MEMORANDUM & ORDER
DEARIE, District Judge.
Plaintiffs bring this diversity action alleging that Similac infant formula caused Alexis Colon to develop Type 1 diabetes. Plaintiffs seek compensatory and punitive damages based on strict products liability, negligence, design defect, failure to warn, breach of warranty, fraudulent misrepresentation, and loss of services. Defendant moves to exclude plaintiffs’ expert’s opinions and for summary judgment pursuant
to Rule 56(c) of the Federal Rules of Civil Procedure. For the reasons stated below, defendant’s motions are granted.
I. BACKGROUND
Except as otherwise noted, the following facts are not in dispute.
Alexis Colon was born to Carmella Stolyar on January 22, 2000. While in the hospital, Ms. Stolyar fed Alexis both breast milk and Similac infant formula. Def. Rule 56.
Stat’t ¶6. Ms. Stolyar gave Alexis formula the day after she was born because she was crying, and Ms. Stolyar “figured [she] wasn’t producing enough” milk to satisfy her hunger. Stolyar Dep. at 177, Vollins Aff. Ex. 'B.
After leaving the hospital, Ms. Stolyar continued to have difficulty breast-feeding and continued to give Similac to Alexis. According to Ms. Stolyar, her breasts hurt a great deal.
Id.
at 193. Alexis “wasn’t sucking hard enough,” and though Ms. Stolyar would “try to squeeze,” “nothing would be coming out.”
Id.
at 193, 196. Ms. Stolyar notes, however, that using “suction cups” to “[p]ump[] breast milk ... was less painful and less uncomfortable ... than having Alexis drink directly from [her] breast.” Stolyar Aff. ¶ 3, Pl.Ex. B.
One month after giving birth, Ms. Stol-yar stopped breast-feeding entirely. Stol-yar Dep. at'190, Vollins Aff. Ex. B. Ms. Stolyar continued to feed Similac to Alexis and at some point added cereal to her diet. In July 2001, Alexis’s pediatrician diagnosed her with the Coxsackie virus. A few weeks later, eighteen-month-old Alexis was diagnosed with insulin dependent Type 1 juvenile diabetes (“T1D”).
This action followed in March 2003. The Court must decide (1) whether plaintiffs’ expert’s testimony is admissible and (2) whether plaintiffs have raised a genuine issue of material fact as to whether Similac caused Alexis’s T1D.
II. EXPERT TESTIMONY
A. Plaintiffs’ Expert Dr. Jack Newman
Dr. Newman, a pediatrician and “breast feeding advocate,” is a fellow at the Royal College of Physicians in Canada and an Assistant Professor on the Faculty of Medicine at the University of Toronto. PI. Opp. at 30; Newman Aff. ¶¶ 1-3, Pl.Ex. H. Dr. Newman has established three hospital-based breast-feeding clinics in Canada and serves on the staff of the Hospital for Sick Children in Toronto. Newman Aff. ¶¶ 1-3, Pl.Ex. H; Curriculum Vitae, Pl.Ex. I. Dr. Newman is not a pediatric endocrinologist and plaintiffs have provided no evidence that he diagnoses or treats patients with T1D. Dr. Newman initially submitted a “Review of Documents Concerning Alexis Colon” dated January 17, 2004 and supplemented his opinions with an affidavit dated April 11, 2005 in response to defendant’s summary judgment motion.
1. Dr. Newman’s “Review of Documents Concerning Alexis Colon”
According to Dr. Newman, when Alexis was born, “it was certainly well known in the medical community, and in the head offices of formula companies that, at the very least, there was a serious concern that the early introduction of cow milk protein could trigger the development of
diabetes mellitus type I.” Newman Review, PLEx. G. Dr. Newman asserts that “in genetically susceptible individuals, the early exposure to cow milk protein may be the ‘trigger’ for a series of events, which eventually results in the child developing diabetes.”
Id.
Dr. Newman admits, however, that “[i]t was not, at the time of Alexis’ birth, proved beyond a doubt, nor is it now, but there was a very strong possibility, with much theoretical, laboratory, and epidemiological evidence to back up the possibility.”
Id.
Dr. Newman states that by 1984 studies were indicating that early exposure to cow’s milk protein “might increase the risk of a child developing diabetes.”
Id.
According to Dr. Newman, a study conducted in Colorado in 1988 “suggested that breastfeeding could reduce the risk of type I diabetes by 2-26%, depending on several factors,” and a study published in 1992 by the New England Journal of Medicine showed that “early exposure to cow milk protein could indeed cause the immunologic changes in babies which would eventually result in type I diabetes.”
Id.
Dr. Newman also notes that a statement published in 1994 by the American Academy of Pediatrics concluded that “the prudent approach to preventing diabetes type I would be to avoid early exposure to cow [sic] (and based on other studies, even soy protein).”
Id.
Dr. Newman argues that from birth until three months of age, infants have “leaky” intestines which allow absorption of whole milk proteins into their bodies.
Id.
Breast milk speeds up gut closure and helps prevent the absorption of whole proteins, Dr. Newman argues, whereas early introduction of cow’s milk damages the gut lining.
Id.
According to Dr. Newman, “bovine serum albumin (BSA), a component of cow’s milk ... present in infant formulas, can enter [a] baby’s body through the gut. BSA is similar to human albumin, but not identical.”
Id.
The body perceives a series of seventeen amino acids (termed ABB OS) in BSA as “foreign” and, particularly in children genetically susceptible to diabetes, produces an antibody to this chain.
Id.
Dr. Newman explains that “[i]n the presence of interferon []stimulated by a viral illness ..., an antigen (p69) is induced on the [beta] cell of the pancreas (the cell which produces the insulin) and the antibody to ABBOS then begins to cross react with this antigen p69 causing damage to the [beta] cell.”
Id.
When enough beta cell destruction occurs, the child becomes diabetic.
Id.
Dr. Newman concludes that “[i]t is likely that the earlier the introduction of cow milk protein (assuming it does trigger diabetes), the earlier the manifestations of diabetes would arise.”
Id.
2.
Dr. Newman’s Affidavit
In response to defendant’s expert reports, Dr. Newman submits an affidavit dated April 11, 2005, which claims that the non-hydrolyzed cow’s milk in Similac causes T1D. Dr. Newman suggests that “some infant formulas on the market today utilize hydrolyzed (broken down) proteins to avoid, among other things, later development of T1D.” Newman Aff. ¶ 10, Pl.Ex. H. Dr. Newman asserts that “the early introduction of Similac [i]nfant formula was a substantial factor in causing Alexis Colon to develop T1D at 18 months of age,” arguing that “the early introduction of non-hydroly[zed] cow’s milk in Similac triggered an immune response to the whole proteins[,] ... [which] rapidly developed into T1D after Alexis Colon ... developed the Coxsackie virus....”
Id.
¶ 6.
Dr. Newman observes that “[t]he question here is not the
exact mechanism
of how early introduction of infant formula causes diabetes. The whole theory of T1D
being an autoimmune disease can be completely disproved without affecting the issue here. The . question remains whether lack of breast feeding and early introduction of foreign protein, such as that found in Similac[,] [sic ] increases the risk of T1D and if so, why?”
Id.
¶ 19. Dr. Newman acknowledges that “[regarding the etiology of T1D, the emphasis has shifted to derangement of the gut mucosal system” and states “there is little debate that the artificially fed infant has an immaturity of the gut mucosal system, which includes a delay in gut closure due to lack of breast milk provided factors ... and the likely result is an aberrant or impaired response to dietary foreign proteins such as [those] found in [non-hydrolyzed] cow’s milk formula.”
Id.
¶ 20.
Dr. Newman further suggests that an ongoing multinational study, the Trial to Reduce Insulin Dependent Diabetes in the Genetically at Risk (“TRIGR”), supports his view. He explains that the goal of the TRIGR study “is to determine whether weaning to a hydrolyzed formula, where the milk protein is broken down, in place of cow’s milk based infant formula, (with intact protein), during the first 6-8 months of life reduces the incidence of T1D in genetically susceptible populations.”
Id.
¶ 11. According to Dr. Newman, “TRIGR is supported by studies that continue to be published showing an association between the early introduction of formula and the development of diabetes.”
Id.
¶ 15. Dr. Newman cites a 1997 study showing “á strong relationship” between the development of T1D and the introduction of infant formula in the first eight days of life, as well as a 2004 study in Sweden and Lithuania in which “longer[,] exclusive and total breast feeding appears as an independent factor against T1D.”
Id.
¶¶ 15-16 (internal quotations omitted).
Dr. Newman contends that most U.S. studies purporting to use exclusively breastfed children as a comparison group are flawed since hospitals routinely provide formula as a supplement.
Id.
¶ 17. He further notes that in 2005 the American Academy of Pediatrics revised an advisory from 1994 to emphasize the benefits of breast-feeding, particularly the “possible protective effect of human milk feeding against ... diabetes.”
Id.
¶¶ 21-23 (internal quotations omitted).
B. Defendant’s Experts
Defendant presents the testimony of the following four experts to discredit the theories advanced by plaintiffs’ expert.
1. Dr. Jill M. Norris
Dr. Norris, a professor in the Department of Preventive Medicine and Biometrics at the University of Colorado Health Sciences Center, has conducted extensive epidemiological research on T1D, focusing on early childhood risk factors. Norris Report ¶¶ 1-3, Vollins Aff. Ex. F. In 1994, she received funding from the National Institutes of Health to conduct the Diabetes Autoimmunity Study in the Young (“DAISY”).
Id.
¶¶ 2-3. According to Dr. Norris, Dr. Newman’s suggestion that Alexis Colon developed diabetes from infant formula “is unproven speculation,” and “recent studies suggest that the alleged association between early introduction of formula and type 1 diabetes is extremely doubtful.”
Id.
¶ 4.
Dr. Norris states that a mother’s inability to recall exactly how long she breast-fed and when she first used formula affects the accuracy of retrospective case-control studies, such as the Colorado study on which Dr. Newman relies.
Id.
¶¶ 5-7. In a retrospective case-control study, mothers of children with diabetes and mothers of children without diabetes are asked when
they first fed their children formula and how long they breast-fed.
Id.
¶ 7. According to Dr. Norris, the 1988 study initially found that “cases (children with diabetes) were 30% less likely to have been breastfed than controls (children without diabetes).”
Id.
¶ 5. However, once the authors limited the population studied to mothers with children five years old or younger to account for problems with long-term recall, they found that “the observed protective effect of breast-feeding was not detectable.”
Id.
As a result, the authors of the 1988 study concluded that “replication of their findings in different populations was critical before the hypothesis of a .protective effect of breast-feeding against type 1 diabetes should be accepted.”
Id.
According to Dr. Norris, since 1988 twenty-three additional retrospective studies have been conducted. Of these studies, two-thirds failed to reveal evidence that breast-feeding protected against T1D in any statistically significant way, and the majority “did
not
find that early introduction of formula increased [the] risk of type 1 diabetes.”
Id.
¶ 6. Dr. Norris emphasizes that the “lack of consistency across studies argues against the hypothesis that lack of breast-feeding or early introduction of formula increases risk of type 1 diabetes.”
Id.
Dr. Norris contends that “prospective” studies such as the DAISY study, which follow children with an increased risk for T1D from birth, eliminate recall bias and ensure higher accuracy of data than retrospective studies.
Id.
¶¶ 12-14. One of the prospective studies cited by Dr. Norris— the Diabetes Prediction and Prevention (“DIPP”) project — follows over 7,100 children in Finland and is “the largest prospective study of diabetes autoimmunity and type 1 diabetes in the world.”
Id.
¶¶ 13-14. Dr. Norris explains that these prospective studies have found no association between the development of diabetes autoimmunity
and early exposure to cow’s milk. According to Dr. Norris, these studies have, however, found that introducing cereals prior to the age of four months increases the risk of diabetes autoimmunity.
Id.
¶ 13.
In her reply affidavit, Dr. Norris adds that the TRIGR study, which began in June 2002, has not produced any reliable data. Norris Reply Aff. ¶ 2, Wilson Aff. Ex. A. In Dr. Norris’s opinion, “the very presence of this [study] suggests that there is not enough evidence to say that early introduction of cow’s milk based infant formula causes type 1 diabetes.”
Id.
As Dr. Norris notes, when TRIGR study investigators published data earlier this year from a pilot trial, they emphasized that “at present there is no justification to change normal infant feeding practices.”
Id.
(internal quotations omitted) (citation omitted). Dr. Norris concludes that “the link between breastfeeding [and type 1 diabetes] is still extremely doubtful, and certainly remains unproven.”
Id.
¶ 5.
2. Dr. Mark Atkinson
' Dr. Atkinson is the Sebastian Family Eminent Scholar at the University of Florida College of Medicine and also serves as director of the University’s Center of Immunology and Transplantation, co-Director of the Diabetes Center of Excellence, and Professor of Pathology, Immunology, and Laboratory Medicine. Atkinson Report ¶ 1, Vollins Aff. Ex. G. For five years, Dr. Atkinson chaired the immunology and genetics study section of the Juvenile Diabetes Research Foundation.
Id.
¶ 3. Since
July 2003, Dr. Atkinson has served as chair of the Medical Science Research Committee of the Juvenile Diabetes Research Foundation.
Id.
Dr. Atkinson opines that “the overall lack of consistency and reproducibility of the research backing [Dr. Newman’s] hypothesis renders [his] opinions unreliable and speculative.”
Id.
¶ 5. According to Dr. Atkinson, Dr. Newman’s notion of molecular mimicry between the beta cell protein of the pancreas and ABBOS is “out of date, unsupported, and well refuted.”
Id.
¶¶ 6-7. The idea of molecular mimicry originated after a 1992 study found that “humans with type 1 diabetes were reported to uniformly show immune system reactivity to BSA and the ABBOS peptide.”
Id.
¶ 7. But follow-up studies, including one by the senior author of the 1992 study, failed to reproduce the findings of the original report.
Id.
¶ 8. Dr. Atkinson concludes that the results of two recent prospective studies “failed to support a linkage between breastfeeding/infant formula introduction and type 1 diabetes.”
Id.
Dr. Atkinson is unpersuaded by Dr. Newman’s unexplained shift away from molecular mimicry to the gut mucosal system as the basis for the link between T1D and infant formula. In his reply affidavit, Dr. Atkinson notes that Dr. Newman’s opinions regarding the “derangement of the gut mucosal system” rely on a 1999 “opinion piece,” not on any data proving a causal relationship between T1D and cow’s milk. Dr. Atkinson emphasizes that the 1999 article clárified that “[available data do not allow the conclusion that cow’s milk itself is pathogenic in type 1 diabetes.” Atkinson Reply Aff. ¶ 6, Wilson Aff. Ex. D.
3. Dr. Ronald E. Kleinman
Dr. Kleinman is Professor of Pediatrics at Harvard Medical School and Associate Chief of Pediatrics at Massachusetts General Hospital. Kleinman Report ¶ 1, Vol-lins Aff. Ex. H. From 1989 until 1993, Mr. Kleinman served as Chairman of the American Academy of Pediatrics (“AAP”) Committee on Nutrition.
Id.
¶ 3.
Citing the recommendations of the AAP’s Work Group on Cow’s Milk Protein and Diabetes Mellitus from the early 1990s, Dr. Kleinman explains that “commercial infant formulas utilizing cow’s milk protein remain the approved alternative]” to breast milk.
Id.
¶¶ 11-12. According to Dr. Kleinman, the AAP’s Pediatric Nutrition Handbook from that same period states that “[b]eeause of [the] conflicting data, the relationship between type 1 diabetes and early ingestion of cow milk formula is unclear.... Current data are too preliminary and not conclusive.”
Id.
¶ 15. According to Dr. Kleinman, results of recent studies “cast even further doubt on the alleged relationship between milk-based infant formula and T1D.”
Id.
¶ 16. For example, the DAISY study found “no association between exposure to cow’s milk and risk of [diabetes autoimmunity] overall or in any of the risk subgroups.”
Id.
¶¶ 17-18. The findings of the prospective BABYDIAB study, discussed in the October 2003 Journal of the American Medical Association, also “argue against a role for cow’s milk in promoting islet autoimmunity, and even suggest that early exposure may be beneficiál in selected individuals.”
Id.
As a result, Dr. Kleinman states that “any theory alleging a link between milk-based infant formula and T1D is not con
sistent with the current opinion of the relevant medical community.”
Id.
¶ 19.
In his reply affidavit, Dr. Kleinman notes that the AAP’s February 2005 Policy Statement “[i]n no way ... concluded] that milk-based infant formula has been found by the medical community to cause TID.” Kleinman Reply Aff. ¶¶ 3-4, Wilson Aff. Ex. C. Although “some studies suggest that there is a reduction of TID in older children and adults who were breastfed, compared with individuals who were not breastfed,” the AAP states that additional research is necessary.
Id.
(internal quotations omitted). Dr. Kleinman adds that the ongoing TRIGR study does not alter his opinion.
Id.
¶ 5.
4. Dr. Robert Rapaport
Dr. Rapaport is Professor of Pediatrics and Pediatric Endocrinology and Diabetes at the Mount Sinai School of Medicine and Chief of the Division of Pediatric Endocrinology and Diabetes at Mount Sinai. Ra-paport Report ¶¶ 1-2, Vollins Aff. Ex. I. He is also on the Medical Advisory Board of the Juvenile Diabetes Research Foundation, New York Chapter.
Id.
¶ 3.
According to Dr. Rapaport, the medical literature does not provide “a reliable basis” for “the contention that exposure to cow’s milk formula causes TID. There is therefore no support for the claim that exposure to cow’s milk formula caused Alexis Colon’s TID. At most, it is just a theoretical possibility, similar to many other unproven medical hypotheses about environmental causes of a whole host of diseases.”
Id.
¶ 15. Dr. Rapaport points to the results of the prospective BABYDIAB and DAISY studies — the “most recent significant studies on the subject”-which “did not find that there was an association between milk-based infant formula and au-toantibodies to TID.”
Id.
¶ 16. Dr. Rapa-port also observes that the AAP, of which he is a member, “has found that there is insufficient evidence to indicate that infant formula causes TID” and “that commercial infant formulas remained the approved alternate to breastmilk.”
Id.
Dr. Rapaport notes that Alexis was diagnosed with the Coxsackie virus, which has “been implicated as a causative factor for diabetes,” and that she was exposed at an early age to cereal, which may also be associated with TID.
Id.
¶¶ 17-18. Dr. Rapaport states that Dr. Newman “is simply incorrect when he opines that the Coxsackie virus alone could not be responsible for Alexis Colon’s TID.” Rapaport Reply Aff. ¶ 4, Wilson Ex. B. According to Dr. Rapaport, “any number of triggering viruses, including the Coxsackie virus, could have contributed to the development of Alexis Colon’s TID.”
Id.
Dr. Rapaport further observes that Dr. Newman refers to the concept of a “genetically susceptible child” but does not consider whether Alexis Colon was such a child.
Id.
¶ 6. Dr. Rapaport found “no indication that Alexis Colon would have been considered ‘genetically susceptible’ [to][ ] TID.”
Id.
According to Dr. Rapaport, if sufficient evidence supported Dr. Newman’s hypothesis regarding a link between non-hydrolyzed cow’s milk formula and TID, the TRIGR study would be unnecessary.
Id.
¶ 5. Moreover, Dr. Rapaport contends that
“it would be unethical to continue to enroll infant patients randomly in an arm of [a] study that would expose them to such an alleged diabetogenic agent.”
Id.
Dr. Ra-paport adds that if Dr. Newman’s theory about the link between exposure to formula and T1D were correct, then Sweden— where 97% of babies are exclusively breastfed on discharge from the hospital— would have a very low incidence of T1D. But studies show, on the contrary, that Sweden has one of the highest incidence rates of T1D in the world.
Id.
¶ 7. Dr. Rapaport concludes that “milk-based infant formula is a reasonable alternative to breastfeeding where the mother, as here, considered and attempted breastfeeding. There is no reason to consider that any different nutritional advice should have been provided for the care of Alexis Colon.” Rapaport Report ¶ 19, Volitas Aff. Ex. I.
III. DISCUSSION
A. Summary Judgment Standard
Summary judgment is appropriate “if the pleadings, depositions, answers to interrogatories, and admissions on file, together with the affidavits ... show that there is no genuine issue as to any material fact and that the moving party is entitled to a judgment as a matter of law.” Ped.R.Civ.P. 56(c). “Only when no reasonable trier of fact could find in favor of the nonmoving party should summary judgment be granted.”
White v. ABCO Eng’g Corp.,
221 F.3d 293, 300 (2d Cir. 2000) (quoting
Taggart v. Time, Inc.,
924 F.2d 43, 46 (2d Cir.1991)). The court must draw all inferences in favor of the party opposing the motion.
Cronin v. Aetna Life Ins. Co.,
46 F.3d 196, 202 (2d Cir. 1995). If, after drawing all inferences, the court finds that an issue of material fact remains, the motion for summary judgment must be denied.
When expert testimony is inadmissible, “the summary judgment determination is made on a record that does not include that evidence.”
Cacciola v. Selco Balers, Inc.,
127 F.Supp.2d 175, 180 (E.D.N.Y.2001) (citing
Raskin v. Wyatt Co.,
125 F.3d 55, 66-67 (2d Cir.1997)). If plaintiffs’ expert opinions are admissible, the Court “remains free to ... grant summary judgment for defendant,” where the evidence is insufficient to permit a rational juror to find in favor of the plaintiff.
Amorgianos v. Nat’l R.R. Passenger Corp.,
303 F.3d 256, 267 (2d Cir.2002).
For the reasons discussed below, the record — even with plaintiffs’ expert opinions — fails to provide a basis for a reasonable juror to conclude by a preponderance of the evidence that defendant’s infant formula caused Alexis Colon’s diabetes.
B. Admissibility of the Testimony of Plaintiffs’ Expert Dr. Newman
The trial judge has broad discretion regarding the admissibility of expert testimony.
See McCullock v. H.B. Fuller Co.,
61 F.3d 1038, 1042 (2d Cir.1995). Under Rule 702 of the Federal Rules of Evidence, testimony must be (1) “based upon sufficient facts or data” and (2) “the product of reliable principles and methods.” Fed.R.Evid. 702. The expert must also apply “the principles and methods reliably to the facts of the case.”
Id.
The trial court acts as a gatekeeper to ensure that any technical or scientific expert testimony admitted is both relevant and reliable.
See Kumho Tire Co. v. Carmichael,
526 U.S. 137, 141, 119 S.Ct. 1167, 143 L.Ed.2d 238 (1999);
Daubert v. Merrell Dow Pharm., Inc.,
509 U.S. 579, 589, 113 S.Ct. 2786, 125 L.Ed.2d 469 (1993);
Nimely v. City of New York,
414 F.3d 381, 395-97 (2d Cir.2005). The trial court must analyze whether the proffered
expert testimony “ha[s] any tendency to make the existence of any fact that is of consequence to the determination of the action more probable or less probable than it would be without the evidence.”
Amorgianos,
303 F.3d at 265 (internal quotations omitted). Scientific evidence requires “more than subjective belief or unsupported speculation.”
Daubert,
509 U.S. at 590, 113 S.Ct. 2786.
Defendant repeatedly emphasizes that Dr. Newman is neither an epidemiologist nor an immunologist and therefore does not have the education or experience to support his opinions. This observation alone is sufficient to cast considerable doubt on the reliability of Dr. Newman’s opinions. But even assuming that Dr. Newman is a qualified expert, his speculative opinions do not meet the requirements of Rule 702 and
Daubert
and therefore must be excluded.
See, e.g., Raskin,
125 F.3d at 67 (affirming summary judgment where expert opinions were “scarcely ‘helpful’ to a jury within the meaning of [Rule] 702”);
Cacciola,
127 F.Supp.2d at 184 (excluding expert testimony and granting summary judgment where opinion evidence was based on “unsubstantiated generalizations, speculative hypotheses and subjective evaluation”) (citations omitted).
1. Dr. Newman’s General Theory of Causation
Dr. Newman’s general theory that T1D is linked to milk-based infant formula is speculative. He merely asserts that “early exposure to cow milk protein
may be
the ‘trigger’ for a series of events, which eventually results in [a genetically susceptible] child developing diabetes.” Newman Review, Pl.Ex. G (emphasis added). He claims that there is a “very strong possibility” that infant formula causes T1D and that there is “theoretical, laboratory, and epidemiological evidence to back up the possibility.”
Id.
But Dr. Newman’s opinions are not based on sufficient facts or data as required by Rule 702. Fed. R.Civ.P. 702;
see McClain v. Metabolife Int’l, Inc.,
401 F.3d 1233, 1247 (11th Cir. 2005) (noting that the expert’s “inclination to draw overreaching conclusions from self-limiting medical articles[ ] show[s] the speculative nature of his opinions”).
Dr. Newman relies on a 1988 retrospective epidemiological study but ignores the problems of bias and inaccuracy associated with such studies. Dr. Newman also cites an immunological study from 1992, but he fails to note that repeated attempts to replicate its findings, including one study by the senior author of the article, have failed. Atkinson Report ¶ 8, Vollins Aff. Ex. G. Although Dr. Newman notes that 97% of babies in Sweden are exclusively breastfed on discharge from the hospital, he does not address the fact that Sweden has one of the highest incidence rates of T1D in the world. Rapaport Reply Aff. ¶ 7, Wilson Aff. Ex. B.
Dr. Newman fails to provide a reliable explanation for his theory that cow’s milk infant formula and T1D are linked. He initially espoused the bovine serum albumin (BSA) mimicry theory as one possible explanation. Defendant’s experts, however, emphasized that the scientific community has repudiated this theory and the evidence before the Court supports this view. In response, Dr. Newman shifted his focus away from the molecular mimicry theory to the gut mucosal system, citing a single “opinion piece” from 1999 in his April 2005 affidavit.
Plaintiffs contend that “the mechanism of toxicity need not be known to support an inference of causation.” Pl. Opp. at 37-38 (quoting
Golod v. Hoffman La Roche,
964 F.Supp. 841, 858 (S.D.N.Y.1997)). This argument is unavailing, however. Unlike the expert opinions in
Golod,
Dr.
Newman’s inference of causation is not based on “accepted clinical methods of diagnosis.” 964 F.Supp. at 858. Significantly, the article on which Dr. Newman relies cautions that “[a]vailable data do not allow the conclusion that cow’s milk itself is pathogenic in type I diabetes.” Def. Reply Mem. at 8; Atkinson Reply Aff. ¶ 6, Wilson Aff. Ex. D. Accordingly, Dr. Newman has not set forth a reliable basis for his opinions as required under
Daubert,
509 U.S. at 597, 113 S.Ct. 2786;
see also Meister v. Med. Eng’g Corp.,
267 F.3d 1123, 1132 (D.C.Cir.2001) (affirming judgment as a matter of law where plaintiffs’ experts were unable to show that breast implants caused scleroderma and epidemiological evidence failed to show a causal relationship).
Dr. Newman’s opinion regarding hydrolyzed versus non-hydrolyzed milk, set forth for the first time in his April 2005 affidavit, is also unreliable.
Dr. Newman opines that hydrolyzed formula may prevent T1D and may provide a safer alternative to non-hydrolyzed milk-based formula. Yet Dr. Newman provides no support for this hypothesis other than the incomplete TRIGR study. It is well established that “[i]n analyzing the reliability of an expert’s testimony, the ‘key question’ is Whether it can be (and has been) tested.’ ”
Colon v. BIC USA, Inc.,
199 F.Supp.2d 53, 75 (S.D.N.Y.2001) (quoting
Daubert,
509 U.S. at 593, 113 S.Ct. 2786);
see also Rypkema v. Time Mfg. Co.,
263 F.Supp.2d 687, 692-93 (S.D.N.Y.2003). As a result, this opinion too must be excluded.
See Ruffin v. Shaw Indus., Inc.,
149 F.3d 294, 297-300 (4th Cir.1998) (affirming exclusion of expert testimony where expert’s method was not independently replicated or widely accepted in the scientific community).
Furthermore, Dr. Newman’s critique of recent scientific findings that contradict his opinions is unpersuasive.
See Norris v. Baxter Healthcare Corp.,
397 F.3d 878, 884-85 (10th Cir.2005) (affirming summary judgment where plaintiffs experts failed, to explain “the many epidemiological studies which found no medically reliable link between silicone breast implants and systemic disease”). Dr. Newman argues that the prospective BABYDIAB and DAISY studies are of limited value due to their small sample sizes, citing a criticism of the studies set forth in an editorial by defendant’s expert Dr. Atkinson in 2003. Dr. Atkinson’s editorial, however, also clearly states that both studies “fail to provide support for the hypothesis that early exposure to cow’s milk might be responsible” for T1D. Atkinson Reply Aff. ¶ 4, Wilson Aff. Ex. D (internal quotations omitted). Moreover, the editorial emphasizes that “given current knowledge ... current infant feeding guidelines should not be changed.” M. Atkinson
&
E.A.M. Gale, Editorial, 290 Journal of the American Medical Association 1771 (2003), Wilson Aff. Ex. D.
Dr. Newman also claims that the DAISY study mischaracterized infants who were fed formula in U.S. hospitals as exclusively breastfed. In her reply affidavit, defendant’s expert Dr. Norris clarifies that, contrary to Dr. Newman’s assumption, -the study in fact accounted for exposure to formula in U.S. hospitals. Norris Reply
Aff. ¶ 4, Wilson Aff. Ex. A. In the DAISY study mothers reported any infant formula use in the hospital, and those children who used formula were not classified as exclusively breastfed.
Id.
Therefore, Dr. Newman’s critiques of these studies are simply unfounded.
Because Dr. Newman fails to provide a reliable basis for his opinion regarding an alleged association between cow’s milk infant formula and T1D, his theory of general causation must be excluded.
2. Dr. Newman’s Theory of Specific Causation
Dr. Newman cannot establish that Similac caused Alexis’s T1D. Dr. Newman is not a pediatric endocrinologist and has provided no indication that he diagnoses or treats patients with T1D. Even if Dr. Newman were qualified to provide opinions as to specific causation, a significant analytical gap remains between the possibility that infant formula may cause T1D and Dr. Newman’s conclusion that Similac was actually a substantial factor in causing Alexis’s diabetes.
See Amorgianos,
303 F.3d at 266 (“ ‘A court may conclude that there is simply too great an analytical gap between the data and the opinion proffered.’ ”) (quoting
Gen. Elec. Co. v. Joiner,
522 U.S. 136, 146, 118 S.Ct. 512, 139 L.Ed.2d 508 (1997));
Rosen v. Ciba-Geigy Corp.,
78 F.3d 316, 319 (7th Cir.1996) (“[T]he courtroom is not the place for scientific guesswork, even of the inspired sort. Law lags science; it does not lead it.”).
Statements by Dr. Newman that “[t]here is no doubt that to develop type I diabetes at 18 months of age is a much younger age than ... usual;” “[i]t is likely that the earlier the introduction of cow milk protein (assuming it does trigger diabetes), the earlier the manifestations of diabetes would arise;” and “[t]he early introduction of non-hydrolysated cow’s milk in Similac triggered an immune response ... [which] rapidly developed into T1D” are insufficient to establish specific causation. Newman Review, Pl.Ex. G; Newman Aff. ¶ 6, Pl.Ex. H. A district court is not required to admit opinion evidence which is “ ‘connected to existing data only by the
ipse dixit
of the expert.’ ”
See, e.g., In re Rezulin Prods. Liab. Litig.,
309 F.Supp.2d 531, 540 (S.D.N.Y.2004) (quoting
Joiner,
522 U.S. at 146, 118 S.Ct. 512);
Gilks v. Olay Co.,
30 F.Supp.2d 438, 443 (S.D.N.Y.1998) (“Mere use of the product and subsequent injury ... are not a sufficient basis from which to infer causation.”).
Contrary to plaintiffs’ contentions, statements by the AAP’s Work Group on Cow’s Milk Protein and Diabetes Mellitus do not support Dr. Newman’s opinions. Plaintiffs point to the Work Group recommendation from the early 1990s that “strongly encourag[ed]” families with a “strong history” of T1D to breast-feed and avoid cow’s milk protein during the first year of life. Plaintiffs emphasize that defendant’s expert Dr. Kleinman includes this Work Group recommendation in his expert report and contend that his report supports their argument. But plaintiffs fail to acknowledge that Alexis Colon does not have a “strong” family history of T1D. Defendant points out that although Alexis’s great-grandfather may have had either type 1 or type 2 diabetes, Alexis would still not be considered genetically susceptible to T1D.
See
Kleinman Report ¶ 13, Vollins Aff. Ex. H; Rapaport Reply Aff. ¶ 6, Wilson Aff. Ex. B. Dr. Newman cannot and does not make any claim to the contrary.
Dr. Newman’s opinions as to specific causation are speculative.
See Rosen,
78 F.3d at 319-20 (granting summary judgment where expert testified that a nicotine patch caused a heart attack without recognizing other potential causes of the heart attack, including high blood pressure and high fat diet). Under
Daubert,
scientific conclusions must be “ ‘more than subjective belief or unsupported speculation.’ ”
McCullock v. H.B. Fuller Co.,
61 F.3d 1038, 1043-44 (2d Cir.1995) (quoting
Daubert,
509 U.S. at 590, 113 S.Ct. 2786). Dr. Newman has failed
to
provide a reliable basis for his conclusion that Similac caused Alexis’s T1D. Accordingly, his opinions must be excluded.
C. Plaintiffs’ Causation Theory
Resolution of this summary judgment motion does not, however, require the exclusion of Dr. Newman’s opinions. Even if admitted, the opinions would not permit a fact finder to conclude that it is more likely than not that Similac causes Type 1 diabetes. Beyond Dr. Newman’s speculative opinions, plaintiffs’ causation theory rests only on short abstracts and articles which describe, at best, a tenuous connection between cow’s milk and T1D.
See
PI. Exs. O & P.
For example, an article plaintiffs submit from November 1999 notes that “[d]ue to conflicting reports in the literature, any association between the early ingestion of cow’s milk products and IDDM is controversial and unresolved.” Pl.Ex. O.
An article from 2001 also states that “the causal role of breastfeeding or of early exposure to cow’s milk in the destruction of beta cells remains controversial.” M
An article published in 2000 observes that focusing on early exposure to cow’s milk as a trigger of T1D is “most likely an oversimplification.” M
At oral argument, plaintiffs cited an article titled “Ischaemic Heart Disease, Type 1 Diabetes and Cow Milk Al Beta Casein,” published in the January 24, 2003 New Zealand Medical Journal. Plaintiffs emphasized that the article confirmed a direct correlation between Al beta casein consumption and T1D.
See
May 31, 2005 Hrg. Tr. at 31-33. Plaintiffs’ unequivocal statement embraced by counsel at oral argument fails in fact to deliver the promised
punch. The article states that “extreme caution is required in interpreting correlations involving dietary factors” and “[c]or-relation, even when statistically significant, does not prove causation, though it may raise th[e] possibility.” Pl.Ex. O.
According to the authors, the “main value [of this study] [wa]s to focus further research, and [the study] [wa]s not by itself a basis for public policy.”
Id.
Although future research may show a link between T1D and cow’s milk formula, plaintiffs cannot, at this time, establish a scientific basis for their arguments.
IV. CONCLUSION
After a review of the literature before the Court and the interpretations and commentaries of the opining physicians, including Dr. Newman, it cannot now be said that a causal relationship has been demonstrated. What the scientific community will in the course of time prove or disprove remains to be seen, but asking a lay jury to do what science has yet to do, and may never do, would be clearly inappropriate. It may well be, as defendants’ experts suggest, that science has overtaken plaintiffs’ theory of causation; or, perhaps, the ongoing studies will provide critical evidence to establish a connection. But the trial court is no place for pure theory, hypothesis, or even sincerely held opinion. And, for the moment at least, that is all plaintiffs are in a position to offer. Accordingly, the defendant is entitled to judgment in its favor. The Clerk of the Court is directed to close the case.
SO ORDERED.