1 Most drug injury lawsuits are pled as failure to warn claims. Steven Garber, Rand Inst. for Civil Justice, Economic Effects of Product Liability and Other Litigation Involving the Safety and Effectiveness of Pharmaceuticals, xiii (2013) (noting the “lion’s share” of mass tort claims are brought as failure to warn cases). Some drug injury lawsuits involve manufacturing defects, as in the case of Fentanyl lawsuits discussed herein, involving a defect in a transdermal patch.

2 Id. at 24; see also Hensler, Deborah R. & Peterson, Mark A., Understanding Mass Personal Injury Litigation: A Socio-Legal Analysis, 59 Brook. L. Rev. 961, 965 (1993)Google Scholar (explaining the structure of mass tort claims); Silver, Charles & Miller, Geoffrey P., The Quasi-Class Action Method of Managing Multi-District Litigations: Problems and a Proposal, 63 Vand. L. Rev. 107, 115 (2010)Google Scholar (“The largest [multidistrict litigations] encompass thousands of cases filed by legions of attorneys.”).

3 In mass tort claims, judges select managerial attorneys. However, the “legion” of other lawyers representing the thousands of clients continue to participate. Their fees are reduced to reflect their client's share of the managing attorneys' fees, and judges sometimes set fee caps on the contingency fees of non-managerial attorneys. See Garber, supra note 1, at 80.

4 This assertion is based on a docket search on Bloomberg Law for the firm’s name and product liability cases. The search revealed any cases where that firm appeared on a case docket in federal court and certain state courts. A docket search on Bloomberg Law for product liability cases revealed that five of the ten most prolific advertisers listed in Table 5, infra, produced more than 100 results. The other five advertisers produced twenty or fewer results.

5 Garber, supra note 1, at xv (“The potential cost of a mass tort to a drug company is in the billions of dollars.”).

6 See Julie Beck, Bayer Designates $1.5 Billion for Litigation over Yasmin and Yaz Contraceptives, Inside Couns. (Mar. 4, 2013), http://www.insidecounsel.com/2013/03/04/bayer-designates-15-billion-for-litigation-over-ya.

7 See supra text accompanying note 4. Of these rare litigators, one firm produced no results, and two firms produced only two results on the Bloomberg docket search. It is possible, however, that these firms litigate in state courts excluded from Bloomberg’s database.

8 See Engstrom, Nora Freeman, Sunlight and Settlement Mills, 86 N.Y.U. L. Rev. 805, 807 (2011)Google Scholar (discussing the rise and benefits of “settlement mills”).

9 See Hensler & Peterson, supra note 2, at 1026 (“Many law firms that advertise [in the mass tort context] serve only as referring lawyers….”); Witt, John Fabian, Bureaucratic Legalism, American Style: Private Bureaucratic Legalism and the Governance of the Tort System, 56 DePaul L. Rev. 261, 286 (2007)Google Scholar (“Many lawyers who advertise as personal injury specialists are little more than referral mills. They serve as intake officers for claims that they then farm out to specialized lawyers in return for a contingent referral fee…. [T]hey seem to thrive in an uncompetitive legal fees market that sometimes functions a little like a gold rush–locating and signing up clients is the key.”).

10 For example, a disclaimer in an ad from The Goldwater Law Firm states “Robert Goldwater (15849 N. 71^st St., Suite 100, Scottsdale, AZ) is licensed to practice law only in Arizona, but associates with attorneys throughout the country. Principle responsibility for cases belongs to associate counsel although Robert Goldwater maintains joint responsibility. Not available in all states.” The Goldwater Law Firm’s website includes a section called “Co-Counsel Opportunities”, which states that it is “one of the largest national advertising law firms in the United States and generate in excess of 10,000 new client inquiries per month. We co-counsel with experienced lawyers throughout the country and are always looking for new law firms to work with. In fact, most of our national advertising has been the direct result of other law firms approaching us with ideas on specific niche campaigns. If you would like to generate more tort cases for your law firm, please call attorney Bob Goldwater….” Co-Counsel Opportunities, Goldwater L. Firm, P.C., http://www.bobgoldwater.com/Co_Counsel_Opportunities.aspx (last visited Mar. 25, 2015).

11 See Model Rules of Prof'l Conduct R. 7.2(b)(1) (2011).

12 See Goldwater L. Firm, supra note 10.

13 See Bayer, Yaz Full Prescribing Information 8 (2012), available at http://labeling.bayerhealthcare.com/html/products/pi/fhc/YAZ_PI.pdf?WT.mc_id=www.berlex.com.

14 See generally Hansen, Richard A. et al., Relationship of Pharmaceutical Promotion to Antidepressant Switching and Adherence: A Retrospective Cohort Study, 61 Psychiatric Services 1232 (2010)CrossRefGoogle ScholarPubMed (examining the effects of direct-to-consumer advertising for antidepressants); Hosken, Daniel & Wendling, Brett, Informing the Uninformed: How Drug Advertising Affects Check-up Visits, 31 Int’l J. Indus.Org. 181 (2013)CrossRefGoogle Scholar (studying whether advertisements encourage consumers to seek treatment); Law, Michael R. et al., Effect of Illicit Direct to Consumer Advertising on Use of Etanercept, Mometasone, and Tegaserod in Canada: Controlled Longitudinal Study, 337 Brit. Med. J. 557 (2008)CrossRefGoogle ScholarPubMed (examining how direct advertisement in the United States affects Canadian prescribing rates); Mintzes, Barbara et al., Influence of Direct to Consumer Pharmaceutical Advertising and Patients’ Requests on Prescribing Decisions: Two Site Cross Sectional Survey, 324 Brit. Med. J. 278 (2002)CrossRefGoogle ScholarPubMed (reviewing the relationship between direct-to-consumer marketing and patient pharmaceutical requests); Bennett Parnes et al., Lack of Impact of Direct-to-Consumer Advertising on the Physician-Patient Encounter in Primary Care: A SNOCAP Report, 7 Annals Fam. Med. 41 (2009) (considering the effects of direct-to-consumer advertisements on clinical encounters); Wosinska, Marta, Direct-to-Consumer Advertising and Drug Therapy Compliance, 42 J. Marketing Res. 323 (2005)CrossRefGoogle Scholar (finding direct-to-consumer advertisements can increase compliance with a drug regimen).

15 See generally Harris Interactive, Pharmaceutical Liability Study Report on Findings 39 (2003), available at https://www.uschamber.com/sites/default/files/legacy/press/rx_pharmaceutical_liability_study_report.pdf; Juurlink, David N. et al., The Effect of Publication on Internet-Based Solicitation of Personal-Injury Litigants, 177 Can. Med. Assoc. J. 1369, 1369 (2007)CrossRefGoogle ScholarPubMed (examining the relationship between the publication of major scientific studies and attorney solicitations on the internet); Koski, Michelle et al., Patient Perception of Transvaginal Mesh and the Media, 84 J. Urology 575, 575 (2014)CrossRefGoogle ScholarPubMed; Judyth Pendell, The Adverse Side Effects of Pharmaceutical Litigation 1 (2003), available at https://www.mysciencework.com/publication/read/2199471/the-adverse-side-effects-of-pharmaceutical-litigation#page-null; Tenggardjaja, Christopher F. et al., Evaluation of Patients’ Perceptions of Mesh Usage in Female Pelvic Medicine and Reconstructive Surgery, 85 J. Urology 326, 326 (2015)CrossRefGoogle ScholarPubMed.

16 See generally Koski, supra note 15; Tenggardjaja, supra note 15.

17 Tenggardjaja, supra note 15, at 328.

18 Koski, supra note 15, at 578.

19 Harris Interactive, supra note 15; Pendell, supra note 15.

20 Harris Interactive, supra note 15, at 14.

21 Id.

22 New Survey Shows Product Liability Litigation May Jeopardize Treatment Outcomes for People with Severe Mental Illnesses, Eli Lilly & Co. (June 13, 2007), https://investor.lilly.com/releasedetail.cfm?releaseid=248836.

23 Daniel M. Schaffzin, Warning: Lawyer Advertising May Be Hazardous to Your Health! A Call to Fairly Balance Solicitation of Clients in Pharmaceutical Litigation, 8 Charleston L. Rev. 319, 343-46 (2013-2014).

24 Alan Schwarz, One Drug or 2? Parents See Risk but Also Hope, N.Y. Times (Nov. 14, 2014), http://www.nytimes.com/2014/11/15/us/one-drug-or-2-parents-see-risk-but-also-hope.html?_r=0 (“Then she saw a television commercial in which class-action lawyers asked parents if their son had developed breasts while on the drug, a rare but disturbing side effect.”).

25 See e.g., David Oliver, Bipolar Disorder Warning About Lawsuits and Attorneys, Bipolar Cent. (June 30, 2007, 7:32 AM), http://www.bipolarcentral.com/supporterblog/2007/06/bipolar-disorder-warning-about-lawsuits.html (describing mental illness blogger's conversation with acquaintance who stopped taking a drug after seeing a drug injury ad; comments to post included reports of similar behavior); Yaz & Gall Bladder Failure, MyFitnessPal (Sept. 10, 2009), http://www.myfitnesspal.com/topics/show/65554-yaz-gall-bladder-failure (describing one commenter's plan to consult her doctor about the risks, while another reported that she “stopped taking [the drug] the SECOND [she] heard that”).

26 See Brief of the FTC as Amicus Curiae Supporting Arguments to Vacate Opinion 39 of the Committee on Attorney Advertising Appointed by the Supreme Court of New Jersey at 11, In re Petition for Review of Committee on Att'y Adver. Op. 39, No. 60,003 (F.T.C. May 8, 2007), 2007 WL 7947967, at *4 [hereinafter Brief of the FTC] (citing empirical evidence that advertising for professional services is associated with lower prices and improved quality); Abel, Richard L., The Real Tort Crisis–Too Few Claims, 48 Ohio St. L.J. 443, 448-57 (1987)Google Scholar (describing the social importance of tort claims, and the tendency of tort victims to avoid bringing claims); Engstrom, Nora Freeman, Attorney Advertising and the Contingency Fee Cost Paradox, 65 Stan L. Rev. 633, 665-66 (2013)Google Scholar (arguing that attorney advertising has not reduced the cost of legal services); Hazard, Geoffrey C., Jr. et al., Why Lawyers Should be Allowed to Advertise: A Market Analysis of Legal Services, 58 N.Y.U. L. Rev. 1084, 1101 (1983)Google Scholar (arguing that advertising may be especially suitable for clams that benefit from economies of scale); Lidsky, Lyrissa Barnett & Peterson, Tera Jckowski, Medium-Specific Regulation of Attorney Advertising: A Critique, 18 U. Fla. J.L. & Pub. Pol’y 259, 263 (2007)Google Scholar; Mensoian, Christopher M., Bates, the Model Rules and Attorney Advertising, 32 McGeorge L. Rev. 77, 80 (2000)Google Scholar (describing a 1984 FTC study that found that competition in advertising reduced the price of legal services).

27 Zauderer v. Office of Disciplinary Counsel of the Supreme Court of Ohio, 471 U.S. 626, 634 (1985) (recounting testimony by an expert witness that attorney advertising is socially valuable in informing the public “of the potential health hazards associated with the Dalkon Shield”). An internet-based study of search result hits for websites soliciting plaintiffs for drug injury claims found that publication of a medical study highlight the risk of an adverse event was associated with “an immediate and sustained increase in [such] websites.” Juurlink et al., supra note 15, at 1369. A marketing company specializing in mass tort advertising advises attorneys interested in pursuing a drug injury practice to monitor the FDA website and the Journal of the American Medical Association for potential adverse events that could form the basis of a mass tort. See Risk vs. Reward: The Mass Tort Advertising Cycle, MCM Servs. Grp. (Aug. 13, 2013, 3:13 PM), http://mcmservicesgroup.com/home/attorney-marketing-blog-mcm-services-group.

28 A minority of states recognize a cause of action for “medical monitoring.” This claim allows attorneys to sue on behalf of individuals never injured by the drug or device to recover the costs of medical monitoring through “diagnostic tests to facilitate the early detection of adverse health effects among exposed individuals.” David M. Studdert et al., Medical Monitoring for Pharmaceutical Injuries: Tort Law for the Public's Health? 289 JAMA 889, 889 (2003). In that sense, uninjured plaintiffs could potentially form part of an advertiser’s target audience. However, one advertising attorney that reviewed a draft of this study reported that uninjured plaintiffs are not the target audience in drug injury advertising because they are considered unprofitable.

29 Schaffzin, supra note 23, at 325. Social science research suggests that individuals are more likely to pay attention to an advertisement when it is “personally relevant.” Stewart, David W. & Martin, Ingrid M., Intended and Unintended Consequences of Warning Messages: A Review and Synthesis of Empirical Research, 13 J. Pub. Pol'y & Marketing 1, 5 (1994)Google Scholar (synthesizing research on warnings). A consumer taking a drug featured in a drug injury would likely find it relevant to his or her health, even if the consumer has not suffered an adverse event. Social science research also suggests that consumers pay more attention to warnings involving information that is not well known, which may be the case for adverse events associated with drugs and medical devices. Id. at 6.

30 See Hosken & Wendling, supra note 14, at 182 (noting that direct-to-consumer pharmaceutical advertising is associated with an increase in likelihood that consumers visit the doctor); Menon, Ajit M. et al., Consumers’ Attention to the Brief Summary in Print Direct-to-Consumer Advertisements: Perceived Usefulness in Patient-Physician Discussions, 22 J. Pub. Pol’y & Marketing 181, 181 (2003)CrossRefGoogle Scholar (summarizing policy arguments in favor of direct-to-consumer drug advertising); see also Hazard et al., supra note 26, at 1094 (arguing that attorney advertising spurs information seeking by consumers).

31 This is known in social science as the “mere exposure” effect, where repeated exposure to a product increases positive responses to that stimulus. Grimes, Anthony & Kitchen, Philip J., Researching Mere Exposure Effects to Advertising, 49 Int'l J. Market Res. 191, 193 (2007)Google Scholar. Even if the product appears in a negative context, repeated exposure to the product’s name may nevertheless increase positive associations through the mere exposure effect.

32 Daniel Schaffzin argues that drug injury ads are also problematic because they erode trust between patients and physicians. Schaffzin, supra note 23, at 341. Similar trust-based arguments have been made in the context of requiring drug manufacturers to convey lengthy warnings to patients about the risks of a drug. See Lars Noah, This is Your Products Liability Restatement on Drugs, 74 Brook. L. Rev. 839, 898 n.260 (2009).

33 Perry, Joshua E. et al., Direct-to-Consumer Drug Advertisements and the Informed Patient: A Legal, Ethical, and Content Analysis, 50 Am. Bus. L.J. 729, 730 (2013)CrossRefGoogle Scholar (“In the case of direct-to-consumer (DTC) advertising of prescription drugs, the link between advertising exposure and purchase is often especially circuitous, in part because consumers … must first consult with a health-care provider, who then decides whether to prescribe the advertised product, an alternative treatment, or nothing at all.”).

34 I was able to identify no instances in which the FDA has weighed in on drug injury advertising or asserted that such ads are subject to drug advertising regulation. The FDA likely considers drug injury advertising outside of its jurisdiction. The legislation prohibiting “misbranded” drugs, from which the FDA derives its authority, presumes that the entities responsible for misbranding a drug are “the manufacturer, packer, or distributor [of]” prescription drugs. See 21 U.S.C. § 352(n) (2012) (“A drug or device shall be deemed to be misbranded … [i]n the case of any prescription drug distributed or offered for sale in any State, unless the manufacturer, packer, or distributor thereof includes in all advertisements” various information required by the FDA, such as side effects.) (emphasis added); id. § 331. FDA regulations also suggest that their scope is limited to promotional advertising. 21 C.F.R. § 202.1(e)(1) (2014) (“All advertisements for any prescription drug … shall present a true statement of information in brief summary relating to side effects, contraindications … and effectiveness.”) (emphasis added). Creative arguments could be made that the FDA has jurisdiction. Indeed, Lars Noah asserts that “[t]he FDA has shown tremendous creativity in construing the reach of its authority” in order to pursue extra-statutory goals. Noah, Lars, The Little Agency that Could (Act with Indifference to Constitutional and Statutory Strictures), 93 Cornell L. Rev. 901, 917 (2008)Google Scholar.

35 The FTC’s power over attorney advertising is somewhat uncertain. As one commentator described, “Congress neither expressly restricted nor expressly delegated to the FTC rulemaking authority over state laws or regulations.” Cohen, Tod H., Double Vision: The FTC, State Regulation, and Deciding What’s Best for Consumers, 59 Geo. Wash. L. Rev. 1249, 1256 (1991)Google Scholar. In any event, the FTC has not inserted itself into the regulation of attorney advertising beyond commenting on proposed attorney advertising rules that it deems overly restrictive. See, e.g., Brief of the FTC, supra note 26. State consumer protection laws also protect consumers against false advertising. State attorneys general have sued drug manufacturers for misleading advertising. See, e.g., Press Release, Fla. Office of the Att'y Gen., McCollum Announces $8 Million Multi-State Settlement with Bayer (Jan. 23, 2007), http://web.law.columbia.edu/sites/default/files/microsites/attorneys-general/files/McCollum%20Announces%208%20Million%20Dollar%20Multi-State%20Settlement%20with%20Bayer_0.pdf. I was, however, unable to find any examples of state attorneys general taking action against attorney advertisers for defective drug advertising.

36 See infra Section IV(A).

37 Zacharias, Fred C., What Lawyers Do When Nobody’s Watching: Legal Advertising As a Case Study of the Impact of Underenforced Professional Rules, 87 Iowa L. Rev. 971, 985-86 (2002)Google Scholar.

38 Zauderer v. Office of Disciplinary Counsel of the Supreme Court of Ohio, 471 U.S. 626 (1985). It is possible that state bars have taken action but that records of that action were not detectable on Westlaw. For example, state bars frequently negotiate with advertisers informally, especially in states that require attorneys to submit advertising to the state. Additionally, secrecy in disciplinary proceedings limits the availability of information on state bar practices. See Levin, Leslie C., The Case for Less Secrecy in Lawyer Discipline, 20 Geo. J. Legal Ethics 1, 15-16 (2007)Google Scholar; Rhode, Deborah L. & Woolley, Alice, Perspectives on Lawyer Regulation: An Agenda for Reform in the United States and Canada, 80 Fordham L. Rev. 2761, 2767 (2012)Google Scholar (noting that most states do not disclose disciplinary complaints absent a subsequent finding of a violation or probable cause).

39 Isaac M. Lipkus, Numeric, Verbal, and Visual Formats of Conveying Health Risks: Suggested Best Practices and Future Recommendations, Med. Decision Making 696, 698 (2007).

40 Nielsen divides the United States into 210 television market regions. See DMA Regions, Nielsen, http://www.nielsen.com/intl-campaigns/us/dma-maps.html (last visited Mar. 25, 2015) (defining “designated market area”).

41 The Boston and Atlanta markets include the cities themselves and large portions of the surrounding geographic area. See Nielsen, Nielsen Station Index: ZIP Codes by DMA (2010) (on file with author). The Boston market also includes Manchester, New Hampshire. Id. Boston and Atlanta represent the seventh and eighth largest media markets in the country, respectively, as of 2009-2010. See Nielsen, Local Television Market Universe Estimates 1, 1 (2009), available at http://www.nielsen.com/content/dam/corporate/us/en/newswire/uploads/2009/08/2009-2010-dma-ranks.pdf. In 2009, the Atlanta designated media market encompassed an estimated 2.37 million homes and the Boston media market encompassed an estimated 2.41 million homes. Id. The two cities were selected on the basis of their geographic distance, on the assumption that they might support somewhat different advertising markets. The very largest media markets —New York, San Francisco and Los Angeles, for example, were ruled out to avoid a strong likelihood that the legal or media market were so large in those regions as to be idiosyncratic. See id.

42 Based on a dataset of national drug injury advertising for the same period, the volume of targeted advertising compares favorably to the volume of national network advertising, at 859 units, compared to 1,258 units. Over the same period, 6,460 additional units of drug injury advertising were broadcast over national cable, scattered throughout an array of channels with varying ratings. Relative pricing of network versus cable advertising within that dataset indicates that on average, a network ad costs more than twice the cost of a cable ad, which likely reflects average viewership. The targeted advertising in this sample would have been broadcast over local affiliates of national networks—for example, the local ABC or NBC station.

43 Ads relating to personal care products (e.g., denture cream) were excluded, as were ads relating to herbal supplements (e.g., the herbal supplement known as Hydroxycut).

44 When content coding or transcriptions for two or more ads produced identical or substantially identical results, the ads were played side by side for any observable differences. Where an ad appeared identical in all respects but the phone number, it was deemed a duplicate. I hypothesize that firms sometimes broadcast different phone numbers with the same ads to optimize their advertising strategy. The different phone numbers may be associated with different broadcast locations, times, or channels. A data-driven advertiser could then use the volume of calls associated with a single number to determine the yield associated with a particular advertising strategy.

45 A few states require attorneys to submit a copy of their advertisements to the state bar. See, e.g., Tex. Disciplinary Rules of Prof'l Conduct R. 7.07 (2005).

46 See Ga. Rules of Prof'l Conduct R. 7.1 (2001); Mass. Rules of Prof'l Conduct R.7.1 (1998). Unlike the ABA Model Rules of Professional Conduct, Georgia’s Rule 7.2 explicitly requires that attorneys disclose the identity and physical location of the sponsoring attorney, as well as the firms referral practices if a majority if cases will be referred. Ga. Rules of Prof'l Conduct R. 7.2 (2001).

47 The coding instrument for the first round was focused primarily on whether and when the ad identified an attorney as the source of the ad. The second round was focused on content relating to medical information, such as adverse events, benefits, or advice to consult a doctor. The third round was focused on disclaimers relating to the geographic location and licensing status of the advertising attorney.

48 Different law students coded each round, with the exception of one student, who coded both the second and third rounds.

49 This definition is from the second coding instrument, the results of which are used for the analysis in this Article. A definition also appears in the first coding instrument and was refined for the second instrument to reduce inter-rater disagreement.

50 This definition is from the second coding instrument, the results of which are used for the analysis in this Article. A definition also appears in the first coding instrument and was refined for the second instrument to reduce inter-rater disagreement.

51 Following the independent coding, one coder was assigned a legal research project relating to attorney ethics and informed that the research was unrelated. During a debriefing discussion following the conclusion of the research project, the coder guessed that the purpose of the research related somehow to attorney ethics. Two of the other coders guessed that the project related to attorney ethics, one thought it was an assessment of the quality of the information in the ads, and another surmised that the study compared the processing of audio and visual information.

52 The “N” for attorney identification (55) is lower because the coders were unable to open one of the files during the first round of coding.

53 Discussions of adverse events typically occurred for a large proportion of the ad but would start and stop intermittently. This led to the accumulation of short timing discrepancies when the discussion started and stopped. Adding to the complexity, coders were instructed to include any audio or text discussion of adverse effects in their “duration” calculation, but the audio and text frequently diverged. Were one coder to identify a short text discussion that the other coder overlooked, it produced noticeable differences in results. In reviewing the coders' independent transcriptions of the text and audio relating to adverse events, timing differences appeared primarily attributable to minor discrepancies in the inclusion or exclusion of conditional clauses such as “if your loved one has died while wearing the pain patch.” The coders, however, generally captured the advertisement's core discussion of adverse events in their transcription.

54 Dataset on file with the author.

55 Table 3 may somewhat underestimate the number of unique ads relating to drugs and medical devices. Duplicate ads relating to drug and medical devices were identified as part of the in-depth content analysis described in this Article. See supra Part II(B). However, other attorney ads were not scrutinized for duplication for this research and instead relied on the unique identifiers Kantar associated with each ad. Therefore, the count for “all ads” likely includes some duplicates.

56 The over-representation of drug/device ads broadcast in both markets reflects the methodology by which duplicate advertising was identified. For drug/device ads, identical advertising in both markets was identified through (1) matching unique identifies originating from Kantar; or (2) identical or near identical content coding, followed by a side-by-side review of the ad. Due to practical constraints, all other attorney ads were only identified as jointly broadcast using the unique identifiers originating from Kantar. Therefore, the second row in Table 4 likely understates the volume of “All Ads” broadcast in both media markets.

57 Kantar uses “units” to refer to the volume of advertising—the number of ads multiplied by the number of times the ad each ad was broadcast.

58 There have also been changes in the law relating to failure to warn claims since 2009. See Pliva v. Mensing, 131 S.Ct. 2567, 2578-80 (2011) (limiting “failure to warn” claims against generic drug manufacturers under state law).

59 Certain drugs were commonly featured together, such as Yaz, Yazmin & Ocella (birth control pills), or Advil, Ketek, Motrin and others (all allegedly associated with Stevens Johnson Syndrome). Each grouping of drugs was treated as a “type.”

60 Four of these firms are captured in the category “other firms.”

61 See supra p. 2.

62 Results for “benefits” and “consult a doctor” are skewed downward by the absence of such content in many of the ads. When averages were calculated based on the subset of ads containing some content in each category, the average word count for adverse events continues to dwarf other categories. The average words devoted to benefits among ads that discuss benefits is ten words of text, and twelve words of audio. Among ads that advise viewers to consult a doctor, an average of eleven words are devoted to that subject.

63 Results displayed in Figure 8 include advertisements that did not reference drug benefits, or advise viewers to consult a doctor. The usefulness of Figure 8 is limited by the low inter-rater reliability for the calculated duration of adverse event discussions.

64 Because only half of advertisements made any reference to the benefits or use of the drug, the excerpts below do not correspond to the ads in Table 9.

65 The drug label for Yaz estimates the risk of blood clots associated with Yaz at 5-20 per 10,000 women taking the drug (compared with 5-13 per 10,000 for women taking other forms of oral contraceptives). See Bayer, supra note 13, at 8.

66 No such content was identified. The coders’ transcriptions of content relating to adverse events and benefits revealed periodic references to information relating to relative risk, which are discussed herein. It is possible that relative risk was described qualitatively elsewhere in the ad and not detected by the coders. For that reason, discussions of relative risk factors is not discussed quantitatively.

67 Ads involving Fentanyl were subsequently reviewed by the author for information identifying the manufacturer of the recalled patch.

68 See ANI Pharm., Reglan Drug Label 8-9 (2011), available at http://www.accessdata.fda.gov/drugsatfda_docs/label/2011/017854s058lbl.pdf

69 See id. The Reglan drug label provides: “It is not known if Reglan is safe and works in children.” Id. at 14. The reference to children may reflect potential liability associated with off-label usage. See id.

70 Id. at 4.

71 The FTC’s “clear and conspicuous” standard for disclaimers in television advertising recommends that the information be presented in both audio and video portions of the ad to improve the consumer’s processing of the information. See FTC, Commission Enforcement Policy Statement in Regard to Clear and Conspicuous Disclosure in Television Advertising 1 (Oct. 21, 1970), available at http://www.ftc.gov/system/files/documents/public_statements/288851/701021tvad-pr.pdf. The FTC’s position is supported by social science research into knowledge, recall, and comprehension of information. See Hoy, Mariea Grubbs & Andrews, J. Craig, Adherence of Prime-Time Televised Advertising Disclosures to the “Clear and Conspicuous” Standard: 1990 Versus 2002, 23 J. Pub. Pol'y & Marketing 170, 172 (2004)Google Scholar.

72 The FTC also recommends that the letters on the screen be “of ‘sufficient size' so that it can be easily seen and read on all television sets, regardless of [screen size].” FTC, supra note 71, at 1.

73 The relative importance of consulting a doctor is somewhat context specific. For example, the potential harm associated with the absence of advice to consult a doctor is limited in connection with this particular ad. The shoulder pain pump consists of a medical device delivering anesthetic post-surgery. Because the pump is a one-time treatment, it is unlikely that a patient would make decisions about the treatment without consulting a doctor.

74 Several ads used the word “warning” at the start, without reference to the FDA. If these ads did not also use the phrase “medical alert” or “consumer alert” at the start, they were not included in the “warning/alert” category, as they were less strongly associated with a failure to identify the attorney source of the ad early in the advertisement.

75 Consumer Alert Segment (WENY TV News television broadcast Mar. 21, 2014), available at http://www.youtube.com/watch?v=Mi7wKrhPUr0 (describing a local fraud scheme).

76 Only a small number of ads extended past the thirty second mark. Data was analyzed in three second increments, which accounts for the jagged shape of the lines.

77 For example, the approximately 40% of ads that identify the attorney at the 23 second mark is calculated as a proportion of all ads at least 23 seconds long.

78 The competing information at the start of these ads could limit viewers' ability to process all of the information presented.

79 Historical records of the Internet using the “Wayback Machine” provide only limited snapshots of the appearance of websites. The Internet Wayback Machine takes snapshots of webpages periodically. See Internet Archive WaybackMachine, Internet Archive, archive.org/web (last visited Mar. 25, 2015).

80 The firm provided a YouTube link to the ads that it claimed it ran during its Yaz campaign, which was 10 seconds longer than the ad in the Kantar dataset and identified the firm in the final 10 seconds.

81 The Kantar dataset included three different copies of the ad at issue, none of which disclosed the name of the advertising firm. However, it is possible that the three copies originated from a single flawed file.

82 Considerable scholarly attention has been devoted to the history of Supreme Court jurisprudence on attorney advertising. See generally Calvani, Terry et al., Attorney Advertising and Competition at the Bar, 41 Vand. L. Rev. 761, 762-74 (1988)Google Scholar; Rossi, Jim & Weighner, Mollie, Contemporary Studies Project: An Empirical Examination of the Iowa Bar's Approach to Regulating Lawyer Advertising, 77 Iowa L. Rev. 179, 186-94 (1991)Google Scholar. For the sake of brevity, the history is not recounted here.

83 See Schaffzin, supra note 23, at 354.

84 Model Rules of Prof'l Conduct R. 7.1 (2011).

85 Forty states incorporate the language in the model rules, or very similar language. See, e.g., Alaska Rules of Prof'l Conduct R. 7.1 (2009); Ark. Rules of Prof'l Conduct R. 7.1 (2014); Haw. Rules of Prof'l Conduct R. 7.1 (2014); Idaho Rules of Prof'l Conduct R. 7.1 (2013); Ky. Rules of Prof'l Conduct R. 3.130(7.15) (2009); Me. Rules of Prof'l Conduct R. 7.1 (2009); Md. Rules of Prof'l Conduct R.7.1 (2015).

86 Many of these Model Rule states include additional examples of false or misleading advertising in their rule, such as statements “likely to create an unjustified expectation about results the lawyer can achieve” or that “compare the lawyer's services with other lawyer's services, unless the comparison can be factually substantiated.” See, e.g., Kan. Rules of Prof'l Conduct R. 7.1(b)-(c) (2014); Ky. Rules of Prof'l Conduct R. 3.130(7.15)(b)-(c) (2009); La. Rules of Prof'l Conduct R. 7.1 (2014); Md. Rules of Prof'l Conduct R.7.1(b)-(c) (2015). Some states also depart from the Model Rules by defining “false or misleading” communications to include statements that cannot be substantiated or verified. See e.g., D.C. Rules of Prof'l Conduct R. 7.1 (2014); Fla. Rules of Prof'l Conduct R. 4-7.2(c) (2014); Iowa Rules of Prof'l Conduct R. 32:7.1 (2014); Ohio Rules of Prof'l Conduct R. 7.1 (2014).

87 See Bates v. State Bar of Ariz., 433 U.S. 350, 363 (1977); see also Backer, Daniel, Note, Choice of Law in Online Legal Ethics: Changing a Vague Standard for Attorney Advertising on the Internet, 70 Fordham L. Rev. 2409, 2415 (2002)Google Scholar; Callendar, Daniel, Attorney Advertising and the Use of Dramatization in Television Advertisements, 9 UCLA Ent. L. Rev. 89, 92-95 (2001)Google Scholar.

88 See Bates, 433 U.S. at 383; see also Fla. Bar v. Went For It, Inc., 515 U.S. 618, 623-24 (1995); In re R.M.J., 455 U.S. 191, 201 (1982); Cent. Hudson Gas & Elec. Corp. v. Pub. Serv. Comm’n, 447 U.S. 557, 566 (1980); Hyland, William G. Jr., Attorney Advertising and the Decline of the Legal Profession, 35 J. Legal Prof. 339, 358 (2011)Google Scholar (“[T]he Bates decision established two principles regarding lawyer advertising: first, the informational function of lawyer advertising is entitled to First Amendment protection. Second, states have the power to protect the public from harmful commercial speech, such as false or misleading advertising.”).

89 Were a state to define “false or misleading” more broadly, then speech prohibited by the bar but protected under the First Amendment would be subject to the Central Hudson test, which has historically proved difficult for state bars to overcome. See, e.g., Peel v. Att'y Registration & Disciplinary Comm'n of Ill., 496 U.S. 91, 115-16 (1990) (holding state prohibition on advertising references to certifications or specializations unconstitutional); Shapero v. Ky. Bar Ass'n, 486 U.S. 466, 479 (1988) (holding state prohibition on targeted direct mail advertising unconstitutional); In re R.M.J., 455 U.S. at 201 (holding rule restricting advertising to ten categories of information unconstitutional).

90 Zauderer v. Office of Disciplinary Counsel of the Supreme Court of Ohio, 471 U.S. 626, 645 (1985) (“[D]istinguishing deceptive from nondeceptive advertising in virtually any field of commerce may require resolution of exceedingly complex and technical factual issues and the consideration of nice questions of semantics.”).

91 Id. at 652.

92 Id.

93 Id.

94 See id. Where the misleading quality of the advertisement is not apparent, the Supreme Court expects some evidence of actual harm to support arguments that the advertisement is potentially misleading. See Peel, 496 U.S. at 101-04 (noting the absence of any evidence of actual harm where Supreme Court was unpersuaded by state's argument as to the potentially misleading nature of a statement); see also Ibanez v. Fla. Dept. of Bus. & Prof'l Regulation, Bd. of Accountancy, 512 U.S. 136, 145 (1994) (finding state must proffer some evidence of deception, or more than the “possibility of deception in hypothetical cases”); Mason v. Fla. Bar, 208 F.3d 952, 957 (11th Cir. 2000) (finding “common sense” arguments insufficient to justify disclaimer requirement absent empirical evidence that public was misled).

95 Peel, 496 U.S. at 102.

96 For example, in Peel, the Supreme Court observed the question of whether or not a certification is misleading depends in part on the methodology used to certify the attorney. An identical statement about a certification could be non-misleading where the certifying agency imposes rigorous requirements, but misleading if the certifying agency accepts applicants indiscriminately or for anyone willing to pay a price. Id.

97 See Farrin v. Thigpen, 173 F. Supp. 2d 427, 437 (M.D.N.C. 2001).

98 See id.

99 Id.

100 See Kraft, Inc. v. FTC, 970 F.2d 311, 314 (7th Cir. 1992) (interpreting Zauderer to mean that implied claims can also be the basis of false or misleading speech outside the protection of the First Amendment).

101 S.C. Bar Ethics Advisory Comm., Ethics Advisory Op. 13-05 (2013), 2013 WL 7196338, at *1 (cautioning lawyers against using phrases such as “Legal Helpline” or “Injury Hotline” because they misleadingly suggest an affiliation with the state bar, a charity or the government).

102 Conn. Bar Assoc. Comm. on Prof'l Ethics, Informal Op. No. 01-03 (2001), 2001 WL 694581, at *3 (characterizing referral service that described itself as “A Free Public Service” as misleading because it “could create a misleading perception on the part of the lay reader that ABC … [is a] nonprofit referral service[]”).

103 State Bar of Cal. Standing Comm. on Prof'l Responsibility & Conduct, Formal Op. 2004-167 (2004), 2004 WL 3079032, at *2 (deeming practice as misleading because it suggested that the firm itself was an official government office, or that it awarded benefits rather than offering legal representation in seeking benefits).

104 N.J. Supreme Court Advisory Comm. on Att'y Adver., Internet Adver., Misleading Content, & Impermissible Referral Servs., Op. No. 43 5 (2011), available at 2011 WL 2691355.

105 See Farrin v. Thigpen, 173 F. Supp. 2d 427, 445 (M.D.N.C. 2001).

106 See Model Rules of Prof'l Conduct R. 7.1 cmt. 3 (1983); see also N.C. State Bar, Adver. for Legal Emp't in Non-Practicing Areas, 2010 Formal Ethics Op. 6 (2011), 2011 WL 665760, at *2 (“Previous ethics opinions have determined that an appropriate disclaimer may cure an otherwise misleading advertisement.”).

107 Farrin, 173 F. Supp. 2d at 445.

108 Id.

109 Pa. Bar Assoc. Comm. on Legal Ethics & Prof'l Responsibility, Informal Op. No. 93-124 (1993), 1993 WL 851228, at *2 (citation omitted) (concluding that firm name containing the names of unaffiliated lawyers was not cured by accompanying disclaimer stating “individual practitioners, not a partnership”); see also N.J. Op. No. 43, supra note 104, at 9-10 (warning advertiser not to undermine a disclosure about the attorney selection process with contradictory statements, like telling users their zip code is needed because “the law varies from state to state” or requesting financial information in order to suggest that such information is used to select a suitable attorney).

110 See Farrin, 173 F. Supp. 2d at 445-46.

111 The legal standard is more stringent when the attorney challenges a prophylactic rule that bars certain categories of content as inherently misleading or deceptive. Such prophylactic rules must withstand intermediate scrutiny under the First Amendment, as outlined in Central Hudson. Cent. Hudson Gas & Elec. Corp. v. Hub. Serv. Comm'n, 447 U.S. 557, 564 (1980); see also Fla. Bar v. Went For It, Inc., 515 U.S. 618, 624 (1995). Such rules rarely survive intermediate scrutiny. See, e.g., Peel v. Att'y Registration & Disciplinary Comm'n of Ill., 496 U.S. 91, 115 (1990) (finding state prohibition on advertising references to certifications or specializations unconstitutional); Shapero v. Ky. Bar Ass'n, 486 U.S. 466, 479 (1988) (holding state prohibition on targeted direct mail advertising unconstitutional); In re R.M.J., 455 U.S. 191, 201 (1982) (finding rule restricting advertising to ten categories of information unconstitutional).

112 See Peel, 496 U.S. at 108 (“Whether the inherent character of a statement places it beyond the protection of the First Amendment is a question of law over which Members of this Court should exercise de novo review.”); see also Ibanez v. Fla. Dept. of Bus. & Prof'l Regulation, Bd. of Accountancy, 512 U.S. 136, 145 (1994) (finding that attorney successfully argued that a CPA designation in commercial communications was not misleading).

113 See Zauderer v. Office of Disciplinary Counsel of the Supreme Court of Ohio, 471 U.S. 626, 653 (1985); see also Kraft, Inc. v. FTC, 970 F.2d 311, 321 (7th Cir. 1992) (“Zauderer teaches that consumer surveys are not compelled by the first amendment when the alleged deception although implied, is conspicuous.”).

114 See Ibanez, 512 U.S. at 145.

115 ABA Comm. on Prof'l Ethics & Grievances, Formal Op. 178 (1938).

116 See Fla. Bar v. Doe, 634 So.2d 160, 161 (Fla. 1994).

117 Id. at 162-63.

118 Id. at 161.

119 FTC, Frequently Asked Questions: A Guide for Small Business, General Advertising Policies (2008), available at 1998 WL 207800, at *10; see also Press Release, FTC, FTC Seeks to Halt 10 Operators of Fake New Sites form Making Deceptive Claims About Acai Berry Weight Loss Products (Apr. 19, 2011), available at http://www.ftc.gov/news-events/press-releases/2011/04/ftc-seeks-halt-10-operators-fake-news-sites-making-deceptive (describing restraining orders filed in federal court to enjoin “fake news websites” that promoted acai berry weight-loss products).

120 See Doe, 634 So.2d at 161.

121 In this respect, the deception is structurally similar to a bankruptcy referral website that the New Jersey State Bar deemed misleading. The website “offer[ed] to connect visitors … with a bankruptcy attorney” by asking them to fill out information about their situation, including their zip code. See N.J. Op. No. 43, supra note 104, at 1. Unbeknownst to users, the website simply matched users to an attorney based on whichever law firm had paid for the exclusive right to receive referrals in that zip code. See id. The New Jersey State Bar deemed the use of the phrase “free evaluation” misleading, because it “connotes a nuanced, fact-specific consideration of the User's financial condition,” rather than referral based on a paid arrangement. Id. at 10. Just as the New Jersey website might mislead a visitor to assume that the attorney produced is a curated match for his or her particular circumstances, a viewer of a “consumer alert” might mistakenly assume that the medical information presented represents a dire risk upon which they should base their decision-making.

122 See Conn. Bar Assoc. Comm. on Prof'l Ethics, Informal Op. No. 01-03 (2001), 2001 WL 694581, at *3 (describing referral service as a “public service” misleading because it obscures the advertiser's “concealed financial interest”).

123 Id.

124 See State Bar of Cal. Standing Comm. on Prof'l Responsibility & Conduct, Formal Op. 2004-167 (2004), 2004 WL 3079032, at *2.

125 Citing an FDA warning as authority for statements in an advertisement is not necessarily misleading where the purpose of the advertisement is otherwise apparent. In fact, Daniel Schaffzin argues that drug injury attorneys should be more transparent about the sources of their medical information by citing them in their advertisements. See Schaffzin, supra note 23, at 340.

126 See FTC, Frequently Asked Questions, supra note 119, at *10. The Supreme Court has opined that state bars can require attorneys to correct potentially misleading advertising through a disclosure identifying the content as an advertisement. See Shapero v. Ky. Bar Ass'n, 486 U.S. 466, 477 (1988) (citing Bates v. State Bar of Ariz., 433 U.S. 350, 384 (1977); In re R.M.J., 455 U.S. 191, 206 n.20 (1982)); see also Mead, Terrence C., Writing the Law of Lawyer Advertising, 23 Ariz. St. L.J. 191, 210 (1991)Google Scholar (“The Supreme Court, in various dicta discussing less restrictive alternatives to tested prohibitions, has suggested that lawyer advertising be labeled as such.”).

127 See Cowley, Elizabeth & Barron, Chris, When Product Placement Goes Wrong: The Effects of Program Liking and Placement Prominence, 37 J. Advertising 89, 90 (2008)CrossRefGoogle Scholar.

128 See id.

129 Id.

130 Id. at 91.

131 The extent to which this principle applies to drug injury advertising is unknown. It is also possible that repeated exposure to attorney advertising allows consumers to immediately identify all attorney advertising and apply this “change of meaning” at the start of the ad.

132 See supra note 94 and accompanying text.

133 See Model Rules of Prof'l Conduct R. 7.2(c) (2011).

134 See, e.g., Relion Grp., http://www.reliongroup.com (last visited Mar. 25, 2015).

135 See e.g., Change Content Depending on Visitor IP Address, Geolify, http://geolify.com/display-content-on-website-based-on-visitor-ip/ (last visited Mar. 25, 2015). A variation of this business structure is described in a New Jersey advisory opinion. Although it did not involve television ads, visitors to an apparent attorney-client matching service for bankruptcy claims were told to enter information about their claim, including their zip code. In fact the website was a form of advertising, which simply matched the visitor to whichever attorney paid for exclusive rights to that zip code. See N.J. Op. No. 43, supra note 104, at 9-10; see also discussion supra note 109.

136 See Relion Grp., supra note 134.

137 A disclaimer at the conclusion of a Relion YouTube ad states “Relion Group is an advertising group that represents lawyers jointly advertising their services; it is not a law firm or a lawyer referral service.” See Relion Group, Xarelto Lawsuit, YouTube (Jan. 21, 2015), https://www.youtube.com/watch?v=W6CfVXChqao.

138 The involvement of non-attorney marketing entities raises questions about permissible attorney referral arrangements and unauthorized practice of law rules, which are beyond the scope of this study.

139 Lipkus, supra note 39, at 696.

140 Social science research suggests that warnings in general “may create the impression that a product is more dangerous than it is in reality or may be more dangerous than an alternative that does not contain a warning.” Stewart & Martin, supra note 29, at 3.

141 Model Rules of Prof'l Conduct R. 7.2 (2011).

142 Zauderer v. Office of Disciplinary Counsel of the Supreme Court of Ohio, 471 U.S. 626, 652 (1985).

143 See id. at 626.

144 See id. at 633, 639-40.

145 The relevant medical statement is visible at the very start of the text of the ad and reads, “The Dalkon Shield Interuterine Device is alleged to have cause serious pelvic infections resulting in hospitalizations, tubal damage, infertility, and hysterectomies.” See id. at 631.

146 At the time the device was introduced to the market, it was not subject to FDA authority and was never reviewed for safety or effectiveness. See Andrea Tone, Devices and Desires 277 (2001). The device was recalled several years prior to the ad’s dissemination. See Zauderer, 471 U.S. at 630. Even if viewers of the ad inferred that the Dalkon Shield was very dangerous, it is not at all clear that such a risk inference would have been exaggerated. It caused 18 deaths and over 200,000 other adverse events. See Tone, supra, at 279. These factual distinctions would be relevant to a First Amendment analysis, as the Supreme Court has previously emphasized the central importance of context in evaluating whether a statement is false or misleading. See Peel v. Att'y Registration & Disciplinary Comm'n of Ill., 496 U.S. 91, 114 (1990).

147 See Peel, 496 U.S. at 101.

148 See supra pp. 2-3.

149 Mather, Lynn, What Do Clients Want? What Do Lawyers Do?, 52 Emory L.J. 1065, 1068 (2003)Google Scholar.

150 See id. at 1084.

151 Id. at 1085.

152 See generally Zacharias, supra note 37.

153 The FTC may be disinclined to act based on its previously stated preference for state agencies regulate any advertising within their jurisdiction. “How does the FTC decide what cases to bring? … FTC jurisdiction. Although the FTC has jurisdiction over ads for most products and services … [t]he FTC concentrates on national advertising and refers local matters to state, county or city agencies…. State or local consumer protection agencies or private groups such as the Better Business Bureau (BBB) often are in a better position to resolve disputes involving local businesses or local advertising.” FTC, Frequently Asked Questions, supra note 126, at *2. Advertisers bring informal claims against other advertisers through the Advertising Self-Regulatory Council’s National Advertising Division. A review of NAD Case Reports dating back to 1997 did not reveal any cases reports involving attorney advertisers. See Case Reports, Advertising Self-Reg. Council, http://case-report.bbb.org/search/search.aspx?doctype=1&casetype=1 (last visited Mar. 25, 2015).

154 Fred Zacharias found that more than 30% of the 857 Yellow Pages advertisements he reviewed did not comply with California's detailed ethics rules regarding advertising. See Zacharias, supra note 37, at 985-86. He also found fewer than ten cases of attorney discipline in California that directly addressed advertising violations over a thirteen year period, which he found comparable to discipline rates in five other jurisdictions. Id. at 987-88. As a result, Zacharias concluded that state bars “seem to have made a conscious decision not to enforce advertising rules stringently.” Id. at 996.

155 Id. at 997-98.

156 Id. at 996, 1004 (noting a heightened level of awareness about the constitutionality of rules, and that state bars may prefer to spend their enforcement dollars on “less controversial violations”).

157 Id. at 1002.

158 Advertising attorneys insist that they have received no complaints from consumers, and focus on the thousands of injured consumers they have helped.

159 References to the arguments from advertising lawyers are based on email correspondence and telephone discussions with advertising attorneys and drug injury litigators that were provided a draft of this study and an opportunity to respond or comment. At least one advertising firm requested anonymity. I therefore have not specifically identified the individual firms and attorneys.

160 One advertising attorney even described a client who continued taking the prescribed medication even after experiencing an adverse event.

161 Moynihan, Ray & Henry, David, The Fight Against Disease Mongering: Generating Knowledge for Action, 3 PLoS Med. 425, 425 (2006)CrossRefGoogle Scholar (describing the problem of disease mongering: “[T]he selling of sickness [by pharmaceutical companies] that widens the boundaries of illness and grows the markets for those who sell and deliver treatments”). Drug injury advertisements also have interesting parallels to “disease-mongering” pharmaceutical sponsored ads that promote a particular disease, without mentioning the specific drug at issue. Depending on the circumstances, such ads are subject to considerably less FDA oversight than direct-to-consumer pharmaceutical advertising. Mintzes, Barbara, Disease Mongering in Drug Promotion: Do Governments Have a Regulatory Role?, 3 PLoS Med. 461, 461 (2006)CrossRefGoogle ScholarPubMed (noting that the FDA has no authority over “disease awareness” advertisements unless they are “perceptually similar” to the branded ads, or the advertiser manufactures the only drug in its class).

162 See supra note 34 (discussing statutes and regulations governing drug advertisements).

163 State bars might be tempted to enact a prophylactic rule barring the appearance of certain types of content, such as the words “medical alert” at the start of an advertisement. Regardless of the efficacy of such a prohibition, prophylactic rules are subject to much more stringent constitutional scrutiny. For a prophylactic rule to survive First Amendment scrutiny, the state must demonstrate a “substantial governmental interest justifying the restriction” and “demonstrate that the restriction vindicates that interest through the least restrictive available means.” Zauderer v. Office of Disciplinary Counsel of the Supreme Court of Ohio, 471 U.S. 626, 627 (1985). This standard is very demanding by design—the Supreme Court expressly prefers for state bars to make individualized determinations regarding false and misleading advertisements. Id. at 649.

164 See Lipkus, supra note 39, at 707-09.

165 Id. at 697 (“Most recent conceptualizations of risk view risk as a combined function, often multiplicative, of the probability of loss and consequence of loss….”).

166 See Stewart & Martin, supra note 29, at 1 (noting that “long-term hazard information is difficult to quantify meaningfully”).

167 See Bayer, supra note 13, at 8 fig.2.

168 In the case of Yaz, for example, some studies have shown they do not present an elevated risk of blood clots compared to other forms of birth control, while others suggest that they double or triple the risk. Id. at 6.

169 See Actavis Inc. Issues a Voluntary Recall of 18 Lots of Fentanyl Transdermal System 25 mcg/h, FDA (Oct. 21, 2010), http://www.fda.gov/Safety/Recalls/ucm230498.htm; Fentanyl Transdermal System Patch Recall, FDA (Aug. 8, 2008), http://www.fda.gov/Safety/MedWatch/SafetyInformation/SafetyAlertsforHumanMedicalProducts/ucm126727.htm; Pricara Recalls 25 mcg/hr Duragesic (Fentanyl Transdermal System) CII Pain Patches, FDA (Feb. 12, 2008), http://www.fda.gov/safety/recalls/archiverecalls/2008/ucm112374.htm.

170 See supra note 66 (regarding methodological limitations as it related to the identification and coding of relative risk information).

171 See ANI Pharm., supra note 68, at 1.

172 FDA, Ctr. for Drug Evaluation & Research, Memorandum: Metoclopramide: Drug Use Data Review 1 (2005), available at http://www.fda.gov/ohrms/dockets/ac/05/briefing/2005-4167B1_02_11-FDA-Tab11-Review.pdf.

173 See ANI Pharm., supra note 68, at 1 (concerning discussion of relative risks in Reglan ads).

174 See Lipkus, supra note 39, at 695.

175 See Schwartz, Lisa M. et al., The Drug Facts Box: Providing Consumers with Simple Tabular Data on Drug Benefit and Harm, 27 Med. Decision Making 655, 659 (2007)CrossRefGoogle Scholar.

176 Id. at 655, 659.

177 Id. at 659. For the statistical information about the blood clot, 89% later identified the proportion of patients taking the drug that would experience a blood clot.

178 Id.

179 See Lipkus, supra note 39, at 701.

180 Id. at 699; see also Slater, Michael D. et al., Developing and Assessing Alcohol Warning Content: Responses to Quantitative Information and Behavioral Recommendations in Warnings with Television Beer Advertisements, 17 J. Pub. Pol’y & Marketing 48, 55 (1998)Google Scholar (finding that quantitative information in a warning was “recalled substantially better than nonquantitative”).

181 See Lipkus, supra note 39, at 699.

182 Id. at 701.

183 See supra note 159 regarding the source(s) of this information.

184 To the extent the attorneys responsible for the advertising are not those litigating the mass tort claims, the advertisers can consult the counsel to whom they refer their cases for additional factual context.

185 See Juurlink et al., supra note 15, at 1369-70 (finding that Internet hits for websites soliciting plaintiffs for lawsuits increased substantially following online publication of study identifying new risk associated with a drug).

186 For example, the revised Yaz labeling includes extensive statistical information about absolute and relative risk. See Bayer, supra note 13, at 6-8.

187 See 21 C.F.R. § 202.1(e)(1) (2014).

188 Id. § 202.1(e)(5)(ii).

189 See supra note 159 regarding the source(s) of this information.

190 See, e.g., Pub. Citizen, Inc. v. La. Att'y Disciplinary Bd., 632 F.3d 212, 228 (5th Cir. 2011) (deeming minimum font size and duration requirements for disclaimers unduly burdensome under the First Amendment absent empirical evidence that they effectively prevented consumer deception).

191 The coding instrument did, however, include instructions to identify any content referencing attorneys, law firms, or lawsuits. See infra Table 3.

192 See infra Figures 10-11.

193 See Advertising FAQ's: A Guide for Small Business, FTC, http://www.ftc.gov/tips-advice/business-center/advertising-faqs-guide-small-business (last visited Mar. 25, 2015).

194 The FDA regulation applicable to direct to consumer pharmaceutical drug advertising requires that manufacturers present a “fair balance” between discussions of a drug's effectiveness and discussions of its side effects and contraindications. See 21 C.F.R. § 202.1(5)(ii) (2014); Schaffzin, supra note 23, at 369-70.

195 The reverse application in this context would be that a discussion of adverse events should not be in greater depth than the benefits.

196 Slovic, Paul, What's Fear Got To Do with It? It's Affect We Need to Worry About, 69 Mo. L. Rev. 971, 972 (2004)Google Scholar.

197 Id. at 976.

198 Id.

199 Id. at 977 (“This process, which we have called ‘the affect heuristic' … suggests that, if a general affective view guides perceptions of risk and benefit, providing information about benefit should change perception of risk and vice-versa …. For example, information stating that benefit is high for a technology such as nuclear power would lead to more positive overall affect which would, in turn, decrease perceived risk….”).

200 See Ibanez v. Fla. Dept. of Bus. & Prof'l Regulation, Bd. of Accountancy, 512 U.S. 136, 146 (1994) (noting that absent evidence of harm, a disclaimer was unjustified where it effectively rules out the advertiser’s ability to convey otherwise non-misleading message); Zauderer v. Office of Disciplinary Counsel of the Supreme Court of Ohio, 471 U.S. 626, 651 (1985) (“[U]njustified or unduly burdensome disclosure requirements might offend the First Amendment by chilling protected speech.”).