Opinion
Case No: 8:04-MD-2523-T-30TBM, MDL 1626.
May 2, 2007
ORDER
THIS CAUSE comes before the Court upon the Defendant Hoffmann-LaRoche, Inc.'s and Roche Laboratories Inc.'s Motion to Exclude Evidence Regarding Causality Assessments (Dkt. #408) and Plaintiffs' response to the same (Dkt. #450). Upon review and consideration and for the reasons detailed below, this Court finds that Defendants' Motion should be granted.
Defendants seek to exclude any evidence relating to "causality assessments." Plaintiffs contend this evidence is an admission by Defendants that the drug Accutane causes inflammatory bowel disease (hereinafter "IBD") and notice to Defendants of the potential harm caused by the drug. Defendants assert such evidence violates Rules 402, 403, and 702 of the Federal Rules of Evidence.
BACKGROUND
Defendants are a pharmaceutical company. In the research and development of their medications, Defendants conduct extensive post-marketing safety surveillance of its prescription medications, including Accutane. This surveillance includes collecting information regarding reported adverse drug experiences (hereinafter "ADRs") which are possibly connected to the drug, and providing that information to the Food and Drug Administration (hereinafter "FDA"). While post-marketing surveillance is increasingly becoming an important component in the research and evaluation process of many pharmaceutical companies, the FDA does not require domestic companies to conduct or submit causality assessments. Certain European countries, however, do require assessments of possible drug/event relatedness, and because Defendants and its affiliates distribute Accutane in these European countries, Defendants conduct causality assessments at all of its companies and affiliates, foreign and domestic. Despite their name, these causality assessments are not used to assess actual causation, but for regulatory reporting and safety surveillance.
Defendants input data from the ADRs into their ADVENT database. This database contains various fields which are completed for each reported adverse event. These fields include reporter (the person who submitted the ADR) assessment of relatedness, company assessments of relatedness and CIMOS II.
CIMOS II is the narrative portion of the database wherein Defendants' drug safety physicians provide their overall assessment of whether there is a possible relationship between the drug and the adverse event.
Reporter Assessments of Relatedness
All ADRs contained in the ADVENT database originate from a reporter. This reporter may be a physician, a patient, a family member, or even a plaintiff's lawyer. The ADRs are received by Defendants through a call center and may be in the form of a letter, telephone call, or news report. Upon receiving the ADR, Defendants record the reporter's assessment of the potential relationship between Accutane and the adverse event in Defendants' database. A response of "yes" is entered in the reporter relatedness field if the reporter says there is a possibility the ADR was related to Accutane (or any other drug Defendants manufacture). For example, if a mother of a child who was taking Accutane calls the call center and reports her belief that her son's taking of Accutane caused his IBD, the mother's assessment would be reported in the ADVENT database and a "yes" would be listed in the reporter relatedness field.
Company Assessments of Relatedness
In addition to the reporter assessments of relatedness, the ADVENT database also contains a field which reflects Defendants' assessment of relatedness between the drug and the ADR. A "yes" is inserted in the field if there is a reasonable probability that the adverse event is drug related. "Yes" is also automatically entered into the company relatedness field if the reporter evaluated the relationship as a "yes." The only exception to this policy is if the drug safety specialist provides a specific explanation for over-riding the reporter's "yes" evaluation.
Prior to 1997, Defendants used one of five terms in the company causality assessment field:
* Not Related: indicated that the company physician felt there was not a reasonable possibility that the adverse event reported was drug related.
* Remote/Unlikely: indicated that the company physician felt there was a remote or distant possibility that the adverse event reported was drug related.
* Possible: indicated that the company physician felt there was a likely or potentially reasonable possibility that the adverse event that was reported was drug related.
* Probable: indicated that the company physician felt there was a presumably reasonable possibility that the adverse event reported was drug related.
* Highly Probable: indicated that the company physician felt there was a possibility that the adverse event reported was drug related.
After 1997, however, Defendants ceased use of the above five referenced categories, and began using "yes/no" responses in the company relatedness field.
In or around late 2000 to early 2001, Defendants adopted the "Naranjo" algorithm to complete the company relatedness field. After adoption of the Naranjo algorithm, Defendants' company relatedness field was based completely on the reporter's assessment and the algorithm score. For example, if the reporter assessed causality as either "yes" or "unknown," Defendants automatically entered "yes" in its company assessment field regardless of the low algorithm score. Alternatively, if a reporter answered "no" to relatedness, Defendants would still answer "yes" in its company assessment field if the algorithm score was five or greater. As a result, the only instance in which the company relatedness field could be answered "no" was when the reporter's assessment was "no" and the algorithm score was four or less.
This algorithm consisted of ten "yes" or "no" questions relating to factors such as temporal association, rechallenge/dechallenge results, and alternative explanations for the adverse event. Points were added or subtracted based on the answer to each of the ten questions, generating a total score range for each adverse event from zero to ten. Zero equated to "unlikely;" one to four equated to "possible;" five to eight equated to "probable;" and nine or more equated to "definite." The algorithm generated a score in the "probable" range in those situations where a patient developed a documented case of IBD while using Accutane. However, this score did not take into account any evidence of alternative reasons for the adverse event.
CIOMS II
The ADVENT data base also contained a CIMOS II field, which was used to input narrative discussion of various factors, including the drug/adverse event relationship. Notably, the company would only have knowledge of the factors reported to them or that they were able to learn through follow-up which sometimes occurred. Generally, the company would have limited knowledge of other potential causes or biases in the information.
Unlike the other fields in the data base which were completed by associates in the drug safety department, the CIOMS II field was to be completed by a drug safety physician. This field was often completed with an overall assessment of the drug/adverse event relationship and took into account the information from other causality assessment fields such as the Naranjo score. According to Defendants, the goal of the overall assessment was to determine whether Defendants determined there was a possible relationship between the drug in question and the event.
DISCUSSION
Plaintiffs seek to admit evidence of Defendants' causality assessments to prove that Accutane causes IBD through the "admission" of a party opponent, and to establish that Defendants had notice and/or knowledge of the potential dangers of Accutane. Plaintiffs contend this evidence is admissible under Rules 402, 702, 703 and 801(d)(2) of the Federal Rules of Evidence. Defendants argue that the causality assessments and any testimony regarding the assessments are inadmissible as they are not an admission, lack the reliability necessary to support an expert opinion, and are unduly prejudicial.
A. Admission by Party Opponent.
Rule 801 of the Federal Rules of Evidence permits the admission of statements offered against a party that is the party's own statement in either an individual or representative capacity. See F.R.E. 801(d)(2)(A). Plaintiffs seek to admit the causality assessment reports to establish that the company's assessments in the database are Defendants' admissions of causation of the reported events. Plaintiffs argue that Defendants' Director of Drug Safety, Dr. Daniel Reshef, acknowledged that the company's assessments are fully investigated articulations of these adverse events. To the contrary, Dr. Reshef testified in his deposition that the information or response of "yes" under the company assessment field did not mean the company had determined the adverse events reported were caused by Accutane, but rather expressed the company's belief, based on the algorithm and the information in the reporter's assessment field, that the adverse events reported could be related to the use of Accutane. Specifically, he testified:
Q: And you then at that point review the company assessment of causality, do you not?
A: Yes.
Q: All right. And at that point you make a judgment, a medical judgment based on your experience as to whether the company assessment is correct?
A: Yes, with a caveat that everything we refer to as causality in this discussion really means relatedness.
Q: Okay. But it says causality, does it not?
A: I agree, but it's really not the best terminology.
(Dr. Daniel Reshef's deposition, page 118).
It is clear from the above testimony that Dr. Reshef's statements do not constitute a non-hearsay admission by a party opponent. At no time did Dr. Reshef testify or allude to the company's assessment of "yes" being construed as the company's position that the reported events were actually caused by Accutane. The admission of his testimony purportedly to prove the same would be misleading.
A good example of why the reported event is not an admission, is the ADR of Mary Farr, a Plaintiff in this case. Mary Farr's ADVENT report includes a company assessment of "yes" for Crohn's disease. But this assessment reflects nothing but the assessment of the reporter of "yes." The reporter was her attorney. Her Naranjo score is only 2 out of 10, and Defendants' physician assessment in the CIOMS field states that the causal relationship between the reported events and the treatment with isotretinoin are assessed as "possible." As with many of the "yes" responses in the database, the "yes" in this case, reflects nothing more than an assessment of a possible relationship, not an actual relationship, and is based solely on her attorney's assessment of what caused her Crohn's disease. Clearly, this is not reliable proof that Accutane caused any of the ADRs alleged by Plaintiffs.
B. Expert Testimony.
Pursuant to Rule 702 of the Federal Rules of Evidence, an expert is permitted to testify in the form of an opinion if "(1) the testimony is based upon sufficient facts or data, (2) the testimony is the product of reliable principles and methods, and (3) the witness has applied the principles and methods reliably to the facts of the case." F.R.E. 702. Additionally, an expert must base his or her opinion or inference on those facts or data reasonably relied upon by experts in the particular field in forming opinions or inferences on the subject. The facts and data need not, however, be admissible in order for the expert's opinion to be admissible.See F.R.E. 703.
Plaintiffs' experts base their opinions, in part, on the information contained in the causality assessments. Plaintiffs contend that Rules 702 and 703 permit the admission of Defendants' causality assessments through expert testimony because their probative value outweighs any prejudicial effect, particularly in light of the fact that the causality assessments were not the only factors used in formulating the experts' opinions. This Court disagrees.
While Rule 703 permits an expert to base his or her opinions on facts or data that may otherwise be inadmissible, the expert must "employ in the courtroom the same level of intellectual rigor that characterizes the practice of an expert in the relevant field," meaning the expert's opinion must be based upon sufficient facts or data, be the product of reliable principles and methods, and be applied reliably to the facts of the case.See McClain v. Metabolife International, Inc., 401 F.3d 1233, 1237 (11th Cir. 2005) (citing Daubert v. Merrell Dell Pharmaceuticals, Inc., 509 U.S. 579, 590 (1993)). The causality assessments in this case do not contain data which is reliable and upon which an expert opinion of causation can be based. The causality assessments reflect a synopsis of complaints from various reporters regarding their subjective beliefs as to the causes of their particular ailments. These reports are not presented in the form of a medical file, but rather are narratives from the reporter describing his or her opinion as to causality. They do not contain information regarding the patient's medical history, family medical history, or use of other medications or drugs.
At most, an unbiased review of the assessments may support a conclusion that there is an association between Accutane and IBD. But " an association is not equivalent to causation." Reference Manual on Scientific Evidence at 336 (2d. Ed. Federal Judicial Center 2000) (emphasis in original). In fact, just such a review was made in 2006. The conclusion of the authors? "In a subgroup of patients, isotretinoin might serve as a trigger for IBD." Deepa Reedy, M.D., et al, Possible Association Between Isotrtinoin and Inflammatory Bowel Disease, Am. J. Gastroenterol, 1569 (2006) (footnote added). This conclusion may serve as a basis for forming an hypothesis, but certainly does not support an opinion on causation.
Isotretinoin is the generic name for Accutane.
Under Daubert, the reasoning or methodology underlying the testimony of an expert must be scientifically valid. See Daubert, 509 U.S. at 593-94. Moreover, the proposed testimony must "derive from the scientific method; good grounds and appropriate validation must support it." Id. at 590. Any testimony elicited from Dr. Fogel or any of Plaintiffs' experts regarding the causality assessments will lack this scientific validity. In fact, Dr. Fogel acknowledged in his deposition that in reviewing the causality assessments, he did not make any independent inquiry as to methodology used by Defendant in creating the assessments. Specifically, Dr. Fogel testified:
Q: Talking about the Roche documents that you say Roche concluded Accutane causes inflammatory bowel disease. What did you do to investigate what that label on that column meant?
A: I did not do any further investigations.
Q: So it said highest causality. Did you ask for the coding manual that would have told you what highest causality meant?
A: No, I did not.
. . .
Q: Did anybody give you the coding manual for the adverse event forms that lead to the generation of those tables?
A: No.
. . .
Q: So you didn't need to know the methodology by which those boxes were checked or what they meant, did you?
A: I did not look that up.
Q: Okay. All right. Do you have any information about the methodology that's used at Roche to make any assessment of causality of adverse event reports?
A: I have no information about Roche methodology.
Q: So you didn't receive any manuals, guidelines or other documents that provide guidance to Roche employees about making assessments with regard to adverse event reports, true?
A: No, because it strikes me if somebody says that something is causally related then it's causally related.
(Dkt. 408, Ex. 23 at pp. 252-253, 255) (emphasis added).
As evidenced by Dr. Fogel's testimony, his opinion regarding the causality assessments is speculative and conclusory and lacks the indicia of reliability necessary under Daubert. He simply opines that when the assessments say that an ADR is causally related to the use of Accutane, that the event is in fact related. As the Eleventh Circuit stated in McClain, "simply because a person takes drugs and then suffers an injury does not show causation. Drawing such a conclusion from temporal relationships leads to the blunder of post hoc ergo propter hoc fallacy." See McClain, 401 F.3d at 1243.
C. Notice.
Plaintiffs also argue that testimony regarding causality assessments should be admitted under Rule 401. Plaintiffs argue the assessments will assist the jury in determining whether Defendants knew or should have known of the serious side effects associated with the use of Accutane. First, if Defendants admit they received notice that some users thought their use of Accutane might be related to their complaints, the reports themselves are unnecessary. Second, to admit the causality assessments for the purpose of establishing notice would be more prejudicial than probative, and would inevitably confuse the jury.
If the Defendants during trial do not admit they received notice of possible adverse events, this issue may be revisited at that time.
While this Court has attained an understanding of the purpose and use of causality assessments, it is likely a lay juror would have difficulty distinguishing that the term "causality assessment," as the term relates to safety surveillance, is not the same as "causation." "[T]he procedures commonly used in [safety surveillance] for the purpose of establishing public health guidelines . . . are often . . . of marginal relevance to estimating causation in an individual."McClain, 401 F.3d at 1249 (citations omitted). In fact, it is highly probable that a juror would perceive the company's "yes" response in the causality assessment field as an admission by Defendants' physicians that Accutane did in fact cause the adverse events reported. Therefore, the potential prejudice outweighs the probative value of the reports. See Soldo v. Sandoz Pharmaceuticals Corp., 244 F. Supp. 2d 434, 546 (W.D. Pa. 2003) (finding that admitting causality assessments, which were prepared for a different purpose other than causation would be grossly misleading to a fact finder, and the likelihood of misleading the finder of fact greatly outweighs any probative value).
It is therefore ORDERED AND ADJUDGED that Defendant Hoffmann-LaRoche, Inc.'s and Roche Laboratories Inc.'s Motion to Exclude Evidence Regarding Causality Assessments (Dkt. #408) is GRANTED.
DONE and ORDERED in Tampa, Florida.