891 F.Supp. 12
United States District Court, D. Massachusetts.
13*13 James P. Keane, Michael J. Coleman, Keane & Klein, Boston, MA, for plaintiffs.
Michael R. Heyison, Hale & Dorr, Boston, MA, for defendant.
MEMORANDUM OF DECISION AND ORDER ON DEFENDANT'S MOTION IN LIMINE
STEARNS, District Judge.
This case arises out of the death of Gary Whiting from acute lymphocytic leukemia (ALL). Whiting was employed as a manual laborer at the defendant Boston Edison Company's Pilgrim Nuclear Power Station (Pilgrim) in Plymouth, Massachusetts, between 1977 and 1980. He died on December 2, 1983. Gary Whiting was 31 years old when he died. Judith Whiting, the administratrix of his estate, alleges that exposure to radiation at Pilgrim caused Gary Whiting's ALL. She alleges negligence on Boston Edison's part, and seeks damages under the Massachusetts Wrongful Death Statute, G.L. c. 229, § 2.
The plaintiff is required to show that the dose level of ionizing radiation received by Gary Whiting while employed at Pilgrim was a proximate cause of his illness and death. To meet that burden, the plaintiff intends to offer the expert testimony of Dr. Stuart L. Shalat, an epidemiologist, and Dr. Thomas A. Winters, an internist, as proof of both general and specific causation. See DeLuca v. Merrell Dow Pharmaceuticals, Inc., 911 F.2d 941, 958 (3d Cir.1990). Plaintiff also relies on Dr. Shalat's testimony to prove that Boston Edison breached the duty of care that it owed to Gary Whiting.
Before the court is Boston Edison's motion in limine to exclude the testimony of both witnesses. The motion challenges the expert 14*14 qualifications of Dr. Shalat and Dr. Winters and the reliability of the scientific methods and theories on which their opinions are based. See Daubert v. Merrell Dow Pharmaceuticals, Inc., ___ U.S. ___, ___ - ___, 113 S.Ct. 2786, 2794-2798, 125 L.Ed.2d 469 (1993). More specifically, Boston Edison contends that there is no scientific support for Dr. Winters' opinion (seconded by Dr. Shalat) that low doses of radiation can cause ALL and that the occupational radiation dose Gary Whiting received at Pilgrim more likely than not caused his ALL. Moreover, this court has previously determined that Boston Edison's duty of care to Gary Whiting was to avoid exposing him to occupational radiation in excess of the amounts permitted by federal regulations. Because there is no material dispute that Whiting's recorded dose level was within the allowed tolerances, plaintiff seeks by way of Dr. Shalat's testimony to prove that a dispute exists as to whether Gary Whiting's actual exposure exceeded his recorded dose. Boston Edison challenges Dr. Shalat's qualifications to perform dose reconstruction and the soundness of his methodology. Six days of hearings were held on the motion in limine. Testimony was taken from plaintiff's expert witnesses and rebuttal evidence was offered by the defendant. Seventy-one exhibits and fourteen affidavits were received in evidence.
THE FEDERAL DOSE LIMITS
The Federal Permissible Dose Limits establish the maximum radiation dose exposure permitted for workers at nuclear power plants licensed by the Nuclear Regulatory Commission (NRC). The regulations are published in 10 C.F.R. part 20. The whole body external dose limit for non-natural radiation exposure during Whiting's employment at Pilgrim was 3 rem to the whole body per calendar quarter, so long as the worker's prior occupationed dose, if any, was known, and his cumulative annual dose did not exceed 5 rem during each working year past age eighteen. 10 C.F.R. § 20.101. This court (by Mazzone, J.) has previously ruled that the whole body external dose limits established by 10 C.F.R. § 20.101 define the duty of care owed by Boston Edison to workers at the Pilgrim plant. See O'Conner v. Commonwealth Edison Co., 748 F.Supp. 672 (C.D.Ill.1990), aff'd, 13 F.3d 1090 (7th Cir. 1994).
Workers likely to be exposed to occupational radiation must be monitored by the licensee. Boston Edison used two types of monitor to measure workers' exposure, the Thermoluminescent Dosimeter (TLD) and the Self-Reading Pocket Dosimeter (SRPD). The TLD is a nonmechanical device that stores energy imparted by ionizing radiation in a phosphor chip. The TLD is accepted by health physicists as a reliable barometer of whole body exposure to radiation. Its disadvantage is that it can only be read by heating the phosphor to measure the rem equivalent of emitted light, a complex process that is typically performed only at quarterly intervals or when there is reason to believe that a worker has been exposed to a harmful dose of radiation. The SRPD, although less accurate, can be read visually at will, and is thus used to make daily estimates of whole body radiation exposure. Boston Edison, consistent with industry practice, maintained permanent records of workers' doses using the TLD. The TLD readings are considered a worker's "dose of record" and are accepted as such by the NRC. According to Boston Edison's records, the whole body dose accumulated by Gary Whiting during his employment at Pilgrim from August 7, 1977 to May 5, 1980, was 6.249 rem. The records also show that in none of the eight calendar quarters of Gary Whiting's employment did his exposure exceed the 3 rem permitted by 15*15 federal regulations. Without evidence to dispute the accuracy of Whiting's dose record, plaintiff will be unable to prove that Boston Edison breached its duty of care. Plaintiff intends to address this issue through the testimony of Dr. Shalat.
RADIATION AND ACUTE LYMPHOCYTIC LEUKEMIA
ALL is a type of naturally occurring cancer. ALL is one of four recognized histopathological types of leukemia. It is primarily a disease affecting children and the elderly. Its occurrence is extremely rare among adults ages 20 to 60. (Eighty percent of cases of ALL occur in children). Of the 29,300 cases of leukemia diagnosed each year in the United States, ALL comprises approximately 12,600 cases. Some 700 cases of ALL are diagnosed yearly in Massachusetts, slightly more than one of which on average occurs in men ages 30 to 34. Tr. 2, at 34-35. In most cases involving adults the etiology of ALL is unknown, although there is no dispute that ALL can be caused by ionizing radiation. Regrettably, "[c]ancers induced by radiation are indistinguishable from those occurring naturally; hence, their existence can be inferred only on the basis of a statistical excess above the natural incidence." BEIR III, at 137.
The most authoritative assessments of the health effects on humans of ionizing radiation are the periodic reports issued by the National Research Council of the National Academy of Sciences Committee on the Biological Effects of Ionizing Radiation (BEIR). The most recent fifth report (BEIR V) was published in 1990. As its authors observed, "[t]he human data on cancer induction by [high dose] radiation are extensive." BEIR V, at 2. The same is not true, however, with regard to low doses of radiation. "Assessment of the carcinogenic risks that may be associated with low doses of radiation entails extrapolation from effects observed at doses larger than 0.1 Gy [10 rads] and is based on assumptions about the relevant dose-effect relationships and the underlying mechanisms of carcinogenesis." Id. at 7. An additional problem lies in the fact that ALL is an extremely rare disease. "[T]he rarer the disease and the smaller the effect of exposure, the larger the sample needs to be." Id. at 49. Because researchers do not have access to a large population exposed to controlled doses of low-level radiation, estimates of the excess risk of ALL can only be generated "by extrapolation from smaller and less homogeneous groups who have been exposed to larger doses by using statistical dose-response models." Id. at 50.
The methodology employed by the BEIR V Committee involved three steps: (1) an 16*16 evaluation of existing epidemiological studies to determine the exposure levels at which excess incidences of cancer have been reliably demonstrated; (2) a downward extrapolation from the levels at which excess cancers have been actually observed to form estimates of the risks at levels for which no reliable data exists; and (3) an attempt to validate the predictions obtained.
Of existing studies, the BEIR V Committee deemed those involving the Japanese atomic bomb survivors and patients treated with x-rays for ankylosing spondylitis (spinal arthritis) as the most comprehensive and reliable. BEIR V, at 242. In the Japanese cohort, the lowest dose level at which a statistically significant excess risk of leukemia has been observed is 40 rem. Id. In the ankylosing spondylitis studies a slight increase in ALL over normal levels was detected; the average patient's dose, however, was 321 rad [3.21 Gy].
In extrapolating downward, BEIR V identified two statistical models as predictors of excess risk, one pertinent to generic cancer, the other to leukemia. "The dose-dependent excess of mortality from all cancer other than leukemia," according to BEIR V, "shows no departure from linearity in the range below 4 sievert (Sv) [400 rem], whereas the mortality data for leukemia are compatible with a linear-quadratic dose response relationship." Id. at 5. Below 40 rem exposure, the BEIR V Committee concluded that excess risk, if any, cannot be established with any degree of scientific certainty, merely hypothesized. Testimony of Dr. David G. Hoel, Tr. 6, at 48. Neither sufficient data nor understanding of the biological mechanism by which radiation induces cancer yet exist to validate either the linear or the linear-quadratic model as a measure of excess risk at low dose levels. Dr. David G. Hoel, Tr. 6, at 42-43, 48.
The conclusions of the BEIR Committee pose a second hurdle for the plaintiff in providing a jury with reliable evidence of causation. On this issue plaintiff relies primarily on the testimony of Dr. Winters, and to a lesser extent, on that of Dr. Shalat.
DR. STUART SHALAT
Dr. Shalat is an Associate Professor of Epidemiology at Texas A & M University where he also serves as the Director of the Institute of Occupational and Environmental Medicine. He earned a Sc.D. in epidemiology and occupational health at the Harvard School of Public Health in 1986 and the following year did post-graduate study at 17*17 Yale. Shalat's background in dosimetry is limited to two years experience as a health physics technician at Morrison Knudsen and a few weeks as a junior health physics technician for a subcontractor at Pilgrim. Tr. 2, at 72.
Dr. Shalat's theory that Gary Whiting was exposed to radiation in excess of that permitted by the federal dose regulations is based on his reconstruction of the total dose that he believes Whiting received while at Pilgrim (approximately three times the figure that Boston Edison recorded). Dosimetry (the science of determining the radiation dose received by persons who work with radioactive materials) is, as Dr. Shalat acknowledged, a scientific tool used by health physicists. It is not an epidemiological technique. Health physics is a specialized branch of science dedicated to protecting humans from harmful exposure to radiation. Dr. Shalat is not a health physicist nor does he have formal training in the field. He has not published, nor does he teach health physics as part of his faculty work at Texas A & M. Tr. 2, at 69. Although Dr. Shalat claims to have gained practical experience in dosimetry from prior employment as a health physics technician, I find his claims exaggerated, and in any event, not inclusive of dose reconstruction, as he admits. Tr. 2, at 72. As was made painfully clear in his deposition testimony, Dr. Shalat has difficulty with even elementary concepts of health physics.
Dr. Shalat's opinion is premised on the possibility that Gary Whiting's whole body dose could have been several times higher than the 6.249 rem measured by his TLD badges. This possibility rests on Dr. Shalat's questioning of the reliability and accuracy of the TLDs used at Pilgrim during Whiting's employment. Tr. 2, at 120-121. Dr. Shalat's evidence is derived from a 1980 study 18*18 conducted by Philip Plato of the University of Michigan for the NRC. According to Dr. Shalat, the Plato study determined that 79 percent of the TLDs then in use had inaccuracy rates in excess of 50 percent. Tr. 1, at 140. The study, however, also concluded that the TLDs supplied by Pilgrim's vendor, Yankee Atomic Electric Company, performed within a +/- 10 to 15 percent margin of error, all of it in the nature of over-reporting.
According to Dr. Shalat, he employed an "historical method" to recalculate Gary Whiting's radiation exposure. Tr. 1, at 35. As Shalat explained it, he relied on Boston Edison's internal records, primarily radiation work permits and radiation surveys, to recreate Whiting's work history. Tr. 1, at 44-45. Shalat "assumed" that as a manual laborer, Whiting would have been assigned to decontamination work and the placement of temporary shielding. Tr. 1, at 46. Shalat explained his method as follows:
"What I did was I looked at the general area of exposures that were listed and looked at the hot spots that were present in the area, and essentially took the average of the general area radiation exposures and then took the hot spots that were possible and made calculations based upon assumptions of the amount of time—I knew the amount of time listed on the records Mr. Whiting spent in the area. I deducted time for entry and exit of the area. I assumed a very small proportion of time in proximity to the hot spots, and based upon that did a reconstruction of his whole body external exposure." Tr. I, at 67-68.
From this methodology, Dr. Shalat arrived at a "point estimate" of Whiting's whole body dose of 18.031 rem, to which he attributed an error rate of +/- 25 to 30 percent. Tr. 1, at 73, 78. On the assumption that bone marrow dose (the relevant dose for leukemia) is typically 80 percent of whole body dose, Shalat converted his estimate of 18.031 rem to a bone marrow dose of 14.4 rem. Tr. 1, at 80.
Dr. Shalat's presentation was replete with factual and mathematical errors that cast significant doubt on the integrity of his purported reconstruction. His report (Trial Exhibit 5) attributed doses of 1,325 millirads to Whiting that were allegedly received between August 11, 1979 and August 21, 1979, when Whiting was not working at Pilgrim. Tr. 2, at 79-80. Similarly, Shalat attributed doses totaling 175 millirads to Whiting on days after he had permanently quit work at Pilgrim. Tr. 2, at 74. The calculations for the year 1978 mistake cumulative quarterly 19*19 dose totals for distinct quarterly measurements. Tr. 2, at 83.
In estimating Gary Whiting's total dose, Shalat used a formula of his own invention (Tr. 2, at 86) according to which total estimated wholebody dose is the sum of minutes of exposure multiplied by the sum of the average dose rate per minute enhanced for exposure to "hot spots" and adjusted for entry times greater or lesser than thirty minutes. While the formula is superficially impressive, it incorporates arbitrary assumptions involving Gary Whiting's presumed entry and exit times, the diffusion of radiation within his working areas, his actual movements, the time it took him to suit up and desuit, and his proximity to "hot spots." Tr. 2, at 84 et seq. The data used in the compilations are similarly flawed. As Boston Edison demonstrated, Dr. Shalat identified "hot spots" that do not appear on the supporting work permits and placed Whiting in areas of the plant when the records on which he claims to have relied have Whiting elsewhere. See Tr. 4, at 46 et seq. In applying his formula, Dr. Shalat made simple mathematical errors while performing forty-four of his seventy-four calculations. Tr. 2, at 88, 94. In sum, putting aside the issue of whether Dr. Shalat is qualified to offer an opinion as to Whiting's whole body exposure to radiation, his explication of the basis of that opinion is so riddled with factual inaccuracies and unproven assumptions that no reasonable jury could give his opinion credence.
DR. THOMAS WINTERS
Thomas Winters, the plaintiff's expert on causation, is a 1973 graduate of Tufts University School of Medicine. Dr. Winters is board certified in internal and occupational medicine. He directs a clinic on occupational medicine at Carney Hospital in Dorchester. As a sideline, Dr. Winters provides consulting services to lawyers seeking evidence of causal links between certain diseases and exposures to chemical substances. Dr. Winters has no training as an epidemiologist and does not "promote" himself as an expert in that field. Tr. 3, at 50. Dr. Winters acknowledged that he does not "do cancer research per se." Tr. 3, at 46. He is not 20*20 trained in hematology or oncology, the medical disciplines involved in the treatment of leukemias. Tr. 3, at 82-86. Dr. Winters has never conducted a study or published a paper on the subject of ionizing radiation and cancer induction. See Plaintiff's Exhibit 24. His pertinent scholarly efforts consist of unpublished secondary research on the association between ionizing radiation and cigarette smoke. Tr. 5, at 18-19. Dr. Winters is a former board member of Physicians for Social Responsibility and has campaigned actively against what he perceives as the dangers of nuclear power. He is an advisor to and former director of the Center for Atomic Radiation Studies, an organization formed "to alert the public about the hazards of ionizing radiation." Tr. 3, at 120. In 1988, Dr. Winters appeared in a television advertisement sponsored by Citizens for Safe Energy lending his support to a public campaign advocating the closing of the Pilgrim and Yankee Rowe nuclear power stations. In the advertisement, Dr. Winters claimed that twenty workers at Pilgrim had been injured and one had died as a result of radiation exposure. Dr. Winters repeated the claim in an October 26, 1988 affidavit, asserting that two workers had died. Dr. Winters conceded on cross-examination that the claim was false, and that the worker cited in the television commercial had in fact died of a heart attack. Tr. 3, at 135-137. Dr. Winters admits that he has no expertise in dosimetry and no background in radiation safety practices. Tr. 3, at 138-141. (Unlike Dr. Shalat, Dr. Winters considers the TLD to be the "gold standard" of radiation measuring devices. Tr. 5, at 70.) He has never visited the Pilgrim plant or studied its safety procedures (or the safety procedures of any other nuclear power facility). Tr. 5, at 22-23.
Dr. Winters does not believe that cancer occurs naturally, but rather is induced by external environmental factors like diagnostic x-rays and electromagnetic radiation. More saliently, Dr. Winters is convinced that any exposure to radiation "from zero up," even radiation occurring naturally in the environment, increases the risk of cancer. Tr. 3, at 122-123. Dr. Winters subscribes to a linear non-threshold model of the association between radiation and cancer, a model which he contends has been scientifically "proven." Tr. 5, at 10. While he relies on BEIR V in this regard, he disagrees with the BEIR V Committee's conclusion that in the specific instance of leukemia, a linear quadratic model is a more accurate predictor. Tr. 5, at 26. Dr. Winters concedes that there are no data contradicting the BEIR V determination that exposure to chronic low-level radiation cannot be correlated in any statistically meaningful way to an increase in the risk of cancer, but he believes that existing studies may not account for chromosomal effects over "several generations." Tr. 5, at 31. Dr. Winters disagrees with Dr. Arthur Upton, whose authority he otherwise accepts, that epidemiological data do not define the shape of dose-incidence curves for leukemia at levels of exposure below 50 rem, although he uses Dr. Upton's chapter on the subject in Levy's Occupational Health in teaching residents 21*21 at Carney Hospital. Tr. 5, at 63. Dr. Winters also disagrees with Dr. Upton's conclusion that ALL is rarely if ever induced by exposure to radiation after age twenty. Tr. 5, at 66.
According to Dr. Winters, he used a "clinical method" and a species of "differential diagnosis" to reach his conclusion that it was "more probable than not" that Gary Whiting's ALL was caused by exposure to radiation at Pilgrim. Tr. 3, at 43, 61-62, 79-80. Beginning with the hypothesis that Whiting's ALL was in fact caused by radiation, Dr. Winters sought to eliminate any possible alternative explanation. Tr. 5, at 71-72. He relied on a social history provided by Mrs. Whiting in a telephone conversation to exclude possible environmental, lifestyle or occupational causes. Tr. 3, at 61. He also reviewed Whiting's medical records, observing that his blood counts in 1978 and 1979 were normal. Tr. 3, at 65-66. He then looked to the scientific literature for support for the proposition that Whiting's exposure to approximately 6¼ rem at Pilgrim more likely than not induced his ALL.
Dr. Winters identified twelve studies that had led him to conclude to "a reasonable degree of epidemiological certainty" that Gary Whiting's ALL was more probably than not caused by his exposure to radiation at Pilgrim. Tr. 3, at 75-78. Although Dr. 22*22 Winters acknowledges that the relationship between cancer and ionizing radiation is influenced by a number of variables, including the type of cancer, the latency period, the age and gender of the patient, the type of radiation, and the patient's social history, Tr. 5, at 60-61, he did not factor any of these considerations into his evaluation of the studies on which he relied.
As defendant points out, leukemia comes in markedly different types which can fairly be classified as distinct diseases, some of which can be induced by radiation and some of which, like chronic lymphocytic leukemia, cannot. Several of the studies on which Dr. Winters relied do not differentiate among leukemia's subtypes, and must be considered suspect for that reason. Of the high dose studies Dr. Winters cited, only the ankylosing spondylitis study made the differentiation. In that study, where patients received an average dose of 321 rad to the bone marrow, only two cases of ALL were observed (where one was expected). Similarly Dr. Winters conceded that in the Japanese atomic bomb survivor studies no statistical correlation to excess incidence of leukemia was demonstrated below a dose level of 40 rem to the bone marrow, and 100 rem in the case of ALL. Tr. 5, at 83-84.
Dr. Winters also alluded to a number of low dose studies that he claims support his opinion that Whiting's relatively low (6¼ rem) wholebody exposure to radiation had caused his ALL. The studies cited by Dr. Winters are Stewart's studies of pregnant women and of workers at the Hanford nuclear facility, Caldwell's study of veterans exposed to radiation at the Shot Smoky nuclear tests, Stevens' study of children exposed to atomic testing fallout in southeastern Utah, Lyons' study of the population of southeastern Utah, Wing's study of nuclear workers at Oak Ridge, Tennessee, and Wilkinson's reanalysis of seven previously published studies of workers at nuclear plants. Although Dr. Winters did not specify the particulars about these studies that he believes support his opinion, they either do not support his opinion that ionizing radiation in the 5 to 10 rem dose range can cause ALL in adults, or they are marred by generally acknowledged flaws in their methodology. Stewart's Hanford study, for example, showed no excess risk of leukemia in the Hanford worker population as Dr. Winters conceded. Tr. 5, at 92. The Wing study of workers at Oak Ridge revealed no evidence of an excess risk of ALL, nor did the Caldwell study of soldiers exposed to radiation at the Shot Smoky tests. The Stevens study found a statistical 23*23 correlation between radiation and ALL only among children under age twenty and cautions that this may be attributable "to chance or to some methodological bias." The Lyons study, as Dr. Winters acknowledged, has been heavily criticized for its methodology (including by the BEIR V Committee). Tr. 5, at 107. Wilkinson's reanalysis is limited to studies of workers in defense facilities published prior to 1989 and is infected with the errors incorporated by its primary sources.
In the absence of validating data, Dr. Winters' opinion reduces itself to faith in a linear non-threshold model of the causal relationship between ionizing radiation and leukemia. In layman's terms, the model assumes that if a lot of something is bad for you, a little of the same thing, while perhaps not equally bad, must be so in some degree.The model rejects the idea that there might be a threshold at which the neutral or benign effects of a substance become toxic.While this may be true, say of cigarettes, it is not true of, for example, aspirin. Whether it is true of ionizing radiation and ALL was one of the major challenges posed to the BEIR panels. The BEIR V Committee decided to adopt a linear quadratic non-threshold model as an experimental predictor for ALL. What must be kept in mind is that both models, although radically different in their estimates of risk, are attempts to extrapolate from the known to the unknown, something that neither model does very well with ALL.
As Dr. Hoel explained when asked why, if scientists do not know, they guess anyway:
The speculation or the purpose for this is that we have to have some rational way of saying what is a permissible exposure level to ionizing radiation, how much do we allow in the environment. The EPA has to set those levels. With regard to toxic levels, we need to set use for medical uses; we need to set levels for radiation workers. And these levels, in my opinion — and I think in most public health people's opinion — should be health based.
And to do this, we need to estimate the risk at levels where we may not see them. If you think about it, if you just set it where you would see them, you're going to have a problem on your hands. In other words, if everyone in the environment could have 40 rem or 30 rem, something 24*24 like that, you may very well have a large effect. So you need to estimate these low dose effects. Tr. 6, at 48-49.
No testimony could better illustrate the difference between the quests for truth in law and science remarked on by the Supreme Court:
Scientific conclusions are subject to perpetual revision. Law, on the other hand, must resolve disputes finally and quickly.The scientific process is advanced by broad and wide-ranging consideration of a multitude of hypotheses, for those that are incorrect will eventually be shown to be so, and that in itself is an advance.Conjectures that are probably wrong are of little use, however, in the project of reaching a quick, final, and binding legal judgment— often of great consequence—about a particular set of events in the past.We recognize that in practice, a gatekeeping role for the judge, no matter how flexible, inevitably on occasion will prevent the jury from learning of authentic insights and innovations.That, nevertheless, is the balance that is struck by Rules of Evidence designed not for the exhaustive search for cosmic understanding but for the particularized resolution of legal disputes.
Daubert v. Merrell Dow Pharmaceuticals, Inc., supra at ___ - ___, 113 S.Ct. at 2798-2799.
THE DAUBERT STANDARD
In Daubert, the Supreme Court abandoned the general acceptance test of Frye v. United States, 293 F. 1013 (D.C.Cir.1923), noting the considerable academic criticism of Frye's ultra-orthodox approach, and finding it superseded by the more liberal relevancy test of Rule 702. "That the Frye test was displaced by the Rules of Evidence does not mean, however, that the Rules themselves place no limits on the admissibility of purportedly scientific evidence....The subject of an expert's testimony must be `scientific ... knowledge.' The adjective `scientific' implies a grounding in the methods and procedures of science.Similarly, the word `knowledge' connotes more than subjective belief or unsupported speculation....Proposed testimony must be supported by appropriate validation — i.e., `good grounds,' based on what is known."Id. at ___ - ___, 113 S.Ct. at 2794-2795. "Ordinarily, a key question to be answered in determining whether a theory or technique is scientific knowledge that will assist the trier of fact will be whether it can be (and has been) tested.`Scientific methodology today is based on generating hypotheses and testing them to see if they can be falsified; indeed, this methodology is what distinguishes science from other fields of human inquiry.'"Id. at ___, 113 S.Ct. at 2796 (citation omitted).
Daubert establishes the duty of a trial judge to play the role of a "gatekeeper" insuring that the fact-finding process does not become distorted by what is popularly called "junk science."See Wilson v. City of Chicago, 6 F.3d 1233, 1238 (7th Cir.1993). This role is especially sensitive in cases "where the plaintiff claims that exposure to a toxic substance caused his injury, [because a] jury may blindly accept an expert's opinion that conforms with their underlying fears of toxic substances without carefully understanding or examining the basis for that opinion."O'Conner v. Commonwealth Edison Co., 807 F.Supp. 1376, 1391 (C.D.Ill. 1992), aff'd, 13 F.3d 1090 (7th Cir.1994).
Two gateposts frame the exercise of a judge's discretion. First, a witness must be shown to be sufficiently qualified by "knowledge, skill, experience, training, or education" before he will be permitted to give expert testimony.Fed.R.Evid. 702. This means that a witness must be qualified in the specific subject for which his testimony is offered.Just as a lawyer is not by general education and experience qualified to give an expert opinion on every subject of the law, so too a scientist or medical doctor is not presumed to have expert knowledge about every conceivable scientific principle or disease.Second, the Federal Rules of Evidence require that a court "ensure that any and all scientific testimony or evidence admitted is not only relevant, but reliable."Daubert, supra at ___, 113 S.Ct. at 2795. "[T]he trial judge must determine at the outset, pursuant to Rule 104(a), whether the expert is proposing to testify to (1) scientific knowledge that (2) will assist the trier of fact to understand 25*25 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."Id. at ___, 113 S.Ct. at 2796 (footnotes omitted).
Applying Daubert's strictures, I conclude that neither Dr. Shalat nor Dr. Winters is qualified to testify in the fields for which their expertise is proposed. More specifically, I conclude that Dr. Shalat is not qualified to opine on the subjects of health physics and radiation epidemiology. As I have previously ruled, Dr. Shalat's reconstruction of Gary Whiting's whole body dose at Pilgrim is not science, but a figment of convenient invention. Dr. Shalat has no qualifications whatever to testify as to the specific cause of Gary Whiting's ALL. I also conclude that Dr. Winters is not qualified to testify on either the general or the specific causal relationship between radiation and the induction of ALL. Nor is he qualified to testify on the subject of health physics. I find that the opinion of Dr. Shalat and Dr. Winters that low doses of nuclear radiation in the range at issue in Gary Whiting's case (6 to 20 rem) can cause ALL is not based on scientific knowledge, but is grounded on speculation shaped by result-oriented biases.It fails all of the Daubert reliability factors. The linear non-threshold model cannot be falsified, nor can it be validated. To the extent that it has been subjected to peer review and publication, it has been rejected by the overwhelming majority of the scientific community.It has no known or potential rate of error. It is merely an hypothesis. In sum, it has no capacity to be of assistance to a jury in resolving the ultimate issues in this case.
For the foregoing reasons, defendant Boston Edison's motion in limine is ALLOWED. Because plaintiff will be unable to meet her burden of proof at trial on the issues of breach of duty and proximate cause, Boston Edison's motion for summary judgment is also ALLOWED.
 Plaintiff has offered no evidence that would implicate the internal (inhalation and ingestion) dose limits of 10 C.F.R. § 20.103. Gary Whiting's whole body counts indicated a total internal dose of radionuclides less than 2 percent of the maximum concentration permitted by the regulation.
 SRPD readings tend to "drift upscale" and are not considered "official" for record keeping purposes. They are, however, recorded and used as a check on the accuracy of the official TLD readings. Testimony of Thomas Sowdon, Chief Radiological Scientist, Pilgrim Nuclear Power Station, Tr. 4, at 14, 142-143.
 Thomas Sowdon, Tr. 4, at 16, 145.
 See DeVita, et al., Cancer: Principles & Practice of Oncology (4th ed.1993) 1938. (Exhibit 17).
 There is no (known) radiogenic cancer signature that would allow a clinical determination that a particular cancer was caused by radiation. BEIR V, at 50.
 The BEIR Committee was created by the National Academy of Sciences at the request of the United States government to periodically assess and advise federal authorities as to the hypothetical risks of low doses of radiation. The BEIR assessments are relied upon by the federal government in formulating safety standards protecting workers exposed to radiation. The government also consults the recommendations of the National Committee on Radiation Protection and Measurements [NCRP] and the International Commission on Radiation Protection [ICRP] in forging a consensus. Plaintiff's and defendant's witnesses mutually acknowledged the BEIR reports as the foremost source of reliable scientific opinion on the effects of ionizing radiation.
 One rad for all practical purposes is equal to one rem.
 Excess risk is calculated by epidemiologists in the form of a ratio derived by dividing the number of cases of a disease observed within a defined group by the number of cases expected in the general population. "Epidemiologists generally agree that excess risks of less than 50% are difficult to interpret causally." BEIR V, at 50. Interpretation is complicated by the so-called "small numbers problem" inherent in any study involving a cohort insufficiently large to generate statistically reliable data.
 The BEIR V panel identified two other problems that compel resort to statistical modeling, the need to take into account the many factors other than radiation that are known to contribute to cancer risks, and the fact that "direct estimates of lifetime risk can only be obtained after an exposed population has been followed for a lifetime." BEIR V, at 50.
 The Japanese Life Span Study Cohort consists of a population of approximately 76,000 persons. The ankylosing spondylitis study involved 14,106 patients. Among the 20,000 Japanese survivors who received doses of 1 to 5 rem, no excess cases of any kind of cancer have yet been observed. Dr. David G. Hoel, Tr. 6, at 44.
 The fact that "the biological effects of x-rays or gamma rays decrease with fractionation or reduction in the dose rate" is generally accepted. BEIR V, at 30. BEIR V estimates that a single acute exposure to ionizing radiation, such as that received by the Japanese survivors, is twice as harmful as the same dose level absorbed at low rates over an extended period. BEIR V, at 22-23.
 Second Affidavit of Stuart C. Finch, M.D., September 6, 1990.
 Dr. Hoel was one of the seventeen scientists who served on the BEIR V Committee. He is presently the Chairman of the Department of Biometry and Epidemiology at the Medical University of South Carolina and the Associate Director for Cancer Epidemiology at the Hollings Cancer Center. He is the former Director of the Division of Biometry and Risk Assessment of the National Institute of Environmental Health Sciences (a unit of the National Institute of Health). Dr. Hoel served as a researcher and director in the ongoing study of the incidences of cancer in the Japanese atomic bomb survivors. He has published at least fifteen articles and six books on the topic of radiation and cancer and is a member of the National Academy of Sciences Institute of Medicine. I find his testimony in all respects to be authoritative.
 The most recent study of Japanese atom bomb survivors concludes that there is no statistically significant increase in the incidence of ALL (as opposed to other types of leukemia) at exposure levels of less than 100 rem (although noting that "the small number of ALL cases precluded precise estimation"). Preston, et al., "Cancer Incidence in Atomic Bomb Survivors. Part III: Leukemia, Lymphoma and Multiple Myeloma, 1950-1987," Radiation Research 137 (1994), Tables IV and VI. (Exhibit 9).
 Dr. Shalat acknowledged that he left Yale after his "inadequacy in administrative over-sight" caused "problems" with "a very large, very complex" project. Tr. 2, at 134.
 Dr. Shalat conceded that if Boston Edison's records are correct, no breach of the duty of care can be shown. Tr. 2, at 124-125.
 "Health Physics and dosimetry are complex scientific fields which require years of specialized training, education, knowledge and experience before competency can be achieved. Health Physicists are board certified just as medical doctors are board certified. Health Physics Technicians (also known as HP techs or rad cons) are not professional health physicists.... They have a lower level of knowledge in the field sufficient to allow them to perform rather basic tasks such as reading radiation detection meters. They would be similar to the x-ray technician at the local hospital who is competent to operate the x-ray machine to take the x-ray but who is not competent to interpret the resulting image as would be a radiologist at that local hospital." Affidavit of Dr. John W. Poston, Sr., Head of the Department of Nuclear Engineering, Texas A & M University, May, 1994, paras. 8-12.
 The experience was supposedly gained during employment as a health physics technician at Morrison-Knudsen in 1974-1975, and for a few weeks at Pilgrim in 1976. Although Dr. Shalat testified that he had held a position as a supervisor of radiation dosimetry at Morrison Knudsen, I find credible the testimony of Samuel T. Williams, his supervisor at Morrison Knudsen, that Dr. Shalat never held a position above that of an ordinary health physics technician (performing essentially clerical functions). Dose reconstruction was not a function of any job Dr. Shalat held at Morrison Knudsen. Tr. 3, at 5-8. Similarly, Dr. Shalat's supervisor at Pilgrim, Jose L. Lucero, contradicted Dr. Shalat's testimony that he had served as a supervisor of radiation for Vikem, a Pilgrim contractor. Rather he was hired as a junior technician. Dr. Shalat was barred from access to the Pilgrim plant after disagreements with management over safety issues. Again, dose reconstruction was not part of his job during his brief tenure at Pilgrim. Tr. 3, at 19-23.
 Dr. Shalat, for example, wrongly characterized Cesium 137 and Cobalt 58 as bone marrow seeking beta emitters in attempting to explain why Gary Whiting's actual dose level might have exceeded that registered by his TLD badges. See Second Affidavit of Dr. Stuart L. Shalat, August 2, 1990. Dr. Shalat conceded the error. Tr. 2, at 10. Dr. Shalat also admitted that in his deposition he had inaccurately described the SRPD as a vacuum based device. Tr. 2, at 22.
 In addition to criticizing the accuracy of the TLD itself, Dr. Shalat suggested that the nature of Gary Whiting's work might have exposed his hands and legs to more concentrated doses of radiation than his upper body where he wore his TLD. Tr. 1, at 61-63. 10 C.F.R. § 20.101, however, requires that the TLD be worn on the torso. Dr. Shalat also appears to have misinterpreted an NRC report criticizing Pilgrim for "bad housekeeping" as a reference to Boston Edison's recordkeeping. The NRC was commenting on the untidy appearance of the plant. See Exhibit 69.
 Performance Testing of Personnel Dosimetry Services, Nureg CR-1064. A supplemental report to the study (which Dr. Shalat conceded he had not read) reported that 84 percent of all TLDs analyzed passed the first test; 92 percent passed the second. Tr. 3, at 117.
 "An epidemiologist should not act as his own dose reconstructionist because it is all too tempting to make assumptions which will support the epidemiologist's prior opinion.... A more proper scientific method would be to have the dose reconstructed by a qualified Health Physicist, so that the epidemiologist cannot influence that reconstruction." Second Affidavit of Dr. Glyn G. Caldwell, June 15, 1994, para. 16.
 These are required when plant workers are assigned to tasks in areas with abnormally high radiation levels. In Dr. Shalat's view, Boston Edison's records are incomplete. Tr. 1, at 66-67.
 As Boston Edison points out, the methodology developed by the science of health physics for determining worker doses relies in the first instance on primary data (TLD and SRPD dosimetry readings). When these are unavailable, secondary data, the readings taken from the subject's most proximate co-workers, are utilized. Only in the absence of primary or secondary data is dose reconstruction performed. Tr. 4, at 25-29. Dr. Shalat rejected both of these sources of data because of his mistrust of the accuracy of the TLD and SRPD devices used at Pilgrim and because of his doubts about the integrity of Boston Edison's recordkeeping. Tr. 2, at 123. Plaintiff has not brought to the court's attention any peer support for Dr. Shalat's departure from the accepted methodology of dose reconstruction or any evidence of a convincing nature justifying a lack of confidence in the relevant dose records. Dr. Winters, I note, accepted Boston Edison's records as essentially accurate for purposes of his testimony.
 Dr. Shalat mistakenly recopied the 1977 figures in his calculation for 1979.
 Dr. Shalat stated that he relied on a "summary" prepared by plaintiff's counsel in wrongly cumulating the totals. Tr. 2, at 82-83.
 A hot spot is a radioactive area where the contact dose is six times higher than the ambient dose measured at eighteen inches. Thomas Sowdon, Tr. 4, at 28. The term obviously is a relative one.
 See also Second Affidavit of Dr. Ray D. Lloyd, June 16, 1994, at paras. 15-16. Dr. Lloyd is a Certified Health Physicist.
 As defendant points out, Dr. Shalat's credibility is further undermined by the revisions he has been forced to make over the course of this litigation in his estimate of Gary Whiting's radiation exposure. Dr. Shalat has variously opined 5 to 20 rem (affidavit, July 1990), 5 to 30 rem (deposition, April 26, 1994), and 18.031 rem (hearing, June 6, 1994). Dr. Shalat's fourth dose opinion, offered in an affidavit filed after the hearing to correct the calculation errors he concedes having made in formulating his third opinion, is 18.293 rem. Sixth Affidavit of Dr. Stuart L. Shalat, June 13, 1994.
 Although Dr. Shalat testified that Gary Whiting's ALL was "caused" by radiation exposure at Pilgrim, I find that by training and experience he is not qualified to render an opinion as to whether a specific person's disease is a result of radiation carcinogenesis. (Dr. Shalat claims no medical expertise beyond that possessed by a general epidemiologist.) Epidemiology, as Dr. Shalat acknowledges, does not pretend to make person specific diagnoses; the best it can do is assign a probability (odds) that a particular disease in a particular person was caused by a given agent. Tr. 2, at 33; Affidavit of Dr. Glyn G. Caldwell, May, 1994, para. 15. Dr. Shalat did not attempt such a calculation in Gary Whiting's case. Tr. 2, at 55. Dr. Hoel, who developed the probability of causation tables for leukemia relied upon in BEIR V, testified that any probability of causation calculation below 30 rem to the bone marrow is purely speculative. Tr. 6, at 47-49. I also do not find Dr. Shalat qualified to opine on the subject of general causation. Dr. Shalat has not published on radiation as a cause of leukemia nor has he conducted any studies on the subject. Tr. 2, at 25-26. He has taken three courses touching on the field of radiation carcinogenesis, but has no training in the area of radiation epidemiology. Dr. Shalat's view, that low doses of radiation create an excess risk of cancers, including leukemia, is based not on independent epidemiological evidence but on the linear non-threshold theory propounded by Dr. Winters, see Tr. 2, at 38, although Dr. Shalat is far less conversant with the relevant scientific literature. The discussion of the weaknesses of Dr. Winters' testimony applies with even greater force to Dr. Shalat's.
 Dr. Winters' only published "report" on his research is a February 11, 1982 letter printed in the correspondence section of the New England Journal of Medicine. (Exhibit 27).
 Dr. Winters participated in a similar campaign in Indiana. Tr. 3, at 50.
 Dr. Winters wrote a letter to the editor of the Boston Globe, January 6, 1989, complaining about worker safety at the Pilgrim plant and stating that three workers at Pilgrim had died "probably secondary to their exposure to radiation." (Exhibit 47).
 The affidavit also stated that Cobalt 60 is "generated only in nuclear reactors," a statement that is untrue, as Dr. Winters admitted. Tr. 3, at 135.
 BEIR V did not find evidence of increased cancer risk in studies of populations chronically exposed to elevated levels of natural background radiation. Id. at 385.
 Dr. Winters believes that "some" members of the BEIR V Committee are too biased in favor of nuclear power to be credible. Tr. 5, at 28-29.
 Dr. Winters admits that a link between ionizing radiation and chromosomal mutation is a theory that has not been proven. Tr. 5, at 36-37.
 Dr. Upton, who is Chairman of the Department of Environmental Medicine at New York University Medical Center, was also Chairman of the BEIR V Committee.
 Upton, et al., Radiation Carcinogenesis (1986) 253. (Exhibit 6B). Dr. Upton noted that in the Japanese studies, radiation induced ALL has only "rarely" been observed in survivors over age twenty, while no cases have been observed among ankylosing spondylitis patients, "very few of whom were under age 20 when treated."
 "The increase in credibility in a given diagnosis varies inversely with the belief that the finding will be present before verification." Delp & Manning, Major's Physical Diagnosis: An Introduction to the Clinical Process (9th ed. 1981) 50, Table 3-4. (Exhibit 48).
 Dr. Winters acknowledges that 90 percent of ALL cases are of unknown cause (Tr. 5, at 56), giving added reason to doubt the validity of both his method of "differential diagnosis" and his conclusion that radiation "more likely than not" was responsible for Gary Whiting's illness. Differential diagnosis, as the technique is used in the medical profession, consists of the comparison of a patient's symptoms to symptoms associated with a known set of diseases.The idea is to find the disease that matches the symptoms. If 90 percent of the causes of a disease are unknown, it is impossible to eliminate an unknown disease as the efficient cause of a patient's illness.
 Dr. Winters did not learn from Mrs. Whiting or observe in the medical records of the Beth Israel Hospital that Gary Whiting had a three pack-year smoking history. See Tr. 3, at 105. Cigarette smoke is a known leukemogen. Heath, "Cigarette Smoking and Hematopoietic Cancer," 82 Journal of the National Cancer Institute, No. 23 (December 5, 1990) 1800. (Exhibit 28). Smoking three packs of cigarettes a day for one year exposes the bronchial epithelium to 16 to 32 rem of radiation. Dr. Winters has published at least one article remarking on the association. Dr. Winters also did not consider the fact that Gary Whiting's lifetime exposure to background radiation was approximately two and one-half times the amount he received at Pilgrim, although Dr. Winters professes to believe that background radiation can cause cancer. Dr. Winters also did not consider the possibility that Gary Whiting might have been exposed to benzene in his previous occupation as a carpenter or to a virus. (Dr. Winter's investigation did not include a detailed work history. Tr. 3, at 113-114.) There is substantial evidence for the leukemogenicity of benzene and certain types of viruses. DeVita, et al., Cancer: Principles & Practice of Oncology (4th ed. 1993) 1939. (Exhibit 17).
 Unlike Dr. Shalat, Dr. Winters believes that Boston Edison's records (although in terrible shape) are "essentially" accurate. Tr. 3, at 67. Consequently, he accepted for purposes of his testimony the figure of 6.249 rem as an accurate representation of the whole body dose of radiation Gary Whiting received at Pilgrim. Dr. Winters did testify that Whiting's dose had exceeded federal permissible dose limits in the third and fourth quarters of 1977, but only after plaintiff's counsel erroneously asked him to assume that the federal regulations established a quarterly limit of 1.25 rem. Tr. 3, at 73-74. He testified that Whiting's exposure could have exceeded 3 rem in the fourth quarter of 1977 if his TLD readings were assumed subject to a 10 to 90 percent margin error. Tr. 3, at 75. There is no evidence to support this assumption.
 Plaintiff offered in evidence the report of an "investigation" conducted by the Massachusetts Department of Public Health purporting to show that prior to 1984, persons living near the Pilgrim power station had a higher risk of contracting leukemia than persons living further away. Morris & Knorr, Southeastern Massachusetts Health Study 1978-1986 (1990). (Exhibit 66). I have not given this study any weight. As the authors themselves make clear, the study was marred by problems of self-selection, refusal to participate, possible interviewer bias, inability to screen environmental factors or accurately recreate work histories, and by prejudicial local publicity. (The study attempted to identify subjects by interviewing relatives of persons whose medical records indicated leukemia as the cause of death.) The authors noted that their results were difficult to explain given what other conflicting studies had shown and what is known about leukemia's latency period. Id. at 62-76. As the authors concluded, "[t]he results presented here should be interpreted cautiously, because they fall far short of proving that Pilgrim 1's operation has caused the association demonstrated between residence and worksite location and leukemia incidence." Id. at 77-78.
 For example, the observed latency period between exposure and the expression of ALL at dose levels of 100 rem or above is twenty-one years in adult males ages 20-39. Preston, et al., "Cancer Incidence in Atomic Bomb Survivors. Part III: Leukemia, Lymphoma and Multiple Myeloma, 1950-1987," Radiation Research 137 (1994), Tables IV and VI. (Exhibit 9). Dr. Winters did not give any weight to the fact that Gary Whiting was diagnosed with ALL within five and one-half years of his initial exposure to occupational radiation.
 Plaintiff offered only the Stevens, Wing and Wilkinson studies into evidence.
 Stewart's study of pregnant women dealt with the effects of ionizing radiation on fetal tissue which, as Dr. Winters acknowledged, is particularly susceptible to radiation exposure. Tr. 5, at 91-92.
 According to one of its authors, "[t]he Wing study does not provide scientific data to support a claim that 6.249 rem (or 5 to 30 rem) of occupational radiation causes ALL in adults." Affidavit of Dr. Donna L. Cragle, Director of the Center for Epidemiological Research, Oak Ridge Institute for Science and Education, May, 1994.
 In an affidavit Dr. Caldwell states that National Academy of Sciences studies following up his original research have found no excess rates of leukemias, and that the one case of ALL found in his original study of 3,741 veterans "is not epidemiologically meaningful." Affidavit of Dr. Glyn G. Caldwell, May, 1994. One of these follow-up studies was offered in evidence. Robinette, et al., Studies of Participants in Nuclear Tests, National Research Council (1985). (Exhibit 60).
 Affidavit of Dr. Ray D. Lloyd, co-author of the Stevens study, May 17, 1994.
 Dr. Winters could point to no underlying study in the Wilkinson group that found an increase in ALL in adult male workers. Tr. 5, at 91. Defendant offered two studies published since 1993 that show no incidences of excess leukemia among nuclear plant workers. See Gilbert, et al., "Mortality of Workers at the Hanford Site: 1945-1986," Health Physics 64 (June 1993) (Exhibit 58); Gilbert, et al., "Updated Analyses of Combined Mortality Data for Workers at the Hanford Site, Oak Ridge National Laboratory, and Rocky Flats Weapons Plant," Radiation Research 136 (1993) (Exhibit 59). See also BEIR V, at 380. Dr. Winters agreed that the first of the 1993 Gilbert studies had found no excess risk of leukemia; he had not read the second. Tr. 5, at 96, 98.
 Dr. Winters (and Dr. Shalat) propound the argument that because ALL is extremely rare in adult males, and because Gary Whiting was exposed to radiation before he contracted ALL, his ALL must have been caused by radiation exposure. See, e.g., Tr. 5, at 67. This is a classic illustration of the logical fallacy post hoc ergo propter hoc. It ignores the fact that ALL can occur (and most often does) in adult males who have no history of occupational exposure to radiation, as well as the fact that adult males who are exposed to radiation at levels similar to Gary Whiting's have no higher incidence of ALL than do unexposed adult males.
 Dr. Winters was mistaken in his belief that BEIR V had endorsed a linear non-threshold model for leukemia. Dr. Shalat, I note, made the same error. See Tr. 2, at 11-13.
 The linear quadratic model predicts at low dose levels one-half the excess risk estimated by the linear model. Tr. 2, at 8-9. Dr. Shalat testified, mistakenly, that the two models produce "essentially the same results at low doses." Tr. 2, at 2. In fact the linear quadratic model only intersects on an upward plane with the linear model at high dose levels.
 Neither model, for example, can explain why children under age twenty present eighty percent of ALL cases despite their lower lifetime exposure to radiation.
 "As the name itself notes, this [linear hypothesis] is not proven fact. It is only a hypothesis. Webster's Dictionary tells us that any hypothesis is `a tentative assumption made in order to draw out and test its logical or empirical consequences.' A hypothesis is synonymous with a theory. Consequently, any hypothesis or theory is not fact until it has been scientifically proven. Anyone who has been trained in the scientific method realizes that a hypothesis is a scientist's educated speculation ... It is important to underscore again that a court of law is not a scientific experiment. When a court of law determines responsibility for human suffering and awards damages, it must do so based upon reasonable evidence, not just speculation or hypothesis. Just because scientists use hypotheses to describe something they really don't know for sure does not justify a court of law in using speculative hypotheses to determine that one person has caused harm to another."Johnston v. United States, 597 F.Supp. 374, 393-394 (D.Kan. 1984) [Kelly, J.].