US Media Coverage of Tobacco Industry Corporate Social Responsibility Initiatives

Abstract

Media coverage of tobacco industry corporate social responsibility (CSR) initiatives represents a competitive field where tobacco control advocates and the tobacco industry vie to shape public and policymaker understandings about tobacco control and the industry. Through a content analysis of 649 US news items, we examined US media coverage of tobacco industry CSR and identified characteristics of media items associated with positive coverage. Most coverage appeared in local newspapers, and CSR initiatives unrelated to tobacco, with non-controversial beneficiaries, were most commonly mentioned. Coverage was largely positive. Tobacco control advocates were infrequently cited as sources and rarely authored opinion pieces; however, when their voices were included, coverage was less likely to have a positive slant. Media items published in the South, home to several tobacco company headquarters, were more likely than those published in the West to have a positive slant. The absence of tobacco control advocates from media coverage represents a missed opportunity to influence opinion regarding the negative public health implications of tobacco industry CSR. Countering the media narrative of virtuous companies doing good deeds could be particularly beneficial in the South, where the burdens of tobacco-caused disease are greatest, and coverage of tobacco companies more positive.

Tobacco control is highly newsworthy in the US,[1] and the media play key roles in advancing the tobacco control agenda. By choosing what issues to cover, the media help illuminate tobacco issues for the public and policymakers, a process known as agenda setting.[2] Agenda setting communicates the relative importance of various issues based on the amount of media attention they attract.[3] It can increase public discourse about an issue and increase the likelihood of a policy response.[4] Media also frame news, defining issues to convey a certain causal interpretation or problem definition and an implied solution.[5, 6, p. 52] The volume of news coverage of tobacco issues and how such issues are framed are associated with government action on tobacco.[7–9] Regional variation in news coverage may play a role in regional differences in public and policymaker opinion and action.[10, 11]

Because of the media’s role in influencing policy, tobacco control advocates and the tobacco industry vie to shape media coverage. Advocates employ media advocacy techniques, including framing strategies, to promote coverage that generates public support for tobacco control;[12–14] likewise, the tobacco industry seeks to generate favorable media coverage by, for example, recruiting journalists to promote its public relations messages.[15–19] Studies have found news coverage to be favorable or neutral toward many tobacco control policies, including smokefree laws and voluntary smokefree policies, point-of-sale restrictions, and tobacco taxes;[14, 20–25] however, less is known about media coverage of the tobacco industry, including its efforts at image improvement via corporate social responsibility (CSR) initiatives.[26–28] Researchers have noted that, if successful at image enhancement, CSR initiatives ultimately thwart public health progress by facilitating tobacco industry access to policymakers and influence over tobacco policymaking,[28–30] and creating new industry allies.[31, 32]

We sought to learn whether tobacco company CSR efforts received US media coverage and, if so, how such efforts were covered (positively or critically). We also explored the content of media coverage and identified characteristics of media items that were associated with positive coverage.

METHODS

We searched two online media databases (Lexis Nexis and Access World News) for media items published from 1998 to 2014 concerning tobacco company CSR. (We chose 1998 as our starting point because it was the year Philip Morris added CSR to its toolkit of strategies.)[33] The two databases covered 3,246 US news sources, including local, college and university, and national newspapers; magazines; newswires; web-only news sources; and transcripts of major network and cable news broadcasts (e.g., CBS, NBC, CNN and MSNBC) and National Public Radio news broadcasts. We used a variety of search terms to locate items, starting with general terms such as “philanthropy” or “grant,” combined with names of major US tobacco companies (e.g., Philip Morris/Altria, RJ Reynolds/Reynolds American, Lorillard) or the phrase “tobacco industry.” We reviewed retrieved items to identify more specific search terms, such as names of particular CSR projects (“Philip Morris in the 21^st Century”) or organizations partnering with tobacco companies. We stopped searching once no new items were found. We included items with nearly-identical content that were published in multiple news outlets in order to reflect the reach of media coverage.

We identified 2,131 relevant media items, and randomly selected 30.5% for additional coding (n=649). To avoid the possibility of selecting no items from a year with fewer media items, we stratified the items by year and randomly selected 30.5% of the items from each year. We coded media items through a collaborative, iterative process. Using an adaptation of a codebook from an earlier project that examined African media coverage of tobacco industry CSR,[34] two coders (including the first author) created an initial coding sheet and piloted it on 25 media items. After discussion, we refined and edited the coding sheet and drafted coding instructions. Next, the coders independently coded a randomly selected overlapping set of 20% (n=129) of the sampled items. We assessed inter-coder reliability of the overlapping sample using Gwet’s AC1 statistic, an improvement on the kappa (κ) statistic, which becomes unreliable without sufficient variety in coding.[35] Like the κ statistic, AC1 has a value of 0–1, and is interpreted similarly. Average inter-coder reliability for all non-static variables was 0.94.

After confirming inter-coder reliability with the overlapping sample,[35] the coders independently coded the remaining media items. We also recoded items coded early in the process to be consistent with the final codebook. We coded story characteristics (i.e., news source, story type, date, photo, page number, word length, etc.) and content. Allowing for multiple mentions, we coded for the presence (“yes”) or absence (“no”) of content; for the purposes of this paper, we focused our analysis on story characteristics and on content related to the sources quoted in media items, tobacco control themes (content that reflected support for tobacco control, such as references to tobacco-caused death and disease), tobacco interest themes (content that reflected support for the tobacco industry or tobacco industry CSR, such as a reference to tobacco industry generosity as critical to nonprofits), and overall impression (slant) of tobacco industry CSR (positive, negative, or neutral). In determining slant, we assessed support or not for tobacco company CSR as reflected in each news item as a whole; thus, for example, an item that included one critical statement and four statements of support was coded as positive.

We conducted a descriptive analysis of story characteristics and content of media items. To assess the association between type of CSR program mentioned in media items and the likelihood of a positive (versus negative or neutral) slant towards tobacco industry CSR, we performed chi-square tests. To assess the association between other story characteristics and the likelihood of a positive slant, we performed a multivariate logistic regression. We examined the type of CSR program in a separate model due to collinearity with other story characteristics.

Our methods have limitations. Although they covered a large number of national and local newspapers, the news databases are not comprehensive. Moreover, our search terms may not have been exhaustive; thus, we may not have identified and drawn a sample from the population of all relevant media items. We chose to include in our sampling frame nearly identical content published by different news sources, reflecting the breadth or reach of news coverage; as a result, some similar content was coded more than once. Therefore, our findings are representative of all coverage that appeared, not unique stories.

RESULTS

Characteristics of media items and trends over time

Most of the 649 sampled media items were published by local newspapers (64.1%) or wire services (28.5%) (table 1). News stories or features and press releases comprised the majority of items (50.8% and 23.7%, respectively); there were fewer op-eds and letters to the editor (7.6% and 1.7%, respectively) (table 1). Most press releases were issued by organizations announcing receipt of tobacco company funding (71.6%) or by tobacco companies publicizing their financial support for charitable organizations or events (18.2%). Tobacco control advocates or organizations were less well-represented as authors of press releases (11/154, 7.1%); they also wrote only a handful of op-eds and letters to the editor (8/60, 13.3%). Item length ranged from 20–4,503 words, with a median of 513 words.

Table 1.

Characteristics of US media items on tobacco industry corporate social responsibility programs, 1998–2014 (n=649, except where noted).

Variable	N	%
Publication type
Local newspaper	416	64.1
National newspaper	28	4.3
News wire	185	28.5
Web-based	11	1.7
Magazine	7	1.1
Radio	2	0.3
Story type
News/feature	330	50.8
Press release	154	23.7
Blurb/brief*	105	16.2
Editorial/op-ed/column	49	7.6
Letter to the editor	11	1.7
Geographic region
West	62	9.6
Midwest	60	9.2
Northeast	43	6.6
South	251	38.7
National	233	35.9
Prominence (newspapers only, excluding missing data)
Front page	26/351	7.4
Front section	66/344	19.2
Photo	123/444	27.7
Type of CSR program mentioned
Food aid	143	22.0
Education	139	21.4
Arts funding	128	19.7
Youth smoking prevention	90	13.9
Minority organizations/issues	78	12.0
Youth-focused organization/issues	70	10.8
Philanthropy/CSR in general	70	10.8
Domestic violence/women’s issues	69	10.6
Environmental	67	10.3
Partnerships w/government	65	10.0
Disaster relief	61	9.4
Employee volunteering	54	8.3
Scientific research	44	6.8
CSR-related award	27	4.2
Smoking cessation	8	1.2
Disease prevention	5	0.8
Illicit trade	5	0.8
Child labor	3	0.5
Harm reduction/e-cigarettes	1	0.2
Other	27	4.2
Tobacco company
Philip Morris/Altria**	498	76.7
RJ Reynolds/Reynolds American	95	14.6
Lorillard	45	6.9
Brown and Williamson	14	2.2
Santa Fe Natural Tobacco Company	18	2.8
Smokeless tobacco company/other	6	0.9
Tobacco industry in general	13	2.0
Sources or evidence cited
Statement from beneficiaries of the CSR program	320	49.3
Statement from tobacco industry representative	193	29.7
Statement from tobacco control or public health representative	118	18.2
Statement from a government official	60	9.2

^*Brief announcement, often included in summaries of current events

^**Includes instances (n=31) when CSR programs sponsored by subsidiaries Kraft or Miller were also mentioned

News items on tobacco industry CSR appeared in media throughout the US, although the majority appeared in southern (38.7%) or national (35.9%) news media (national newspapers, National Public Radio, magazines, news wires, and websites) (table 1). Two states accounted for the majority of coverage in the South -- North Carolina (37.8%) and Virginia (37.5%). These states are home to several tobacco company headquarters (Lorillard, Brown and Williamson, Philip Morris USA and its parent company, Philip Morris Companies (renamed Altria in 2003), and RJ Reynolds and its parent company, Reynolds American Inc.).

In newspapers, issues considered editorially important are likely to be given greater prominence – placed on the front page, front page of a section, or accompanied by a photograph.[36] In our study, among newspaper items, 7.4% appeared on the front page, 19.2% on the first page of a section (other than the first/front section), and 27.7% had accompanying photos (table 1). Among front page newspaper articles, the majority (17/26) were published in the South; only 1 national newspaper devoted a front page article to the topic. Similarly, most news articles appearing on a section’s first page were published in the South (35/66), and fewer were published in national newspapers (5/66).

Media items concerned a variety of tobacco industry-sponsored CSR programs, with food aid (22.0%), educational programs (21.4%), and arts funding (19.7%) together receiving over half of the coverage (table 1). The majority of media items concerned CSR programs sponsored by Philip Morris/Altria (76.7%) (table 1).

The volume of news coverage of tobacco industry CSR programs varied from 1998–2014 (figure 1), ranging from 21–60 media items annually. The surge in coverage from 1998 to 1999 and 2000 may be explained by Philip Morris USA’s “Philip Morris in the 21^st Century” image improvement campaign, which featured corporate advertising focused on the company’s contributions to organizations focused on domestic violence, hunger, and disaster relief.[28, 32] In 1999, 95% of media items mentioned Philip Morris, with most coverage devoted to its involvement in food aid (56.7%) and domestic violence programs (30.0%); coverage in 2000 followed a similar pattern. Later years that saw increases in news coverage (e.g., 2007 and 2012) also focused largely on Philip Morris or Altria (e.g., in 2012, 92% of items mentioned Philip Morris and 64% mentioned Altria); however, different CSR programs received the most media attention. For example, in 2007, media items most often mentioned tobacco company funding for research (33.3%), the arts (25.5%), and government partnerships (21.6%).

Figure 1.

Number of US media items concerning tobacco industry CSR, per year, 1998–2014

Sources

Media items frequently included statements from those who benefited from CSR programs (49.3%) (table 2). For example, in a November 15, 2005 article in the Contra Costa Times, a spokesperson from the Berkeley Repertory Theater noted that “without the early and significant commitment from Altria [a sponsor of the production], this show would never have happened.”[37] Tobacco industry representatives were quoted in nearly one-third of media items (29.7%) (table 2). In an October 11, 1999 press release distributed by Business Wire, for example, Ron Milstein, Director of Lorillard’s Youth Smoking Prevention Program, asserted that Lorillard’s program was “aggressive, highly creative, and effective in getting the message to kids that they should not smoke.”[38] Statements from public health or tobacco control organizations critical of the tobacco industry or tobacco industry CSR were less common (18.2%), as were statements from government officials, including legislators and representatives of government agencies (9.2%) (table 2). When commenting specifically on tobacco industry CSR (42/60 comments), they most often expressed support for or mixed feelings about it (27/42, 64.3%) rather than opposition (15/42, 35.7%). For example, in a December 17, 2004 article in the US Federal News newswire, Representative Marty Meehan (Democrat; Massachusetts) commented favorably on Philip Morris USA’s $350,000 grant to the Lawrence, Massachusetts Boys and Girls Club, noting that “This funding will allow the [organization] to educate children at a young age about the dangers and health problems that are linked to smoking.”[39]

Table 2.

Content of US media items concerning tobacco industry corporate social responsibility programs, 1998–2014 (n=649, except where noted)^*

Content	Total	%	Example
Sources quoted
CSR initiative beneficiary	320	49.3	“‘We are grateful for Philip Morris’s continued leadership in the national fight against hunger,’ said Connie Benton Wolfe, executive director of the National Meals on Wheels Foundation. ‘Their generous support will literally put more than a million meals on the tables of thousands of frail elderly people across the United States.’”[57]
Tobacco industry representative	193	29.7	“‘The ad campaign is about telling the story of who Philip Morris is,’ says Karen Brosius, director of corporate affairs for Philip Morris Cos. ‘It shows what we believe in and what our values are.’”[58]
Public health or tobacco control organization representative	118	18.2	“‘We urge the National School Boards Association to immediately end its harmful partnership with R.J. Reynolds to promote the company’s sham ‘youth tobacco prevention’ program,’ said Matthew L. Myers, President of the Campaign for Tobacco-Free Kids.”[59]
Government official	60	9.2	“U.S. Senator Richard Blumenthal (D-Conn.) today called on the National School Boards Association to abandon its planned partnership with the R.J. Reynolds Tobacco Company to promote Right Decisions Right Now, a supposed youth tobacco prevention program.”[59]
Tobacco control themes
Tobacco-caused disease or death	130	20.0	“Smoking as an adolescent can stunt growth, stain teeth, and cause premature wrinkles, health officials say, and smokeless tobacco can cause mouth, esophagus and pancreas cancers. Studies have estimated that every cigarette smoked shaves 11 minutes off your life.”[40]
CSR as image improvement	109	16.8	“This is really more of an image campaign for Philip Morris. This is not Philip Morris suddenly deciding that it doesn’t want kids to smoke. Philip Morris knows that if kids don’t smoke, their new customers dwindle. They need to attract kids.”[60]
Policies to regulate industry	109	16.8	“In 2009, the Tobacco Control Act became law, giving the FDA the authority to regulate tobacco products to protect public health, including manufacturing, distribution and marketing.”[40]
Negative impacts of tobacco or tobacco industry other than health	99	15.3	“Litter experts say cigarettes are, by far, the country’s most tossed object. … In addition to being unsightly, [cigarette butts] can be dangerous.”[61]
Youth smoking	80	12.3	“The hundred million dollars will fund research into youth smoking and develop school-and community-based programs aimed at keeping kids from starting to smoke. Tobacco control advocates are dubious about the company’s motives.”[60]
Advertising/targeting youth	62	9.6	“The issue of marketing to youth – which cigarette makers vehemently deny – has long been at the heart of the smoking debate.”[62]
Mentioned person/group that refused tobacco industry funds	50	7.7	“A spokesman for the American Lung Association said several public universities, including Harvard and Johns Hopkins, have voted to turn down tobacco research money.”[63]
Tobacco’s economic toll	18	2.8	“In San Francisco, Mayor Gavin Newson…is seeking a 33-cents-a-pack tax to cover the $11 million that the city spends annually to remove cigarette litter.”[64]
Tobacco interest themes
Tobacco industry generosity as critical to nonprofits	37	5.7	“Philip Morris USA is pulling out of United Way, taking with it … $2 million. That amount represented almost 9% of the $22.6 million raised by United Way, which has had to make deep cuts for two years in its grants to local nonprofit organizations.”[65]
Tobacco industry as source of jobs	28	4.3	“‘If Reynolds was gone, Winston-Salem would die. They put a lot of money into Winston-Salem and a lot of jobs into Winston-Salem.’”[66]
Informed choice/adult consumers	12	1.8	“‘Philip Morris is guided by a mission to be the most responsible marketer of consumer products intended for adults.’”[67]
Legal/legitimate business	9	1.4	“Winston-Salem residents…see tobacco as a legal product that some people choose to use.”[66]
Taxes paid by company as economic benefit	5	0.8	“While productivity is important to Altria, Berlind says corporate responsibility goes beyond paying property taxes and keeping people on the payroll.”[68]
Tobacco farming as lucrative crop	5	0.8	“‘If they were to go under, [following the $145 billion verdict in Florida’s class action suit] we’re [Richmond VA] going to have major unemployment here’…noting the potential impact on factory workers and the state’s 8,000 tobacco farmers.”[69]
Overall impression
Positive	501	77.2	“10 employees of Philip Morris USA delivered meals to the homebound. … Philip Morris also lifted campaign spirits through the performance of its company choir.”[70]
Negative	84	12.9	“Regarding your column expressing disgust with recent Philip Morris do-gooder ads, the one I like best is how they stopped bottling beer so they could bottle water instead for flood victims. They should stop one of their aerosol plants and bottle oxygen for the emphysema ward victims, as well as those surviving with only part of one lung after cancer surgery.”[71]
Neutral/mixed	64	9.9	“The American Lung Association and other smoking opponents called on California community groups to reject donations from Philip Morris Inc., accusing the tobacco company of using charity to deceive the public and curry favor with politicians … Philip Morris … spokeswoman Peggy Roberts … chastised the groups for asking organizations to refuse Philip Morris’s charity when they often already have so little. Requests for assistance far exceed what the company doles out, Roberts said.”[72]

^*News items were coded for multiple responses in each category; the percentages reported in each section reflect the percent of items coded as “yes.”

Tobacco control vs. tobacco interest themes

Although approximately one-quarter of media items (27.7%) offered no information beyond describing a CSR project and the sponsoring tobacco company, the remainder included additional information that we coded under the categories “tobacco control” versus “tobacco interest” themes. The most common tobacco control theme referenced tobacco-caused disease, death, or addiction, mentioned in 20.0% of all media items (table 2). For example, a February 4, 2014 CNN newswire contrasting a tobacco company youth smoking prevention campaign with a Food and Drug Administration (FDA) youth-focused media campaign noted that the latter featured former smoker Terrie Hall, “who had her larynx … removed as a result of the ravages of oral and throat cancers.”[40] Other tobacco control themes included the non-health-related negative impacts of the tobacco industry (such as environmental degradation) (15.3%), negative references to CSR as an attempt by tobacco companies to improve their public image (16.8%), and discussion of public policies, such as FDA tobacco product regulation, aimed at preventing tobacco use or regulating the tobacco industry (16.8%) (table 2). Media items more commonly mentioned support for such policies (74/109, 67.9%), rather than opposition.

Tobacco interest themes were mentioned less often than tobacco control themes. Indeed, only 12.0% of items had any tobacco interest theme, vs 31.6% that had any tobacco control theme. The most common tobacco interest theme was a reference to tobacco industry philanthropy as critical to the survival of non-profit organizations (5.7%) (table 2).

Slant of media items

The majority of all media items conveyed a positive impression of tobacco industry CSR initiatives (77.2%) (table 2). The type of CSR program receiving coverage was associated with slant (table 3). Items mentioning education or arts funding, Philip Morris’s PM21 program, and tobacco company partnerships with local governments or volunteer programs were significantly more likely to have a positive slant than items not mentioning such programs (table 3). Mention of CSR programs related to youth smoking prevention and other youth issues or disease prevention (including smoking cessation, health-related research, and harm reduction), however, decreased the likelihood of a positive slant (table 3).

Table 3.

Odds ratios for the association of type of CSR program mentioned in media items with positive slant towards tobacco industry CSR, 1998–2014.

Type of CSR program mentioned	ORcrude	95% CI
Education or arts funding	2.04	1.35–3.07
Youth smoking prevention and other youth issues	.37	0.25–0.55
PM21 (domestic violence, disaster relief, food aid)	2.51	1.60–3.93
Environment/supply chain (environmental issues, child labor, illicit trade)	1.33	0.72–2.45
Disease prevention (smoking cessation, health- related research, harm reduction, disease prevention)	.14	0.08–0.25
Partnerships with government/tobacco company employee volunteering	3.15	1.64–6.04
Minority issues	1.07	0.60–1.89
General philanthropy/CSR award	.75	0.45–1.23

Story type, region, and sources cited were associated with the slant of news items, even when controlling for other items that also predicted slant (table 4). Compared to opinion pieces (editorials, op-eds, columns, and letters to the editor), all other story types were significantly more likely to have a positive than a negative or neutral slant. Publication in a Southern versus a Western state was also associated with a positive slant. Media items that included statements from beneficiaries of tobacco industry CSR programs were more likely to have a positive slant than those without such statements; however, media items that contained all other types of sources, including tobacco industry representatives, were less likely to have a positive slant.

Table 4.

Adjusted odds ratios for the association of selected article characteristics with positive slant towards tobacco industry CSR among news items, 1998–2014.

Variable	ORadj	95% CI
Story type
Editorial/op-ed/column/Letter to the editor	ref
News/feature	2.77	1.19–6.45
Brief/blurb	13.67	3.22–58.00
Press release	20.64	6.22–68.52
Geographic region
West	ref
Midwest	1.39	0.49–3.94
South	4.24	1.72–10.48
Northeast	1.82	0.52–6.41
National	0.87	0.33–2.23
Prominence
Photo	1.34	0.62–2.89
Sources and evidence
Statement from beneficiaries of the CSR program	2.63	1.42–4.87
Statement from tobacco industry representative	0.53	0.29–0.95
Statement from tobacco control or public health representative	0.28	0.14–0.54
Statement from a government official	0.33	0.15–0.77

DISCUSSION

During the 17-yearperiod of our study, tobacco industry CSR initiatives garnered media attention in the US, particularly at the local level. Media in the South, the region that is home to several tobacco company headquarters, were also particularly likely to cover this issue, a reflection of the key role that geographic proximity plays in attracting media attention.[41] Moreover, among newspapers, tobacco industry CSR initiatives were sometimes considered newsworthy enough to garner prominent coverage, which enhances the perceived importance of an issue and increases the size of the reading audience.[41, 42] Prominent coverage indicated that such initiatives were an important story for local communities, perhaps because disadvantaged members of the community stood to benefit.

Among tobacco companies, Philip Morris received the most coverage, likely reflecting its status as the largest US tobacco company, with the most sophisticated and sustained CSR initiatives.[28, 32, 33, 43, 44] Two of the three components of its PM21 campaign (food aid and domestic violence programs) received the majority of news coverage in the two years after the campaign was launched, indicating that the company was successful in generating free media coverage to supplement its own PM21-related media expenditures. Indeed, the monetary value of such coverage may have exceeded the $150 million Philip Morris had spent (as of 2001) to publicize its PM21-related donations of $115 million.[45]

Across all tobacco companies, CSR initiatives receiving the most coverage -- food aid, education projects, and arts funding – were those having no connection to tobacco, with appealing, non-controversial beneficiaries likely to generate public sympathy: students, the elderly, and arts organizations. However, it is unknown if the dominance of these CSR initiatives in media coverage reflected their popularity among tobacco companies, or media preference for covering certain CSR topics over others. Local media, for example, may prefer to report on these types of programs because they have more local impact than programs focusing on illicit trade or scientific research.

While the majority of media coverage took the form of news stories, nearly one-quarter consisted of press releases. This finding is consistent with studies showing that American media rely heavily on press releases.[46, 47] Tobacco control advocates were rarely authors of press releases; moreover, they rarely authored opinion pieces. The editorial section of newspapers is the most commonly read section after the front page, and draws legislators’ attention[48, 49]; thus, the absence of tobacco control advocates from these pages indicates a missed opportunity to influence public and policymaker opinion.

Tobacco control advocates were also less likely to be represented as sources in news items compared to tobacco company spokespersons and beneficiaries of CSR initiatives. Journalists may not regard tobacco control advocates as an obvious choice when seeking comment on tobacco company CSR, particularly in the local communities that stand to benefit from tobacco company largesse. Yet even without their voices, media items that included any type of tobacco control or tobacco interest theme were more likely to include tobacco control themes. This suggests that the news media may be sympathetic to such themes, if not to critiques of tobacco industry CSR. However, this failed to blunt the overwhelmingly positive, or pro-tobacco industry slant of news items.

Not every CSR initiative was equally likely to receive positive coverage. Tobacco-company sponsored initiatives that research has shown are ineffective or harmful -- youth smoking prevention, smoking cessation, harm reduction, and health-related research –[50–53] were among those less likely to be associated with a positive slant. This indicates that some degree of concern or controversy about these programs registered in media accounts, and underscores the importance of framing seemingly uncontroversial CSR programs, such as education and arts funding, in a similar manner, highlighting their potential negative impacts.

Other characteristics of news items associated with the slant of news coverage also highlighted the importance of media advocacy for tobacco control. For example, press releases, which reflect the values of their organizational authors,[47] were more likely to have a positive slant than opinion pieces. Nearly all authors of press releases in our study were either recipients of tobacco industry funds or industry representatives; thus, increasing the number of press releases issued by tobacco control advocates could serve as a potential counterweight to pro-industry perspectives. Our finding that statements from tobacco control advocates were less likely to be associated with a positive slant towards tobacco industry CSR offered further evidence of the potential impact of including tobacco control voices in media coverage of this topic.

The absence of tobacco control perspectives was seen most acutely in Southern media. Media items published in the South were more likely than those published in the West to have a positive slant. Southern media in our sample were comprised largely of local newspapers from cities that were home to tobacco company headquarters; as others have noted, local media are generally “supportive of and dependent on local institutions,”[54, p. 77] including, presumably, tobacco companies. This dependency, however, comes at a cost – the South has the highest rate of tobacco-related invasive cancer cases and deaths of any US region.[55] Local corporate philanthropy is unlikely to offset the costs associated with early deaths from tobacco-caused disease.

CONCLUSION

Despite some critical coverage, tobacco industry CSR was largely portrayed by the media as a case of virtuous companies doing good deeds for deserving members of local communities; the deadliness of the companies’ products and the negative public health implications of their apparent largesse were, for the most part, absent from media narratives. Countering this narrative will require media advocacy, a tactic already employed by tobacco control and public health practitioners. One challenge in formulating a persuasive counter narrative is the complexity of the message. The harm of CSR is indirect: an image enhanced via CSR makes it easier for policymakers to engage with the tobacco industry, which, in turn, better enables the industry’s to promote its interests, thus prolonging the tobacco epidemic and its devastating impact on individuals and communities. By contrast, the prevailing narrative regarding tobacco industry CSR is simpler, focusing on worthy organizations and individuals receiving tangible benefits from generous tobacco companies. One solution is to highlight the paltry sums tobacco companies devote to CSR compared to the public costs of tobacco-caused disease. A similar approach has been successful in the past: the revelation that Philip Morris spent more money advertising its PM21 campaign than investing in PM21-sponsored charities generated critical news coverage with a simple message that it was inexcusable to “brag more than you give.”[45]

News media remain important channels for conveying information and promoting public discourse. Controversy often generates additional media attention to issues.[56] Media coverage of tobacco company CSR has, to date, been largely uncritical, contributing to the normalization of tobacco companies as ordinary businesses. Problematizing this normalization through advocacy may help promote continued discussion of the need for additional tobacco control measures.