According to the World Health Organization (WHO), an infodemic is “an overabundance of information and the rapid spread of misleading or fabricated news, images, and videos” (19). This information overload intensified during the COVID-19 pandemic significantly shaped public perception and decision-making regarding vaccination (13, 14, 29). For instance, Singh et al. (15) found that the excessive circulation of information across media platforms weakened public trust in vaccines and affected their decision to inoculate, while Ouyang et al. (17) identified that greater exposure to social media content increased vaccine hesitancy as individuals struggled to process the overwhelming volume of both, accurate and inaccurate information. Beyond influencing vaccination decisions, the infodemic also had a psychological toll, as the constant exposure to contradictory and excessive information led to anxiety, confusion, and emotional fatigue, further complicating people’s ability to make informed health choices (14, 30).

The sheer volume of information circulating during the COVID-19 pandemic, as described in the concept of the infodemic, presented a complex challenge. While the term broadly captures the overwhelming influx of information, it is crucial to dissect the nature of this information to understand its precise impact. Simply acknowledging an ‘overabundance’ does not address the varying degrees of accuracy and intent behind the disseminated content. It becomes necessary to differentiate between various forms of information disorder to move beyond the general description of the infodemic and delve into the specific mechanisms that shaped public perception.

The proliferation of information disorder has led to an excessive eagerness to oversimplify the phenomenon under the label of “fake news,” contributing to confusion and a failure to distinguish among the various subtleties of the issue (27, 31–33). Following Wardle and Derakhshan (33), this research conceptualizes “information disorder” as a broad phenomenon of disinformation, misinformation, conspiracy theories, false rumors, and misleading content (33, 34). Wardle and Derakhshan (33) identified three types of information disorder: “Disinformation,” false information created to cause harm; “Misinformation,” false information spread without harmful intent, and “Malinformation,” sharing private, real information with the intent to cause harm.

Previous studies have extensively documented the adverse impact of vaccine information disorders on society (35–38). The correlation between this phenomenon has been confirmed by Montagni et al. (26), who found out that “being ‘anti-vaccination’ or ‘hesitant’, rather than ‘pro-vaccination’, was higher among individuals reporting bad detection of fake news, respectively (…), [and that] the risk of being in ‘hesitant’, rather than ‘pro-vaccination’ was higher among individuals having a bad health literacy score” (26, p. 1). These issues align the WHO decision in 2019 to term “vaccine hesitancy” as a noxious experience and classify it as one of the top ten significant threats to global public health (39).

Globally, it has been demonstrated that social media platforms are the primary channels for disseminating disinformation (12, 27). During the COVID-19 pandemic, the WHO proposed concrete measures for social media companies to address and control false and/or misleading content, such as reporting incorrect information by users themselves (40).

To overcome this threat, the journalistic role performance of fact-checkers has been pivotal (41, 42). Meanwhile, traditional media mostly privileged a locally scoped domain; digital online communication widely opened instant dissemination of information on a broader scale across a diverse range of topics (41, 43–45). Therefore, to regulate information disorders on traditional and social media that could harm citizens by adopting incorrect decisions, fact-checkers news verification task is essential (23, 46). In the case of vaccination processes, it is imperative to scrutinize the phenomenon of information disorder within specific contexts, such as the COVID-19 pandemic.

In the case of vaccine disinformation in the media, there is a lack of studies that analyze this phenomenon. In contrast, a plethora of research has examined the characteristics of information disorder within the context of the COVID-19 pandemic. For example, studies have focused on identifying differences in inaccurate information in several countries, along with components of the type of information disseminated, such as format employed, means of dissemination, primary actors, participants studied, and techniques used, among others (47–49). As a result, scholars agree that the phenomenon may adhere to certain behavioral patterns, such as the election of written text as the most used resource for disseminating inaccurate information (11, 22, 47–49). In addition, the authors agree that vaccine disinformation varies according to particularities in each context studied (11, 50).

On the other hand, scholars have found no consensus regarding the primary social media platform used, the main figure in information disorder, and other variables to consider. Firstly, concerning the leading platform for disseminating information disorder, findings have shown diverse results among Facebook (11, 51), WhatsApp (52), and multiple social media platforms (47, 48). Additionally, scholars have observed that the main social platform varies depending on the country studied (11, 51). Secondly, regarding the leading figures in information disorder, discrepancies have been found between governmental actors (11, 47) and public healthcare actors (49). Thirdly, numerous other variables to consider include analysis levels —including local, national, or global scales—; variable choices for specific targets —such as celebrities, professionals from the public or private sector, government entities, and more—; information format, such as sources used —anonymous, fictional, impersonated, or real—; and topics under discussion —ranging from politics and health to sports and battle conflicts—, among others (52).

A total of 226 fact-checking articles were initially obtained. Only those fact-checks classified as containing inaccurate, false, or decontextualized information were selected from them. Therefore, the final sample was composed of a total of 140 texts. Articles that were not fact-checked were not considered in this research, as they do not align with the research questions.

The selected articles were analyzed based on variables aimed to identify the main characteristics related to false information on vaccination that circulated during COVID-19 in Chile. The operationalization of the variables used to analyze this information disorder was based on previous research during the COVID-19 pandemic (8, 11, 12, 27, 49, 52, 67). The variables considered for this study were operationalized as follows:

The extracted data was analyzed according to the variables operationalized by two coders, who used each fact-checking article as their analysis unit. First, we constructed a coding sheet, including all the operationalized variables. Second, to accomplish this task reliably, we calculated inter-coder reliability (ICR) standards (68) from a selected sample of 26 articles, corresponding to 10% of the original sample. As a result, both coders met a coder agreement ranging from 90.6% (Krippendorff = 0.80) to 100% (Krippendorff = 1.0).

Once the coding phase was completed, we performed statistical analysis focusing on the research questions of this study. Research questions 1, 2, 3, and 4 were answered through descriptive statistics analyses. Meanwhile, the Chi-square test was used to examine the association between variables. When the assumptions of the Chi-square test were violated, the Monte Carlo test was used to ensure the accuracy of the results. In the next section, we will present our study results and, later, discuss the implications of our findings.

Results of our analysis showed that most of the topics came from international sources, corresponding to 73.6% of the sample; meanwhile, the remaining percentage correlated with Chilean sources.

The most used format for the circulation of informational disorders was text, corresponding to 32.9% (n = 49) of the sample, followed by videos (28.6%, n = 40) and photography (18.6%, n = 26). More than one format for the diffusion of inaccurate information corresponded to 19.3% (n = 27) of the cases related to the vaccine and/or the vaccination process against COVID-19. More details are presented in Figure 1.

Concerning the frequency distribution of sources that disseminate inaccurate information, we found out that: (a) 62.9% (n = 88) of the sources corresponded to real people who did or said what was stated; (b) 25.7% (n = 36) of the cases were impersonated sources and, finally, (c) 11.4% (n = 16) of the sources used were anonymous.

No cases of malinformation were registered from the total sample of 140 verifications carried out by Fast Check and Mala Espina. On the contrary, 91.4% (n = 128) of the rectified notes responded to disinformation cases; the remaining cases belonged to misinformation. To illustrate this finding, in Table 1, we present the most representative statements that were found in the fact-checking articles studied.

Regarding the subject of information disorder, 70% (n = 98) of the articles studied corresponded to the field of science and health, which included topics related to vaccination and/or people’s health. In contrast, 17.8% (n = 25) of the cases were on politics and/or government —such as political parties, their members, and governmental matters—; meanwhile, the remaining 9.3% (n = 13) and 2.9% (n = 4) corresponded to inaccurate information linked to elite groups —which considered individuals or groups of people (companies) associated with economic and private power but not from the public sector—, and other areas, respectively.

Finally, the most used information disorder strategy was (a) deception, accounting for 64.3% (n = 90), and (b) decontextualization, corresponding to 32.9% (n = 46) of the sample. Pranks and exaggerations accounted for 1.4% (n = 2), and although the literature acknowledges that both techniques can be highly utilized in disinformation contexts, they were scarcely identified in this study. For more information, see Figure 2.

We additionally analyzed the association between the information source variables and the most used information disorder strategy through chi-square tests. As a result, we identified a significant association when a real source was used with the deception strategy (32.9%, n = 46), followed by the decontextualization of information (27.9%, n = 39). Besides, the supplant and anonymous source information also tended to be more commonly used with the deception technique, accounting for 20.7% (n = 27) and 10.7% (n = 15), respectively.

Results showed that 45% (n = 63) of fact-checking articles had as main topic vaccination side effects. Mainly, in those articles where scientific evidence was not provided, adverse reactions such as fainting or even death were attributed. Secondly, findings revealed that 20% (n = 28) of the articles addressed different topics than those proposed in our variable’s classification. However, these topics were diverse and did not follow a specific trend. Thirdly, 16.4% (n = 23) of the articles addressed the topic of conspiracies by power elites. In them, the most recurring themes were affirming that the pandemic and vaccines were a global plan orchestrated by different power groups. In addition, attacks were constantly carried out on pharmaceutical companies and its managers, accusing them of hiding relevant information. Fourthly, 10.7% (n = 15) of fact-checking articles focused on the components of the vaccines. In this topic, it was recurrent to find disinformation claiming that vaccines were composed of graphene or aborted fetal cells. Finally, 7.9% (n = 11) of the articles studied presented more than one topic. In conclusion, 80% of the articles analyzed were related to the topics of side effects, components, and elite conspiracy, indicating their relevance during the COVID-19 vaccination process. More details are presented in Figure 3.

In addition, a Chi-square test revealed a significant association between the variables topic and the type of technique used. The results indicated that when the main theme of the stories was secondary effects, the most used disinformation technique was deception (25%, n = 35), followed by the decontextualization of the information (20%, n = 28). The trend was maintained in the other areas, although to a lesser extent. Specifically, when the topic was vaccine components, the deception technique was used (8.6%, n = 12), followed by decontextualization (2.1%, n = 3); and when the theme was a conspiracy by power elites, the most used technique was deception (11.4%, n = 16), followed by decontextualization (4.3%, n = 6).

We found that 66.4% (n = 93) of the cases corresponded to non-public health professionals, revealing that key actors were individuals representing a pharmaceutical company or private sector medical professionals. In contrast, individuals coming from the public sector, such as the Ministry of Health or any state institution, accounted only for 10% (n = 14) of the total sample. In addition, 9.3% (n = 13) of the sample were non-political public figures, such as athletes or actors; meanwhile, 8.6% (n = 12) of the time, political actors were the main characters. In 1.4% (n = 2) of the cases, the main figure was not an individual person but a collective, and in 4.3%, no protagonist of any kind was detected within the shared story. More details are presented in Figure 4.

In addition, Chi-square tests revealed a statistically significant association between key actors and topic variables, revealing that the most relevant characters disseminated through misinformation were non-public medical professionals, such as researchers, pharmaceutical professionals, or healthcare personnel. These last characters tended to have the greater presence in all the topics addressed as follows: 35% (n = 49) in topics related to side effects, 7.9% (n = 11) in vaccine components, 8.6% (n = 12) in elite conspiracy, 7.9% (n = 11) in other topics, and 7.1% (n = 10) when more than one of the topics was mentioned.

We identified that 52.9% (n = 74) of the information disorder tended to circulate through more than one media outlet, following this trend: 13.6% (n = 19) circulated exclusively on Twitter, 12.1% (n = 17) on Instagram, 8.6% (n = 12) on Facebook, 5.7% (n = 8) through other types of media like Telegram or websites, 4.3% (n = 6) on WhatsApp, 2.1% (n = 3) through traditional media, and only 0.7% (n = 1) via YouTube. More details are presented in Figure 5.

Firstly, the selected sample is small, reflecting the low number of verifications conducted in the country during the studied period. Therefore, it would be interesting to reply to this same research in other Latin American countries with more vaccine disorder information to study. And secondly, this work focused only on quantitative research areas. Therefore, it is also necessary to examine the qualitative context or frames in which vaccination processes, as in the Chilean case, were lived. Thus, this research needs to be deepened by other health communication scholars.