Opinion
No. 87 Civ. 3712 (WCC).
December 1, 1987. As Amended December 4 and 7, 1987.
Patterson, Belknap, Webb Tyler, New York City, for McNeilab, Inc.; David F. Dobbins, Gregory L. Diskant, Frederick B. Campbell, Janet Cohn, of counsel.
Arnold Porter, Washington, D.C., for American Home Products Corp.; Stuart J. Land, Steven P. Lockman, of counsel.
Morrison Foerster, New York City, for American Home Products Corp.; Jack C. Auspitz, Kim J. Landsman, Charles F. Hagan, William P. Woods, Egon E. Berg, American Home Products Corp., of counsel.
OPINION AND ORDER
This action under § 43(a) of the Lanham Trademark Act of 1946, 15 U.S.C. § 1125(a), and the New York General Business Law, charging false claims of safety in the advertising of defendant's over-the-counter ("OTC") internal analgesic Advil, is before the Court on plaintiff's motion for preliminary injunction. For the reasons stated hereinafter, the motion is granted.
Factual and Procedural Background
Plaintiff McNeilab, Inc. ("McNeil"), a wholly-owned subsidiary of Johnson Johnson, manufactures and sells a variety of pharmaceutical and health-care products including an OTC internal analgesic, Tylenol, whose principal active ingredient is acetaminophen.
Defendant American Home Products ("AHP") manufactures and sells a line of competing products, including an OTC internal analgesic, Advil, whose principal active ingredient is ibuprofen.
Since the commercial introduction of Advil in 1984, the parties have battled fiercely both in the marketplace and in the courts, each flooding the media with advertisements claiming equal or superior efficacy and safety of its product, and the courts with claims of false advertising by its adversary.
When the present action was brought on May 29, 1987, the parties were already before the Court in American Home Products Corp. v. Johnson Johnson, 654 F. Supp. 568 (S.D.N.Y. 1987), in which each party charged the other with manifold false or misleading claims of efficacy and safety. On February 25, 1987, the Court filed its opinion in that case, finding that certain of the advertising claims of each party were false or misleading. The damage issues still remain to be tried.
In its original complaint in this action, McNeil charged that shortly after the Court's decision in that case, AHP began a nationwide advertising campaign in which a consumer (customarily portrayed by a professional actor or actress) is pictured saying, in haec verba or in substance, "Like Tylenol, Advil doesn't upset my stomach."
McNeil charged that such claims are false and/or misleading in that they create the erroneous impression that, insofar as concerns the hazard of all types of adverse gastrointestinal effects, Advil is the equal of Tylenol. McNeil further alleges that, as this Court found in the prior action, the mechanisms of action of acetaminophen and ibuprofen are entirely different, with the latter posing a substantial threat of serious, though frequently asymptomatic, gastrointestinal damage, such as ulcers and internal bleeding, in a limited but nonetheless significant number of users who are predisposed thereto.
In its answer, AHP contended that its challenged advertising merely stated what this Court also found in the prior action: that insofar as concerns the likelihood of minor, temporary symptoms of stomach distress (heartburn, queasiness, flatulence and the like) Advil and Tylenol, in recommended OTC dosages, are substantially equivalent.
On July 2, 1987, McNeil filed a First Amended Complaint adding similar charges with respect to a new advertisement in AHP's campaign.
In both its original and amended complaints, McNeil sought a preliminary injunction to prevent repetition of the advertisements in question. The Court conducted an evidentiary hearing on such a motion on July 27 and 28, 1987. On the first day of the hearing, McNeil introduced evidence of a survey designed to determine whether a substantial number of consumers understand the expression "stomach upset," as used in AHP's advertisements, to refer to all adverse effects on the stomach, including serious, objective, possibly asymptomatic effects, and not merely to minor, temporary subjective symptoms.
At the beginning of the third day of the hearing, AHP's counsel, acting in response to a suggestion made by the Court at the end of the preceding day, filed with the Court a written undertaking to discontinue all advertising in which Advil is claimed to be the equivalent of Tylenol in the respect of freedom from "stomach upset," without modifying that term by such adjectives as "minor," "temporary," "occasional," "simple" or "mild." Based upon that undertaking the Court denied the motion for preliminary injunction, finding that there was no evidence before the Court on which it could make the findings necessary to support such an injunction: likelihood of success on the merits or a balance of hardships tipping decidedly in favor of the party seeking such relief.
The Court took pains to point out that its decision concerned only the modified form of advertising claim which AHP had undertaken to use. All of the survey evidence which McNeil had introduced at the hearing, of course, involved the advertising claims which were being discontinued. The Court specifically emphasized that this was not a decision on the merits of the case, but merely a decision that, on the evidence then before the Court, it could not conclude that there was a likelihood that McNeil would succeed in establishing that AHP's modified claims were false or misleading.
On July 31, 1987, the Court granted McNeil leave to file a Second Amended Complaint to cover AHP's modified advertising claims, when they appeared. AHP furnished McNeil advance copies of the modified advertisements and on August 14, 1987, McNeil filed its Second Amended Complaint, charging that both the original and modified advertising claims were false and/or misleading and seeking a preliminary injunction preventing their continued use. An evidentiary hearing on McNeil's new motion for preliminary injunction was conducted in October 1987. This opinion incorporates the Court's findings of fact and conclusions of law thereon, pursuant to Rule 52(a), Fed.R.Civ.P.
Discussion
Since 1984, when AHP introduced Advil to the U.S. market, a significant feature of Advil advertising has been an assurance of its safety and gentleness to the stomach. Frequently this took the form of a claim that Advil is "gentler to the stomach than aspirin," a claim which this Court has specifically found to be true. Another common version was a television commercial in which an actor stated, "Advil didn't upset my stomach." The public is accustomed to such dramatized personal claims, and recognizes that the actor speaking these lines may never actually have taken Advil but is merely conveying the message that typical users do not suffer gastric distress as a result of taking Advil — a message which this Court has likewise found to be true. McNeil has never complained about such advertising and does not do so now.
In the spring of 1987, however, AHP began the new television campaign in which the actor proclaimed, "Like Tylenol, Advil doesn't upset my stomach" (emphasis added). McNeil brought the present action to enjoin this comparison of Advil to Tylenol in the respect of their tendency to cause "stomach upset," contending that this term is understood by a substantial portion of the consuming public to refer not only to the subjective symptoms of minor and temporary gastric distress, but also to more serious problems such as ulceration and occult bleeding which in the early stages may not produce symptoms sensible to the user.
As previously noted, after the first preliminary injunction hearing, AHP voluntarily modified the script of the Advil commercials to read instead, "Like Tylenol, Advil doesn't give me minor stomach upset," or, "And like Tylenol, with Advil I don't have to think about occasional stomach upset," or, "And like Tylenol, with Advil minor stomach upset isn't a problem for me." (emphasis added).
McNeil charges that these new commercials are no better than the ones they replaced — that most listeners either do not notice the qualifying adjectives "minor" or "temporary" or, if they do, take away the message that Advil doesn't cause any type of stomach distress, because if it doesn't cause even "minor" or "temporary" problems, it is logical to assume that it won't cause major or permanent problems. This is what has been referred to as the "if it won't nick, it won't cut" rationale.
At the second preliminary injunction hearing, McNeil introduced various types of evidence designed to show that the term "stomach upset" is widely understood to cover all types of gastric distress. Dr. David Graham, a gastroenterologist, testified that his patients use the term to describe every type of gastric problem including ulcers. McNeil also introduced dictionaries defining "upset" as meaning "disturb normal functioning" or "disturb physically," popular medical guides using the expression interchangeably with stomach ache, heartburn and nausea, and press clippings in which the term is used with reference to the symptoms of stomach ulcers.
McNeil also submitted what it terms "evidence of actual confusion" in the form of a letter which a woman had written to AHP complaining that the television commercials claiming that Advil is "as gentle as Tylenol to the stomach" had caused her to use Advil for sinus headaches and that "After one month I landed up at the Doctors" because "Advil was eating the lining of my stomach." She added that "I do have an ulcer but I switched from Tylenol to Advil because of your advertising" and that she "had to be put on Zantac" to alleviate the pain that the commercials had caused. As AHP points out, the writer of this letter, Mrs. Evelyn Hadoulis, testified on deposition that she switched from Tylenol to Advil in 1986, long before the "like Tylenol" commercials began, so that she could not possibly have been induced to switch by those particular commercials. Nevertheless, the letter has some significance because it illustrates the type of injury that may result from misplaced reliance on generalized claims of stomach safety.
As other evidence of "actual confusion" McNeil points to the fact that all three major television networks initially refused to broadcast AHP's "like Tylenol" commercials, citing this Court's decision of February 25, 1987 and other evidence that Advil and Tylenol are not "like" but pharmacalogically different. It was only after considerable pressure from AHP that the networks relented and agreed to air these commercials. This is obviously not evidence of actual confusion of consumers, but nonetheless tends to show that the "like Tylenol" claim has the potential for misleading those who hear it into the belief that Advil's effects on the stomach are no different from those of Tylenol.
However, McNeil places its principal reliance on three public opinion surveys in which respondents were shown AHP's revised television commercials and asked open-ended questions as to the message they communicated. In answer to these questions, 21% to 34% of the respondents stated, in substance (though they chose various wording), that they understood the commercials to mean that Advil did not adversely affect the stomach, and that it was safe to use. Almost none of the respondents mentioned the modifying adjectives "minor" or "temporary" in their playback of the message they received.
In one of the three surveys (the "tan" survey, PX 102), all of the respondents were persons who had been treated for stomach ulcers. To the open-ended question as to the message they took from the Advil commercials, 31% of them answered, in substance, that Advil would not harm the stomach. And to a closed-end question whether they understood that Advil was safe for them to use, 85% responded affirmatively.
In the three surveys, between 34% and 45% of the respondents made the surprising statement that they got the further message that Tylenol upsets the stomach!
The last figures, indicating a perception that Advil is not merely equal but superior to Tylenol in the respect of adverse gastric effects, is obviously not attributable to the "like Tylenol" claim itself, but to other aspects of the commercials such as the "rolling pills" sequence. Each of the commercials shows first an aspirin tablet with the voice-over "I used to take aspirin." Then the aspirin tablet rolls away and its place is taken by a Tylenol tablet while the voice-over continues ". . . or Tylenol." The Tylenol tablet in turn rolls away to be replaced by an Advil tablet as the sound track announces, "But today it's Advil." In the longer (30-second) commercials, the sound track continues, "I learned about Advil from my doctor," or "My doctor told me to try Advil for muscle aches," and "It contains the same medicine as the prescription brand Motrin. You know, just one Advil is as effective as two regular strength Tylenol." Each commercial concludes, "Advil. Advanced medicine for pain."
The overall impression given, and obviously the one intended, is that Advil represents the current state of the art, a definite advance over Tylenol, which had been a step forward from aspirin.
McNeil does not complain of such comparative claims per se, perhaps because they are typical of the kind of commercial puffery which the purchasing public is conditioned to view with skepticism. However, McNeil charges that Advil commercials which not only create the impression that Advil represents an advance over Tylenol but also make a specific reference to gastric side effects, tend to give many listeners the erroneous impression that Advil is superior to Tylenol in that respect among others. The McNeil surveys tend to support that charge.
AHP, while conceding that properly conducted opinion surveys are the best means of determining how advertising is perceived by consumers, challenges the reliability of McNeil's surveys on a multiplicity of grounds.
AHP first argues that the "tan" survey (PX 102) should be rejected outright for lack of a "reasonably representative" sample. AHP does not complain that the respondents in that survey consisted entirely of persons who had been treated for gastric ulcers, nor could it effectively do so. Obviously ulcer patients are not typical OTC analgesic users; their preoccupation with their gastric problems would likely cause them, when hearing a commercial for an OTC internal analgesic, to pay heightened attention to any claims respecting the likelihood of adverse gastric effects. Moreover, persons having a predisposition to irritation of the gastric mucosa are the very ones who would suffer if they were misled into believing that they could use Advil as safely as Tylenol, and it is particularly desirable, if not indispensable, to know how those especially at risk perceive the advertising claims in issue.
But AHP does complain that the persons interviewed in the "tan" survey were selected from the records of previous surveys in the files of the survey organization. Thus they were not "reasonably representative of the target universe." AHP argues that, having previously agreed to be interviewed, these persons are likely to be "yea-sayers" who will say whatever they think the interviewer wants to hear. However, everyone who participates in any survey must consent to be interviewed and there is no reason to expect that a person who participated in a prior survey would more likely be a "yea-sayer" than one consenting to participate in the present survey. Moreover, since the respondents are not informed of the identity of the survey sponsor, they have no way of knowing what answers the interviewer wants.
AHP further complains that the survey sample — 149 persons — was too small a base from which to project conclusions to the entire universe of ulcer patients, and that the selection of interviewees from in-house files could not have produced a typical cross-section of that universe. AHP insists that, at the least, a random "mall intercept" or telephone directory selection process should have been used. For these reasons, AHP contends, the tan survey must be disregarded altogether.
The Court cannot agree. It is true that the sample was relatively small; however, this does not require that the results be discarded outright; it merely means that the results cannot be treated as a relatively precise numerical determination which can be reliably extrapolated to the target universe, but only as an indication whether the commercials have the tendency to mislead a substantial number of viewers.
It is also true that the use of "household lists" is not the optimum sampling process, but the time between two preliminary injunction hearings was severely limited and McNeil was looking for a narrowly-defined group: persons who had been diagnosed by a licensed physician as having stomach ulcers within the past five years and who were users of OTC internal analgesics. Those who had participated in any market survey within the past six months or in any OTC analgesic study within the past year were excluded. It would surely have taken many thousands of random telephone calls to locate 149 eligible persons who would consent to be interviewed. And there is no reason to expect that the persons so selected would have been any more representative of the target universe than those chosen from the household lists, which had presumably been selected originally in a random fashion, either by a mall intercept or telephone directory method. Regional bias was avoided by selecting persons from seven cities scattered around the country. In any event, this criticism applies only to the tan survey. The other two were typical mall intercept surveys, as to which AHP has not criticized the sampling process. Significantly, the results of the other two surveys are not markedly different from those of the tan survey.
AHP also makes much of the fact that McNeil, in its computation of the results of the tan survey (PX 117), counted among the respondents who were "misled" by AHP's commercials those who stated that they had received the message that Advil does not "irritate" the stomach. AHP points out that in the original tabulation of the results of this survey, McNeil's surveyor, acting on McNeil's specific instructions, had included among those "misled" only respondents who had used in their answers one of the four specific words, "hurt," "harm," "damage" or "safe." This was apparently done with the conservative objective of limiting the "misled" group to those who had received a message which clearly denoted serious, long-term damage.
This almost certainly resulted in understatement of the number who were misled, because the term "irritate" might appropriately be used to describe the exacerbation of a pre-ulcerous or even an ulcerous condition. However, AHP argues that only survey results which are unambiguous should be relied upon in granting a preliminary injunction.
Moreover, as AHP points out, it was not until after the Court had criticized, as leading, questions 7 and 8 of the McNeil survey questionnaires that McNeil recomputed the survey results to eliminate the responses to these question. AHP argues that this reduced the percentage of respondents who were "misled" to such an unimpressive level that McNeil found it necessary to expand the group by including respondents who took away the message that Advil does not "irritate" the stomach.
AHP also challenges all three of McNeil's surveys on the ground of "order bias," contending that the answers to question 6 (as to whether the commercial communicated a message concerning "Tylenol and the stomach") were influenced because that question was preceded by question 5 (a similar question about "Advil and the stomach"). AHP argues that because question 5 was usually answered by saying that the commercial conveyed the message that Advil does not upset (or harm or irritate) the stomach, there was a natural tendency to say something different in answer to question 6, for example, that Tylenol does upset (or harm or irritate) the stomach.
However, in response to questions 1, 2 and 4 of the three surveys, before Tylenol had been mentioned in any of the questions, between 8% and 13% of the respondents stated that the commercials communicated to them that Advil is superior to Tylenol with respect to their effects on the stomach. AHP argues that these percentages are statistically insignificant, particularly in surveys involving so few interviews.
AHP's argument would be more convincing if McNeil's case depended upon showing that a substantial fraction of the audience received the message that Advil is superior to Tylenol insofar as gastric effects are concerned. But McNeil does not have to prove that the commercial conveys a message of superiority. All it needs to do is to prove that a substantial number of listeners take away the message that Advil is the equal of Tylenol in the respect of their overall effects on the stomach. This Court has already found that any such message would be false.
AHP also points out that an earlier McNeil survey (the "yellow" survey, DX AA), question 5 was an open-ended question whose wording had been expressly approved by the Court before the survey was conducted. It asked simply, "What does the term `stomach upset' as used in the commercial communicate to you?" In their answers, virtually all of the respondents used terms which were consistent with minor, temporary stomach conditions. This caused the Court to remark at the July hearing, when the yellow survey was introduced, that it tended to show, if anything, the "opposite" of what McNeil contended, that is, "that the average man in the street thinks of stomach upset as a fairly mild and transitory discomfort." AHP contends that these earlier survey results prove the unreliability of the later surveys.
AHP's criticisms of McNeil's later surveys are not without significant merit. However, even discounting the results of these surveys in view of their flaws, they nevertheless establish, although not in precise or even approximate quantitative terms, that a substantial fraction of those who hear the Advil commercials in question get the impression that Advil is comparable to Tylenol in the respect of freedom from adverse effects on the stomach, including not only minor and temporary discomfort, but also more serious and lasting harm. The survey evidence is supported by the other evidence adduced by McNeil and by common sense. The "won't nick, won't cut" rationale, despite AHP's belittling of it, is no doubt a factor which contributes to the misleading effect of the commercials. It is logical to assume that if Advil won't cause even minor, temporary upset, it surely won't cause serious, chronic damage.
If it were not for their "like Tylenol" claim, the Advil commercials would probably fall within the acceptable range of commercial puffery. But, as the Court has already found after a full trial, Advil is not "like Tylenol" in the respect of their effects on the stomach. The evidence introduced at the two preliminary injunction hearings has convinced the Court that at the least, there is a probability that at a trial on the merits, McNeil can establish that AHP's current television advertising campaign tends to mislead a "not insubstantial" fraction of the consumer audience to the erroneous belief that the use of Advil poses no greater hazard of adverse effects on the stomach than the use of Tylenol. Such misleading advertising is a violation of § 43(a) of the Lanham Act. American Home Products Corp. v. Johnson Johnson, 577 F.2d 160, 165-66 (2d Cir. 1978). See also Vidal Sassoon, Inc. v. Bristol-Myers Co., 661 F.2d 272, 275-76 (2d Cir. 1981).
The legal requirements for a preliminary injunction are firmly established: (a) irreparable harm and (b) either (1) a likelihood of success on the merits or (2) sufficiently serious questions going to the merits to make them a fair ground for litigation and a balance of hardships tipping decidedly toward the party requesting the preliminary relief. Jackson Dairy, Inc. v. H.P. Hood Sons, Inc., 596 F.2d 70, 72 (2d Cir. 1979).
Where false or misleading advertising in violation of the Lanham Act is shown, irreparable harm is presumed. Coca-Cola Co. v. Tropicana Products, Inc., 690 F.2d 312, 317 (2d Cir. 1982). And where, as here, the false or misleading advertising claims create a danger to public health, the presumption of irreparable harm is particularly appropriate. McNeilab, Inc. v. American Home Products Corp., 501 F. Supp. 517, 540 (S.D.N.Y. 1980); Syntex Laboratories, Inc. v. Norwich Pharmacal Co., 315 F. Supp. 45, 53 (S.D.N.Y. 1970), aff'd, 437 F.2d 566 (2d Cir. 1971).
Furthermore, as discussed above, McNeil has established the strong probability of its ultimate success on the merits. Thus the requirements of a preliminary injunction are clearly satisfied. The injunction shall restrain the continued publication of any advertisements stating or implying that Advil is "like Tylenol" in the respect of adverse effects on the stomach. However, the injunction shall not prevent AHP from advertising that Advil does not cause stomach upset, without a comparison to Tylenol.
McNeil's motion for preliminary injunction is granted to the extent indicated. Settle order on notice.
SO ORDERED.