This systematic review will be conducted in accordance with the guidelines provided in the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) statement [33] (Multimedia Appendix 1). This protocol will be registered with PROSPERO after peer review. The literature search had already commenced at the time of acceptance, and the analysis is planned to be completed by March 30, 2025.

Searches will be conducted across several databases, including PubMed, MEDLINE, CINAHL, PsycINFO, and Web of Science. A combination of keywords, informed by previous studies [34,35], will be used to search the abstracts within these databases: (narrative OR story OR storytelling OR testimonial OR anecdote OR exemplar OR episode OR vignette OR experience) AND (debunk* OR prebunk* OR correct* OR refut* OR rebuttal* OR counter* OR combat* OR retract* OR belief correction OR misinformation rejection) AND (misinformation OR disinformation OR misperception OR mislead* OR misbelief* OR fake OR conspiracy* OR myth* OR rumo* OR fals* OR fraud* OR scam* OR wrong OR inaccurate OR incorrect OR gossip OR hoax* OR fallac* OR infodemic OR misperception OR misconception OR suspicion* OR skeptic* OR unverified OR unscientific OR lie OR anti-vaccin* OR anti-vax* OR anti-scien*). No restrictions will be applied to the publication year during the searches. Additionally, the reference lists of the relevant studies will be reviewed to identify other potentially eligible publications.

Intervention studies that quantitatively evaluated the impact of narrative-based corrections for health-related misinformation will be included. For this systematic review, misinformation will be defined as “information that asserts or implies claims that are inconsistent with the weight of accepted scientific evidence at the time” [36], following the definition proposed in the 2025 Consensus Study by the US National Academies of Sciences, Engineering, and Medicine. This definition emphasizes the scientific status of misinformation, irrespective of the intent behind its dissemination, and provides a consistent foundation for evaluating studies included in this review. Adopting the World Health Organization’s definition of health as “a state of complete physical, mental, and social well-being and not merely the absence of disease or infirmity” [37], a comprehensive interpretation of health-related misinformation will be applied. Therefore, this review will cover topics related to health, such as physical well-being (eg, infectious diseases, obesity, reproductive health) as well as mental and social well-being (eg, nutrition, developmental health); relevant literature in these areas will be included. The eligibility criteria are summarized in Table S1 of Multimedia Appendix 2.

The Rayyan QCRI software [38] will be used for the screening process. Duplicate entries will be automatically identified and removed by this tool. Initially, the titles and abstracts of the studies will be reviewed to determine eligibility based on the selection criteria. The first author (TO) and the second author (HO) will independently assess the titles and abstracts. Any disagreements between the reviewers will be resolved through discussion until a consensus is reached. In cases where consensus cannot be reached, the third author (RY) will be consulted to assist in resolving the issue. Subsequently, the full texts of papers that meet the initial inclusion criteria will be reviewed independently by the first (TO) and second (HO) reviewers. If there are disagreements, the third reviewer (RY) will facilitate resolution through discussion. The screening procedure will be presented using a PRISMA flow diagram.

The first reviewer (TO) will extract descriptive information from the eligible studies into Microsoft Excel, and the second reviewer (RY) will review the extracted data to identify errors. The data to be extracted are outlined in Table S1 of Multimedia Appendix 3. The outcomes to be extracted will include distinct constructs reflecting the effects of corrections, including (1) belief change (eg, shifts in factual accuracy judgments), (2) attitude change (eg, evaluations of health behaviors or interventions), (3) behavioral intention (eg, willingness to vaccinate), and (4) misinformation rejection (eg, explicit rejection or skepticism toward misinformation), following previous studies that differentiate these constructs as separate outcome categories [39-41].

The results will be summarized in a compact table and synthesized through a descriptive and narrative review. Given the anticipated heterogeneity in study designs, narrative intervention formats (eg, testimonial, anecdote, story, vignette), outcome measures (eg, belief change, attitude, behavioral intention), and types of comparators (eg, non-narrative text, factual correction, or control), this review will not conduct a meta-analysis. The methodological and conceptual diversity across studies is expected to preclude meaningful effect size pooling. Instead, a narrative and descriptive synthesis will be conducted to explore patterns in the effectiveness of narrative corrections and to address the predefined research questions, focusing on message-, sender-, and recipient-related factors. The findings will be discussed along with their implications for future research and practice as each of the research questions is addressed.

To evaluate the methodological quality of the eligible studies, the first reviewer (TO) will apply the Cochrane Risk of Bias 2 (RoB-2) tool for assessing the risk of bias in randomized controlled trials [42]. This tool allows for the assessment of the overall risk of bias in each study by examining areas such as randomization processes, allocation concealment, handling of missing data, outcome measurement, and outcome reporting. Each of these domains will be categorized as low risk, some concerns, or high risk based on the responses to several signaling questions. For nonrandomized studies, the Risk Of Bias In Non-randomized Studies - of Interventions (ROBINS-I) tool will be applied to ensure appropriate assessment of methodological quality [43]. The second reviewer (HO) will verify the quality assessment results by using the RoB-2 tool and ROBINS-I tool, and a third reviewer (RY) will be consulted if required. The results of the quality assessment will not be used as a basis for excluding studies from the review.

This systematic review aims to be the first comprehensive examination of previous studies comparing the effectiveness of narrative and nonnarrative approaches in correcting health-related misinformation. However, a simple comparison between these 2 approaches is insufficient. Although the use of narratives to correct health-related misinformation holds theoretical promise, the results of previous studies have been inconsistent [26]. This inconsistency is likely to be due to various factors that influence the effectiveness of narratives. Previous studies have explored message-related factors that may affect narrative effectiveness, such as the inclusion of a refutational ending [30] and conversion narratives [32], in which the narrator undergoes a change in belief. This review will examine the types and effects of these message-related factors. Additionally, studies have identified factors that moderate or mediate the effectiveness of narratives in correcting health-related misinformation. These include sender-related factors such as source credibility [28] and the type of endorser [27] and recipient-related factors, including motivations for using social media [32] and issue involvement [30]. This systematic review will provide insights into the types and effects of moderating and mediating factors on the effectiveness of narrative corrections. In summary, this review will be a comprehensive overview of the effectiveness of narrative interventions and the various factors influencing their success in correcting health-related misinformation, thereby contributing to the development of narrative persuasion theories and more effective future correction strategies.

This systematic review will have limitations. Despite employing a comprehensive search strategy and using search terms based on previous systematic reviews, the possibility of incomplete retrieval of relevant literature cannot be entirely excluded. In addition, this review may overlook relevant studies published in languages other than English. Although a quality assessment of the studies included will be conducted, the findings of this review may still be influenced by the risk of biases present in the studies. Moreover, the limitations inherent in the qualitative synthesis of evidence as opposed to employing meta-analytical techniques should also be recognized.

Over the past few decades, the infrastructure for communication technologies has become increasingly complex, with social media platforms facilitating the easy dissemination of accurate and inaccurate information. Addressing health-related misinformation has become a critical concern in medicine and public health. This systematic review will examine the potential of narratives as a theoretically promising approach to correcting health-related misinformation. It will provide a comprehensive overview of the effectiveness of narrative-based interventions and explore the various message-, sender-, and recipient-related factors that may influence their success. The synthesis of these findings will contribute to theoretical advances in future research on the use of narratives to combat health-related misinformation and inform more effective corrective practices.