(Chest. 2005;128:746-754.)
© 2005
American College of Chest Physicians
Smoking in Contemporary American Cinema*
Karan Omidvari, MD, FCCP;
Klaus Lessnau, MD, FCCP;
Jeannie Kim, MD;
Donald Mercante, PhD;
Ann Weinacker, MD, FCCP and
Carol Mason, MD, FCCP
* From the Pulmonary and Critical Care Section, St. Michael’s Medical Center, Newark, NJ (Dr. Omidvari), Lenox Hill Hospital, New York, NY; New York University School of Medicine (Dr. Lessnau), New York, NY; the Section of Pulmonary and Critical Care Medicine (Dr. Kim), University of California, San Diego, San Diego, CA; the Section of Pulmonary and Critical Care Medicine (Dr. Mason), Louisiana State University Health Science Center (Dr. Mercante), New Orleans, LA; and the Section of Pulmonary/Critical Care Medicine (Dr. Weinacker), Stanford University Medical Center, Stanford, CA.
Correspondence to: Karan Omidvari, MD, FCCP, 268 King Blvd, Newark, NJ 07102; e-mail: komidvari{at}yahoo.com
 |
Abstract
|
|---|
Objective: The true prevalence of smoking among characters portrayed in the movies is unknown. This study examines this prevalence objectively.
Design: The top 10 movies on the weekly box office charts were reviewed. Whether or not the top five characters in these movies smoked, was documented. It was determined prior to the start of the study that 300 male characters and 300 female characters were needed to detect any significant difference. A total of 447 movies, composed of 193 movies rated restricted (R) [children < 17 years of age must be accompanied by an adult], 131 movies rated PG13 for parental guidance suggested for children < 13 years of age (PG) and 123 movies rated PG for parental guidance suggested, were examined until the sample size was reached.
Results: Smoking prevalence is the same in contemporary American movies and in the general US population (23.3% vs 24.8%, respectively). However, there was more smoking in these movies among men than among women (25.5% vs 20.5%, respectively; p < 0.006), among antagonists than among protagonists (35.7% vs 20.6%, respectively; p < 0.001), lower vs middle vs upper socioeconomic class (SEC) [48.2%, 22.9%, and 10.5%, respectively; p < 0.001], among independent vs studio movies (46.2% vs 18.2%, respectively; p < 0.001); and among R-rated vs PG13-rated vs PG-rated movies (37.3%, 16.2%, and 8.1%, respectively; p < 0.001). In R-rated movies, and in both subcategories of R-rated studio movies and R-rated independent movies, smoking prevalence is higher than in the US population (37.3%, 30.5%, and 50.6% vs 24.8%, respectively; p < 0.001 for all). Additionally, compared to the US population, men, women and lower SEC members smoke more in R-rated movies, R-rated studio movies, and R-rated independent movies. In R-rated movies, antagonists smoke more than protagonists (43.9% vs 35.8%, respectively; p < 0.001), and whites smoke more than nonwhites (38.3% vs 26.4%, respectively; p < 0.001). In R-rated studio movies, antagonists smoke more than protagonists (42.6% vs 26.6%, respectively; p < 0.001), and men smoke more than women (32.0% vs 27.9%, respectively; p = 0.03). In R-rated independent movies, whites smoke more than nonwhites (51.8% vs 40.5%, respectively; p < 0.001). Smoking prevalence is higher in R-rated independent movies than in R-rated studio movies (50.6% vs 30.5%, respectively; p < 0.001). Smoking prevalence is also higher in R-rated independent movies than in R-rated studio movies in subcategories of men (32.0% vs 49.8%, respectively; p < 0.001), women (21.8 vs 51.8%, respectively; p < 0.001), protagonists (26.6% vs 51.6%, respectively; p < 0.001), whites (31.5% vs 51.8%, respectively; p < 0.001), nonwhites (24.7% vs 40.5%, respectively; p < 0.001), and all three SECs.
Conclusions: In contemporary American cinema, the smoking prevalence is higher for men, antagonistic characters, lower SEC, independent movies, and R-rated movies. Smoking prevalence is higher than in the general US population in R-rated movies, and in both its subcategories of R-rated studio movies and R-rated independent movies. There is more smoking in R-rated independent movies than in R-rated studio movies. Smoking in contemporary American cinema is associated with male sex, lower SEC, and antagonistic (ie, bad) characters.
Key Words: motion pictures • smoking • tobacco • tobacco industry
|
Introduction
|
|---|
The film industry has been criticized because the frequency of smoking in movies has been reported to be increasing since 1990.1234 The depiction of tobacco as sexy and cool may be a reason for increased smoking initiation among adolescents.156789 Movies, with far-reaching entertainment appeal, have the potential to influence behavior in the viewing population and especially in the youth.10 Surprisingly, the public believes that the prevalence of smoking in the movies is decreasing.211 However, there has been no attempt to study the prevalence of smoking in the movies objectively. This study evaluates the prevalence of smoking in the movies and introduces the application of an objective methodology to an art form that has previously been difficult to study.
|
Hypothesis
|
|---|
The null hypothesis is that the prevalence of cigarette smoking is the same among the leading characters in contemporary American movies and in the general US population.
|
Materials and Methods
|
|---|
Contemporary American movies released after 1990 were included. Only movies rated as parental guidance suggested (PG), parental guidance suggested for children < 13 years of age (PG13), and restricted (R) [ie, children < 17 years of age must be accompanied by an adult] were included. G-rated and science fiction movies were excluded as they may not portray real life. The majority of the story in the movie had to occur in the 1990s. Movies that did not take place in the 1990s were excluded because a movie about a different era may not reflect contemporary life. For example, the movie Titanic was excluded even though both main characters smoked. For the same reason, parts of a movie that did not take place in the 1990s, such as "flashbacks," were excluded. Whether or not US citizens smoked was recorded, and only while the character was in the United States. Smoking in a foreign country or by a foreign citizen in the United States was excluded.
In order to avoid viewer bias, we watched all the weekly box office top 10 movies that met these criteria. These movies are viewed by the highest number of people and have the greatest potential impact on smoking habits. We only watched movies unedited in the theaters or on video. Movies on video were watched only in the normal "play" mode. Movie versions aired on network or cable television, and special versions such as rereleases or "director’s cuts" were excluded. To avoid bias, we excluded movies that had cigarette smoking as a central motif or were making intentional comments about smoking, such as The Insider, Blue in the Face, or Smoke. We excluded science fiction and animated movies as it can be argued that they do not portray real life.
We observed whether the top five characters smoked cigarettes at any time during these movies. Prior to the start of the study, it was determined that 300 male characters and 300 female characters had to be included in each of three categories ie, (PG, PG13, and R-rated) to detect a statistical difference. Between January 1, 1996, and December 31, 1998, we watched every American movie rated in the top 10 on the weekly box-office list that met the inclusion criteria in theaters. To reach the 600 necessary characters, we reviewed videotapes of the weekly top-10 movies rated PG and PG13 dating back to January 1, 1990.
"Hollywood studio system" was defined as, "the set of practices by which a few companies control most of the American movie industry."12 There are eight big companies (Warner Brothers, Universal, Sony, Fox, Buena Vista, MGM/UA, Paramount, and Dreamworks SKG) that comprise the Hollywood studio system. Independent movies are those made outside the Hollywood studio system. New Line Cinema, Fine Line Features, Lions Gate Films, Artisan Entertainment, and Destination Films are some of the smaller distribution companies.
An upper socioeconomic class (SEC) member is defined as a person in a household with an income of > $50,000 annually, usually a white collar worker. A lower SEC member is usually a blue collar worker from a household with an income of less than $25,000 annually. A middle SEC member is a person from a household with an income between $25,000 and $50,000 annually.13
Data were analyzed using a statistical software package (SAS, version 3; SAS Institute; Cary, NC) in the Section of Statistics and Biometry of the Louisiana State University Health Science Center. 
2 goodness-of-fit test analysis was used to compare smoking prevalence to the published data for the US population. 2 contingency table analysis was utilized to test the associations between smoking and other independent variables.
|
Results
|
|---|
A total of 447 movies were reviewed (R-rated, 193; PG13-rated, 131; and PG-rated, 123). Of the 447 movies, 68 (15%) had been reviewed by more than one reviewer. Of those 68 movies, 23 (one third) were reviewed by more than two reviewers. There was complete concordance among different reviewers as to whether or not the top five characters smoked cigarettes at any time during these movies.
All Ratings Combined
Smoking prevalence in contemporary American movies and in the general US population14 is similar. When subcategories were analyzed, smoking prevalence in American movies was not higher than that in the general US population for men or women. However, in American movies, men smoke more than women (Fig 1
).
Although trending toward significance, smoking prevalence is not higher among whites than nonwhites. When subcategories were analyzed for ratings, smoking prevalence was higher among whites than nonwhites only in R-rated movies, but not in PG-rated movies or in PG13-rated movies. When subcategories were analyzed for gender, smoking prevalence was higher for white men than for white women, but was not higher for nonwhite men than for nonwhite women, or higher for white men than nonwhite men, or higher for white women than for nonwhite women (Table 1
).
Smoking prevalence is higher in R-rated movies than in PG13-rated movies, and the prevalence was higher in both than that in PG-rated movies. When subcategories of race were analyzed, both whites and nonwhites smoked more in R-rated movies than in PG13-rated movies, PG-rated movies, and both men and women smoked more in R-rated than in PG13-rated movies, and more in both of these than in PG-rated movies. Last, men smoked more than women in PG-rated movies, but not in PG13-rated or R-rated movies (Table 2 ).
Smoking prevalence is higher in antagonists ("bad guys") than in protagonists ("good guys") overall, and in all three subcategories of PG-rated, PG13-rated, and R-rated movies (Fig 2
). Smoking prevalence in American movies is significantly higher in the lower SEC than in the middle SEC, which in turn is higher than that in the upper SEC. This difference is also observed in the subcategory of R-rated movies, but not in PG-rated or PG13-rated movies (Table 3
). Smoking prevalence was markedly higher in independent movies than in studio movies overall, and in all three subcategories of movies rated PG, PG13, and R (Fig 3
).
R-Rated Movies
Smoking prevalence in R-rated American movies is higher than that in the general US population overall and in both subcategories of men and women (Table 4
). In R-rated American movies, smoking prevalence was not higher for men than for women, but was higher for antagonists than for protagonists, for whites than for nonwhites, for white women than for nonwhite women, for lower SEC than for middle SEC, which were higher in turn than for upper SEC, and for independent movies than for studio movies (Table 4).
R-Rated Studio Movies:
Smoking prevalence was higher in R-rated American studio movies than in the general US population overall, and in both subcategories of men and women. In R-rated American studio movies, smoking prevalence was higher for men than for women, for antagonists than for protagonists, for lower SEC than for middle SEC, which in turn were higher than for upper SEC, for white women than for nonwhite women, but not for white men than for nonwhite men, and for whites than for nonwhites overall. Smoking prevalence in these movies was also higher for men than for women in the lower SEC, but not in the middle or the upper SECs (Table 5
).
R-Rated Independent Movies:
Smoking prevalence was higher in R-rated American independent movies than in the general US population, and in both subcategories of men and women. In R-rated American independent movies, smoking prevalence was not different for men than for women, for antagonists than for protagonists, for white women than for nonwhite women, but was higher for whites than for nonwhites overall, for white men than for nonwhite men, and for lower SEC than for middle SEC, which in turn were higher than for upper SEC. There was more smoking in these movies among men than among women in the upper SEC, but not in the lower or in middle SECs (Table 6
).
R-Rated Studio Movies vs Independent Movies:
Smoking prevalence in R-rated independent movies was higher than in R-rated studio movies overall, and in subcategories of men, women, protagonists, whites, white males, white women, nonwhites, and nonwhite women, but not in nonwhite men and antagonists. Subanalyzed for SECs, lower SEC members, middle SEC members, and women in all three SECs smoked more in R-rated independent movies than in R-rated studio movies. However, smoking prevalence in upper SEC members and men in any of the three SECs was the same in R-rated independent and R-rated studio movies (Table 7 ).
|
Discussion
|
|---|
This is the first study to objectively quantify the prevalence of smoking in contemporary American movies. This methodology can be applied to study smoking in other art forms, such as video, television shows, or theater. Equally as important, this study challenges many of the myths about smoking in the movies.
The number of movies released theatrically in the United States each year has varied from a high of 491 in 1988 to a low of 370 in 1995.15 Only 6% of these movies are rated G16 (ie, general admission, suitable for all audiences). We excluded G-rated movies even though Goldstein and colleagues17 have reported that, "more than two thirds of animated children’s films feature tobacco or alcohol use in story plots." The majority of these movies are animations and can be defended as not portraying real life. Science fiction movies were excluded for the same reason. It has been postulated that since adolescents are avid moviegoers5 and are at a stage of life in which they are experimenting with their own image,181920 "movies influence adolescents experimentation with smoking."101920 The three categories of movies rated PG, PG13, and R comprise over 93% of the movies released each year in the United States and are the ones seen by the overwhelming majority of the population, including teenagers.
Previous investigators156 have indicated that 90% of movies and approximately 30% of all movie scenes contain at least one instance of smoking. However, the criteria for documenting smoking in the movies have been arbitrary at best. For instance, Hazan et al10 defined a "tobacco event" as "implied or actual consumption of tobacco, paraphernalia, ashtrays or matches, talking about tobacco, no smoking signs, and tobacco product logos." They selected 2 of the top 20 movies at the box office for each year from 1960 to 1990 and recorded whether any tobacco event had occurred within any 5-min intervals of these movies.6 They concluded that 41% of the 5-min intervals showed tobacco use.6 The same methodology was adopted by other investigators to conclude that from 1985 to 1995, 98% of the top 10 grossing American movies "had evidence to support tobacco use."21
Some authors222324 have argued that Hollywood is intentionally collaborating with the tobacco industry to influence the behavior of the public or a specific segment of the population, for instance, men, women, blacks, or black women.222324 We have demonstrated that contemporary American movies do not have a higher prevalence of smoking than the general US population, overall. Independent movies have a much higher prevalence of smoking than the Hollywood studio movies (Fig 3), in all categories, but most remarkably among R-rated movies (Fig 3).
Escamilla and colleagues24 reported that, "male actors were 2.5 times more likely to smoke in R-rated movies." The current study demonstrates that men smoke 2.5 times more than women only in PG-rated movies. In contrast, men do not smoke more than women in movies rated PG-13 or R (Table 2). And while men smoke more than women (32% vs 27.9%, respectively; p < 0.03) in R-rated studio movies, it is certainly not 2.5 times more (Table 5). Moreover, in R-rated independent movies, women actually smoke more than men, although the difference does not reach statistical significance (Table 5).
Some investigators have reported22232425 that Hollywood is attempting to get minorities, and specifically black women, to smoke. Our data contradict these findings. Men smoke more than women (Fig 1) and white men smoke more than white women (Table 1). However, there is no significant difference in smoking prevalence between whites and nonwhites (Table 1). In addition, neither white men nor white women smoke more than their nonwhite counterparts (Tables 1and 2). Additionally, lower SEC members smoke more than middle SEC members, who in turn smoke more than upper SEC members, overall and in movies rated PG-13 and R (Table 3). Therefore, smoking is a habit of the lower SECs.
Over 65% of movies released each year are rated R.12 These are the movies that are seen by the majority of people, especially teenagers,4 and, as such, they have the greatest potential to influence behavior.10 Therefore, R-rated movies require a detailed analysis on their own.
Whites in general and white women in particular smoke more than nonwhites in general and nonwhite women in particular in R-rated movies (Table 4). In R-rated studio movies, only white women smoke more than nonwhite women (Table 5). In R-rated independent movies, whites and white men smoke more than nonwhites in general and nonwhite men in particular (Table 6). These data suggest that whites smoke more than nonwhites and that white women smoke more than nonwhite women except in R-rated independent movies. These findings contradict previous reports222324 of conspiracy of the Hollywood studio movies to influence (female) minorities to smoke. It also demonstrates that if there is a conscious attempt to influence public smoking habits, it is being orchestrated by independent movies, and not by Hollywood.
Lower SEC members smoke more than middle SEC members, who in turn smoke more than upper SEC members in R-rated movies in general (Table 4), and in subcategories of R-rated studio movies (Table 4) and R-rated independent movies (Table 6). Additionally, in the lower SEC subgroup, men smoke more than women in R-rated studio movies (Table 5). Again, these findings contradict those from previous reports221 that movies portray smokers as middle class and personally and professionally successful. The only discrepancy in our findings occurred in the R-rated independent movies, in which women smoked more than men in the upper SEC subgroup (Table 6).
Antagonists smoked more than protagonists in R-rated movies (Table 4), and in the subcategory of R-rated studio movies (Table 5), but not in R-rated independent movies (Table 6). One explanation may be that while studio movies seem to be utilizing smoking as an antagonistic characteristic, independent filmmakers appear to be dealing with more mature human interests, utilizing smoking to denote inner conflict, regardless of the smoker’s character.
Compared to R-rated studio movies, there is a remarkably higher prevalence of smoking in R-rated independent movies, and in subcategories of men, women, protagonists, whites, nonwhites, both white and nonwhite women, white men, lower and middle SEC members, and in women in all three SEC subgroups (Table 1). One reason for this may be the nature of independent movie making as antiestablishment or free-spirited. Another explanation may be the much-written-about, albeit in nonscientific literature, product placement. It may be that the independent filmmakers are financing their movies with the aid of the tobacco industry. Obviously, this phenomenon merits further investigation, which is beyond the scope of this article. At the same time, the portrayal of smoking in studio and independent movies is not different for nonwhite men, antagonists, men in all three SECs, and upper SEC members. In other words, the majority of R-rated movies, produced both by studios and independently; portray smoking as a male, lower-middle SEC, and, most importantly, antagonistic phenomenon. It is also possible that in some of these subgroups (eg, nonwhite men) the sample size was too small to reach statistical significance.
In this study, many measures were used to avoid bias. We watched every movie on the weekly top 10 box office list that met the inclusion criteria. It can be argued that the study discriminated against movies that were not on the weekly top 10 box office list. But those movies are not seen by many viewers and therefore have a minimal effect on the behavior of the general public. It can be argued that young people are avid fans of science fiction movies and that smoking in those movies can especially signify the survival of the tobacco industry into the future. However, animated movies and science fiction movies can be defended as not portraying real life and therefore we excluded them from the study.
|
Conclusion
|
|---|
The scientific study of an art form such as motion pictures is fraught with difficulty. As such, previous studies have used arbitrary means to quantify the prevalence of smoking in the movies. This article introduced a methodology that can be applied to quantitatively study such notions such as prevalence, objectively.
We demonstrated that, although the smoking prevalence is not higher in movies than in the general population overall, it is higher in movies that are targeted at youth (ie, the R-rated movies). Furthermore, smoking prevalence is markedly higher in R-rated independent movies than in R-rated studio movies.
It was previously reported6 that, "films portrayed a smoker who is typically white, male, middle class, successful, and attractive, a movie hero who takes smoking for granted." In contrast, we have demonstrated that contemporary American films portray a smoker who may be white and male, but is likely to be of lower SEC, unsuccessful, and a villain.
|
Footnotes
|
|---|
Abbreviations: G = general admission; PG = parental guidance suggested; PG13 = parental guidance suggested for children < 13 years of age; R = restricted; SEC = socioeconomic class
Received for publication September 18, 2003.
Accepted for publication February 10, 2005.
|
References
|
|---|
- Stockwell, TF, Glantz, SA (1997) Tobacco use is increasing in popular films. Tob Control 6,282-284[Abstract]
- Dalton, MA, Tickle, JJ, Sargent, JD, et al The incidence and context of tobacco use in popular movies from 1988 to 1997. Prev Med 2002;34,516-523[CrossRef][ISI][Medline]
- Mekemson, C, Glantz, SA How the tobacco industry built its relationship with Hollywood. Tob Control 2002;11(suppl),I81-I91[Medline]
- Shield, DL, Carol, J, Balbach, ED, et al Hollywood on tobacco: how the entertainment industry understands tobacco portrayal. Tob Control 1999;8,378-386[Abstract/Free Full Text]
- Terre, L, Drabman, RS, Speer, P Health-relevant behaviors in media. J Appl Soc Psychol 1991;21,1303-1319[CrossRef][ISI]
- Hazan, AR, Lipton, HL, Glantz, SA Popular films do not reflect current tobacco use. Am J Public Health 1994;84,998-1000[Abstract/Free Full Text]
- Klein, JD, Brown, JD, Childers, KW, et al Adolescents’ risky behavior and mass media use. Pediatrics 1993;92,24-31[Abstract/Free Full Text]
- Centers for Disease Control and Prevention.. Youth tobacco surveillance: United States, 2000. MMWR CDC Surveill Summ 2001;50,1-84[Medline]
- Madden, PA, Grube, JW The frequency and nature of alcohol and tobacco advertising in televised sports, 1990 through 1992. Am J Public Health 1994;84,297-299[Abstract/Free Full Text]
- Distefan, JM, Gilpin, EA, Sargent, JD, et al Do movie stars encourage adolescents to start smoking? Prev Med 1999;26,1-11[CrossRef]
- Burns, D, Pierce, JP Tobacco use in California 1990–91. 1992 Department of Health Service. Sacramento, CA:
- Katz, E The film encyclopedia 4th ed. 2001 Harper Collins Publishers. New York, NY:
- Duncan, G, Smeeding, T, Rogersa, W W(h)ither the middle class? A dynamic view. Papadimitriou, DP Wolff, EN eds. Poverty and prosperity in the USA in the late twentieth century 1996 St. Martin’s Press. New York, NY:
- Center for Disease Control and Prevention, National Center for Health Statistics.. MMWR; 2000;49,977-982
- Gimmy, A, Gates, M The business of show business: the evaluation of movie theaters; 1999 MAI. the Appraisal Institute. Chicago, IL:
- Motion Picture Association of America. Lagniappe column. The Times-Picayune, July 13, 1999
- Goldstein, AO, Sobel, RA, Newman, GR Tobacco and alcohol use in G-rated children’s animated films. JAMA 1999;281,1131-1136[Abstract/Free Full Text]
- Snyder, S Movies and adolescents: an overview. Adolesc Psychiatry 1992;18,74-90[Medline]
- Johnston, L National survey results on drug use from the monitoring the future study, 1975–1995. Johnston, LD O’Malley, PM Bachman, JG eds. Monitoring the Future Series 1996 National Institute on Drug Abuse, US Department of Health and Human Services. Rockville, MD: NIH Publication No. 96-4139
- Sargent, JD, Dalton, JD, Beach, ML, et al Viewing tobacco use in movies: does it shape attitudes that mediate adolescent smoking? Am J Prev Med 2002;22,137-145[CrossRef][ISI][Medline]
- Everett, SA, Schnuth, RL, Tribble, JL Tobacco and alcohol use in top-grossing American films. J Community Health 1998;4,317-324
- Tickle, JJ, Sargent, JD, Dalton, MA, et al Favourite movie stars, their tobacco use in contemporary movies, and its association with adolescent smoking. Tob Control 2001;10,16-22[Abstract/Free Full Text]
- US Federal Trade Commission.. Staff report on the cigarette advertising investigation. 1981 US Federal Trade Commission. Washington, DC:
- Escamilla, G, Cradock, AL, Kawachi, I Women and smoking in Hollywood movies: a content analysis. Am J Public Health 2000;90,412-414[Abstract/Free Full Text]
- Roe, K School achievement, self-esteem and adolescent’s video use. Levy, MR eds. The VCR age: home video and mass communication 1989 Sage Publications. Beverly Hills, CA:
This article has been cited by other articles:
|
|
|
|
 
S. A. Glantz, J. Polansky, K. Omidvari, K. Lessnau, C. Mason, and A. Weinacker
Smoking in movies.
Chest,
February 1, 2006;
129(2):
495 - 496.
[Full Text]
[PDF]
|
|
|
|
|
|
|
A. Charlesworth and S. A. Glantz
Smoking in the Movies Increases Adolescent Smoking: A Review
Pediatrics,
December 1, 2005;
116(6):
1516 - 1528.
[Abstract]
[Full Text]
[PDF]
|
|
|
eLetters:
Read all eLetters
- Response to: Smoking in Contemporary American Cinema
- Madeline A. Dalton, Ph.D., et al.
- Chest Online, 1 Sep 2005
[Full text]
TOP
Abstract
Introduction
