A Content Analysis of News Coverage of Skin Cancer Prevention and Detection,1979 to 2003

Objective  To analyze newspaper coverage between 1979 and 2003 to understand howprint coverage may affect primary and secondary skin cancer prevention inthe US population.

Design  Content analysis of 921 skin cancer articles released by the AssociatedPress during the study period.

Main Outcome Measures  Amount of attention given to primary and secondary prevention practicesand to risk communication.

Results  Media attention to skin cancer has not increased since 1986. Neitherprevention (31.8% of all stories) nor detection (24.4% of all stories) receivedas much attention as treatment (47.0% of all stories). Specific sun protectionpractices were mentioned infrequently. Dermatologic detection (6.6%) or self-detection(5.5%) of skin cancer was rarely discussed. Risk communication about skincancer was suboptimal: articles rarely presented absolute and relative risk.

Conclusions  The media pay little attention to skin cancer, and, in general, storiesdo not contain important educational information. Strategies for generatingincreased media attention are discussed.

Skin cancer is the most common cancer in the United States, affectingapproximately 1 million Americans every year.¹ Fortunately,skin cancer is highly preventable and treatable. Sun exposure is a significantrisk factor for the 3 most common skin cancers (melanoma, basal cell carcinoma,and squamous cell carcinoma)^2-12;hence, the promotion of sun avoidance, use of sunscreens and sun protectiveclothing, and avoidance of indoor tanning could produce a significant decreasein the incidence of these cancers. Early detection is critical for the optimalmanagement of cutaneous cancers; if diagnosed early through screening, themost common skin cancers are highly curable by surgical removal, with a lowprobability of metastasis.¹³

Public education of primary and secondary prevention measures shouldplay a vital role in skin cancer prevention efforts. It has been proposedthat simple use of sunscreen during childhood could decrease the lifetimeincidence of nonmelanoma skin cancer by 78%.¹⁴ Importantcomponents of early detection of skin cancers include regular screening bya dermatologist and skin self-examinations. As a primary source for healthinformation,^15-23 thenews media could be a powerful tool for skin cancer prevention and detectioneducation, ultimately decreasing the incidence of skin cancer and reducingthe cost and morbidity of treatment in the United States. There is increasingevidence that news coverage affects primary and secondary preventive healthbehaviors.

A compelling event, such as a celebrity diagnosis of cancer, can generatesubstantial news coverage capable of producing temporary changes in healthbehavior. For example, screening rates increased immediately following the1974 surgical treatments for breast cancer of former First Lady Betty Fordand of Margaretta “Happy” Rockefeller, the wife of former VicePresident Nelson Rockefeller,²⁴ and women reportedstronger intentions to get a mammogram after former First Lady Nancy Reagan’sannouncement that she had breast cancer in 1987.²⁵ Similarly,colorectal cancer screening rates increased after former President RonaldReagan underwent surgery to remove an intestinal tumor.²⁶ Morerecently, news anchor Katie Couric’s public crusade to educate individualsabout colon cancer after her husband’s death, including the live broadcastof her colonoscopy on The Today Show in 2000, resultedin increased colonoscopy rates after the program aired.²⁷ Althoughmedia attention to celebrities’ experiences with disease can stimulateincreased public awareness and preventive action, coverage tends to wane inthe absence of additional news events, often corresponding with similar declinesin health behaviors.

There is also evidence that general media attention to an issue overtime can affect long-term secular trends in health behavior. News coveragehas been associated with changes in different health behaviors, includingdiscontinued use of intrauterine devices and oral contraceptives,²⁸ discontinuance of aspirin and other salicylates inchildren,²⁹ smoking cessation and initiation,³⁰ drunk driving,²¹ mammographyscreening,^16,31 cocaine use,³² binge drinking,³³ andmarijuana use.³⁴ In sum, evidence is accumulatingthat news media messages, including celebrity experiences with disease andmore general disease coverage, affect participation in prevention and detectionbehaviors.^24,26,35,36

The objectives of our study were to conduct a formative descriptiveanalysis of skin cancer news coverage and to evaluate the context within whichprimary and secondary skin cancer prevention was discussed. One importantcontextual factor was how risk was communicated. Perceptions of risk are oftenan important motivation for adopting preventive behaviors.^37,38 Althoughexperts have yet to reach a consensus about optimal reporting of risks, thereis some agreement about how mass media should communicate cancer risk, includingproperly contextualizing risk by presenting absolute and relative risk inthe same story.^39-46 Althoughnews coverage of other cancer types, particularly breast cancer, has beenstudied extensively,^{16,22,25,47-52} thereare no studies addressing news coverage of skin cancer, to our knowledge.Herein, we review the coverage of skin cancer, speculate about the implicationsof such news coverage, and suggest considerations for media advocacy efforts.

Media coverage was measured using news from the Associated Press (AP).The AP was selected because it is representative of the national news environment,including radio, television, and newspapers.^32,53,54 TheAP is used by more than 6000 broadcast stations and almost 90% of newspapersin the United States.^53-55 Althoughthe specific content of news stories may vary across media formats as a functionof time or space and other institutional normative differences, the topicsbeing covered in newspapers and television are often similar.^22,23 TheAP sets the agenda for newspapers and television, which in turn set the agendafor less traditional news venues, such as magazines or televised news magazineprograms.^32,53,54,56 Therefore,it can be inferred that the amount of attention paid to skin cancer by theAP is representative of the amount of attention paid to skin cancer by mostof the US news media.

The unit of analysis was a story that gave substantial attention toskin cancer, identified using a process of successive filtration. First, usingthe AP file in the Lexis-Nexis database, an open search term was developedto capture every story that was primarily about skin cancer. Based on formativeresearch we have been conducting, we found that search terms that refer tocancer at least 3 times retrieve 86% of all stories that are primarily aboutcancer. To be more inclusive, the open search term we used retrieved all storiesthat twice referred to skin cancer, melanoma, basal cell, or squamous cell.The open search term retrieved 1657 AP stories between 1979 and 2003. Becausewe analyzed the content of the census of AP skin cancer stories rather thana sample, our results are not estimates of the “true” values;hence, the point values do not require confidence intervals.

Articles were reviewed for their relevance, and 2 coders (J.E.S. andB.A.S.) applied the coding instrument to any article in which skin cancerwas given substantial attention. Intercoder reliability was established ona sample of 80 stories; for all variables (including whether an article wasprimarily about skin cancer), κ was at least 0.70. Of the 1657 articlesretrieved by the open search term, 921 (55.6%) paid substantial attentionto skin cancer.

All of the articles were coded as to whether they included informationabout 1 of 4 primary topics, including celebrity experience, new research,policy, and other. Articles were also coded as to the types of skin cancerdiscussed, presentation of risk information, and presentation of informationrelated to skin cancer prevention, detection, diagnosis, or treatment. Ifrisk was discussed, articles were coded for the presence or absence of relativeand absolute risk and for references to particular risk factors (eg, dysplasticnevi). If prevention was discussed, articles were coded for the presence orabsence of specific sun protection behaviors. If detection or diagnosis wasdiscussed, articles were coded for references to dermatologists, skin self-examinations,and detailed information regarding how to perform such examinations.

There was a mean ± SD of 37.0 ± 15.6stories per year primarily about skin cancer between 1979 and 2003, althoughthere was substantial annual fluctuation. Skin cancer coverage peaked in 1985(n = 70), associated primarily with 2 significant events: (1) RonaldReagan had a basal cell carcinoma removed from his nose for the second timeand (2) scientists discovered that the ozone layer (which shields UV radiationand, hence, helps protect against skin cancer) was diminishing. In contrast,other peaks in coverage (eg, in 1988 and 1997) were not attributable to isolatedevents but rather were due to the co-occurrence of multiple events duringthe same year.

Less than half of all stories were dedicated exclusively to one kindof skin cancer (41.0%), most often melanoma (30.0%) (Table 1). A much smaller percentage of stories were dedicated exclusivelyto basal cell carcinoma (6.0%), and virtually none made exclusive or explicitreference to squamous cell carcinoma (1.0%). The remaining stories that focusedexclusively on one cancer were about other forms of skin cancer, such as Kaposisarcoma. The stories that were not specifically about one type of skin cancer(56.0%) discussed skin cancer more generally (46.0%) or made reference tomultiple types of skin cancer (10.0%).

New reports about skin cancer research received the most media attentionof the possible topics (35.3%) (Table 1),with a mean±SD of 13.0±6.3 stories per year. Coverage of celebrityexperiences with skin cancer was also high (32.5%), with a mean±SDof 12.0±11.3 stories per year. Media attention was particularly highfor politicians and their families: many news stories detailed the diagnosisor treatment of Nancy Reagan (1982), Ronald Reagan (1985 and 1995), formerPresident George H. W. Bush (1986), former First Lady Barbara Bush (1990-1991),Ronald Reagan’s daughter Maureen Reagan (diagnosed as having melanomain 1996 and treated until her death in 2001), presidential candidate JohnMcCain (2000), Al Gore and Dick Cheney (when the health records of presidentialand vice presidential candidates for the 2000 election were released), andformer President Bill Clinton (2000-2001).

Thirteen percent of stories reported information about policy recommendationsor guidelines about skin cancer (Table 1).Most of these stories detailed the estimated effect of the depleting ozonelayer on skin cancer incidence. Other important policies that received mediacoverage concerned regulations about the use of indoor tanning lamps, Foodand Drug Administration guidelines about the use of sunscreen on newborns,and National Institutes of Health approval of gene therapy research for melanoma.There was a mean±SD of 4.8±5.7 stories per year about skin cancerpolicy.

The remaining 19.3% of studies (Table 1) were about different topics. Many of these articles were aboutthe dangers of the sun or the importance of skin cancer prevention. Therewere numerous stories about people with xeroderma pigmentosum, likely reflectingthe news media’s preference for the unusual given the rarity of thedisease.^57-59 Therewas a mean±SD of 7.1±4.7 skin cancer stories per year that werenot about research, celebrities, or policy.

Despite the importance of safe sun practices for reducing skin cancerincidence, primary prevention was discussed in only 31.8% of skin cancer stories,and secondary prevention (ie, detection) was discussed in only 24.4% of stories(Table 2). In contrast, treatment wasdiscussed in approximately 47.0% of all skin cancer stories. Scant attentionwas paid to prevention until 1982, when Nancy Reagan had a basal cell carcinomaremoved from her upper lip and vowed to avoid the sun in the future. Althoughprevention was discussed less frequently in stories about policy (16.0%),prevention was mentioned approximately equally in stories about celebrities(23.9%), new research (29.4%), and other topics (30.7%).

When skin cancer prevention was discussed (n = 293) (Table 2), specific sun protection practices werenot always mentioned. Approximately 11% (32/293) of the articles about preventiondid not mention any of the primary prevention behaviors (minimizing sun exposure,wearing protective clothing, using sunscreen, and avoiding indoor tanning)but rather focused on environmental risks (eg, ozone depletion).

Using sunscreen was the most frequently cited strategy of primary prevention(21.4% of all skin cancer stories and 66.4% of stories discussing prevention)(Table 2). Cited somewhat less commonlywas avoiding the sun (18.1% of all stories and 56.5% of prevention stories)and wearing protective clothing to shield the skin from the sun’s harmfulrays (12.5% of all stories and 38.4% of prevention stories). Avoidance ofindoor tanning lamps or tanning beds was mentioned in only 3.9% of all skincancer stories (12.3% of prevention stories).

Despite the role that skin cancer warning signs can play in early detection,few articles discussed skin examinations by a dermatologist or by self-examination.Only 6.6% of all stories (27.5% of screening stories) explicitly mentionedscreening by a dermatologist, and fewer mentioned skin self-examinations (5.5%of all stories and 21.5% of screening stories) (Table 2). Although passing references to dermatologic examinationsor self-examinations were uncommon, it was less likely that secondary preventionstories detailed specific educational information about warning signs forskin cancers (2.2% of all stories and 8.1% of screening stories). For example,of the 36 stories that mentioned self-detection in the context of melanoma,17 stories reviewed the ABCDs (asymmetry, border irregularity, color variegation,and diameter >6 mm) of self-examination or articulated the warning signs formelanoma. Similarly, only 15.1% (59/390) of articles about melanoma mentionedthat moles are a risk factor. Even when moles were mentioned, only 23.7% (14/59)of the articles reviewed the warning signs for melanoma.

Risk was discussed in 45.1% (415/921) of all stories. Discussions ofrisk included measures of absolute risk and relative risk equally often (60.5%and 56.7%, respectively). However, if one type of risk was mentioned in askin cancer article, then the other type was significantly less likely tobe mentioned: an article that mentioned absolute risk was 84% (odds ratio,0.17; 95% confidence interval, 0.10-0.26) less likely to mention relativerisk than an article that did not mention absolute risk.

This content analysis reveals the lack of media attention to skin cancerin general and to prevention in particular. Despite the fact that skin cancerincidence has been steadily rising in the United States, and has the fastestgrowing prevalence of all cancer types, media attention to skin cancer hasnot increased over time (Figure). Althoughskin cancer is largely preventable, and curable if detected early, primaryprevention and secondary prevention are not featured prominently within storiesabout skin cancer. There was little to no skin cancer coverage in the earlyyears of the period covered by our analysis. Although there has been an increasein media attention to skin cancer since 1980, current levels of coverage areroughly equivalent to levels from the mid 1980s. To contextualize the amountof media attention to skin cancer, we conducted a similar search for breastcancer coverage. The AP produced more stories in 1 year regarding breast cancerthan in 25 years regarding skin cancer. To further contextualize the quantityof skin cancer coverage, we determined that in 2003 the AP released 191 517stories; 30 of these were about skin cancer.

Although the effect of skin cancer news coverage is unknown, evidencefrom news coverage of preventive health behaviors and agenda-setting researchtells us that increased media attention to skin cancer should increase thelikelihood of behavioral effects; that is, more coverage is better. As a primarysource for health information, the news media could be a powerful tool forskin cancer prevention and detection education, ultimately decreasing skincancer incidence and improving prognosis in the United States. It is possiblethat the lack of media attention to skin cancer is a contributing factor tothe low rates of sun protection practices in the United States. A recent nationalsurvey of youth aged 11 to 18 years revealed that 72% reported having at least1 summer sunburn.⁶⁰ A 2003 survey by the AmericanAcademy of Dermatology reported that only 34% of people younger than 25 yearssay that they use sunscreen while outdoors.⁶¹ Tenpercent of youth and 8% of their parents reported using indoor tanning sunlampsin the past year.⁶² Few adults from the 1998National Health Interview Survey reported that they were very likely to performsun protective behaviors: 23% were very likely to wear protective clothing,27% were very likely to seek shade, and 30% were very likely to use sunscreen.⁶³ The American Academy of Dermatology’s morerecent survey revealed that adult sunscreen use rates have not increased since1998.⁶¹ In addition, the 1998 National HealthInterview Survey indicated that only 21% of adults had ever had a skin cancerexamination.⁶³

Efforts should focus on increasing the amount of skin cancer media coveragegenerally and the amount of educational information about specific componentsof primary and secondary prevention, including sun protection practices andself-detection. It has been hypothesized that media attention to breast canceris higher than attention to other cancer types because several groups (eg,special interest groups and celebrities) have created a successful mobilizationeffort.^22,64 By building suchconstituencies for skin cancer, forging lasting relationships with news producers,and taking strides to help shape the content of news coverage, a similar mobilizationeffort could be made for skin cancer. Known as media advocacy,⁶⁵ thisstrategy holds great promise for catapulting an issue onto the media’sagenda. Although media advocacy traditionally focuses on the strategic useof news media to advance a social or policy initiative,⁶⁶ weare applying the term more broadly to include any purposive effort to helpshape news coverage to positively affect health.

There are numerous obstacles to garnering and sustaining media attentionto an issue over time. There is a limited amount of time and space for healthnews; hence, skin cancer coverage must compete with other health issues forroom on the media’s agenda. Moreover, the media tend to exhibit whathas been termed the “issue-attention cycle,” whereby any givenissue tends to receive substantial media attention for a limited time only,because journalists are motivated to constantly introduce new issues in aneffort to maintain the interest of their audiences.⁶⁷ Yet,the historical success of breast cancer lobbying efforts, and more recentlyof Katie Couric’s increasing colon cancer awareness and screening,²⁷ suggests that the media can be mobilized to coverskin cancer. A manual published by the American Public Health Association⁶⁸ provides detailed strategies for media advocacy efforts.

Our content analysis revealed that almost 70% of skin cancer storieswere reports of new research or celebrity experiences with skin cancer. Researchersmust make themselves available to the news media, serving as a resource toreporters and increasing the likelihood that the content of skin cancer newsstories can be shaped by experts. In addition, part of the lobbying effortmust include enlisting celebrities to serve as spokespersons for skin cancerprevention and encouraging these celebrities to provide specific preventionmessages as sound bites when interviewed. The fact that every president andall but 1 vice president of the United States since 1980 have been diagnosedas having skin cancer suggests that prominent and powerful people might bewilling to promote the cause.

Correspondence: Jo Ellen Stryker, PhD, Departmentof Speech Communication, University of Illinois at Urbana-Champaign, 244 LincolnHall, 702 S Wright St, Urbana, IL 61801 (jstryker@uiuc.edu).

Accepted for Publication: August 3, 2004.

Previous Presentation: This study was presentedat the 53rd Annual Conference of the International Communication Association;May 26, 2003; San Diego, Calif.

Financial Disclosure: None.