Daubert hearing, findings of fact, sufficiency of causation in scientific evidence, and learned intermediary instruction in medical negligence and products liability case involving punitives: HYMAN & ARMSTRONG PSC V. SANDOZ PHARM. (SC 4/24/2008)



In this well-publicized case, a Jefferson County jury awarded Gunderson’s estate, her surviving husband and two minor children a total of almost $19.1 million ($6 million for loss of parental consortium and approx. $1.85 million for loss of earning power and services along with $11.25 million in punitives) in February 2004 against Dr. Lynn Armstrong and Sandoz Pharmaceuticals (90% of compensatorys and all of punitives assessed to Sandoz) stemming from the sudden death of Gunderson in October 1993 while taking Parlodel postpartum (to stop lactation) as prescribed by Dr. Armstrong and manufactured by Sandoz. On appeal, the COA affirmed the compensatory damages award, but vacated the punitives award based on its determination that the TC failed to instruct the jury that punitives could not be based on conduct of Sandoz that occurred outside Kentucky. By opinion dated October 21, 2005, the COA remanded for a new trial on the amount of punitives only. Dr. Armstrong and Sandoz filed separate motions for discretionary review, which were granted and consolidated for the Supreme Court’s review.

In its 48-page opinion, the SC addresses a number of issues raised by one or both Appellants as follows:

Lack of Daubert Hearing by TC

Both Appellants argue that the TC never conducted a formal Daubert hearing per their request to determine the admissibility of Gunderson’s causation experts, which was error. The SC makes note that a hearing is only required if the record is not complete enough to measure the proffered testimony against the proper standards of reliability and relevance. In this case, the SC notes that the TC had a "mountain" of discovery material, reports, affidavits, scientific studies, etc. as well as extensive briefing by the parties, and devoted an entire day addressing the parties’ motions in limine, many of which related to the admissibility of the same scientific evidence being challenged. Thus, the Court felt that the TC did not abuse its discretion in conducting its review of the Daubert motions without a formal hearing.

Lack of Express Findings of Fact on Daubert Ruling by TC

While acknowledging the lack of any written findings of fact to support the TC’s decision, the Court reiterated recent Kentucky decisions determining that the only requirement is that the record clearly show the TC effectively conducted a Daubert inquiry. While the SC would prefer trial courts to include findings of fact in their rulings, the failure to do so does not automatically render the rulings arbitrary, unreasonable or unfair and is not grounds for reversal. An appeals court’s review standard is clear error to see if the TC’s ruling is supported by substantial evidence in the record.

Reliability and Relevance of Causation Evidence

The general basis for the Appellants’ challenges to the various pieces of scientific evidence (case reports, animal studies, chemical analogies) relied upon by Gunderson’s causation experts was that they were unreliable and/or irrelevant because the evidence failed to prove that Parlodel causes seizures in women taking the drug specifically for postpartum lactation suppression (PPLS). Instead, the Appellants suggested that the only reliable method of proving this theory is an epidemiological study, and since the only 2 such studies done found no significant link between the drug and such seizures the evidence cited by Gunderson’s experts was unreliable. In response, the SC spends a good deal of time in tracing the history of Parlodel, the reported cases of new mothers suffering seizures, strokes and heart attacks while taking the drug (especially those women who had hypertension during the pregnancy), and pointing out the bias and unreliability of the two referenced epidemiological studies. The Court acknowledged the fact that the great majority of courts from other jurisdictions addressing the same Daubert issues relating to the same scientific evidence presented by the Gunderson’s have ruled that such evidence was inadmissible. However, the Court felt the reasoning expressed by the U.S. District Court in Alabama was more on point in its determination that the majority courts had created too high a standard of admissibility under Daubert by equating reliability with scientific certainty. The SC highlighted the Alabama court’s view that while an epidemiological study may be the best evidence on the issue at hand, Daubert requires only that reliable evidence be presented and that the evidence at issue (animal studies, case reports, medical literature reviews) met that standard as being routinely used by scientists. In the end, the Court reiterated the TC’s view that while any one of the individual pieces of evidence may not definitely prove that the drug causes postpartum seizures, when considered together as an aggregate body of evidence all of it was reliable enough to put before the jury.

Sufficiency of Causation Evidence

Dr. Armstrong next argued that the Appellants were entitled to a direct verdict because there was insufficient reliable evidence that Parlodel caused the decedent’s alleged seizure (there was some issue about the exact cause of her death since the state medical examiner found no anatomic cause of death initially and only after further research into Parlodel and its reported effects on PPLS women concluded that death was attributed to seizure ). The SC again highlights generally the nature of the pieces of evidence relied upon by Gunderson’s expert and agreed with the TC that when viewed together they tend to show that the drug can cause seizures in women taking it for PPLS, and even referenced Sandoz’ internal memos from 11 years pre-death that acknowledged a connection between the drug and postpartum hypertension and seizures.

The second part to this issue was whether the Gunderson’s presented enough evidence of specific causation – that death was actually caused by a seizure due to the ingestion of Parlodel. The SC noted that all 3 of their medical experts used the differential diagnosis methodology to reach their independent conclusions that the drug caused the seizures that led to Gunderson’s death, and analyzed the respective investigations and bases for the opinions of each of the experts before concluding that the Gunderson’s had met their burden of proof on this issue submit for a jury to determine.

Failure of TC to Give Learned Intermediary Instruction

3 months after the trial in this case, the SC adopted the learned intermediary doctrine from the Restatement (Third) of Torts in Larkin v. Pfizer, Inc., which relives a drug manufacturer from liability to the ultimate consumer if it provides adequate warning about the drug to the prescribing physician. Sandoz argued information in the package insert for Parlodel and the PDR, as well as direct letters to doctors, sufficiently noted the risks and seizure and hypertension for postpartum patients and therefore constituted an adequate warning entitling it to the instruction. Despite acknowledging the evidence presented of Sandoz’s alleged deception and continued marketing of Parlodel even after the FDA asked it to withdraw its indicated use for PPLS, the SC held that the TC erred in its failure to give the instruction since Sandoz had presented enough evidence of an adequate warning to Dr. Armstrong of the risks of the drug. However, the SC ruled that the Gunderson’s had met their burden of demonstrating no prejudice resulted from the error given the "abundant" evidence they presented of Sandoz’s efforts to deliber
ately conceal or downplay the drug’s risks and when considering the jury’s assessment of almost all of the liability to Sandoz, a fact that severely undercut Sandoz’s argument on the drug warning’s adequacy.

Failure of TC to Grant Directed Verdict on Medical Malpractice Claim

Dr. Armstrong next argued the TC erred by not directing a verdict on the medical malpractice claim filed against him since the Gunderson’s failed to show his treatment of the decedent violated the applicable standard of care. The SC agreed that something more than mere injury from a drug must be shown in order to have a viable med mal claim, but noted that Gunderson’s experts had presented sufficient evidence that it was a deviation of the standard of care for Dr. Armstrong to prescribe Parlodel in 1993 (even though it was not when he prescribed it to the deceased after her first pregnancy in 1989) because of the additional information about the known risks of the drug when used for PPLS that had come forth during those four years and the fact that she had suffered from gestational hypertension during the second pregnancy.  The SC reaffirmed the majority view among courts that while information about the drug in the package insert and PDR is relevant and useful information regarding the prescribing doctor’s standard of care, it is not the sole determinant of this standard. The Court therefore concluded that the Gunderson’s had presented enough evidence for the jury to consider this claim.

Mitigation Evidence on Loss of Parental Consortium Claim

Appellants attempted unsuccessfully to present evidence of the two surviving sons’ close relationship with their father’s girlfriend of 4 years (at trial time) to mitigate the damages on their loss of parental consortium claims, and argued on appeal that it was an abuse of the TC’s discretion to deny the proffered evidence. The TC had felt that the evidence could not be presented on the parental consortium claim where it was arguably admissible without running the risk that it would be considered in connection with the estate’s wrongful death claim where it clearly was not admissible. The COA agreed with the TC since the relationship between the decedent’s husband and new girlfriend was not of sufficient duration and stability to be admissible.

The SC began its review by noting that the admissibility of such relationships on a parental consortium claim was one of first impression in Kentucky, and then traced the distinctions between spousal and parental consortium claims through Kentucky case law. While not wanting to minimize the loss of a spouse’s consortium, the SC held that it is undeniable that a child’s loss of parental consortium is often greater and more difficult to recover from over time. The SC thus concluded that the proffered evidence in this case was not admissible on the parental consortium claims (even though it would be on a spousal consortium claim).

Admissibility of Gunderson’s Mental Health Counselor Testimony

Appellants challenged the admissibility of any testimony from Dr. Bower, a mental health counselor who had performed a psychological assessment on both children in the lawsuit and concluded that both had been deeply affected by their mother’s death. Her testimony included the showing of drawings the children had made and their significance to their mother’s death. Appellants filed a motion in limine on grounds that her testimony was too speculative, subjective and inflammatory, and not based on good science or methodology as required by Daubert (since Bowers had not performed any form of objective, standardized testing during her analysis). The SC concluded that Bowers’ extensive education and experience in child counseling and the limited purpose of her testimony rendered her testimony sufficiently reliable and therefore admissible on the parental consortium claims under KRE 702.

Admissibility of Evidence of Dr. Armstrong’s Cross-Claim against Sandoz

In 1998, Dr. Armstrong’s estate (he passed away while the case was pending) filed a cross-claim against Sandoz alleging fraudulent misrepresentation and gross negligence in its marketing of Parlodel and sought damages for injury to his reputation and for indemnification on the Gunderson’s claims. Prior to trial, Sandoz entered an indemnification agreement with Dr. Armstrong settling his claims and taking over his defense. The cross-claim was dismissed and Dr. Armstrong offered no evidence against Sandoz. However, the TC ruled that the cross-claim could be introduced as evidence of the Defendants’ non-adverse relationship per KRE 408. The Gunderson’s discussed the cross-claim during their opening and closing statements and was mentioned during voir dire. Sandoz cites error in the TC’s decision since the cross-claim was inadmissible hearsay introduced as substantial evidence against it in violation of KRE 801. The SC notes that while Gunderson had argued at trial that the cross-claim was admissible to show Dr. Armstrong’s change in position and the collusive nature of his relationship with Sandoz while on appeal it was argued that the cross-claim was non-hearsay as an admission of a party-opponent (KRE 801A(b)) since Dr. Armstrong admitted therein that the drug was unsafe and caused Gunderson’s death, which contradicted his position at trial.

The SC held that while the settlement agreement between the Defendants was admissible per KRE 408 to show the potential bias of the Defendants who were previously adversaries, the cross-claim was not. Although KRE 801A(b)(1) allows introduction of an adverse party’s admissions, such evidence can only be used against the declaring party. Here, the SC noted that the Gunderson’s clearly used the cross-claim to try and prove Sandoz’s liability instead, and even conceded to the TC that the cross-claim tended to exculpate Dr. Armstrong. Thus, the SC ruled that the TC erred in admitting the cross-claim. However, the SC determined the error was not palpable and that no manifest injustice resulted in light of the other voluminous and "quite damning" evidence that certainly was persuasive to the jury.

Punitive Damages

Sandoz appealed the TC’s refusal to tender a jury instruction pursuant to the US Supreme Court’s decision in State Farm v. Campbell (2003) that specifically precluded the jury from using punitive damages to punish Sandoz for conduct outside Kentucky. The COA ruled this was error and remanded for a new trial on the amount of punitive damages only. On appeal to the SC, Sandoz argued that the COA’s decision deprived it of the right to a fair trial because it improperly presumed the Gunderson’s were entitled to such damages in the first place. In response, the SC agreed with the COA that the Gunderson’s had presented sufficient evidence that Sandoz acted with wanton or reckless disregard for the decedent thereby justifying a punitive damage instruction. However, in accordance with its earlier decision in Sand Hill Energy v. Smith (2004), the SC ruled that Sandoz was entitled to a new trial on liability for punitive damages rather than just a determination of the amount of such damages to be awarded.


While the Supreme Court deemed it error for the TC to not give a learned intermediary instruction and to allow the cross-claim’s admission, it adjudged both to be harmless errors. It therefore affirmed the COA decision. All Justices concurred (except Justice Abramson who did not sit), with Justice Scott concurring in result only since he did not believe Sandoz was entitled to a learned intermediary instruction.

Digested By Chad Kessinger
Schiller Osbourn Barnes & Maloney

Please note: I reserve the right to delete comments that are inappropriate, offensive or off-topic.

Leave a Reply

Your email address will not be published.