Towards Electronic Medical Record (EMR) utilization in developing countries: a conceptual model based on systematic literature review Authors: Abolfazl Rezaei, Maryam Jahanbakhsh, Hamid Reza Dehghan Published: 2026 ABSTRACT backgroundPaper-based medical records remain prevalent in many developing countries, limiting the accuracy and efficiency of healthcare delivery. Electronic Medical Record (EMR) offer significant potential to improve care quality, yet their adoption faces infrastructural, financial, and contextual challenges in resource-limited settings.MethodsA systematic review of five databases was conducted to identify studies on EMR design, development, and implementation. Eligible articles underwent qualitative analysis to identify key components and challenges. These findings were synthesized into a conceptual model, using the System Development Life Cycle (SDLC) as a comprehensive Conceptual map to cover all system phases rather than a linear execution plan.ResultFrom 80 eligible studies, qualitative analysis yielded 204 refined codes, synthesized into a three-layered framework (Foundation, Core Cycle, and Optimization) comprising seven conceptual categories: strategic foundations, system architecture, user experience, interoperability standards, implementation processes, ethical considerations, and emerging technologies. Together, these categories outline the key dimensions of EMR system development in resource-limited settings.ConclusionsThe proposed model offers a constructive and iterative approach for EMR implementation in resource-limited settings. By distinguishing between core necessities and optimization features, the framework allows for a flexible, scalable adoption strategy that addresses local infrastructural constraints better than rigid, linear models.Supplementary InformationThe online version contains supplementary material available at https://doi.org/10.1186/s12911-026-03488-z. Introduction In many developing countries, clinical data are still maintained in paper-based records. The continued reliance on this traditional approach is attributed to its ease of use, limited awareness of modern technologies, infrastructural constraints, and physicians’ efforts to accelerate the provision of healthcare services. However, this method not only fails to benefit patients but also poses a significant barrier to the accurate, timely, and safe provision of healthcare services [1]. With advancements in digital technologies, many activities within the healthcare domain have shifted toward digitalization [2]. Digital health refers to the use of technology in delivering care services and enhancing the quality of healthcare. The digitalization of healthcare contributes to the standardization of care delivery, reduction of medical errors, improvement in diagnostic processes, and enhancement of quality of life [3]. An important early step in hospital digitalization and a fundamental requirement for expanding digital maturity is the implementation of an electronic medical record (EMR) [4]. EMR stands out as one of the most critical components in enhancing the quality of healthcare services. EMR is a digital record that provides stakeholders with immediate access to patient health information, facilitates care coordination, and improves physician–patient communication [5]. Each access to paper-based records incurs a cost of nearly $3 for the organization, whereas EMR significantly reduces these expenses [6]. Additional benefits of EMR include improved legibility of clinical notes, alerts for allergies and drug interactions, treatment reminders, notifications of abnormal laboratory results, and support for program monitoring and clinical research [7]. Moreover, many governments are exploring the integration of EMR data through cloud-based platforms to enhance diagnostic accuracy, improve treatment quality, and reduce medical costs [8]. Nevertheless, the successful implementation of EMR in developing countries faces numerous challenges, including demographic characteristics, geographical location of healthcare facilities, limited access to electricity and internet, high costs, lack of information standards, security concerns, and budget constraints [9]. This gap between planning and practical execution—especially in countries affected by crises such as epidemics, civil wars, and natural disasters—has hindered the optimal utilization of EMR [6, 10]. To address these challenges, the conceptual model of this study was informed by the System Development Life Cycle (SDLC). SDLC provides a systematic process for planning, designing, implementing, and evaluating software systems, and its stages offer a practical lens through which EMR development can be analyzed [11]. Furthermore, by adapting this framework to the specific conditions of developing countries, the proposed model seeks to reduce health disparities and promote equity in healthcare [12]. Although the systematic review encompassed studies from across the world, the conceptual model was specifically designed to address the unique challenges of EMR implementation in developing countries, where resource constraints, infrastructure limitations, and interoperability issues are more pronounced. While EMR design principles are globally relevant, this study emphasizes LMIC contexts in which constraints such as limited power supply, connectivity challenges, and governance structures shape unique design requirements [13]. Accordingly, the primary research question guiding the systematic review was formulated as follows: What empirically supported EMR elements, governance mechanisms, and architectural considerations have been reported in LMIC contexts that can inform the development of a context-appropriate EMR architecture? It is important to note that the systematic literature review was utilized to identify and synthesize these core components. Subsequently, a distinct conceptual design phase was undertaken by the research team to structure, integrate, and operationalize these findings into the proposed conceptual model. Method This systematic review was conducted and reported in accordance with the PRISMA 2020 statement. The completed checklist is provided in Appendix 1. Registration and search strategy A systematic review of the literature was conducted across PubMed, Embase, Scopus, ProQuest, and Web of Science. Search strategies were developed using combinations of keywords related to the design, development, and implementation of electronic medical record (EMR) systems, linked with Boolean operators (“AND”, “OR”) and, when necessary, wildcard symbols (*). core keywords included terms such as component, interoperability, challenge, solution, implementation, development, design, cloud computing, and digital medical record. Additional relevant studies were identified through manual searches in Google Scholar. Detailed search strategies for each database are provided in Appendix 2. Inclusion and exclusion criteria Studies were eligible for inclusion if they were written in English, published between January 1, 2015, and March 1, 2025, and directly addressed the design, development, or implementation of EMR systems. Qualitative, quantitative, and mixed-methods studies were considered relevant. Studies were excluded if they focused only on general health information systems without specific reference to EMR, lacked practical or conceptual modeling information, or were non-research items (e.g., perspectives, commentaries, expert opinions, conference abstracts without full text, debates, reports, letters to the editor, and editorials). Records without abstracts and duplicate entries retrieved across databases were also excluded. Selection process All retrieved articles were imported into EndNote to remove duplicates. After deduplication, two reviewers independently screened the titles and abstracts against the predefined inclusion criteria. Any discrepancies between the reviewers were resolved through discussion, with arbitration by the corresponding author. Full texts of potentially relevant studies were then assessed for eligibility. The overall number of records identified, screened, excluded, and included was documented and presented in a PRISMA flow chart in the Results section. Data extraction and framework for synthesis A structured template was developed to extract relevant information from each eligible study, including first author, year of publication, type of article, study design, and key findings. Data extraction was independently performed by two reviewers and subsequently verified by the corresponding author to enhance reliability and minimize bias. To systematically analyze the data, qualitative content analysis was conducted using NVivo Enterprise 20 (Lumivero). The full texts of the selected articles were coded independently, resulting in 787 initial codes. To reduce redundancy and enhance conceptual clarity, codes that represented overlapping or closely related concepts were collaboratively merged and grouped into broader thematic categories. Through this iterative process, the research team refined the dataset to 204 unique codes. A third reviewer contributed to the categorization and consolidation to ensure conceptual coherence and consistency across domains. To ensure methodological rigor, the coding process was structured into multiple levels. Initial codes were generated directly from the textual data (specific codes). These codes were then grouped into clusters based on conceptual similarity (code clusters), forming higher-order categories. From these clusters, thematic axes were developed to represent the central ideas emerging across studies. Finally, these thematic axes were synthesized into broader conceptual domains that captured the overarching dimensions of EMR system development. This hierarchical process (specific code → code cluster → thematic axis → conceptual domain) provided a transparent pathway from raw data to the final conceptual model. Extracted and synthesized data were then organized to identify key components, implementation challenges, proposed solutions, and technical and organizational requirements for EMR system development. The findings were integrated into a conceptual model reflecting the contextual realities of developing countries. The specific codes and themes, categorized across seven core conceptual domains and subsequently synthesized into the three-layered model (depicted in Fig. 3), are detailed in Appendix 3. To bridge the gap between the systematic review findings and the proposed framework, a traceability approach was employed. Table 1 demonstrates how specific empirical evidence and identified challenges from the SLR directly informed the design logic and shaped the specific structural components of the proposed multi-layered framework. Table 1Traceability matrix: from SLR evidence to EMR framework design componentsSLR Evidence / Identified ChallengeDesign Logic & RationaleResulting Component in Proposed FrameworkFragmented systems and proprietary silos prevent data exchange (High frequency in SLR).Data exchange protocols and coding standards must be established as non-negotiable prerequisites before technical deployment begins.Foundational Layer: Standards & Interoperability (ICD-10, HL7, DICOM) established as core prerequisites.High implementation failure rates linked to poor user interfaces and lack of clinical staff involvement.End-users must be central to the development cycle, requiring continuous feedback loops rather than a linear, one-time handover.Core Functional Cycle: Iterative User Evaluation and User Interface & Experience (UI/UX) modules integrated into the design loop.Varied infrastructural readiness across different LMIC hospital sizes (e.g., rural clinic vs. urban center).The system architecture cannot be one-size-fits-all; it must offer scalable options depending on local hardware and connectivity constraints.Core Functional Cycle: Adaptable Architecture selection (ranging from flat files to relational databases or cloud systems).Patient data breaches and lack of institutional trust hinder EMR adoption.Security cannot be an afterthought; fundamental safeguards must be explicitly engineered into the basic operational layer.Core Functional Cycle: Security & Privacy module (Authentication, Encryption, Access Control) mandated for basic operation.Desire for advanced tech (AI, Blockchain) exists, but basic digitization struggles to survive power and resource limits.Advanced features must not drain resources needed for basic clinical documentation. They require a distinct, subsequent implementation phase.Optimization Layer: Advanced tech isolated into a distinct Optimization & Advanced Integration phase, pursued only after core stability. Table 1Traceability matrix: from SLR evidence to EMR framework design componentsSLR Evidence / Identified ChallengeDesign Logic & RationaleResulting Component in Proposed FrameworkFragmented systems and proprietary silos prevent data exchange (High frequency in SLR).Data exchange protocols and coding standards must be established as non-negotiable prerequisites before technical deployment begins.Foundational Layer: Standards & Interoperability (ICD-10, HL7, DICOM) established as core prerequisites.High implementation failure rates linked to poor user interfaces and lack of clinical staff involvement.End-users must be central to the development cycle, requiring continuous feedback loops rather than a linear, one-time handover.Core Functional Cycle: Iterative User Evaluation and User Interface & Experience (UI/UX) modules integrated into the design loop.Varied infrastructural readiness across different LMIC hospital sizes (e.g., rural clinic vs. urban center).The system architecture cannot be one-size-fits-all; it must offer scalable options depending on local hardware and connectivity constraints.Core Functional Cycle: Adaptable Architecture selection (ranging from flat files to relational databases or cloud systems).Patient data breaches and lack of institutional trust hinder EMR adoption.Security cannot be an afterthought; fundamental safeguards must be explicitly engineered into the basic operational layer.Core Functional Cycle: Security & Privacy module (Authentication, Encryption, Access Control) mandated for basic operation.Desire for advanced tech (AI, Blockchain) exists, but basic digitization struggles to survive power and resource limits.Advanced features must not drain resources needed for basic clinical documentation. They require a distinct, subsequent implementation phase.Optimization Layer: Advanced tech isolated into a distinct Optimization & Advanced Integration phase, pursued only after core stability. Traceability matrix: from SLR evidence to EMR framework design components Finally, the synthesis of findings was guided by the SDLC framework, structured not as a linear checklist but as a layered framework. The conceptual model presented in Fig. 3 illustrates this stratified approach: A Foundational Layer (Objectives, Standards) that establishes the necessary infrastructure;A Core Functional Cycle (Design, Functionality, Privacy, Implementation) representing the iterative execution of the EMR system; and.An Optimization Layer (Advanced IT), depicted as a distinct component to indicate that technologies like AI and Blockchain are supplementary enhancements rather than immediate prerequisites for LMICs. A Foundational Layer (Objectives, Standards) that establishes the necessary infrastructure; A Foundational Layer (Objectives, Standards) that establishes the necessary infrastructure; A Core Functional Cycle (Design, Functionality, Privacy, Implementation) representing the iterative execution of the EMR system; and. A Core Functional Cycle (Design, Functionality, Privacy, Implementation) representing the iterative execution of the EMR system; and. An Optimization Layer (Advanced IT), depicted as a distinct component to indicate that technologies like AI and Blockchain are supplementary enhancements rather than immediate prerequisites for LMICs. An Optimization Layer (Advanced IT), depicted as a distinct component to indicate that technologies like AI and Blockchain are supplementary enhancements rather than immediate prerequisites for LMICs. This structure directly addresses the need for prioritization in resource-constrained settings. Figure 4 subsequently maps these components to the SDLC phases to demonstrate operational alignment. Methodological Rationale: A Global-to-Local Approach. To address this limitation, this study adopted a Systematic Literature Review (SLR) to synthesize globally validated EMR components and design principles derived largely from mature health information systems. Unlike scoping reviews, which primarily map broad conceptual landscapes, the SLR approach enabled a rigorous, evidence-based synthesis of empirically supported EMR elements, governance mechanisms, and architectural considerations. Importantly, these globally identified components were not transferred directly into the proposed framework. Instead, a contextual filtering and adaptation process was applied, whereby each component was critically examined against well-documented LMIC constraints, including infrastructural instability, workforce shortages, limited financial resources, and evolving governance capacities. This process allowed the framework to retain global best practices while explicitly accounting for local feasibility and sustainability. To address the wicked nature of EMR implementation in LMICs, this study adopts an iterative and agile-compatible interpretation of the SDLC, explicitly rejecting the rigid, linear assumptions of the traditional Waterfall model. In our framework, the SDLC is not employed as a prescriptive or sequential implementation plan, but rather as a foundational conceptual scaffold. It incorporates essential mechanisms for complex environments, including stakeholder co-design, continuous feedback loops, and dynamic adaptation. By utilizing this iterative SDLC approach, the model ensures that requirement analysis and system design are not static, but continuously evolve in response to the shifting infrastructural, socio-technical, and governance realities of developing countries. Critical appraisal / Risk of bias assessment Two reviewers independently conducted quality appraisals of the included studies using the Quality Assessment with Diverse Studies (QuADS) tool. This instrument comprises 13 common criteria (e.g., theory, aims, setting, design, sampling, data collection, analysis, stakeholder involvement, and discussion of strengths/limitations), each rated on a scale from 0 (no mention) to 3 (complete) [14]. Discrepancies between the two reviewers were resolved through discussion, with input from a third author. In line with recommendations for the QuADS tool, no overall score was calculated for individual studies; instead, findings were reported narratively across individual criteria rather than as a composite score. A detailed consensus table is provided in Appendix 4. Results A total of 1,810 records were identified through database searches. After removing 1,507 duplicates, 303 unique records remained for screening. Of these, 114 were excluded at the title and abstract stage for not aligning with the study objectives. Full texts were sought for 189 reports, but 24 could not be retrieved. The remaining 165 full-text reports were assessed for eligibility, of which 85 were excluded after detailed review. The main reasons for exclusion were the absence of concrete EMR design components, insufficient technical implementation detail, exclusive focus on clinical outcomes without system-level analysis, or lack of relevance to resource-limited settings. Ultimately, 80 studies met the inclusion criteria and were incorporated into the review. The detailed selection process is illustrated in the PRISMA 2020 flow diagram (Fig. 1). Fig. 1Prisma 2020 flow diagram for the current systematic review based on database searches only Prisma 2020 flow diagram for the current systematic review based on database searches only Quality assessment Across the 80 included studies, clarity of aims (240) and appropriateness of study design (238) consistently emerged as the strongest domains. Most studies also demonstrated robust analytic approaches (238) and provided adequate format of data collection tools (226). Theoretical or conceptual underpinnings (220) and the rationale for data collection (201) were generally present, though often described with moderate depth rather than comprehensive detail. By contrast, several domains were notably weaker. Sampling [15] and recruitment [16] represented the lowest scoring criteria, with many studies offering only minimal information on participant selection or dataset inclusion. The description of data collection procedures (185) and justification of analytic methods (167) were also frequently limited, reflecting a lack of transparency in methodological reporting. Evidence of stakeholder involvement (136) was moderate overall, typically restricted to end-users or clinicians, with relatively few studies engaging multiple stakeholders or incorporating their perspectives into system design and interpretation of findings. Similarly, the discussion of strengths and limitations (159) was often brief, with many studies providing only a basic reflection on methodological constraints. In line with recommendations for the QuADS tool, no overall score was calculated for individual studies. Instead, aggregated counts across criteria were used to illustrate relative strengths and weaknesses, and findings were reported narratively. Taken together, the results indicate that while most studies demonstrated clarity in aims, appropriate design, and sound analytic approaches, important gaps remain in sampling, recruitment, stakeholder engagement, and the articulation of strengths and limitations. Addressing these areas would strengthen transparency, reproducibility, and the overall quality of EMR-related research. To better understand the scope and diversity of the reviewed studies, key characteristics of the 80 selected articles—including publication year, country of origin, article type, and thematic focus on EMR design, development, or implementation—are presented in Appendix 5. The final codes were organized into seven conceptual categories, each addressing a distinct aspect of the research topic and contributing to a coherent structure of the findings: Foundations and Objectives: Encompassing the necessity, strategic goals, and rationale behind the design of Electronic Medical Records (EMR).System Design and Architecture: Covering technical structures and design patterns.Functionality and user experience: Including system performance, ease of use, and user interaction.Standards and Integration: Focusing on data exchange and compliance with interoperability standards.Implementation, Evaluation, and Maintenance: Addressing execution phases and maintenance challenges.Privacy, security, and ethics: Highlighting legal considerations and ethical principles.Enhancing data usability using advanced information technology: Exploring the role of artificial intelligence, cloud computing, and machine learning. Foundations and Objectives: Encompassing the necessity, strategic goals, and rationale behind the design of Electronic Medical Records (EMR). Foundations and Objectives: Encompassing the necessity, strategic goals, and rationale behind the design of Electronic Medical Records (EMR). System Design and Architecture: Covering technical structures and design patterns. System Design and Architecture: Covering technical structures and design patterns. Functionality and user experience: Including system performance, ease of use, and user interaction. Functionality and user experience: Including system performance, ease of use, and user interaction. Standards and Integration: Focusing on data exchange and compliance with interoperability standards. Standards and Integration: Focusing on data exchange and compliance with interoperability standards. Implementation, Evaluation, and Maintenance: Addressing execution phases and maintenance challenges. Implementation, Evaluation, and Maintenance: Addressing execution phases and maintenance challenges. Privacy, security, and ethics: Highlighting legal considerations and ethical principles. Privacy, security, and ethics: Highlighting legal considerations and ethical principles. Enhancing data usability using advanced information technology: Exploring the role of artificial intelligence, cloud computing, and machine learning. Enhancing data usability using advanced information technology: Exploring the role of artificial intelligence, cloud computing, and machine learning. Figure 2 presents a sunburst chart illustrating the distribution of extracted codes across seven conceptual categories. Fig. 2Distribution of extracted codes across conceptual categories Distribution of extracted codes across conceptual categories Each category represents a distinct dimension of EMR design and implementation as reported in the literature. Foundations and Objectives – Studies described the transition from paper-based to digital records, highlighting the role of information technology in organizing medical data and enabling secure platforms for data sharing [17, 18]. Reported design objectives included the development of storage capacity and integrated data sharing [19–24], comprehensive patient record management [22, 25–30], support for clinical decision-making and health research [17, 23, 31–36], and the promotion of equity in access to healthcare services [7]. Reported benefits encompassed improved record structure and usability [37–39], active patient–physician engagement [26, 40], enhanced quality of care and clinical safety [22, 27, 31, 41–44], preservation of privacy [25, 34, 36, 37, 45–49], resistance to tampering [50], increased speed and efficiency in managing patient information [29, 36, 37, 51–56], and reduced healthcare costs [48]. System Design and Architecture – Studies addressed architectural design, logical structures, and platform selection in the conceptual modeling domain [26, 31, 36, 39, 47, 49, 53, 57–64]. Notable examples included the use of the Chord protocol in peer-to-peer networks [60], XDS profile-based implementation with RESTful services on Azure [65], and identification of key actors in data-sharing frameworks such as IPFS and blockchain technologies [41, 66]. Several studies adopted the Dual Model approach, comprising a reference model and archetypes, while semantic web technologies and the OWL language were emphasized in the development of open software platforms [67].In terms of functional and non-functional requirements, articles highlighted data structures and unique identifiers [30, 62, 64, 68, 69], hybrid development approaches [16, 27, 33, 62], and success factor analysis [34, 36, 44, 45, 70]. User-centered design [30, 34, 36, 44, 46, 47, 57, 62], stakeholder engagement [30, 34, 40, 44, 46, 47, 57, 62, 71], and infrastructure standardization [18, 27, 34, 42, 45, 72, 73] were also frequently reported. Ontology was introduced as a key tool for semantic data extraction [74, 75], with studies addressing quality assessment, resource enrichment [67], and educational challenges related to its implementation [74]. Functionality and User Experience – Reported functionalities included educational tools [44, 50], real-time processing, data validation, and secure information sharing [37, 44, 50, 61, 76]. User experience and interface design encompassed client-side and server-side technologies, data presentation templates [53, 61, 63], and the development of applications and standardized APIs [21, 26, 57, 62–64, 77, 78]. Effective interaction, ease of learning, and alignment with real-world tasks were considered key components of successful design [79]. Some studies also highlighted the consequences of poor design, including medical errors and professional burnout [40]. In terms of system modules, studies explored extended functionalities, data retrieval mechanisms, patient interaction features [21, 30, 60, 76, 80], and role-based management systems [21, 23, 32, 36, 42, 64]. Decision support systems included intelligent diagnostic tools [29, 77, 81, 82], personalized recommendations [56, 77, 83], and alerts and reminders aimed at enhancing the quality of care [15, 24, 32]. Standards and Integration – Interoperability standards such as HL7, FHIR, SNOMED CT, CDA, and LOINC were widely cited for enhancing semantic consistency of medical data [15, 16, 23, 32, 49, 74, 84]. Studies also described the use of data dictionaries and exchange protocols to support modeling, storage, and secure transmission of health information [16, 47, 50, 65, 72]. FHIR was frequently highlighted for its ease of implementation and support for RESTful architecture, overcoming limitations of HL7 V2 and V3 [16, 17, 20, 23, 27, 31, 32, 38, 39, 49, 54, 58, 60, 63, 73, 78, 80, 84, 85]. Reported challenges included heterogeneity of medical data [52], investment instability [31], and concerns regarding data quality and consent [16, 31, 39, 43, 63, 65, 72, 78, 81, 86]. Implementation, Evaluation, and Maintenance – Reported barriers to adoption included cost-related concerns [17, 28, 29, 34, 43, 55, 61, 68, 71], interoperability, and security issues, as well as the absence of comprehensive approaches [16, 17, 25, 32, 35, 43, 50, 58, 73, 74, 87]. In terms of infrastructure, studies emphasized agile methodologies, active stakeholder engagement [58], and participatory data collection techniques such as observation and workshops [33, 44, 71]. Evaluation methods encompassed usability assessment, learning curves, and technology acceptance [46, 55, 85], alongside performance monitoring and data mining model evaluation [52, 79, 88, 89]. Risk management and continuous monitoring throughout development and implementation phases were also highlighted [18, 34, 51, 57, 62]. Maintenance strategies focused on continuous training [45, 71], staff empowerment [30, 43, 47, 55, 64], participatory development approaches [43, 57, 71], and recognition of cultural and legal challenges [54]. Privacy, Security, and Ethics – Reported strategies for protecting medical data included network infrastructure and risk management [39, 56, 72, 79, 90], security algorithms [18, 19, 22, 39, 41, 48, 72, 79], and technologies such as distributed ledger and smart contracts [20, 39, 41, 48, 54, 64, 72, 91]. Studies also noted threats arising from software vulnerabilities and human error [20], as well as the need for staff training to safeguard medical information [18, 29, 32, 48, 54]. Authentication mechanisms focused on secure access and precise user control [24, 39, 41, 51, 90], identity-based stealth addresses [41], mutual authentication and anonymization techniques [87, 90], and highlighted vulnerabilities in biometric schemes (63,93). Digital signatures such as ElGamal [56], PKI, and S/MIME were proposed to ensure non-repudiation and signature integrity [39]. Encryption methods investigated included ECC [42], ABE [19, 38, 51, 91, 92], CP-ABE with the SM4 algorithm [19], KP-ABKS [91], and symmetric encryption approaches [42, 90, 91]. Reported challenges involved inefficiencies of traditional algorithms [48] and technical complexities in implementation [22]. Emerging Technologies – Advanced technologies such as machine learning, cloud computing, and blockchain have been proposed to enhance data efficiency and strengthen the management of health information[83]. Reported applications include disease prediction, data integration, and secure storage [21, 77, 93]. While these approaches offer promising pathways for EMR development, challenges such as technical complexity, limited transaction speed, and constraints in updating records remain. Overall, These approaches highlight the growing role of advanced technologies in digital health[22]. In the content analysis of scholarly articles related to EMR, thematic categories were re-ranked based on the number of unique codes to reflect research priorities. The highest number of unique codes was identified in Enhancing Data Usability Using Advanced Information Technology (77 codes from 89 total codes across 33 articles). The second largest category was Privacy, Security, and Ethics (33 unique codes from 83 total codes across 41 articles). Foundations and Objectives ranked third (22 unique codes from 118 total codes across 66 articles). The subsequent categories included Standards and Integration (20 unique codes from 62 codes across 38 articles), System Design and Architecture (19 unique codes from 68 codes across 40 articles), Implementation, Evaluation, and Maintenance (15 unique codes from 60 codes across 39 articles), and Functionality and User Experience (17 unique codes from 56 codes across 37 articles). Discussion The findings of this review confirm that EMR systems play a pivotal role in enhancing care quality, improving information accessibility, reducing medical errors, and increasing operational efficiency across diverse healthcare settings. However, these benefits are not automatic; they depend on context-sensitive design choices that reflect the specific needs of clinical environments. Unlike generic database systems, such as those used in financial transactions, EMR systems must accommodate complex clinical workflows, patient diversity, and sensitive health information. This includes balancing trade-offs between simplified interfaces for clinicians and extensive data structures required for research and analytics, as well as ensuring privacy while enabling patient access and secure third-party data exchange [94]. In low-income countries, pilot implementations have demonstrated that EMR deployment is feasible when design strategies are tailored to local constraints, including limited infrastructure, human resources, and financing models. The case of SEPAS, Iran’s national EMR initiative [95], illustrates the challenges of bridging policy-level ambitions with clinical utility. Although originally designed to collect patient data from all hospital encounters in a centralized repository, its real-world application remains largely confined to administrative domains. The limited engagement of healthcare professionals in clinical settings reflects a functional gap between the system’s intended goals and its actual performance [96]. To address these challenges, the proposed conceptual model in this study is mapped to the Software Development Life Cycle (SDLC), offering a structured framework for EMR design and implementation. While SDLC provides a useful foundation [97], The proposed model presented in this study, as illustrated in Fig. 3, conceptually and operationally aligns with the SDLC phases, emphasizing an iterative development process, and can serve as a practical framework for clinical environments in resource-limited countries. Fig. 3Proposed conceptual model for the design and implementation of EMR in developing countries. Note: The model is structured into three layers: Foundation, Core Cycle, and Optimization. The central loop indicates the iterative nature of system design and implementation. The “Advanced Information Technology” domain is connected via a dashed line, signifying that it is an advanced optimization layer to be integrated only after the core functional stability is achieved Proposed conceptual model for the design and implementation of EMR in developing countries. Note: The model is structured into three layers: Foundation, Core Cycle, and Optimization. The central loop indicates the iterative nature of system design and implementation. The “Advanced Information Technology” domain is connected via a dashed line, signifying that it is an advanced optimization layer to be integrated only after the core functional stability is achieved The alignment between the proposed EMR model and the SDLC framework offers a structured lens for understanding system development. The Foundations and Objectives component corresponds to the Planning phase, where strategic goals, organizational requirements, and overarching visions are defined [98]. In LMIC contexts, this phase is particularly critical, as systems must be tailored to local constraints such as limited infrastructure, workforce shortages, and fragmented governance structures (65). Strategic planning must also address dual imperatives: ensuring data privacy and security while enabling fast and reliable access to clinical information [51]. The subsequent components—Standards and Integration, Privacy, Security, and Ethics, Advanced Information Technologies, and Functionality and User Experience—reflect the Analysis phase, which involves identifying functional requirements, interoperability standards, and user expectations [98]. One persistent challenge in LMICs is the limited awareness and adoption of communication protocols such as HL7, which impedes system integration and slows interface development [53]. Addressing these gaps requires not only technical solutions but also capacity-building and stakeholder engagement. Emerging technologies such as blockchain offer promising alternatives to centralized architectures. Their potential benefits—privacy preservation, auditability, decentralization, and automation via smart contracts—are well documented [99]. However, their implementation in LMICs must be carefully evaluated against infrastructure realities, transaction speed limitations, and governance capacity. Similarly, semantic modeling through openEHR templates can enhance data quality and interoperability [67], but requires localized training and adaptation. User-centered design is another critical factor in clinical effectiveness. Studies show that EMR tools aligned with clinician workflows—such as those supporting atrial fibrillation management—receive more positive feedback and improve decision-making [30]. Yet, usability remains underprioritized in many implementations, suggesting a need to rebalance technical innovation with human-centered approaches. The System Design and Architecture component maps to the Design phase of SDLC, encompassing technical structures and conceptual modeling [98]. Modular architectures such as three-tier designs support scalability and maintainability [100], while integration profiles like IHE XDS enable secure inter-organizational data exchange [101]. However, these design choices must be adapted to LMIC realities, including intermittent connectivity and limited technical support. The final model component—Implementation, Evaluation, and Maintenance—spans the remaining SDLC phases. Successful EMR deployment requires not only technical rollout but also staff training, usability evaluation, and long-term support. Mixed-methods approaches have been used to assess workflow integration and clinical impact [30], while decision-making frameworks help identify key performance indicators [45]. Post-deployment sustainability hinges on continuous feedback, incremental updates, and local empowerment [62, 64]. In addition to presenting the seven key themes, we have elaborated on their implications for the design and development of EMR systems in developing countries: Privacy: This theme highlights the necessity of multi-level access control. In practice, EMR systems must differentiate between patient, provider, and administrator roles, ensuring that sensitive data is protected while still accessible to authorized users. Interoperability: The review emphasized the importance of data exchange. To address this, EMR systems should adopt international standards such as HL7 or FHIR, which enable seamless integration with external systems and reduce fragmentation. Simplicity: A recurring concern was usability. For clinical providers, EMR systems must offer user-friendly interfaces aligned with workflows, minimizing cognitive load and reducing documentation time. Data duplication: Several studies noted duplication of patient records. A practical design implication is the implementation of unique patient identifiers to prevent multiple or conflicting records, thereby improving data reliability. Security: Beyond privacy, robust security measures are essential. Encryption protocols and strong authentication mechanisms must be embedded to safeguard health information against unauthorized access. Patient-centered care: The literature underscored the importance of involving patients. EMR systems should incorporate patient portals that allow individuals to access their records, while maintaining confidentiality and respecting cultural sensitivities. Scalability: Finally, scalability emerged as a critical theme. EMR systems in developing countries should adopt modular architectures that can expand or adapt to diverse care contexts and resource constraints. From linear structure to iterative implementation While the proposed model utilizes the SDLC phases as a structural map to categorize EMR domains comprehensively, we acknowledge that EMR implementation in healthcare—especially in LMICs—is a wicked problem characterized by shifting requirements and unstable environments. A rigid, linear execution of SDLC phases (Waterfall model) is often ill-suited for such contexts. Therefore, the alignment presented in Fig. 4 represents a conceptual mapping rather than a prescriptive chronological sequence. Our model advocates for an adaptive and iterative approach. The design and implementation phases should not be viewed as isolated steps but as continuous cycles. For instance, feedback from pilot implementations (a key component in the ‘Transition’ phase) must loop back to refine the ‘Design’ requirements dynamically. This hybrid strategy ensures that while the components of the model are standardized via SDLC, the process of deploying them remains agile enough to respond to the unpredictable infrastructural and organizational constraints typical of developing countries. Stratification of components: core vs. advanced functionalities To address the challenge of an overly broad scope, it is critical to distinguish between foundational components and advanced enhancements. For LMICs starting their digitalization journey, the priority must be establishing the Core Functional Layer—comprising secure patient registration, basic clinical documentation, and offline data buffering. These elements address immediate infrastructural gaps. Advanced technologies identified in this review, such as Blockchain for integrity or AI for predictive analytics, constitute an Optimization Layer. While valuable, these should be viewed as secondary phases to be deployed only after the core infrastructure achieves stability. Treating these layers as simultaneous requirements often leads to implementation failure due to resource exhaustion. To demonstrate the practical applicability of the proposed framework, Table 2 explicitly maps the identified infrastructural, organizational, and resource constraints typical of LMIC settings to the specific architectural features and design strategies integrated into our conceptual model. Table 2Mapping of LMIC constraints to EMR model architectural featuresLMIC Constraint / ChallengeManifestation in Healthcare SettingsProposed Model Feature / Architectural ResponseInadequate IT Infrastructure & Intermittent ConnectivityUnreliable internet access; frequent network drops during clinical workflows.Core Functional Layer Focus: Prioritization of offline data buffering capabilities and resilient architectures before cloud reliance.Power InstabilityFrequent electricity outages interrupting system access and data entry.Adaptable Architecture: Support for lightweight solutions (e.g., standalone systems or flat files) for remote areas requiring low power overhead.Limited Financial ResourcesInability to afford expensive proprietary software licenses or high-end servers.Scalable & Open-Source Prioritization: Recommendation for open-source frameworks to minimize initial capital expenditure while allowing future scalability.Human Resource Capacity & IT LiteracyResistance to unfamiliar technologies; lack of highly trained on-site IT support staff.Stratified Implementation: Delaying advanced tech (AI, Blockchain) to the Optimization Layer; focusing first on basic documentation and secure registration to avoid user cognitive overload.Weak Policy & Standardization GovernanceFragmented health policies and lack of standardized data protocols across regions.Foundational Layer Prerequisites: Mandating the establishment of clear objectives, standards, and governance protocols before technical deployment.Geographic & Demographic VariabilityDiverse needs ranging from small, isolated rural clinics to large urban hospital networks.Context-Aware Scaling: Flexible system selection (lightweight databases for small clinics vs. robust coded databases for larger networks). Table 2Mapping of LMIC constraints to EMR model architectural featuresLMIC Constraint / ChallengeManifestation in Healthcare SettingsProposed Model Feature / Architectural ResponseInadequate IT Infrastructure & Intermittent ConnectivityUnreliable internet access; frequent network drops during clinical workflows.Core Functional Layer Focus: Prioritization of offline data buffering capabilities and resilient architectures before cloud reliance.Power InstabilityFrequent electricity outages interrupting system access and data entry.Adaptable Architecture: Support for lightweight solutions (e.g., standalone systems or flat files) for remote areas requiring low power overhead.Limited Financial ResourcesInability to afford expensive proprietary software licenses or high-end servers.Scalable & Open-Source Prioritization: Recommendation for open-source frameworks to minimize initial capital expenditure while allowing future scalability.Human Resource Capacity & IT LiteracyResistance to unfamiliar technologies; lack of highly trained on-site IT support staff.Stratified Implementation: Delaying advanced tech (AI, Blockchain) to the Optimization Layer; focusing first on basic documentation and secure registration to avoid user cognitive overload.Weak Policy & Standardization GovernanceFragmented health policies and lack of standardized data protocols across regions.Foundational Layer Prerequisites: Mandating the establishment of clear objectives, standards, and governance protocols before technical deployment.Geographic & Demographic VariabilityDiverse needs ranging from small, isolated rural clinics to large urban hospital networks.Context-Aware Scaling: Flexible system selection (lightweight databases for small clinics vs. robust coded databases for larger networks). Mapping of LMIC constraints to EMR model architectural features Fig. 4Conceptual mapping of the proposed EMR model components to the SDLC taxonomic structure (Note: Arrows indicate logical grouping, not a strictly linear workflow) Conceptual mapping of the proposed EMR model components to the SDLC taxonomic structure (Note: Arrows indicate logical grouping, not a strictly linear workflow) In developing countries, EMR implementation faces persistent challenges, including limited human and financial resources, inadequate IT infrastructure, and resistance among staff toward unfamiliar technologies. As Khalifa notes, human and financial factors are among the most significant barriers to adoption, despite the potential benefits of EMR in improving access to information, enhancing care quality, and reducing medical errors [10]. Studies recommend that low-income countries prioritize open-source and scalable systems that align with local economic and infrastructural realities [2, 94]. System selection must account for demographic and geographic factors. For small or remote populations, lightweight solutions such as flat files or standalone systems may be more feasible, while larger networks require coded databases and resilient architectures capable of operating under intermittent connectivity [9]. In such contexts, system design must be robust enough to accommodate limitations in electricity, internet access, and technical support. Organizational readiness—technical, human, cultural, managerial, and financial—is a critical determinant of post-implementation success. While comprehensive readiness planning is ideal, some LMICs may need to pursue parallel strategies, initiating deployment while gradually building capacity [96]. Active involvement of healthcare providers, especially physicians, is essential for system acceptance and sustained use. Yet, many LMICs face shortages in clinical leadership, funding, and ICT infrastructure, which hinder continuity of care and effective technology utilization [6, 10]. At the national level, integrating EMR systems requires coordinated governance. Establishing platforms for health data sharing demands collaboration among ministries, insurers, and regulators, supported by clear legal frameworks and implementation strategies [57]. Without such coordination, EMR systems risk fragmentation and limited impact. Balancing technical rigor with clinical usability A significant analytical insight from this review is the inherent tension between technical security measures and human-centered usability. While the literature emphasizes robust encryption and complex authentication protocols (technical dimension), field reports indicate that overly stringent security mechanisms can impede clinical workflows, leading to user resistance or workarounds by staff (human dimension). In LMIC settings, where digital literacy may vary, prioritizing complex security architectures without parallel investments in user interface (UI) simplicity creates a usage barrier. Therefore, our model suggests a trade-off analysis where security protocols are adapted to the local threat level to preserve clinician efficiency, rather than applying a one-size-fits-all high-security standard. Comparison with existing frameworks and contribution To contextualize our model’s contribution, it is essential to compare it with established frameworks. Unlike explanatory frameworks such as NASSS (Non-adoption, Abandonment, Scale-up, Spread, and Sustainability)[102], which primarily focuses on diagnosing the complexity of health interventions and predicting barriers to adoption, our model adopts a constructive perspective. While NASSS effectively categorizes the wicked socio-technical challenges, it does not explicitly define the architectural building blocks required to build the system itself. Our study fills this gap by translating complexities identified in frameworks like NASSS into tangible, actionable technical domains (e.g., specific interoperability standards and offline data architectures). Furthermore, several EMR frameworks have been developed and widely adopted in high-income countries, including the WHO EMR and digital health frameworks, OpenEHR, and HL7-based architectures. These frameworks place strong emphasis on interoperability, standardized data exchange, and advanced system architectures designed to support large-scale, mature health information infrastructures[103–105]. While such models provide robust technical and architectural guidance, they largely reflect high-resource contexts and often insufficiently address the infrastructural, organizational, and workforce constraints commonly encountered in low- and middle-income countries[106]. Consequently, unlike existing frameworks, our conceptual model integrates generalizable components with strategies tailored to resource-limited settings. This dual focus adds value by bridging the gap between universal design principles and the specific challenges of developing countries. By synthesizing common EMR elements with LMIC-specific strategies—such as phased digital transformation, change management, and participatory implementation—the model offers a more context-sensitive pathway for sustainable EMR adoption. Research limitations Despite our rigorous efforts to develop a comprehensive conceptual model for EMR design and implementation in low- and middle-income countries (LMICs), this systematic review has several limitations that should be acknowledged. First, source heterogeneity and contextual applicability: The review synthesized evidence from both developed and developing countries. While we applied a “contextual filter” to adapt global findings to LMIC constraints, the foundational studies vary significantly in terms of IT infrastructure, health policies, and organizational cultures. Therefore, the applicability of specific model components requires local customization to align with unique realities such as power stability, connectivity, and workforce capacity. Second, nature of the evidence and publication bias: A notable limitation in the extant literature—which inevitably influenced our synthesis—is the predominance of technical and architectural studies over socio-technical inquiries. Our coding process revealed a heavy concentration on security algorithms, blockchain, and structural frameworks, with comparatively less empirical evidence on the “lived experience” of clinicians, change management, and human-computer interaction (HCI). While we attempted to balance this in our discussion, the model may still reflect this underlying skew toward technical solutions over usability and organizational dynamics. Third, lack of empirical validation: The proposed model is currently theoretical, derived from a synthesis of secondary data. It has not yet undergone real-world validation through methods such as Delphi techniques with experts or pilot implementation in a hospital setting. Consequently, its feasibility, user acceptance, and impact on clinical outcomes remain to be empirically tested. Finally, economic scope: While implementation costs are acknowledged as a barrier, this study does not provide a detailed economic evaluation. A comprehensive cost-benefit analysis, return on investment (ROI) assessment, and financial sustainability modeling were beyond the scope of this review but are essential for informing policy decisions in resource-limited environments. Conclusion EMR represent a strategic priority not only for high-income countries but also for developing nations, particularly in resource-constrained environments that require scalable, cost-effective, and locally adaptable solutions. The findings of this review provide a conceptual framework—mapped to SDLC principles—that highlights potential strategies for EMR design, implementation, and evaluation in LMICs. However, this framework should be interpreted as a conceptual guide rather than a prescriptive blueprint, given the heterogeneity of evidence and the absence of real-world validation. Lessons drawn from this synthesis emphasize the importance of tailoring EMR strategies to local conditions, including infrastructural limitations, workforce capacity, and governance structures. While successful experiences from other institutions and vendor-recommended practices may inform future frameworks, country-specific customization remains essential. Moreover, the findings underscore that technical innovation alone is insufficient; usability, clinician engagement, and iterative development cycles must be prioritized to ensure adoption and sustainability. Future research should focus on empirical testing of the proposed model in LMIC contexts, incorporating pilot studies, economic evaluations, and participatory approaches. Such efforts will be critical to confirm feasibility and applicability, and to advance EMR systems that not only meet technical standards but also align with the practical realities of diverse healthcare environments. Supplementary Information Below is the link to the electronic supplementary material. Supplementary Material 1 Supplementary Material 1