This is an open-access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work, first published in JMIR Public Health and Surveillance, is properly cited. The complete bibliographic information, a link to the original publication on https://publichealth.jmir.org, as well as this copyright and license information must be included.
Short, animated story-based (SAS) videos are a novel and promising strategy for promoting health behaviors. To gain traction as an effective health communication tool, SAS videos must demonstrate their potential to engage a diverse and global audience. In this study, we evaluate engagement with a SAS video about the consumption of added sugars, which is narrated by a child (a nonthreatening character), a mother (a neutral layperson), or a physician (a medical expert).
This study aims to (1) assess whether engagement with the sugar intervention video differs by narrator type (child, mother, physician) and trait proneness to reactance and (2) assess whether the demographic characteristics of the participants (age, gender, education status) are associated with different engagement profiles with the sugar intervention video.
In December 2020, after 4013 participants from the United Kingdom completed our randomized controlled trial, we offered participants assigned to the placebo arms (n=1591, 39.65%) the choice to watch the sugar intervention video (without additional compensation) as posttrial access to treatment. We measured engagement as the time that participants chose to watch the 3.42-minute video and collected data on age, gender, education status, and trait reactance proneness. Using ordinary least squares regression, we quantified the association of the demographic characteristics and trait reactance proneness with the sugar video view time.
Overall, 66.43% (n=1047) of the 1576 participants in the 2 placebo arms voluntarily watched the sugar intervention video. The mean view time was 116.35 (52.4%) of 222 seconds. Results show that view times did not differ by narrator (child, mother, physician) and that older participants (aged 25-59 years, mean = 125.2 seconds) watched the sugar video longer than younger adults (aged 18-25 years, mean = 83.4 seconds). View time remained consistent across education levels. Participants with low trait reactance (mean = 119.3 seconds) watched the intervention video longer than high-trait-reactance participants (mean = 95.3 seconds), although this association did not differ by narrator type.
The majority of participants in our study voluntarily watched more than half of the sugar intervention video, which is a promising finding. Our results suggest that SAS videos may need to be shorter than 2 minutes to engage people who are young or have high trait proneness to reactance. We also found that the choice of narrator (child, mother, or physician) for our video did not significantly affect participant engagement. Future videos, aimed at reaching diverse audiences, could be customized for different age groups, where appropriate.
German Clinical Trials Register DRKS00022340; https://www.drks.de/drks_web/navigate.do?navigationId=trial.HTML&TRIAL_ID=DRKS00022340
RR2-10.2196/25343
Engaging the public in the care and maintenance of their own health constitutes a longstanding challenge for health communicators, health educators, and public health agencies worldwide [
As with all persuasion strategies, optimal engagement with SAS videos may be limited by a motivation to reject the health message—a phenomenon known as reactance [
The Intertwined Process Cognitive-Affective Model of reactance [
In a recent study, we investigated whether a child narrator reduced reactance to a SAS video about the consumption of added sugars [
In this study, we investigate the role of trait reactance proneness and demographic factors in voluntary engagement with the sugar intervention video narrated by the child, the mother, or the physician. The participants (n=1576) are those who were initially randomized to the content placebo (the sunscreen SAS video) or the placebo (the earthquake SAS video) arms in the main trial and who were then offered the intervention video as posttrial access to treatment [
This was a randomized controlled trial (RCT) with posttrial access to the treatment stage [
Flowchart depicting the study methodology.
Both trials (main and post) were hosted and run on the Gorilla platform (Cauldron Science Limited) [
The Gorilla algorithm randomly assigned participants to the 5 arms in the main trial and to the 3 intervention arms in the posttrial stage. Since the recruitment took place on the Prolific platform, it was not possible to identify or link data back to the participants. Participants responded to the survey questions and submitted their responses anonymously through the Gorilla platform. Both the study subjects and the investigators had no knowledge regarding the allocation status of the participants.
All participants underwent a process of informed consent on the Prolific platform. The consent form explained the purpose of the study, the risks and benefits of the research, and how to contact the study investigators. By clicking the link, participants agreed to participate in our study and were redirected to the Gorilla platform, where additional information was given. Participants could leave the research study at any time.
Here, we provide some basic details of the main trial to give context to our posttrial study. Further details of the main trial and its procedures can be found elsewhere [
At the beginning of the main trial, participants were asked to answer demographic questions about their age, gender, and highest educational attainment. Participants were then randomized to the sugar intervention arm, the content placebo arm, or the placebo arm, where they watched a SAS video from start to finish.
The sugar intervention video was narrated in English, with a duration of 3 minutes and 42 seconds. Its aim was to boost knowledge about the health consequences of consuming added sugars [
For this study, participants who were randomized to the content placebo video or placebo video were then given the option to watch the sugar intervention video (posttrial access to treatment). Participants could watch the sugar video or end the study without watching the sugar video. If participants chose to watch the sugar video, they were asked on the next page to click the Play button or click the Finish button at any time to end the survey. The participants were informed that they would not be compensated for the additional time taken to watch the sugar video.
The primary outcome of this study was participant engagement, measured as the total time (in seconds) spent watching the SAS sugar video. We also collected data on the participants’ age, gender, and educational status. We further considered the role of the participants’ propensity toward reactance and its effect on view time. To measure trait reactance proneness, participants answered 11 questions based on the Hong Psychological Reactance Scale [
To quantify the participants’ engagement, we used the graphical experiment builder in Gorilla that records a timestamp whenever a new screen is displayed. In our case, Gorilla registered the moment when the participant reached the instruction screen of the final task as the first timestamp, the moment when they entered the video screen as the second timestamp, and the moment when they ended the experiment as the third timestamp. Gorilla also recorded loading delays of more than 10 seconds.
Participants who spent less than 3 seconds on the video screen were grouped together with participants who did not watch the SAS video. Among participants who chose to watch the SAS video, we quantified the length of time spent watching the sugar video. We defined the dependent variable
We dropped observations that had missing values and performed all statistical analyses using Stata software version 14.2.
Between December 9, 2020, and December 11, 2020, we recruited 4159 participants for the main RCT. The main trial design is shown in
Characteristics of 1576 participants from the United Kingdom, with data on engagement with a short, animated video about added sugars in a web-based RCTa, December 2020.
Demographics | Content placebo arm (790 observations) | Placebo arm (786 observations) | |||||||
|
|
Narrator 1 (child), n (%) | Narrator 2 (mother), |
Narrator 3 (physician), |
Narrator 1 (child), n (%) | Narrator 2 (mother), |
Narrator 3 (physician), |
||
|
|||||||||
|
18-24 | 57 (22.8) | 64 (22.5) | 63(24.7) | 69(25.1) | 68(28.2) | 71 (26.3) | ||
|
25-34 | 83 (33.2) | 93 (32.6) | 80 (31.4) | 89 (32.4) | 74 (30.7) | 85 (31.5) | ||
|
35-44 | 57 (22.8) | 67 (23.5) | 51 (20.0) | 58 (21.1) | 50 (20.7) | 59 (21.8) | ||
|
45-54 | 37 (14.8) | 41 (14.4) | 49 (19.2) | 41 (14.9) | 36 (14.9) | 39 (14.4) | ||
|
55-59 | 16 (6.4) | 20 (7.0) | 12 (4.7) | 18 (6.5) | 13 (5.4) | 16 (5.9) | ||
|
N/Ad | .77 | N/A | N/A | .99 | N/A | |||
|
|||||||||
|
Female | 155 (62.0) | 167 (58.6) | 153 (60.0) | 179 (65.1) | 142 (58.9) | 161 (59.6) | ||
|
Male | 95 (38.0) | 117 (41.0) | 98 (38.4) | 92 (33.4) | 98 (40.7) | 107 (39.6) | ||
|
Other | 0 (0.0) | 1 (0.3) | 4 (1.6) | 4 (1.4) | 1 (0.4) | 2 (0.7) | ||
|
N/A | .19 | N/A | N/A | .31 | N/A | |||
|
|||||||||
|
Primary school | 6 (2.4) | 4 (1.4) | 3 (1.2) | 2 (0.7) | 3 (1.2) | 5 (1.8) | ||
|
High school | 37 (14.8) | 45 (15.8) | 38 (14.9) | 39 (14.2) | 42 (17.4) | 45 (16.7) | ||
|
BA, some college | 155 (62.0) | 176 (61.7) | 166 (65.1) | 184 (66.9) | 152 (63.1) | 160 (59.3) | ||
|
MA/PhD | 52 (20.8) | 60 (21.0) | 48 (18.8) | 50 (18.2) | 44 (18.3) | 60 (22.2) | ||
|
N/A | .19 | N/A | N/A | .54 | N/A |
aRCT: randomized controlled trial.
bThe
cThe
dN/A: not applicable.
A total of 1047 (66.43%) of the 1576 participants chose to watch the sugar video. Among these participants, the average time spent watching the sugar video was 116.35 (52.4%) of 222 seconds.
To see view time as a function of trait reactance proneness for the child, mother, and physician narrators, adjusted for education level, please see
Participant view times (n=1047) of a short, animated video about sugar intake by narrator’s voice, sociodemographic characteristics, and trait proneness reactance. Note: The whiskers represent the 95% CIs of view time.
Linear regression coefficients of factors (narrator’s voice, sociodemographic characteristics, reactance proneness) associated with engagement with a short, animated video about added sugars (n=1047).
Factors | Model 1 (SE, |
Model 2 (SE, |
Model 3 (SE, |
Model 4 (SE, |
Model 5 (SE, |
|
|
||||||
|
Narrator 2: mother | 5.217 (7.362, .48) | 4.196 (7.250, .56) | 4.820 (7.255, .51) | 4.807 (7.265, .51) | 3.995 (7.262, .58) |
|
Narrator 3: physician | 0.023 (7.429, .99) | –1.080 (7.306, .88) | –0.616 (7.307, .99) | –0.849 (7.304, .91) | –1.582 (7.295, .83) |
|
||||||
|
25-34 | —b | 33.44 (7.873, <.001) | 33.60 (7.857, <.001) | 33.42 (8.022, <.001) | 32.92 (8.026, <.001) |
|
35-44 | — | 46.62 (8.546, <.001) | 47.41 (8.546, <.001) | 47.22 (8.622, <.001) | 46.96 (8.624, <.001) |
|
45-54 | — | 49.28 (9.609, <.001) | 48.99 (9.591, <.001) | 48.77 (9.575, <.001) | 47.50 (9.592, <.001) |
|
55-59 | — | 47.80 (13.959, .001) | 48.68 (13.886, <.001) | 47.77 (13.977, .001) | 46.78 (13.897, .001) |
|
||||||
|
Female | — | — | 10.21 (6.109, .095 | 10.47 (6.134, .09) | 10.02 (6.129, .102) |
|
Other | — | — | –6.822 (33.142, .84) | –6.874 (33.296, .84) | –7.537 (34.051, .83) |
|
||||||
|
High school | — | — | — | 28.63 (22.411, .202) | 25.43 (22.257, .25) |
|
BA, some college | — | — | — | 21.54 (21.230, .31) | 19.61 (21.037, .35) |
|
MA/PhD | — | — | — | 27.02 (21.977, .22) | 23.90 (21.844, .27) |
|
— | — | — | — | –10.86 (5.871, .07) | |
|
n | 1047 | 1047 | 1047 | 1047 | 1047 |
aRef: reference group.
bNot applicable.
Predicted view times (n=1047) by narrator’s voice of a short, animated video about sugar intake. Note: The whiskers represent the 95% CIs at predicted trait proneness values.
In this web-based RCT, we assessed participant engagement with a SAS video about added sugar consumption. We hypothesized that participants with higher levels of trait reactance proneness would watch the SAS video for a shorter period and that younger participants (aged 18-24 years) would have higher engagement with the sugar intervention when compared with older participants (aged 25-59 years). Overall, 66.43% (1047/1576) of the participants voluntarily watched the sugar intervention video with an average view time of 116.35 (52.4%) of 222 seconds. We observed that participants with low levels of trait reactance proneness watched the video longer, whereas contrary to our expectations, older participants watched the intervention video longer than younger participants.
As stated, our results show that the majority of the 1576 participants chose to engage with the intervention video and watched, on average, more than half of the video. In recent years, an ever-increasing number of offerings, including high-budget entertainment productions, have competed to occupy our leisure time. The degree of voluntary engagement seen in this study, despite a comparatively low-budget, SAS health video, underscores the potential for this health communication modality. The engagement documented in our study also far exceeds patient engagement with print-based health communication materials distributed in health care settings [
We examined the role of trait reactance proneness on participants’ view time, which has been shown to be an obstacle to successful health promotion campaigns [
In the main trial, we investigated the role of the narrator on reactance to the sugar video. Since previous studies have shown that individuals may perceive doctors as coercive or overly directive [
Since the SAS video was designed for rapid distribution on social media channels, we expected higher participation from younger participants (aged 18-25 years). Surprisingly, we found that older people (ages 25 years and more) watched the video, on average, longer than younger participants. This result might be explained by the perceived vulnerability among older adults, that they are more likely to suffer from health problems that are associated with an excessive consumption of added sugar. Furthermore, longer viewing times in older adults might also be connected to differences in information processing or the perceived seriousness and involvement in the study. Younger participants (ie, emerging adults in this case) are less risk averse and more accustomed to engaging with extremely short forms of content [
A key strength of this study was the use of an RCT design, which allowed us to reduce any systematic differences and bias through randomization. In addition, the use of an online recruitment platform helped us reach a large sample size, ensuring the quality and reliability of the results. We are not aware of any other study that had such a large sample size and used a similar experimental approach to examining participant engagement in the field of public health. Arguably, this posttrial stage of our RCT enabled us to capture participants’ voluntary willingness to watch a SAS video without any financial compensation. Although this condition is similar to the real world, we acknowledge that participants’ responses may have been affected by their awareness of being in a scientific study and that their actions were being recorded for scientific purposes. Nevertheless, outside of a scientific study setting, we report anecdotal evidence of willingness to engage in our sugar intervention video. After our RCT, the child-narrated version of the video was posted on the creator’s (author MA) YouTube channel, where it reached 3700 views in the first 48 hours after its release.
Our study had several limitations. Given the online setting of our study, we were not able to determine whether participants actively watched the intervention video (it may have been playing in the background while the participant was engaged in other activities). Given the posttrial phase of our study, we were only able to evaluate the role of demographic factors and trait proneness reactance on participant engagement with the sugar intervention video. We acknowledge that other factors could have affected the time that participants spent watching the sugar intervention video, such as the perceived threat of the message, the perceived threat to health, the perceived risk of adopting an alternative behavior, and anger and negative cognition toward the sugar message (see
SAS videos demonstrate potential for engaging diverse audiences and thereby enhancing the distribution of health education messages. Designed to be emotionally arousing and culturally neutral, SAS videos can facilitate public health efforts to promote healthy behaviors and meet audiences where they are across the media landscape. The evidence from this study demonstrates promising engagement with the SAS health messaging modality, across diverse audiences. As these audiences spend increasingly more time online, the need for innovative approaches to engaging them also increases. Even the most accurate and clear health messages have little value if they fail to reach their target viewers. For this reason, researchers and health communicators of the future will need to understand how to optimally engage their audiences and research in this field should be a high priority.
Supplementary Materials.
CONSORT-EHEALTH checklist (V. 1.6.1).
randomized controlled trial
short, animated story-based
This study was funded by the Alexander von Humboldt University Professor Prize awarded to TB.
CF, MA, and AV wrote the paper. CF, AV, and VH undertook the statistical analysis. MA designed, produced, and created all 3 sugar videos (child, mother, physician). AV, VH, and CF contributed to the questionnaire development. AV and TB designed the trial. All authors provided comments and feedback. The data that support the findings of this study are available from the corresponding author upon request.
The authors declare no competing interests. The funder of the study had no role in study design, data collection, data analysis, data interpretation, or writing of the report. The corresponding author had full access to all the data in the study and had final responsibility for the decision to submit for publication.