Characteristics of adverse events following immunization reporting systems: A scoping review

Abstract

Objective

This study aimed to identify characteristics of adverse events following immunization (AEFI) reporting systems and vaccine adverse events reporting systems (VAERS) including technical platforms, user groups, data elements, functional and non-functional requirements.

Methods

In this scoping review, various databases were searched from 1^st January 2015 to 31^st December 2024, and all types of studies that explained AEFI reporting system/VAERS characteristics were considered. The findings were reported descriptively.

Results

Different technical platforms including web-based, mobile-based, or hybrid platforms were used by multiple user groups. Data elements included personal, clinical, vaccination, and adverse events data. The functional requirements included recording vaccination and adverse events data as well as generating reports. Non-functional requirements were related to system security, data privacy, etc.

Conclusion

This review presented a set of characteristics that has been considered for different AEFI reporting systems and VAERS. The results can be used for designing more comprehensive AEFI reporting systems in different countries. These features along with new digital technologies and analytical tools including artificial intelligence offer more potential to enhance efficiency and effectiveness. Future research should focus on AI-driven methodologies, including natural language processing, machine learning techniques, and predictive analytics, while addressing ethical, regulatory, and practical challenges.

Introduction

Implementing immunization programs globally is one of the most significant public health achievements.^1–3 These programs have proven to significantly reduce the spread of infectious diseases and the associated mortality rates.^3–5 However, vaccines can also be associated with adverse events, as with other medical interventions. In this context, reporting systems for adverse events following immunization (AEFI) have become increasingly significant component of the immunization information system.^6,7 These systems help to ensure vaccine safety, maintain public confidence in vaccination programs, and promote evidence-based decision-making in public health policy.^8–10

While the US vaccine adverse events reporting system (VAERS) represents the most well-established and frequently cited model, various countries have developed their own AEFI reporting systems with different nomenclatures, technical platforms, and specifications (e.g., CNAEFIS in China, VigiFlow in Brazil, SIMUNDU in Indonesia).^11–13 Therefore, in this review, AEFI reporting system has been used as a general term along with VAERS.

Managing adverse events following vaccination reports is essential for thoroughly evaluating vaccine safety across different population groups, including, children, the elderly and people with underlying health conditions.^8,14 By collecting and analyzing data, these systems enable rapid identification of potential threats and ultimately improve public health.^15–17 Given the variety and high volume of vaccinations provided, the ability of these systems to detect rare or unforeseen events is of particular importance, achieved through reporting standards and modern technologies^18,19. Moreover, recent advances in information technology, in particular the use of artificial intelligence and big data analyses, have enabled faster and more accurate processing of adverse events reports.^20,21 Specific applications include natural language processing (NLP) for extracting structured data from unstructured adverse events reports in AEFI reporting systems. As demonstrated by Baer et al., using this technology may lead to a substantial reduction in review time, automated signal detection in large-scale adverse events databases, and predictive analytics for identifying populations at elevated risk of adverse events. ²²

The results showed that the best structure for AEFI reporting systems and VAERS should include clinical information, demographic data, medical history and specific information on the type and timing of vaccination to allow for scientific and accurate interpretation of events.^19,23 In addition, the effective design of these systems requires a deep understanding of the different needs of users, including healthcare workers, patients and regulatory authorities ²⁴ The collected data elements must be accurate and complete to form a basis for scientific analysis and the detection of risk patterns. ²⁵

In addition to the data elements, it is crucial to consider functional requirements such as interoperability with other systems, high volume of data entry and processing efficiency and the ability to create analytical reports. Non-functional requirements, such as information security, privacy protection, usability and high accessibility play an important role in the success and effectiveness of these systems. ²⁶ According to the literature, users are dissatisfied with the complexity of reporting forms and prefer a simple, understandable environment with suitable guidance. ²⁷

One of the fundamental challenges in this field is underreporting, in which only a small proportion of actual adverse events following immunization are recorded. ²⁸ Even in systems designed specifically to record serious incidents, up to 95% of cases go unreported.^29,30 Delays in reporting and a lack of user-friendly interfaces for doctors and patients severely reduce the efficiency of these systems. ²⁹ Another challenge is the use of different definitions and classifications of adverse events in various systems, which makes it difficult to analyze and compare collected data. ³⁰ Insufficient interaction between national vaccination information systems and adverse events reporting systems is another challenge reducing the effectiveness of data analysis.^25,31 Overall, challenges such as weaknesses in data coordination, underutilization of technical capacity and lack of advanced analytics may hinder the effectiveness of these systems. ²¹

Despite ongoing efforts, significant gaps remain in the systematic design of AEFI reporting systems. The most critical deficiency is the lack of a comprehensive analysis of end-user requirements, particularly regarding essential data elements and the balance between functional and non-functional requirements.^23,31,32 Furthermore, many existing systems have been developed without sufficient adherence to data standardization or user-centered design principles,^33,34 which may lead to persistent challenges such as underreporting and poor data interoperability. ³⁵

To our knowledge, previous reviews have primarily focused on specific aspects of immunization information systems^35–37 or adverse events reporting mechanisms,^20,38 often with an emphasis on high-income countries. The current study aimed to complement the existing body of literature by simultaneously investigating technical platforms, user groups, data elements, functional requirements and non-functional requirements of AEFI reporting systems and VAERS in diverse geographical settings. This approach provides a more comprehensive understanding of requirements for AEFI reporting systems that may be applicable across different global contexts, and offer a practical guideline for system design and implementation in future.

Methods

This scoping review was conducted according to Arksey and O’Malley’s framework. ³⁹ Before conducting the research, ethics approval was obtained.

Stage 1: Identifying the research question

A comprehensive understanding of characteristics and user requirements of AEFI reporting systems and VAERS, including data elements and functional and non-functional requirements is essential to enhance quality of adverse events reporting. Therefore, the research question was as follows:

What are the characteristics of AEFI reporting systems and VAERS?

Stage 2: Identifying relevant studies

To identify relevant studies, 7 databases, including Web of Science, PubMed, Scopus, ProQuest, the Cochrane Library, IEEE Xplore, and Google Scholar were searched. The time frame was between 1st January 2015 to 31st December 2024. The search strategy included three main concepts: namely, “Vaccination”,“Vaccine adverse events reporting system”, “Adverse events following immunization reporting systems” and “system design”. MeSH terms, synonyms and other related keywords were also used in developing search strings (Supplementary Table 1). To identify relevant papers, three fields of each database, including title, abstract and keywords were searched. The reference lists of the included articles were also searched to identify additional studies that met our eligibility criteria.

Stage 3: Study selection

This study included all quantitative, qualitative, and mixed-method studies that explained AEFI or VAERS characteristics and were published in English between 2015 and 2024. Papers published in languages other than English, abstracts without access to the full texts, letters to the editor, and papers that did not primarily focus on VAERS or AEFI reporting system design and characteristics were excluded. The retrieved papers were imported into the EndNote version 21, and after removing duplicates, the remaining articles were assessed for title and abstract relevance to the study objective. After removing irrelevant articles, the full texts of the remaining ones were examined separately by two authors, and any disagreements were resolved by the third author.

Stage 4: Charting the data

Data were extracted using a data extraction form that included author(s)’ name(s)/year of publication, country of the study, research objective, research methodology, technical platform, user groups, data elements, functional and non-functional requirements, and results.

Stage 5: Collating, summarizing and reporting the results

A qualitative content analysis approach was used to analyze the extracted data. The charted data were systematically grouped and categorized according to the predefined domains of system technical platforms, user groups, data elements, functional and non-functional requirements. The results were then tabulated to facilitate comparison across studies and summarized descriptively to address the research question. In this study, conducting a meta-analysis was not feasible due to the inherent heterogeneity of the studies; therefore, we used a structured data extraction form and coding process.

Results

In total, 1608 papers were retrieved from databases. After completing the screening process and applying the inclusion and exclusion criteria, 19 studies were identified for the final review. The detailed process of article selection is illustrated in Figure 1.

Figure 1.

Selecting papers based on the PRISMA-ScR checklist. ⁴⁰

Characteristics of the selected studies

The results showed that the highest frequency of the studies (n = 6) was conducted in the United States.^{22,37,41–44} Other studies were completed in Indonesia (n = 3),^35,45,46 Brazil (n = 3),^12,47,48 Canada (n = 3),^36,49,50 Nigeria (n = 1), ⁵¹ Germany (n = 1), ⁵² Kenya (n = 1), ⁵³ China (n = 1). ¹¹ The highest frequency of papers (n = 6) were published in 2023.^{35,37,45,46,49,51}

Table 1 summarizes the articles included in this review, while detailed study characteristics and extended textual descriptions are provided in Supplementary Table S2.

Table 1.

Summary of the selected studies.

No.	Authors/Year	Country	Technical platform	Key findings
1	Staras et al./2024 44	USA	Web-based application integrated with a vaccination registry	High usability and acceptability; improved communication between parents and healthcare providers; real-time integration with vaccination registry
2	Rahmadhan & Handayani/2023 45	Indonesia	Web-based application	Good usability and performance; supported vaccination registration, and vaccination certification
3	Rahmadhan & Handayani/2023 35	Indonesia	Not specified	Fourteen key implementation challenges identified; interoperability and data quality were most frequently reported.
4	Bota et al./2023 49	Canada	Mobile-based application; web-based platform	Higher completion rates observed for the web-based platform (64%) compared with the mobile-based application (45%).
5	Sulistyawati et al./2023 46	Indonesia	Web-based application	Demonstrated scalability and sustainability; interoperability and offline functionality supported successful implementation.
6	Shragai et al./2023 51	Nigeria	Web-based application (national dashboard)	Data integration and triangulation enabled improved national-level decision making on COVID-19 AEFI.
7	Donckels et al./2023 37	USA	Web-based immunization information system	Improved quality and timeliness of vaccination data; challenges included policy heterogeneity and limited interoperability.
8	Huang et al./2021 42	USA	Web-based application	Successfully integrated COVID-19 vaccine data and enabled comparative analysis of adverse events.
9	Atkinson et al./2020 36	Canada	Web-based application	Improved accessibility, interoperability, and public participation in immunization and AEFI reporting
10	Vogler et al./2020 12	Brazil	Web-based application	Implementation of VigiFlow resulted in a 62.6% increase in reported adverse events nationwide.
11	Guimarães et al./2020 47	Brazil	Web-based immunization information system	Moderate usability; navigation and efficiency issues were identified but they were largely correctable.
12	Silva et al./2020 48	Brazil	Web-based immunization information system	Low overall implementation level; major issues related to internet instability and data management.
13	Nguyen et al./2020 54	Germany	Mobile-based application	Mobile application was well accepted (63% positive evaluation); data protection and usability were critical adoption factors.
14	Namageyo-Funa et al./2018 53	Kenya	Web-based electronic health information system	Immunization functionality was insufficient; systems could not fully replace paper-based reporting.
15	Baer et al./2016 22	USA	Text mining system	Natural language processing reduced review time by 58% and achieved high agreement with manual review.
16	Wilson et al./2016 50	Canada	Mobile-based application	Consolidated fragmented immunization records and improved the quality of AEFI reporting.
17	Munoz et al./2015 43	USA	Not specified	Identified and prioritized maternal and fetal/neonatal AEFI; recommended standardized definitions and terminology.
18	Liu et al./2015 11	China	Web-based immunization information system (CNAEFIS)	Collected over 300,000 AEFI reports and strengthened national-level vaccine safety surveillance
19	Baker et al./2015 41	USA	Web-based Electronic health record–based immunization information system	Significantly increased AEFI detection and reporting; improved timeliness and reduced physicians’ workload.

About half of the studies used mixed-methods approaches (n = 10). The remaining studies were categorized as quantitative (n = 6) or qualitative (n = 3). Among the quantitative studies, the methods included randomized controlled trial, ⁴⁹ cross-sectional design, ⁴⁸ quasi-experimental intervention^22,41,50 and descriptive study. ¹¹ The qualitative studies comprised a systematic review ³⁵ and case studies.^12,53

Technical platforms

The primary findings indicated that the most frequently used platforms were web-based applications, mobile-based applications, and hybrid technical platforms. Web-based systems constituted the predominant technical approach, functioning as the core infrastructure for vaccine safety surveillance, adverse events reporting, and comprehensive data management.^{12,36,42,44–46,49,51}

These systems were typically designed to enable centralized data collection, real-time reporting, and analytical functionalities accessible through standard web browsers.^36,42,51

Mobile applications were reported with lower frequency and were predominantly implemented as complementary tools rather than standalone surveillance systems.^49,50,54 In these studies, mobile applications primarily supported data entry, user notifications, or field-level reporting, frequently operating in conjunction with broader web-based infrastructures.^49,50,54

Hybrid systems combined multiple technical components, such as web-based interfaces linked with Electronic Health Records (EHRs) or national registries.^22,41,53 The objective of these platforms was to substantially enhance data interoperability and streamline information exchange across clinical, administrative, and surveillance settings.^22,41 Furthermore, a limited subset of studies described systems, including Electronic Health Record-integrated (EHR-integrated) or text mining–based platforms. These approaches were specifically designed to support automated adverse events reporting and data extraction.^22,41,55

User groups

Seven categories of user groups included healthcare providers, public/vaccine recipients, healthcare managers and policymakers, data entry staff, system administrators, researchers and safety evaluators. Healthcare providers were responsible for recording vaccinations, reviewing adverse events, and monitoring clinical alerts,^{11,35–37,41,43–48,50–53} Physicians,^22,35,43,44 nurses and midwives,^45,47,53 health center staff,^35,51,52 and vaccinators^11,36,46 were the main healthcare providers involved in vaccine safety activities.

The public/vaccine recipients, particularly parents, played a key role in reporting adverse events through mobile-based applications.^{11,36,43–45,49,50,52} Public reporting provided faster and more comprehensive data compared to other reporting channels.^49,52 Healthcare managers and policymakers leveraged data from AEFI reporting systems to evaluate vaccination performance, identify under-vaccinated populations, and formulate policies.^{11,12,35,37,43,46,50,51} Data entry staff and system administrators were responsible for data entry and supporting AEFI reporting systems.^{41,46–48,53} Researchers and safety evaluators were involved in identifying safety signals, analyzing large-scale data, supporting policy development, creating predictive models, and assessing long-term vaccine safety.^{12,22,37,42,43}

Data elements

The results showed that the data elements related to VAERS and AEFI reporting systems could be categorized into four main groups: personal, clinical, vaccination and adverse events data. More than half of the included studies (n=15) reported that personal data were the core data elements of AEFI reporting systems.^{11,22,36,41–46,51,53} These data included the full name, national identification number, date of birth, sex, age, residential address, place of birth, contact number, vaccination location, and pregnancy status. Such characteristics enabled patient identification,^12,36,37 adverse events monitoring, and analysis of vaccination coverage at the community level.^49,51 Some studies highlighted that national identification numbers and vaccine identifiers played a key role in preventing duplicate entries and ensuring data integrity.^37,41,47,48 Clinical data included vaccination history^{22,45,46,49,50} and clinical records including laboratory tests and medications that were considered important for managing and monitoring adverse events. ³⁷

Vaccination data were essential for preventing unnecessary doses.^44,48,49,51 These data included vaccine lot numbers,^22,36,41,50 as well as vaccine name, type, and brand.^11,36,42,45 Some studies indicated that integrating vaccine production data with AEFI reporting systems enabled the identification of safety patterns.^36,42,51 Huang et al. demonstrated that data such as brand name, ⁵¹ manufacturer, ⁴² production and expiration dates, ³⁶ and clinical trial information ⁴² were critical for advanced analyses, such as comparing the efficacy of COVID-19 vaccines. ⁴² Vaccine inventory data were considered crucial for supply chain management and preventing shortages.^36,37

Regarding adverse events data, items such as symptoms, severity, type, and outcomes of adverse events were essential for vaccine safety assessment and regulatory decision-making.^{11,12,41–43} Reporting the type of adverse events (e.g., fever, dizziness),^12,41,43,50 classifying events as serious or non-serious,^22,50,51 and analyzing geographical vaccination coverage ⁴⁸ contributed to vaccine safety evaluation.

Functional requirements

The functional requirements included recording vaccination and adverse events data as well as generating reports. Twelve studies highlighted that recording vaccination data was one of the core functional requirements of the reporting systems.^{36,37,42,45–50,52} This functionality supported the management of immunization programs, the calculation of vaccination coverage rates, and the identification of vaccination gaps. Accurate recording of vaccination histories was also important to increase vaccination rates and reduce unnecessary vaccine doses. ³⁷ Reporting adverse events following immunization was identified as a critical functional requirement for community-level surveillance, policymaking, and evaluation of vaccination program effectiveness.^36,41–53 Several studies described different types of adverse events reports, including classification of events based on type, severity, and outcomes,^{11,12,50–52} periodic reporting, ⁵³ and comparative reporting of adverse events across different vaccines. ⁴² In addition, recording the date and time of adverse events, the reporting user, and the reporting location was considered essential for adverse events analysis.^11,43,50 Eighteen studies reported that management of reporting forms (e.g., creating, deleting, and editing) and analytical dashboards were essential for reporting systems.^{11,12,22,36,37,41–53}

Data exchange and interoperability between AEFI reporting systems and other systems, such as electronic health records (EHR) ^36,37,41,53 and national immunization information systems,^11,45,50 were also found essential. Such integration enables creating comprehensive patient records and supports more accurate data analysis. Data aggregation from multiple data sources,^43,51 data transmission to global adverse events databases, such as WHO VigiBase, ⁵¹ as well as automated ⁴⁴ and scheduled reminder notifications ⁴⁵ were other important functions reported for enhancing performance of reporting systems.

Non-functional requirements

Non-functional requirements of AEFI reporting systems and VAERS included system security and data privacy protection, usability, interoperability with other health information systems, reliability, and accessibility. Fourteen studies reported that system security and data privacy were among the most important concerns for AEFI reporting systems.^{22,35–37,44,47,49–53} Key requirements in this domain included secure data transmission, ⁴⁴ data security,^11,36,51,52 privacy preservation,^11,36,50 user authentication^37,51 and data encryption.^50,53 Several studies emphasized the importance of improving user interface of AEFI reporting systems.^{11,36,41,44,45,47,50–53} Interoperability and integration with other health information systems were identified essential for reducing errors and improving data management.^{36,37,44,45,48,51}

Reliability refers to the ability of a system to maintain functionality and data integrity during internet disconnections or low-quality connectivity. This requirement was considered critical for regions with weak or intermittent communication infrastructure.^{12,46,48,49,52} Accessibility refers to the continuous and stable availability of central services for users. This requirement was considered essential for regions with weak communication infrastructure.^{12,46,48,49,52} Studies indicated that high accessibility significantly increased community participation in AEFI monitoring and reporting.^{11,36,47–49,52} Additional non-functional requirements included data standardization and compliance with national and international regulations,^11,12,43,50 regular data updates,^37,42 scalability,^22,36 data backup, ⁵³ data quality,^43,49 and accuracy.^22,48

Discussion

This scoping review examined the characteristics of AEFI reporting systems and VAERS. According to the results, systems were different in terms of technical platforms, user groups, data elements and functional and non-functional requirements. Together, these features point to the importance of context-sensitive system design that reflects differences in health system maturity, regulatory environments, and digital infrastructure.^{11,12,22,36,37,41–53} Systems in high-income settings typically rely on stable connectivity, established standards, and stronger institutional capacity, enabling richer data capture, web-based workflows, and closer integration with electronic health records (EHR).^{11,22,36,37,41,42,44,49,50,52} In contrast, implementations in middle-and low-income settings prioritise resilience to infrastructural instability, adopting hybrid or mobile approaches to maintain reporting under intermittent connectivity and limited hardware capacity.^{12,35,45–48,51,53}

According to the results, AEFI reporting systems and VAERS were commonly created using web-based, mobile-based, or hybrid technical platforms. Mobile applications were associated with increased user engagement through reminders and notifications, whereas web-based platforms offered broader accessibility and, for some populations, greater perceived ease of use.^32,56 Although Wilson et al. reported an overall preference for mobile applications, ⁵⁰ other findings suggested that web-based systems remain more feasible for specific user groups. ⁵⁷ Similarly, Bettinger et al. showed that web-based VAERS was faster and more practical than telephone-based approaches. ⁵⁸ An example for the hybrid systems was the integrated web and mobile-based VigiPharmacoVax system, which successfully captured pediatric AEFI data. ⁶ Non-digital options, such as health booklet cards and telephone interviews can be complementary approaches in settings with limited digital access. ³⁸

The user groups of these systems were different, each bringing distinct priorities. Healthcare providers, who serve as the primary reporters in most AEFI systems, prioritize clinical relevance, workflow integration, and data accuracy, and often prefer comprehensive data collection forms.^28,59 Vaccine recipients and the public increasingly engage through mobile-based reporting and underline system ease of use and data privacy. However, simplified interfaces may sacrifice clinical details. ²⁹ Healthcare managers and policymakers emphasize standardization and population-level completeness, which may conflict with the local adaptation needs of frontline users. These diverging priorities underscore the need for systems with flexible, role-based interfaces that accommodate clinical, public, and patient perspectives simultaneously.^21,60 System managers and data entry staff prioritize security, reliability, and maintainability. These competing demands must be carefully balanced in system design to achieve optimal outcomes.^{26–29,45,47,53,54,59}

Regarding essential data elements and consistent with Brighton Collaboration and VAERS analyses,^5,61–64 the results showed that patient identifiers, vaccination details, and detailed adverse events descriptions are necessary for reliable monitoring. High-quality vaccination records and inclusion of lot numbers support safety evaluation and supply tracking,^65–68 while integration of post-marketing and clinical trial data broadens evidence for long-term safety. ⁴² Liu et al. showed that classifying adverse events such as distinguishing serious from non-serious events along with statistical and analytical reporting and documentation of events timing and type, provided critical information for regulatory decision-making at the national level. ¹¹ However, strict adherence to externally defined standards may overlook local terminology, regulatory requirements, or health system structures.^25,30 Brazil’s adaptation of the VigiFlow system illustrates how flexible frameworks can preserve core standard data elements while allowing contextual customization. ¹²

In terms of functional requirements, automation continues to be a key requirement for timely detection and regulatory action. According to Chen et al., among the design priorities for low and middle-income settings, automation must be considered as a priority. ⁶⁹ Obviously, barcode-based data entry, EHR interoperability, and duplicate prevention mechanisms remain central to accurate and scalable reporting.^70–73 Shragai et al. demonstrated that web-based dashboards in Nigeria improved decision-making related to adverse events associated with the COVID-19 vaccine. ⁵¹

Non-functional attributes, such as data privacy, usability, interoperability, accessibility, and data quality were found fundamental to sustainable, high trust AEFI systems.^64,74–76 Liu et al. demonstrated that data privacy was essential for maintaining trust and ensuring accurate reporting. ¹¹ Encryption, multi-factor authentication, and audit trails are critical for safeguarding public trust,^8,10 but may inadvertently restrict access for older adults or individuals with limited digital literacy.^52,53 Similarly, Nguyen et al. indicated that privacy concerns reduced public participation in mobile-based reporting. ⁵² Therefore, there is a need for authentication mechanisms that balance data protection with inclusive usability.^25,45

Usability was another important requirement. Staras et al. demonstrated that usability was a key requirement for improving communication between physicians and parents. ⁴⁴ Similarly, Rahmadhan and Handayani reported highlighted the role of usability in improving data management and AEFI reporting. ⁴⁵ While comprehensive data entry is essential for clinical assessment and regulatory reporting, long and complex forms reduce reporting compliance.^28,29 Design strategies such as adaptive forms, progressive disclosure, and intelligent defaults can help maintain data details without overwhelming users.^26,27

In terms of interoperability, the findings of the current study are in line with other studies. Donckels et al. showed that interoperability with electronic health records enhanced monitoring and decision-making in AEFI report analysis. ³⁷ Atkinson et al. demonstrated that the absence of interoperability was a barrier to completing demographic data. ³⁶ In another study, Rahmadhan and Handayani reported that offline functionality directly improved reporting coverage among general users and healthcare workers. ⁴⁵ Silva et al. mentioned that network instability on the user side posed a major challenge, and the lack of offline capability led to data entry delays and reduced data quality. They also indicated that the most frequently reported problems by users were instability and downtime of central servers. ⁴⁸ In another study, Liu et al. noted that improved vaccine safety monitoring in China’s national AEFI system was associated with increased system accessibility. ¹¹

Regarding data quality, manual data entry was reported to increase the risk of input errors, such as incorrect vaccination dates. ⁴⁸ To address this challenge, several studies suggested that the use of technologies such as two-dimensional barcodes to improve the accuracy of vaccination records.^47,50

It is also notable that with respect to the advancement of information technology, artificial intelligence is increasingly being incorporated into vaccine safety surveillance, bringing measurable efficiency, and raising important design and governance questions. ²¹ For example, natural language processing has reduced VAERS case review time by approximately 58%. ²² In parallel, machine-learning and predictive approaches have been used to surface latent safety patterns and to flag population groups that may require closer monitoring. ⁷⁷ Implementation, however, shapes by ethical and regulatory constraints. Concerns over transparency, data bias, and model interpretability continue to influence trust among clinicians and policymakers. ⁷⁸ Moreover, the lack of established validation standards for adaptive models, together with the challenge of integrating AI tools into existing surveillance frameworks, limits broader adoption. ⁷⁹ From a practical perspective, workforce training, compatibility with legacy systems, and ongoing data quality oversight are critical to maintaining system reliability. ²¹ Taken together, these findings suggest that AEFI reporting systems and VAERS can be designed based on the contextual priorities and requirements, and all recommended items can be used or customized to enhance the design quality and efficiency.

Research implications

The findings of this review can be translated into actionable recommendations for different groups of stakeholders. System designers and health IT developers need to prioritise user-centred design with simplified interfaces to enhance the usability of the designed systems and encourage users to report adverse events with adequate data.^44,47 Systems should incorporate offline functionality in regions with unreliable Internet connectivity to ensure continuity of reporting. ⁴⁵ They should also be interoperable with electronic health records or other health information systems through standard protocols, enabling seamless data exchange and reducing duplicate entry.^36,37,41 Barcode scanning can enhance data accuracy by minimizing manual entry errors, ²⁸ and role-based interfaces can adapt system views to different user categories for example, simplified dashboards for patients versus comprehensive ones for clinicians.^27,44

Healthcare managers and policy-makers need to pay more attentions to national standards for core data elements are needed to enable cross-jurisdictional comparison and support signal detection.^8,11 In addition, investments in scalable digital infrastructure are essential for managing mass vaccination campaigns and surge reporting events.^35,36 Clear data-governance and privacy regulations should balance security with accessibility, as overly stringent requirements may discourage participation.^52,53 Furthermore, continuous training programmes for healthcare providers can improve the quality and consistency of reporting.^28,29

Future studies should focus on rigorous evaluations that measure not only system usability, but also effectiveness and health outcomes.^64,80 Researchers should investigate optimal reporting modalities for population groups, such as older adults or those with limited digital literacy.^49,52 Finally, exploration of artificial intelligence applications should systematically address ethical challenges and validation standards.

Research Limitations

There were several limitations in this study. First, certain databases were searched and only studies published in English were included mainly due to time and resources constrains. There might be other articles indexed in different databases and published in non-English languages that were not included in this study. Second, the full texts of some relevant studies were inaccessible. Third, heterogeneity in study designs and methodologies precluded meta-analysis, necessitating a narrative synthesis consistent with scoping review methodology. Fourth, there might be potential publication bias towards technologically successful systems, and research including system failure may not be retrieved. However, we believe that the current results derived from the most relevant papers and can be used as a practical model for system development. In addition, most included studies focused on system features rather than effectiveness and challenges. Future research should prioritize generating evidence on system effectiveness, outcomes and strategies to overcome challenges.

Conclusion

This scoping review synthesized evidence from diverse global contexts to offer a set of technical platforms, user groups, data elements as well as functional and non-functional requirements involved in AEFI reporting systems and VAERS. Rather than identifying a single optimal model, the findings highlight requirements which can be considered in different systems based on the contextual and user needs. In addition, this review provided a practical guideline for upgrading AEFI reporting systems and VAERS across heterogeneous healthcare systems. New systems can be designed using advanced digital technologies and analytical tools including artificial intelligence; however, their value depends on robust data quality, governance, and alignment with user workflows. Future research should therefore prioritize evaluation studies that explicitly link system features to measurable performance and public health outcomes, particularly in resource-constrained environments.

Supplemental material

Supplemental material - Characteristics of adverse events following immunization reporting systems: A scoping review

Supplemental material for Characteristics of adverse events following immunization reporting systems: A scoping review by Hassan Asadi, Haleh Ayatollahi, Seyed Mohsen Zahraei and Sana Eybpoosh in Health Informatics Journal.