Designing Skin Cancer Prevention Messages: Should We Emphasize Gains or Losses? Message Framing,Risk Type,and Prior Experience

Abstract

Purpose:

To test whether message framing (ie, gain vs. loss) and risk type (ie, health vs appearance risk) in skin cancer prevention messages interact with one’s prior experience. Design: Two experiments with a 2 (message framing: gain vs loss) × 2 (risk type: health vs appearance risk) factorial design were conducted.

Setting:

The participants were given a URL to the experiment website via e-mail. On the first page of the website, the participants were told that they would be asked to evaluate a skin cancer print public service announcement (PSA): Online experiments.

Participants:

A total of 397 individuals participated (236 for experiment 1 and 161 for experiment 2).

Apparatus:

Four versions of the skin cancer print PSAs were developed. Four PSAs were identical except for the 2 manipulated components: message framing and risk type.

Measures:

Measures were adopted from Cho and Boster (message framing), Jones and Leary and Kiene et al. (risk type), De Vries, Mesters, van’t Riet, Willems, and Reubsaet and Knight, Kirincich, Farmer, and Hood (prior experience), and Hammond, Fong, Zanna, Thrasher, and Borland and Hoffner and Ye (behavioral intent).

Analysis:

General linear models were used to test hypotheses.

Results:

Three-way interactions among message framing, risk type, and prior experience were found: When the intent of the message was to encourage sunscreen use, the effects of message framing and risk type were shown to be the exact opposite directions from when the intent was to discourage indoor/outdoor tanning.

Conclusion:

To discourage tanning among those with prior experience, messages emphasizing losses in terms of one’s health will work better. For those with no prior experience, messages emphasizing potential appearance losses will work better for discouraging tanning while messages emphasizing gains like improving appearance will do a better job in encouraging sunscreen use.

Keywords

message framing risk type prior experiences skin cancer prevention message design

Purpose

Skin cancer is one of the most common types of cancer in the United States.¹ More than 3.5 million people were diagnosed for skin cancer each year.² The incidence rates of melanoma, particularly among white women younger than age 44 have been increasing by 6.1% every year³ and became one of the most common cancers among young adults.⁴ Different from other skin cancers, melanoma grows fast and survival rates decline dramatically.⁵ Despite its danger and severity, however, many young adults have engaged in behaviors that increase the likelihood of developing skin cancer.⁶

One critical factor for developing skin cancer is ultraviolet (UV) radiation exposure through sunbathing/tanning.^7,8 Simply avoiding UV rays from the sun/indoor tanning itself could prevent more than 1 million people from developing skin cancer per year.³ However, both tanning and sunbathing are quite popular in America, and health promotion campaigns to discourage indoor tanning/excessive sunbathing have not been successful.⁹

One prominent approach to enhance health message effectiveness is the use of message framing. Message framing is known to influence people’s reactions to the given messages.¹⁰ According to prospect theory (PT), individuals’ decisions to take or avoid certain risks depend on how messages are framed.^11,12 The main tenant is that people have a tendency to take a small, but more probable gain over a large but less probable gain, but when facing a loss, people tend to take a chance for a bigger, but less probable loss to avoid a small but more probable loss.

Much effort has been placed to identify the best message framing type (ie, gain vs loss) for health promotion. However, the previous findings of framing effects have not been consistent to draw any meaningful/practical guidelines to adopt from. Some found gain-framing to be more effective in yielding desired responses (ie, increasing one’s intention to adopt a recommended preventive behavior).^13,14 Others found the opposite results in that loss-framing is more effective than gain-framing.^15
–17 Some found no significant framing effect,^18,19 increasing skepticism toward the applicability of PT in the field of health communication.²⁰

Individuals’ motivations to adopt or reject certain behaviors depend on not only how messages are framed but also types of risk (eg, physical health, appearance risk) they perceive as relevant to themselves.^21

–25 Moreover, what behavior individuals perceive as risky should depend on one’s prior experience with that targeted behavior.²⁶ Thus, it is logical to assume that message framing may interact with risk type, and this interaction may depend on individuals’ prior experience with that particular risky behavior (ie, indoor/outdoor tanning). Therefore, we tested and compared a 3-way interaction among message framing, risk type, and prior experience on individuals’ acceptance of skin cancer prevention messages between when a targeted behavior is encouraged (ie, sunscreen use) and discouraged (ie, indoor/outdoor tanning).

Message Framing in Health Communication

According to PT, people are likely to stay in certainty and avoid “risk” when facing gains while opting for risk/uncertainty when facing losses, particularly when the risk is paired with the possibility of avoiding a loss.¹² Two distinctive types of message framing were shown to influence this decision-making process: gain versus loss.¹¹ Gain-framing is defined as messages focusing on the “gains” to be attained with the adoption of the recommended behavior while loss-framing focuses on the “losses” from nonadoption of the behavior.¹²

One area where PT has been applied is health communication/promotion.²⁷ Health promotional messages can be designed to focus on either gains, by emphasizing the positive physical/social consequences of accepting the recommended behavior, or losses, by focusing on the negative physical/social consequences of not following the recommendation.²⁶ Some suggested that framing effects might differ by 2 types of outcome measures: prevention versus detection behaviors; loss-framing would be more persuasive for detection behaviors because they involve a potential risk of discovering a health problem, while gain-framing would be more effective for promoting prevention behaviors because they are associated with certain positive outcomes like a decreased risk for illness and improving health.²⁶

However, the findings have been neither consistent nor conclusive. For prevention behaviors alone, some found gain-framing more effective in yielding desired responses,^13,14 but others found the opposite results.^15
–17 Some found no significant framing effect based on meta-analysis studies.^18,19 Previous findings regarding framing effects in skin cancer prevention messages are also puzzling: some found that gain-framing is more effective in encouraging skin cancer prevention behaviors while no framing effects were found for detection behaviors.²⁸ Some found no significant framing effects, but found a significant message effect.²⁹ In other words, the intention to use sunscreen in either the gain-framing or the loss-framing was higher than that of the control group. On the other hand, some found that loss-framing was more effective for discouraging indoor tanning among young adults with prior experience with tanning.^30,31

Furthermore, some investigated how other variables such as body consciousness,²¹ motivation type,³² and level of risk³³ interacted with message framing. Hevey et al reported that gain-framing increased the intention to use sunscreen only among those with high body consciousness, compared to their counterparts.²¹ In addition, Hwang, Cho, Sands, and Jeong found 2 variables (ie, perceived effectiveness in preventing risk and perceived susceptibility to the risk of skin cancer) interacted with message framing: for those with low perceived effectiveness in averting risk and with high susceptibility to skin cancer, loss-framing worked better in increasing their intention to use sunscreen.³⁴ On the other hand, gain-framing was more effective for those with high perceived effectiveness in averting risk and with low susceptibility to skin cancer. Then, it begs the question: who would be likely to perceive high susceptibility of skin cancer while perceiving less in terms of effectiveness of averting the risk and vice versa?

Risk Type: Health Versus Appearance Risk in Skin Cancer Prevention Message Design

Another critical factor that influences individuals’ responses to health messages is risk type.²⁴ There are different types of risk one considers relevant/important in making a decision about their health behaviors, such as health risk^24,25 and appearance risk.^22,23,35 Health-risk appeal emphasizes how one would suffer from a serious health consequence (eg, disease or death) if a specified recommendation is not followed, whereas appearance-risk appeal emphasizes how people would suffer from negative consequence of nonadoptive behaviors on one’s appearance.

Addressing one’s concerns for appearance, rather than health, to intervene against the strong normative influences for tanning among young adults, was suggested and found to be more effective than using conventional health risk.^{22,23,35

–38} Mahler et al^22,23 and Williams et al³⁸ found that appearance-based intervention like the photographs/information about photo-aging (eg, wrinkles and age spots) and UV photo intervention showing negative consequences of UV exposure were more effective in increasing sun protection intentions than the control group, suggesting that appearance risk may work well for encouraging sun protection behaviors among young adults. Hevey et al²¹ tested whether message framing interacted with risky type (health vs appearance) in skin cancer prevention messages and found no 2-way interaction effect between message framing and risk type, but found a significant interaction effect between message framing and individuals’ body consciousness, suggesting individual differences may play a role in their reactions to health promotional messages.

Cafri et al found that physical attractiveness and skin aging concerns were the major motivations, highly correlated with female college students’ intentions for sun protection.³⁹ However, people’s perceptions of a seemingly “risky” behavior are heavily influenced by their prior experience²⁶ which may reflect their underlying motivations for behaviors and consequent decision-making processes about indoor/outdoor tanning. For instance, Greene and Brinn⁴⁰ reported that prior tanning bed users showed higher intention to tan and lower intention to protect skin than nontanning bed users. Tanning bed users place a strong emphasis on one’s physical appearance while exhibiting more willingness to take risks associated with indoor/outdoor tanning. Accordingly, those with prior experience of indoor/outdoor tanning are more likely to be concerned with physical attractiveness and less likely to be concerned about susceptibility of skin cancer than those with no prior experience. Therefore, in this study, we infer that for those with prior indoor/outdoor tanning experience, attaining better appearance (ie, gain-framing) or losing health (ie, loss-framing) would be a stronger motivation than potential losses of appearance or gains of health. At the same time for those who have no prior experience, avoiding negative consequences of appearance or health (ie, loss-framing of appearance risk or health risk) would be a stronger motivation not to tan in the future. Therefore, the following hypotheses and research questions were investigated.

Hypothesis 1a: For those with prior experience of indoor/outdoor tanning, their intention to avoid tanning will be greater in the gain-framing of appearance risk (Hypothesis 1b: in the loss-framing of health risk) than the loss-framing of appearance risk (Hypothesis 1b: the gain-framing of health risk).

Hypothesis 2a: For those with no prior experience of indoor/outdoor tanning, their intention to avoid tanning will be greater in the loss-framing of appearance risk (Hypothesis 2b: health risk) than the gain-framing of appearance risk (Hypothesis 2b: health risk).

Research Question 1: Are there different effects on message framing and risk type based on the type of advocated behaviors (ie, avoiding indoor/outdoor tanning vs sunscreen use)?

Method

Design and sample

This study is comprised of 2 experiments with a 2 (message framing: gain vs loss) × 2 (risk type: health vs appearance risk) factorial design. Study 1 (S1) investigated behavioral intent of avoiding indoor/outdoor sunbathing, and Study 2 (S2) measured behavioral intent of using sunscreen. A total of 397 individuals participated (236 for S1 and 161 for S2). College students were recruited from a large university. Participant’s ages ranged from 18 to 32 years (S1: M = 20.39, SD = 1.63; S2: M = 20.58, SD = 2.11). More female students (S1: 75%, n = 176; S2: 71%, n = 114) participated than male students (S1: 25%, n = 60; S2: 29%, n = 47).

Four different versions of the skin cancer print public service announcements (PSAs) were developed: 2 (message framing: gain vs loss) × 2 (risk type: health risk vs appearance risk). Four PSAs were identical except for the 2 manipulated components: message framing and risk type. Consulted by Jones and Leary’s study,³⁵ the title at the top of each ad reflected its content: “Tanning: A risk to your health” (health risk) and “Tanning: A risk to your natural beauty” (appearance risk). The 4 ads were similar in length and structure, with identical beginning and ending sentences. The loss-framed message contained negative terms such as “harm” and “ruin,” and the gain-framed message contained positive terms such as “protect” and “saving”. All 4 ads provided the guidelines to prevent skin cancer.

The health-risk message discussed the skin cancer incidence linked with tanning bed use/sunbathing and offered information related to the risk of UV rays that can cause skin cancer. Health-risk messages were framed as loss- or gain-framed: the ad with a loss-framing of health risk discusses the negative health risks associated with indoor tanning or sunbathing such as “using tanning beds or sunbathing can be like asking for skin cancer.” The ad with gain-framing of health risk discusses the positive health consequences of avoiding tanning beds or sunbathing. The appearance-risk message discusses the negative effects of using tanning beds or sunbathing on one’s appearance, especially skin damage such as wrinkles, age spots, and skin burn, and the possibility of ruining “natural beauty”. The ad with loss-framing of appearance risk discusses the negative physical appearance of using tanning beds or sunbathing and information related to the risk of UV rays that can ruin “natural beauty.” The ad with gain-framing of appearance risk discusses the positive physical appearance of avoiding tanning/sunbathing.

The identical print PSAs from S1 were used except for the topic itself (ie, sunscreen use): the loss-framed health-risk message discusses the negative health consequences of not using sunscreen while the gain-framing of health risk contains the positive health consequences of using sunscreen. The loss-framing of appearance risk discusses the negative physical appearance of not using sunscreen while the gain-framing of appearance risk contains the positive physical appearance of using sunscreen (ie, “using sunscreen can decrease your chance of having skin damage wrinkles, scars, age spots, and skin burn.”).

Participants were recruited from various undergraduate classes for an extra credit opportunity. Once the invitation was sent to the instructors who agreed to offer extra credit, the instructors circulated/e-mailed the invitations to their students. Upon receipt of the announcement, the participants were given a URL to the experiment website via the invitation and asked to participate in the study within a week after the receipt of the e-mail invitation. On the first page of the website, the participants were told that they would be asked to evaluate a skin cancer print PSA. Once informed of their rights as participants, they were instructed to go to the next page by clicking the “next” button. The experiment was conducted through Qualtrics.com, an online survey tool designed to facilitate online data collection through websites.

On the experiment website, the participants were randomly assigned to 1 of the 4 experimental conditions: loss-framing of health risk (S1: n = 53; S2: n = 47), gain-framing of health risk (Study1: n = 64; S2: n = 42), loss-framing of appearance risk (S1: n = 63; S2: n = 31), and gain-framing of appearance risk (S1: n = 56; S2: n = 41). The random assignment was preprogramed in the program, Qualtrics. After exposure to one of the conditions, when participants clicked the “next” button, the program led to a questionnaire containing measures for the dependent variables, manipulation check items, and demographic information. Both experiments took approximately 15 minutes. Figure A1 provides a pictorial representation of the experimental process (Appendix A).

Measures

Message framing

Gain-framing was operationalized as a message containing positive consequences in terms of one’s health or appearance obtained by avoiding indoor tanning beds or sunbathing. In contrast, loss-framing was operationalized as a message containing negative consequences in one’s health or appearance caused by using indoor tanning beds or sunbathing. Two 7-point Likert-scale items used by Cho and Boster¹⁵ were used in S1 (r = 0.46 for gain-framing and r = 0.21 for loss-framing). However, due to the low Pearson correlation scores between these 2 items, the four 7-point Likert-scale items from Baek, Shen, and Reid⁴¹ were adopted in S2 (α = .91 for gain-framing and α = .87 for loss-framing).

Risk type

Health-risk appeal was operationalized as a message appealing to skin cancer disease and death. Appearance-risk appeal was operationalized as risk of ruining one’s “natural beauty” and the possibility of skin damage like wrinkles, age spots, and skin burn. Eight 7-point Likert-scale items from Jones and Leary³⁵ and Kiene et al⁴² were slightly modified and used. Health-risk appeal was measured using 4 items (ie, “in general, the message is about health risk”; S1: α = .57; S2: α = .68). Appearance risk was measured with the other 4 items (ie, “In general, the message is about appearance threat”; S1: α = .63; S2: α = .65).

Prior experience with indoor/outdoor tanning (S1) or with sunbathing (S2)

Prior experience with tanning beds or sunbathing was measured using items by De Vries, Mesters, van’t Riet, Willems, and Reubsaet⁴³ and Knight, Kirincich, Farmer, and Hood⁴⁴: “Have you ever used tanning beds or sunbathing?”. If the participants answered “yes,” they were asked a follow-up question, “How many times have you used tanning beds or sunbathing per year? Approximately 65.3% (n = 154) reported that they have used tanning beds or sunbathing with an average of 6 times (SD = 4.29) per year. About 35% (n = 82) reported that they have never done it while 31% (n = 74) said they had done so more than 10 times. In S2, prior experience with sunbathing was measured with the items that were slightly modified from S1 (ie, “Have you ever sunbathed”). About 77.6% (n = 125) reported they have sunbathed before with an average of 7 times per year (SD = 4). About 22% (n = 36) reported that they never did and approximately 37% (n = 60) said they had done so more than 10 times.

Behavioral intent of avoiding tanning beds or sunbathing (S1) or using sunscreen (Study 2)

The five 7-point Likert-scale items from Hammond, Fong, Zanna, Thrasher, and Borland⁴⁵ were used (ie, “In the future, I do not intend to use indoor tanning beds”; “In the future, I will be more careful going out sunbathing”; α = .85). Behavioral intent of using sunscreen was measured with the four 7-point Likert-scale items, adopted from Hoffner and Ye²⁹ (ie, “In the future, I intend to use sunscreen during daily activities”; α = .79).

Manipulation Check

Message framing

Independent Samples t tests indicated the success of the manipulation in both studies. For gain-framing, participants exposed to the gain-framing message perceived it more as gain-framing (S1: M = 4.51, SD = 1.39; S2: M = 4.98, SD = 1.42) than loss-framing (S1: M = 3.22, SD = 1.53; S2: M = 3.90, SD = 1.62) (S1: t (234) = −6.79, P < .01; S2: t (159) = −4.47, P < .01). Participants exposed to the loss-framing message perceived it as more loss-framing (S1: M = 5.51, SD = 1.18; S2: M = 6.21, SD = .86) than gain-framing (S1: M = 4.87, SD = 1.12; S2: M = 5.39, SD = 1.27) (Study1: t (234) = 4.32, P <.01; S2: t (159) = 4.26, P < .01).

Risk type

Eight 7-point Likert-scale items were used to measure risk type (ie, “in general, the message is about health risk”). Participants exposed to the health-risk condition perceived it more as a health-risk message (S1: M = 5.83, SD = 0.79; S2: M = 4.98, SD = 1.42) than an appearance message (S1: M = 5.24, SD = 0.94; S2: M = 3.90, SD = 1.62) (S1: t (234) = 5.21, P < .01; S2: t (159) = 4.47, P < .01). Participants exposed to the appearance-risk message perceived it more as an appearance-risk message (S1: M = 5.77, SD = 0.86; S2: M = 6.12, SD = 0.86) than a health-risk message (S1: M = 4.63, SD = 1.02; S2: M = 5.39, SD = 1.27) (S1: t (234) = −9.29, P <.01; S2: t (159) = 4.26, P < .01). The manipulation of risk type was also successful.

To test the success of random assignment, χ² tests were conducted to check the distributions of gender across the conditions and found no significant differences in both experiments, P > .05 (S1: χ² = 5.52, df = 3, P = .14; S2: χ² = 1.50, df = 3, P = .68). Therefore, the random assignment in both experiments was successful. To analyze the data, a general linear model (GLM) was used to test anticipated interaction effects, followed by simple effect tests.

Results

A GLM revealed a significant 3-way interaction among message framing, risk type, and prior experience on participants’ behavioral intent of avoiding tanning beds or sunbathing, F(1, 228) = 4.268, P < .05, ή_p ² = 0.02, observed power = 0.54 (S1). Hypothesis 1a predicted the gain-framing of appearance risk would result in a higher level of intention to avoid tanning/sunbathing than the loss-framing of appearance risk among those with prior experience of tanning/sunbathing. However, no significant framing effect was found in appearance risk, P = .23, although the anticipated direction was shown. Thus, Hypothesis 1a was not supported.

However, Hypothesis 1b predicted that the loss-framing of health risk would result in a higher level of intention to avoid tanning/sunbathing with this group and found that the loss-framing of health risk, M = 4.75, SE = 0.22, results in a higher level of intent to avoid tanning beds/sunbathing than those with the gain-framing of health risk, M = 4.19, SE = 0.22, t (77) = 1.58, P = .06. Thus, Hypothesis 1a was supported (Figure 1a). Although S2 also yielded a significant 3-way interaction on the participants’ behavioral intent of using sunscreen, F(1, 153) = 3.96, P < .05, ή_p ² = 0.02, observed power = 0.51 (Research question 1), no significant difference was found between framing and risk type among those with prior experience of sunbathing, P > .05 (Figure 1b).

Figure 1.

A, Behavioral intent of avoiding tanning beds or sunbathing for those who have used tanning beds or sunbathing. B, Behavioral intent of using sunscreen for those who have sunbathed before.

In contrast, among those who have not used tanning beds or sunbathing, the effects of framing were found only among messages with appearance risk: the loss-framing of appearance risk (M = 6.05, SE = 0.29) produced a higher level of behavioral intent of avoiding tanning beds or sunbathing than the gain-framing of appearance risk (M = 5.21, SE = 0.30), t(42), = 2.84, P < .01 (Figure 2a; Hypothesis 2a). Therefore, Hypothesis 2a was supported. However, in terms of increasing one’s intention to use sunscreen (S2), among those who have not sunbathed, the result was quite the opposite from the result in S1: the gain-framing of appearance (M = 5.83, SE = 0.38) produced a higher level of behavioral intent of using sunscreen than the loss-framing of appearance risk among those without prior experience (M = 3.96, SE = 0.47), t(16) = −2.74, P < .01 (Figure 2b; Research question 1). Descriptive statistics are summarized in Table 1 (Appendix B).

Figure 2.

A, Behavioral intent of avoiding tanning beds or sunbathing for those who have not used tanning beds or sunbathing before. B, Behavioral intent of using sunscreen for those who have not sunbathed before.

Discussion

We investigated how message framing and risk type in skin cancer prevention messages interacted in influencing young adults’ behavioral intentions to avoid indoor/outdoor tanning (S1) and to use sunscreen (S2). We found a significant 3-way interaction effect among message framing, risk type, and prior experience in both experiments. As expected, the effects of message framing and risk type varied based on prior experience and the directions of the 3-way interactions appear to be the opposite, depending on whether indoor tanning is discouraged or sun screen use is encouraged. Among those with prior tanning/sunbathing experience, only the loss-framing of health-risk message exhibited a higher level of intention to avoid tanning/sunbathing than the gain-framing of health risk. In contrast, for those with no prior experience of tanning/sunbathing, the framing effect was shown only in the appearance-risk condition: the loss-framing of the appearance risk produced a higher level of intention to avoid future tanning/sunbathing while the gain-framing of the appearance risk was shown to be more persuasive in getting people to use sunscreen.

When PT, originally developed in testing of financial risk-taking tendency, was introduced in the field of health communication, much confusion arose in its basic tenet and applicability. Prospect theory originally explicates individuals’ tendency to take a small, but more probable gain over a less probable, but bigger gain, but take the chance for a more probable, bigger loss rather than taking a small, but less probable loss. The basic tenet mainly adopted from the original theory was regarding people’s tendency to avoid “risks” when facing gains and to take “risks” when facing losses⁴⁶ without elements of chance that were operationalized in the original PT literature. Some argued that the “nature” of risk (uncertainty) and behavioral alternatives in PT have been misapprehended.²⁰ Consequently, Rothman and Salovey’s²⁶ misguided speculation that the effects of framing would depend on prevention versus detection behaviors led many researchers’ pursuits to verify its assertion, but yielded conflicting results, not only casting doubt on the applicability of PT in the field²⁰ but also leading much effort placed to find the best message framing in general.

Here, we found several 3-way interactions that may explain why the previous discrepancies may exist. The effects of message framing depend on risk type, individuals’ prior experience, and targeted behaviors (ie, discouraging indoor/outdoor tanning vs encouraging sunscreen use). Whether the messages are to encourage sunscreen use or to discourage indoor/outdoor tanning had an impact on individuals’ willingness to follow a recommendation. Still, many questions remain: why the message framing effects were shown in the appearance risk, not in the health-risk condition, among those who did not have any prior experience of indoor/outdoor tanning? Why did the message framing and risk type interact among those with prior experience for avoiding future tanning/sunbathing, but no such effect was detected for using sunscreen? Is the finding that loss-framing worked better in discouraging a “risky” behavior (ie, indoor/outdoor tanning) while gain-framing worked better in encouraging a preventive measure (ie, sunscreen use) among those without prior experience in the appearance risk, applicable in other contexts?

It is well-documented that people who tan, particularly young adults, are less likely to be concerned about the risks of tanning^6,47 and more likely to be motivated by their belief about the benefits of a tan such as looking healthy, better, or more attractive.⁴⁸ Using one’s reverse psychology and appealing to the same motivation (ie, the motivation to improve their appearance) that led them to tan in the first place, we predicted that gain-framing of appearance risk like obtaining “natural beauty” could motivate them to avoid future indoor/outdoor tanning. Although the anticipated direction was detected, the result was inconclusive at this time, calling for further investigations. Perhaps those with prior tanning experience may as well know of risks and benefits associated with tanning already, so the loss-framing of health risk may have been perceived more convincing than the loss-framing of appearance risk. Thus, individuals’ risk perception of a targeted behavior should be explored further to understand its full effects and underlying mechanism(s).

A handful of previous studies revealed that using appearance warnings was more effective in promoting sunscreen use than using health warnings^{22,23,36
–38} and evoked lower intentions to engage in tanning behavior.³⁵ However, we found that it depends on message framing, risk type, and individuals’ prior experience. In practice, our findings indicate that to discourage future indoor/outdoor tanning among those with prior experience, messages emphasizing losses in terms of one’s health will motivate the avoidance tanning behaviors. For those with no prior experience, messages emphasizing potential appearance losses will work better for discouraging future indoor/outdoor tanning while messages emphasizing gains like improving appearance will do a better job in encouraging sunscreen use.

Some limitations should be acknowledged. Stimuli used in this study cannot fully mimic exposure to a PSA in a real-life setting, and we used only 1 message to represent each of the four message categories without a control group. Thus, future replications of the findings should be put forward with careful consideration for using multiple messages for each condition with a control group. The measure of prior experience of tanning/sunbathing used in S1 raised some concern of conflation despite the facts that the items were from previously published studies and amended in S2. Although college students are a very relevant age-group, the findings from this study may not be generalizable to the general public since we used a student sample. Therefore, future studies should expand to use diverse and more samples to corroborate these findings in a larger context. Despite these limitations, the findings from both experiments provide a better understanding of the interplay of message framing, risk type, and one’s prior experience.

With the high frequency of media coverage promoting the benefits of tanning⁴⁸ and in turn influencing people’s perceptions toward tanning, this study contributes to public health promotion studies by examining how people respond to different kinds of skin cancer prevention messages. Investigating whether or not characteristics of health promotion messages (ie, message framing and risk type) bear any persuasive power and influence individuals’ decisions to adopt a health recommendation is critical for garnering the success of future health promotion campaigns. Given the importance of approaching different message strategies on PSAs based on target audience characteristics, the findings provide health advocates and campaign designers much needed insight into effective use of message framing and risk type and provide a meaningful step in the right direction.

Skin cancer prevention public health campaigns/educational programs should be tailored depending on individuals’ prior experience and perhaps could be launched earlier in targeting young adults before they are exposed to risk factors such as promotional messages for indoor tanning or before tanning becomes a routine behavior. Such public health campaign messages should emphasize the risk of losing “natural beauty” to discourage excessive indoor/outdoor tanning or in obtaining “positive appearance” to encourage preventive measures like sunscreen use. On the other hand, messages emphasizing loss of health (eg, developing melanoma) should be adopted in deterring further tanning behaviors among individuals with prior tanning experience. Perhaps such messages can be delivered through popular media programs including social media channels that are specialized in beauty/fashion tips or popular among young adults. Such embedded messages within media programing could set new norms among young adults for skin cancer prevention practices. However, finding additional methods to deliver tailored messages to appropriate target publics is a must. Therefore, quests for finding ways to tailor messages, contexts, and framings to communicate with young adults about skin cancer prevention should be continued through systematic investigations of message effects as well as message deliberation methods.

SO WHAT?

What is already known on this topic?

The main question is, should we emphasize gains or losses in skin cancer prevention campaign messages? The answer to this question is neither simple nor easy, knowing how complex individuals’ desire, motivation, and cognition interplay with their own prior experience with this particular, risky behavior in terms of rejecting the risky behavior or adopting a preventive measure.

What does this article add?

At least, we’ve learned that an individual’s decision to either avoid a risky behavior (ie, indoor/outdoor tanning) or adopt a preventive measure (ie, using sunscreen) depends on types of message framing, risk type, and their own prior experience with that particular behavior (ie, indoor/outdoor tanning).

What are the implications for health promotion practice or research?

Effective promotion of skin cancer prevention depends on what risk type these young people consider the most relevant as well as message framing where messages emphasize benefits or losses that will resonate with people from different backgrounds and prior experience.

Footnotes

Appendix A

Appendix B

Table A1.

Behavioral Intent of Avoiding Tanning Beds/Sunbathing (S1)/Wearing Sunscreen (S2).

			n	Mean	SD
Message framing	Loss		116/78	4.98/4.63	1.43/1.23
Message framing	Gain		120/83	4.77/4.88	1.48/1.09
Risk type	Health		117/89	4.85/4.73	1.54/1.18
Risk type	Appearance		119/72	4.90/4.80	1.37/1.16
Two-way interaction	Loss/health Gain/health Loss/appearance Gain/appearance		53/47 64/42 63/31 56/41	5.00/4.62 4.73/4.85 4.96/4.65 4.82/4.91	1.46/1.25 1.61/1.09 1.41/1.21 1.34/1.11
Three-way interaction	Experience—Yes	Loss/health	38/38	4.75*/4.68	1.43/1.21
		Gain/health	39/30	4.19*/4.87	1.66/0.99
		Loss/appearance	41/25	4.38/4.81	1.31/1.09
		Gain/appearance	36/32	4.61/4.65	1.44/0.98
	Experience—No	Loss/health	15/9	5.65/4.36	1.39/1.46
		Gain/health	25/12	5.58/4.81	1.10/1.34
		Loss/appearance	22/6	6.05/3.96	0.85/1.54
		Gain/appearance	20/9	5.21/5.83	1.05/1.12

Note. S1: F(1, 228) = 4.268, P < .05.

S2: F(1, 153) = 3.96, P < .05.

*P < .05. **P < .01.

Authors’ Note

This study used 2 experiments with a 2 (message framing: gain vs loss) × 2 (risk type: health vs appearance risk) factorial design to test whether message framing (ie, gain vs loss) and risk type (ie, health vs appearance risk) in skin cancer prevention messages interact with one’s prior experience. The individuals included 397 participants (236 for experiment 1 and 161 for experiment 2). The authors found 3-way interactions among message framing, risk type, and prior experience. When the intent of the message was to encourage sunscreen use, the effects of message framing, and risk type were shown to be the exact opposite directions from when the intent was to discourage indoor/outdoor tanning. Based on these results, the authors concluded that in order to discourage tanning among those with prior experience, messages emphasizing losses in terms of one’s health will work better. Additionally, for those with no prior experience, messages emphasizing potential appearance losses will work better for discouraging tanning while messages emphasizing gains like improving appearance will do a better job in encouraging sunscreen use. Hannah Kang is now associated with Ewha Institute for Age Integration Research, Ewha Womans University, Seoul, South Korea.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

References

Centers for Disease Control and Prevention (CDC). Skin Cancer Statistics. http://www.cdc.gov/cancer/skin/statistics/. Accessed August 30, 2017.

Rogers

Weinstock

Feldman

Coldiron

. Incidence estimate of nonmelanoma skin cancer (keratinocyte carcinomas) in the US population, 2012. JAMA Dermatol. 2015;151(10): 1081–1086. Doi:10.100/jamadermatol.2015.1187

Little

Eide

. Update on the current state of melanoma incidence. Dermatol Clin. 2012;30(3):355–361. https://doi.org/10.1016/j.det.2012.04.001

American Cancer Society. Key Statistics for Melanoma Skin Cancer 2015. https://www.cancer.org/cancer/melanoma-skin-cancer/about/key-statistics.html. Accessed August 30, 2017.

American Cancer Society (ACS). Melanoma Skin Cancer. https://www.cancer.org/cancer/melanoma-skin-cancer.html. Accessed August 30, 2017.

Monfrecola

Fabbrocini

Posteraro

Pini

. What do young people think about the dangers of sunbathing, skin cancer and sunbeds? A questionnaire survey among Italians. Photodermatol Photoimmunol Photomed. 2000;16(1):15–18.

Elwood

Gallagher

Davison

Hill

. Sunburn, suntan and the risk of cutaneous malignant melanoma—the western Canada melanoma study. Br J Cancer. 1985;51(4):543–549.

Koh

Geller

. Melanoma control in the United States: current status. Recent Results Cancer Res. 1995;139:215–224.

Wakefield

. Use of mass media campaigns to change health behavior. Lancet. 2010;376(9748):1261–1271.

10.

Tewksbury

Scheufele

. News framing theory and research. In: Jennings

Oliver

, eds. Media Effects: Advances in Theory and Research. 3rd ed. New York, NY: Routledge; 2009:17–33.

11.

Kahneman

Tversky

. Prospect theory: an analysis of decision under risk. Econometrica. 1979;47(2):263–292.

12.

Tversky

Kahneman

. The framing of decisions and the psychology of choice. Science. 1981;211(4481):453–458.

13.

Gray

Harrington

. Narrative and framing: a test of an integrated message strategy in the exercise context. J Health Commun. 2011;16(3):264–281.

14.

Ahern

Connolly-Ahern

Shen

. Communicating the risks of fetal alcohol spectrum disorder: effects of message framing and exemplification. Health Commun. 2010;25(8):692–699.

15.

Cho

Boster

. Effects of gain versus loss frame antidrug ads on adolescents. J Commun. 2008;58(3):428–446.

16.

Goodall

Appiah

. Adolescents’ perceptions of Canadian cigarette package warning labels: investigating the effects of message framing. Health Commun. 2008;23(2):117–127.

17.

Maheswaran

Meyers-Levy

. The influence of message framing and issue involvement. J Mark Res. 1990;27(3):361–367.

18.

O’Keefe

Jensen

. The advantages of compliance of the disadvantages of noncompliance? A meta-analysis review of the relative persuasive effectiveness of gain-framed and loss-framed messages. In: Beck

, ed. Communication Yearbook 30. Mahwah, NJ: Lawrence Erlbaum Associates; 2006:1–43.

19.

O’Keefe

. Gain-framed messages do not motivate sun protection: a meta-analytic review of randomized trials comparing gain-framed and loss-framed appeals for promoting skin cancer prevention. Int J Environ Res Public Health. 2012;9(6):2121–2133.

20.

O’Keefe

. From psychological theory to message design: lessons from the story of gain-framed and loss-framed persuasive messages. In: Cho

, ed. Health Communication Message Design: Theory and Practice. Thousand Oaks, CA: Sage; 2012:3–20.

21.

Hevey

Pertl

Thomas

Maher

Craig

Ni Chuinneagain

. Body consciousness moderates the effect of message framing on intentions to use sunscreen. J Health Psychol. 2010;15(4):553–559.

22.

Mahler

HIM

Kulik

Gibbons

. Effects of appearance-based intervention on sun protection intentions and self-reported behaviors. Health Psychol. 2003;22(2):199–209.

23.

Mahler

HIM

Kulik

Harrell

Correa

Gibbons

Gerrard

. Effects of UV photographs, photoaging information, and use of sunless tanning lotion on sun protection behaviors. Arch Dermatol. 2005;141(3):373–380.

24.

Rogers

Cognitive and Physiological Processes in Fear Appeals and Attitude Change: A Revised Theory of Protection Motivation. In: Cacioppo

Petty

, eds. Social Psychophysiology. New York, NY: Guilford Press; 1983:153–176.

25.

Unger

Stearns

The Use of Fear and Guilt Messages in Television Advertising: Issues and Evidence. In: Murphy

Laczniak

, eds. American Marketing Association Educators’ Proceedings. Chicago, IL: American Marketing Association; 1983:16–20.

26.

Rothman

Salovey

. Shaping perceptions to motivate healthy behavior: the role of message framing. Psychol Bull. 1997;121(1):3–19.

27.

Meyerowitz

Chaiken

. The effect of message framing on breast self-examination attitudes, intentions, and behavior. J Pers Soc Psychol. 1987;52(3):500–510.

28.

Gallagher

Updegraff

. Health message framing effects on attitudes, intentions, and behavior: a meta-analytic review. Ann Behav Med. 2012;43(1):101–116.

29.

Hoffner

. Young adults’ responses to news about sunscreen and skin cancer: the role of framing and social comparison. Health Commun. 2009;24(3):189–198.

30.

Mays

Zhao

. The influence of framed messages and self-affirmation on indoor tanning behavioral intentions in 18-to 30-year-old women. Health Psychol. 2016;35(2):123–130.

31.

Mays

Tercyak

. Framing indoor tanning warning messages to reduce skin cancer risks among young women: implications for research and policy. Am J Public Health. 2015;105(8):e70–e76.

32.

Hevey

Dolan

. Approach/avoidance motivation, message framing and skin cancer prevention: a test of the congruency hypothesis. J Health Psychol. 2014;19(8):1003–1012.

33.

Van’t Riet

Cox

. Does perceived risk influence the effects of message framing? A new investigation of a widely held notion. Psychol Health. 2014:29(8):933–949.

34.

Hwang

Cho

Sands

Jeong

. Effects of gain- and loss-framed messages on the sun safety behavior of adolescents: the moderating role of risk perceptions. J Health Psychol. 2012;17(6):929–940.

35.

Jones

Leary

. Effects of appearance-based admonitions against sun exposure on tanning intentions in young adults. Health Psychol. 1994;13(1):86–90.

36.

Cornelis

Cauberghe

De Pelsmacker

. Being healthy or looking good? The effectiveness of health versus appearance-focused arguments in two-sided messages. J Health Psychol. 2014;19(9):1132–1142.

37.

Tuong

Armstrong

. Effect of appearance-based education compared with health-based education on sunscreen use and knowledge: a randomized controlled trial. J Am Acad Dermatol. 2014;70(4):665–669.

38.

Williams

Grogan

Clark-Carter

Buckley

. Appearance-based interventions to reduce ultraviolet exposure and/or increase sun protection intentions and behaviours: a systematic review and meta-analyses. Br J Health Psychol. 2013;18(1):182–217.

39.

Cafri

Thompson

Roehrig

van den Berg

Jacobsen

Stark

. An investigation of appearance motives for tanning: the development and evaluation of the Physical Appearance Reasons For Tanning Scale (PARTS) and its relation to sunbathing and indoor tanning intentions. Body Image. 2006;3(3):199–209.

40.

Greene

Brinn

. Messages influencing college women’s tanning bed use: statistical versus narrative evidence format and a self-assessment to increase perceived susceptibility. J Health Commun. 2003;8(5):443–461.

41.

Baek

Shen

Reid

. Effects of message framing in anti–binge drinking PSAs: the moderating role of counterfactual thinking. J Health Commun. 2013;18(4):442–458.

42.

Kiene

Barta

Zelenski

Cothran

. Why are you bringing up condom use now? The effect of message content on framing effects of condom use messages. Health Psychol. 2005;24(3):321–326.

43.

De Vries

Mesters

van’t Riet

Willems

Reubsaet

. Motives of Belgian adolescents for using sunscreen: the role of action plans. Cancer Epidemiol Biomarkers Prev. 2006;15(7):1360–1366.

44.

Knight

Kirincich

Farmer

Hood

. Awareness of the risk of tanning lamps does not influence behavior among college students. Arch Dermatol. 2002;138(10):1311–1315.

45.

Hammond

Fong

Zanna

Thrasher

Borland

. Tobacco denormalization and industry beliefs among smokers from four countries. Am J Prev Med. 2006;31(3):225–232.

46.

Millar

Promoting safe driving behaviors: the influence of message framing and issue involvement. J Appl Soc Psychol. 2000;30(2):853–866.

47.

Abroms

Jorgensen

Southwell

Geller

Emmons

. Gender differences in young adults’ beliefs about sunscreen use. Health Educ Behav. 2003;30(1):29–43.

48.

Cho

Hall

Kosmoski

Fox

Mastin

. Tanning skin cancer risk, and prevention: a content analysis of eight popular magazines that target female readers, 1997-2006. Health Commun. 2010;25(1):1–10.