The search terms were oriented according to the Population, Intervention, Comparison and Results (PICOS) approach, as shown in Table 1. Studies published between December 2019 and the present were included to ensure that the research focused on misinformation during the COVID-19 pandemic. Language: Only studies published in English were considered due to accessibility and consistency in analysis. Peer-reviewed journal articles, conference papers, and reputable preprints were included to ensure academic rigour. Studies specifically examining misinformation related to COVID-19, including its sources, spread mechanisms, psychological and social impacts, and mitigation strategies, were included. Studies published before December 2019 was excluded as they do not pertain to COVID-19 misinformation. Non-English studies were excluded due to language barriers and potential translation inconsistencies. Opinion pieces, editorials, blog posts, and non-peer-reviewed sources were excluded to maintain academic reliability. Studies that addressed misinformation in general but did not specifically focus on COVID-19 were excluded from the review.

We designed the search strategy with an information specialist using medical subject headings and specific keywords (Table 2). We included articles published in English from December 2019 to December 2024, focusing on the infodemic during the COVID-19 pandemic. Non-English papers were excluded due to language bias, resource constraints for translation, limited accessibility, and inconsistent quality control across languages, which may affect the reproducibility and comparability of findings. We searched four databases (PubMed, Scopus, Web of Science, and Google Scholar) and explored the included studies reference lists. Potential limitations of including only these four databases can include database, exclusion of grey literature, publication bias and indexing limitations. Boolean operators like AND is used to narrow a search by including all specified terms, ensuring that results contain each keyword and OR is used to broaden a search by retrieving results that contain at least one of the specified terms. We first conducted the search on 4 December 2024, and we re-ran the search on 6 Jan 2024. After removing duplicates, two authors independently screened the title, abstract and full text of articles and included eligible articles for evaluation. An independent third author resolved any disagreements. We performed the screening process in Gurugram (Gurugram, Haryana, India).

Two independent researchers extracted the general characteristics of each study and classified them into seven major themes: 1. The Role of Digital Platforms in Amplifying Misinformation, 2. Interventions to Combat the Infodemic. 3. The Role of Trusted Messengers 4. Proactive Regulation of Digital Platforms, 5.Enhancing Health and Media Literacy, 6.Bridging the Digital Divide, 7.Technological Innovations in Misinformation Management. We clustered articles based on similar properties associated with the stated objective and the reported outcomes. Although infodemics were primarily defined as the overabundance of information, usually with a negative connotation, we decided to report data from articles that also described the potential beneficial effects of the massive circulation of information and knowledge during health emergencies. We summarised the challenges and opportunities associated with infodemics and misinformation. A third author verified the retrieved data, and another author resolved any disagreements between the inter-reviewers.

Two authors independently appraised the quality of the included articles using the AMSTAR 2 tool, which consists of 16 domains (Shea et al., 2017). Both reviewers conducted the screening and data extraction independently and in a blinded manner to minimize bias. Each categorical domain was rated using an online platform, and an overall assessment of critical and non-critical domains was obtained. Any inter-rater discrepancies were initially resolved through discussion, and if consensus could not be reached, a third reviewer was consulted for arbitration.

Data extraction followed a clear and structured process, with records initially identified through database searches and additional sources such as hand-searching and references. The PRISMA Flow Diagram outlined the screening process, beginning with 495 records identified through database searches, with an additional 127 duplicate records from other sources. After removing duplicates, 495 records were screened for relevance based on title and abstract. Of these, 310 were excluded because 216 were found to be irrelevant to the research topic, 48 were excluded due to language and accessibility barriers, and 46 lacked empirical data. Following this step, 185 reports were sought for retrieval, but 27 could not be accessed, leaving 158 reports for eligibility assessment.

At the eligibility stage, a detailed evaluation of the 158 reports was conducted, leading to the exclusion of 82 reports. These were removed due to inappropriate study design (Ahmed et al., 2020), lack of an integrated study (Islam et al., 2020), or lack of relevance to the research objectives (Guess et al., 2020). After this rigorous selection process, 76 studies met the inclusion criteria and were incorporated into the final review. The diagram visually represents the systematic approach used in study selection, ensuring transparency and reproducibility in the research process (Figure 3).

A narrative synthesis approach was used to categorize studies into key themes: misinformation spread, impact, and mitigation. Findings were analyzed thematically, integrating qualitative insights and quantitative summaries. Contradictory results were examined for methodological or contextual variations. Intervention effectiveness—fact-checking, media literacy, and regulations—was compared across studies. Insights were mapped to existing misinformation frameworks, providing a comprehensive understanding of its dynamics and implications.

The COVID-19 infodemic, a term describing the rapid and widespread dissemination of misinformation and disinformation during the pandemic, is a complex issue influenced by societal, technological, and psychological factors (WHO, 2020a). While digital platforms played a pivotal role in amplifying misinformation, the effectiveness of interventions to mitigate its impact remains debatable. A critical analysis of the mechanisms driving misinformation spread, the limitations of current strategies, and areas requiring further research is necessary to formulate a more comprehensive response.The amplification of misinformation on digital platforms can be attributed to algorithmic biases that prioritize engagement over accuracy. Studies by Cinelli et al. (2020) and Vosoughi et al. (2018) suggest that emotionally charged misinformation spreads more rapidly than factual content (Cinelli et al., 2020; Vosoughi et al., 2018). However, these studies primarily focus on Western social media landscapes, raising concerns about their generalizability to regions with different digital ecosystems and media consumption patterns. Additionally, while social media platforms have introduced measures to curb misinformation, such as fact-checking partnerships, the effectiveness of these interventions is inconsistent.

While AI can rapidly identify disinformation campaigns, its reliance on pattern recognition increases the likelihood of false positives, particularly when distinguishing between satire and harmful misinformation (Pennycook and Rand, 2022). This contrasts with human fact-checking efforts, which, though slower, provide nuanced contextual understanding. The interplay between these approaches remains a contentious debate, with some scholars arguing for a hybrid model that combines AI efficiency with human oversight to balance speed and accuracy (Zhang et al., 2023). Others highlight the susceptibility of AI systems to adversarial manipulation, where misinformation creators adapt content to evade automated detection, raising concerns about long-term sustainability. Meanwhile, human fact-checking, despite its strengths in contextual analysis, faces challenges related to scalability and biases introduced by individual or institutional perspectives. The debate between AI-driven and human-led approaches underscores the need for a more integrated strategy that considers the strengths and weaknesses of both methodologies. For instance, Pennycook and Rand (2021) found that flagged misinformation was less likely to be shared, yet Guess et al. (2020) demonstrated that such efforts had minimal impact on users entrenched in misinformation echo chambers (Pennycook and Rand, 2021; Guess et al., 2020).

Additionally, fact-checking is not always effective in changing the beliefs of individuals deeply embedded in misinformation echo chambers. Echo chambers, where people are repeatedly exposed to like-minded opinions and selective information, reinforce pre-existing biases and make individuals resistant to correction, even when presented with credible evidence. Psychological factors play a key role in this resistance (WHO, 2020b). Confirmation bias leads people to seek, interpret, and remember information in ways that align with their existing beliefs while dismissing contradictory facts. The backfire effect can also occur, where direct confrontation with fact-checks strengthens rather than weakens false beliefs. Emotional investment in misinformation, particularly when linked to identity or ideology, further reinforces resistance to correction (Pennycook and Rand, 2022). Technological factors also contribute to this challenge. Social media algorithms prioritize engagement, often amplifying misleading content and reinforcing belief systems within closed networks. When fact-checks appear in such environments, they may be rejected outright or perceived as biased attacks, especially if they come from sources that individual’s distrust. Research suggests that while fact-checking remains a valuable strategy, it must be combined with other approaches for greater impact. Media literacy programs can help people critically evaluate information before they form rigid beliefs. Narrative-based corrections, where misinformation is debunked through storytelling rather than direct contradiction, have shown promise in overcoming resistance. Encouraging open dialogue and trust-building within communities may also help reduce misinformation’s grip (Kbaier et al., 2024; Cinelli et al., 2020). This contradiction suggests that fact-checking alone is insufficient, particularly when cognitive biases and ideological predispositions influence information consumption.

Another major intervention—trusted messengers—has shown promise in countering misinformation, yet its success is contingent on cultural and contextual factors. Research by MacKay et al. highlights the credibility of healthcare professionals and community leaders in disseminating accurate information (MacKay et al., 2022). However, the assumption that trust in these figures translates to behavioral change is problematic. A case study from India by Sundaram et al. demonstrated that community health workers effectively addressed vaccine hesitancy through direct engagement (Sundaram et al., 2023). Nevertheless, this approach may not be scalable in urban or digitally interconnected populations, where misinformation circulates rapidly and personal interactions are limited (Journal of Primary Care Specialties, 2021). Furthermore, there is insufficient research on whether trust in experts extends to digital platforms, where misinformation thrives.

Regulatory measures targeting digital platforms have also been proposed to curb the infodemic, yet their implementation remains contentious. Germany’s NetzDG law mandates the removal of illegal content within 24 h, a model cited as effective in reducing hate speech (Library of Congress, 2021; Human Rights Watch, 2018a). However, concerns about censorship and freedom of expression complicate its adoption on a global scale. Furthermore, the lack of transparency in how platforms determine what constitutes misinformation raises ethical and practical dilemmas. A standardized, international approach to content moderation is necessary, yet the feasibility of such a framework remains uncertain given the divergent regulatory environments across countries (Trengove et al., 2022).

The contradiction between the effectiveness of digital literacy campaigns and the persistence of misinformation-related beliefs presents a critical challenge in misinformation research. Some studies advocate for digital literacy as a key intervention, arguing that training individuals to critically evaluate information sources reduces their susceptibility to false claims (Finland Toolbox, 2024). Programs like Finland’s media education initiative have been highlighted as promising models that integrate critical thinking into curricula, fostering long-term resilience against misinformation. However, other studies suggest that belief persistence—where individuals cling to preexisting views despite corrective information—undermines the impact of such initiatives. Cognitive biases, such as the backfire effect and motivated reasoning, may lead people to reject or reinterpret corrective messages in ways that reinforce their existing beliefs. This discrepancy raises questions about whether digital literacy efforts can significantly alter misinformation consumption patterns or whether they merely benefit those already inclined toward critical engagement. Furthermore, there is a need to examine how digital literacy interventions interact with different sociocultural and psychological factors, as well as their scalability in diverse populations. Addressing these contradictions requires a more nuanced approach that accounts for cognitive resistance to factual corrections and the broader social dynamics of misinformation spread.

Furthermore, inconsistencies in research findings highlight the need for adaptive policy frameworks. Policymakers must consider variations in audience responses, the influence of social media algorithms, and the trustworthiness of fact-checking organizations when designing interventions. A rigid, one-size-fits-all approach may fail to address the complexity of misinformation spread. Instead, policies should be evidence-driven and flexible, incorporating ongoing research to refine strategies over time.

Artificial intelligence models have shown promise in identifying disinformation campaigns. However, their reliance on pattern recognition can lead to challenges, such as distinguishing between satire and harmful misinformation. For instance, a study revealed that existing fake news detectors are more likely to flag AI-generated content as false, while often misclassifying human-written fake news as genuine, indicating a bias in detection mechanisms (Ghiurău and Popescu, 2025). The advent of generative AI technologies has facilitated the creation of deepfakes—highly realistic but fabricated content—which poses significant threats to information integrity. These AI-generated media have been implicated in spreading false information across various domains, including politics and health, complicating efforts to maintain information accuracy (Sunil et al., 2025). Automated content moderation systems, while efficient, can inadvertently perpetuate biases present in their training data. This can lead to unjust outcomes, such as the disproportionate removal of content from certain groups. Therefore, incorporating human oversight is crucial to mitigate these biases and ensure fair content moderation practices.

Despite the various strategies employed to combat the COVID-19 infodemic, significant gaps remain in understanding the psychological mechanisms that drive misinformation adoption and resistance to correction. Future research should prioritize comparative studies that examine the effectiveness of interventions across different sociocultural contexts. Additionally, longitudinal studies assessing the durability of fact-checking, media literacy programs, and regulatory measures would provide deeper insights into sustainable solutions. Regulatory measures targeting digital platforms have also been proposed to curb the infodemic, yet their implementation remains contentious. Germany’s NetzDG law mandates the removal of illegal content within 24 h, a model cited as effective in reducing hate speech. However, concerns about censorship and freedom of expression complicate its adoption on a global scale.

Here is a comparative table (Table 3) of different misinformation countermeasures, detailing their effectiveness and limitations (Table 4).

Research into algorithm transparency should aim at developing frameworks for ethical algorithm design. This could involve making algorithmic processes more understandable and accessible to the public, ensuring that platforms are held accountable for the content they promote. Platform-driven interventions, such as X’s Community Notes and Facebook’s misinformation labels, aim to curb misinformation through algorithmic adjustments and user-driven corrections (Tan, 2022; Yu et al., 2024; Dujeancourt and Garz, 2023).

Understanding the psychological factors behind the belief and sharing of misinformation is essential for designing interventions that target the root causes of these behaviors. Psychological theories of belief formation, cognitive biases, and social influence can provide crucial insights into why people are so easily influenced by misinformation. For instance, cognitive biases such as confirmation bias, where individuals seek out information that confirms their pre-existing beliefs, play a critical role in the spread of falsehoods. Emotional responses to misinformation, such as fear or anger, also contribute to its virality, as these emotions increase engagement with content (Pennycook and Rand, 2021; Munusamy et al., 2024).

Misinformation is not bound by borders, and its effects are global. The widespread nature of misinformation, especially on platforms like Facebook, Twitter, and YouTube, means that disinformation can easily cross geographical, political, and cultural boundaries. As a result, tackling the infodemic requires global cooperation and coordination among researchers, policymakers, and technology companies. Future research should focus on fostering international partnerships to share data, research findings, and best practices in combating misinformation (Adams et al., 2023; Desai et al., 2022; New WHO Review Finds, 2022).

While global and national interventions are crucial, community-based strategies are also vital for combating misinformation, especially in regions with limited access to digital literacy resources. Misinformation spreads rapidly in local communities through word-of-mouth, local media, and interpersonal interactions. Therefore, it is essential to evaluate the efficacy of grassroots efforts in building trust and promoting accurate information within communities (Oxford Academic, 2022; Borges Do Nascimento et al., 2022; Stover et al., 2024).

Regulatory and policy measures have been among the most widely discussed approaches to tackling misinformation (Tan, 2022). However, the effectiveness of these measures remains a subject of debate. Some countries have implemented stringent laws aimed at curbing the spread of false information. For example, Singapore’s Protection from Online Falsehoods and Manipulation Act (POFMA) empowers authorities to issue correction orders to platforms and individuals found spreading falsehoods. While such measures have been successful in curbing some forms of misinformation, they have also raised concerns about censorship and the suppression of dissenting voices (Human Rights Watch, 2018b; Protection from Online Falsehoods and Manipulation Act, 2021; Nannini et al., 2024).

However, the future research avenues discussed above may face feasibility challenges. Finally, by assessing strategies to enhance public resilience against misinformation, this research highlights the importance of digital literacy, institutional collaboration, and proactive policy frameworks. These findings provide a foundation for developing robust misinformation mitigation strategies applicable to future health crises, reinforcing the necessity of a multidisciplinary approach in addressing digital misinformation.