When considering content, it is worthwhile to address the phenomenon of distracting imagery during the presentation of risk information in pharmaceutical advertisements and how it might violate FDA rules. Sullivan et al. (2017) studied distracting imagery's effects by creating hypothetical DTCAs with low-distraction or high-distraction features and showing them to participants during the presentation of the major statement. In the low-distraction hypothetical DTCA, a male narrator read the risk information as he stood still and spoke to the camera directly. The high-distraction ad mimicked the format of actual DTCAs, with an emergent or ongoing narrative presented during communication of the risk information. The authors used eye-tracking technology and found that the distracting elements of the ad (the visual narrative displayed in the background) drew attention away from the risk information, even when it was presented in dual modality (audio and text). In addition, questionnaire responses indicated that the distracting elements reduced the viewers' retention of the drug's risks and that they paid more attention to the narrative than the risk information. Similarly, FDA researchers confirmed that distractions during the presentation of risk information affect consumers' attention and recall of the side effects U.S. Food and Drug Administration, (2023).

Although DTCAs are pervasive and there has been notable research on their effects, research that has focused solely on narratives in DTCAs is still emerging. Ball and Applequist (2019) were among the first to examine narratives in broadcast DTCAs and look exclusively at the way information was conveyed through first-person, second-person, or third-person perspectives. Doing so allowed them to identify specific styles of narratives (classic drama or vignette). They categorized classic drama as one character (or set of characters) that acts out a single continuous story, while vignettes include multiple actors with individual stories that culminate in multiple storylines. Although it was not their goal, they determined that all DTCAs include some type of narrative. Other research on narratives in DTCAs has focused specifically on transportation and identification, which are components of narrative persuasion. Montgomery (2019) confirmed the findings of previous research that all DTCAs include some form of narrative, and also found that the majority of narratives are presented visually with ads that begin with a healthy exemplar (implying positive results of using the medication). Delving into the theoretical framework of the use of the narrative underpins its use in DTCAs and lays a foundation for this research's focus on analyzing the content of health messages in DTCAs.

Visual-verbal redundancy (VVR) is a form of vivid cue that warrants individual attention. This cue involves communicating the same information via two modes of delivery: text on the screen is spoken verbatim simultaneously. This draws attention to and highlights the information that is being communicated. Work on VVR has illuminated its positive effect on memory retention, learning, and comprehension (Adesope and Nesbit, 2014). Interestingly, it has been shown to decrease cognitive load and perhaps serves as a heuristic cue (e.g., see Albers et al., 2023; Russell et al., 2018; Trypke et al., 2023). There is evidence that when the visual images and verbal text are contradictory, individuals defer to the visual image for understanding. For example, Grabe (2020) pointed out the contradiction and preference to dismiss side-effect information in DTCAs when visual images of smiling, healthy people were presented at the same time. She coined the contradiction “non-redundancy.” However, DTCAs use narratives as a persuasive mechanism and VVR use is not limited to the presentation of side effects. Positive information about the drug is also conveyed through VVR as it is a vivid way to attract a viewer's attention (Montgomery, 2019). Although it has not been studied for its transportability, it is an integral part of each advertisement. DTCAs use VVR during the presentation of the major statement (as FDA regulations require) and at other times throughout the advertisement. When VVR is positioned as a vivid visual cue presented during a story, it is important to understand the way that it intersects with the ongoing narrative, particularly given VVR's propensity to increase learning, comprehension, and retention.

To address the specific ways in which VVR is used in DCTAs, the following research question is posed: RQ3: In what ways is visual-verbal redundancy used during an ongoing visual narrative in DTCAs?

First, the top 10 network and cable programs in March of the 2016 and 2018 seasons were identified and recorded according to Nielsen ratings reports. The programs chosen were: The Big Bang Theory, The Voice (Monday and Tuesday nights), NCIS, Little Big Shots, Empire, Dancing with the Stars, Blue Bloods, 60 Min, The Good Doctor, Grey's Anatomy, Young Sheldon, Bull, and NCIS New Orleans. The DTCAs from each recording were isolated and yielded a total of 62. Of these, 10 ads were duplicates, which left 52 remaining for analysis. However, after the COVID-19 pandemic, three of the ads were edited to include COVID-19 information, and thus, they were removed from the dataset as well. The ads were saved on a computer and then uploaded to NVIVO, which is ideal because of its ability to organize and code large sets of videos at specific time points. Emergent themes were documented and each code's frequency and percentage were calculated.

The coding methodology was used to address the following research question:

RQ1: What message(s) about health are communicated in a DTCA's visual stories? To answer this question, the exemplars' visual health in the 49 DTCAs was assessed. They were identified as completely healthy or symptomatic at the beginning of the ad and healthy by the end. The sample ads were watched and data were recorded in an Excel spreadsheet, which was then imported into an NVIVO project. The information recorded about the ad included the title, a summary of the visual story, the narrative type, the activity in which the exemplar was engaged, and the exemplar's health condition during the narrative. The advertisement title was listed first. The advertiser creates the advertisement titles, which include the name of the medicine advertised and a general title that reflects the content in the ad (e.g., Cosentyx See Me; Cosentyx is a medicine that treats psoriasis). Then, the ads were summarized and key information was identified. The narrative type was categorized as “classic drama” or “vignette” and the exemplar's observable health was identified as healthy (showed no health symptoms related to the condition that the medication treats—asymptomatic) or restored health (symptomatic at the beginning of the ad, but asymptomatic as the ad progressed). In vignette advertisements, the health condition was categorized according to the condition(s) represented in the ad: healthy if none of the exemplars was symptomatic, restored health if all of the exemplars were symptomatic at the beginning and their health was restored by the end of the ad. Wellness and restored health were categorized if some exemplars were asymptomatic and some were symptomatic. Trained coders coded the setting and exemplar's actions.

Realism and vividness were coded to answer RQ2: What transportation cues are used in the visual stories in DTCAs?

The next step was segmenting the individual scenes in the sample DTCAs. Wang et al. (2000) explained that multimedia text is complex and must be segmented into scenes so that each “corresponds to a story unit” (p. 12). A scene was identified in this research by an exemplar's (may include other actors) continuous movement through a particular setting during a singular event. There were multiple scenes in each of the DTCAs (687 scenes overall), and isolating each of them allowed for an in-depth examination of each unit of analysis. Each advertisement was saved in a folder on the computer and then imported into a project in NVIVO. After all of the advertisements were imported to a file entitled “video files for visual information,” the timespans for each scene were created. NVIVO has a preset column for timespan, content, and speaker; editing functions to create timespans allowed each scene to be marked at its beginning and end. Some of the scenes overlapped, and in those cases, it was necessary to adjust the timestamp manually. Such precise measurements were important so that data would be coded to the correct scene.

Codes were developed using an a priori deductive approach, and were derived from research in the literature that discussed the components of transportation in narratives, including realism, vivid imagery, and vivid sensory cues (see Table 1). After completing the framework for the codes, five ads were selected randomly to test the framework. After several iterations of those five ads were observed, the emergent codes were tested on five additional ads, so that all codes were applied to a total of 10 ads. During the sample coding, it became apparent that typicality and novelty were redundant, and although realism includes typicality and novelty, only novelty was used. Novelty is defined as a divergence from expectations that is interesting and unanticipated or presents information in an unexpected way. Therefore, novelty codes were determined by visual information that was unique, unanticipated, engaging, and interesting. For example, showing a split screen with the sick person on one side in bed as they watch their family eat out with friends at a restaurant on the other would be considered a novel approach. As defined in the literature, novelty appears to be subjective. There are no clearly defined steps in the literature to determine novelty in an image or video. However, coders were instructed to code scenes that appeared unique, interesting, innovative, engaging, surprising, or unanticipated. The data coded as novel were downloaded and described in a Word document, and then categorized.

Vivid cues are images that appeal to the senses or images used in such a way that they promote a clear image or powerful feeling. Vivid transportation cues are identified by information that highlights sensory information or those with bright colors, intense or graphic imagery, and close-ups. Sensory appeals are defined as images that appeal to the sense of sight, touch, taste, smell, sound, balance (body orientation), temperature, or pain specific to the exemplar. Only sensory information that was highlighted obviously to draw attention, awareness, or focus was coded. For example, squinting eyes, rubbing the back of one's neck, eating food in which the emphasis was placed on eating, shivering from the temperature, an exemplar on a seesaw evoking a feeling of balance, hands on head as if in pain, or any combination of these. Temperature is a type of touch sensation, but was considered separately in this research because of its association with diminished health (fever). Other vivid images that do not necessarily appeal directly to the senses, but are powerful and emphasized nonetheless, include a dog or cat shaped like the words “sleep” or “awake,” charts, graphs, simulated drawings, and information on a split screen with an exemplar, or the lose-up of the product or exemplar, or emphasis placed on either (see Table 1 for an overview of cues).

The following a priori coding framework was used to answer RQ2: What transportation cues are used in the visual stories in DTCAs?

A detailed codebook was created and definitions were provided for each code and subcode (see Table 2). The researcher trained two coders, and their coding analysis was compared to the researcher's. Their questions were answered and they were set up with an NVIVO account, and given access to the DTCAs collected. To assess interrater reliability, the same 10 ads were uploaded (20% of the entire dataset) to each NVIVO account for collaborative coding. Once the 10 ads were coded and were over 80% consistent, the remaining ads were distributed evenly for coding. The literature provides generalities only about novelty: unique; interesting; engaging, and innovative. Therefore, scenes coded as novel were isolated and an explanation of the novel imagery was recorded separately in a Word document to be categorized.

All VVR was coded to answer RQ3: In what ways is visual verbal redundancy used during an ongoing visual narrative in DTCAs?

The file “Video files for visual information” created in NVIVO for RQ2 was copied and entitled “Video files with transcription.” It was necessary to have a separate file with the transcription to eliminate confusion between codes and narration. The scenes were transcribed manually and each scene was watched as many times as necessary to ensure accuracy. All 49 advertisements contained VVR, but only nine advertisements also used VVR during the presentation of risk information. Because guidelines related to the dual modality in the presentation of side effects changed recently, only the advertisements that included VVR with the side effects were analyzed, and all of the information communicated by VVR in those ads was included.

In the NVIVO file with transcription, coders watched each scene, and highlighted and coded words that were spoken and appeared on the screen simultaneously. VVR from every ad was then transferred to a table in Word with four columns: the advertisement title; the narrated text (VVR); vividness of text in the VVR, and a still image of the VVR. The text's vividness was previously assessed in the codes for vivid images in RQ2 (i.e., colorful, bright, large, or close-up) and used to answer RQ3. Still images were snipped with a snipping tool the moment the narrator began to speak a word, or the moment that the entire phrase was spoken when there was more than one word.

All of the advertisements used visual narratives with clear messages that ended in wellness and health. The sample contained 144 exemplars: 13% (19) exemplars in classic dramas and 87% (125) in 30 vignettes. The number of exemplars in the vignettes varied; the largest number of exemplars featured in a vignette was 8 and the fewest 2. Of the 144 exemplars, 74% (107) of them were asymptomatic and 37 began the ad symptomatic, but were asymptomatic by the end of the advertisement. The majority of the ads in this sample were vignettes with multiple narratives. Only 41% of the ads showed individual(s) whose health was restored after the intervention of the medicine (see Figure 1), and the remaining 59% began with a healthy and happy exemplar (see Gilead Sciences, Inc., 2018, 00m:01s). The latter strategy compels the viewer to infer that the medicine has been approved, consumed, and has worked.

The majority of the ads, 30 (61%), were vignettes with multiple exemplars and stories all of which culminated in a unifying message that the medicine restores health. For example, Entyvio, a medication for Crohn's disease and ulcerative colitis, features several exemplars who are unable to participate in social events. The ad begins with different exemplars looking at their calendar/notes for an event they are uncertain they will be able to attend.

New exemplars are introduced in the advertisement that includes the scene of a woman visiting her doctor. The doctor's visit is the catalyst in the ad and all stories after that point show happy, energized, and socially involved exemplars.

The first scene after the doctor's visit shows a man sitting on a park bench looking at his phone. What he is looking at is not visible, but as his son (assumed) walks up behind him, it shows that he is seen purchasing tickets. The scene changes to the man and boy walking up the street, then toward the theater. Other exemplars appear in the remainder of the ad who are smiling and looking up or forward (see examples in Table 3).

Other visual narratives featured only one exemplar (37%) navigating a setting through an event or series of events (see Table 4). The exemplars in classic dramas are engaged in multiple activities throughout their day and 84% (16 of 19) engage in some sort of social activity with either family or friends.

There were 794 transportation cues coded. They included realism that identified actual patients, scenarios that appear unscripted, recognizable figures, interesting and innovative imagery, vivid colors, close-up shots, taboo glimpses into private moments, and a preponderance of intentional touch.

The visual stories in these advertisements predominantly communicated positive information about using the drug advertised. First, the majority of advertisements used multiple exemplars, which contributed to a collective perception of the drug's effectiveness. Moreover, all of the advertisements ended with messages of wellness and the majority of visual narratives began with wellness stories. Thus, the ads began with a healthy and fully functioning exemplar. These wellness stories presented the exemplars going through their daily routines unimpeded by a health issue. Many of the narratives implied that the medicine had been taken at some point in the past because the exemplar was shown to have no symptoms. Cohn (2019) explained that viewers use bridging inferences to fill in gaps based upon current, as well as prior, information. Petty and Cacioppo (1986) stated that prompting an individual to make inferences about content is an effective persuasive strategy.

The remainder of the stories about health involved restored health after an intervention with a doctor or medicine. Restored health stories also rely on inference. The chronology in these scenes shows an unhappy, struggling exemplar, while the climax to the stories develops with a doctor's visit or a screen shot with the name of the drug in the scene. Thereafter, the exemplar is no longer symptomatic and is shown living a life unhampered by the woes that they experienced previously. The viewer is once again encouraged to infer that the medicine is the catalyst. These positive visuals create a distraction that uses the cognitive resources necessary to make critical judgments and assessments (see Niederdeppe et al., 2012; Petty and Cacioppo, 1986; Rubinelli et al., 2008; Shapiro and Kim, 2012). They also imply that the medicine is the means to achieve good health. Lastly, the preponderance of positive information throughout indicates a gross imbalance between the benefits vs. risks of using the product, in direct violation of FDA regulations of adequate provision.

Isolating the transportation cues allows the nuanced details included in these ads to be examined closely. Transportation cues such as realism, novelty, and vividness are constructed to captivate viewers' attention. Transportation into the narrative through these persuasive strategies can result in more positive perceptions of the ad and brand, and more favorable emotions about the drug advertised. In addition, the same strategy used to transport viewers into a narrative can result in disengaged critical evaluations and fewer counter-arguments (Ching et al., 2013). There were several clear cues present in the ads analyzed (see Table 9 for cue frequency).

Although VVR includes audio components, it does serve as a transportation cue. The combination of bright colors, large fonts, and strategic placement of the visual text, as well as the audiovisual effects make it a vivid element in an advertisement. The positive vs. risk information is not balanced in the VVR portions of the ads. Information that is both seen and stated audibly increases learning, memory, and comprehension (Adesope and Nesbit, 2014). Therefore, the abundance of positive information in the ad, the repetition of the drug's name in vivid colors or size, contributes to the dominance of alleged positive information over the risks. As new regulations are mandated, future research should expand to all VVR included in DTCAs.