Using Tobacco-Industry Marketing Research to Design More Effective Tobacco-Control Campaigns

SPECIAL COMMUNICATION Using Tobacco-Industry Marketing Research to Design More Effective Tobacco-Control Campaigns Pamela M. Ling, MD, MPH Stanton A. Glantz, PhD MOST ANALYSES OF TObacco-industry marketing have been designed to demonstrate that the industry targeted youth. 1-3 The industry also conducts extensive research on adults, particularly young adults aged 18 to 24 years. The industry s interest in young adults is not surprising, 4 since they compose its youngest legal marketing target, vastly outnumber teen smokers, 5-7 are making the transition from smoking initiation to becoming fully addicted smokers, 8-10 and are role models for teens. 10-12 Tobacco companies have used detailed studies of young adult smokers motivations, aspirations, activities, and environment to guide cigarette marketing and advertising campaigns. Tobacco marketers divide actual and potential smokers into segments: groups of consumers who respond similarly to a given set of marketing messages on the basis of needs, characteristics, or behavior. 13 Market segments can be defined by many criteria, including demographics, geography, attitudes, preferred product benefits, usage, and competitive products. 14 In contrast to public health, which designs tobacco-control programs according to demographics or constituencies, the industry defines targetmarket segments primarily by using consumer desires and needs. See also pp 2990 and 3001. To improve tobacco-control efforts by applying tobacco-industry marketing research and strategies to clinical and public health smoking interventions, we analyzed previously secret tobacco-industry marketing documents. In contrast to public health, the tobacco industry divides markets and defines targets according to consumer attitudes, aspirations, activities, and lifestyles. Tobacco marketing targets smokers of all ages; young adults are particularly important. During the 1980s, cost affected increasing numbers of young and older smokers. During the 1990s, eroding social acceptability of smoking emerged as a major threat, largely from increasing awareness of the dangers of secondhand smoke among nonsmokers and smokers. Physicians and public health professionals should use tobacco-industry psychographic approaches to design more relevant tobacco-control interventions. Efforts to counter tobacco marketing campaigns should include people of all ages, particularly young adults, rather than concentrating on teens and young children. Many young smokers are cost sensitive. Tobacco-control messages emphasizing the dangers of secondhand smoke to smokers and nonsmokers undermine the social acceptability of smoking. JAMA. 2002;287:2983-2989 Throughout the past 40 years, tobacco marketers have paid increasing attention to young adults psychological and attitudinal profiles, physical environment, and activities. Advertising images were integrated into environments where young adults work, study, and socialize, and smoking was presented as a normal adult behavior, with brand images designed to appeal to specific psychological profiles. Since the 1980s, market segments have increasingly reflected smokers concerns about price and the eroding social acceptability of smoking. METHODS We used standard methods 4,15 to search tobacco-industry document archives on www.jama.com the Internet (http://www.library.ucsf.edu/tobacco, http://legacy.library.ucsf.edu, http://www.tobaccoarchive.com, and http://www.tobaccodocuments.org) between August 2000 and February 2002. Search terms included young adults, younger adult, new smokers, marketing, advertising, segment, college, bars, industry terms for young adult smokers (eg, YAFS [young adult Author Affiliations: Center for Tobacco Control Research and Education, Institute for Health Policy Studies, Cardiovascular Research Institute (Dr Glantz) and Traineeships in AIDS Prevention Studies, Center for AIDS Prevention Studies (Dr Ling), University of California, San Francisco. Corresponding Author and Reprints: Stanton A. Glantz, PhD, Box 0130, University of California, San Francisco, CA 94143 (e-mail: glantz@medicine.ucsf.edu). 2002 American Medical Association. All rights reserved. (Reprinted) JAMA, June 12, 2002 Vol 287, No. 22 2983

female smokers]), lifestyle, motivation, strategy, and the most popular brands for young adult smokers (Marlboro, Camel, and Newport). We conducted a detailed analysis of about 100 marketing research reports, contractors presentations, strategic planning documents, consumer data files, memorandums, and brand reviews focused on young US adult smokers. Although the industry sometimes refers to teens as young adults, the context indicated the documents discussed young adults. RESULTS Tobacco Market Segments Before the 1970s: Product Characteristics Dominate Philip Morris presentations of the historic trends in the cigarette market describe the major market segments according to cigarette characteristics: nonfilters, filters, menthol, length ( 80 mm, 100 mm, 120 mm), low tar, and discount. 16 Philip Morris also studied smokers demographics, including age, sex, income, geographic location, education, and occupation. 17 R.J. Reynolds documents from the 1970s also refer to more than 20 market segments defined by product characteristics, called styles, such as amount of tar or nicotine, menthol or nonmenthol, strength, or shape or length. 18 It was difficult to use segments solely according to product characteristics to analyze changes in the market over time. Philip Morris researcher John Tindall 19 noted that as company researchers paid more attention to the consumer s point of view, they better understood why the market changed. Tindall s retrospective analysis of the market in the 1990s attributes the overall change from unfiltered to filtered cigarettes and the subsequent proliferation of low tar and light brands to consumers increased concern about health and attributes the growth of brands with women s imagery (slimmer and menthol cigarettes) to increased smoking acceptance among women. 16,19 The 1970s and Early 1980s: More Attention to Young Smokers and the Use of Psychographics During the 1970s, R.J. Reynolds investigated alternate ways to define segments that took more consumer characteristics into account. 20 They tried defining market segments according to cigarette taste, social values, and health concerns and found that consumer segments labeled liberal strong flavor, liberal medium flavor, conservative strong flavor, conservative light flavor, and worriers (about health) preferred different cigarette brands. 20,21 R.J. Reynolds studies increasingly focused on the psychological motivations for smoking and smokers lifestyles, 22 which included their leisure activities, jobs, values, and purchase behavior, such as a tendency to buy on impulse. 22 Their 1981 Brand Family Segmentation Study prioritized consumers point of view. Brand style marketing tended to group styles of brands into the same category whether or not the brands competed with each other. For example, Winston 100s and Virginia Slims were in the same category, but they did not compete for the same smokers, and in fact the brands stood for different things. 23 R.J. Reynolds grouped brands that appeared similar to customers, regardless of product characteristics. For example, Camel smokers often switched to Marlboro or Winston; these 3 brands defined a single market segment. To consumers, they shared a macho, rugged image; the segment was nicknamed Virile. 23 The market segments that emerged from this study each matched a key consumer need: Virile (masculinity), Stylish (stylish image), Coolness (alleviation of hot harsh taste [menthol brands]), Moderation (alleviation of tar concerns), and Concerned (more health concerns). 24 R.J. Reynolds examined demographic characteristics after the segments were defined by consumer desires. Significantly, the Virile segment contained the most young smokers. 25 The company analyzed the changes in each segment over time (FIGURE); the Virile segment had the largest and most stable market share from the 1960s through the 1980s. 26 The Virile segment s stability and the success of R.J. Reynolds Virile segment competitor, Philip Morris Marlboro was attributed to Marlboro s large share of beginning smokers. R.J. Reynolds tailored its marketing to the youngest smokers and the Virile segment psychological profile during the 1980s with its Joe Camel campaign. 27,28 Philip Morris tested segmenting the cigarette market along nondemographic lines by using smokers attitudes in 1968 but did not redefine segments then. 17,29 The company continued to monitor demographic trends by using public health data and their annual tracking studies of about 25000 smokers. 19 In 1983, Philip Morris discussed developing new segmentation strategies based on smokers leisure activities, political views, media use, attitudes, and goals. 30-33 One report reviewing many market-segmentation techniques recommended that Philip Morris hire an expert to develop a new market-segmentation method internally. 30 The 1990s: Psychographic Segmentation of Young Adults During the 1990s, Philip Morris conducted several market-segmentation studies solely on young adults aged 18 to 24 years. They hired Michael Normile Marketing to study young adults personalities and smoking motivations in 1990. The goal was to help the company maximize current brand positioning and define new areas of opportunity. Philip Morris was interested in understanding this target market group in more detail, particularly the different personality types among these young adult male smokers, the degree to which the current brands in the market fit the needs of different personality types, the images each type perceived about the major brand, and their social behavior, interests, and concerns for the 1990s. 34 Normile s studies included questions about brand popularity, trial, and expectations; car or motorcycle, job, income, vacation, night out, and party activities; friends 2984 JAMA, June 12, 2002 Vol 287, No. 22 (Reprinted) 2002 American Medical Association. All rights reserved.

and girlfriends; living space; cologne; brand of soft drink, beer, and jeans preferred; and overall personality. 35 Between December 1991 and January 1992, Philip Morris conducted a Young Adult Smokers segmentation study of male and female smokers aged 18 to 24 years. Subjects described their smoking attitudes, general attitudes and goals, the people they socialized with, and main activities. The generalattitude data were used to segment young male smokers into 4 groups with distinct personalities: Macho Hedonists, 50s Throwbacks, Enlightened Go- Getters, and New Age Men. 36-38 This study was used to increase understanding of young-adult smoking behavior and analyze the threat posed by R.J. Reynolds Joe Camel campaign. 39 The Enlightened Go-Getters and New Age Men were less attracted to a traditional masculine image, 40,41 and the Marlboro man appeared isolated, outdated, and old compared with the carefree, social Joe Camel. 40 The Enlightened Go-Getters and New Age Men were also more price sensitive and interested in Camel s promotions. 36,39,41 After this study, Philip Morris sought to develop new brands that combined a social image with cost savings to target these segments 42 and update the Marlboro image to appeal to more modern views of masculinity. 40,43 The same methods used in the 1992 Young Adult Male Smoker segmentation study were used to study young adult female smokers. The 4 female segments, 90s Traditionalists (25%), Uptown Girls (24%), Mavericks (25%), and Wallflowers (21%), 36 differed in attitudes, social patterns, and cigarette brand choice. The 90s Traditionalists tended to be home and community oriented, married, and bargain hunters who were easily tempted by discounts and had no concern about the negative image of cheaper cigarettes. They were less likely to smoke for social reasons, such as to feel more comfortable in a group. 36 Uptown Girls were success oriented and status conscious, had higher income, were single, and had a high value of brand images. They were not interested in discounts or premiums. 36 The TABLE summarizes the evolution of market segmentation. Marketing strategies that emerged from lifestyle and attitude segmentation studies emphasized reaching young adults through activities and environments, with messages tailored to appeal to their attitudes and values. For example, the ideals and activities of the Virile segment led R.J. Reynolds to expand sponsorships in racing and sports events and develop promotions at places where young adults learn new social behaviors, such as bars, clubs, and springbreak parties. 44-49 Philip Morris followed with its own promotional activities designed to appeal to fun-seeking young adults in social environments. 50-54 Figure. R.J. Reynolds Illustration of Changing Major Segments From 1960 to 1985 The major trends noted in the report were the disappearance of the Traditional segment, which has lost 90% of its share since the 1950s, the rise and stability of the Virile segment, which has been stable at 40% since the 1960s, the growth of the Coolness (menthol) segment, which peaked in the 1970s and then began to decline, the decline of the Moderation (low-tar) segment since 1980, and the sharp growth of the Savings segment since its introduction in 1980. 26 2002 American Medical Association. All rights reserved. (Reprinted) JAMA, June 12, 2002 Vol 287, No. 22 2985

Tobacco Industry Weaknesses: Price During the 1980s, R.J. Reynolds observed an unprecedented growth in a new Savings segment, particularly among young smokers, a group that had been willing to pay extra for premium brand names. 26 The company found that it could attract young Savings-segment smokers by offering savings opportunities with youthful appeal: instant gratification offers, such as buy 1, get 1 free; instant coupons; and free promotional items. In 1985, R.J. Reynolds focus groups reinforced that price tactics were relevant to young adults: Although it is cool to say you are not materialistic, the FUBYAS [first usual brand young adult smokers] of today are when you take into consideration their love for expensive cars and audio/video equipment. Any savings on cigarettes or opportunities to save on important items, therefore, could be relevant: 30 cigarettes/pack. Priced lower than Marlboro. Money back refund at store if not pleased. Coupons for free junk food (Wendy s etc). I.O.U. s on pack to be used when funds are low, eg just before payday. Coupons for record, tapes, MTV... Promotion that lets respondent think he is ripping off the company or getting away with something. Call an 800#. Operator flirts and offers you three free packs instead of one. 55 [emphasis in original] The Savings segment was also important. In 1987, it was the only segment with sustained growth in an overall declining US cigarette market. 56 By the 1990s, R.J. Reynolds classified one third of the market as savings motivated. 57 They increasingly used promotional items and instant-gratification deals to attract the growing numbers of young adults entering the Savings segment. 44-47,55 Philip Morris also observed a growth in the Discount segment starting in the 1980s. Its psychographic segmentation studies on young smokers in the 1990s found that a surprising number of young smokers were price vulnerable. An August 1990 study found that one third of young adult male smokers were price sensitive and concluded that it could become a problem because they were a target for discount brands. 34 Philip Morris s 1993 Young Adult Female Smoker study of female vulnerability to discount brands found that the 90s Traditionalists segment was remarkably price sensitive. 36 Tobacco Industry Weaknesses: Declining Social Acceptability Tobacco companies closely monitored the social acceptability of smoking. Philip Morris conducted segmentation studies based on attitudes about smoking issues (smoking and health, the dangers of secondhand smoke, social pressures against smoking, or opinions about government regulation) among both smokers and nonsmokers in the 1960s and in 1988, 1990, 1991, and 1994. 29,58-61 They consistently showed that about half of smokers felt uncomfortable about smoking, largely because of price and smoking s effects on other people. 58,60,61 Philip Morris studies of young adult smokers also showed that different segments expressed social pressures differently. Wallflowers were likely to feel pressured not to smoke. 36 In contrast, 90s Traditionalists were the least likely Table. Philip Morris and R.J. Reynolds Tobacco Companies Cigarette Market Segmentation Strategies Basis for Market Segments Examples of Market Segments Major Brands in Segment 1960s Product characteristics and demographics Politics and flavor strength (R.J. Reynolds, 1976) Brand switching and core consumer desires (R.J. Reynolds s 1985 Segment Description Study) Smoking career and 4 basic consumer desires (R.J. Reynolds strategic plan, 1991-1995) Psychographic segments of young adults (Philip Morris studies in 1994) Menthol unfiltered, filtered, 70-mm, 80-mm, 100-mm, low-tar/nicotine 1970s Liberal strong flavor (16%); liberal medium flavor (27%); conservative strong flavor (14%); conservative light flavor (26%); worriers (17%) 1980s Traditional (5% older adults not image oriented), Virile (40% masculine image), Coolness (19% menthol), Stylish (11% older female upscale image), Moderation (10% conflicting feelings about smoking), Concerned (8% social pressure and health), Savings (7% price sensitive) 1990s Personal concern (33%); social guilt (24%); price: quality shopper (14%), smart shopper (10%), price shopper (9%); young smokers: irreverent approach to life (14%); originality and status (12%) Segmented 18- to 24-year-old male smokers: Macho Hedonists (20%), 50s Throwbacks (24%), Enlightened Go-Getters (28%), New Age Men (22%); segmented 18- to 24-year-old female smokers: 90s Traditionalists (25%), Uptown Girls (24%), Mavericks (25%), Wallflowers (21%) Brands belong to multiple segments Brands belong to multiple segments Traditional: Pall Mall, Lucky Strike, L&M; Virile: Camel, Marlboro, Winston; Coolness: Kool, Newport, Salem; Stylish: B&H, More, Virginia Slims; Moderation: Vantage, Merit; Concerned: Kent, Now, Carlton; Savings: generic, Doral, Century Personal concern: Vantage, Now; social guilt: Horizon, Vantage, new brands; quality shopper: Winston; smart shopper: Sterling, Magna; price shopper: Doral; irreverent: Camel, Dakota, Salem, Marlboro, Newport; originality: Salem, new brands Brands belong to multiple segments but are more applicable to the following: 50s Throwback: Marlboro; Enlightened Go-Getters: Camel, Newport; New Age Men: Camel, discount; 90s Traditionalists: Marlboro Lights, 100s, discount; Uptown Girls: Newport, Virginia Slims; Mavericks: Marlboro Red, Camel; Wallflowers: Marlboro 100s, B&H, Eve, Salem 2986 JAMA, June 12, 2002 Vol 287, No. 22 (Reprinted) 2002 American Medical Association. All rights reserved.

to agree with statements expressing the so-called social benefits of smoking. 36 A Philip Morris study of young adult male smokers found that Marlboro lowflavor (all versions of Marlboro Lights and Mediums) smokers were less comfortable with smoking than Marlboro full-flavor and Camel smokers. 62 R.J. Reynolds also studied social forces, such as peer influence, affecting young adult smokers in the late 1980s. 9 Its strategic planning documents from the 1990s suggest that the company modified its market segmentation strategy to incorporate growing social concerns about smoking. 57,63 The 1990 to 1992 strategic plan identified 4 major consumer desires used to define market segments: taste or image, reduced personal smoking negatives (such as health concerns), reduced smoking negatives for nonsmokers, and reduced price. 63 More smokers were concerned about the effects of secondhand smoke on others. R.J. Reynolds defined a new segment, Social Guilt, which accounted for 24% of the market. 57 Philip Morris also used attitude and lifestyle segmentation to plan marketing outside the United States. Its contractor, Michael Normile Marketing, performed segmentation studies of young smokers (generally aged 18 to 24 years but younger in some countries) to identify new marketing opportunities in Canada, Germany, Switzerland, Singapore, Hong Kong, and Japan. The study results presented in 1986 show that most countries had a segment that was concerned about the social and health effects of smoking; almost every country had a youthful, sociable segment, and most had a price-sensitive segment. 64 COMMENT Tobacco Industry Targets Attitudes and Lifestyle, Not Only Demographics Public health surveillance generates demographic data, and funding streams and public health interventions tend to target groups according to demographic characteristics, such as age, sex, and race. Although these factors predict health outcomes, they can lead to programs for demographic targets rather than groups with similar attitudes or activities. Some tobacco-industry marketing programs have been aimed at specific racial and ethnic groups (such as Philip Morris and R.J. Reynolds Hispanic- and black-focused campaigns 47,65 ), but these efforts are part of larger marketing programs defined by brand or smoking behavior. In contrast to public health programs, tobacco marketers define targets primarily by consumer behavior: they study large groups of smokers and nonsmokers or smokers of specific brands and compare them. Attitude and lifestyle questionnaires define market segments; the demographic characteristics of the segment are noted after these segments are defined. The tobacco industry s tailored messages to different market segments who are often reached through different communications channels explain how smoking can be promoted with logically contradictory messages (eg, as a way to rebel among one group and a way to conform among another). Targeting tobacco control efforts to vulnerable attitudinal segments may be more effective and efficient than targeting to demographic groups who may have redundant or conflicting attitudes, peer profiles, or lifestyle activities. Attitude segmentation studies can also guide priorities for countermarketing programs. For example, industry studies * suggest that interventions that increase price and reduce social acceptability will affect large segments of young adult smokers. These studies also suggest that, although young adult smokers are less concerned about the long-term health consequences of smoking, many (such as Philip Morris Enlightened Go-Getters men or Wallflowers women) are sensitive to the social pressures against smoking. 36 Industry studies also reveal that smokers and nonsmokers attitudes affect smoking behavior and that both overwhelmingly agree on the dangers of smoking and secondhand smoke. 60,61 *References 26, 28, 36, 38, 56-59, 62, 63, 66. Clinical interventions and public health campaigns targeted only to smokers ignore a large pool of nonsmoking allies that can be turned into a powerful constituency to undermine the social acceptability of smoking. 67 Some tobaccoindustry studies suggest that, in particular, people who successfully quit smoking may be a strong source of social pressure to encourage smokers to quit. 11 Efforts addressing the social acceptability of smoking and the social acceptance of the tobacco industry should focus on smokers and nonsmokers. Exposing Tobacco-Industry Manipulation Appeals to Different Attitudinal Segments Tobacco-control efforts that expose tobacco-industry manipulation also decrease the social acceptability of smoking and are effective in teen and adult tobacco control. 68-70 The irreverent tone and bold statements in these campaigns were designed to appeal to teens; they also appeal to the rebellious and excitement-seeking segments of young adults (such as Philip Morris Macho Hedonists or Mavericks). Campaigns that encourage activism against the tobacco industry provide an alternative to smoking as an act of rebellion: to fight corporate manipulation rather than be seduced by it. Exposing specific manipulative targeting in tobaccoindustry campaigns, such as Philip Morris program to reach black smokers with so-called inner-city bar nights 65 or R.J. Reynolds plan to target homosexual and homeless young people with Project SCUM 71 (subculture urban marketing), may be useful in reaching other rebellious smokers. The tobacco industry s public relations campaigns that started in 1999 72-74 underscore the importance of corporate image the social acceptance of the industry itself. The industry s understanding of this fact probably explains why it works so hard to prevent education campaigns about the industry s dishonesty. 70,75,76 Exposing manipulative marketing techniques may also decrease the social acceptance of tobacco-sponsored activities and remove their masquerade of harmless fun. 2002 American Medical Association. All rights reserved. (Reprinted) JAMA, June 12, 2002 Vol 287, No. 22 2987

Focusing on Environmental Change Is Particularly Effective for Young Adults Tobacco marketers know that the youngest smokers tend not to think about their smoking behavior much. The younger the target, the more the act of smoking is automatic, not reasoned. 66 The habitual presence of smoking in the physical environment is a powerful inducement to smoke among young people. Tobacco promotions and advertisements have saturated areas where young people adopt new behaviors, such as work, military service, college, and bars, 48,49,77,78 and aggressively established a brand presence in young adult lifestyle contexts, including music, sports, racing, and dance. 9,28,44-46,50-54,79 These arenas also provide important opportunities for implementing tobacco-control efforts related to secondhand smoke and reducing social support for smoking. Clean indoor air policies on school campuses, at workplaces, in the military, at bars and clubs, and at sports events may discourage young adults from making the transition from being occasional smokers to being confirmed daily smokers. Making homes smoke free is also a powerful deterrent of teen smoking initiation 80 and motivates attempts to quit among adults 81-84 and teens. 80 Limitations There are approximately 40 million pages of internal tobacco-industry documents available to the public. The volume and the often poor quality of indexing provided by the industry make locating all relevant documents difficult. However, the industry marketing activities portrayed are consistent and replicated by several tobacco companies. Although some advertising campaigns refer to objectives stated in industry marketing plans, the relationship is not always explicit. The success of individual campaigns is not always known and cannot be attributable solely to segmentation strategies. The studies nonetheless provide insight into the way tobacco marketers view and approach targets for multimillion-dollar campaigns. The industry may have used other tactics. CONCLUSION Young adults have been targeted by the tobacco industry and deserve attention from public health. Industry marketing techniques provide an alternative framework for planning public health interventions that define new targets and suggest new strategies for tobacco-control interventions. Market segmentation strategies based on the attitudes, lifestyle, aspirations, and activities of young adults may be more useful than demographic data alone. Physician counseling and public health campaigns that identify with the psychological needs and values of smokers and nonsmokers may improve smoking prevention and cessation efforts. Interventions that affect cigarette prices, acceptance of the tobacco industry, the social acceptability of smoking, and secondhand tobacco smoke particularly threaten the industry. Funding/Support: This work was supported by National Institutes of Health grant T32 MH-19105 and National Cancer Institute grant CA-87472. REFERENCES 1. Perry CL. The tobacco industry and underage youth smoking. Arch Pediatr Adolesc Med. 1999;153:935-941. 2. Pollay RW. Targeting youth and concerned smokers. Tob Control. 2000;9:136-147. 3. Hastings G, MacFadyen L. A day in the life of an advertising man. BMJ. 2000;321:366-371. 4. Ling PM, Glantz SA. Why and how the tobacco industry sells cigarettes to young adults. Am J Public Health. In press. 5. Johnston ME, Daniel BC, Levy CJ. Young smokers prevalence, trends, implications, and related demographic trends [Philip Morris USA Research Center article]. March 31, 1981. Bates No. 1000390803/ 0855. Available at: http://www.pmdocs.com/getallimg.asp?docid=1000390803/0855. Accessibility verified 6. Mowery P, Brick P, Farrelly M. First Look Report 3: Pathways to Established Smoking: Results From the 1999 National Youth Tobacco Survey. Washington, DC: American Legacy Foundation; 2000:1-24. 7. Conjoint simulation model [Philip Morris Tobacco Company Web site]. February 23, 1993. Bates No. 2045199142/9311. Available at: http://www.pmdocs 8. 1981 Segmentation study [RJ Reynolds online litigation document repository]. Bates No. 501233021-3038. Available at: http://www.rjrtdocs.com. Accessibility 9. Business Information Analysis Corporation. RJR young adult motivational research [RJ Reynolds online litigation document repository]. January 10, 1985. Bates No. 502780379-0424. Available at: http://www.rjrtdocs 10. Teague CE. Research planning memorandum on some thoughts about new brands of cigarettes for the youth market [RJ Reynolds online litigation document repository]. February 2, 1973. Bates No. 502987357-7368. Available at: http://www.rjrtdocs 11. Hugh Bain Research. The psychology and significant moments and peak experiences in cigarette smoking [University of California, San Francisco, Tobacco Control Archives Web site]. November 1993. Bates No. 500237804-7890. Available at: http://www.library.ucsf.edu/tobacco. Accessibility 12. Friedman L, Lazarsfeld P, Meyer A. Motivational conflicts engendered by the on-going discussion about cigarette smoking [Philip Morris Web site]. January 1972. Bates No. 1003291740/1916. Available at: http://www.pmdocs 13. Kotler P, Armstrong G. Principles of Marketing. 8th ed. Upper Saddle River, NJ: Prentice Hall; 1999. 14. Urban GL, Hauser JR. Design and Marketing of New Products. 2nd ed. Englewood Cliffs, NJ: Prentice Hall; 1993. 15. Malone RE, Balbach ED. Tobacco industry documents. Tob Control. 2000;9:334-338. 16. US Cigarette Market [Philip Morris Web site]. February 1994. Bates No. 2057977028/7063. Available at: http://www.pmdocs.com. Accessibility verified 17. The cigarette market national tracking study [Philip Morris Web site]. August 1970. Bates No. 2040071212/1262. Available at: http://www.pmdocs 18. Segment fact sheets [RJ Reynolds online litigation document repository]. Bates No. 500098477-8499. Available at: http://www.rjrtdocs.com. Accessibility 19. Tindall J. Cigarette market history and interpretation and consumer research [Philip Morris Web site]. February 13, 1992. Bates No. 2057041153/1196. Available at: http://www.pmdocs.com. Accessibility 20. Cigarette market segmentation study [RJ Reynolds online litigation document repository]. February 27, 1976. Bates No. 501087046-7084. Available at: http://www.rjrtdocs.com. Accessibility verified 21. Cigarette market segmentation study: Los Angeles Metro [RJ Reynolds online litigation document repository]. January 27, 1977. Bates No. 500995624-5629. Available at: http://www.rjrtdocs.com. Accessibility 22. Strategic research report: the 1981 brand family segmentation study [RJ Reynolds online litigation document repository]. April 14, 1982. Bates No. 500455035-5124. Available at: http://www.rjrtdocs 23. Fields TF. 1981 segmentation study [RJ Reynolds online litigation document repository]. June 11, 1982. Bates No. 501743228-3232. Available at: http://www.rjrtdocs 24. McCann R. 1981 Market overview: product perspective by segment [RJ Reynolds online litigation document repository]. October 28, 1982. Bates No. 500119052-9057. Available at: http://www.rjrtdocs 25. Analysis of the Virile segment [University of California, San Francisco, Tobacco Control Archives Web site; Mangini Collection]. Bates No. 507242216-2280. Available at: http://www.library.ucsf.edu /tobacco/. Accessibility 26. 1985 segment description study: overview presentation [RJ Reynolds online litigation document repository]. Bates No. 506203246-3327. Available at: http:// www.rjrtdocs 27. Younger adult smoker opportunity: purpose [RJ Reynolds online litigation document repository]. April 1988. Bates No. 506653159-3218. Available at: http://www.rjrtdocs 28. Question: are young adult smokers important [RJ Reynolds online litigation document repository]? Bates 2988 JAMA, June 12, 2002 Vol 287, No. 22 (Reprinted) 2002 American Medical Association. All rights reserved.

No. 507515518-5625. Available at: http://www.rjrtdocs 29. Marplan Research. Pilot study on cigarette market segmentation [Philip Morris Tobacco Company Web site]. January 1968. Bates No. 2001256105/ 6171. Available at: http://www.pmdocs.com. Accessibility 30. Beane T. Market segmentation in a technology push environment: recommendations [Philip Morris Tobacco Company Web site]. June 23, 1983. Bates No. 2022151950/1954. Available at: http://www.pmdocs 31. Beane T, Ennis D, Rowe C. Market segmentation in a technology push environment [Philip Morris Tobacco Company Web site]. July 1, 1983. Bates No. 2051989551/9652. Available at: http://www.pmdocs 32. Johnston M. Market segmentation [Philip Morris Tobacco Company Web site]. August 11, 1983. Bates No. 1003285101/5104. Available at: http://www.pmdocs 33. Mendell S. Market segmentation [Philip Morris Tobacco Company Web site]. August 19, 1983. Bates No. 1003659204/9205. Available at: http://www.pmdocs 34. Michael Normile Marketing. Young adult smokers opportunity profiles [Philip Morris Tobacco Company Web site]. August 1990. Bates No. 2043102805/ 2944. Available at: http://www.pmdocs.com. Accessibility 35. Michael Normile Marketing. Proposal for new brand identification and concept development [Philip Morris Tobacco Company Web site]. March 1990. Bates No. 2049398247/8258. Available at: http://www.pmdocs 36. Ansberry C, Oboyle T, Roberts J. Today s adult young male smoker [Philip Morris Tobacco Company Web site]. August 1992. Bates No. 2060127742/ 7919: 60-178. Available at: http://www.pmdocs 37. Marketing Information Systems International. YAS market structure study number 83-77-91 [University of California, San Francisco, Legacy Tobacco Documents Library Web site. Bates No. 2049266352/ 6697. Available at: http://legacy.library.ucsf.edu. Accessibility 38. YAS segmentation study [Philip Morris Tobacco Company Web site]. May 13, 1993. Bates No. 2040838100/8198. Available at: http://www.pmdocs 39. Analysis of Camel threat to Philip Morris [Philip Morris Tobacco Company Web site]. May 1988. Bates No. 2024462782/2931. Available at: http://www.pmdocs 40. Bruce Eckman Inc. The viability of the Marlboro Man among the 18-24 segment [Philip Morris Tobacco Company Web site]. March 1992. Bates No. 2045060177/0203. Available at: http://www.pmdocs 41. Camel smoker profile demographics [Philip Morris Tobacco Company Web site]. December 1990. Bates No. 2044124506/4515. Available at: http://www.pmdocs 42. Young and Rubicam. Brand X: a qualitative exploratory among young adult male smokers [Philip Morris Tobacco Company Web site]. February 1994. Bates No. 2040174998/5045. Available at: http://www.pmdocs 43. Millward Brown awareness, advertising, and image tracking study [Philip Morris Tobacco Company Web site]. March 1993. Bates No. 2045169011/ 9099. Available at: http//www.pmdocs.com. Accessibility 44. Kevin By and Associates. Trend influence marketing program [Galen II Tobacco Control Archives Web site]. April 14, 1994. Bates No. MB(2)25655-MB(2) 25687. Available at: http://www.library.ucsf.edu /tobacco/mangini. Accessibility 45. Entertainment Marketing and Communications International Ltd. Camel s make your own music video [University of California, San Francisco, Tobacco Control Archives Web site]. Bates No. 507366721-6731. Available at: http://www.library.ucsf.edu/tobacco /mangini. Accessibility 46. Promotional Marketing Inc. Camel spring resort program Daytona Beach, 1989 [Galen II Tobacco Control Archives Web site]. Bates No. 507240802-0812. Available at: http://www.library.ucsf.edu/tobacco /mangini. Accessibility 47. Promotional Marketing I. Camel Hispanic program [University of California, San Francisco, Tobacco Control Archives Web site]. February 15, 1989. Bates No. 507133229-3239. Available at: http://www.library.ucsf.edu/tobacco/mangini. Accessibility verified 48. Sepe E, Ling PM, Glantz SA. Smooth moves: bar and nightclub tobacco promotions target young adults. Am J Public Health. 2002;92:414-419. 49. Sepe E, Glantz SA. Tobacco promotions in the alternative press. Am J Public Health. 2002;92:75-78. 50. Marlboro country 1994 dance club night program [Philip Morris Tobacco Company Web site]. Bates No. 2041942879/2885. Available at: http://www.pmdocs 51. Marlboro racing bar nights [Philip Morris Tobacco Company Web site]. August 1993. Bates No. 2040993113/3116. Available at: http://www.pmdocs 52. Gennaro MJ. Marlboro country nights dance showdown 94 [Philip Morris Tobacco Company Web site]. May 31, 1994. Bates No. 2061803735/3737. Available at: http://www.pmdocs.com. Accessibility 53. GMR Marketing. Parliament party zone 1995 program [Philip Morris Tobacco Company Web site]. August 11, 1994. Bates No. 2040575026/5043. Available at: http://www.pmdocs.com. Accessibility verified 54. Howe K. Special markets 91 Marlboro regional event marketing [Philip Morris Tobacco Company Web site]. February 15, 1991. Bates No. 2043685428/ 5441. Available at: http://www.pmdocs.com. Accessibility 55. Nassar SC. Marlboro vulnerability [University of California, San Francisco, Tobacco Control Archives Web site]. January 21, 1985. Bates No. 503043738-3746. Available at: http://www.library.ucsf.edu/tobacco /mangini/. Accessibility 56. RJ Reynolds Tobacco Company. Input to the long term strategic plan for 1988-90 [University of California, San Francisco, Tobacco Control Archives Web site]. June 29, 1987. Bates No. 509179045-9138. Available at: http://www.library.ucsf.edu/tobacco/mangini/html /u/029/. Accessibility 57. RJ Reynolds Tobacco Company. RJRTC 1991-1995 strategic plan [R.J. Reynolds online litigation document repository]. August 1990. Bates No. 513330214-0338. Available at: http://www.rjrtdocs 58. How today s smokers feel about smoking issues [Philip Morris Tobacco Company Web site]. March 1988. Bates No. 2001260000/0114. Available at: http://www.pmdocs 59. Non-smokers views toward smoking issues [Philip Morris Tobacco Company Web site]. March 1989. Bates No. 2045184104/4188. Available at: http://www.pmdocs 60. The 910000 [1991] study of smokers and nonsmokers attitudes [Philip Morris Tobacco Company Web site]. January 1992. Bates No. 2024790388/ 0472. Available at: http://www.pmdocs.com. Accessibility 61. Smoker/non-smoker segmentation study [Philip Morris Tobacco Company Web site]. February 24, 1994. Bates No. 2045165478/5583. Available at: http://www.pmdocs 62. Profile of the young adult Marlboro smoker part I: males, 18 to 24 years old [Philip Morris Tobacco Company Web site]. August 1994. Bates No. 2061804612/ 4673. Available at: http://www.pmdocs.com. Accessibility 63. Strategic plan 1990-1992 [University of California, San Francisco, Tobacco Control Archives Web site]. Bates No. 506831898-1930. Available at: http://www.library.ucsf.edu/tobacco. Accessibility 64. Normile M, Bolleschweiller F, Drummond D, et al. Market research seminar 860900, volume IV [Philip Morris Tobacco Company Web site]. September 1986. Bates No. 2043101172/1425: 218-254. Available at: http:// www.pmdocs 65. Menthol: event sponsorship Marlboro menthol inner city bar nights [Philip Morris Tobacco Company Web site]. Bates No. 2048678695-8697. Available at: http:// www.pmdocs 66. Young adult male smokers and young adult female smokers [Philip Morris Tobacco Company Web site]. November 1993. Bates No. 2045180640/ 0775. Available at: http://www.pmdocs.com. Accessibility 67. California Department of Health Services Tobacco Control Section. A Model for Change. Sacramento: California Dept of Health Services; 1998. 68. Bauer UE, Johnson TM, Hopkins RS, Brooks RG. Changes in youth cigarette use and intentions following implementation of a tobacco control program. JAMA. 2000;284:723-728. 69. Goldman LK, Glantz SA. Evaluation of antismoking advertising campaigns. JAMA. 1998;279:772-777. 70. Healton C. Who s afraid of the truth? Am J Public Health. 2001;91:554-558. 71. Project SCUM [RJ Reynolds online litigation document repository]. December 12, 1995. Bates No. 518021121-1129. Available at: http://www.rjrtdocs 72. Elliott S. Tired of being a villian, Philip Morris works on its image. New York Times. November 11, 1999; sect C:12. 73. Gugliotta G. Tobacco firms try to clear smoke from their images. Washington Post. October 13, 1999; sect A:A16. 74. Mann J. Selling responsibility. Kansas City Star. December 17, 1999;Business section:c1. 75. Glantz S, Balbach E. Tobacco War. Berkeley, Calif: University of California Press; 2000. 76. Balbach ED, Glantz SA. Tobacco control advocates must demand high quality media campaigns. Tob Control. 1998;7:397-408. 77. Magzamen S, Glantz SA. The new battleground. Am J Public Health. 2001;91:245-252. 78. Magzamen S, Charlesworth A, Glantz SA. Print media coverage of California s smokefree bar law. Tob Control. 2001;10:154-160. 79. Layne S, Murphy K. Daytona promotions [University of California, San Francisco, Tobacco Control Archives Web site]. March 27, 1989. Bates No. 506868845-8847. Available at: http://www.library.ucsf.edu/tobacco /mangini. Accessibility 80. Glantz SA, Jamieson P. Attitudes toward secondhand smoke, smoking, and quitting among young people. Pediatrics. 2000;106:E82. 81. Farkas AJ, Gilpin EA, Distefan JM, Pierce JP. The effects of household and workplace smoking restrictions on quitting behaviours. Tob Control. 1999;8: 261-265. 82. Gilpin EA, White MM, Farkas AJ, Pierce JP. Home smoking restrictions. Nicotine Tob Res. 1999;1:153-162. 83. Pierce JP, Gilpin EA, Farkas AJ. Can strategies used by statewide tobacco control programs help smokers make progress in quitting? Cancer Epidemiol Biomarkers Prev. 1998;7:459-464. 84. Pierce JP, Gilpin EA, Emery SL, et al. Tobacco Control in California. La Jolla: California Dept of Health Services; 1998. 2002 American Medical Association. All rights reserved. (Reprinted) JAMA, June 12, 2002 Vol 287, No. 22 2989