Abstract
This study examines the impact of message framing and visual cues on the effectiveness of social and mass media advertising promoting omega-3 screening programs for pregnant people and their families. (Taking omega-3 supplements during pregnancy may reduce the risk of premature birth for pregnant people who are deficient in this nutrient.) Women of childbearing age responded better to positively framed messages with emotional visual cues, whereas the general audience had high screening intentions regardless of the advertising condition. Processing fluency directly affected respondents’ intentions to screen, attitudes, and awareness. A dose-response relationship between attitudes, awareness, and intentions to screen was observed, whereas no moderation or moderated-mediation pathways were detected. Advertisements promoting omega-3 screening should emphasize its benefits and consider the interplay between advertising attitudes and awareness.
- Health promotion campaigns
- public health messaging
- social and mass media
- processing fluency
- regulatory focus
- need for cognition
MANAGEMENT SLANT
Advertisement pretesting is important for ensuring that the messages are well received, inclusive, and effective in promoting uptake of health behaviors among the target audience.
Basic demographic segmentation can be a cost-efficient and effective way to develop evidence-based segmentation strategies for social and mass media advertising campaigns, instead of relying on highly tailored advertisements.
The use of gain-frame messages and emotional visual cues can be effective in promoting positive attitudes and driving behavioral intentions in health-related advertising campaigns.
Advertisers should focus on developing such messaging strategies that resonate with their target audience to increase engagement and encourage positive behavioral change.
INTRODUCTION
Premature birth is the leading cause of death for children under age five worldwide (Australian Institute of Health and Welfare, 2016). Omega-3 supplementation has been endorsed by the Australian Pregnancy Care guidelines (Department of Health, 2020) as a way to reduce the risk of prematurity for pregnant people who are low in this nutrient (Makrides, Best, Yelland, et al., 2019; Simmonds, Sullivan, Skubisz, et al., 2020). In South Australia, this evidence is being translated into routine antenatal care through a statewide omega-3 screening and targeted treatment program that detects low total omega-3 status in early pregnancy and advises targeted omega-3 supplementation to reduce the risk of early premature birth. Promoting the program to ensure equitable access to omega-3 screening for at-risk pregnant people, however, requires a population-appropriate strategy targeting a diverse audience.
Mass media campaigns—in particular, social media—are one means to disseminate such information and may serve as a cost-efficient and wide-reaching solution to promote screening programs, allowing people to make an informed decision about their health-care opportunities. Also, mass marketing strategies facilitate inclusion and awareness to reach pregnant people at risk of low omega-3, a population group that is otherwise hard to target precisely. Mass media may also help to overcome streams of inequity among individuals who have restricted access to resources or health care information by providing readily available, evidence-based information through mass exposure at no cost to them.
Campaigns have been previously used in attempts to evoke health behavior change across several domains (Wakefield, Loken, and Hornik, 2010). Most notably, these campaigns have sought to curb alcohol and other drug use (Douglass, Early, Wright, et al., 2017; Durkin, Brennan, and Wakefield, 2012; Farrelly, Nonnemaker, Davis, and Hussin, 2009; Solomon, Bunn, Flynn, et al., 2009; Stephenson, Palmgreen, Hoyle, et al., 1999), increase uptake of screening (Anderson, Mullins, Siahpush, et al., 2009; Worthington, Feletto, Lew, et al., 2020) and vaccination (Shilo, Rossman, and Segal, 2021; Zimicki, Hornik, Verzosa, et al., 1994), encourage cancer prevention behaviors (Montague, Borland, and Sinclair, 2001), and promote healthy dietary and exercise practices (Northcott, Curtis, Bogomolova, et al., 2021; Samson, Nanne, and Buijzen, 2021). Often, campaigns seek to deliver a specific public health message and are usually executed across several traditional media—more recently, digital and social media—with a key goal to drive repeated, passive exposures to audiences over time to encourage behavioral change (Wakefield et al., 2010).
An important first step in designing messages for public health campaigns is pretesting, which involves refining messages through pilot studies with a smaller audience before disseminating information to the population at large (O’Keefe, 2018). This process increases message acceptance and inclusivity, and it avoids unintended boomerang effects that may lead to the rejection of public health advice (Brown, Lindenberger, and Bryant, 2008; Parvanta, Gibson, Forquer, et al., 2013). Additionally, message pretesting can inform scalable mass media strategies underpinned by formative evidence (O’Keefe, 2018).
When developing materials, advertising practitioners must consider both the creative execution and individual traits of the target audience to ensure congruency and acceptance of messages (Eisend, Muldrow, and Rosengren, 2022). Message-framing strategies that highlight the benefits or losses of partaking in omega-3 screening, for example, may achieve various levels of effectiveness depending on the audience’s involvement and relevance (Meyerowitz and Chaiken, 1987). Similarly, the effectiveness of visual elements may differ on the basis of age, gender, or relevance to the audience, eliciting various behavioral intentions and attitudes toward screening (Finn, 1988; Kisielius and Sternthal, 1984). Such strategies should be viewed in the context of the audience’s ability to process the material (i.e., processing fluency; Alter and Oppenheimer, 2009) and their psychographic orientation, such as whether they are prevention or promotion oriented (i.e., regulatory focus; Gomez, Borges, and Pechmann, 2013; Kees, Burton, and Tangari, 2010), or their level of need for information (i.e., need for cognition; Haugtvedt, Petty, and Cacioppo, 1992).
The authors of this study, therefore, aimed to inform the development of a statewide mass media campaign promoting an omega-3 test-and-treat program for pregnant people in South Australia by undertaking a 2 (Message Frame) × 2 (Visual Cue) full-factorial, between-subjects, randomized pretesting experiment. The authors investigated how message frames and visual cues in omega-3 advertisements affect behavioral intentions, attitudes, and awareness related to omega-3 screening. Processing fluency was assessed as a mediator, whereas regulatory focus and need for cognition were examined as moderators. This information is useful for advertisers who want to increase the uptake of health behaviors among unaware, hesitant, or low-health-literate populations (Nielsen-Bohlman, Panzer, and Kindig 2004; Okuhara, Ishikawa, Okada, et al., 2017).
THEORETICAL FRAMEWORK AND BACKGROUND
Benefit to Target Audience
Rates of premature birth continue to rise concerningly (World Health Organization [WHO], 2023). Premature birth, especially early premature birth, is associated with many health risks, including long-term neurological disability (e.g., cerebral palsy) and increased risk of chronic lung disease and breathing problems, neurodevelopmental issues, and perinatal mortality (AIHW, 2016; Centers for Disease Control and Prevention, 2018; Granese, Gitto, D’Angelo, et al., 2019). Premature babies are at increased risk of developing chronic degenerative diseases in adulthood, including heart disease, stroke, and diabetes mellitus (Saigal and Doyle, 2008). A systematic review of 70 randomized controlled trials found a 42 percent relative reduction in early premature birth for omega-3 fatty acid supplementation (Middleton, Gomersall, Gould, et al., 2018); additionally, analysis of data from the largest randomized controlled trial found that only people with low omega-3 fatty acid status benefited from omega-3 supplementation to reduce their risk of early premature birth (Makrides et al., 2019; Simmonds et al., 2020).
Effective messaging for promoting an omega-3 test-and-treat program during pregnancy should primarily target pregnant people while also recognizing the importance of engaging partners and families as secondary audiences to achieve mass inclusion (Dunkel-Schetter, Sagrestano, Feldman, and Killingsworth, 1996). When designing effective health messages to promote the uptake of a screening program, it is essential that the message fidelity of these materials is well understood and communicated clearly to both primary and secondary audiences to promote awareness and uptake of the program.
Effective messaging for promoting an omega-3 test-and-treat program during pregnancy should primarily target pregnant people while also recognizing the importance of engaging partners and families as secondary audiences to achieve mass inclusion.
Message pretesting through formative research is a crucial step in designing and evaluating effective health messages (O’Keefe, 2018). For a new omega-3 test-and-treat program, the primary message for pregnant individuals is to undergo omega-3 screening and, if necessary, take supplements until 37 weeks of pregnancy. This differs from common advice for pregnant individuals to take various supplements without knowing whether, or which, supplements will be of use to them. This is particularly relevant, because most pregnant individuals have adequate omega-3 levels, and only 17.5 percent have low levels (Simmonds et al., 2020). A misinterpretation of the message by those who already have an adequate omega-3 status could lead to unnecessary supplementation and potential harm. Effective health messages are therefore needed to increase behavioral intentions to screen for omega-3, improve advertising attitudes and awareness about omega-3 screening during pregnancy, and mitigate possible misinterpretation.
Message Framing
Message framing is a technique that recognizes the potential for a message to be presented positively or negatively (Quick, Reynolds-Tylus, Fico, and Feeley, 2016). In the context of promoting an omega-3 test-and-treat program, a health message could either highlight the positive outcomes of adhering to antenatal screening advice (gain-framed), such as having a healthy baby, or emphasize the negative outcomes of not adhering to the advice (loss-framed), such as premature birth (Meyerowitz and Chaiken, 1987; Van’t Riet, Cox, Cox, et al., 2016). This approach is based on prospect theory, which suggests that people are more likely to take risks to avoid potential losses than to achieve potential gains, even when the outcomes are objectively the same (Kahneman and Tversky, 1979; Shimul, Cheah, and Lou, 2021). The effectiveness of message framing depends on the specific context, and it has been found that loss-framed messages are generally more effective in promoting detection-based behaviors, whereas gain-framed messages are better suited for prevention-based behaviors (Suls and Wallston, 2008). In earlier research, for example, negative messages about breast cancer screening, a detection-based behavior, led to more positive attitudes and intentions toward the behavior than positive messages (Meyerowitz and Chaiken, 1987), whereas positive messages about exercise, a prevention-based behavior, were more effective at promoting intentions to exercise than negative messages (Robberson and Rogers, 1988).
Promoting the uptake of an omega-3 screening program during pregnancy, however, is more complex than simply categorizing the behavior as detection or prevention focused. The program requires pregnant people to both screen for omega-3 levels and subsequently treat, if necessary, or maintain their usual dietary practices if their levels are adequate to reduce their risk of premature birth. The presence of both detection-based and prevention-based behaviors within this screening program therefore introduces intricacies in devising effective messaging strategies. In contrast to prior studies focused on singular behaviors, such as breast cancer screening (Meyerowitz and Chaiken, 1987) or exercise (Robberson and Rogers, 1988), the coexistence of these behaviors in the current study adds a further component. This novel dynamic—which, to the best of the authors’ knowledge, has not been extensively explored in existing literature—constitutes a distinct and original contribution to the field.
Understanding the nuanced relationship between these behaviors and tailoring messaging approaches accordingly will be crucial in effectively promoting the program. By considering the interconnected nature of detection- and prevention-based behaviors, messaging strategies can be designed to address the unique challenges and motivations associated with both aspects. Whereas other studies have focused exclusively on either detection- or prevention-based behaviors and their link with message framing, this study contributes key insights into the complex interplay between these behaviors simultaneously in the context of gain- and loss-framed messages; accordingly, conducting an empirical investigation becomes essential in bridging the knowledge gap (Maheswaran and Meyers-Levy, 1990; Rothman, Salovey, Antone, et al., 1993).
Visual Cues
Similarly, visual cues within an advertisement may have the potential to enhance or inhibit a message’s perceived effectiveness. One study (Kisielius and Sternthal, 1984) suggested that images related to a message can enhance consumer elaboration, whereas other research (Finn, 1988) noted that congruent images may further elicit consumer attention, thus enhancing positive attitudes. In the context of omega-3 messaging, advertisers who use visual cues might seek to enhance the uptake of omega-3 testing through providing relevant informational cues, such as the inclusion of a pathology request form, acting as a visual aid to reinforce the message of omega-3 testing. This rational approach is based on the information-processing model, in which consumers make decisions that are based on logical reasoning and careful thought (Albers-Miller and Stafford, 1999). By contrast, an emotional appeal might be used to elicit a positive response toward screening. A visual depiction of a happy and healthy mother and baby, for instance, can help to elaborate on the emotional significance of maternal health and healthy outcomes. The effectiveness of these approaches can vary, with evidence suggesting that messages adopting a rational appeal might be more effective because of their ability to articulate direct information (Coulson, 1989), whereas other research suggests that emotional appeals are better able to elicit feelings and produce a greater consumer response (Page, Thorson, and Heide, 1990). Inconsistencies in past research, however, suggest that the effectiveness of visual cues may be situational and therefore provide limited guidance in the development of public health campaigns. Empirical testing of different visual cues is needed in the context of promoting such screening programs. Thus:
RQ1: Is there a difference in outcomes for behavioral intentions, attitudes, and awareness between the message frame and visual cue conditions for the primary and secondary target audiences regarding omega-3 screening?
RQ2: Is there a correlation between behavioral intentions, attitudes, and awareness among the general target audience?
Moderation and Mediation
Despite the importance that message framing and visual cues might have in effectively driving public health behaviors, many of the existing investigations largely overlook how differences in processing style or individual traits of the target audience play in predicting their effectiveness (Jensen, King, Carcioppolo, et al., 2012). There is, therefore, a need to explore the possible role of mediating and moderating pathways to provide light on why, how, or when a phenomenon occurs. Mediators can help explain the relationship between independent and dependent variables, and moderators can affect the strength of this relationship (Bennett, 2000).
Processing Fluency
Processing fluency is the perceived ease with which one is able to process external information (Alter and Oppenheimer, 2009), characterized by the degree of effort and speed required to interpret the message (Winkielman, Schwarz, Fazendeiro, and Reber, 2003). Individuals may evaluate information on the basis of subjective feelings of ease or difficulty in processing information (Schwarz, 2004). Within the context of consumer behavior, processing fluency is thought to mediate the way in which consumers evaluate messaging stimuli, demonstrating that stimuli that is well understood and easy to process may elicit greater positive judgment and more favorable attitudes toward a given stimulus as a result of greater processing fluency (Labroo and Lee, 2006; Okuhara et al., 2017). Thus:
RQ3: Does processing fluency mediate the effects of message effectiveness among the general target audience?
Regulatory Focus
Regulatory focus within the context of health describes the concept that individuals have a natural tendency to be prevention or promotion oriented when pursuing health goals (Gomez et al., 2013; Kees et al., 2010). Consumers are said to experience a natural fit when message strategies, such as gain-frame or loss-frame messages, emphasize consequences that are congruent with one’s regulatory orientation (Aaker and Lee, 2006; Zhao and Pechmann, 2007), in turn, enhancing the persuasiveness of the message. The implication is that a gain-frame message (highlighting the positive outcomes of screening) for promotion-focused individuals, or a loss-frame message (highlighting the negative consequences of not screening) for prevention-focused individuals, is likely to achieve a greater regulatory fit and, therefore, message acceptance (Zhao and Pechmann, 2007). Exploring the influence of individuals’ regulatory orientation on message acceptance and persuasiveness may shed light on the potential variability in regulatory focus and its relevance for selecting suitable message frames. The findings will help reveal the importance of the congruence between message strategies, such as gain-frame or loss-frame messages, and individuals’ regulatory orientation in designing advertisements that effectively cater to both prevention-focused and promotion-focused individuals if such variability is apparent.
Need for Cognition
When developing stimuli, consideration must be given to need for cognition, where individuals enjoy a varying level of need for information (Haugtvedt et al., 1992). People who score high on the need for cognition scale are said to intrinsically enjoy thought, compared with people who score low on the scale, who tend to avoid effortful, cognitive work (Haugtvedt et al., 1992). It is important to note that an individual’s need for cognition is thought to contribute to the level of involvement that the person gives toward a message and, hence, the motivation to process these messages (Andrews, Durvasula, and Akhter, 1990; Zhang and Buda, 1999). It is possible that the argument strength, subliminally represented through the different informational or emotional visual cues, may be processed at various levels, depending on the cognitive predisposition of the individual, and thus moderate a message’s overall effectiveness. For individuals with a high need for cognition, exposure to informational cues that are high in argument strength may be more desirable; likewise, for individuals with a low need for cognition, exposure to emotional cues with little information may be more effective (Batra and Stayman, 1990; Haugtvedt et al., 1992). Thus:
RQ4: Does regulatory focus or need for cognition moderate the effects of processing fluency on message effectiveness among the general target audience?
METHOD
Study Design
The study was a 2 (Message Frame) × 2 (Visual Cue) full-factorial, between-subjects, randomized experimental survey design, examining message frames and visual cues on behavioral intentions to screen for omega-3, the primary outcome of this study, and attitudinal and awareness outcomes, the secondary outcomes. The authors of this study also explored the relationship of the primary outcome of interest, behavioral intentions to screen for omega-3, with attitudinal and awareness outcomes. The authors additionally explored the potential mediating role of processing fluency and the moderating roles of regulatory focus and need for cognition on behavioral, attitudinal, and awareness outcomes. Two message frames (one gain frame and one loss frame) and two visual cues (one emotional and one informational) were examined, resulting in a total of four experimental conditions (See Figure 1). A full-factorial study design was used to comprehensively assess the effects of different messaging factors and their interactions on behavioral, attitudinal, and awareness outcomes, thereby highlighting the most appropriate components for promoting and encouraging the uptake of an omega-3 screening program and eliminating the need for a separate control group.
Participants and Procedure
This study was designed in line with the guideline for studies reporting mediation analyses (Lee, Lamb, Hopewell, et al., 2021) and the National Health and Medical Research Council’s National Statement on Ethical Conduct in Research Involving Humans (NMMRC, 2007). This study received approval from the Women’s and Children’s Health Network Human Research Ethics Committee (HREC/20/WCHN/138). This survey captured the views of the general South Australian population (>18 years) and their perspectives regarding omega-3 stimuli to develop a public health mass media campaign to promote an omega-3 screening program in South Australia. The campaign for which these advertisements will be later used seeks to target a primary target audience, women of childbearing age, and to capture secondary target audiences who support pregnant people, including women older than childbearing age (e.g., mothers and grandmothers of women of childbearing age) and the general adult male population in South Australian (e.g., partners, fathers, and grandfathers of women of childbearing age), to achieve mass awareness.
Participant Recruitment and Consent
Participants were recruited through a paid panel provider, Pure Profile, an Australian platform that distributes surveys and sources participants to partake in research. Eligible participants were invited to complete the survey after reading the participant information sheet and agreeing to participate in the survey. Participants were able to withdraw from the study at any time and were reimbursed for their participation.
Sample Size
An a priori power analysis was conducted using GPower 3.1 for the primary outcome of interest, behavioral intentions (Faul, Erdfelder, Buchner, and Lang, 2009), adopting a type I error rate of .05 and power of .80. Using an analysis of covariance (ANCOVA) with an effect size of .25, the study required a minimum of 210 participants to be statistically powerful; thus, the authors aimed to recruit at least 260 participants to account for dropouts and inflation of multiple comparisons and secondary outcome measures (n = 65 per experimental condition).
Study Setting and Procedure
The advertisement-pretesting experiment was programmed and self-administered online through Qualtrics, an online survey platform (Qualtrics XM, n.d.). The survey flow was developed on the basis of the recommended flow for experimental survey designs in advertising research (Geuens and De Pelsmacker, 2017). Participants were first presented with a short introduction detailing the purpose of the survey and were then asked a series of eligibility questions. Next, they were asked about their prior knowledge regarding premature birth and omega-3 behaviors before being simply randomly allocated and presented to one of four experimental conditions. Next, participants were exposed to a set of delay questions before being met with a quality check and were then asked to recall the key messages in the public health communication to which they were previously exposed. After this, participants’ attitudes and behavioral intentions toward omega-3 screening were measured. Message-processing fluency, regulatory focus, and need for cognition were then captured, followed by a manipulation check and a set of demographic questions.
Independent Variables
Message Frame. Two message frames were explored: one gain-frame message and one loss-frame message. The gain-frame message depicted the benefits of omega-3 screening and treatment and the reduced risk of premature birth. The loss-frame message depicted the consequences of not partaking in omega-3 testing and treatment and, subsequently, the increased risk of premature birth if pregnant people have a low omega-3 status.
Visual Cue. Two visual cues were also explored: one emotional visual cue and one informational visual cue. The emotional visual cue illustrated a mother and newborn baby, in an attempt to appeal to emotional, experiential side of consumers—particularly, the primary target audience. The informational visual cue illustrated a pathology request form detailing the omega-3 screening test available to consumers.
The study design included four experimental conditions:
Dependent Variables
Using a 5-point Likert scale, the authors measured behavioral intentions by assessing the participants’ willingness to undergo omega-3 screening. Attitudes toward the message were examined by asking participants about their feelings regarding the public health communication, which were assessed on a 9-point bipolar Likert scale ranging from 1 (dislike) to 9 (like). Mood was examined by obtaining participants’ emotional responses to the public health communication, evaluated using a three-item scale on a 9-point Likert scale. Perceived advertising message effectiveness was examined by asking participants about their perceptions of how convincing and influential the message was, which was measured using a three-item scale on a 4-point Likert scale. Advertisement free recall was assessed through participants’ unaided recall of the message’s arguments, scored on the basis of correctness against predefined answers. Advertisement key message comprehension was evaluated by assessing participants’ understanding of the main message conveyed by the communication, coded as correct or incorrect against a predefined answer.
Mediator. For the mediator, processing fluency, the authors assessed participants’ ease in processing the public health communication using a four-item scale on a 7-point Likert scale, calculating a mean processing fluency score.
Moderators. For regulatory focus, participants’ motivational orientation (promotion or prevention focused) was measured using a 10-item scale, generating mean composite scores for each focus.
For need for cognition, participants’ preference for cognitive engagement was measured using a six-item scale on a 5-point Likert scale, resulting in a mean composite score.
An in-depth discussion on each of these validated measures, including dependent variables, mediators, and moderators and how they were operationalized is provided (See Appendix).
Descriptive Variables
Demographics and Prior Knowledge and Behavior. Respondent’s age, gender, language, country of origin, and level of education were captured. Ages were categorized into six groups: 18–24 years, 25–34 years, 35–44 years, 45–54 years, 55–64 years, and 65+ years. Information regarding respondents’ prior knowledge of premature birth, knowledge of omega-3, and prior omega-3 testing behaviors was also captured.
Manipulation Check
Message Frame. A single-item question to assess participants’ perceptions regarding the message frame was examined. Participants were asked to indicate on a 7-point bipolar scale, ranging from −3 (negatively framed) to +3 (positively framed), the extent to which they perceived the message to elicit a negative, neutral, or positive frame, with higher scores indicating greater positive message perceptions.
Visual Cue. A single-item question was also used to test participants’ perceptions of the image. Participants were asked to indicate on a 7-point bipolar scale, ranging from −3 (emotional based) to +3 (informational based), the extent to which they perceived the image to be emotional, neutral, or informational, with higher scores indicating greater informational image perceptions.
Data and Statistical Analyses
Data were analyzed quantitatively using statistical analysis software SPSS (IBM SPSS Statistics for Macintosh, 2019). Descriptive statistics and randomization checks were undertaken using chi-square analysis and analysis of variance. Manipulation checks were assessed using a series of t tests. Inferential statistics, including group differences, were undertaken using ANCOVAs, with post hoc interaction contrasts undertaken using pairwise comparisons based on z-tests and p values adjusted using a Bonferroni correction; prior behavior was included in each model as a covariate. ANCOVA models were run separately for the primary target audience, women of childbearing age, and again for secondary target audiences, which included older women and the general male adult population. This distinction was necessary to provide greater granularity in advertising performance results among women of childbearing age and will ensure that the design of future omega-3 public health advertisements appeal to this primary audience in the first instance before the results of the secondary target audiences are considered. Intercorrelation analyses were undertaken with the whole sample using Spearman’s rank-order correlations. The proposed mediation and moderated mediation models were also undertaken with the whole sample and examined using the PROCESS macro for SPSS (Hayes, 2022), applying Model 4 and Model 8, respectively, with 5,000 bias-corrected bootstrap samples (95 percent confidence interval [CI]).
RESULTS
Participant Characteristics and Randomization Check
A total of 346 participants completed the survey, with 25 not passing the quality control check and thus being excluded from analysis. This resulted in a final sample of 321 participants, with a total of 157 (48.9 percent) identifying as female; 162 (50.5 percent), as male; and two (0.6 percent), as nonbinary. The age of the sample was relatively balanced, with the age groups of 18–24 years accounting for 17.8 percent of the sample; 25–34 years, for 17.1 percent; 35–44 years, for 15.9 percent; 45–54 years, for 15.0 percent; 55–64 years, for 14.0 percent; and 65+ years accounting for 20.2 percent of the sample.
Participants were randomly allocated to one of four experimental conditions: Gain Frame × Emotional Cue, 24.3 percent (n = 78); Gain Frame × Informational Cue, 24.3 percent (n = 78); Loss Frame × Emotional Cue, 24.6 percent (n = 79); and Loss Frame × Informational Cue, 26.8 percent (n = 86). To assess whether the sample data were equally distributed at baseline, a randomization check was performed, confirming that each condition did not differ significantly with respect to age, χ2(15) = 12.465, p = .644; or gender, χ2(3) =.52, p = .914; or on the basis of language, education, need for cognition, and regulatory focus, with no significant differences detected (See Table 1).
Manipulation Check
A manipulation check was performed to examine whether each experimental condition was manipulated as intended. Respondents in the gain-frame condition reported higher positive ratings (M = 1.58, SD = 1.09) than loss-frame respondents (M = 0.75, SD = 1.43), a mean difference of 0.84 scale points (95 percent CI, 0.56 to 1.19), t = 5.88, p < .001. Respondents in the informational cue condition reported higher informational ratings (M = 1.51, SD = 1.12) than respondents in the emotional cue condition (M = 0.61, SD = 1.47), a mean difference of 0.90 scale points (95 percent CI, −1.19 to −0.61), t = −6.199, p < .001.
Descriptive Statistics
Table 1 outlines the highest level of education attained by respondents, highlighting that the sample is relatively diverse in terms of education level, in line with the general Australian population, with the majority having completed further studies, including a trade, certificate, apprenticeship, or diploma (28.3 percent). Tertiary qualifications also accounted for a large portion, with 25.2 percent of respondents indicating that they held a tertiary (undergraduate) qualification and 21.5 percent indicating that they held a tertiary (postgraduate) qualification. Just under a quarter of respondents (23.1 percent) indicated that secondary school was their highest level of education, whereas only six respondents noted primary school as their highest level of education (1.9 percent).
Most respondents (95.3 percent) had heard of premature birth, although only 22.7 percent correctly defined premature birth as “birth before 37 weeks of gestation.” Response options ranged from “before 32 weeks” to “birth before 42 weeks,” and the option “a baby born with a low birthweight,” as per de Seymour, Simmonds, Gould, et al., 2019. Knowledge of omega-3 was highly prevalent, with 94.7 percent indicating that they had heard of omega-3 fatty acids, although only 14.6 percent noted having (or knew of someone having) an omega-3 test during pregnancy (See Table 2). These results align with the findings of de Seymour et al.’s survey conducted in 2019, where 84 percent of the population had heard of preterm birth. The authors of the current study, however, observed a slightly lower proportion of individuals correctly defining premature birth (37.9 percent in 2019 versus 22.7 percent in 2021). Likewise, in the 2019 survey, 90.6 percent of respondents indicated that they had heard of omega-3 fatty acids.
Message Frame and Visual Cue: Primary and Secondary Target Audiences
Analysis among the primary target audience (n = 85) revealed a statistically significant interaction effect between message frame and visual cue on respondent recall, F(1, 80) = 4.04, p = .048 (See Table 3). Post hoc interaction contrasts revealed that advertisements that used a gain-frame message with an emotional cue produced significantly greater recall than gain-frame messages that used an informational cue, with a statistically significant mean difference of .56 in recall points (95 percent CI, .113 to 1.009), p = .015 (post hoc follow-up for interaction effects not shown in a table). A statistically significant main effect of message frame on advertisement key message comprehension was also detected, F(1, 80) = 4.93, p = .029; with gain-frame messages achieving greater message comprehension scores than loss-frame messages. The adjusted marginal mean and standard error for message comprehension scores were 1.38 ± 0.07 for gain-frame messages and 1.61 ± 0.07 for loss-frame messages, a statistically significant mean difference of −.23 (95 percent CI, −.44 to .24), p = .029 (main effects not shown in a table). No further interaction or main effects were observed for this group.
Analysis among the secondary target audiences (n = 236) revealed no significant interaction effects between the message frame and visual cue (See Table 3) or any main effects (not shown in a table) on all measures of behavioral intentions, attitudes, and awareness (all ps > .05). As such, these results indicate that neither condition was more effective with respect to respondents’ behavioral intentions, attitudes, and awareness in this audience. It is important to note, nevertheless, that the mean intention to screen for omega-3 after viewing the advertisements was high across all conditions for all audiences (M = 4.08; SD ± 0.94), suggesting that, on average, most people responded positively to the advertisement and intended to get an omega-3 test, regardless of which advertisement they saw. All analyses controlled for prior behavior, suggesting that these results are without regard to whether respondents had previously undertaken omega-3 testing.
Intercorrelations between Behavioral Intentions and Attitudinal and Awareness
A series of Spearman’s rank-order correlations were run to assess the relationship between the primary outcome of interest, behavioral intentions to screen for omega-3 on attitudinal and awareness outcomes. A visual inspection of a scatterplot revealed the relationship to be monotonic. There was a statistically significant correlation between behavioral intentions and attitudinal and awareness measures (See Table 4). Notably, there was a positive correlation between behavioral intentions with advertising attitudes, r(319) = .603, p < .001; advertising message effectiveness, r(319) = .611, p < .001; and advertisement free recall, r(319) = .401, p < .001; highlighting the finding that as respondents’ attitudes, perceptions of advertising message effectiveness, and advertising free recall scores improved, so too did their likelihood to engage in omega-3 screening. Similarly, there was a positive correlation between behavioral intentions with advertising mood, r(319) = −.276, p < .001; and advertisement key message comprehension, r(319) = −.324, p < .001; whereby as respondents’ mood grew increasingly “bad,” “unpleasant,” and “sad,” and advertising key message comprehension difficulty increased, respondents’ behavioral intentions to screen for omega-3 decreased. A summary of further intercorrelation scores by experimental condition is highlighted (See Table 4).
Mediation and Moderated Mediation Analyses
Simple mediation analyses were undertaken using Hayes PROCESS Model 4 (Hayes, 2022), with processing fluency as a mediator. First, the effect of the mediator, processing fluency, on all advertising outcomes—b path—was assessed. A significant interaction effect was observed, suggesting that, regardless of advertising condition, processing fluency plays an influential role in respondents’ behavioral intentions, attitudes, and awareness. Next, whether processing fluency mediates the relationship between advertising condition (message frame or visual cue) and advertising outcomes (intentions, attitudes, and awareness) was assessed. As highlighted in Tables 5 and 6, there was no evidence of a mediation effect.
To further explore whether the relationship between advertising condition and advertising outcomes, by means of the indirect path of processing fluency, is contingent upon regulatory focus or need for cognition, conditional process modeling was undertaken by means of moderated mediation using Hayes PROCESS Model 8 (Hayes, 2022). The overall index of moderated mediation for each outcome variable was not significant; accordingly, there was no evidence of any moderated mediation (See Tables 5 and 6).
DISCUSSION
Advertising Appeals and Visual Cues
In Research Question 1, the authors sought to explore whether the message frame or visual cue influenced respondents’ behavioral intentions, attitudes, and awareness. Both awareness measures—free recall and message comprehension—differed among the primary target audience, women of childbearing age. Advertising that used a combination of gain-frame messages using an emotional cue produced significantly greater free recall scores. Similarities exist between the findings of this study and recent research by Ghosh, Sreejesh, and Dwivedi (2022), who reported greater recall of logos over brand names in video game advertising. These results validate the concept of a picture-superiority effect, which refers to better recall of visual stimuli that was due to deeper cognitive processing and encoding. Emotional cues may be more effective than highly informational cues such as screening forms for producing a picture-superiority effect and, in turn, perhaps eliciting greater recall. Additionally, women exposed to gain-framed messages had better message comprehension. Prior studies have ordinarily explored these results within the context of behavioral intentions and attitudes (Detweiler, Bedell, Salovey, et al., 1999; Meyerowitz and Chaiken, 1987; Rothman and Salovey, 1997), but few studies have explored the application of message framing or visual cues on recall, although such studies failed to demonstrate any meaningful differences (Meyerowitz and Chaiken, 1987; Mollen, Engelen, Kessels, and van den Putte, 2017). The current results seem plausible, nevertheless, and perhaps corroborate an earlier study’s findings that reported a gain-frame advantage for behavioral intentions primarily among highly involved participants (Rothman et al., 1993).
Conversely, behavioral, attitudinal, and awareness measures were high among the general target audience, regardless of the advertising condition. Notably, respondents exhibited a mean behavioral intention to screen score of 4.05 out of 5, slightly higher than the mean intention to take an omega-3 fatty acid supplement score, as reported in de Seymour et al.’s (2019) research (3.68 out of 5). Although intentions to screen and intentions to supplement for omega-3 are distinct measures, they provide a meaningful basis for comparing the findings of the two studies. These practical differences highlight the efficacy of advertising techniques in promoting positive intentions and desired actions. By demonstrating the effectiveness of advertising interventions in fostering intentions to screen for omega-3, the current study further supports the potential impact of such interventions in encouraging beneficial health practices such as adhering to screening and supplementation advice.
Behavioral–Attitudinal Paradigm
Research Question 2 explored the relationship between the primary outcome of interest, behavioral intentions to screen for omega-3 on attitudinal and awareness measures among the general target audience. The findings suggested a dose-response relationship between measures, with higher scores in respondents’ attitudes, perceived message effectiveness, and free recall related to greater intentions to screen for omega-3. This finding was also evident for respondents’ mood and message comprehension, where a decline in mood and message comprehension mirrored a decline in behavioral intentions.
The theory of reasoned action may begin to explain these findings, highlighting a behavioral–attitudinal paradigm in play. According to this theory, a person’s intention to engage in any given behavior is, in part, a function of one’s attitudes (Fishbein and Ajzen, 1975), with several prior studies demonstrating this idea across various health domains (Stead, Tagg, MacKintosh and Eadie, 2004), including health screening (Rutter, 2000), nutrition (Dennison and Shepherd, 1995), gambling (Bouguettaya, Lynott, Carter, et al., 2020), smoking (Norman, Conner, and Bell, 1999), and alcohol and drug consumption (Marcoux and Shope, 1997).
It is worthwhile considering other underpinning concepts of the theory of reasoned action and the theory of planned behavior, which could also be relevant in further exploration of the influence of behavioral intentions to screen for omega-3, aside from one’s own attitudes. The potential influence of social norms, for instance, may have played a significant mediating or moderating role in motivating individuals to comply with omega-3 screening advice. Subjective norms represent an individual’s perception of social pressure, and perhaps exploring how the attitudes and opinions of close family members or relevant others—in this case, the secondary target audience—might affect the primary target audience’s intentions to undergo omega-3 testing may be relevant when designing appropriate messaging stimuli to drive greater intentions (Ajzen, 1991; Ajzen and Fishbein, 1980).
Regardless of the mechanisms used within advertisements, making the advertisements easier to process may enhance their overall effectiveness.
Likewise, one’s perceived behavioral control, which refers to an individual’s belief in their ability to perform a behavior, could certainly be a factor in assessing how individuals perceive their control over omega-3 testing intentions. Considering factors such as self-efficacy or perceived barriers could provide insights into the relationship between perceived control and intentions to screen for omega-3 to further develop appropriate messaging strategies for the target audiences (Ajzen, 2002). Taken together, the current findings broadly lend support to Ajzen’s theory, highlighting its usefulness in explaining an individual’s greater intentions to screen for omega-3 as a function of greater attitudes and awareness toward omega-3 advertisements.
Influence of Processing Fluency
A significant direct effect of processing fluency on behavioral intentions and attitudinal and awareness outcomes was observed. This finding suggests that, regardless of the mechanisms used within advertisements (e.g., message frame or visual cue), making the advertisements easier to process may, in turn, enhance their overall effectiveness. These findings are consistent with that of Li (2021), who explored the role of processing fluency and regulatory focus within the context of narrative versus nonnarrative advertising in excess sugar consumption.
Next, Research Questions 3 and 4 explored the proposed mediation of processing fluency and moderated mediation of regulatory focus and need for cognition. Results indicated no mediation or moderated mediation pathways to exist between measures, suggesting no underlying mechanism in how respondents may process advertising stimuli on the basis of advertising conditions. These results may be explained by the fact that advertisements scored relatively well in terms of processing fluency, regardless of the advertising condition being manipulated (processing fluency: M = 2.77, SD = 1.39). Li, too, demonstrated no evidence of a moderated mediation effect among advertisement form and processing fluency, yet a significant direct effect between processing fluency and behavioral intentions to lower sugar intake was found (Li, 2021).
Limitations and Future Research
This study has some limitations that affect the generalizability of its findings. For one, the research was conducted under artificially controlled conditions rather than real-world market advertising conditions. The authors of the study, however, used a random-sampling approach from a well-defined population group, with balanced respondent demographics for increased statistical validity. The experimental design and survey flow also allow for strong causal inferences and protect against biases or priming effects and used a robust factorial design, allowing for the exploration of several interrelated advertising factors simultaneously and efficiently (Sedgwick and Greenwood, 2015). Although further research is needed to test the effectiveness of the advertising stimuli in real-world settings, pretesting is an important initial step in developing evidence-based health promotion advertising campaigns. The study’s scope is also limited to examining a few advertising factors related to an omega-3 test-and-treat program, and future research could investigate other factors such as advertising appeal, delivery medium, or audience temporal orientation. Future research could also include the potential mediating or moderating effects of concepts posited by the theories of reasoned action and planned behavior in its design, to gain deeper insights into their influence in promoting omega-3 screening.
Theoretical and Managerial Implications
Given its limitations and strengths, the current study makes several theoretical and managerial contributions. First, this research contributes to the growing body of literature extending the concepts of processing fluency, regulatory focus, and need for cognition within the context of precision nutrition, and maternal health promotion. In particular, this study contributes additional evidence to suggest the role of processing fluency in enhancing advertising effectiveness (Li, 2021). Further research is needed to understand the specific mechanisms that influence processing fluency to draw meaningful implications to enable the development of future health campaigns. This study suggests neither mediation nor moderated mediation pathways exist between advertising measures of processing fluency, regulatory focus, and need for cognition.
From a practical standpoint, the findings of this study will inform the development of an evidence-based segmentation strategy for an omega-3 marketing campaign to be disseminated on mass media platforms. The results suggest that highly tailored advertisements that are based on consumer archetype or orientation may not be necessary; instead, marketers can utilize basic demographic segmentation, which is more cost-efficient and more timely (Sharp, 2017).
The study suggests that using gain-frame messages and emotional visual cues can encourage pregnant people and their families to undertake omega-3 screening. This finding has important managerial implications. By pretesting and incorporating these messaging strategies into promotional campaigns, advertising practitioners can create more effective messages that will resonate better with the target audience. This, in turn, can lead to greater engagement with the campaign and a positive impact on behavioral intentions, ultimately resulting in improved public health outcomes. These findings, moreover, have broader implications for health-related advertising campaigns more generally, particularly when it comes to promoting prevention-based behaviors. Advertisers can use the messaging strategies identified in this study to create effective campaigns that encourage healthy behaviors and improve public health outcomes beyond omega-3 screening. By tailoring their messaging to resonate with their target audience, advertisers can increase engagement and encourage positive behavioral change.
Finally, this study highlights a behavioral–attitude paradigm, which suggests that positive attitudes toward omega-3 advertisements are associated with greater intentions to screen for omega-3. This has important managerial implications for advertising practitioners, emphasizing the need to develop advertisements that elicit favorable attitudes to drive greater behavioral intentions toward a given health behavior. The findings can inform the design of more effective health promotion advertising campaigns by prioritizing the development of messaging strategies that positively influence attitudes toward health behaviors. By doing so, advertisers can increase the effectiveness of their campaigns and ultimately improve public health outcomes.
Taken together, these results are important in highlighting the value of pretesting: They will ensure that marketing messages aimed at encouraging the uptake of omega-3 screening during pregnancy are well accepted and inclusive of primary and secondary target audience groups and that they facilitate inclusion and awareness to allow pregnant people to make an informed decision about their health care. More broadly, the study’s findings have wider implications for informing the development of comprehensive strategies beyond a mass media campaign. Policymakers can utilize these insights to create educational materials and communication collateral that offer clear and accurate information on the importance of health screening, its benefits, and practical steps for accessing screening services for consumers. Strategies also can be developed to enhance health care professionals’ knowledge and communication skills, particularly among general practitioners, enabling them to effectively discuss and promote health screening among their patients. These strategies can leverage the use of positive messaging techniques and provide clearer informational collateral that is easy to process, ensuring effective communication and understanding and ultimately strengthening public health approaches.
ABOUT THE AUTHORS
Celine Northcott is a research fellow at the South Australian Health and Medical Research Institute (SAHMRI) and PhD candidate in medicine (pediatrics) at the University of Adelaide. Her predominant focus is within implementation science, with a cross-disciplinary academic background in business marketing and population health practice. At SAHMRI, Northcott leads such initiatives as the state-wide Omega-3 Test-and-Treat Program to prevent preterm births.
Philippa Middleton is a professor who co-leads the Pregnancy and Perinatal Care Group at SAHMRI and is affiliated with the University of Adelaide. She is a National Health and Medical Research Council fellow. As a perinatal epidemiologist and implementation scientist, Middleton specializes in research on preterm birth, nutrition, stillbirth, and diabetes in pregnancy, using randomized controlled trials, systematic reviews, and research synthesis (quantitative and qualitative). She has published in journals such as The Lancet, BMC Pregnancy and Childbirth, PLoS Medicine, and BJOG: An International Journal of Obstetrics & Gynaecology.
Maria Makrides is executive director of SAHMRI and a National Health and Medical Research Council leadership fellow, is a clinical nutritionist committed to enhancing the nutrition and health of mothers and their babies through research. Her work has influenced international policy on infant foods, pregnancy diet, and infant feeding guidelines. She has published in journals such as the New England Journal of Medicine, The Lancet, Journal of the American Medical Association, and American Journal of Clinical Nutrition.
Lucy Simmonds is an assistant professor of marketing and health care management at Flinders University. Her research focuses on advertising and consumer behavior particularly in health settings. Simmonds’ industry background includes roles at the Australian Science Media Centre and the Department of Health. She serves on editorial review boards for various journals and publishes in marketing and health sciences, including Journal of Business Research, Psychology & Marketing, and JAR.
APPENDIX Supplementary Text
DEPENDENT VARIABLES
Behavioral Intentions
Intentions to undertake omega-3 screening were the primary outcome of this study and were assessed using a single-item scale, adapted from van den Berg, Timmermans, Knol, et al. (2008), who assessed intentions to undergo prenatal screening. Respondents were asked, “if you or your partner became pregnant, would you intend to have the omega-3 screening test done?” on a 5-point unipolar Likert scale with items anchored from (1) “absolutely not” to (5) “absolutely.”
Attitudes
Attitudes toward the public health communication were assessed using a single-item scale adapted from Rossiter and Bergkvist (2009) on advertising attitudes. Respondents were asked “how would you rate the public health communication?” on a 7-point bipolar Likert scale with responses anchored from (−3) dislike to (+3) like.
Mood
Mood was measured on a three-item scale, adapted from Wegener, Petty, and Smith (1995), with respondents asked to rate how the public health communication made them feel on a 9-point unipolar Likert scale with items anchored from (1) “good,” “pleasant,” and “happy” to (9) “bad,” “unpleasant,” and “sad” respectively. A mean mood score was then calculated using the three items to provide a composite measure of overall mood, with lower scores indicating a more positive mood. Internal consistency for this scale was excellent (Cronbach’s α = .94) (George and Mallery, 2003).
Perceived Advertisement Message Effectiveness
Perceived advertisement message effectiveness was measured on a three-item scale, adapted from Jensen et al. (2002) and Fishbein, Hall-Jamieson, Zimmer, et al. (2012), where respondents were asked to rate the following items on a 4-point unipolar Likert scale, item 1: “the message was convincing,” item 2: “people your age who are pregnant or planning to become pregnant are more likely to get screened after seeing the advertisement,” and item 3: “the advertisement would be helpful in convincing your friends to be screened for omega-3 levels.” Response options were (1) “definitely not,” (2) “no,” (3) “yes,” and (4) “definitely yes,” with a mean score of the four items generated to indicate a perceived level of ad message effectiveness. Higher scores indicated a greater level of perceived advertisement message effectiveness. Internal consistency for this scale was considered good (Cronbach’s α = .81), as per George and Mallery (2003).
Advertisement Free Recall
Advertisement free recall (unaided recall), an awareness measure, was measured as per Cacioppo and Petty (1979). Respondents were asked to list as many of the arguments they recalled from the public health communication as possible. A total recall score was given for each respondent by scoring their listed open-ended arguments against a set of four predefined answers for correctness. Scores were summed to create a total recall measure out of four. Two coders independently assessed recall scores, with disagreements resolved by a third independent coder. Intraclass correlation coefficient (ICC) estimates were calculated using a two-way mixed-effects model (single measure) to examine intra-coder reliability. ICC was 0.99, indicative of excellent reliability (95 percent CI 0.99–0.99) as per Cicchetti and Sparrow (1981).
Advertisement Key Message Comprehension
Advertisement key message comprehension, another awareness measure, was measured by coding correct responses to the question “what is the main message that you think the public health communication was trying to convey?” adapted from McElrath and Wakefield, Ruel, et al., 2005. Open-ended responses were coded and dichotomised as (1) “correct” or (2) “incorrect” based on whether respondents correctly identified the key argument of the public health communication against a predefined answer. Two coders independently assessed advertisement key message comprehension, with disagreements again resolved by a third independent coder. ICC was 0.89 (95 percent CI 0.86–0.91) and is considered to have good reliability (Cicchetti and Sparrow, 1981).
MEDIATOR
Processing Fluency
Processing fluency was assessed using a four-item scale, as per Kostyk, Leonhardt, and Niculescu (2021), with respondents asked to rate their level of agreement with the following items on a 7-point unipolar Likert scale, item 1: “the public health communication was difficult to process,” item 2: “the public health communication was difficult to read,” item 3: “the public health communication takes a long time to process,” and item 4: “the public health communication was difficult to understand.” Response options were anchored from (1) “strongly disagree” to (7) “strongly agree.” A mean processing fluency score was then calculated using all four items, providing a composite measure of overall processing fluency. Greater scores indicated a greater degree of difficulty in fluently processing advertising stimuli. Internal consistency for this scale was excellent (Cronbach’s α = .93) (George and Mallery, 2003).
MODERATORS
Regulatory Focus
Regulatory focus was assessed using a 10-item scale, comprising five promotion and five prevention-focused items, with respondents asked to rate their level of agreement with each item on a seven-point unipolar Likert scale as per Haws, Dholakia, and Bearden (2010). Promotion-focused items included item 1: “when it comes to achieving things that are important to me, I find that I don’t perform as well as I would ideally like to do,” item 2: “I feel like I have made progress toward being successful in my life,” item 3: “when I see an opportunity for something I like, I get excited right away,” item 4: “I frequently imagine how I will achieve my hopes and aspirations,” and item 5: “I see myself as someone who is primarily striving to reach my ‘ideal self’—to fulfill my hopes, wishes, and aspirations.” Prevention-focused items included item 1: “I usually obeyed rules and regulations that were established by my parents,” item 2: “not being careful enough has gotten me into trouble at times,” item 3: “I worry about making mistakes,” item 4: “I frequently think about how I can prevent failures in my life,” and item 5: “I see myself as someone who is primarily striving to become the self I ‘ought’ to be—fulfill my duties, responsibilities and obligations.” Negatively worded items (promotion focus item 1 and prevention focus item 2) were then reverse-coded. Internal consistency for the regulatory focus scale was acceptable (Cronbach’s α = .70) (George and Mallery, 2003). A mean composite score for both promotion- and prevention-focused items was then generated, and data were transformed in order to create a measure of dominant regulatory focus. Responses were dichotomized as either (1) “promotion centric” or (2) “prevention-centric” regulatory focus on the basis of a median split after subtracting prevention-focused scores from promotion-focused scores (Lee and Koo, 2012).
Need for Cognition
Need for cognition was assessed using a six-item scale as per Lins de Holanda Coelho, Hanel, and Wolf (2020), with respondents asked to rate their level of agreement with the following items on a 5-point unipolar Likert scale, item 1: “I would prefer complex to simple problems,” item 2: “I like to have the responsibility of handling a situation that requires a lot of thinking,” item 3: “thinking is not my idea of fun,” item 4: “I would rather do something that requires little thought than something that is sure to challenge my thinking abilities,” item 5: “I really enjoy a task that involves coming up with new solutions to problems,” and item 6: “I would prefer a task that is intellectual, difficult, and important to one that is somewhat important but does not require much thought.” Negatively worded items 3 and 4 were then reverse-coded before generating a mean composite measure of need for cognition using all six items. Internal consistency for this scale was acceptable (Cronbach’s α = .77) (George and Mallery, 2003).
- Received April 19, 2023.
- Received (in revised form) September 28, 2023.
- Accepted October 3, 2023.
- Copyright © 2024 ARF. All rights reserved.
REFERENCES
ARF MEMBERS
If you are a member of the Advertising Research Foundation, you can access the content by logging in here
Log In
Pay Per Article - You may access this article (from the computer you are currently using) for 30 days for US$20.00
Regain Access - You can regain access to a recent Pay per Article purchase if your access period has not yet expired.