Effective public health communication plays a crucial role in empowering individuals with information and skills to have more control over the decisions and behaviors that impact their health. Modifiable behavioral risk factors, such as sedentary behavior, unhealthy diets, not attending health screenings, and high-risk sexual behaviors, contribute to between a quarter and a half of the global burden on health and social care, as well as deaths [1-3]. Health communication science can critically inform the strategies required to ensure that accurate and accessible information and support are provided to individuals, communities, and populations to promote health and positive health behaviors, and reduce subsequent morbidity and mortality [4].

Effective health communication strategies need to address health literacy, defined as the knowledge and competencies needed to access, understand, and use health information [5]. Including visual stimuli within health communication is encouraged, particularly when trying to reach individuals disadvantaged by lower literacy and education levels; knowledge, understanding, attention, and recall are improved when visual images are used in health communication compared with using text alone [6]. This is supported by the picture superiority effect from cognitive psychology, which shows that visual information is more memorable than textual information. This suggests that visual information is more effectively encoded into memory, as it activates both verbal and visual memory systems, strengthening memory formation and subsequent recall [7].

Health animations, defined as simulated motion pictures depicting the movement of drawn (or simulated) objects [8] related to health, are a highly prevalent approach to visual health communication. Due to their digital nature, animations can be easily and inexpensively shared via the internet, apps, and social media, catalyzing their widespread global application as a digital health solution [9] to deliver public health information, promote health behavior change, and prevent ill health. Compared with paper-based or in-person health communication, animations have greater reach and scale, overcoming geographical barriers and exploiting the rapid rise and evolution of social media platforms (eg, TikTok, WhatsApp, and Instagram) and online mediums (ie, local and international governmental, nongovernmental, voluntary sector, charity organizations, or websites).

Emerging reviews and meta-analytical evidence of animations suggest that there are benefits to learning and cognitive load from animations as educational tools for patients [10,11], health care professionals [12], and in non–health care settings [13]. However, evidence of health animations in the promotion of public health preventive behaviors, such as lifestyle, hygiene, and preventive health appointment uptake—which can prevent and control communicable and noncommunicable diseases—remains unclear. To date, evidence relies on individual empirical studies [14-16] investigating the impact of health animations on behaviors and social cognitive determinants (eg, recall, knowledge, and intention). These individual studies exhibit substantial heterogeneity in study design, comparators, target health behaviors, target audiences, and contextual complexities in the animation design and content. Therefore, rather than attempting to synthesize evidence for the effectiveness of health animations promoting preventive health behaviors from a pool of heterogeneous studies, we adopt a realist review approach to iteratively identify and synthesize evidence to uncover the specific causal mechanisms being triggered by animations that, in turn, facilitate certain outcomes.

The aim of this realist review was to identify and synthesize evidence about health animations to understand how, why, for whom, to what extent, and in which contexts they are expected to produce their effects. We aimed to generate knowledge about the causes of outcomes in particular circumstances in terms of the context-mechanism-outcome configurations (CMOCs) of realist reviews and syntheses. By understanding how and why health animations and the processes underpinning them work (or fail to work), we aimed to translate key learnings into a set of recommendations to inform the design, development, and implementation of future public health animations.

This realist review and synthesis was conducted in accordance with the Realist and Meta-Narrative Evidence Syntheses: Evolving Standards (RAMESES) [17-19] and a published protocol [20]. Textbox 1 provides definitions of key realist concepts. The realist review methodology and structure are well-suited to the complexity and heterogeneity of health animations and the methods used to evaluate them. Health animations are used across a variety of health behaviors and therefore can produce multiple varied outcomes. While conventional systematic reviews focus on outcomes, realist reviews focus on the contexts and the mechanisms they trigger to produce those outcomes. Furthermore, realist methodology does not seek to separate and categorize the elements of the evidence being explored but instead finds causation and configurations that lead to a better understanding of how and why effects are produced [21]. This is achieved by identifying contexts and mechanisms and examining how they interact to produce outcomes; once identified, CMOCs can be developed. The following 5 iterative steps were undertaken with some elements within the steps occurring simultaneously.

Theories and models across multiple disciplines were explored for insights and understanding of what may influence how health animations are perceived, interpreted, or experienced. These insights were discussed within the review team to develop initial program theories. Exploratory background searches were conducted to identify literature on the use of health animations and their impacts and effects, helping to define the review’s scope.

Initial searches of electronic databases were conducted from database inception until July 10, 2023, with ongoing searching throughout the review process (in accordance with the iterative nature of realist reviews). A final search of electronic databases was conducted on April 27, 2025. The databases searched were MEDLINE, Scopus, CINAHL, EMBASE, PsycINFO, ProQuest Social Science, Web of Science, and the Cochrane Library. The search strategy, combining relevant Medical Subject Headings (MeSH) terms and keywords for concepts of animation, health behavior, and behavior change, is available in Multimedia Appendix 1. Reference lists were checked, and websites of governmental, nongovernmental, and third-sector organizations known to promote public health were searched to identify additional health animations. Google and YouTube were also searched using keywords to identify additional health animations and organizational sources of animations.

Rayyan (Rayyan Systems, Inc) was used to manage the records, including deduplication. Title and abstract screening was conducted by 1 reviewer (KMC), with a random 10% sample screened independently by another reviewer (NO’B). Full-text screening was conducted independently by 2 reviewers (KMC and NO’B). The authors were contacted to provide animations when they were not publicly available. Textbox 2 presents the inclusion and exclusion criteria.

Evidence was selected using the criteria outlined in step 2 of the protocol [20]. Peer-reviewed and gray literature publications were included where they could potentially contribute to building the program theory. The relevance, rigor, and richness of the data were used to assess the quality of evidence [22]. Questions asked of the data in relation to relevance included: Does the evidence help develop CMOCs and contribute to their advancement? Is the building and testing of program theory supported by this evidence? Does it speak to the research question? Questions asked in relation to richness included: Is there a sufficient level of description about the development of the intervention and the theories and concepts underpinning it? Was there discussion about factors that impacted the functioning or outcomes of the intervention? Were details provided about how the intervention was meant to work? Was this information general or specific to the intervention? Questions asked in relation to rigor included: What methods were used to generate the data? What are the credibility and trustworthiness of these methods?

Data extraction was conducted by 1 reviewer (KMC) and checked by a second reviewer (NO’B). Bespoke data extraction sheets were developed to record study and animation characteristics. Study characteristics included information about participants, outcomes, findings, health behaviors targeted, and theories influencing animation design. Animation characteristics included length, use of sound, color, language, and text, development information, composition, character profiles, use of cultural indicators, symbols, and metaphors.

Data synthesis and theory refinement were developed through multiple viewings of the animations, discussions within the review team, and consideration of the identified theories and models. The initial synthesis and CMOC development involved KMC, NO’B, and GM reviewing the data to identify patterns (ie, demiregularities) and elements that the animations shared to define the “if” (ie, contexts) and to list the “then” (ie, outcomes). Data were then interrogated to define the “because” (ie, mechanisms), resulting in if-then-because (context-outcome-mechanism) statements [23]. The integrity of the if-then-because statements was questioned as discrete statements by considering the included data and their potential interpretations in relation to the initial program theories identified in step 1. In doing so, an emerging program theory was developed.

Four stakeholder workshops were held to sense check preliminary findings and to discuss and refine the if-then-because statements and the emerging program theory. Two international online workshops were conducted with professionals in April 2024 (n=21), allowing for the testing and refinement of the program theory with individuals who develop, commission, and/or promote animations or who work with individuals or systems that may benefit from health animations. Two in-person workshops were held with members of the public in the United Kingdom in May 2024 and June 2024 (n=18), enabling the testing and refinement of the program theory with public stakeholders targeted by and exposed to health animations. Following these workshops, feedback was used to review the if-then-because statements and translate them into CMOCs. The CMOCs were tested against the data and refined as required. They were then used to develop a set of recommendations to inform the design, development, and implementation of future public health animations. Both the CMOCs and the recommendations were presented to professional and public stakeholders for feedback during a final online workshop in September 2024 (n=17).

Ethical approval for the stakeholder workshops was granted by the Northumbria University Research Ethics Committee (reference #6803). Stakeholders provided informed consent before the workshops, and their privacy and confidentiality were ensured. Public stakeholders were remunerated for their time and contribution with a £25 (US $33.8) shopping voucher.

The final sample for this review included 48 unique single animations or series of animations; 39 were reported in 47 peer-reviewed publications [14-16,24-67] and the remaining 9 were gray literature [68-76]. Figure 1 displays the flow of the selection of animations. Categories of health behaviors targeted were communicable diseases and hygiene behaviors (n=11), vaccination (n=10), general health promotion (n=8), screening (n=5), substance use (n=5), sexual health (n=4), nutrition (n=3), and oral health (n=2). Study designs included: randomized controlled trials (n=18), randomized studies (n=12), pre-post studies (n=8), mixed methods studies (n=5), and cross-sectional studies (n=4).

Of the 48 animations included, 28 were single animations and 20 were animation series; for the purposes of data synthesis, animation series were considered as single animations. Outcome measures included knowledge, behavioral intentions, attitudes, skills, acceptability, and viewer engagement; 8 studies also included behavioral outcomes. Many animations reported details of co-design processes, for example, in a study by Willis et al [65], and theories and models (eg, social cognitive theory [47], elaboration likelihood model [24], and theory of planned behavior [50]) that informed animation development, evaluation, and outcome measures, and provided key insight and information in the building of the program theory.

Most animations used storytelling to convey messages (n=22), while others used entertainment elements (n=18), storytelling elements (n=4), or no storytelling/entertainment elements (n=4). Textbox 3 provides an explanation of the different categories of entertainment that were used. Further study design details, including the demographic and geographical target audience of the animations, as well as the animation characteristics, such as the use of sound, language, text, and length, are presented in Multimedia Appendices 2 and 3 [14-16,24-76].

The main characteristics contributing to understanding why, how, for whom, to what extent, and in which contexts health animations are expected to promote preventive health behaviors were considered within 3 high-level constructs: design, content, and delivery. These constructs did not operate in isolation but often worked together to enhance effects. It is within these constructs that the contexts, mechanisms, and outcomes identified in the analysis were situated (Figure 2). Consequently, the following 9 CMOCs were developed and tested with stakeholders (Table 1). The final column of Tables S1 and S2 in Multimedia Appendices 2 and 3, respectively, details which CMOCs were associated with each animation included. Additional detail on the constructs, contexts, and mechanisms is provided in Multimedia Appendix 4.

The construct design refers to the ways in which health animations are constructed to express the ideas and messages they are trying to communicate. How an animation looks, feels, and sounds, as well as its length, use of symbols and metaphors, character profiles, and other features, are all products of the design.

The construct content refers to the message itself, the information it contains, and how it is expressed.

The construct delivery refers to how health animations are experienced, including who is delivering the message, what platforms are being used, and how and through what modes animations are accessed. The delivery of health animations also includes the intensity of exposure.

This realist review identified and synthesized evidence from 48 health animations providing an understanding of how, why, for whom, to what extent, and in which contexts health animations designed to promote preventive health behaviors work or fail to work. With input from public and professional stakeholders, we developed a program theory comprising high-level constructs and CMOCs, which were translated into a set of recommendations to inform the design, development and implementation of future public health animations.

The review produced 4 key findings. First, there are three interrelated high-level constructs (ie, design, content, and delivery) that often work synergistically. Furthermore, when exploring these constructs and their use in health animations, they were broken down into specific contexts thought to trigger causal mechanisms.

Second, entertainment was identified as a key context used in many health animations, through storytelling and its elements. Storytelling is based on the theory that humans understand and give meaning to their life experiences by telling stories [77]. Moyer-Guse et al [78] discussed the role of the cognitive mechanisms of persuasion and transportation by a story in relation to childhood vaccination attitudes and further discussed the power of narratives and entertainment to affect attitudes and beliefs. More broadly, “entertainment education” approaches have historically been used for health communication globally and to good effect across a range of health domains [79]. In this review, the terms “storytelling” and “entertainment” were used instead of “narrative” and “didactic,” as conversations with the research team and stakeholders revealed that the terms “narrative” and “didactic” had the capacity to take on multiple meanings and cause confusion.

Third, the entertainment context often operated in conjunction with other contexts identified in the review. For example, when emotionally evocative content was present, storytelling, and/or entertainment were also present. This finding is in line with evidence that stories can capture audiences’ attention, educate and entertain them, and evoke emotions [80]. Concurrent with evoking emotions, is the possibility of psychological reactance, which, in the context of health communication refers to an audience’s resistance to act upon a health message in the intended way due to a perceived threat to their freedom [81].

Fourth, contexts that facilitated accessibility were deemed particularly valuable to stakeholders, who frequently discussed the important challenge for health communications to address health literacy. Health literacy is a key social determinant of health [82]. Literature reviews consistently show that low levels of health literacy are associated with worse health outcomes, including higher mortality rates, more hospitalizations and emergency department visits, and poor engagement with healthy behaviors, health messages, and preventive health care services [83,84]. Educational attainment has been shown to be the most important determinant of health literacy, followed by other socioeconomic factors such as financial resources [85]. Therefore, within the current context of health animations, it is of paramount importance to consider the role of these factors in determining the impact of an animation. The contexts that facilitate accessibility to and comprehension of the information and resources provided in health animations were present across all three of the high-level constructs identified in this review and therefore equally require full consideration in efforts to enhance health literacy.

Based on the findings of this review, we suggest a set of recommendations for researchers and practitioners to consider when designing and developing future health animations. The relative weight or focus that can be given to each recommendation may depend on multiple factors including the health topic, audience, time, and financial resources available to develop the health animation. For example, when developing an animation designed for a global or multilingual audience, to maximize its potential reach without requiring additional resources to develop multiple versions in different languages, it would be of paramount importance that the animation does not rely on written or spoken language to be understood, thereby reducing language and literacy barriers. Instead, storytelling, entertainment, and emotional elements could be exploited to ensure that the animation visuals alone can work to promote the target health behaviors. It would also be important to consider the most acceptable and utilized modes for animation delivery (eg, social media, instant messaging, and screens in public places) that meet the differing needs of subgroups in the target population. Researchers and practitioners should also consult these recommendations when reviewing data from individual studies evaluating health animations to appraise the specific features of the animation design, content, and delivery.

Few evaluations of animations provided data on behavioral outcomes and therefore, our program theory largely reflects the contexts and mechanisms at play when producing effects on key evidence-based behavioral determinants. The behavioral determinants that the animations were found to impact included evidence-based correlates of subsequent health behavior, namely behavioral skills, attitudes, and intentions [86]. Critically, interventions targeting behavioral skills and attitudes are known to produce small effects on health behavior [87], and therefore animations we identified as evidencing impact on these determinants may also yield positive behavioral changes. Moreover, many of the animations identified in our review reported knowledge gains, and even though interventions targeting knowledge are known to produce only minimal effects on behavior [87], knowledge often plays a critical role in health behavior change in terms of raising awareness of the health problem or topic, the need to change behavior, and its link to behavioral and self-regulation skills [88,89].

Future studies are urged to collect both short- and long-term behavioral data when evaluating health animations to expand the evidence base regarding their effect on behavioral outcomes. Furthermore, future research is needed to investigate the role animations could play as part of wider behavior change interventions, potentially targeting not only the more impactful individual determinants of behavior but also sociostructural factors [86,87]. Finally, investigations into potential dose-response effects through repeated exposure to an animation (or a series) are required to further our knowledge of the use of animations, especially since they can be easily and inexpensively shared digitally [9].

Limitations in the type of data, depth, and quality of reporting in studies on the design and implementation of public health interventions within published literature are well known [90]. Consequently, the pool of available literature for this review was reduced due to studies failing to provide access to the animation. Nevertheless, increasing interest in and application of health animations mean that the scientific evidence base will continue to expand.

Despite the aforementioned limitations, this review also has the following strengths. This is the first realist review of health animations designed for preventive health behavior promotion in the general population. We strictly adhered to the standards and guidance for conducting and reporting realist reviews [18,19,22], and as such, the review team engaged in regular discussion and reflection on the evidence identified. We actively and authentically included public and professional stakeholders throughout the review process, seeking to ensure that we involved international professional stakeholders from different sectors and public stakeholders from diverse ethnic and sociodemographic backgrounds. The resultant program theory extends the knowledge base by identifying and refining thinking around the questions of why, how, for whom, to what extent, and in which contexts health animations work as a digital form of health communication. The iterative, cyclical nature of realist reviews meant that searching for new evidence continued throughout the review process; therefore, noteworthy new evidence was captured and included in the synthesis. This new evidence reinforced initial findings shared with stakeholders and contributed to the currency of the review outcomes.

Animations provide a compelling digital approach to support the promotion of preventive health behaviors. Their visual and entertaining nature, with reduced reliance on language, means that they can help reduce health literacy barriers in the target audience and address health inequities in populations. The findings of this realist review of health animations designed to promote preventive health behaviors, along with the program theory it has generated, provide novel insights into the specific contexts and mechanisms influencing evidence-based determinants of behavior and behavior change. It is critical that the contexts and mechanisms found to drive animation impacts in the evidence we reviewed, as well as the associated recommendations we provide, are considered by those designing and developing health animations in the future.