Defending against a failure-to-warn claim always presents unique challenges. Plaintiffs’ attorneys often file  design-defect claims, tacking on a failure-to-warn count as an afterthought.  It is common for plaintiffs’ attorneys to focus almost exclusively on developing their cases for defective design, which in turn puts the manufacturer in the position of focusing their defense on that particular claim. If manufacturers are not careful, those pesky warnings claims may still be there by the time trial rolls around. Thus, it is critical that manufacturers and their counsel are prepared to remind courts of their gatekeeping function under Daubert, and to close the gate on unreliable warnings opinions. This article takes a look at some successful attempts to exclude “warnings experts.”

Lack of Qualifications

While not always the most successful avenue for excluding a proposed expert, the first area to consider attacking is the expert’s qualifications. The focus should be on whether the expert is qualified to render an opinion about the particular product or warning method. For instance, in Calisi v. Abbott Labs., No. 11-1010671, 2013 U.S. Dist. LEXIS 139257 (D. Mass. Sept. 27, 2013), the plaintiff sued drug manufacturer Abbott Laboratories for failure to warn about the alleged risks of developing lymphoma from taking the drug Humira, a drug designed to treat rheumatoid arthritis. The plaintiff proffered their warnings expert as an expert in FDA regulations “with a specialization in drug labeling,” who was to testify regarding the adequacy of Humira’s label while the plaintiff was taking the drug.

At first blush, the proposed expert appeared qualified; he was a biochemist with a Ph.D. in pharmacology, had completed a postdoctoral fellowship in cancer research, and had even worked for the FDA as a pharmacologist reviewer and served as chief of regulatory affairs at the National Institutes of Health. But the court found that, under Daubert and Federal Rule of Evidence 702, he was not qualified to render an opinion about the adequacy of the warning. First, he was not a medical doctor, so he lacked “qualifications to opine on what is clinically appropriate in terms of treating patients.” Second, he was neither an expert nor had he received specialized training in psychology, cognitive decision making, or behavioral science. Thus, he could not show “how the label’s relevant target audience would interpret the Humira labels.” The court granted the manufacturer’s motion to exclude the warnings expert, and, as a result, awarded the manufacturer summary judgment.

Similarly, in Robertson v. Norton Co., 148 F.3d 905 (8th Cir. 1998), the plaintiff was injured when a power sander’s ceramic grinding wheel exploded. The plaintiff sued the manufacturer, claiming that the warnings were inadequate. The grinding wheel warned that “[i]mproper use may cause grinding wheel breakage and serious injury.” At trial, the plaintiff offered a ceramics expert who opined that the ceramic wheel’s warning label did not explain “improper use,” and its cross-reference to an ANSI publication was ineffective. The manufacturer’s attempt to exclude the expert failed at trial.

The circuit court reversed on appeal and provided the following reasoning:

Whether or not a given warning is adequate depends upon the language used and the impression that it is calculated to make upon the mind of an average user of the product. Questions of display, syntax, and emphasis are involved in evaluating a warning, or set of directions, and upon those matters [many experts in other areas] are not necessarily qualified to speak.

Thus, while the ceramics expert was qualified to testify about a manufacturing defect in the exploding ceramic grinding wheel, that did not qualify him as a warnings expert. His ceramics knowledge did not provide the necessary expertise on “questions of display, syntax, and emphasis” to assist the jury.

Lack of Reliable Methodology Due to Insufficient Testing of Causation Theory

Under Daubert, the first factor for determining the reliability of expert testimony is whether the expert’s theory can be and has been tested.  Product manufacturers should focus on this factor, because it often helps smoke out the ipse dixit, or “because I said so,” opinions. In Parker v. Brush Wellman, Inc., Nos. 1:04-CV-606, 1:08-CV- 2725, 2010 U.S. LEXIS 97702 (N.D. Ga. Sept. 17, 2010), the plaintiffs were employees of Lockheed Martin Corp., where they used beryllium-containing alloys to manufacture component parts for Lockheed. The plaintiffs allegedly developed beryllium-related sicknesses and sued the manufacturer of the alloys for failure to provide Lockheed with adequate warnings pertaining to the health risks associated with beryllium.

It is generally recognized that beryllium is the cause of the sickness alleged by the plaintiffs. Therefore, the manufacturer focused its Daubert attack on whether the plaintiffs’ expert had established specific causation as to the particular plaintiffs – i.e., could the expert show that the plaintiffs’ actual exposure level at Lockheed was sufficient to cause the sickness? The court agreed with the manufacturer, and found that the proposed expert could not establish causation stemming from the plaintiffs’ exposure at Lockheed, and thus, could not establish causation resulting from the manufacturer’s alleged failure to warn Lockheed. Specifically, the expert did not sufficiently test his theory, stating that “only industrial hygiene sampling for a period of time can document the actual exposure.” Since Lockheed had since implemented controls that were not in place at the time of the alleged beryllium exposures, and the sampling did not occur until after those exposures, those samples were not an appropriate test. Furthermore, the plaintiffs’ proposed expert could have placed the products in an environmental chamber and reenacted the relevant machining procedures that the plaintiffs performed, but failed to do so. Because the expert did not perform sufficient testing to establish causation due to elevated beryllium exposure at Lockheed, his theory that the failure to warn Lockheed of beryllium’s health risks caused the plaintiffs’ illnesses was deemed unreliable and excluded.

Failure to Establish Known Error Rate

Akin to the lack-of-testing factor is Daubert’s consideration of whether an expert’s theory has a known error rate. Manufacturers should be quick to point out that if a plaintiff’s warnings expert has not established an error rate for his theory, his opinions are unreliable. For instance, in Shepherd v. Michelin Tire Corp., 6 F. Supp. 2d 1307 (N.D. Ala. 1997), a laborer was fatally injured when he attempted to place a 16-inch tire on a 16.5-inch rim, causing the tire to explode. Even though the existing warning on the tire warned against placing smaller tires on larger rims, the estate brought a warning claim. The plaintiff’s expert was eminently qualified, having written several scholarly articles on the subject of warnings, and having studied and testified regarding the theory of  human factors related to warnings. In this case, he opined that the tire lacked adequate warnings and that if the tire contained his proposed alternative, or “candidate” warning, the injury would not have occurred.

The trial court held a Daubert hearing and excluded the testimony. The expert’s proposed warning system was, in part, to include posters and pamphlets describing the potential dangers of inflating tires on mismatched rims. However, he did not sufficiently test his theory, and thus, the court was concerned that he could not provide an error rate for his methodology. As a result, the plaintiff’s expert “[could] not say that the degree of effectiveness of his warning scheme would exceed 50% in cases such as this one.” The court further reasoned that “[s]imply concluding, as is reasonable, that a more prominent warning would decrease the danger, although plausible, does not make such a conclusion ‘relevant’” in a case where the plaintiff bears the burden to prove causation.

Lack of “Fit” With the Facts of the Case

The Daubert factors are not exhaustive, and courts must customize their inquiries to the facts of each particular case. Indeed, so must the proposed warnings expert. In Jaurequi v. Carter Mfg. Co., 173 F.3d 1076 (8th Cir. 1999), the plaintiff brought design and warning claims against the manufacturer of a corn harvesting tool after an accident in which his legs were amputated by one of the tool’s components. In support of his warning claims, the plaintiff offered a human factors engineer with a doctorate in experimental psychology to opine that the tool was defective because the original warning signs were “too small, too far from the point of danger, and oriented at an angle which made them difficult to read.”  He testified that the warnings should have been accompanied by stripes and chevrons to draw the user’s attention.

But the Eighth Circuit affirmed the district court’s decision to exclude the proposed warnings expert.  According to the court, there was a lack of “fit” between the testimony and the facts of the case. First, the proposed expert failed to take into account that the warnings that originally accompanied the tool had been painted over prior to the plaintiff’s use. When pressed on this issue, the proposed expert could not say whether his suggested warnings would have also been painted over. As a result, the plaintiff could not establish that the allegedly defective warnings proximately caused the accident.


Even if plaintiffs focus almost exclusively on developing their design-defect claims, manufacturers should not be caught off guard when plaintiffs offer a warnings expert prior to trial.  While this article focused on a select number of Daubert factors, be prepared to vigorously apply each of the Daubert reliability criteria to the proposed expert’s opinions. By doing so, courts will be more inclined to slam the gate on unreliable warnings opinions.