court declines to decide between Eveready and Squirt and lets a linguist in too

Therapeutics MD, Inc. v. Evofem Biosciences, Inc., 2022 WL 1013285, No. 20-CV-82296-RUIZ/REINHART (S.D. Fla. Mar. 30, 2022) (R&R)

The parties cross-moved to exclude each other’s experts who conducted consumer surveys to determine the likelihood of confusion between plaintiff’s mark (IMVEXXY) and defendant’s mark (PHEXXI).

An Eveready survey doesn’t prompt respondents with the plaintiff’s mark, only the accused product; it is “appropriate for testing alleged infringement [of] a ‘top of mind’ mark: one that is ‘highly accessible ... in memory, enhancing the likelihood that it will be cognitively cued by a similar junior use.’ ” It uses open-ended questions to detect assocation. A Squirt survey, by contrast, presents the consumer with multiple competing products, including the parties’ marks, and asks consumers whether they believe any two of the products are offered by the same company. The Squirt test’s “use of closed-ended and leading questions has been criticized by both courts and commentators because of the suggestive nature of those questions.” In addition, a Squirt study is less effective “when the consumer is presented with the competing products in a manner different than how those products appear in the actual marketplace.” “Nevertheless, experts who have conducted Squirt surveys are routinely permitted to present their findings to juries.”

Plaintiff’s Squirt survey found net confusion among providers/patients of approximately 20% between IMVEXXY and PHEXXI. Defendant’s Eveready survey found no material likelihood of confusion.

The Squirt survey: 300 prescribers and 800 patients saw plaintiff’s mark, IMVEXXY, “as seen in the real-world marketplace.” Then the survey posed “distraction questions” intended to “creat[e] a buffer period” and “reduce[ ] any potential order-effect bias” so that IMVEXXY receded from the respondents’ immediate short-term memory before being asked questions about other products. The prescriber test group was shown product samples of PHEXXI followed by two control brands (Lo Loestrin and Premarin), while the prescriber control group was shown LEPHEL2 followed by the same control brands. The patient-respondents saw the same product samples and product brochures. Questions asked whether they believed the products were “made by”; were “affiliated with or sponsored or approved by”; and “[r]equires permission or authorization from” “the company whose product you were shown in the first section of the survey.”

Defendant argued that Squirt was inappropriate given that prescription drugs are not presented to patients in the marketplace in close proximity to each other. It also argued that the survey failed to replicate market conditions by showing respondents actual product packaging, failed to randomize the order of the stimuli which “artificially inflated the level of confusion,” asked leading questions, and failed to use control names similar to the marks at issue. Defendant also argued that the only relevant consumers were physicians and pharmacists.

Eveready: 209 qualifying pharmacists were first asked to review a particular name in the context of “a pharmaceutical drug for women’s sexual and reproductive health.” The test cell used PHEXXI and the control used LEPHEL. Questions asked respondents to identify any brand or company that they believed (a) to be the source of the named drug, or (b) to be affiliated/connected with the named drug, or (c) needed to give permission or approval to the named drug. The survey also asked the pharmacists to identify any other pharmaceutical drugs they believed were put out by the same company or brand that puts out, or were affiliated/connected with, or needed to give permission/approval to the named drug. None of the participants named plaintiff or IMVEXXY. Pharmacists were chosen because, the surveyor reasoned, prescription drugs are controlled by prescribers, and pharmacists review and dispense the prescription.

In a rebuttal report, defendant’s expert did the same Eveready survey again, with 200 prescribers of contraceptive drugs, still with zero confusion.

Plaintiff argued that Eveready was inappropriate because despite IMVEXXY being a “strong brand” that has been “out for a few years in the market,” it has “not achiev[ed] a top-of-mind awareness” because the “level of marketing ... is minute.” In addition, defendant’s survey didn’t show product packaging to the survey participants and, plaintiff argued, used leading questions that wrongly limited the respondents’ thinking to drugs prescribed only for “women’s sexual and reproductive health.”

Finally, defendant’s expert did a modified Squirt survey with different randomization procedures, showing respondents the product packaging for IMVEXXY and PHEXXI as they are sold commercially (as opposed to product samples), and showing all six sides of the packaging. This, defendant argued, found “the absence of confusion” among prescribers, though plaintiff argued that it showed the prescribers packaging that they do not encounter in the marketplace.

The court decided to leave it all for the jury. Courts have accepted both methodologies, and the flaws in each one will be presented to the jury because both types will be offered to them. “Simply put, each party has satisfied its burden of showing that its expert should not be excluded based on Daubert.”

The court also declined to exclude the testimony of  a linguist with over 35 years of experience about the alleged orthographic and phonetic similarities of IMVEXXY and PHEXXI; she opined that the marks share numerous similarities, “including (i) similar letters; (ii) similar sounds; (iii) similar word stresses; (iv) interchangeability of some of the sounds; (v) unique distribution of other sounds (such as -xx-); and (vi) auditory weakness of the first syllable – [that] make it difficult to distinguish PHEXXI from IMVEXXY.”  [Ah, drugs with X in them, reminds me of my favorite case; search the opinion for "Mexican" to find the best anecdotal takedown of a statistical model you will ever read.] She was qualified to testify to those things and provided an adequate basis for her conclusions, and her testimony could help the jury, though her criticisms of the FDA and USPTO’s processes for approving drug names were stricken. Plaintiff didn’t show she was qualified to do that and her testimony wouldn’t help the factfinder.

Plaintiff was also allowed to present an expert on the costs of corrective advertising claimed as damages. Defendant’s expert could counter that, and was qualified to do so notwithstanding his lack of familiarity with judicial precedents on corrective advertising. He could also provide an alternative damage calculation based on an imputed royalty, which addressed the same subject matter—damages—to which plaintiff opened the door.