stanton has been a consultant on public safety and health for almost three decades and worked for the New Jersey Department of Labor as Chief of Occupational Health for four years. Mr. Stanton has published numerous articles on health and safety including publications concerning hazardous chemicals and controlling toxic substances and hazardous materials. Mr. Stanton is amply qualified to opine in this case on the issues involving recognition, evaluation and control of health hazards in the workplace.

2. Mr. Stanton's Opinion

Mr. Stanton opines both on the level of PCE to which Plaintiff was exposed and on the adequacy of the warnings given by the various Defendants. According to Mr. Stanton, Plaintiff was regularly exposed to at least 200ppm of PCE. This exposure estimate was determined by using an "odor threshold methodology."

The preferred method for determining the concentration of a chemical in the air is through air sampling. However, despite some evidence that as early as 1977 it was good industrial hygiene practice to regularly sample the air in dry cleaning establishments, no air sampling was performed at One Hour Martinizing during the relevant period. Thus, Defendants can provide no air sampling data.

Mr. Stanton opined that, in the absence of air sampling, the concentration of PCE can be estimated through the odor level tables provided in the industrial hygiene literature and the technical literature issued by Defendant Dow Chemical Company and various dry-cleaning trade associations. In support of his opinion he cited Patty's Industrial Hygiene and Toxicology, and publications by Dow Chemical Company, the International Fabricare Institute and the Laundry and Cleaning Allied *614 Trains Association to establish PCE's odor threshold. The testimony of Plaintiff Kathy Magistrini, the observations of Plaintiff's father and the testimony of the PCE distributor's salesman were used to determine the odor level in this case.

The methodology relied on by Mr. Stanton has been subject to peer review and is a generally accepted way of estimating exposure levels in the absence of air sampling. This methodology is not testable per se and the rate of error may be higher than one would like, given the subjective nature of an individual's ability to detect odors. However, even Defendants' expert, Dr. David Garabrant testified that a proxy level of exposure is sometime relied on by epidemiologists where actual exposure information is lacking. Dr. Garabrant indicated that an exposure estimate, albeit an imprecise one, could be ascertained using information concerning the agent involved, the cubic footage of the workspace and the odor threshold. While Mr. Stanton's methodology is not the most reliable method, Defendants should not benefit from their own past failure to perform the necessary air sampling to deprive Plaintiff of the ability to offer evidence of PCE exposure levels. Thus, the Court finds that Mr. Stanton's testimony would assist the trier of fact in determining Plaintiff's exposure to PCE and Mr. Stanton's exposure will be admissible.

Mr. Stanton also opines that it was industry practice to provide warnings and instructions regarding PCE and thus Plaintiff should have been warned of dangers of PCE and the consequent adverse health effects. This Court is inclined to permit Mr. Stanton to testify regarding industry practice and the state of industry knowledge. However, many of the opinions offered by Mr. Stanton encroach upon the province of the jury.[31] If these opinions are redacted, Mr. Stanton may also testify as an expert in this area. Here, again, the Court will rely on Defendants' thorough and knowledgeable counsel for "[v]igorous cross-examination [and] presentation of contrary evidence," Daubert, 509 U.S. at 596, 113 S. Ct. 2786, to counter testimony on industry practice and the state of industry knowledge.


For the foregoing reasons it is on this 4th day of January, 2002 hereby

ORDERED that Defendant Dow's Motion to Strike the Affidavits of Landigran and Robbins is GRANTED; and it is further

ORDERED that Defendant Dow's application that this Court not consider Plaintiff's amended conclusions of law is DENIED; and it is further

ORDERED that Plaintiff's Motion to Exclude the testimony of Drs. Garabrant Green and Jandl is DENIED;

ORDERED that Defendants' Motion to Exclude the Testimony of Dr. David Ozonoff M.D. is GRANTED; and it is further

*615 ORDERED that Plaintiff's Motion to Exclude the Testimony of Dr. James Jandl is DENIED; and it is further

ORDERED that Defendants' Motion to Exclude the Testimony of Mr. George Stanton is DENIED.

So ordered.


[1] Daubert v. Merrell Dow Pharmaceuticals, Inc., 509 U.S. 579, 113 S. Ct. 2786, 125 L. Ed. 2d 469 (1993).

[2] Plaintiff also moves to exclude the testimony of Drs. Laura Green and David Garabrant, which is offered by Defendants solely for the purposes of the Daubert hearing. Plaintiff proffers the testimony of Dr. Sander Greenland, a statistician, in an attempt to discredit the methodology used in affidavit jointly submitted by Drs. Garabrant, Green and Jandl (hereinafter the "Joint Affidavit") in support of Defendants' Motion to Exclude Dr. Ozonoff. The admissibility of Dr. Greenland's testimony is unchallenged.

[3] PCE is a chlorinated solvent that has been commercially produced since the early 1900's. PCE has been used in dry cleaning and textile processing for many years and was widely used by dry cleaners in 1977, when Plaintiff started working at One Hour Martinizing. In 1990, PCE accounted for 75% of the cleaning fluid used by dry cleaners.

[4] Plaintiff has also moved to exclude the pretrial hearing testimony of Dr. James Jandl as it relates to the opinion he proffered in the Joint Affidavit.

[5] Also before the Court are Dow's motion to strike the affidavits of Landigran and Robbins and Dow's motion to bar consideration of the untimely amended conclusions of law in Plaintiff's post-hearing submissions. The affidavits of Drs. Philip Landigran and Anthony Robbins are admitted for the purpose of demonstrating Dr. Ozonoff's reputation and not for the truth of their scientific content. Neither Landigran nor Robbins was named by Plaintiff in conjunction with Judge Chesler's July 31, 2000 Order and the Final Pretrial Order was not otherwise amended to permit their substantive testimony for the purposes of the Daubert hearing or for trial. Defendants have had no opportunity in this case to depose these two expert witnesses and they were not called upon to give testimony at the Daubert hearing. Defendant Dow's application that this Court not consider Plaintiff's amended conclusions of law is denied.

[6] Two other types of epidemiological studies, cross-sectional studies and ecological studies are not discussed here.

[7] Different types of epidemiology studies express the relative risk using different terms, including the standardized mortality ratio (SMR), the standardized incidence ratio (SIR), and the odds ratio (OR). The risk estimates may sometimes be expressed as percents, such that the null value is 100.

[8] A confounding factor is "[a] factor that is both a risk factor for the disease and a factor associated with the exposure of interest. Confounding refers to a situation in which the effects of two processes are not separated. The distortion can lead to an erroneous result." Green, supra, at 389.

[9] These factors, first set forth by Sir Austin Bradford Hill, also have been referred to as "viewpoints," emphasizing that one or more of the factors may be absent even where a causal relationship exists and that no factor is a sine qua non of causation.

[10] Even where the Court itself acts as the ultimate trier of fact at a bench trial, the Court's role as a gatekeeper pursuant to Daubert is arguably less essential. See e.g., Gibbs v. General American Life Ins. Co., 210 F.3d 491, 500 (5th Cir.2000); Volk v. United States, 57 F. Supp. 2d 888, 896 n. 5. (N.D.Cal.1999).

[11] Dr. David Garabrant is a medical doctor and an epidemiologist. He has designed and conducted numerous epidemiology studies involving occupational exposures, including exposure to PCE. He earned a Bachelor's degree with high honors in chemical engineering from Tufts University in 1972, a Medical Doctor degree from the Tufts University School of Medicine in 1976, a Master of Public Health degree from the Harvard School of Public Health in 1979 and a Master of Science degree in physiology from the Harvard School of Public Health in 1980. Dr. Garabrant is a Professor of Occupational Medicine at the University of Michigan School of Public Health. He is also an Associate Professor of Medicine, Department of Medicine, at the University of Michigan School of Medicine. Dr. Garabrant is licensed to practice medicine in four states and the District of Columbia. He is board certified in internal medicine, preventive medicine and occupational medicine and serves as an attending physician for the Occupational Medicine Clinic at the University of Michigan Medical Center.

[12] Dr. Laura C. Green is a toxicologist. She earned a Bachelor's degree with honors in Chemistry from Wellesley College in 1975, and a doctorate from the Department of Nutrition and Food Science at the Massachusetts Institute of Technology in 1981. Dr. Green is Senior Scientist and President of Cambridge Environmental and a Lecturer in the Division of Toxicology at M.I.T. Dr. Green's current work is in the fields of environmental health and toxicology, including quantitative risk assessment. She is certified in General Toxicology by the American Board of Toxicology.

[13] See infra at 611.

[14] The testimony and Joint Affidavit of Drs. Garabrant, Green and Jandl (hereinafter the "Joint Affiants") are largely offered to rebut the proffered reliability of Dr. Ozonoff's methodology. The statistical methods that they use are testable, have been peer reviewed, are subject to standards controlling the technique's operation (which largely have been followed), are generally accepted and are relied on much in the same way in many non-judicial contexts. Moreover, Plaintiff was permitted to cross-examine them and present contrary evidence through Dr. Sander Greenland, a statistician and epidemiologist. In Daubert, helpfulness to the trier of fact is the touchstone of admissibility. Velasquez, 64 F.3d at 850. This Court finds that the testimony of the Joint Affiants is indeed helpful and thus will not exclude it. To the extent that any one or all of the Joint Affiants relies on peer-reviewed literature to opine on the state of the knowledge, the propriety of doing so is discussed below.

[15] "Genotoxic chemicals alter the cell's DNA, thus initiating cancer ...." (Initial Report at 26 n. 54.) It is unclear from the Initial Report whether Dr. Ozonoff asserts that PCE is genotoxic. Instead he asserts that he has carefully considered arguments regarding genotoxicity and has "concluded, in agreement with standard practice and federal agencies like the National Toxicology Program, that the bioassays for PCE are relevant for human cancer risks at doses encountered by Ms. Magistrini." (Initial Report at 26-27.)

[16] Lymphohematopoietic cancers are cancers of the lymphatic and blood systems.

[17] In Paoli, 35 F.3d at 743, the Court found that "the requirement of reliability, or `good grounds,' extends to each step in an expert's analysis all the way through the step that connects the work of the expert to the particular case." The viability of these two underlying assumptions is necessarily discussed below.

[18] The International Agency for Research in Cancer IARC has developed a system for classifying chemicals based on what is known about their carcinogenicity. The four categories under which a particular chemical may be classified are (1) group one known human carcinogen; (2) group two probable or possible human carcinogen; (3) group three not carcinogenic to humans; and (4) group four insufficient data for classification. (Testimony of Dr. Laura Green, 5/30/01 Tr. at 138-140.) A chemical is classified as a "known" human carcinogen if there is sufficient epidemiological data indicating a causal relationship between the chemical and cancer in humans. The classification of "probable" carcinogen is given to a chemical if there is limited information indicating that it is carcinogenic to humans and sufficient information indicating a causal relationship between the chemical and cancer in animals. A chemical may be classified as a probable carcinogen based on animal studies alone. "Possible" carcinogens are those chemicals for which there is limited information indicating carcinogenicity to animals and an absence of human data. IARC categorizes probable and possible carcinogen together. Other agencies, such as the EPA and the National Toxicology Program, have developed similar systems of classification.

[19] As respects the human epidemiological studies, Dr. Ozonoff stated, "Evaluating these studies for internal validity, they present a mixed picture, not unexpectedly. Some are very well done, others have limitations. Most suffer from exposure misclassifications that would be expected to underestimate the risk. Chance has been dealt with by the usual methods and is evaluable in each study." (Initial Report at 31.)

[20] This Court does not find that these two premises are wrong, rather that they were not sufficiently supported here. Thus, Dr. Ozonoff did not present the Court with the "good grounds" necessary to find them reliable. It may be that Dr. Ozonoff gave them more thorough treatment in some other forum. However, it is Plaintiff's burden, by a preponderance of the evidence, to show that his opinion is reliable and, in turn, that the basic underpinnings of his opinion are reliable. Plaintiff has not done so here.

[21] Although styled as a motion pursuant to Rule 702, many of Defendants' arguments could have been brought as a Rule 703 challenge.

[22] Dr. Ozonoff concedes that his conclusion could be different if certain of the studies were excluded.

[23] But see Testimony of Dr. Sander Greenland, 5/23/01 Tr. at 85.

[24] Serious questions as to the internal validity and applicability of the NTP's 1986 rat study were raised by Defendants in the Joint Affidavit and by Dr. Green at the Daubert hearing. (Joint Affidavit at 16-17.) The Joint Affiants stated that the type of leukemia observed in the rat study is not an analogue to human AMML and is almost entirely unique to the Fisher rat. The Joint Affiants further stated that the PCE bioassays (and another bioassay being conducted at the same time, in the same place) yielded higher than expected rates of leukemia incidence in the control male rats. Additionally, the Joint Affiants stated that an audit of the study revealed (1) an inordinate number of animal escapes from cages; (2) lack of ear tags on the animals, which would have permitted them to be returned to their proper group upon recapture; (3) unusual stresses on the animals, such as higher than recommended temperatures, high levels of ammonia and routine eye injuries; and (4) deficiencies in the way the animals were autopsied. The auditor of the study found serious deficiencies in the study conduct, calling into question the findings in their entirety.

[25] Dr. Ozonoff's testimony regarding statistical methods for determining the meaning of the studies he relied upon is somewhat troubling. Dr. Ozonoff equivocates about whether ascertaining the role that chance played in an observed association is even necessary. While the Court understands that no study is perfect and that some (if not many) scientists espouse the drawbacks of using statistical significance as the touchstone of reliability, there must be some way to judge the results of a particular study. There must be some objective way to put a value on what that study says or shows. To be sure, even if some sciences don't require it (Testimony of Dr. David Ozonoff, 5/21/01 Tr. at 125-127), this Court does.

[26] A study is statistically significant if the p-value attributed to that study is less than .05. That is, the study shows that there is only 5% probability that an observed association is due to chance alone. The p-value is the probability of getting a test statistic as large or larger than observed, given that the null hypothesis is correct, and given that all the assumptions that went into creating the test statistic are correct. A large p-value is not evidence of the null hypothesis. (Testimony of Dr. Sander Greenland, 5/23/01 Tr. at 16-17.) Both Drs. Ozonoff and Greenland caution against using statistical significance rigidly, as it is a somewhat arbitrary measure of the import of a particular study, which provides little information about the value of the study.

[27] Still, it is notable that many courts confronted with determining the reliability of expert testimony look at whether or not the studies relied upon by the expert are statistically significant. See, e.g., General Elec. Co. v. Joiner, 522 U.S. 136, 145-46, 118 S. Ct. 512, 139 L. Ed. 2d 508 (1997) (evaluating reliability of epidemiological data based on statistical significance); In re TMI Litig., 193 F.3d 613, 711 (3d Cir.1999) (finding epidemiology study unreliable because results not statistically significant); see also Green, et al., supra, at 361 (noting that where the confidence interval encompasses a relative risk of 1.0, the study is not statistically significant at significance level chosen).

[28] The testimony of Dr. Garabrant is relied on not to dispute the conclusions drawn by Dr. Ozonoff, but rather to highlight what is critical to, yet absent from, Dr. Ozonoff's methodology and reasoning.

[29] It is unclear how quickly Plaintiff's smoking increased from a few cigarettes a week to one to two packs a day.

[30] It should be noted that experts historically have been permitted to testify in order to assist the fact-finder in areas beyond the ken of the ordinary person. For such generalized testimony, the Rule 702 inquiry requires that (1) the expert is qualified; (2) the subject matter of the testimony will assist the trier of fact; (3) the testimony is reliable; and (4) the testimony fits the facts of the case. See Fed. R.Evid. 702 advisory committee's notes. Dr. Jandl's qualifications and stature as a prominent hematologist have been established. The general principles of hematology and leukemia underpinning Dr. Jandl's opinion that there is no scientific basis establishing PCE as a cause of Plaintiff's leukemia are reliable in light of the breadth of knowledge, experience, and expertise relied on by Dr. Jandl. Thus, even if this Court had excluded Dr. Jandl's testimony insofar as it is specific to the cause of Plaintiff's illness, the Court would allow him to testify about the state of medical knowledge concerning hematology, lymphohematopoietic disorders, leukemogenesis and chemical leukemogenesis.

[31] The following opinions offered by Mr. Stanton bear on the ultimate issue of liability and on questions of fact for the jury: (1) Defendants' knowledge of the health hazards of PCE; (2) that PCE is defective and unreasonably dangerous when not accompanied with warnings; (3) that Defendants failed to provide adequate warnings; (4) that the Martin Defendants failed to provide proper inspections; and (5) that Defendants were negligent. As to these issues, Mr. Stanton offers no expertise, but only the information imparted upon him by Plaintiff. Additionally, Mr. Stanton's opinion that the absence of adequate labels and warnings was a causative factor in Plaintiff's overexposure is presumed in New Jersey where no adequate warning has been provided. James v. Bessemer Processing Co., Inc., 155 N.J. 279, 298, 714 A.2d 898, 908 (1998).