Drivers of vaccine mis/disinformation in the media: from personal beliefs to cultural dimensions

Abstract

Introduction:

The unabated spread of vaccine mis/disinformation poses a great challenge to the achievement of the SDG 3 and Universal Health Coverage (UHC) goals. This systematic review synthesizes the drivers of vaccine mis/disinformation in the media and how geography shapes these drivers through the lens of Hofstede’s cultural dimensions theory.

Methods:

A search was conducted in Scopus, Web of Science, and PubMed for studies between 2011 and 2024, arriving at a final sample of 27 studies. Emerging drivers of vaccine mis/disinformation identified were categorized into four levels- individual, message, platform, and societal levels with the individual-level (personal-related) drivers dominating the spread of vaccine mis/disinformation.

Results:

Results reveal that though individual-level drivers such as being right-wing authoritarian, religious, or being an older male drive the spread of vaccine mis/disinformation on the demand side, message-level drivers including emotional framing and introduction of expert cues in messages also significantly drive the spread of vaccine mis/disinformation from the supply side. Further findings revealed that the prevalent cultural dimension in different climes played significant roles in the prevalence of drivers across certain geographies.

Discussion:

The high-power distance culture of developed societies such as North America reflected the prevalence of the message-level driver given the mature and robust research and media ecosystem. Conversely, African and Asian societies which are tilted to the collectivism dimensions of Hofstede’s dimensions theory showed a higher propensity for individual-level drivers, given that the social identity in a collectivist society shapes the behaviors of individuals. The study concluded that cultural theories predict the dominance of how vaccine mis/disinformation spreads in different geographies. Further findings revealed an overlapping complementary relationship between drivers. It was thus recommended that future reviews and studies should deeply explore these relationships and how they shape vaccine mis/disinformation discourse across geographies.

Systematic review registration:

https://www.crd.york.ac.uk/PROSPERO/recorddashboard, CRD42024601978.

Introduction

Vaccines are among the safest preventive medical interventions in human history, averting an estimated 3.5 to 5 million deaths annually, markedly improving global health outcomes (WHO, 2024). Despite this, immunization rates continue to decline in many regions of the world, partly driven by vaccine hesitancy- the delay or outright refusal of vaccines despite availability (WHO SAGE Working Group on Vaccine Hesitancy, 2014). This continuously undermines global public health and immunization efforts (WHO, 2019) at achieving universal health coverage and the sustainable development Goal 3 (SDG3) targets. Declining immunization uptake due to vaccine hesitancy has dire implications for global health, evidenced by the recent resurgences of Measles in developed regions of the world, with a global mortality of over 140,000 in 2018 alone (Carrieri et al., 2019; CDC, 2024).

A key contributor to vaccine hesitancy is the twin problem of vaccine misinformation and disinformation -the spread (intentionally and unintentionally) of false information about vaccines. Recent studies (Nwachukwu et al., 2024; Serge Andigema and Tania Cyrielle, 2024; Morejón- Llamas, 2023) have linked vaccine hesitancy to widespread mis/disinformation and conspiracy theories about vaccines. Although this circulation of falsehoods about vaccines, with its far-reaching consequences has always existed (Eddy et al., 2023; Eichman and Bichianu, 2024; Schwartz, 2012), the recent COVID-19 pandemic marked a new era, an all-time high of widespread circulation of misleading vaccine narratives and aggressive anti-vaccine messaging. This proliferation has led to a distorted understanding of immunization facts among the public, prompting public skepticism toward vaccines (Posetti and Bontcheva, 2020), a critical challenge that negatively influences public health decision-making.

Public health practitioners, policymakers as well as researchers have advanced varying interventions- including fact-checking and debunking mechanisms - to combat vaccine mis/disinformation. Although the deployment of these interventions and efforts have yielded significant positive results as recorded in literature (Whitehead et al., 2023; Schmid and Betsch, 2022; Xue et al., 2022), vaccine mis/disinformation continues to spread, necessitating the quest for impactful interventions to combat this spread (Schmid and Betsch, 2022). Understanding how these messages circulate, particularly the factors that motivate widespread circulation becomes crucial for designing effective interventions that directly respond to these drivers.

The propagation of vaccine mis/disinformation is prompted by different influences including- contextual, individual/group, and vaccine-specific across different parts of the world (World Health Organization SAGE Working Group, 2014). Studies have increasingly focused on a diverse range of drivers behind the spread of vaccine mis/disinformation- ranging from the quest for financial gain (Tokojima Machado et al., 2020), to the emotional triggers, low media literacy, religious beliefs, right-wing authoritarian attitudes, and affordances of social media platforms through echo chambers (Dunn et al., 2015; Lundy, 2023; Moran et al., 2022). Even though these studies have explored drivers individually across different contexts, with different study designs, and studying different vaccine types, they have mainly created a fragmented view of these drivers in literature. In the quest for an all-encompassing intervention that provides broad-level view across all vaccine types, media types, and contexts, it becomes critical to synthesize available evidence regarding the drivers of vaccine mis/information. In the same development, the complex interplay of these drivers and regional/cultural orientation using established cultural theories additionally remains underexplored. Our review, therefore, addresses this gap in literature by synthesizing the existing studies while also evaluating the global and geography-specific prominence of the drivers of vaccine mis/disinformation, guided by the lens of the cultural dimensions theory.

Existing reviews have also provided different focus in literature, with some focusing on the broad health misinformation landscape (Wang et al., 2019), others narrowly focusing on the COVID-19 pandemic and vaccines (Malik et al., 2023; Skafle et al., 2022), focusing on single media types (social media) and specific vaccines, limiting their relevance to the vaccine mis/disinformation discourse in general. While these research efforts exist, none has holistically analyzed vaccine mis/disinformation drivers across all vaccine types, media types, disaggregated by geographical contexts. Based on the foregoing, this review, following PRISMA guidelines, systematically synthesizes published studies (between 2011 and 2024) to examine drivers of vaccine mis/disinformation across traditional and digital media, while examining the cultural dimensions of these drivers across geographic contexts.

Literature review

Key concepts

Misinformation and disinformation are key concepts in this review that guide the extraction of relevant data. Whereas Misinformation and disinformation both point to the dissemination of false and/or inaccurate information to the public by different actors (Praveenkumar, 2024), there is a difference in the intent. Misinformation is oftentimes not deliberate, and hence not intended to deceive or achieve preconceived sinister goals, while disinformation refers to the deliberate creation, presentation dissemination of verifiably false information to deceive the public intentionally, cause public harm, or for economic gain (European Commission, 2018). Equally, a UNESCO report (Posetti and Bontcheva, 2020) has broadly defined disinformation as “content that is false and has potentially negative impacts” (P.1). Further, they aver that the goal of the person producing or sharing such inaccurate content differentiates disinformation from misinformation.

Despite these stated differences in the use of both terms, the distinction appears to be more plausible theoretically, because it is difficult to differentiate in practice (Wang et al., 2019). Similarly, it has been shown that there exists an issue of conceptual clarity and distinction among and between these terms (Broda and Strömbäck, 2024), hence some studies explore these terms and conduct a general analysis without making major distinctions among them (Skafle et al., 2022). The incumbent study is not aimed at establishing the different ways the terms have been studied distinctly rather, it adopts all-encompassing approach to view the subject matter. Consequently, our review categorizes both terms in the same block of information disorder while not differentiating between both concepts as the impact of false content is potentially the same, irrespective of the intentions (Posetti and Bontcheva, 2020). Whereas drivers of false vaccine content could be outcomes of misinformation or disinformation, our goal in this review is to provide a broad perspective about the dissemination of misinformation and disinformation, the agent, the message, and the interpreter (Wardle and Derakhshan, 2017).

Other key concepts we have used in this review include drivers, and media. ‘Drivers’ has been used in previous research (Wang et al., 2019) to refer to the facilitators of the spread of misinformation across the media. Drivers in this study would range from message content to characteristics of individuals, to platform characteristics, that enable the diffusion of vaccine mis/disinformation in the media. The media includes studies about the social media and traditional media. Existing reviews have significantly focused on social media in health and vaccine mis/disinformation discourses (Malik et al., 2023; Skafle et al., 2022), however, the circulation of vaccine misinformation predates the widespread use of the social media. It is based on this that we have deemed it fit to include studies from both media types.

Theoretical underpinning

This review is anchored on Hofstede’s (2011) cultural dimensions theory and Wardle and Derakhshan’s (2017) framework of information disorder. The Cultural dimensions theory explains the relationship between culture and the behaviour of the members of that culture. It describes the “effects of culture on the values of its members and how these values relate to the behaviour of people who live within a culture” (Nickerson, 2023). The theory goes beyond the collective nature of culture to espousing the idea that this phenomenon is connected to different collectives called dimensions. Hofstede initially identified four dimensions in his seminal work, namely; individualism and collectivism, power distance, uncertainty avoidance, and masculinity and femininity. Two more dimensions emerged as products of validation studies- long-term or short-term orientation (see Hofstede and Bond, 1988); and Indulgence versus Restraint (see Hofstede et al., 2010). A critical objective of our review is to examine the interplay of vaccine mis/disinformation drivers across different cultures marked by geographies, hence the adoption of this theory. Through the lens of these dimensions, we analyze and advance reasons for geographical variations in the drivers of vaccine mis/disinformation, showing why some drivers are more dominant in some cultures than others. Although the studies that led to the theory examined national cultures as the units of analysis, our review takes a broader geographical standpoint at the continental level.

In their elements of information disorder, Wardle and Derakhshan provide a lens through which disinformation and misinformation -both are components of information disorder- can spread (see Figure 1). They provide three major elements- the agent, the message, and the interpreter- to explain how the process of creating, producing, and reproducing mis/disinformation works. The agent for our study is like the interpreter who disseminates the message, however, our focus is on the characteristics and motivations of these actors as developed by Wardle and Derakhshan (2017), which could be financial, political, social, and psychological. These motivations provide a framework to clearly conceptualize the human characteristics and motivations that drive vaccine mis/disinformation. The message element in the framework provides an avenue to examine the framing and construction of vaccine mis/disinformation content and how the message is shaped to induce a certain kind of reaction by the spreaders- agents and interpreters. The framework guides our view of the major factors that drive vaccine mis/disinformation.

Figure 1

The cultural dimensions theory enables us to interpret how geographical/cultural variations affect the spread of vaccine mis/disinformation in different contexts, explaining why certain drivers are more dominant in a context than others. The agent-message-interpreter model provides a narrower lens, focusing on how individuals (not national cultures) and message characteristics drive the spread of vaccine mis/disinformation.

Existing evidence

Vaccine mis/disinformation, as major contributors to vaccine hesitancy has long been a subject of discourse among scholars, policymakers, and public health workers alike, given its continued influence on immunization programs across the world, limiting the potentials of achieving the SDG3 goal of universal health coverage (UHC). While the WHO has noted that vaccine hesitancy is a continuum between acceptance and outright refusal of vaccines (WHO SAGE Working Group on Vaccine Hesitancy, 2014), the proliferation of vaccine mis/disinformation contribute to each phase of the hesitancy spectrum.

Reviews have explored issues around misinformation and disinformation in health with limited focus on vaccines in general. While some of these reviews have a narrow focus, if they are about vaccines, others treat broader issues around health misinformation. The closest review to the current study (Zhao et al., 2023), in our opinion focused on misinformation evidence related to COVID-19 vaccines alone, without geographical insights explained through the lenses of a cultural theory. The review synthesized evidence on the prevalence, features, influencing factors, impacts, and solutions as regards COVID-19 misinformation from January 2020 to August 2022. In the same vein, a similar study (Skafle et al., 2022) - a rapid review of COVID-19 misinformation on the social media was conducted in 2021, with strict focus on COVID-19 as a pandemic and disease, not a specific focus on vaccines. On one hand, these reviews were narrow in scope, given that they were limited to the COVID-19 pandemic or vaccines, leaving out evidence about studies that focused on other vaccines, while on the other hand they omitted studies that were published before 2020 and beyond 2022. It has been noted that the spread of vaccine misinformation did not commence with the COVID-19 vaccines (Eddy et al., 2023; Schwartz, 2012), and such, there is a critical need to explore earlier studies about other vaccines before the start of the COVID pandemic.

Another review (Malik et al., 2023) recently explored the factors related to the sharing of COVID-19 misinformation on social media. The researchers discovered five major factors associated with COVID-19 misinformation sharing on social media, including socio-demographic characteristics, financial considerations, political affiliation or interest, conspiracy ideation, and religious factors. This review provided a blueprint for creating categories from the drivers of misinformation in the media; however, the review was significantly targeted at the COVID-19 pandemic in general, with less emphasis on vaccines, which is the focus of the current study. While this review focused on the disease, it provides a concrete blueprint for separating drivers or factors into varying categories.

Other studies have researched the spread of health misinformation generally. A review (Moran et al., 2022) explored the drivers of health-related misinformation between 2012 and 2018. The review revealed how immunization and infectious diseases were prevalent in health misinformation discourses, with a significant number of studies from the social media. This study, however, paid little attention to the drivers of vaccine misinformation in the media.

Current literature about the drivers of mis/disinformation, illustrated by the identified systematic and rapid reviews, falls short of proffering solutions to the continued spread of vaccine mis/disinformation across the media. These studies have either generally focused on COVID-19 vaccines alone, focused on only social media, focused on shorter timeframes, focused on the broad health misinformation spectrum, or laid emphasis on the interventions advanced against the spread. This reveals a gap, which this current study fills by synthesizing evidence about all vaccines, with a geographical nuanced perspective, across both traditional and social media, and as well as an extended timeframe which stretches from the decade of vaccines (World Health Organization, 2019) to the aftermath of the COVID-19 pandemic. This is with the aim of providing a single, concise evidence.

Methods

We conducted a systematic literature review of peer-reviewed published articles on three major public health databases- Web of Science (WoS), Scopus, and PubMed to retrieve relevant articles following PRISMA guidelines. The research team unanimously developed and agreed on a protocol and search strategy for the review, pre-registered on Prospero with reference number CRD42024601978.

Database search

Our team searched the databases in September 2024 to retrieve relevant articles for the study. These databases have been employed in a range of previous reviews encompassing vaccination misinformation (Skafle et al., 2022; Zhao et al., 2023). The search included articles published in 14 years (01/01/2011 to 30/06/2024) at the intersection of vaccines, mis/disinformation, media, and drivers. The study period was decided upon to ensure that our review covers a crucial period in global immunization that witnessed an increased relevance of the new media in health communication dynamics (Huo and Turner, 2019; Putri et al., 2023; Yao, 2024). This period witnessed a surge in vaccine discourses with 2010–2020 being labeled as the decade of vaccines (World Health Organization, 2019). Studies related to the pandemic, the Ebola epidemic, as well as the development and adoption of a wide range of vaccines in different countries are targeted. The researchers also retrospectively extended the timeframe to be able to ascertain trends beyond the immediate COVID-19 pandemic, which resulted in an infodemic as cautioned by the WHO Director General (WHO, 2020). Search languages were limited to English language and Spanish since the research team has a combination of proficiency in both languages.

Search strategy

Studies were included regardless of their methodological quality and risk of bias. The focus on describing and synthesizing patterns rather than establishing causation influenced the decision to include studies despite their Risk of Bias. Our searches included MeSH terms as well as keywords and synonyms relevant to the study objectives which bother on the media, vaccines, drivers (or influencers), as well as mis/disinformation. The search keywords/terms were generated from previous similar studies as well as from initial pilot searches conducted by the study team members (Table 1).

The study included original observational and intervention studies published in peer-reviewed journals. Other inclusion criteria were:

• Articles with a focus on vaccination (all types of vaccines) and dis-misinformation spread.

• Studies that have any objective(s) that deal with drivers (implicitly or explicitly) of vaccine mis/disinformation.

• Articles written in English or Spanish languages.

• Studies from all fields of knowledge. Not limited to health or communication sectors.

Our strategy excluded review articles of all types as well as opinion papers, position, conceptual, or argumentative papers without original empirical evidence, book chapters, theses, clinical trials, conference papers, reports, letters, editorials, comments, and textbooks. We also excluded studies with no objective focused on drivers of dis/misinformation in the media or those that were not related to vaccine misinformation, disinformation, fake news, or conspiracy theories. All the authors agreed on the inclusion and exclusion criteria applied in the studies and two actively participated in the entire article selection process.

First, searches were conducted on individual strings concerning the four major components of the study (vaccine, media, mis/disinformation, and drivers). After these four initial searches were conducted using the ´´or´´ Boolean operator and the ´´*´´ truncator operator, a fifth search combining these four initial searches using the ´´and´´ Boolean operator was conducted to arrive at the results.

Data screening and extraction

Generated search results from the different databases were exported in Zotero-compatible file types for further reference management. We used Zotero reference management software to process the generated results and automatically detect duplicates. A researcher conducted abstract and title screening by applying the stated eligibility criteria to identify potentially relevant studies with active guidance from a second researcher. The full texts of these potentially eligible articles were retrieved and initially reviewed by one of the research team members. This was followed by another phase where this was discussed by a second team member, going over any potential concerns to arrive at full consensus. A plan was in place to resolve any emerging disagreements over the eligibility of studies through discussion with the third team member (see Figure 2).

Figure 2

Synthesis and analysis

The included articles were inductively coded using Microsoft Excel data management and cleaning software. These codes were eventually categorized into different themes based on the level of society to provide a manageable list for discussion of drivers of vaccine misinformation in the media. The major categories for this theme were the drivers, while other categories focused on geography of study, methods employed, the media analyzed, as well as types of vaccines studied.

Data was extracted from the objectives, methods, results, and discussion sections of included studies. The Microsoft Excel data management application was used to organize, systematize, and code studies. Table 2 shows a summary of extracted data grouped according to study, relevant objective, study method, country/continent, media/population group analyzed, sample size, vaccine studied, and conclusion. Qualitative narrative synthesis was employed to discuss the key findings and results. After data extraction, summaries in the form of narrative answers were developed with the review objectives in mind. Qualitative narrative synthesis has been used in previous studies (Catalán-Matamoros et al., 2019) and has proven to provide an effective means of producing an actionable knowledge base to inform further policy and practice (Denyer and Tranfield, 2006).

Results

The search included 27 studies which were fully analyzed, with data inductively extracted using an Excel-based code sheet.

General description of the study sample (analyzed studies)

Our sample reveals an almost stable trend in the number of studies from 2011 with a significant dramatic upward shift in publication from 2022. The majority of the studies (n = 24) were after the pandemic started in 2020, while three were conducted before the COVID-19 (see Table 2, Figure 3).

Figure 3

Types of media studied

The most studied mass medium is the social media (n = 18). No study analyzed strategies or drivers employed in traditional media, including their online versions, which suggests that the traditional media has minimum to no contribution to the diffusion of vaccine mis/disinformation. The most analyzed social media platform in exploring the drivers of mis/disinformation in our sample was Twitter (now X) (n = 12) out of the 18. Other social media platforms studied were Instagram (n = 3), TikTok (n = 1), YouTube (n = 1), and Facebook (n = 1). A study (Okuhara et al., 2018) analyzed the contents of Japanese anti-vax campaigners´ websites. No study in our sample analyzed content from multiple social media platforms.

Geographies studied

In the total sample, roughly 75% (n = 18) were linked to a country or geographic region, while three articles (Hoffman et al., 2019; Schulte-Cloos and Anghel, 2024; Unfried and Priebe, 2024) out of these focused on more than one country. These three included a survey (Unfried and Priebe, 2024) conducted among participants from six African countries- Ghana, Kenya, Nigeria, South Africa, Tanzania, and Uganda. From Europe, Schulte-Cloos and Anghel (2024) conducted an online experiment with participants from Hungary and Romania, while Hoffman et al. (2019) analyzed localized American Facebook content that included data contributions from the US and eight other unnamed countries. Other countries studied included the US (n = 6), while a few other countries appeared once – Iran, France, China, Jordan, Japan, Bangladesh, Finland, Brazil, and Spain.

At the continental level, North America was studied most (n = 7). Other continents studied included Europe (n = 4), Asia (n = 5), South America (n = 1), and Africa (n = 1).

Figure 4 illustrates an uneven distribution of studies across continents, pointing to the dominating power of individual countries in regional/continental discourse. For example, the North American data is dominated by the US, leading to ass regional perspective shaped by US-originated research orientation, media system, methods, as well as politics. In contrast, even though only one study was included from Africa - (Unfried and Priebe, 2024) - five countries were represented, underscoring the value of multi-country studies in presenting representative insights from one region. This further offers a broader policy and cultural context, serving as an example of how to amplify underrepresented voices.

Figure 4

Methodological characteristics of included studies

Most of the studies (n = 19) adopted content analysis as the major design while other studies employed surveys (n = 3), and quasi-experiments (n = 5). We categorized modeling and linguistic analysis, as well as analyses of websites as content analytical studies seeing that their basic strategy was generating manifest content from social media and/or websites. The content analysis focused significantly on using advanced machine-learning techniques to ascertain the drivers of misinformation in the media. Studies monitored social media content and their diffusion patterns significantly by engagement metrics (comments, shares, tweets, retweets, likes) while also exploring the range of the spread of posted vaccine mis/disinformation messages. It was not stated if the posts in this category were texts or images or videos, however, two studies analyzed video content- one on YouTube (Tokojima Machado et al., 2020), and the other on TikTok (Lundy, 2023). The sample sizes in the selected articles varied according to the research designs adopted. The majority of the content analytical studies analyzed posts ranging from 436 tweets to 6 billion tweets. The studies that analyzed videos had smaller samples- 100 videos on TikTok, and 52 videos on YouTube. The other studies with humans as subjects of study (surveys and experiments) had samples ranging from 202 to 5,307 human participants.

Type of vaccine studied

COVID-19 vaccines (n = 19) dominated the discourse, supporting the earlier finding that most of the studies were conducted after the advent of the COVID-19 pandemic. Other vaccines studied included the influenza (n = 1) and the HPV (n = 1) vaccines. A significant proportion of our sample (5 studies, 19%) did not specify particular vaccines. One study (Unfried and Priebe, 2024) focused on more than one vaccine- HPV, Polio, and COVID-19 vaccines.

Drivers of mis/disinformation in the media

Our sample presented a wide array of drivers and motivators of vaccine mis/disinformation in the media. Our findings show a list of 34 emerging drivers coded from the respective studies. The emerging drivers were categorized into four distinct levels of drivers for reference. Our taxonomy is based on the emerging factors discovered from the review. These broad drivers include individual-level drivers, message-level drivers, network/platform-level drivers, and structural-level drivers.

We have defined these levels of drivers as:

• Message-level drivers- The motivators and appeals in the development and crafting of media content that make for easy spread and dissemination of vaccine mis/disinformation in the media.

• Individual-level drivers- These are factors related to personal characteristics (cognitive, demographic, and psychographic) that render individuals liable to disseminate vaccine mis/disinformation in the media.

• Platform-level drivers- These drivers are about the characteristics of the media platforms that allow for and are manipulated to spread vaccine mis/disinformation.

• Structural and societal-level drivers- These are related to broader society and contextual factors that drive vaccine mis/disinformation in the media.

The majority (n = 17) of the studies showed multiple drivers (more than one driver) while the rest (n = 10) had a single driver. We hence coded the occurrence of each driver as single cases under the categories adopted for the review as listed above. We coded a total of 63 cases/observations for the categories. Individual-level drivers had the most cases (n = 26). Occurrences of other categories were message-level drivers (n = 16), platform-level drivers (n = 14), and structural and societal-level drivers (n = 7). This reveals a dominance of individual-level considerations in the spread of vaccine mis/misinformation in the media. Interestingly, even though more content analytical studies typically focus on message-level factors, the individual-level drivers dominated the discourse, suggesting a broad tendency to interpret message content through individual psychology lens (see Supplementary material 1 for full list of specific drivers).

Figure 5 presents observations for specific drivers across the reviewed studies. The message-framing narrative, which discussed the linguistic components of vaccine mis/disinformation content was coded in more studies others (n = 5). These frames included the use of concrete words (Saini et al., 2022); negative emotions and maneuvers (Unlu et al., 2024). Further, trust in sources (n = 4) and profit motives (n = 4) followed closely. The findings align with results about the methods adopted, where content analysis, typically a message-centered design dominated. The regular citation of trust in sources also points to the continued focus on individuals who would share vaccine mis/disinformation messages if they trust a source, as against the accuracy of the message especially if the source is an expert (Di Domenico et al., 2022; Harris et al., 2024), or if they are personally connected to the source with the source (Samya et al., 2023). The quest for gain also drove vaccine mis/disinformation across the media, showing that “super spreaders driven by financial incentives that allow them to profit from misinformation” (Pierri et al., 2023). The data also highlight a long list of less-frequently cited drivers that appeared only once or twice, underscoring the need for further empirical observation to capture dynamics that are not widely represented in literature (see Supplementary material 2 for full list of drivers across categories).

Figure 5

Data in Figures 6a,b show complementary insights into broader vaccine mis/disinformation driver-levels regarding the studies that discovered at least one driver in a category (Figure 6a) and the frequency of observation (Figure 6b) across the dataset. The data implies that although categories could be identified by equal number of studies, it does not necessarily imply its dominance or depth in literature. For instance, the case of message-level and individual-level drivers. This suggests that individual-level drivers (such as trust, age, beliefs) may be significantly influential in driving vaccine mis/disinformation and should therefore be a key focus for designing interventions.

Figure 6

Drivers according to geographies studied

In responding to objective 2, we sought to ascertain the emerging drivers from the different geographies studied. 18 out of the articles included in our review mentioned a focal country, hence our analysis is based on these 18 studies. This analysis was based on the continents represented, rather than individual countries since countries in the same continents most often share similar characteristics that make them amenable to similar experiences and interventions. The continents identified from the studies included Africa, Asia, Europe, North America, and South America. It should be noted that the US was the only country in our sample that was from North America, while Brazil was the only country from South America. Two –North America and Europe- of these five continents had drivers spread across all 4 categories/levels of drivers (Figures 7a,b).

Figure 7a illustrates the disparity in thematic diversity in the drivers of vaccine mis/disinformation across continents. Europe and North America stood out as the most thematically diverse geographies with all four driver categories- individual, message, platform, and societal- represented in their findings. This suggests a developed research capacity as well as greater engagement and understanding of the complexity of the mis/disinformation ecosystem in these regions. This wide representation may also be influenced by the volume of studies, seeing that these two geographies were among the most represented in our sample.

Figure 7

In contrast, Figure 7b shows a more continent-specific picture of driver categories. For example, individual-level drivers drive majority of vaccine mis/disinformation in Asia and Africa, pointing to a more human-centered interpretation of vaccine mis/disinformation drivers. Further aligning with cultural frameworks such as Hofstede’s cultural dimensions theory, which links these regions with collectivist inclinations, a dimension which might have affected how mis/disinformation spreads and how researchers from these geographies explore this pattern. The dominance of message-level drivers in North America further points not only to a developed media ecosystem but could have also been shaped by the design of many of the studies from the region (e.g., content analysis). The limited presence of platform and societal-level drivers in Asia and Africa further suggests an under-exploration of deeper network factors in these contexts, showing a research orientation that might be prioritizing individual behaviors over institutional or technological issues (see Supplementary material 3).

In Europe (n = 4), all the levels of drivers were evident in the sampled studies. Antivaccination promoters in Europe made use of significant events in the development of a health event- The COVID-19 pandemic in this case to drive vaccine misinformation. Faccin et al. (2022) discovered that the antivaccination crusaders used the announcement of the Pfizer vaccine, the documentary- hold-up-, AstraZeneca retraction in Denmark, and health pass in Europe to drive vaccine mis/information spread across the media. In Finland, vaccination mis/disinformation spreads in the media through deliberate attacks on pro-vaccine authorities by responding to posts by these authorities (Unlu et al., 2024), further revealing a low power-distance dimension where hierarchies and authorities can be questioned. This in turn amplifies their messages as followers of these pro-vaccine government agencies. The age of social media accounts (older accounts) (Manuel Noguera-Vivo et al., 2023), and possession of right-wing authoritarian attitudes by users of social media (Schulte-Cloos and Anghel, 2024) also promote the dissemination of vaccine/misinformation across the media in Europe.

North American studies explored different drivers of vaccine mis/disinformation cutting across all the four levels of drivers. For example, misinformation posts that were not flagged and labeled as vaccine misinformation (Calac et al., 2022) had the propensity of being shared across social media (platform-level driver). Profit goals were not exactly part of the drivers of vaccine mis/disinformation in North America, however, message-level factors such as appeal to motherhood emotions (Baker and Walsh, 2023) were used to generate profit through alternative healthcare promotion. Perceived expert cues via medical qualifications (individual-level driver) would increase the propensity of a vaccine mis/disinformation post being shared by North Americans (Di Domenico et al., 2022). Objective content, and concrete words (Saini et al., 2022) also proved to be a major determinant of vaccine mis/disinformation spread in North America. The event of the death of a major baseball player (Hank Aaron) after taking the COVID-19 vaccine also shaped the spread of vaccine mis/disinformation in the media as (Calac et al., 2022) discovered.

Discussion

We sought to synthesize existing evidence around the spread and diffusion of vaccine mis/disinformation in the media. Our review is specifically aimed at identifying the potential drivers of vaccine mis/disinformation across the media, exploring the geographies that have been studied for this, and how the different drivers relate to the different geographical landscapes studied. This was to explore existing gaps in the literature and provide actionable insights that will serve as a springboard for future research efforts and interventions to counter vaccine misinformation in the media. 27 studies published between 1 January 2011 and 30 June 2024 were included based on our criteria after a series of processes as shown in Figure 2.

Preliminary findings show a geometric increase in studies conducted during and after the COVID-19 pandemic, further confirming that the COVID-19 pandemic heralded a new era of mis/disinformation’s popularity in vaccine discourse in literature. Our analysis reveals that the majority of the studies analyzed social media platforms, which confirms the tenets of Veblen and McLuhan’s technological determinism theory (Madaki et al., 2024; Marshall, 1962) that the available technology of an era drives all its civilization, including health communication scholarship. The social media continues to be a major driving force in the spread of vaccine mis/disinformation, making it the most studied media type further justifying the focus of earlier reviews on misinformation (Wang et al., 2019; Skafle et al., 2022) on social media alone. The absence of studies focusing on the traditional media points to this fact.

Drivers of vaccine mis/disinformation in the media

The factors that drive vaccine mis/disinformation in the media were categorized into distinct levels. Individual-level drivers emerged as the most frequently identified motivators of vaccine mis/disinformation in our sample, followed by message-level drivers with the other two –platform and societal levels- appearing less frequently. The findings point to the centrality of individual factors in the spread of vaccine mis/disinformation in the media. These drivers included mistrust in institutions (Hoffman et al., 2019), existing biases in health beliefs (Sharevski et al., 2022), conservative right-wing authoritarian attitudes (Schulte-Cloos and Anghel, 2024), avoidance of cognitive dissonance with mental fatigue of platform users (Mønsted and Lehmann, 2022), and trust in expert sources (Samya et al., 2023). Studies have continuously demonstrated how skepticism and existing biases toward government institutions, and pro-vaccination actors fuel the spread of health mis/disinformation (Jaiswal et al., 2020; Lee et al., 2024). Trust in the source of mis/disinformation holds great potential for a user to disseminate such misinformation without paying cognitive attention to the veracity of the details of such messages since it does not demand too many resources from the decision-maker (Unlu et al., 2024; Siegrist, 2021). Even though a majority of the studies adopted content analysis as research design, which is naturally disposed to message-level analysis, the dominance of individual level occurrence points to the centrality of the agent and interpreter in the information disorder ecosystem as seen in Wardle and Derakhshan (2017).

On the other hand, message-level drivers such as emotional and negative framing of messages accompanied with different emotions, and testimonials from ‘past witnesses’ also strongly promote the spread of vaccine mis/disinformation in the media. The dominance of the message, agent, and interpreter components in the propagation of vaccine mis/disinformation aligns with the general postulation of the agent-message-interpreter framework of information disorders. Previous reviews have discovered the dominance of these factors in the spread of misinformation in other similar areas- health (Wang et al., 2019), COVID-19 (Malik et al., 2023), COVID-19 vaccines (Skafle et al., 2022), further showing a similar trend in vaccine-focused studies and the relevance of the interrelatedness of these drivers in the mis/disinformation discourse.

Although the other categories- platform-level and societal level drivers- were found with less frequency, their occurrence reveal a larger challenge with current interventions, the inability of existing automated debunking mechanisms in stopping the spread of all vaccine mis/disinformation (Schmid and Betsch, 2022; Sun and Ma, 2023; Zhang et al., 2021).

The dataset also provides insights into the sustained relevance of individual-level drivers such as mental as well as its interplay with message-level drivers. While both drivers were identified and cited in equal number of studies (10 each), individual drivers emerged with a higher frequency of observations, suggesting not only wider distribution but also greater relevance and depth in vaccine mis/disinformation discourse. Conversely, message-level drivers such as message framing, appeal to emotions, use of scientific sources, use of concrete and vivid expressions, etc., though appearing in multiple studies, were less frequently observed, suggesting a possibility of an interplay between the two highest ranking drivers. This further implies the possibility of the message-level drivers serving as avenues to infer deeper individual-level motivations since the individual reacts to the content based on their own beliefs as it aligns with certain content in the message. Overall, this position reinforces the critical role the individual –agent and interpreter- in the vaccine mis/disinformation discourse.

Drivers and geography studied

Geographical (and by extension, cultural) peculiarities affect the spread of vaccine mis/disinformation across the media. Factors including the development of the media and social media system, economic development, and media literacy level, which vary widely across different geographies impact the spread of misinformation. When examined at a continental level, the drivers of vaccine misinformation slightly vary, giving nuanced insights that can be leveraged in designing tailor-made interventions for different cultural and/or regional contexts. Continents share similar characteristics in cultural and media consumption patterns, as well as economic and literacy levels. This approach prioritizes the significant role of localized responses in a global crisis that thrives on regional peculiarities. Existing literature buttresses the adoption of public health interventions that are culturally relevant to maximize impact (Grover et al., 2024; Pope et al., 2024; Pastrana et al., 2020).

The prominence of the US in geographies studied aligns with the fact that misinformation has largely been popularized from the US, especially since the 2016 presidential elections (Gaultney et al., 2022; Padda, 2020). Studies from Europe and Asia also ranked quite high in the vaccine misinformation drivers literature, corroborating the prevalence of mis/disinformation in developed Western cultures (Skafle et al., 2022; Li et al., 2023). Africa and South America – two of the world’s least developed continents- were the least researched, in line with a recent review (Skafle et al., 2022). The low output from less-developed regions portends a risk-filled future for these societies, seeing that they might be more susceptible to misinformation due to low media literacy. Low immunization uptake, which is correlated to misinformation is highest in least developed societies (World Bank, 2021) where antivaccination campaigners may likely exploit the ignorance of the populace to drive their cause. This further amplifies the recent calls for more health communication research from low-income societies, given that they account for only 0.27% of global health communication research output (Mheidly and Fares, 2020).

In a broad sense, the review reveals the dominance of individual-level drivers across all continents represented except one- South America, underscoring the significance of personal factors –particularly trust in institutions and sources, age and gender- in driving the spread of vaccine misinformation across different cultural contexts. While cultural contexts might differ, these shared psychological factors are universal. The absence of individual-level drivers in the South American study (Tokojima Machado et al., 2020) could be significantly due to the nature of the study- YouTube videos from anti-vaxxers were analyzed.

Throughout our sample, Europe and North America had the most varied drivers of vaccine misinformation in the media. They both had all four level drivers of vaccine misinformation identified, though granular factors slightly differ. Misinformation, particularly health misinformation has long been associated with more developed countries/continents in the global North (Li et al., 2023), hence the identification of all categories of drivers in these contexts. However, in North America –dominated by the US- message-level drivers are more eminent, implying the level of sophistication in the research ecosystem regarding the availability and application of technological tools for natural language processing (Getzoff, 2023), as well as a developed media ecosystem. The developed status means there is relative freedom of expression, and freedom after expression, unlike less-developed countries. This freedom enables citizens and dwellers alike to propagate what they deem fit on social media, which may be vaccine mis/disinformation, aligning with what Jeremy Bentham recognizes as ‘the liberty of doing mischief’ (Phiri, 2023). This liberty is part of what Hofstede refers to as power distance in his cultural dimensions theory, where the US ranks low (Hofstede, 2011).

In Africa, only individual-level drivers were identified. Though only one study was in our sample, data was collected from six African countries. The prevalence of individual-level drivers such as age, gender, and trust in Africa shows how deep-rooted social beliefs play a role in driving vaccine misinformation. Older males in Africa are more prone to sharing vaccine misinformation in the media. This finding aligns with existing literature that older people are less media literate, and are more inclined to share conspiratorial conjectures given their deeply rooted socio-cultural beliefs (Akello, 2024; Osuagwu et al., 2023), owing to widespread poor media literacy and a paucity of media literacy interventions in the continent (Boshoff and Fafowora, 2024; Cunliffe-Jones et al., 2021).

Similarly, the Asian context, dominated by individual-level drivers, reflects similar regional and cultural characteristics with Africa. While the study designs, experiments, survey, and content analysis are different from that of Africa- survey-, the result reflects deeper dimensions of similarity and focus on individuals as major contributors to the spread of vaccine/disinformation in the media. Asia and Africa’s emphasis on individual-level drivers such as age and gender are traceable to established theories such as the cultural dimensions where group roles (collectivism) define and shape the way individuals act on vaccine mis/disinformation. Individual behaviors in these societies are thus deeply rooted in social expectations. Further, the recurrence of trust and mistrust in authorities, an outworking of high-power distance cultures in the cultural dimensions theory could be indicative of the collectivist tendencies in these cultures. Given that individual behaviors are shaped by collective expectations, it becomes easier for these groups to easily convince members to trust or mistrust authorities by spreading vaccine mis/disinformation. These dimensions therefore explain why individual-level drivers are prevalent in collectivist cultures.

There is a complex connection between message-level, platform-level, and structural/societal-level drivers in South America, particularly Brazilian antivaccination YouTube (Tokojima Machado et al., 2020). The use of links to external social media drives followers to these other platforms where they are fed with mis/disinformation and also get alternative health solutions sold. The antivaccination channel owners also use these platforms to collect testimonials from followers and feed them back into the YouTube channel, which amplifies the spread.

The complex mix of drivers from South America is reflective of the prevailing collective culture in the continent, seeing that Brazil ranks low on individualism in the cultural dimensions theory. In this type of culture, the basis of trust is relationship, which includes testimonials from past users of alternative health products, making the use of WhatsApp and Telegram groups veritable tools for community-driven dissemination of mis/disinformation. Interestingly, the socio-economic dynamics that tend to drive the spread of vaccine mis/disinformation reveal how such low-income settings –where legitimate income streams are not sufficient- propel the development of other avenues such as monetizing mis/disinformation for livelihood.

Study limitations

Our review, despite following standard systematic review and evidence synthesis protocols, fell short in some respects, which might have affected the quality of the results. First, we did not include grey literature, conference papers, and pre-prints in this review. This might have limited the options of available evidence to synthesize, hence possibly leaving out some interesting insights that would have enriched the results. Second, the review was limited to articles in English and Spanish, leaving out studies in other languages, which might have affected the results generated. As a result of this exclusion, certain drivers or regions-specific mis/disinformation patterns might have been underrepresented in this review, leading to an incomplete understanding of what is known about the drivers of vaccine misinformation globally and regionally. While we acknowledge this limitation, research has established that the exclusion of non-English articles has little to no effect on systematic review results (Nussbaumer-Streit et al., 2020).

Furthermore, only three databases –WOS, Scopus, and PubMed- were searched to generate the analyzed studies based on available timelines and interests. While the combination of these three databases produces robust enough results, some relevant studies might have been inadvertently omitted from the review. Additionally, our review, particularly the analysis of drivers according to geographies, was skewed toward broad-level factors to provide an overview, thus limiting deeper-level insights that could have been generated from comparing individual-level factors and how they exactly drive vaccine mis/disinformation in different contexts.

Notwithstanding the foregoing limitations, this review has strengths in advancing evidence around the spread of vaccine mis/disinformation with study duration and data spanning years before the COVID-19 pandemic, evaluating how these drivers of mis/disinformation spread vary and are shaped by different contextual realities based on geography.

Conclusion

Our review shows how drivers of vaccine mis/disinformation are not only thematically diverse, but also shaped by distinct cultural and socioeconomic dynamics across various regions of the world. While individual-level drivers are prominent across most of the continents studied, they were more prominent in regions with collectivist and high power-distance tendencies like Asia and Africa, where individuals’ propensity to spread disinformation is largely shaped by group identity. Conversely, the dominance of message-level drivers in North America shows low power distance, and robust research ecosystem that supports complex analysis of content. These regional variations reiterate the need for tailoring interventions to combat vaccine mis/disinformation that go beyond what is being said or shared, but to the how, where, and within which cultural frame the circulation happens.

Future research directions

The findings from our review open up avenues for further exploration in critical areas that would foster better understanding around the spread of vaccine mis/disinformation.

Theoretically, future research could undertake an in-depth exploration of the interplay between cultural theories such as Hofstede’s dimensions and the spread of vaccine mis/disinformation. Our review has applied the Cultural dimensions theory to explain broad-level differences across geographies; future empirical studies should go beyond category-level analyses to test the direction and strength of cultural characteristics in strengthening the prominence of specific drivers across geographies, with their theoretical implications. In addition, empirical studies could mainstream cultural theories in a narrower geographic classification than ours, since our review also did a broad-level classification of geographies to provide an overview based on continents. The theoretical implications of cultural theories would provide further insights necessary for explaining why specific drivers influence the spread and perception of vaccine mis/disinformation in specific geographies.

In terms of geographic context, there is a need for an equitable distribution of research outputs from the global south- particularly Africa, parts of Asia, and South America. Studies from less-developed continents –such as Africa and South America- remain scarce, omitting critical insights from these contexts, which contribute a significant quota to the global human population, and by extension, vaccine mis/disinformation spread. Despite the fact that the survey from Africa provides insights from six countries, it is not sufficient, as the interpretation or the design of the study could be influenced by the research orientation of the researchers. Limited representation from these parts of the world portends a form of looming challenge in global health, given that they contribute a significant percentage of low vaccine uptake. This status makes these societies even more relevant in providing empirical evidence. The lack of studies from these regions shows the increasing divide between the global north and south in terms of knowledge production. Interventions designed on data generated in other climes will be less potent in combating the challenges on the ground in these continents. Hence, there is an urgent need to conduct studies that not only focus on drivers, but also interventions from these regions.

Future studies from a constructs perspective, should undertake a deep dive into how specific drivers that are present in different geographies affect the spread of vaccine mis/disinformation in their respective cultures. For instance, if age drives vaccine misinformation in Africa and Asia for instance, how age drives misinformation in Africa could be largely different from how it does in Asia. It could be that the younger population are more susceptible in Asia, compared to Africa where the older population may be susceptible, showing deeper dynamics that can foster interventions that propel positive change.

Furthermore, our review highlighted the regional distribution of broad vaccine misinformation drivers with little focus on the overlapping nature of these drivers. The finding that message-level drivers could be a lens to eventually infer a connection with individual-level drivers needs further attention. The interconnected and mutually reinforcing nature of these drivers needs to be explored. For instance, how message-level appeals are amplified by platform-level drivers and individual dispositions. The interplay of these relationships in different geographical contexts could present insights that would lead to practical interventions for each society studied. In the same vein, the direction of influence of each specific factor ought to be explored.

From a methodological standpoint, as observed in previous reviews (Whitehead et al., 2023; Skafle et al., 2022; Zhao et al., 2023), the study of health misinformation has largely been conducted using textual and content analysis to identify patterns of spread in the media, with the majority using Twitter data. While this has been because the Twitter API lends itself to research-friendly ends, it is important to note that more insights from other social media platforms would be interesting to compare results between platforms, particularly multimedia-based platforms such as TikTok and YouTube. These platforms have become more popular among the younger population, and if vaccine mis/disinformation must be combated at all levels, a worthy next step would be generating empirical insights from these platforms. The continued dominance of short video content (from YouTube and TikTok) among fact-checked mis/disinformation content points to the urgent need to further explore these platforms (International Fact-Checking Network, 2025). Additionally, more insights need to be generated from locally dominant social media platforms such as the WeChat and Weibo in China.

In the future, ethnographic studies in mis/disinformation-prominent settings that focus on locally motivated drivers of vaccine mis/disinformation, particularly in low-income settings should be a major consideration. This would provide insights beyond self-reported surveys, content analysis, and social or quasi-experiments that have dominated the literature. The ethnographic studies should take a deeper dive into the individual-level drivers identified and explore the connection between cultures and these drivers based on careful observation. Insights generated from these studies, with definitions from the cultural dimensions theory, could provide an avenue where tailored intervention can be designed for specific societies with similar dominant cultural dimensions.

By addressing these gaps in theory, context, construct, and method, future studies can generate evidence that can inform practical interventions that would help combat the spread of vaccine mid/disinformation.

References

• 1

  AkelloT. (2024). Digital literacy and media consumption among different age groups. J. Commun.5, 14–27. doi: 10.47941/jcomm.1973

  • CrossRef
  • Google Scholar

• 2

  AliK.LiC.Zain-ul-abdinK.MuqtadirS. A. (2022). The effects of emotions, individual attitudes towards vaccination, and social endorsements on perceived fake news credibility and sharing motivations. Comput. Hum. Behav.134:107307. doi: 10.1016/j.chb.2022.107307

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 3

  AlievaI.RobertsonD.CarleyK. M. (2023). Localizing COVID-19 misinformation: a case study of tracking twitter pandemic narratives in Pennsylvania using computational network science. J. Health Commun.28, 76–85. doi: 10.1080/10810730.2023.2217102

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 4

  ArgyrisY. A.ZhangN.BashyalB.TanP. N.Using deep learning to identify linguistic features that facilitate or inhibit the propagation of anti- and pro-vaccine content on social media. In: AhamedS.ArdagnaC.BianH.BochicchioM.ChangC.ChangR.et al., editors. 2022 IEEE International Conference on Digital Health (ICDH) (2022). p. 107–116. IEEE: Spain

  • Google Scholar

• 5

  BakerS. A.WalshM. J. (2023). A mother’s intuition: it’s real and we have to believe in it’: how the maternal is used to promote vaccine refusal on Instagram. Inf. Commun. Soc.26, 1675–1692. doi: 10.1080/1369118X.2021.2021269

  • CrossRef
  • Google Scholar

• 6

  BoshoffP.FafoworaB. (2024). Digital media literacy in Africa: towards a research agenda. Afr. J. Stud.45, 259–269. doi: 10.1080/23743670.2025.2478460

  • CrossRef
  • Google Scholar

• 7

  BrodaE.StrömbäckJ. (2024). Misinformation, disinformation, and fake news: lessons from an interdisciplinary, systematic literature review. Ann. Int. Commun. Assoc.48, 139–166. doi: 10.1080/23808985.2024.2323736

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 8

  CalacA. J.HauptM. R.LiZ.MackeyT. (2022). Spread of COVID-19 vaccine misinformation in the ninth inning: retrospective observational Infodemic study. JMIR Infodemiol.2:e33587. doi: 10.2196/33587

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 9

  CarrieriV.MadioL.PrincipeF. (2019). Vaccine hesitancy and (fake) news: quasi-experimental evidence from Italy. Health Econ.28, 1377–1382. doi: 10.1002/hec.3937

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 10

  Catalán-MatamorosD.ParienteA.Elías-PérezC. (2019). What we know about media communication on antibiotics and antimicrobial resistance: a systematic review of the scientific literature. Patient Educ. Couns.102, 1427–1438. doi: 10.1016/j.pec.2019.03.020

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 11

  CDC. (2024) Measles (Rubeola). Measles cases and outbreaks. Available online at: https://www.cdc.gov/measles/data-research/index.html (Accessed November 19, 2024)

  • Google Scholar

• 12

  Cunliffe-JonesP.DiagneA.FinlayA.GayeS.GichungeW.OnumahC.et al. (2021). Misinformation policy in sub-Saharan Africa. London: University of Westminster Press.

  • Google Scholar

• 13

  DaradkehM. (2022). Analyzing sentiments and diffusion characteristics of COVID-19 vaccine misinformation topics in social media: a data analytics framework. Int. J. Bus. Anal.9:22. doi: 10.4018/IJBAN.292056

  • CrossRef
  • Google Scholar

• 14

  DenyerD.TranfieldD. (2006). Using qualitative research synthesis to build an actionable knowledge base. Manag. Decis.44, 213–227. doi: 10.1108/00251740610650201

  • CrossRef
  • Google Scholar

• 15

  Di DomenicoG.NunanD.PitardiV. (2022). Marketplaces of misinformation: a study of how vaccine misinformation is legitimized on social media. J. Public Policy Mark.41, 319–335. doi: 10.1177/07439156221103860

  • CrossRef
  • Google Scholar

• 16

  DunnA. G.LeaskJ.ZhouX.MandlK. D.CoieraE. (2015). Associations between exposure to and expression of negative opinions about human papillomavirus vaccines on social media: an observational study. J. Med. Internet Res.17:e144. doi: 10.2196/jmir.4343

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 17

  EddyJ. J.SmithH. A.AbramsJ. E. (2023). Historical lessons on vaccine hesitancy: smallpox, polio, and measles, and implications for COVID-19. Perspect. Biol. Med.66, 145–159. doi: 10.1353/pbm.2023.0008

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 18

  EichmanD.BichianuD. (2024). From cotton Mather to Dr Fauci: historical markers of vaccine hesitancy. NeoReviews25, e187–e192. doi: 10.1542/neo.25-4-e187

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 19

  European Commission. (2018). Tackling disinformation: a European approach. Available online at: https://eur-lex.europa.eu/legal-content/EN/TXT/PDF/?uri=CELEX:52018DC0236&from=EN (Accessed December 11, 2024).

  • Google Scholar

• 20

  FaccinM.GargiuloF.Atlani-DuaultL.WardJ. K. (2022). Assessing the influence of French vaccine critics during the two first years of the COVID-19 pandemic. PLoS One17:1157. doi: 10.1371/journal.pone.0271157

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 21

  GaultneyI. B.SherronT.BodenC. (2022). Political polarization, misinformation, and media literacy. J. Media Lit. Educ.14, 59–81. doi: 10.23860/JMLE-2022-14-1-5

  • CrossRef
  • Google Scholar

• 22

  GetzoffM.. (2023) Global finance magazine. Most technologically advanced countries in the world 2023. Available online at: https://gfmag.com/data/non-economic-data/most-advanced-countries-in-the-world/ (Accessed November 18, 2024).

  • Google Scholar

• 23

  GroverH.NourR.PowellL.. (2024) Online interventions addressing health misinformation: protocol for a scoping review. Available online at: https://osf.io/preprints/osf/mfujb_v1. (Accessed November 18, 2024)

  • Google Scholar

• 24

  HarrisM. J.MurtfeldtR.WangS.MordecaiE. A.WestJ. D. (2024). Perceived experts are prevalent and influential within an antivaccine community on twitter. PNAS Nexus3:pgae007. doi: 10.1093/pnasnexus/pgae007

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 25

  HoffmanB. L.FelterE. M.ChuK. H.ShensaA.HermannC.WolynnT.et al. (2019). It’s not all about autism: the emerging landscape of anti-vaccination sentiment on Facebook. Vaccine37, 2216–2223. doi: 10.1016/j.vaccine.2019.03.003

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 26

  HofstedeG. (2011). Dimensionalizing cultures: the Hofstede model in context. Online Read. Psychol. Cult.2:8. doi: 10.9707/2307-0919.1014

  • CrossRef
  • Google Scholar

• 27

  HofstedeG.BondM. H. (1988). The Confucius connection: from cultural roots to economic growth. Organ. Dyn.16, 5–21. doi: 10.1016/0090-2616(88)90009-5

  • CrossRef
  • Google Scholar

• 28

  HofstedeG.HofstedeG. J.MinkovM. (2010). Cultures and organizations: Software of the mind: Intercultural cooperation and its importance for survival. New York: McGraw-Hill.

  • Google Scholar

• 29

  HuoJ.TurnerK. (2019). “Social media in health communication” in Social web and health research. eds. BianJ.GuoY.HeZ.HuX. (Cham: Springer International Publishing), 53–82.

  • Google Scholar

• 30

  International Fact-Checking Network (2025). State of the fact-checkers report 2024. Florida: Poynter Institute.

  • Google Scholar

• 31

  JaiswalJ.LoSchiavoC.PerlmanD. C. (2020). Disinformation, misinformation and inequality-driven mistrust in the time of COVID-19: lessons unlearned from AIDS denialism. AIDS Behav.24, 2776–2780. doi: 10.1007/s10461-020-02925-y

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 32

  LeeS.Jones-JangS. M.ChungM.LeeE. W. J.DiehlT. (2024). Examining the role of distrust in science and social media use: effects on susceptibility to COVID misperceptions with panel data. Mass Comm. Soc.27, 653–678. doi: 10.1080/15205436.2023.2268053

  • CrossRef
  • Google Scholar

• 33

  LiX.LyuW.SallehS. M. (2023). Misinformation in communication studies: a review and bibliometric analysis. Jurnal Komun.39, 467–488. doi: 10.17576/JKMJC-2023-3904-25

  • CrossRef
  • Google Scholar

• 34

  LuJ.XiaoY. (2024). Heuristic information processing as a mediating factor in the process of exposure to COVID-19 vaccine information and misinformation sharing on social media. Health Commun.39, 2779–2792. doi: 10.1080/10410236.2023.2288373

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 35

  LundyM. (2023). TikTok and COVID-19 vaccine misinformation: new avenues for misinformation spread, popular infodemic topics, and dangerous logical fallacies. Int. J. Commun.17, 3364–3387.

  • Google Scholar

• 36

  MadakiW. M.KaigamaK. P.AyagwaF. N. (2024). Issues in broadcasting and technological determinism: a review. Int. J. Humanit. Educ. Soc. Sci.2, 231–248. doi: 10.58578/ijhess.v2i2.3301

  • CrossRef
  • Google Scholar

• 37

  MalikA.BashirF.MahmoodK. (2023). Antecedents and consequences of misinformation sharing behavior among adults on social media during COVID-19. SAGE Open13:21582440221147022. doi: 10.1177/21582440221147022

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 38

  Manuel Noguera-VivoJ.del Mar Grandio-PerezM.Villar-RodriguezG.MartinA.CamachoD. (2023). Disinformation and vaccines on social networks: behavior of hoaxes on twitter. Rev. Lat. Comun. Soc.81, 44–62. doi: 10.4185/RLCS-2023-1820

  • CrossRef
  • Google Scholar

• 39

  MarshallM. (1962). The Gutenberg galaxy: The making of typographic man. Toronto: Toronto Press, Scholarly Publishing Division.

  • Google Scholar

• 40

  MheidlyN.FaresJ. (2020). Health communication in low-income countries: a 60-year bibliometric and thematic analysis. J. Educ. Health Promot.9:163. doi: 10.4103/jehp.jehp_384_20

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 41

  MiriA.Karimi-ShahanjarinA.AfshariM.TapakL.BashirianS. (2024). The impact of emotional vs rational message framing on social media users’ detection and sharing of misinformation: an experimental study. J. Inf. Commun. Ethics Soc.22:124. doi: 10.1108/JICES-10-2023-0124

  • CrossRef
  • Google Scholar

• 42

  MønstedB.LehmannS. (2022). Characterizing polarization in online vaccine discourse—a large-scale study. PLoS One17:3746. doi: 10.1371/journal.pone.0263746

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 43

  MoranR. E.GrassoI.KoltaiK. (2022). Folk theories of avoiding content moderation: how vaccine-opposed influencers amplify vaccine opposition on Instagram. Soc. Media Soc.8:4252. doi: 10.1177/20563051221144252

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 44

  MoranR. E.SwanA. L.AgajanianT. (2024). Vaccine misinformation for profit: conspiratorial wellness influencers and the monetization of alternative health. Int. J. Commun.18, 1202–1224.

  • Google Scholar

• 45

  Morejón- LlamasN. (2023). Characteristics and discursive axes of misinformation and the fact-checking process on COVID-19 vaccines in Latin America. Rev. Esp. Comun. Salud43, S47–S61. doi: 10.20318/recs.2023.7005

  • CrossRef
  • Google Scholar

• 46

  NickersonC. (2023) Hofstede’s cultural dimensions theory & examples. Available online at: https://www.simplypsychology.org/hofstedes-cultural-dimensions-theory.html (Accessed June 2, 2025)

  • Google Scholar

• 47

  Nussbaumer-StreitB.KleringsI.DobrescuA. I.PersadE.StevensA.GarrittyC.et al. (2020). Excluding non-English publications from evidence-syntheses did not change conclusions: a meta-epidemiological study. J. Clin. Epidemiol.118, 42–54. doi: 10.1016/j.jclinepi.2019.10.011

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 48

  NwachukwuG.RihanA.NwachukwuE.UdumaN.ElliottK. S.TirunehY. M. (2024). Understanding COVID-19 vaccine hesitancy in the United States: a systematic review. Vaccine12:747. doi: 10.3390/vaccines12070747

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 49

  OkuharaT.IshikawaH.KatoM.OkadaM.KiuchiT. (2018). A qualitative analysis of the beliefs of Japanese anti-influenza vaccination website authors. Heliyon4:e00609. doi: 10.1016/j.heliyon.2018.e00609

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 50

  OsuagwuU. L.MashigeK. P.Ovenseri-OgbomoG.EnvuladuE. A.AbuE. K.MinerC. A.et al. (2023). The impact of information sources on COVID-19 vaccine hesitancy and resistance in sub-Saharan Africa. BMC Public Health23:38. doi: 10.1186/s12889-022-14972-2

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 51

  PaddaK. (2020). Fake news on twitter in 2016 U.S. presidential election: a quantitative approach. J. Intell. Confl. Warfare3:26. doi: 10.21810/jicw.v3i2.2374

  • CrossRef
  • Google Scholar

• 52

  PastranaN.Lazo-PorrasM.MirandaJ. J.BeranD.SuggsL. S. (2020). Social marketing interventions for the prevention and control of neglected tropical diseases: a systematic review. Ekpo UF, editor.PLoS Negl. Trop. Dis.14:e0008360. doi: 10.1371/journal.pntd.0008360

  • CrossRef
  • Google Scholar

• 53

  PhiriC. (2023). Political disinformation and freedom of expression. Painosalama, Turku, Finland: University of Turku.

  • Google Scholar

• 54

  PierriF.DeVernaM. R.YangK. C.AxelrodD.BrydenJ.MenczerF. (2023). One year of COVID-19 vaccine misinformation on twitter: longitudinal study. J. Med. Internet Res.25:227. doi: 10.2196/42227

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 55

  PopeJ.ByrneP.DevaneD.PurnatT.DowlingM. (2024). Health misinformation: protocol for a hybrid concept analysis and development [version 2; peer review: 1 approved, 1 approved with reservations]. HRB Open Res.5:70. doi: 10.12688/hrbopenres.13641.2

  • CrossRef
  • Google Scholar

• 56

  PosettiJ.BontchevaK. (2020). Disinfodemic: deciphering COVID-19 disinformation. London: UNESCO.

  • Google Scholar

• 57

  PraveenkumarB. (2024). Misinformation and disinformation: unravelling the web of deceptive information. J Law Leg Res Dev1, 29–33. doi: 10.69662/jllrd.v1i1.7

  • CrossRef
  • Google Scholar

• 58

  PutriK. Y. S.KuswarnoE.FathurahmanH.MutiaraP. (2023). Interplay health communication in new media in healthy living attitudes in Indonesia. Informasi53, 83–92. doi: 10.21831/informasi.v53i1.58894

  • CrossRef
  • Google Scholar

• 59

  SainiV.LiangL. L.YangY. C.LeH. M.WuC. Y. (2022). The association between dissemination and characteristics of pro−/anti-COVID-19 vaccine messages on twitter: application of the elaboration likelihood model. JMIR Infodem.2:e37077. doi: 10.2196/37077

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 60

  SamyaS. S. R.TonmoyM. D. S. I.RabbiM. D. F. (2023). A cognitive behavior data analysis on the use of social media in global south context focusing on Bangladesh. Sci. Rep.13:4236. doi: 10.1038/s41598-023-30125-w

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 61

  SchmidP.BetschC. (2022). Benefits and pitfalls of debunking interventions to counter mRNA vaccination misinformation during the COVID-19 pandemic. Sci. Commun.44, 531–558. doi: 10.1177/10755470221129608

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 62

  Schulte-CloosJ.AnghelV. (2024). Right-wing authoritarian attitudes, fast-paced decision-making, and the spread of misinformation about COVID-19 vaccines. Polit. Commun.41, 608–626. doi: 10.1080/10584609.2023.2291538

  • CrossRef
  • Google Scholar

• 63

  SchwartzJ. L. (2012). New media, old messages: themes in the history of vaccine hesitancy and refusal. AMA J. Ethics14, 50–55. doi: 10.1001/virtualmentor.2012.14.1.mhst1-1201

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 64

  Serge AndigemaA.Tania CyrielleN. N.. (2024) Addressing vaccine hesitancy: a review of factors contributing to vaccine refusal and effective communication strategies for promoting vaccine acceptance. Available online at: https://www.preprints.org/manuscript/202407.1224/v1 (Accessed November 19, 2024).

  • Google Scholar

• 65

  SharevskiF.HuffA.JachimP.PieroniE. (2022). (Mis)perceptions and engagement on twitter: COVID-19 vaccine rumors on efficacy and mass immunization effort. Int. J. Inf. Manag. Data Insights2:100059. doi: 10.1016/j.jjimei.2022.100059

  • CrossRef
  • Google Scholar

• 66

  SiegristM. (2021). Trust and risk perception: a critical review of the literature. Risk Anal.41, 480–490. doi: 10.1111/risa.13325

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 67

  SkafleI.Nordahl-HansenA.QuintanaD. S.WynnR.GabarronE. (2022). Misinformation about COVID-19 vaccines on social media: rapid review. J. Med. Internet Res.24:e37367. doi: 10.2196/37367

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 68

  SunM.MaX. (2023). Combating health misinformation on social media through fact-checking: the effect of threat appraisal, coping appraisal, and empathy. Telemat. Inform.84:102031. doi: 10.1016/j.tele.2023.102031

  • CrossRef
  • Google Scholar

• 69

  Tokojima MachadoD. F.de SiqueiraA. F.GitahyL. (2020). Natural stings: selling distrust about vaccines on Brazilian YouTube. Front. Commun.5:577941. doi: 10.3389/fcomm.2020.577941

  • CrossRef
  • Google Scholar

• 70

  UnfriedK.PriebeJ. (2024). Who shares fake news on social media? Evidence from vaccines and infertility claims in sub-Saharan Africa. PLoS One19:e0301818. doi: 10.1371/journal.pone.0301818

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 71

  UnluA.TruongS.SawhneyN.SivelaJ.TammiT. (2024). Long-term assessment of social amplification of risk during COVID-19: challenges to public health agencies amid misinformation and vaccine stance. J. Comput. Soc. Sci.7, 809–836. doi: 10.1007/s42001-024-00257-8

  • CrossRef
  • Google Scholar

• 72

  WangY.McKeeM.TorbicaA.StucklerD. (2019). Systematic literature review on the spread of health-related misinformation on social media. Soc. Sci. Med.240:112552. doi: 10.1016/j.socscimed.2019.112552

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 73

  WardleC.DerakhshanH. (2017). Information disorder: toward an interdisciplinary framework for research and policy making. Strasbourg, Cedex: Council of Europe.

  • Google Scholar

• 74

  WhiteheadH. S.FrenchC. E.CaldwellD. M.LetleyL.Mounier-JackS. (2023). A systematic review of communication interventions for countering vaccine misinformation. Vaccine41, 1018–1034. doi: 10.1016/j.vaccine.2022.12.059

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 75

  WHO. (2019) Ten threats to global health in 2019. Available online at: https://www.who.int/news-room/spotlight/ten-threats-to-global-health-in-2019 (Accessed November 19, 2024)

  • Google Scholar

• 76

  WHO. (2020) Working together to tackle the “infodemic”. Available online at: https://www.who.int/europe/news/item/29-06-2020-working-together-to-tackle-the-infodemic-. (Accessed November 19, 2024)

  • Google Scholar

• 77

  WHO. (2024) Vaccines and immunization. Available online at: https://www.who.int/health-topics/vaccines-and-immunization (Accessed November 18, 2024)

  • Google Scholar

• 78

  WHO SAGE Working Group on Vaccine Hesitancy (2014). Appendices to the report of the SAGE working group on vaccine hesitancy. Geneva: World Health Organization.

  • Google Scholar

• 79

  World Bank. (2021) “Absolutely Unacceptable” Vaccination Rates in Developing Countries. Available online at: https://www.worldbank.org/en/news/podcast/2021/07/30/-absolutely-unacceptable-vaccination-rates-in-developing-countries-the-development-podcast (Accessed November 18, 2024)

  • Google Scholar

• 80

  World Health Organization (2019). Global vaccine action plan and decade of vaccines review and lessons learned reports. Geneva: World Health Organization.

  • Google Scholar

• 81

  World Health Organization SAGE Working Group (2014). Report of the SAGE working group on vaccine hesitancy. Geneva: WHO.

  • Google Scholar

• 82

  XueH.GongX.StevensH. (2022). COVID-19 vaccine fact-checking posts on Facebook: observational study. J. Med. Internet Res.24:e38423. doi: 10.2196/38423

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 83

  YaoY. (2024). The role of new media technologies in epidemics – taking the 2014 Ebola outbreak as an example. Highl. Sci. Eng. Technol.109, 249–255. doi: 10.54097/whrnhe60

  • CrossRef
  • Google Scholar

• 84

  ZhangJ.FeatherstoneJ. D.CalabreseC.WojcieszakM. (2021). Effects of fact-checking social media vaccine misinformation on attitudes toward vaccines. Prev. Med.145:106408. doi: 10.1016/j.ypmed.2020.106408

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 85

  ZhaoS.HuS.ZhouX.SongS.WangQ.ZhengH.et al. (2023). The prevalence, features, influencing factors, and solutions for COVID-19 vaccine misinformation: systematic review. JMIR Public Health Surveill.9:e40201. doi: 10.2196/40201

  • Pubmed Abstract
  • CrossRef
  • Google Scholar