Overcoming Big Data Interoperability Challenges Generated by Ubiquitous Devices Outside Traditional Health Systems: A Systematic Review Protocol

ABSTRACT

Rationale

Ubiquitous devices such as smartphones, wearables, and personal gadgets generate large volumes of personalized health data outside of traditional health systems. Despite its abundance, differences in data formats and semantics across hardware, platforms and sectors keep information siloed. As a result, valuable precision health and social insights are lost when training or validating predictive health models. Interoperable data pipelines that operate across non‐clinical settings are needed so this data can be responsibly translated into actionable information for personal and public health.

Aims and Objectives

This systematic review protocol aims to evaluate existing methodologies that address the interoperability challenges when integrating multimodal health data from ubiquitous devices into non‐clinical health infrastructures. The systematic review seeks to identify primary barriers (technical, semantic, organizational, and regulatory), describe current solutions, and pinpoint gaps that hinder seamless exchange and use of such data.

Methods

A comprehensive literature search will be conducted across multiple databases (PubMed, IEEE Xplore, ACM Digital Library, and Web of Science) for peer‐reviewed primary studies from 2014 to 2024. Gray literature and conference proceedings are excluded. Eligible studies empirically evaluate a framework, architecture or tool that enables technical, semantic, organizational, and regulatory interoperability for person‐generated health data outside clinical settings. Two reviewers will conduct title and abstract screening and full‐text screening in Covidence using predefined criteria, with blinding of authors, journal, and year. Disagreements will be resolved by consensus or a third reviewer. Data will be extracted in Microsoft Excel, and methodological quality will be assessed with the Mixed Methods Appraisal Tool (2018). Interoperability methodology synthesis will be primarily narrative with evidence tables. Where comparable quantitative outcomes exist, we will compute effect sizes and consider robust statistically significant findings (p < 0.05).

Anticipated Implications

This systematic review protocol will sufficiently assemble a comprehensive and reproducible evidence base to conduct the review. The completed review will synthesize and critically appraise peer‐reviewed approaches to interoperability for ubiquitous device‐generated personal health data outside of traditional health systems; identify recurrent barriers and current solutions; and outline good practices to enable cross‐sector data usage, thereby increasing the volume of interoperable data available for predictive modeling of personalized health outcomes.

1. Introduction

The COVID‐19 pandemic exposed the fragility and fragmentation of global health systems, directly challenging progress towards the United Nations' third Sustainable Development Goal (SDG 3): ‘Good Health and Well‐being’ [1, 2] In response, health systems rapidly adopted digital health solutions, embracing disruptive digital innovation ecosystems that link economic, academic, political and societal stakeholders [3, 4, 5, 6, 7]. This accelerated digital transformation, however, has led to uneven implementation that has intensified fragmentation rather than resilience [3, 4, 5, 7, 8]. Rapid disruptive innovation emphasizes interoperable digital infrastructures—systems that seamlessly exchange and integrate data across different systems and devices—as a critical enabler of SDG 3, enabling timely, coordinated, and equitable delivery of health services during public health crises [1, 9, 10]. Recent literature has explored interoperability in patient‐centered and decentralized innovations, such as in‐home‐centered healthcare services, digital pathology, and blockchain‐enabled data frameworks that exemplify efforts to enhance care quality and security [4, 5, 6, 11, 12]. Central to all these interoperability efforts are Internet of Medical Things (IoMT)—a network of interconnected devices that collect, transmit, and analyze data within the medical domain [8, 13, 14]. These devices, often integrated into clinical settings like hospitals, are already delivering valuable insights for professional care [8, 10, 14]. However, their potential remains underutilized, especially when considering the abundant data generated by everyday computing devices (ubiquitous devices) such as smartphones, laptops, and wearable sensors [15].

In today's era of pervasive computing, ubiquitous devices generate continuous streams of big health data [8]. This data is reliable, voluminous, valuable, and variable, presenting significant opportunities for personalized health interventions, real‐time monitoring, and data‐driven healthcare solutions [16]. For instance, smartphones and wearables capture diverse data types, including GPS tracking, physical activity patterns, screen time usage, and social media interactions [17]. Such data points can offer unique and early insights into individual health conditions, often preceding formal healthcare interventions [15].

Despite its promise, much of this data remains siloed, segregated from traditional health systems (centralized institutions such as hospitals, clinics, and offices) largely due to incompatible formats, architectures, and standards that hinder data transferability [18, 19]. These fragmented issues limit the integration of ubiquitous device data into clinical and public health workflows and constrain AI‐driven insights that rely on consistent, interoperable datasets [13, 15, 18, 19]. Achieving interoperability is therefore critical, however, privacy concerns, technical incompatibilities and the absence of common standards continue to impede progress [10, 19, 20, 21, 22]. As disruptive innovation accelerates digital change, without coordinated, ecosystem‐level efforts across syntactic, semantic and organizational layers, data silos will persist and restrict the scalability of mHealth solutions designed to extend healthcare beyond traditional boundaries [7, 13, 15, 18, 21, 22, 23].

Previous reviews have explored different frameworks for translating multimodal health data into standardized formats, such as Electronic Health Records (EHR), within clinical settings like hospitals [8, 24]. For example, Rubí et al. proposed an IoMT platform based on oneM2M and openEHR standards to facilitate the sharing of healthcare data through semantic interoperability, allowing for data exchange across systems [8]. Similarly, Ganzha et al. proposed a semantic interoperability framework to harmonize heterogeneous data sources with ontologies and semantic annotations [21]. These frameworks demonstrate the potential of adopting standardized protocols and structured vocabularies to achieve interoperability and improve data integration in healthcare environments. However, there remains a significant gap in the literature on how methodologies can specifically address the integration of ubiquitous, person‐generated multimodal data into health ecosystems. This gap is particularly relevant for innovation ecosystems in digital health, where disruptive technologies (e.g., blockchain, IoMT platforms, AI solutions) must be embedded within coordinated infrastructures to avoid reinforcing silos and uneven access [4, 5, 6, 11, 12]. The objective of this protocol is to outline the workflow and methodological approach that will be undertaken to evaluate how interoperability challenges are being addressed for multimodal health data generated by ubiquitous devices outside traditional health systems. This protocol aims to ensure transparency, reproducibility, and rigor in the conduct of the forthcoming systematic review. The planned systematic review aims to:

• 1.evaluate existing methodologies proposed or validated to enable interoperability of ubiquitous devices data.
• 2.identify the primary barriers (technical, semantic, organizational, and regulatory) that continue to hinder seamless integration.
• 3.synthesize current solutions and highlight persistent gaps that require further innovation within the digital health ecosystem.

By systematically addressing these objectives, the completed systematic review will generate a comprehensive evidence base that clarifies the state of knowledge, supports the development of responsible innovative ecosystems for digital health, and informs future methodological and policy directions.

2. Methods

This systematic review protocol was prospectively registered with the Center for Open Science (OSF) (https://doi.org/10.17605/OSF.IO/QWDA5). This systematic review will be conducted in accordance with the Preferred Reporting Items for Systematic Reviews and Meta‐Analyses (PRISMA) guidelines to ensure a structured, transparent, and replicable methodology [25]. The guiding research question is: ‘How are interoperability challenges being overcome for ubiquitous device‐generated multimodal health data outside traditional health systems?’

To answer this question, a systematic search will be conducted across four major databases. PubMed captures biomedical and health‐related research, particularly studies on interoperability in healthcare and medical data applications. IEEE Xplore provides technical literature on engineering, system design, and emerging technologies such as IoT and wearable devices. The ACM Digital Library contributes foundational and applied research in computer science, including semantic technologies and data processing. Web of Science offers broad scientific coverage, tracking interdisciplinary research and high‐impact studies. Search terms will combine three main themes: interoperability, multimodal health data, and ubiquitous/IoMT devices (Table 1). Boolean operators will be applied to tailor queries for each database, and the search will cover the period from 2014 to 2024 to capture the most recent advances. Additionally, reference lists from relevant reviews will be screened to identify key studies not retrieved through database searches. Together, these sources ensure a robust and balanced exploration of the topic.

Table 1.

Structured search terms for systematic review (combined using ‘AND’).

Theme	Search terms
Interoperability	(‘interoperability’ OR ‘semantic interoperabilit*’ OR ‘syntactic interoperabilit*’ OR ‘technical interoperabilit*’ OR ‘data integration’ OR ‘health data sharing’ OR ‘interoperability framework*’ OR ‘ontology’ OR ‘semantic ontology’ OR ‘semantic technolog*’ OR ‘knowledge representation’)
Multimodal Health Data	(‘multi‐modal data’ OR ‘sensor data’ OR ‘multimodal data’ OR ‘heterogeneous data’ OR ‘biometric data’ OR ‘biometrics’ OR ‘physiological data’ OR ‘voice biometrics’ OR ‘behavioral biometrics’ OR ‘biosensor data’)
Internet of Medical Things/Ubiquitous Devices	(‘platform’ OR ‘app’ OR ‘application’ OR ‘framework*’ OR ‘Internet of Medical Thing*’ OR ‘IoMT’ OR ‘mHealth’ OR ‘smartphone*’ OR ‘wearable device*’ OR ‘handheld device*’ OR ‘connected device*’ OR ‘personal device*’)

Note: The asterisk (*) is used as a truncation symbol to capture all variations of a root term (e.g., ‘interoperabilit*’ retrieves interoperability, interoperable, interoperabilities, etc.).

Studies will be considered eligible if they (1) are peer‐reviewed primary research articles; (2) propose or validate frameworks, methodologies, architectures, or tools that address interoperability of multimodal health data generated by ubiquitous devices; and (3) provide some form of empirical validation or application. Studies focusing exclusively on hospital‐based devices or clinical‐only technologies, as well as reviews, conference abstracts, gray literature, and purely theoretical work without validation, will be excluded. A summary of the inclusion and exclusion criteria is outlined in Table 2.

Table 2.

Inclusion/exclusion criteria.

	Inclusion	Exclusion
Language	English	Non‐English
Study population/context	Studies focusing on person‐generated data or ubiquitous devices (e.g., wearables, smartphones) in health and fitness contexts.	Studies exclusively focused on hospital‐based or clinical contexts.
Target devices	Smartphones, wearable devices, wearable sensors, personal devices	Hospital and clinical devices: patient monitoring devices, imaging and diagnostic devices, laboratory and pathology devices, infusion and therapy devices, ventilation and anesthesia devices, surgical and treatment equipment, health information systems.
Date of publication	2014−2024	Prior to 2014
Publication status	Peer‐reviewed primary research articles	Gray literature, conference proceedings, technical reports, white papers, editorials, opinion pieces and purely theoretical papers without empirical validation.
Type of study	Studies presenting frameworks, applications, methodologies, architectures, designs, or solutions addressing interoperability challenges related to multimodal data and ubiquitous devices.	Review articles, meta‐analyses, systematic reviews, and non‐primary sources.
Outcome	Solutions or methodologies addressing multimodal data sharing and interoperability.	Studies that do not provide procedures for data fusion or focus solely on hospital‐based devices or clinical technologies.
Framework Validation	Studies with empirical validation of frameworks, methodologies, or solutions.	Theoretical or conceptual papers without validation or real‐world application.

All search results will be imported into Covidence for screening and duplicate removal. A two‐stage screening process will be followed. In the first stage, titles and abstracts will be independently screened by two reviewers against the eligibility criteria. In the second stage, full texts will be assessed for final inclusion. Any disagreements will be resolved through consensus to minimize bias.

For each study included, data will be extracted into a structured spreadsheet. Extracted fields will include publication details, study context, type of devices used, nature of multimodal data collected (data input and output), interoperability challenges addressed, proposed solutions, and evidence of validation. The methodological quality of each study will be appraised using the Mixed‐Methods Appraisal Tool (MMAT), ensuring reliable and valid evaluation of included studies [26, 27, 28, 29, 30]. MMAT is a tool designed to assess diverse study designs, including quantitative, qualitative and mixed‐methods explicitly for systematic reviews, as opposed to other assessment tools that are study‐design specific, such as the Quality Assessment of Diagnostic Accuracy Studies (QUADAS), A Measurement Tool to Assess Systematic Reviews (AMSTAR) or Risk of Bias in Systematic Reviews (ROBIS) [30, 31, 32]. The assessment begins with an initial screening to ensure that the research objective is clearly stated and that the data collected are appropriate for addressing the research question. Following this, eligible studies are evaluated using five questions on their methodological approach, with each criterion scored on a scale from zero to five, where five represents the highest level of methodological quality. This approach ensures a thorough and standardized evaluation of study quality across multiple research designs.

The synthesis will primarily adopt a narrative approach, supported by evidence tables to summarize study characteristics, interoperability strategies, and outcomes. Where studies report comparable quantitative measures, effect sizes will be calculated and statistical significance considered (p < 0.05). Figure 1 shows the PRISMA 2020 flow diagram for the proposed systematic review described in this protocol.

Figure 1.

Flow diagram for the proposed systematic review in this protocol. This study will use data from published and publicly accessible sources. As no human participants will be recruited and no identifiable individual‐level data will be collected, ethical approval is not required.

2.1. Anticipated Results

The systematic review outlined in this protocol is expected to produce a comprehensive synthesis of the literature on interoperability challenges and solutions for multimodal health data generated by ubiquitous devices outside traditional health systems.

The systematic review will identify and categorize the types of interoperability barriers documented in the literature. We expect these challenges to span technical issues (heterogeneous data formats, connectivity limitations), semantic gaps (inconsistent data definitions or terminologies), regulatory and privacy constraints (data sharing policies, consent and security requirements), and organizational barriers (lack of integration into clinical workflows). By extracting and stratifying these barriers into these groups, the results will synthesize known obstacles that currently impede seamless data exchange between personal health devices and health information systems. This catalog of challenges will provide a clear overview of what problems are most prevalent across studies specific to person‐generated health data.

In parallel, the systematic review will compile the range of proposed solutions and frameworks aimed at overcoming these interoperability challenges. We anticipate identifying technical solutions such as the adoption of open data standards (e.g., using HL7 FHIR to structure wearable data) and middleware platforms or APIs for data integration. Semantic solutions may include common ontologies or mapping schemes to standardize multimodal data from diverse devices. Regulatory and governance solutions could involve privacy‐preserving data sharing frameworks, consent management systems, or compliance with interoperability mandates (regional data exchange policies). Organizational strategies might include new workflow models or agreements that enable the routine use of device‐generated data. Each solution will be documented along with any information on its development stage (conceptual framework, prototype or deployed system) and whether it has been validated in practice (tested through pilot studies or real‐world implementation). Notably, emerging and disruptive technologies are expected to feature in these solutions. For instance, some studies may propose blockchain or distributed ledger technology to enable secure, decentralized exchange of personal health data or the use of artificial intelligence to harmonize and analyze data streams from multiple sources. These innovative approaches, if identified, will be highlighted as part of the mapping, noting both their potential and the challenges they pose (e.g., the need for rigorous evaluation and validation to ensure they align with interoperability goals).

A crucial component of the results will be the identification of unmet needs and research gaps revealed by MMAT. The evidence from this systematic review is expected to pinpoint areas where interoperability challenges remain unresolved or where proposed solutions have not been validated. For example, the findings may reveal gaps in standardization for certain emerging data types (such as novel biomarker data from wearable biosensors or geolocation data from smartphones that lack integration protocols), or a lack of studies addressing inter‐organizational data sharing frameworks for community‐generated health data. If the systematic review finds that few studies have evaluated real‐world implementation outcomes (impact on clinical workflow efficiency or patient outcomes), this will be noted as an evidence gap. Similarly, any mismatch between the rapid evolution of consumer health technology and the slower development of interoperability standards will be highlighted. These identified gaps will directly inform the discussion on future research and innovation directions. For instance, it might suggest the need for more pilot programs deploying promising frameworks that lack evidence‐based validation of effectiveness and feasibility (such as testing blockchain‐based data exchanges in a healthcare setting or deploying AI algorithms for data normalization at scale). It may also indicate a need for greater focus on equity and accessibility, if the literature shows that interoperability solutions have primarily been tested in high‐resource settings or only among certain populations.

Ultimately, the expected findings will have broader implications for building an inclusive, digitally interoperable health ecosystem. By illuminating current barriers and solutions, the systematic review will shed light on how we can better incorporate person‐generated health data into learning health systems and predictive analytics in an equitable way. The synthesis will likely discuss how disruptive innovations are being leveraged. This includes studies providing evidence that blockchain is proving useful to establish trust and decentralize platforms for data exchange, or how AI might assist in making sense of multimodal data deluge in real time [4, 6, 11, 12]. These insights will highlight what is feasible today and what might be on the horizon, informing strategic planning for a future where data from wearable sensors, home monitors, and health apps can seamlessly augment with health ecosystems. A key anticipated message is that solving interoperability challenges is not merely a technical endeavor, but a socio‐technical one, requiring alignment of technology capabilities with policy, ethics, and user adoption. By emphasizing findings that relate to responsible innovation (such as solutions that incorporate privacy‐by‐design, patient consent, and user‐centered design), the systematic review will promote approaches that safeguard individual rights while advancing technology.

3. Conclusions

3.1. Theoretical Implications

This protocol addresses how our systematic review will identify notable gaps in the literature by consolidating research into a cohesive understanding of how person‐generated health data can be integrated with health ecosystems. One key contribution of our work is a proposed taxonomy of interoperability challenges and corresponding solutions, which categorizes barriers into technical, semantic, organizational and regulatory domains and links each category to potential strategies. This organizing schema deepens theoretical insight by illustrating how different types of barriers interrelate and can be collectively addressed. Furthermore, this protocol is designed to reveal the socio‐technical nature of interoperability, reinforcing the notion that technical innovation must be accompanied by changes in workflows, stakeholder engagement, and supportive policies. By providing a unified perspective on previously disparate findings, the systematic review offers researchers a clearer conceptual map of the interoperability landscape.

3.2. Policy Implications

The findings from the prospective systematic review aim to highlight a need for stronger health data standards and supportive regulations. We expect the systematic review to evaluate the extent to which open interoperability standards (such as HL7 FHIR) can support the integration of wearable device‐generated data into health ecosystems, and to identify conditions under which mandates or incentives prove effective. In parallel, by systematically classifying privacy, security, and consent concerns, the systematic review is intended to indicate where data protection rules may require updating. Our protocol also points to specific gaps that the completed systematic review may confirm as priorities for standards development organizations, professional bodies, and regulators. Finally, the systematic review will assess whether proposed solutions are adaptable to low‐resource or underserved settings, with the goal of avoiding interoperability pathways that exacerbate disparities. If the evidence supports these lines of inference, the final systematic review can offer decision‐makers actionable and evidence‐based insights for funding and collaborative implementation.

3.3. Ideas for Future Research

This protocol acknowledges likely constraints in the underlying literature due to a relatively small pool of eligible studies and a prevalence of conceptual or prototype‐level contributions. Contingent on the systematic review's findings, we anticipate recommending: (1) rigorous, real‐world evaluations of interoperability approaches identified in the synthesis (such as emergent data standards or integration platforms); (2) continued innovation in data standardization, including consensus protocols for novel biosensor and smartphone‐derived streams; (3) robust, testable frameworks for inter‐organizational data sharing that allow person‐generated health information to follow patients across care settings; and (4) targeted assessment of novel technologies that lack extensive empirical validation such as blockchain for decentralized exchange or artificial intelligence for automated harmonization of heterogeneous data. This protocol aims to ensure that the completed systematic review can credibly map these gaps and motivate empirically grounded next steps. Taken together, the protocol lays the methodological foundation for a synthesis that can refine theoretical models of digital health interoperability and provide a stronger evidence base to steer technological innovation and supportive policy in an evolving digital health landscape.

Conflicts of Interest

The authors declare no conflicts of interest.

Banik K., Patel J., Ibrahim S. T., and Katapally T., “Overcoming Big Data Interoperability Challenges Generated by Ubiquitous Devices Outside Traditional Health Systems: A Systematic Review Protocol,” Journal of Evaluation in Clinical Practice 32 (2026): e70410. 10.1111/jep.70410.

Data Availability Statement

No new datasets are generated as part of this protocol. Data extracted from included studies and any derived data (e.g., evidence tables) will be made available in the final review and/or in supporting materials.