Determinants of First-Time Cancer Examinations in a Rural Community: A Mechanism for Behavior Change

Abstract

Objectives. After conducting a media campaign focusing on the importance of oral and pharyngeal cancer (OPC) examinations, we assessed mechanisms of behavior change among individuals receiving an OPC examination for the first time.

Methods. We used data from 2 waves of telephone surveys of individuals residing in 36 rural census tracts in northern Florida (n = 806). The second survey occurred after our media intervention. We developed media messages and modes of message delivery with community members via focus groups and intercept interviews. We performed a mediation analysis to examine behavior change mechanisms.

Results. Greater exposure to media messages corresponded with heightened concern about OPC. Heightened concern, in turn, predicted receipt of a first-time OPC examination, but only among men.

Conclusions. We extended earlier studies by measuring an outcome behavior (receipt of an OPC examination) and demonstrating that the putative mechanism of action (concern about the disease) explained the link between a media intervention and engaging in the target behavior. Improving the quality of media campaigns by engaging community stakeholders in selecting messages and delivery methods is an effective strategy in building public health interventions aimed at changing behaviors.

Racial and gender disparities in disease-specific mortality from oral and pharyngeal cancer (OPC) are not new,^1–5 but recent data show that these inequities are widening.⁶ According to data from the Surveillance, Epidemiology, and End Results Program, Black men are most at risk for dying from OPC, with a 36.0% 5-year relative survival rate (as compared with rates of 66.6% among White men, 68.0% among White women, and 59.7% among Black women).⁷ Moreover, a downward trend in survival rate has been observed among Black men, with the rate decreasing from 48.5% to 36.0% between 2004 and 2006. Although other explanations are plausible,¹ disparities in OPC survival are largely attributed to inequities in stage of diagnosis.⁸

Late-stage diagnosis is associated with low survival rates.⁸ Blacks are more likely than Whites to be diagnosed at the regional and distant stages,^6,9 when the outcome of the disease is most devastating and costly.^10–12 One explanation for this late-stage diagnosis may be lower rates of OPC examinations (sometimes called OPC screenings) among those who self-identify as Black.^13,14 OPC examination rates are low for a number of reasons, including a lack of overall public awareness of risk factors, signs, and symptoms.¹⁵ These low examination rates are believed to place certain groups, such as men, Blacks, and rural people of lower socioeconomic status, at increased risk for late-stage diagnosis of OPC.

Overall, cancer mortality rates are higher among rural residents than urban residents, and rural cancer survivors are more likely than urban cancer survivors to report fair or poor health and health-related unemployment.¹⁶ Thus, it is important to aid rural residents in undergoing examinations for cancer.^16,17 Limited access to health care, especially in rural areas, is one reason why health inequities exist, but access to information and knowledge is also a key factor in health inequalities.^14,18

To date, there is little evidence of the effectiveness of population-wide public health campaigns,^19,20 but there is growing evidence that these campaigns should be carefully and sensitively designed for specific groups and geographic regions.^18,21–23 It seems reasonable that when an individual deems health information relevant, he or she is more likely to pay attention to the message. Viswanath and Emmons have suggested that “framing” messages to target a specific population’s concerns increases attention and salience, leading to behavior change and ultimately improved outcomes.¹⁸ Appropriately reaching different sectors of the public by framing relevant messages according to their health literacy level is a crucial first step in improving overall population health.^24,25

To address Black–White disparities in OPC stage of diagnosis, we propose that messages must be designed to garner the attention of Blacks, must be salient to Blacks, and must contain a call to action. Using these salient messages to increase OPC examinations (and to promote earlier examinations) may have a profound effect on reducing disparities in OPC stage of diagnosis.

We designed a small media campaign targeting rural residents (particularly Black residents) to promote OPC examinations. We used images and facts characterizing the disease and disease threat among Blacks.²⁶ The campaign consisted of posters, brochures, car magnets, and handheld fans placed in different areas of the study communities. Our rationale was that exposure to multiple message delivery modes would be more influential than exposure to a single delivery modality.

We were also interested in understanding why our messages might increase OPC examinations. In an earlier study, we found that concern about OPC partially mediated the relationship between message exposure and intention to undergo a free OPC examination.²⁶ When concern about a health behavior is high, it follows that people are more likely to engage in relevant actions (as predicted by several theories of attitude, persuasion, and behavior change^27–32) such as undergoing a cancer examination.

Because many people are not aware of OPC and especially its deadly consequences,^14,15 we predicted that our media campaign would raise concern about the disease, which would in turn lead to more first-time OPC examinations among our participants. In our earlier work, we found that there were race and gender effects related to whether participants had heard about and undergone examinations, as well as race and gender effects related to knowledge and concern about OPC.^13,15 These results led us to predict that gender and race would moderate our effects (Figure 1). We hypothesized that our use of multiple messages delivered through multiple modalities designed to be noticeable and important to Blacks³³ would lead to higher OPC concern among Blacks and, ultimately, more first-time OPC examinations among Blacks than among Whites.^18,34

FIGURE 1—

The conceptual study model.

Note. OPC = oral and pharyngeal cancer.

METHODS

Our study took place from April 2010 to February 2011 in a pair of rural Florida communities that included all or part of 6 counties. Participants in the intervention community were exposed to a small OPC media campaign, whereas participants in a wait-listed comparison community were not. We observed an unintended spread of the health promotion campaign in that participants from both communities reported similar exposure to all messages. As a result of this spreading of the campaign, we were unable to use community as a grouping variable.

We identified rural census tracts within the study counties and oversampled Blacks to ensure adequate representation. The 6 counties from which the census tracts were drawn (excluding the cities of Tallahassee and Gainesville) had an overall median household income of $37 158, as compared with $45 000 for Florida overall, placing these counties among the lowest in the state with respect to income. We limited our sampling to homes with landlines. Our rationale was that those with landlines might be less mobile and more likely to be available for recontact.³⁵ In addition, we excluded individuals younger than 25 years, who have a lower risk of OPC. We implemented a within-household respondent selection procedure to maximize participation among older men.¹³ Our findings are based on a subset of 806 participants who, at baseline, had never had an OPC examination.

Survey Procedure

A professional survey institute was employed to contact participants and ensure that they provided informed consent; the survey was administered via computer-assisted telephone interviewing methodology. Participants received $15 Wal-Mart gift cards for completing each of the 2 surveys. The baseline survey was administered from November 2009 to March 2010; the follow-up survey was conducted between March and June 2011.

Media Campaign Messages

We designed health promotion messages to appeal to rural Black residents. The campaign included 9 posters, 2 types of handheld fans, 1 type of car magnet, and a trifold brochure, all pilot tested with community members.²⁶ We used focus groups and intercept interviews to gather community feedback and refine the campaign messages and images. The messages were consistent regardless of whether they appeared on the magnet, posters, brochures, or fans. Our goal was to increase knowledge and salience of the consequences of OPC among Blacks.^26,36

The images and text on the posters included facts about OPC and emphasized the seriousness of the disease and Blacks’ susceptibility to the disease (e.g., “African American men are twice as likely to die as other men,” “What you don’t know can kill you,” and “40,000 people will learn they have mouth and throat cancer this year”). All of the images on the posters were of Black actors with the exception of one poster that included images of a racially mixed group of men (Eddie Van Halen, Sammy Davis Jr, Jim Thorpe, and Babe Ruth).

The messages on the 2 types of handheld fans were similar to the posters, and both included images of Blacks. The phrase “do it for you” in reference to getting an OPC examination was prominently displayed. The car magnet was a reproduction of a poster depicting a well-dressed Black man stating “Man, what you don’t know can kill you” and “Have an exam today.” The brochures, revised from an earlier campaign, included statements such as “In north Florida, African Americans are less likely to be diagnosed with mouth and throat cancer at an early stage when it is easier to treat.”³⁷ Of the 7 individuals shown on the brochures, 6 were Black. In 4 separate statements, the brochures encouraged the reader to have an examination or talk to a health care provider about early detection.

Figure 2 shows an image and text from the trifold brochure. In Figure 2a, a Black actor uses a bullhorn to inspire the reader to ask his or her health care provider about getting an OPC examination. This call to action is illustrated (Figure 2b) with the brochure message addressing the possible signs and symptoms of OPC and the importance of early detection.

FIGURE 2—

Sample (a) image and (b) text used in the media campaign brochure: 2 rural Florida communities, 2010–2011.

Distribution of Messages

The posters and brochures were distributed to 45 local businesses. The posters were replaced with different posters once each month over a period of 5 months. Similarly, the brochures were replenished each time the poster display was changed. The fans (in bundles of 25) were placed in 14 churches, one funeral home, and one beauty shop. Forty-five vehicles displayed the car magnet for 5 months. The research team confirmed the presence of the posters, brochures, car magnets, and fans at each site on a monthly basis.

Measures

Oral and pharyngeal cancer examinations.

The main outcome of interest was receipt of an OPC examination in the past year. This outcome was defined on the basis of 2 variables. Participants were asked whether they had ever had an OPC examination and, if so, whether it had occurred in the past year. We categorized participants who reported never having had an OPC examination at baseline into 1 of 2 groups: the first-time-examination group (participants who reported during the follow-up survey that they had received an OPC examination in the past year) and the no-examination group (participants who reported at both baseline and follow-up that they had never had an examination). Two secondary response variables consisted of whether the respondents recognized that they had had an oral cancer examination after such an examination was described and whether the examination had occurred in a medical or dental setting.

Message exposure.

At follow-up, we asked participants whether they had seen messages about OPC from 4 sources: on posters in businesses, on handheld fans, in brochures, or on the sides of cars or trucks. Then, on the basis of their answers regarding the 4 modalities, we created a single message exposure index that ranged from 0 (saw none of these 4 modes of message delivery) to 4 (saw all modes of message delivery). To ensure that this index was a valid measure of message exposure, we asked participants questions related to the OPC information presented on the posters, brochures, car magnets, and church fans (e.g., “How common is OPC among Black men?”). A linear contrast analysis revealed that exposure to more message modes corresponded with correctly recalling more information about OPC (F_1,804 = 6.91; P = .009; d = 0.52). In addition, the more modes a participant saw, the higher the likelihood that he or she would talk to others about the campaign (χ²₄ = 21.6; P < .001).

Concern.

We measured concern about OPC at baseline by asking, “How concerned are you about getting mouth or throat cancer in the future?” (1 = definitely not concerned, 4 = very concerned). At follow-up, we included 2 additional items involving the same 4-step response scale: “How concerned are you about your future health overall?” and “Thinking about the important people in your life, how concerned are they about getting mouth or throat cancer in the future?” We averaged the 3 follow-up items to form a single index (α = 0.69).

Sample characteristics.

We collected sociodemographic information, including age, gender, and race, in the baseline survey. We used 2 items to assess financial security.³⁸ The first item asked participants, “Which of these statements best describes your present financial status?” (1 = I really can’t make ends meet, 2 = I manage to get by, 3 = I have enough to manage plus some extra, 4 = money is not a problem, I can buy about whatever I want). The second item asked, “If you were faced with an unexpected $500 medical bill that was not covered by insurance, how would you best describe your situation?” (1 = not able to pay the bill, 2 = able to pay but with difficulty, 3 = able to pay comfortably). We computed a financial security score (range = 0–2, with 2 indicating the highest level of financial security) from a weighted average of these 2 items, which have been shown to be reliable predictors of health outcomes.^13,39,40 We also categorized education into 3 groups: less than high school, high school or general educational development diploma, and some college or more.

Data Analysis

We used the t test and χ² test to determine whether the examination status groups differed with respect to the sociodemographic measures at baseline. In addition, we conducted regression analyses that controlled for baseline measures (age, education, financial security, and baseline concern) to determine whether media exposure predicted follow-up concern about OPC and whether media exposure and follow-up concern predicted whether participants underwent an examination. In these analyses, we also assessed potential moderation by gender and race.

Finally, we conducted separate mediation analyses for men and women, again controlling for baseline measures, to examine whether follow-up concern mediated the relationship between media exposure and undergoing an examination and whether race was a moderator. All analyses were survey-sampling weighted to account for the complex survey design. We used SAS version 9.3 (SAS Institute Inc, Cary, NC) in performing all of our analyses.

RESULTS

There were no significant differences between groups (first-time examination or no examination) with respect to age, gender, race, community of residence, or baseline concern about OPC (Table 1). Overall, 12.9% of the sample received a first-time OPC examination. Among Black men, 11.2% received an OPC examination for the first time. Among White men, 17.1% received a first-time examination. The respective percentages for Black women and White women were 9.6% and 12.9%. As expected, the first-time-examination group reported greater financial security and more education than the no-examination group (P = .001 and P = .009, respectively).

TABLE 1—

Descriptive Statistics for the Study Population, by Examination Group: 2 Rural Florida Communities, 2010–2011

Characteristic	First-Time Examination (n = 104), Mean ±SD or No. (%)	No Examination (n = 702), Mean ±SD or No. (%)	P
Age, y	57.0 ±13.8	54.7 ±15.6	.144
Gender			.216
Male	49 (47.1)	284 (40.5)
Female	55 (52.9)	418 (59.5)
Race			.084
White	75 (72.1)	449 (64.0)
Black	29 (27.9)	253 (36.0)
Community			.667
Comparison	48 (46.2)	322 (45.9)
Intervention	56 (53.8)	380 (54.1)
Financial security score (range = 0–2)	1.21 ±0.61	1.00 ±0.53	.001
Education			.009
< high school	7 (6.7)	78 (11.1)
High school or GED diploma	20 (19.2)	222 (31.7)
≥ some college	77 (74.1)	400 (57.2)
Baseline level of concern			.981
Definitely not concerned	42 (40.4)	275 (39.3)
A little concerned	38 (36.5)	257 (36.7)
Concerned	11 (10.6)	81 (11.6)
Very concerned	13 (12.5)	87 (12.4)

Note. GED = general educational development.

Predictors of Examination Receipt and Follow-Up Concern

We conducted a regression analysis to explore the effects of message exposure and follow-up concern on getting an examination and whether the effects differed by gender and race. In the analysis, we included follow-up concern, message exposure, and their interactions with gender and race as predictors while controlling for age, education, financial security, and baseline concern. The analysis revealed a significant follow-up concern by gender interaction (χ² = 10.6; P = .001). Follow-up concern predicted receipt of an examination among men (odds ratio [OR] = 1.7; 95% confidence interval [CI] = 1.08, 2.82; P = .024) but not among women (OR = 0.70; 95% CI = 0.45, 1.10; P = .123). No other effects were statistically significant (all Ps > .1). Race did not modify the effect of follow-up concern or message exposure on receipt of an examination.

We also performed a regression analysis to assess whether message exposure predicted follow-up concern and whether gender and race moderated the effect of message exposure. We included message exposure and its interactions with gender and race as predictors while controlling for age, education, financial security, and baseline concern. The analysis showed a significant message exposure by race interaction (Β = −0.11; t = −2.3; P = .019), such that message exposure increased concern among Blacks (B = 0.21; t = 3.8; P < .001) but not among Whites (B = 0.05; t = 1.2; P = .224). No other effects were statistically significant (all Ps > .1). The effect of message exposure on follow-up concern did not differ by gender.

Mediation Model

Given that the regression analyses showed that follow-up concern predicted receipt of an examination among men but not women, we conducted mediation analyses separately for men and women. In addition, our regression results led us to examine whether race moderated any mediation effects. As in our regression analyses, we controlled for all baseline measures (baseline concern, age, education, and financial security).

Our model for men revealed a nonsignificant χ² value, indicating adequate fit (χ² = 2.01; P = .367; comparative fit index = 0.99; root mean square error of approximation = 0.01). As shown in Figure 3, we observed moderated mediation among men. Specifically, our analysis revealed a significant direct path from message exposure to examination, indicating that greater message exposure corresponded with a greater probability of undergoing an OPC examination. In addition, follow-up concern mediated the effect of message exposure on examination receipt. Greater message exposure corresponded with greater follow-up concern, and, in turn, greater follow-up concern corresponded with greater probability of undergoing a first-time OPC examination. Importantly, race moderated the effect of message exposure on follow-up concern. In response to the messages, Black men reported greater follow-up concern about OPC than did White men, although concern was increased among all men.

FIGURE 3—

Results of path models focusing on oral and pharyngeal cancer (OPC) examination behaviors by (a) men and (b) women: 2 rural Florida communities, 2010–2011.

Note. Numbers indicate Standardized Path Coefficients (SEs). Dashed lines indicate nonsignificant paths.

*P < .05; **P < .01; ***P < .001.

By contrast, the mediation model showed a poor fit among women (Figure 3). Although greater message exposure corresponded with increased follow-up concern among women (B = 0.118; P < .001), greater follow-up concern was not related to undergoing an OPC examination (P = .06). In addition, unlike the path model for men, race did not influence follow-up concern among women.

Although we controlled for age, financial security, education, and baseline concern in the mediation models, we observed several interesting relationships between these control variables and media exposure, follow-up concern, and receipt of a first-time examination. These relationships are shown in Table 2 as standardized path coefficients. Among men, more message exposure corresponded with greater follow-up concern (B = 0.231) and with a greater likelihood of undergoing a first-time OPC examination (B = 0.224). We also found that higher educational levels among men corresponded with lower follow-up concern (B = −0.123) and with a greater likelihood of undergoing a first-time OPC examination (B = 0.235).

TABLE 2—

Standardized Path Coefficients: Predictors and Covariates: 2 Rural Florida Communities, 2010–2011

	Men		Women
Predictor or Covariate	OPC Examination, B (SE)	Follow-Up Concern, B (SE)	OPC Examination, B (SE)	Follow-Up Concern, B (SE)
Follow-up concern	0.185* (0.073)	. . .	−0.094 (0.064)	. . .
Message exposure	0.224* (0.055)	0.231*** (0.057)	−0.046 (0.035)	0.168** (0.078)
Race	0.075 (0.151)	−0.043* (0.063)	−0.051 (0.057)	−0.092 (0.052)
Race by message exposure	0.026 (0.029)	−0.157* (0.067)	0.130 (0.069)	−0.069 (0.061)
Age	0.155* (0.069)	0.131** (0.047)	−0.022 (0.052)	0.052 (0.041)
Education	0.235* (0.076)	−0.123* (0.041)	0.036 (0.055)	−0.046 (0.043)
Financial security	0.106 (0.095)	−0.065 (0.061)	0.148** (0.053)	−0.188*** (0.044)
Baseline concern	−0.040 (0.070)	0.489*** (0.047)	0.074 (0.062)	0.364*** (0.042)

Note. OPC = oral and pharyngeal cancer.

*P < .05; **P < .01; ***P < .001.

DISCUSSION

The key findings of our study are that exposure to more message delivery modes corresponded with heightened OPC concern, particularly among Black participants; men reporting heightened concern were more likely to have received an OPC examination for the first time; and increased message exposure was associated with a higher probability of men undergoing an OPC examination in the past year. This latter relationship was partially mediated by increased concern about OPC.

Among men who had not had a prior OPC examination, 11.2% had an examination after our media campaign. By contrast, our previous baseline work showed that, in the absence of an intervention, 7% of Black men had undergone an OPC examination in the past year.¹³ We interpret these data to mean that our intervention was appropriately targeted toward a group of men who traditionally have disproportionately low examination rates. These findings further increase our confidence in the methods we used to develop and implement this small media campaign and show the importance of our study to the applied public health research field.

Overall, we confirmed that concern about the disease is the mechanism by which message exposure leads to receipt of a first-time OPC examination. However, in our analysis, the mechanism was valid only among certain groups of participants, probably as a result of the focus of the messages used in the campaign. As predicted, greater message exposure corresponded with greater concern about OPC. Greater concern, in turn, corresponded with undergoing a first-time examination, but only among men. Among women, concern was not related to undergoing a first-time examination.

Importantly, the effect of message exposure on concern among men (but not women) was moderated by race, such that the increase in concern was greater among Black men than among White men. Of note, Blacks reported exposure to significantly more message delivery modes (data not shown). This moderation effect is sensible given that we specifically constructed our messages to increase message salience and accessibility among Blacks and men.³¹ We placed the posters and brochures primarily in businesses serving Black clients, although we recognize that the businesses also served the broader community, including people traveling through the area. In addition, of the 27 images of people used in this campaign (on posters, brochures, handheld fans, and car magnets), 13 images were of adult Black men, 5 were of Black children, and 3 were of Black women.

It may be that cultural tailoring is especially important in persuading members of minority groups that a message (e.g., disease outcome) is “of concern” to them.^34,41 Supporting this interpretation is evidence that race is more salient for Black Americans than for White Americans. When asked to define themselves, Blacks tend to include some aspect of their race in their responses, whereas Whites do not.³⁴ Perhaps Blacks, relative to Whites, are more sensitive in general to their similarity to the images, making tailored messages more persuasive for Blacks.⁴¹ By contrast, a message that includes minority group members may be interpreted by Whites as also pertaining to Whites. In summary, members of minority groups may need a signal that a message is meant for them (through use of a face-appropriate model), whereas members of the majority group do not.⁴¹

Another vital aspect of our approach to message development is going beyond “surface structure” and addressing “deep structure,” such as the social and psychological factors that affect the health of the target community.⁴² Although it is difficult to address deep structures in a small media campaign such as ours, our messages and modalities were designed by community focus groups. It has been suggested that when deep structures are addressed in health campaigns, the messages are truly salient to the target community.⁴²

The more modes of messages Black men saw, the more concern they had about OPC, and the more likely they were to undergo a first-time OPC examination. Our previous research¹⁵ showed that many people are unaware of the seriousness of OPC and unaware that they need to get an OPC examination. By developing multiple modes of messages targeting Blacks and addressing the seriousness of the disease, we increased the likelihood that Black men became aware of the importance of undergoing an OPC examination.

According to National Cancer Institute guidelines, there is inadequate evidence to establish whether OPC examinations would result in decreases in oral cancer mortality.⁴³ It is important to note, however, that these guidelines are silent with respect to morbidity caused by OPC and that the reference is to general population screenings, without mention of OPC examinations among subpopulations with documented higher risks. Moreover, inadequate evidence is not the same as evidence against OPC examinations. Our own data^9,44 show that individuals in our targeted geographic region have poorer OPC-related outcomes than those in other regions. Thus, we feel justified in attempting to develop parameters for designing effective interventions in this region.

Most important, above and beyond the question of the value of OPC examinations, is the impact of our investigation on health promotion intervention research in general. Our work shows that with the proper level of design, a well-originated intervention can achieve a hypothesized outcome in a vulnerable population. The lessons we provide go well beyond the dependent variable of receipt of an OPC examination. They begin to point to some guidelines for engaging communities and designing interventions based on the concerns of those target audiences.

Public Health Implications

Our findings demonstrate that targeting a specific group with images of and facts about the consequences of a given disease (in this case, OPC) is an effective approach to promoting health behaviors. This is particularly true when the sociodemographic characteristics of a group suggest a low level of health literacy.^45,46 To address the issue of low health literacy, we developed the messages used in this study in conjunction with Black community stakeholders and through focus groups and intercept interviews. As such, the messages were relevant and were physically accessible to the individuals at risk. The implication for public health campaigns is that message content must be salient to members of the at-risk group such that they see the facts, believe the facts, and are motivated to take action.

Our small media campaign was developed specifically for residents of rural areas, where types and available modes of message delivery are more limited than in other areas. We demonstrated in previous research that “bus wraps” (the practice of completely or partially covering the large outer surface of transit buses with an advertising message, creating a mobile billboard) are effective in conveying OPC information in an urban setting.³⁷ However, regularly scheduled transit buses were not available within our communities, ruling out that approach in our small media campaign. We designed our modes of message delivery on the basis of what was available in these communities and input from our community stakeholders. In short, knowing the characteristics of the target group and the environment in which at-risk individuals live and work is an important precursor to designing effective public health campaigns and ensuring that materials are culturally relevant.

We found that the messages raised concern, which in turn led to individuals (primarily men) undergoing OPC examinations. In other words, we identified a mechanism through which exposure to a public health message affects receipt of a first-time cancer examination. The implications for public health messaging are clear. Concerns that may be unique to specific group should be identified, and messages should be built to address those concerns. Our data further suggest that increases in concern among members of the at-risk group can result in their taking the called-for action (i.e., undergoing an examination). Identification of this pathway, in parallel with population-level approaches, offers an avenue to develop interventions aimed at alleviating other health disparities.⁴⁵

Limitations

The findings of this study should be interpreted in the context of its limitations. First, we acknowledge that the spread of the effects of our media campaign to the comparison community was unanticipated. We do not believe, however, that the combining of the 2 communities for the purposes of our analysis weakens our conclusions. Second, concern and receipt of a first-time OPC examination were both assessed at follow-up, and thus it was impossible to determine whether greater concern at follow-up led to receipt of an examination or the reverse. Although we believe that the former is the case, we cannot rule out the latter.

A third limitation is that receipt of OPC exams was based on self-reported data. However, we used multiple items, including items focusing on the physical setting of the examination and whether it took place in the preceding year, to increase the accuracy of the information. In addition, we included a follow-up question designed to confirm or disconfirm whether participants had received an OPC examination and where it occurred. This follow-up confirmation increases our confidence in our self-reported outcomes. Finally, our use of landlines may have biased the original study group selection, as the sample may have included individuals who were more affluent and better educated than the region’s overall population.⁴⁷

Conclusions

Previous studies of the effects of media campaigns on receipt of an OPC examination have examined intentions as opposed to behaviors, have failed to show changes in behavior, or have failed to assess the mechanisms leading to such changes. We moved beyond these studies by assessing whether participants underwent an examination, exploring a potential mechanism of action, and demonstrating that the putative mechanism of action explained the link between media exposure and receipt of an examination. Our small media campaign focused on the consequences of OPC among Black men. This focus was intended to increase concern about the disease in our target audience. Our findings provide the groundwork for future interventions aimed at reducing inequities in receipt of cancer examinations.