The objectives were:

The review followed the PRISMA statement. In line with PRISMA guidance, the main methodological decisions (information sources, eligibility criteria, screening stages, data items and synthesis approach) were defined a priori and applied consistently. The overall identification, screening, eligibility and inclusion process is summarised in a PRISMA flow diagram (Figure 1), which reports the number of records at each stage and the reasons for full-text inclusion.

A comprehensive database search was conducted to identify empirical studies on mHealth applications used in disaster, emergency or humanitarian contexts. The following electronic databases were searched: PubMed, Scopus, Web of Science, IEEE Xplore and Google Scholar. The search covered publications from January 2010 to May 2025, a period that aligns with the rapid global expansion of digital health and mHealth. Search strings combined terms related to mHealth in disaster and humanitarian settings, using Boolean operators and database-specific adaptations. An example of the core search string was:

‘mHealth’ OR ‘mobile health’ OR ‘mobile applications’ OR ‘smartphone applications’ AND ‘disaster’ OR ‘emergency’ OR ‘crisis’ OR ‘humanitarian’ OR ‘natural disaster’ OR ‘man-made disaster’.

Where appropriate, truncation and synonyms were adapted to each database’s indexing terms to maximise sensitivity. To enhance completeness, the database search was supplemented by manual screening of reference lists of all included studies and relevant review articles, as recommended in PRISMA for using multiple information sources. Google Scholar was included in the database search to ensure that important grey literature, organisational reports and white papers were captured.

Eligibility criteria were specified a priori and applied in accordance with PRISMA guidance on clearly defining inclusion and exclusion criteria. The review focused on primary empirical studies, while a small number of review-type papers were retained solely for contextual and conceptual purposes and were not included in the core evidence base.

The study selection process followed the four PRISMA stages: identification, screening, eligibility and inclusion.

The number of records at each stage, together with the reasons for exclusion during full-text assessment, is presented in the PRISMA flow diagram (Figure 1), which visually maps the identification, screening, eligibility and inclusion stages.

In this review, the study population comprised peer-reviewed empirical articles reporting on the use of mHealth or telemedicine in disaster response, recovery or preparedness across diverse settings. The sampling approach was purposive at the study level, based on the systematic application of pre-defined eligibility criteria to all records identified through the search strategy. Of the studies progressing to full-text assessment, 23 empirical investigations satisfied all inclusion criteria and formed the core sample for qualitative synthesis. Two additional review-type papers were purposefully retained to enrich the conceptual and contextual interpretation of findings; these were analysed separately and not pooled with the empirical evidence. This distinction between empirical and contextual sources is consistent with PRISMA’s emphasis on clearly reporting the number and type of included studies and how they were selected. The 23 empirical studies included in this review were published between 2010 and 2025, with the majority published from 2018 onwards, reflecting the accelerating growth of mHealth research in disaster contexts. Geographically, studies were concentrated in low- and middle-income countries (LMICs). Disaster types represented included natural disasters such as earthquakes, floods, cyclones and typhoons, health-related disasters including epidemics and pandemics and conflict-related emergencies. In terms of intervention category, health surveillance systems and emergency medical services (EMS) applications were the most frequently studied, followed by telemedicine solutions and resource coordination platforms. Study designs were predominantly observational, case-based or implementation-focused, with a smaller number of systematic or scoping reviews retained as contextual sources.

In keeping with guidance that systematic reviews should systematically search for, appraise and synthesise research evidence, we undertook a structured quality appraisal of all included peer-reviewed studies. Because the review encompassed diverse empirical designs (quantitative, qualitative and mixed-methods), we used a brief, standardised critical appraisal checklist guided by tools such as the Critical Appraisal Skills Programme (CASP) checklists (Long, French & Brooks 2020) and the Mixed-Methods Appraisal Tool (MMAT) (Hong et al. 2018). The checklist covered core methodological domains commonly recommended for study-level appraisal:

Each study was independently appraised by two reviewers; disagreements were resolved through discussion and, where necessary, by consulting a third, independent reviewer, as recommended to reduce authors’ bias until a consensus was reached on all papers that qualified for inclusion in the study. Consistent with methodological guidance that quality assessment in many narrative syntheses is primarily used to inform interpretation rather than act as an absolute exclusion criterion, no studies were excluded solely based on appraisal scores. Instead, quality ratings were used to weight the contribution of individual studies when synthesising evidence on application types, operational outcomes, governance mechanisms and contextual challenges, with lower-quality studies treated more cautiously in the narrative and inferences about effectiveness, barriers and governance gaps.

Data extraction was undertaken using a standardised extraction form developed by the authors to ensure consistent capture of variables across the included studies. For each study, the following data items were extracted:

The review methods were designed to minimise bias and to be as reproducible as possible, consistent with PRISMA 2020 guidance on transparent reporting of systematic reviews. The research questions, eligibility criteria, databases to be searched and screening procedures were specified in advance and this guided all subsequent stages of the review. Comprehensive search strategies using predefined keywords and Boolean operators were developed for each database, and the full strategy for at least one database is reported in Appendix 1 Table 1-A1 to allow replication.

Records retrieved from all databases were imported into a reference manager and deduplicated before screening to reduce selection bias. Title and abstract screening, followed by full-text screening, were conducted independently by two reviewers using the same inclusion and exclusion criteria; disagreements were resolved through discussion and, where necessary, by consulting a third, independent reviewer, which is recommended to reduce authors’ bias.

As a result of the heterogeneity of study designs, disaster contexts, interventions and outcome measures, a narrative thematic synthesis was undertaken rather than a statistical meta-analysis (Braun & Clarke 2006, 2019), as adapted for systematic reviews by Lucas et al. (2007). The synthesis focused exclusively on the 23 empirical studies, with the two contextual review-type articles used only to triangulate and interpret emergent themes. The analytic process proceeded in several steps:

Categorisation of mHealth applications: Empirical studies were grouped into four main categories, EMS, health surveillance systems, resource coordination platforms and telemedicine solutions, based on the primary function of the application. Comparison of implementation challenges: Across categories and settings, we compared reported implementation challenges, including infrastructure constraints, connectivity and power-supply issues, interoperability limitations, governance and regulatory barriers, user adoption and usability concerns, and sustainability problems. Analysis of governance and policy arrangements: We synthesised how governance frameworks, regulatory instruments, data protection provisions and accountability arrangements were articulated or implied in the studies, with particular attention to gaps relevant to disaster-affected and low-resource contexts. Assessment of reported effectiveness: We analysed how mHealth interventions contributed to triage, outbreak monitoring, logistics and supply coordination, remote consultation, and broader health system performance during disasters, highlighting both demonstrated benefits and areas of equivocal or limited evidence. Identification of gaps and future directions: Finally, we mapped thematic gaps in the empirical evidence, including under-represented disaster types, regions, user groups, and governance issues, and drew on the contextual review-type papers to corroborate or extend these gaps at a higher level. This approach is consistent with PRISMA recommendations for reviews in which meta-analysis is not appropriate, requiring a clear description of the narrative synthesis methods and the rationale for adopting a qualitative, thematic approach.

This review relied exclusively on secondary analysis of data reported in publicly available, peer-reviewed publications and did not involve direct contact with human participants or animals. In line with institutional and journal norms for systematic reviews of published literature, formal ethics committee approval was not required. The review process adhered to principles of academic integrity, appropriate attribution and responsible citation of all included and referenced works, including transparent reporting of study selection and synthesis procedures as recommended by PRISMA 2020.

The final sample comprised 23 empirical studies and two contextual review-type papers published between 2010 and 2025. Most empirical studies were conducted in LMICs, with a concentration in South and Southeast Asia, sub–Saharan Africa and the Middle East, and focused on natural disasters such as earthquakes, floods, cyclones and typhoons, alongside a smaller number related to epidemics and conflict-related emergencies. Across the empirical set, health surveillance systems and EMS applications were the most frequently reported interventions, followed by resource coordination platforms and telemedicine solutions, with a predominance of observational and implementation-focused study designs. The two contextual review-type articles retained to support the conceptual and contextual interpretation of the findings are summarised in Table 3.

The thematic analysis of the included studies identified four primary categories of mHealth applications used in disaster zones: EMS, health surveillance systems, resource coordination platforms and telemedicine solutions. These categories emerged from patterns observed across the selected studies and reflect the main ways in which mHealth technologies are applied in disaster response and recovery. This section presents the findings related to these categories, highlighting their key features, use cases, and examples as reported in the reviewed studies.

Building on analyses of existing governance models, implementation challenges and effectiveness assessments, this section proposes a conceptual framework for effective mHealth governance in disaster zones. This framework, illustrated in Figure 2, integrates key governance components, contextual factors and desired outcomes to provide a holistic approach to mHealth deployment in humanitarian crises.

The conceptual framework for the effectiveness of mHealth in disaster management is a dynamic system designed to optimise the use of mHealth applications in crisis situations. The framework proposes an enabling environment that encompasses essential elements, including strong policy and governance, reliable infrastructure and access, sufficient human capacity, sustainable funding and resources, and established trust and ethical standards. These foundational components create the conditions necessary for the effective deployment of mHealth interventions. The mHealth interventions encompass communication platforms, data and monitoring capabilities, information and education initiatives, resource management tools and mental health support services. These processes are implemented across the various phases of disaster management, preparedness, response and recovery, providing a contextual framework for their application throughout the disaster lifecycle. The framework seeks to achieve several desired outcomes, including improved healthcare access, enhanced disaster response, strengthened public health systems, empowered communities and reduced vulnerability. Supporting the entire framework are guiding principles such as person-centred care, innovation for impact, interoperability and digital health equity. These principles ensure that mHealth solutions remain ethical, effective and centred on the needs of users. A critical feedback loop fosters continuous improvement and adaptability. This loop involves ongoing monitoring and evaluation, incorporation of lessons learned, and the generation of research and evidence. Through this iterative process, insights derived from real-world experiences and new knowledge inform policy adjustments, capacity-building and technological advancements. In doing so, the system is regularly refined, allowing mHealth solutions to evolve in response to the changing challenges of disaster management. This proposed governance framework emphasises a multi-stakeholder approach, with collaboration among government agencies, healthcare providers, technology developers, NGOs, academic institutions and affected communities. It also highlights the importance of adapting governance strategies to the specific context of each disaster, considering factors such as the type of disaster, the affected population, and available resources. The framework emphasises a dynamic and iterative approach to mHealth governance. Contextual factors should be continuously assessed, and governance components should be adapted as needed to achieve desired outcomes. Feedback loops from M&E processes are essential for informing these adaptations and ensuring continuous improvement. Furthermore, the framework highlights the importance of a multi-stakeholder approach that involves collaboration among government agencies, healthcare providers, technology developers, NGOs, academic institutions and affected communities. Each stakeholder group has a unique role to play in shaping and implementing effective mHealth governance. For instance, governments are responsible for establishing enabling policy environments and providing oversight. Healthcare providers are key to ensuring that applications meet clinical needs and are integrated into service delivery. Technology developers contribute expertise in designing and building robust and user-friendly applications. Non-governmental organisations often play a crucial role in implementation and community engagement, particularly in hard-to-reach areas. Academic institutions can contribute through research, evaluation, and capacity-building. Finally, the active involvement of affected communities is paramount to ensure that mHealth solutions are relevant, acceptable and sustainable. This conceptual framework provides a comprehensive roadmap for developing and implementing effective mHealth governance strategies in disaster zones. By addressing both the core governance components and the specific contextual factors, stakeholders can work towards maximising the potential of mHealth to improve health outcomes and strengthen health system resilience in the face of disasters.

Our four-category typology EMS/triage, health surveillance, resource coordination, telemedicine is broadly consistent with Alanazi et al. (2025), who likewise distinguished triage, surveillance, and remote-care application classes in their JMIR mHealth and uHealth review, and with Lokmic-Tomkins et al. (2023), whose systematic review of disaster-management systems clustered tools along similar operational lines. Where our review departs is in (1) explicitly carving out resource coordination platforms as a distinct fourth category. Alanazi et al. (2025) treated logistics as a sub-feature of EMS rather than a stand-alone class. The review also departs in (2) reporting a higher prevalence of sub-Saharan African deployments, where prior reviews were dominated by Asian and high-income settings (Komi et al. 2021). This regional re-balancing is one of the empirical contributions of the present review. On governance, our findings of a fragmented landscape of national strategies in some countries, international guidance from the WHO, but a paucity of disaster-specific frameworks, are convergent with Komi et al. (2021), who reached the same conclusion for LMICs. Our contribution is to advance from description to prescription by proposing a six-component conceptual governance framework that is explicitly disaster-tuned rather than a generic mHealth model. Likewise, prior work has catalogued barriers and facilitators of mHealth adoption during COVID-19 (Lokmic-Tomkins et al. 2023). Our work shows that the same adoption barriers (cultural, infrastructural, data-related) recur across natural, conflict-related and health-emergency disasters, suggesting a generalisable barrier structure rather than pandemic-specific friction. Two divergences are worth flagging. Firstly, prior reviews have reported positive evidence regarding the effectiveness and potential efficiency gains of selected mHealth tools in disaster response contexts (Yarmohammadian, Safdari & Tavakoli 2015); however, our synthesis is more cautious because only a limited number of empirical studies included formal cost–benefit analyses, highlighting an important evidence gap rather than a direct contradiction. Secondly, where earlier studies have often treated interoperability as primarily a technical issue, our analysis suggests that the principal constraints in disaster settings are frequently semantic and organisational in nature rather than network-layer limitations. This finding has important implications for the design of governance and coordination mechanisms (Gastaldi & Horlait 2022). Regarding implementation challenges, our findings are broadly consistent with prior reviews but provide greater geographic and thematic granularity. Zakerabasali et al. (2021) identified infrastructure, digital literacy and cultural factors as principal barriers to mHealth adoption, and our synthesis confirms all three across the four application categories. Where our review extends this literature is in the specific salience of power-supply constraints (as distinct from connectivity) for EMS and telemedicine applications in sub-Saharan African and South Asian settings, and in the prominence of sustainability as a binding post-disaster constraint, a dimension that COVID-19-focused reviews such as Jain et al. (2023) were less well-positioned to capture, given the comparatively sustained funding environment of pandemic response. The finding that 80% of mHealth initiatives reviewed by Agarwal et al. (2015) failed to sustain operations beyond initial funding cycles is mirrored in our sample, reinforcing that sustainability is a structural rather than incidental problem. On effectiveness, our synthesis aligns with the finding that mHealth tools consistently improve process-level outcomes (data collection speed, triage throughput, supply-chain visibility) but that evidence on patient-level health outcomes remains limited and methodologically weak. Our review extends this finding in two ways (Yarmohammadian et al. 2015). Firstly, by disaggregating effectiveness across four application categories, we show that health surveillance systems have the most consistent evidence base (anchored by multi-country deployments such as SPEED and DONS with measurable outbreak-detection improvements), whereas resource coordination platforms have the weakest, partly because effectiveness metrics vary substantially across deployments and are rarely standardised. Secondly, our inclusion of sub-Saharan African deployments surfaces a pattern not prominent in prior reviews: applications adapted for low-bandwidth, offline-first operation consistently outperform those designed for higher-resource environments, suggesting that contextual fit is as important a predictor of effectiveness as technical design quality. This contextual adaptation argument represents a distinctive empirical contribution of the present review. The findings underscore the significant potential of mHealth to transform healthcare delivery in humanitarian crises, while also highlighting the complex array of challenges that must be addressed to fully realise this potential. The categorisation of mHealth applications into EMS, health surveillance, resource coordination and telemedicine solutions provides a useful typology for understanding the diverse ways in which mobile technology is being leveraged in disaster contexts. Each category addresses specific needs and offers unique functionalities, from rapid patient triage and disease outbreak detection to efficient supply chain management and remote specialist consultations. The examples from various disaster settings, including those in South Africa, illustrate the practical application and tangible benefits of these tools. However, the widespread and effective deployment of these applications is consistently hampered by recurring implementation challenges. Infrastructure limitations, particularly around connectivity and power, remain a fundamental barrier in many disaster-affected regions. User adoption issues, stemming from factors such as digital literacy, cultural considerations and the perceived burden of new technologies during high-stress situations, also pose significant hurdles. Furthermore, data management concerns, including quality, interoperability, privacy and security, are critical and often inadequately addressed. Finally, the long-term sustainability of mHealth initiatives, encompassing financial, technical and institutional aspects, is a major concern that frequently undermines the lasting impact of promising interventions. The review of governance frameworks reveals a need for more robust, comprehensive and disaster-specific models. While national eHealth strategies and international guidelines offer some direction, they often lack the specificity and enforcement mechanisms required to address the unique demands of disaster zones. The proposed conceptual framework for mHealth governance in disaster zones attempts to address this gap by integrating essential governance components (leadership, legal/regulatory aspects, technical standards, ethics, capacity-building and M&E) with a keen sensitivity to contextual factors and a clear focus on desired health and system outcomes. This framework emphasises a multi-stakeholder, adaptive and iterative approach, which is crucial for navigating the dynamic and complex nature of disaster environments. The assessment of effectiveness indicates that while mHealth applications can lead to improvements in process measures and, in some cases, health outcomes, the evidence base is still developing. Methodological rigour in evaluation studies needs to be strengthened to provide more definitive conclusions about the impact and cost-effectiveness of these interventions. An understanding of effectiveness requires looking beyond simple health metrics to consider user satisfaction, system efficiencies and broader societal impacts. Several gaps and areas for future research have emerged from this review. Firstly, there is a need for further research on the long-term impact and sustainability of mHealth applications in disaster recovery phases, moving beyond the immediate response. Secondly, comparative studies evaluating different types of mHealth applications for similar purposes in diverse disaster contexts would be valuable for identifying best practices and transferable lessons. Thirdly, research focusing on the specific needs and challenges of vulnerable populations within disaster zones, such as children, the elderly and individuals with disabilities, is crucial for ensuring equitable access and benefit from mHealth interventions. Fourthly, further investigation is needed into effective models for public-private partnerships to develop and deploy mHealth solutions for disasters, leveraging the strengths of both sectors. Finally, the ethical implications of using artificial intelligence and big data analytics in mHealth applications in disaster settings warrant careful exploration to ensure responsible, responsive innovation.

The findings of this review converge with and extend four streams of theory. Firstly, the Unified Theory of Acceptance and Use of Technology (UTAUT-2; Venkatesh, Thong & Xu 2012) is supported: performance expectancy, effort expectancy, social influence and facilitating conditions all surface as drivers of mHealth uptake in the synthesised studies, but our findings extend UTAUT-2 by foregrounding contextual moderators specific to disaster settings, such as infrastructure collapse, security and trust in responder agencies. We believe that these factors warrant explicit incorporation into the model for crisis contexts. Secondly, the Information Systems Success Model (DeLone & McLean 2003) provides a defensible lens for the effectiveness findings in this paper: system quality, information quality and service quality each emerged as determinants of net benefit (e.g. faster triage, fewer stock-outs). However, our results suggest that the model’s ‘use’ construct must be re-interpreted as ‘use under duress’ in disaster settings, where intermittent use, offline operation and surge use are the norm rather than the exception. Thirdly, our governance findings align with the Multi-Level Governance perspective (Liesbet & Gary 2003): effective mHealth governance is shown to be neither purely national nor purely transnational but jurisdictionally layered, consistent with the framework’s central claim. Fourthly, the dual emphasis on semantic and organisational interoperability that was identified resonates with sociotechnical systems theory (Sawyer & Jarrahi 2014; Trist 1981), reinforcing that successful mHealth deployment is a joint optimisation problem across technical and social subsystems rather than a purely technological one.

For policymakers, the most actionable implication is that disaster-specific mHealth governance should be codified in national disaster-management acts rather than left to general eHealth strategy documents that do not anticipate crisis conditions. The proposed six-component framework operationalises this: leadership, legal regime, technical standards, ethics and community engagement, capacity-building, and monitoring and evaluation. For ministries of health in sub-Saharan Africa, the region most represented in our included studies, the study recommends pre-positioning licensing arrangements for telemedicine across provincial and national borders so that response is not delayed by jurisdictional licensing checks during the first 72 h of a disaster. Policymakers should also foster multi-stakeholder collaboration by creating and supporting platforms that bring together government entities, NGOs, private sector technology providers, academic institutions and community representatives to ensure a coordinated and effective mHealth response. For healthcare providers and humanitarian organisations, the practical implication is to embed validated mHealth tools into standard operating procedures and pre-disaster drills rather than treating them as ad hoc surge tools. For technology developers, the implication is to design for offline-first operation, low-bandwidth fallback and language localisation, and to publish interoperability artefacts under open licences to lower the integration cost for downstream responder agencies. For funders, they should fund 5–7-year sustainability envelopes rather than discrete deployment grants, since our review found that sustainability, and not initial deployment, is the binding constraint on long-term impact. Furthermore, healthcare providers and organisations should integrate validated mHealth tools into their disaster preparedness and response plans, embedding these tools into standard operating procedures, and conduct regular training and simulation exercises for their staff. They must champion user-centred design and contextual adaptation by actively participating in the design, testing, and refinement of mHealth applications to ensure these tools are user-friendly, meet clinical needs, and are appropriate for the local context and user capabilities. Healthcare organisations need to prioritise data quality and security by implementing robust protocols for data collection, management and protection. This ensures the reliability of the information and safeguards patient privacy, in line with ethical guidelines and legal requirements. For researchers, the implication is twofold: (1) the absence of randomised or quasi-experimental evaluations in our search window is a methodological gap that should be addressed by pre-registered hybrid implementation-effectiveness designs in the next wave of evaluations; and (2) the under-representation of mental-health-focused mHealth tools in disaster settings warrants targeted primary research, given the documented psychosocial burden of disasters. Researchers and academic institutions are also urged to conduct rigorous evaluations of mHealth interventions, including randomised controlled trials when feasible, to generate strong evidence on their health, system and economic impacts in disaster settings. They should investigate implementation science approaches to better understand the factors that facilitate or hinder the successful scale-up and sustainability of mHealth applications across diverse disaster contexts. Additionally, exploring emerging technologies such as artificial intelligence, the Internet of Things, and blockchain must be done responsibly, with careful consideration of ethical implications and a commitment to a human-rights-based approach to enhance mHealth capabilities for disaster management. Technology developers and the private sector have a crucial role in designing mHealth applications optimised for low-resource and challenging environments. These applications should function effectively under low-bandwidth conditions, offer offline capabilities, be power-efficient and feature intuitive interfaces that require minimal training. Emphasising interoperability and adherence to open technical standards is vital to enable seamless data exchange between different mHealth systems and integration with national health information systems. Developers should also engage in ethical and sustainable partnerships, collaborating with humanitarian organisations and local stakeholders to create mHealth solutions that are contextually appropriate, ethically sound, and supported by clear plans for long-term sustainability and local ownership. Funding agencies and donors need to support long-term sustainable mHealth initiatives by shifting funding models beyond initial development and deployment to include ongoing maintenance, local capacity-building and integration into health systems. They should promote evidence-based investments by prioritising funding for mHealth interventions that demonstrate effectiveness and align with national health and disaster management priorities. Moreover, encouraging innovation and knowledge sharing is essential, supporting initiatives that foster new developments in mHealth for disasters and facilitate the exchange of best practices, lessons learned, and open-source solutions across regions and organisations. Based on the findings of this review, it is important that these frameworks are integrated seamlessly with broader national disaster management and health system strategies. Investing in resilient infrastructure must be a priority, particularly in disaster-prone areas, to ensure the reliable operation of mHealth applications. This includes strengthening telecommunications and power infrastructure and exploring the deployment of rapidly deployable emergency communication systems. In conclusion, mHealth applications are indispensable tools in the modern disaster management toolkit. By addressing the identified challenges through robust governance, collaborative action, and continuous innovation, the global community can harness the full potential of mHealth to save lives, alleviate suffering and build more resilient health systems in the face of ever-increasing disaster risks. The journey requires commitment, investment and a shared vision for a future where technology effectively serves humanity in its most vulnerable moments.