Products Liability Judicial 'Gatekeeping': 'Frye,' 'Foundational Reliability' By Michael Hoenig - New York Law Journal - February 11, 2008	Print article

People v. Wesley¹ can be viewed as a kind of bellwether for restating New York's modern version of the Frye rule. Decided in March 1994 by the Court of Appeals, with only five judges participating, the issue was whether DNA profiling (in 1988 and 1989), as novel scientific evidence, was generally accepted by the relevant scientific community to be admitted against the defendant in a murder and rape case.

Three judges held "yes." Chief Judge Judith S. Kaye (joined by Judge Carmen Beauchamp Ciparick) issued a separate concurring opinion, but it was really a kind of dissent because they concluded that it was error to admit bloodstain analysis in this case. They nevertheless agreed that the conviction should be affirmed because, as a minor part of the People's case, the admission was harmless error.

Why did Chief Judge Kaye write separately? As she put it,

out of concern, for future cases, that the principles governing admission
of novel scientific evidence be correctly articulated and applied.

Wesley established that the Frye inquiry, the "general acceptance" test for novel scientific evidence, is a quest for reliability. All the judges said that, after the Frye question, there is a foundation assessment (the majority called it a distinct "voir dire" foundation question) where the focus moves from general reliability to specific reliability of the foundation. The foundation question is a preliminary matter of law. Chief Judge Kaye called it an inquiry going to admissibility, not simply its weight. It is an issue of law for the judge. If it is a "trustworthiness" issue, it is assessed by the court. Infirmities in collection and analysis not affecting trustworthiness go to "weight," to be assessed by the jury.

Chief Judge Kaye disagreed with the majority and found error in the trial court's relegating to the jury foundational issues such as adequacy of the methodology and procedures to assure the reliability and accuracy of the results. Foundational questions go to admissibility, not weight. Wesley thus says there are separate or distinct reliability quests, headlined by the words, Frye (general acceptance test) and "Foundation." Each of those is a matter for the court, not for the jury. Further, the foundational inquiry is employed in all science cases, not just "novel" science cases. Thus, trial courts are required to gatekeep for reliability in science cases not governed by Frye as well.

'Parker v. Mobil Oil'

The New York Court of Appeals' Oct. 17, 2006 decision in Parker v. Mobil Oil Corp.,2 is of obvious relevance in the sophisticated toxic tort arena. But Parker is also informative on the gatekeeping task in other science settings.

First, the facts in a nutshell. Mr. Parker, a gas station attendant for some 17 years, developed a form of leukemia (AML) his experts attributed to benzene exposure through inhalation of gasoline fumes and through skin contact with gasoline. He sued Mobil Oil and other defendants for toxic consequences and failure to warn. The parties' experts disputed what the specific cause of Mr. Parker's leukemia was. Defendants moved to preclude Mr. Parker's expert testimony on the issue of medical causation. Defendants contended plaintiffs' expert testimony regarding causation was scientifically unreliable under the Frye standard. Defendants also moved for summary judgment dismissing all claims because of the inappropriate causation evidence. Neither side requested a Frye hearing.

Experts' Battle

Defendants' motion was supported by opinions of two experts prepared for other litigations, an epidemiologist employed in a high position at Mobil and a toxicologist. Plaintiff countered the motion with reports from two experts, a board-certified physician in occupational medicine, and a physician expert in toxicology and epidemiology. Both sides referred to studies allegedly supportive of their respective positions.

The trial court denied defendants' motion. The Appellate Division reversed, however, finding the pivotal legal issue to be: "to what extent the plaintiff was required to establish the precise level of his exposure to benzene." Since Mr. Parker's experts neither quantified plaintiff's actual exposure nor the threshold required, they could not show that Mr. Parker's exposure exceeded the minimum disease-causing level. The causation opinions were therefore deemed speculative. The Second Department panel observed that plaintiff's experts did not utilize the three-step causation methodology approved by the World Health Organization and the National Academy of Sciences.

The Court of Appeals affirmed the dismissal but on the ground that plaintiff's experts, although highly qualified in their respective fields, "failed to demonstrate that exposure to benzene as a component of gasoline caused [Mr.] Parker's AML." The dispute was not whether there was an associational relationship between benzene exposure and the risk of developing AML. The key issue was the relationship, if any, between exposure to gasoline containing benzene as a component and the disease. As the Court concluded, the expert "fails to make this connection perhaps because, as defendants claim, no significant association has been found between gasoline exposure and AML. Plaintiff's experts were unable to identify a single epidemiological study finding an increased risk of AML as a result of exposure to gasoline . . . . Thus, the experts' opinions were properly excluded."

Despite affirming the dismissal, the Court rejected the Appellate Division's approach on a need for precise quantification of the dose-response relationship in all cases.

Where we depart from the Appellate Division is that we find it is not always necessary for a plaintiff to quantify exposure levels precisely or use the dose-response relationship, provided that whatever methods an expert uses to establish causation are generally accepted in the scientific community.

The Court referred to "several other ways an expert might demonstrate causation." Thus, for example, "intensity of exposure" to benzene may be more important than a cumulative dose. Or, perhaps, mathematical modeling taking into account a plaintiff's work history could be used to estimate exposure to a toxin. Possibly, "more qualitative means" could be used to express exposure. Thus, comparisons to exposure levels of subjects in other studies might be helpful "provided that the expert made a specific comparison sufficient to show how the plaintiff's exposure level related to those other subjects." Such methods, "along with others, could be potentially acceptable ways to demonstrate causation if they were found to be generally accepted as reliable in the scientific community."

'Gatekeeping' Lessons

Apart from the specific toxic causation issue in this case, the Court in Parker announced a number of other important gatekeeping principles. The parties disputed whether the battle of the expert opinions should be analyzed under Frye. The Court explained that when "novel scientific evidence" is introduced, there must be a determination of its reliability. The Frye "general acceptance" test asks "whether the accepted techniques, when properly performed, generate results accepted as reliable within the scientific community generally."

However, said the Court, the Frye inquiry is "separate and distinct from the admissibility question applied to all evidence - whether there is a proper foundation - to determine whether the accepted methods were appropriately employed in a particular case." Thus, the focus moves "from the general reliability concerns of Frye to the specific reliability of the procedures followed to generate the evidence proffered and whether they establish a foundation for the reception of evidence at trial."

In Parker, there was no particular novel methodology at issue for which the Court needed to determine the question of general acceptance. Rather, said the Court, "the inquiry here is more akin to whether there is an appropriate foundation for the experts' opinions, rather than whether the opinions are admissible under Frye." There was a question whether the methodologies used by plaintiff's experts "lead to a reliable result." Specifically, did they provide "a reliable causation opinion" without using a dose-response relationship and without quantifying Mr. Parker's exposure? Reliability of expert methodologies and opinions is thus a threshold issue, clearly "separate and distinct" from Frye, "more akin to" the exercise of the judge's responsibility to determine "whether there is an appropriate foundation for the experts' opinions."

There is, of course, a robust reliability frame of reference in cases applying the federal Daubert gatekeeping standard.³ But the parties in Parker did not argue for Daubert to apply acknowledging, instead, that Frye was the current New York standard. Nevertheless, the Court of Appeals, when analyzing federal case law on how much precise exposure information a toxic tort plaintiff needs to prove, stated in footnote ⁴: "We recognize that these cases employ a Daubert analysis. However, they are instructive to the extent they address the reliability of an expert's methodology." In other words, the federal Daubert analysis is "instructive" for New York courts addressing reliability concerns about experts' methods and their opinions.

Summary

To recapitulate: Parker first suggests that the Frye "general acceptance" test properly applies to novel scientific evidence. Then, upon appropriate challenge, there may be a separate and distinct reliability analysis that is needed, a foundation inquiry. Here the federal Daubert reliability screening criteria may be "instructive."

Further, in analyzing foundation in toxic exposure cases specifically, a rigid obedience to requiring precise dose-response and exposure information may be inappropriate as it may set an "insurmountable standard that would effectively deprive toxic tort plaintiffs of their day in court." Nevertheless, experts cannot introduce "junk science" or simply give an opinion unreliably or speculatively about general and specific causation issues. In analyzing foundation of toxic exposure cases, "it is necessary to find a balance between these two extremes." There may be alternate means to reliably demonstrate causation but these alternative methods will first have to be found to be "generally accepted as reliable in the scientific community." (A Frye question).

Parker is an important decision. It echoes Chief Judge Kaye's concurring opinion in Wesley. Note that Judge Ciparick wrote the opinion in Parker. She had joined Chief Judge Kaye's separate opinion in Wesley. There is a clear line of continuity between the two opinions.

The 'Nonnon' 'Time Warp'

Nonnon v. City of New York,⁴ a toxic tort case that eventually reached the Court of Appeals, complicates the analytical picture somewhat and creates some uncertainties. The Appellate Division, First Department's decision in Nonnon was published on June 6, 2006. The question in Nonnon concerned plaintiffs who alleged that they were being systematically poisoned by environmental contaminants in and around an 81-acre landfill owned by the city and operated by the Sanitation Department for some 16 years until 1979. The city moved to dismiss for, among other things, plaintiffs' failure to allege a viable causal connection between the landfill and 13 plaintiffs' cancers. The panel split 3-2 issuing bristling majority and dissenting opinions. The majority affirmed denial of the city's motion to dismiss. The dissenting justices wanted to dismiss the claims.

Nonnon is a decision caught in a kind of time warp, a kind of procedural "freeze" zone with important events inside, around and outside the decision creating some tensions. Here is the essence of the critical time frame problem. The Court of Appeals opinion in Parker was issued on Oct. 17, 2006. However, the First Department's split decision in Nonnon was published in June, more than four months earlier. So, the First Department justices in Nonnon did not have the benefit of guidance from the pivotal Parker decision in the state's high court. On the contrary, the First Department justices in Nonnon argued vigorously over whether the Frye gatekeeping standard announced in the Second Department's March 2005 decision in Parker should control the result in Nonnon. The Nonnon majority held "no" while the dissenters argued "yes." Afterwards, Parker in the Court of Appeals affirmed the Second Department's dismissal of the plaintiff's claim but not its reasoning under Frye. Parker held that the controversy there didn't present a novel science Frye issue. Rather, it was a foundational/reliability issue that plaintiffs had flunked.

The Nonnon saga continues. The defendant city sought appellate review and the Appellate Division, on Sept. 29, 2006, granted leave to appeal on a certified question ("whether the order of the Supreme Court, as modified by the Appellate Division, was properly made?"). There was full briefing and oral argument at the Court of Appeals. Much of this focused on whether the majority decision in Nonnon violated the Court of Appeals standard of gatekeeping later articulated in Parker. On June 27, 2007, the state's high court issued a memorandum opinion affirming the First Department's decision but "solely on procedural grounds."

This stunning procedural result stemmed from the following circumstance. The city had moved to dismiss the claims under CPLR 3211(a)(7) for failure to state a cause of action. Both the city and plaintiffs, however, had submitted expert affidavits addressed to causation. But the city's motion had never been converted to one for summary judgment. Thus, the affidavits could not properly be considered for their evidentiary support. Therefore, the Court of Appeals concluded that "plaintiffs were not put on notice of their obligation to make a complete record and to come forward with any evidence that could possibly be considered." Accordingly, said the Court, the city is "not now entitled to dismissal of plaintiffs' complaints for failure to state a cause of action."

So, what is the status of the Nonnon First Department opinion? Well, it is on the books. It is a ruling with some weight. But, if you run Nonnon through Shepard's, it comes up with a yellow caution signal. It says "subsequent appellate history contains possible negative analysis." Clearly, all eyes can and should look for guidance to Parker in the Court of Appeals. Does the earlier Nonnon majority decision square with the standards later announced in Parker? That's likely what the Court of Appeals was considering in Nonnon before the Court decided to affirm "solely on procedural grounds."

In Nonnon, the Appellate Division panel majority held that reports and findings by plaintiffs' expert epidemiologists and toxicologists satisfied the Frye standard because "epidemiology itself is certainly not novel" and "toxicology is likewise not a novel field of science" and, for those reasons, issues of reliability with regard to the methodology used and conclusions reached by plaintiffs' experts are to be resolved by the jury.

But is that approach consistent with Parker where the Court said the "Frye test asks whether the accepted techniques, when properly performed, generate results accepted as reliable within the scientific community generally."⁵ Further, Parker dictates that threshold reliability must be present even when novel science is not involved. Thus, to the extent the Nonnon majority deemed epidemiology and toxicology "not novel" methodologies but left significant foundational reliability issues to go to the jury, is that approach consistent with Parker?

There also may be some question whether simply labeling the overall field of a particular science as "not novel" - for example, epidemiology or toxicology and thereby validating as "generally accepted" any and all procedures, techniques, theories, hypotheses or opinions a witness subjectively argues to be acceptable under that discipline - is consistent with Parker. Could any conclusion by any expert in the field of physics, for example, be admitted no matter how absurd or unfounded, because physics is a long-established field of science? Since, under Parker, gatekeeping for foundational reliability covers methodology as well as procedures and results, and since such foundational issues clearly are for the court, not the jury, how much weight is to be given to Nonnon's approach?

Recent Decisions

Considering the "time warp" described above, by being on the books (with a technical, procedural "affirmance"), the First Department's decision in Nonnon seems to have potential to influence a line of reasoning that may be in tension with some of Parker's fundamentals. For example, consider Diejoia v. Gacioch,⁶ a medical malpractice action, where the Fourth Department quoted Nonnon for the proposition that Frye is "not concerned with the reliability of a certain expert's conclusions" but only with whether the expert's deductions "are based on principles that are sufficiently established to have gained general acceptance as reliable."

However, in Jackson v. Nutmeg Technologies, Inc.,⁷ a claim for injuries based on occupational exposure to certain toxins in a building, the Third Department seems to have cited Nonnon only for limited purposes but followed Wesley and Parker for the fundamentals. Thus, Nonnon was cited for the premise that epidemiological studies "are by no means a novel methodology for demonstrating a causal relationship between a chemical compound and a set of symptoms or a disease." But then the Third Department relied on Parker and Wesley standards to analyze whether an appropriate foundation for the experts' opinions was shown. The court distinguished between foundational reliability going to admissibility and factual disagreements which go to the weight (citing both Wesley and Nonnon).

A particularly striking post-Nonnon decision is the First Department's opinion in Marso v. Novak,⁸ a medical malpractice case. Marso was issued on July 19, 2007, just weeks after the Court of Appeals in Nonnon decided "solely on procedural grounds." In Marso, the First Department referred to Nonnon as "our recent holding" but did not mention the subsequent history in the Court of Appeals. Of interest is that in Marso only one of the justices on the panel had been on the Nonnon panel. The Marso panel thus had four justices who were not involved in the Nonnon decision. Whether or not this was a factor, in Marso, the First Department seems to have backed away from the "methodology-only" approach to Frye articulated in Nonnon.

In Marso, plaintiff argued that his expert used the generally accepted methodology of "differential diagnosis," a process of elimination of other causes. Thus, since the doctor's methodology was generally accepted, plaintiff contended that, according to Nonnon, when properly performed, it should lead to admissible expert conclusions. The Marso panel said, however, this case prompts us to add 'but not when there is a generally or
widely held view in the scientific community rejecting such conclusions outright.'

Here, plaintiff's own expert testified that the result generated, which purportedly confirmed the expert's initial theory, was not accepted in the medical community. The Marso panel said that, to accept plaintiff's "methodology-only, ignore-the-conclusion" approach would circumvent the rationale for the Frye doctrine. Rather, "it is plaintiff's burden to show that his or her expert's theory is generally accepted in the relevant community." "Theory" obviously means more than merely naming the broad category of an established science or invoking the title of an accepted methodology.

Perhaps, the First Department's Nonnon and Marso rulings can be summed up as follows: In Marso, the Appellate Division clarified its earlier holding in Nonnon by noting that, in order to survive a Frye challenge, a proponent of expert testimony would have to demonstrate that not only the "methodology" used by the expert was generally accepted, but that the "conclusions" reached by the expert are generally accepted as well or, at least, that the conclusions have not been rejected.

Wesley and Parker teach us that, even when Frye does not apply because the scientific methodology is not novel, there nevertheless is a need for robust gatekeeping by the court in the form of a foundational inquiry into the specific reliability and trustworthiness of the procedures followed to generate the evidence proffered. This includes reliability of the expert's methodologies and opinions. This foundational screening presents a threshold admissibility question, not a jury question.

Michael Hoenig is a member of Herzfeld & Rubin.

Endnotes:
_________________________________________________________________________________________

1. 83 NY2d 417 (1994).

2. 7 NY3d 434 (2006).

3. See Fed.R.Evid. 702; Daubert v. Merrell Dow Pharmaceuticals, Inc., 509 U.S. 579 (1993); General Elec. Co. v. Joiner, 522 U.S. 136 (1997); Kumho Tire Co. v. Carmichael, 526 U.S. 137 (1999).

4. 32 AD3d 91 (1st Dept. 2006), Aff'd Solely on Procedural Grounds, 9 NY3d 825 (2007). The author of the instant column submitted an amicus curiae brief in the Court of Appeals on behalf of the Product Liability Advisory Council, in support of defendant.

5. Parker, 7 NY3d at 447. See People v. Hughes, 59 NY2d 523, 537 (1983), where the Court said that under Frye, scientific evidence will only be admitted if "the procedure and results are generally accepted as reliable in the scientific community." It seems that results also must conform to the reliability standard.

6. 42 AD3d 977, 979 (4th Dept. 2007).

7. 43 AD3d 599, 601-602 (3d Dept. 2007).

8. 42 AD3d 377 (1st Dept. 2007).