Coburn v. Smithkline Beecham Corp., 174 F. Supp. 2d 1235 (D. Utah 2001)

*1236 Edward B. Havas, Dewsnup King & Olsen, Salt Lake City, UT, Arnold Anderson *1237 Vickery, Richard W. Ewing, Vickery & Waldner, Houston, TX, for plaintiffs.

John A. Anderson, Matthew M. Durham, D. Matthew Moscon, Stoel Rives LLP, Salt Lake City, UT, Chilton Davis Varner, Andrew T. Bayman, King and Spalding, Atlanta, GA, Vernon I. Zvoleff, Preuss Shallangher Zvoleff & Zimmer, San Francisco, CA, for defendant.

KIMBALL, District Judge.

This matter is before the court on Plaintiffs' Motion for Partial Summary Judgment, Defendant's Request to Take Judicial Notice, and Plaintiffs' oral motion to disallow the deposition of Dr. Healy. A hearing on these motions was held on October 31, 2001. At the hearing, Plaintiffs Judith Coburn et al. ("Plaintiffs") were represented by Andy Vickery, and Defendant SmithKline Beecham, dba GlaxoSmithKline ("GSK"), was represented by John A. Anderson. Before the hearing, the court considered carefully the memoranda and other materials submitted by the parties. Since taking the matter under advisement, the court has further considered the law and facts relating to these motions. Now being fully advised, the court renders the following Order.

Plaintiffs seek partial summary judgment on the issues of "general causation" and "fault." Specifically, they seek summary judgment establishing two issues of fact: (1) that Paxil^[1] can cause some individuals to commit homicide and/or suicide, and (2) that GSK is at "fault" within the meaning of Utah products liability law for its failure to warn either prescribing physicians or patients about this risk.

Plaintiffs argue that they are entitled to partial summary judgment based on offensive, non-mutual collateral estoppel, resulting from the jury verdict in Tobin v. GlaxoSmithKline, No. 00-CV-25BEA, a case in the United States District Court for the District of Wyoming. The plaintiffs in Tobin had alleged that Donald Schell ("Mr. Schell") shot and killed his wife, daughter, granddaughter, and then himself after taking Paxil for two days. On the special verdict form, the jury found that Paxil "can cause some individuals to commit homicide and/or suicide" and that it was a legal cause of the individuals' deaths. The jury apportioned "fault" at 80% to GSK and 20% to Mr. Schell. The court entered judgment in the plaintiffs' favor in the amount of $6.4 million.

Plaintiffs argue that this court should apply collateral estoppel because such application would be consistent with the underlying policies of the doctrine of collateral estoppel (i.e., judicial and litigation economy and avoiding inconsistent verdicts) and because they have satisfied the elements of collateral estoppel.

GSK urges the court to deny Plaintiffs' motion, arguing that the elements for collateral estoppel have not been satisfied, and, that even if the elements had been satisfied, this court should exercise its discretion not to apply collateral estoppel because preventing GSK from litigating these issues would be fundamentally unfair.

The application of collateral estoppel requires that: (1) the issue previously decided is identical with the one presented in the action in question; (2) the *1238 prior action has been finally adjudicated on the merits; (3) the party against whom the doctrine is invoked was a party, or in privity with a party, to the prior adjudication; and (4) the party against whom the doctrine is raised had a full and fair opportunity to litigate the issue in the prior action. Dodge v. Cotter, 203 F.3d 1190, 1198 (10th Cir. 2000). The burden of proving the identity of issues rests on the party asserting the collateral estoppel. Hernandez v. Los Angeles, 624 F.2d 935, 937 (9th Cir.1980).

In evaluating the "issue identity" prong of collateral estoppel, the Tenth Circuit, in Dodge determined that it was "not possible to know the compass of the [first] jury's finding of negligence" when the plaintiffs had presented multiple factual theories of what specific acts constituted negligence and the jury had made a non-specific finding that the defendant was "negligent." Id. at 1197. Thus, the Tenth Circuit reversed the trial court's application of offensive collateral estoppel. Id. at 1198-99. Numerous other courts in products liability cases have refused to apply collateral estoppel when there was any ambiguity in discerning precisely what the jury actually and necessarily decided. See, e.g., Anderson v. Bungee Int'l Mfg. Corp., 44 F. Supp. 2d 534, 539 (S.D.N.Y.1999); Setter v. A.H. Robins Co., 748 F.2d 1328, 1331 (8th Cir.1984); Kramer v. Showa Denko K.K., 929 F. Supp. 733, 750 (S.D.N.Y.1996).

This court is not convinced that the "issue identity" element of collateral estoppel has been satisfied because it is not clear what the jury actually decided in Tobin. For example, in the Tobin trial, the plaintiffs offered several theories as to how Paxil "can cause some individuals to commit homicide and/or suicide." Specifically, the plaintiffs offered testimony that violence and/or suicide can result from Paxil-induced akathisia, emotional indifference, or psychological decompensation. Yet, the jury did not specify indeed was not asked on which theory it based its decision. In addition, the so-called "vulnerable subpopulation" of "some individuals" is entirely undefined and undefinable based on the verdict, rendering it impossible for the jury in this case to determine whether Mr. Coburn was among this undefined "vulnerable subpopulation." Because of the ambiguities inherent in the Tobin jury's verdict and because the facts surrounding Mr. Coburn's suicide in the case at bar are much different from the facts surrounding the suicide and homicides in the Tobin case, it would be inappropriate for this court to apply collateral estoppel on the causation issue.

In addition, it is unclear whether the Tobin jury determined whether GSK was "at fault" for failing to test or failing to warn, or both, as the special verdict form asked the jury to determine the amount of fault attributable to GSK and to Mr. Schell. It did not ask the jury to specify the theory upon which it based its finding of fault.^[2] Moreover, Plaintiffs have not demonstrated that the same legal standards *1239 applicable in Wyoming for the Tobin case are applicable in Utah to the instant case.

Furthermore, the duty to warn arises when a manufacturer "knows or should know of a risk associated with its product." House v. Armour of Am., Inc., 929 P.2d 340, 343 (Utah 1996); Grundberg v. Upjohn Co., 813 P.2d 89, 97 (Utah 1991). Thus, the question in the instant case is whether GSK was aware of the alleged risks in November 1996, when Mr. Coburn took Paxil. This court will not dictate the answer to that question by embracing a 2001 jury verdict, which was predicated on post-1996 evidence.

Thus, the "issue identity" element of collateral estoppel has not been satisfied on either the general causation or fault issue. Accordingly, it would be reversible error for this court to apply collateral estoppel to preclude GSK from litigating these issues.

Moreover, even if the elements required for the application of collateral estoppel had been satisfied, this court would still deny Plaintiffs' motion. A court has broad discretion to determine when offensive, non-mutual collateral estoppel should be applied Parklane Hosiery v. Shore, 439 U.S. 322, 332, 99 S. Ct. 645, 58 L. Ed. 2d 552 (1979); Dodge v. Cotter, 203 F.3d 1190, 1198 (10th Cir.2000). In this case, it is abundantly clear that this court should not exercise its discretion to apply collateral estoppel. To preclude GSK from litigating the general causation and fault issues in this case would be fundamentally unfair and would not promote at least one of the policies underlying the doctrine of collateral estoppel judicial economy.

First, the court is doubtful that significant time would be saved by granting Plaintiffs' motion. For example, Plaintiffs argue that, by instructing the jury that "Paxil can cause some people to commit homicide and/or suicide," it would eliminate the need for expert testimony on general causation and the parties could focus on specific causation (i.e., whether Paxil was a cause of Mr. Coburn's suicide). GSK has made clear, however, that, even if this court grants Plaintiffs' motion on general causation, its defense will continue to be that Paxil was not a legal cause of Mr. Coburn's suicide because Paxil cannot cause anyone to commit suicide. Thus, GSK will still elicit general causation testimony from its experts, and this court could not prevent GSK from doing so, as this is its primary defense regarding why it could not have caused Mr. Coburn's suicide. Furthermore, it is unclear to this court how Plaintiffs will establish that Paxil was a cause of Mr. Coburn's suicide without eliciting expert testimony concerning general causation (i.e., that Paxil can cause some people to commit suicide). Thus, it is not clear that significant time will be saved by granting Plaintiffs' motion.

Second, the Tobin jury's verdict was necessarily based, in large part, on experts whose testimony has not consistently been found to be reliable. While the Magistrate Judge in the Tobin case declined, after a Daubert hearing, to strike the testimony of Dr. David Healy, the Tobin plaintiffs' general causation expert, this court finds it material that, in another Paxil case pending in the United States District Court for the District of Kansas, Miller v. Pfizer Inc., No. 99-2326-KHV, the independent experts appointed by the court found that Dr. Healy's opinions rely upon methods that have not been accepted in the relevant scientific community and misapply the scientific postulates upon which he ostensibly relies.^[3] This court recognizes that the *1240 Daubert hearing has not yet occurred in the Miller case, and thus there is no judicial determination regarding Dr. Healy's testimony; however, in light of such a finding by the court-appointed independent experts, this court declines to bind GSK to a jury verdict based largely on Dr. Healy's testimony.

Furthermore, the testimony of another expert for the Tobin plaintiffs, Dr. Maltsberger, has been rejected by another court. See Smith v. Pfizer, Inc., No 98-4156-CM, 2001 WL 968369 (D. Kansas Aug. 14, 2001). The Smith court found that Dr. Maltsberger is not qualified as an expert on the issues of general causation or warnings.

In the instant case, there is a scheduled Daubert challenge to Dr. Healy, the very expert on which the Tobin verdict was largely based. To grant Plaintiffs' motion would be tantamount to a determination that this court is bound by the Magistrate Judge's conclusions on the Daubert challenge in Tobin. Collateral estoppel, however, is not intended to operate as a shield against scrutiny of the reliability of the underlying science and the experts.

Third, this court finds that, for purposes of this motion, there are inconsistent judgments on the issue of causation. While it is true that Tobin was the first jury verdict involving Paxil, it is not the first judgment regarding other selective serotonin re-uptake inhibitors ("SSRIs"), such as Prozac and Zoloft. In Tobin, GSK unsuccessfully moved in limine and otherwise objected to the introduction of evidence regarding other SSRI medications. The Tobin court, however, permitted the plaintiffs' leading causation expert, Dr. Healy, to rely upon internal documents from other SSRI manufacturers and upon studies involving other SSRI medications, namely Prozac and Zoloft, to substantiate his opinion that Paxil can induce violence and/or suicide. In other words, the Tobin plaintiffs' defect and causation theories lumped all SSRI drugs together, making the Tobin trial as much about SSRIs generally as about Paxil specifically.

At least one other SSRI case resulted in a defense verdict prior to the Tobin verdict. See Forsyth v. Eli Lilly & Co., 904 F. Supp. 1153 (D. Hawaii 1995). Additionally, summary judgment in favor of the defendant manufacturer was recently granted in another case against a different SSRI manufacturer. See Smith, 2001 WL 968369 at *10-13. In light of Plaintiffs' reliance upon evidence of the effects of other SSRI medications to indict Paxil, these inconsistent verdicts against other manufacturers of SSRI medications militate against applying collateral estoppel in this case.^[4]

Fourth, it would be unjust to freeze in time the answers to these questions when the resolution of these questions is very time-dependent. For example, the issue of general causation will be decided upon the basis of scientific knowledge as it exists at the time of trial. When a scientific hypothesis is at issue in a trial, the evidence is likely to be different at a later trial. Similarly, as mentioned above, the scope of a duty to warn is dependent on whether the drug manufacturer knew or should have known of the risks at the time *1241 the drug was given to the patient. Because Paxil was prescribed and given to Mr. Coburn and Mr. Schell at different times (November 1996 and February 1998, respectively,) the relevant scientific knowledge would naturally be different a difference that may affect when any obligation to warn arose.

Fifth, it is entirely unclear to this court exactly what would be accomplished by granting Plaintiffs' motion, other than causing prejudice to GSK and confusion to the jury. As discussed above, it is not likely that significant time would be saved. Further, the jury certainly could not be told merely that Paxil "can cause some people to commit homicide and/or suicide," without guidance as to the probability (1 in 10 versus 1 in 1,000,000), or the characteristics, if there are any, of people in this population of "some people," or the circumstances under which such a scenario could occur.

Moreover, such an instruction would inevitably color the jury's decision regarding specific causation in this case. The court cannot discern the bright-line distinction advocated by Plaintiffs between general causation and specific causation. It is basic toxicology that any causation inquiry is dependent on dose, which necessarily includes duration, frequency, and amount of exposure, and for pharmaceuticals, also implicates the effect of other agents and biochemical and metabolic interactions and processes, preexisting medical conditions, and environmental factors. Thus, the risk of prejudice and confusion significantly outweighs any benefit that might be derived from applying collateral estoppel.

Finally, this court finds that it is contrary to public policy to allow a single jury verdict to brand an entire product defective throughout the country, particularly when there exists a significant and ongoing debate in the scientific and medical community about the issues involved in this and other SSRI cases, when the time frames for assessing the answers to the general causation question and the product defect question are different in this case than in the Tobin case, and when the Tobin case was a case that may have incited passion and sympathy for the plaintiffs and/or spite against a drug manufacturer, which, Plaintiffs' attorney is quick to highlight, generates over $6 million per day from Paxil.

This court declines to enshrine a jury verdict in a way that could trump the FDA's determination of safety and effectiveness and jeopardize the availability of a product that is beneficial to many people. As the scientific landscape changes concerning the evidence for or against the causal connection posited by Plaintiffs, GSK should be free to present and juries free to evaluate evidence that makes it more or less likely that Paxil does or does not cause violent reactions in some patients. Simply said, the Tobin verdict should not be and will not be the final word on the subject.

GSK has requested that this court take judicial notice of certain facts for the purpose of GSK's opposition to Plaintiffs' Motion for partial summary judgment. Plaintiffs have not opposed GSK's request, but noted at the hearing that the facts listed by GSK are not all adjudicated facts and that the weight given to these facts should be balanced with that knowledge.

The court will take judicial notice of the facts set forth in GSK's request, recognizing that not all facts are adjudicated facts.

At the hearing, the parties notified the court that they were having a dispute *1242 about the deposition of one of Plaintiffs' experts, Dr. Healy. Plaintiffs argue that GSK is not entitled to depose him because it deposed him for 10 hours in the Tobin case on the same issues. Plaintiffs also note that GSK cross-examined Dr. Healy in the Tobin trial and that GSK has had several other opportunities to hear or otherwise obtain Mr. Healy's testimony on general causation.

GSK asserts that it is entitled to depose Dr. Healy in this case because the facts in this case are different than the facts in the Tobin case. Furthermore, GSK claims that it did not receive voluminous materials in the Tobin case until the eve of Dr. Healy's deposition in that case and thus, it did not adequately depose him on all of those materials. In addition, GSK argues that Dr. Healy's expert report in this case contains material that was not relied upon in the Tobin case and that his report states that he may have additional materials. GSK recognizes, however, that it need not depose Dr. Healy as extensively as it otherwise would, due to the previous deposition.

The court agrees with GSK that it is entitled to depose Dr. Healy. However, GSK's deposition of Dr. Healy is limited to no more than five hours. The court strongly encourages the parties to schedule the deposition during Dr. Healy's visit to the United States for the Miller case in mid-November.

For the foregoing reasons, and good cause appearing, IT IS HEREBY ORDERED that Plaintiffs' Motion for Partial Summary Judgment is DENIED, Defendant's Request to Take Judicial Notice is GRANTED, and Plaintiffs' oral motion to disallow the deposition of Dr. Healy is DENIED. Dr. Healy's deposition, however, is limited to 5 hours and every effort should be made by GSK to take the deposition while Dr. Healy is in the United States in mid-November.

[1] Paxil ("Paroxentine") is an FDA-approved medication indicated for, among other things, the treatment of depression, panic disorder, obsessive/compulsive disorder and other medical conditions.

[2] Plaintiffs also argue that, even if collateral estoppel is not appropriate on the fault issue, they are still entitled to summary judgment on this issue based upon the undisputed facts. They claim that it is undisputed that there is no warning alerting prescribing physicians about Paxilinduced suicide and that it is undisputed that, in both Wyoming and Utah, GSK must comply with the minimum duty to warn under FDA regulations. Plaintiffs ignore that GSK adamantly disputes that it implicitly agreed at the charge conference in the Tobin case that, because it has never warned about Paxil's ability to induce violent behavior, GSK would necessarily be at fault for this failure if a jury found that Paxil can cause violent behavior and that it caused Mr. Schell's violent behavior. Moreover, Plaintiffs have failed to demonstrate that, under Utah law, GSK would necessarily be at fault for failing to warn if the jury determines that Paxil caused Mr. Coburn's suicide.

[3] According to GSK-and not contradicted by Plaintiffs Dr. Healy's opinions in Miller are the same or substantially similar to the opinions he offered at trial in Tobin. For example, at the trial in Tobin, Dr. Healy relied upon his own healthy volunteer study using Zoloft to indict Paxil. In Miller, Dr. Healy also relies upon his own healthy volunteer data. The independent experts, however, found that Dr. Healy's study "did not produce statistically significant results." See GSK's Mem. in Opp'n to Mot. for Partial Summ. J. at n. 7 & Ex. 7.

[4] The court recognizes that GSK will continue to argue that it is both scientifically and legally improper to lump all SSRIs together.