Authentic Digital Interaction with E-Government: A Systematic Review of Key Determinants

Abstract

Authentic Digital Interaction (ADI) refers to citizens’ direct, secure, and independent engagement with e-government services without reliance on intermediaries. This systematic literature review applies ADI as an organizing lens to synthesize recent empirical evidence on determinants shaping citizen interaction with e-government. Following PRISMA, 178 peer-reviewed studies published between January 2020, and October 2025 were identified across five databases, and 43 met the inclusion criteria. Descriptive mapping and ADI-guided narrative synthesis were used to consolidate related determinants and interpret their associations and contextual conditions. The review identifies three dominant patterns: perceived usefulness and performance expectancy are most frequently associated with intention, use, and continuance; trust and confidence shape whether perceived benefits translate into engagement; and policy and governance condition service consistency and the effects of usability and accessibility. Theoretically, the review shows that ADI provides a useful lens for interpreting e-government research beyond adoption and satisfaction by emphasizing direct, trustworthy, inclusive, and independent citizen interaction. Practically, the findings suggest that public agencies should prioritize accessible design, transparent processes, visible safeguards, and supportive governance arrangements. However, no formal risk-of-bias assessment was conducted. In addition, the evidence base remains limited by the sparse examination of participation, value co-creation, autonomy, and empowerment, and the review protocol, although prepared in advance, was not registered.

1. Introduction

Governments have expanded e-government and digital government services to improve efficiency, transparency, and access to public services [1,2]. In the Gulf Cooperation Council (GCC) and many developing contexts, these initiatives are part of national digital transformation agendas and smart-city programs that position digital channels as a key interface between citizens and the state [2,3]. Although comparative assessments report improvements in service availability and maturity, progress remains uneven across agencies, service types, and countries [4,5].

Despite wider availability, citizens do not always experience digital services as clear, safe, or easy to use. Across contexts, studies report recurring problems such as unclear procedures, weak usability and information design, accessibility barriers for older adults and people with disabilities, and continuing concerns about privacy and data misuse [6,7,8]. As a result, citizens may complete an online task but still feel uncertain, excluded, or exposed to risk. This means that usage can increase while interaction quality remains weak.

Much of the e-government literature explains citizen behavior using well-established information systems models, including TAM, UTAUT, and the DeLone and McLean IS Success model [9,10,11,12]. These approaches provide strong evidence that factors such as perceived usefulness, ease of use, service quality, and trust relate to intention, satisfaction, and continued use [13,14,15]. However, these models are often used in ways that treat adoption, intention, or satisfaction as the main indicators of success. In public-service settings, citizens may use online channels because they are required, because alternatives are limited, or because the digital route is the quickest. In such cases, adoption may reflect convenience or compliance rather than a confident and legitimate interaction.

To address this limitation, this review uses ADI as an organizing lens for synthesis. ADI refers to citizens’ direct, secure, and independent use of digital public services without intermediaries [16]. In this review, ADI guides how determinants reported in prior studies are grouped and interpreted so that the evidence speaks to independent, secure, and confidence-building interaction, not only adoption.

ADI should therefore be understood as distinct from several related but narrower constructs commonly used in e-government research. Adoption and intention indicate whether citizens are willing to use a service, but they do not show whether the interaction is clear, trustworthy, or independently manageable. Satisfaction reflects an evaluation after use, but it does not necessarily capture whether citizens understood the process, felt secure, or could complete the task without assistance. Trust, usability, accessibility, and service quality are also important, but in this review, they are treated as contributing conditions rather than substitutes for authentic interaction itself. ADI is used here to capture the broader quality of citizen–state digital interaction, especially whether citizens can engage directly, securely, confidently, and with minimal dependence on intermediaries.

Several systematic reviews and meta-analyses have already synthesized determinants of e-government adoption and use [17,18,19]. However, these reviews largely remain adoption-centric in their conceptual framing. They aggregate predictors of intention or usage, but they do not clearly separate use from authentic interaction, meaning whether citizens experience digital government as transparent, inclusive, safe, and responsive. Without this distinction, synthesis can overstate what current evidence implies about the real quality of citizen experiences and the legitimacy of digital service delivery.

An ADI-focused synthesis is timely for countries undergoing rapid digital transformation, including Saudi Arabia and other GCC states. In these contexts, digital government is central to reform agendas and citizens are increasingly expected to use online channels for essential services [2,3,20]. This increases the importance of understanding not only whether people use digital services, but also whether those services support independent, secure, and confident interaction.

Accordingly, this review makes three contributions. Conceptually, it reframes citizen digital engagement beyond adoption by distinguishing between simple use and authentic interaction. Methodologically, it synthesizes recent empirical evidence across socio-technical and institutional dimensions, rather than treating determinants as isolated effects. Practically, it identifies priorities for digital government design and governance that support authentic interaction, including reducing friction, ensuring accessibility by default, making safeguards and service status visible, and strengthening the capacity needed to deliver consistent outcomes. The remainder of this paper is structured as follows. Section 2 presents the review scope, research question, and methods. Section 3 reports the results of descriptive mapping and ADI-guided narrative synthesis. Section 4 discusses the findings, implications, limitations, and future research directions. Section 5 concludes the paper.

2. Materials and Methods

2.1. Research Scope and Review Question

Existing systematic reviews and meta-analyses provide valuable synthesis on e-government adoption, intention, satisfaction, and continuance, but they largely treat these outcomes as sufficient indicators of success [17,18,19]. As highlighted in the Introduction, this adoption-centric framing can overlook an important issue documented in primary studies: citizens may use digital public services while still experiencing the interaction as unclear, inaccessible, or unsafe, or while relying on intermediaries to complete tasks. Consequently, the current evidence base has not been synthesized in a way that explains which conditions support independent, secure, and confidence-building interaction across different contexts.

To address this conceptual gap, this review uses ADI as an analytic lens for synthesis. ADI is used here to organize and interpret determinants reported in the literature so that findings speak to the quality of citizen interaction, not only whether citizens adopt or use digital services. ADI is therefore applied as a lens derived from existing studies, rather than as a new theory being proposed. In this sense, ADI is not treated as a substitute for adoption, satisfaction, trust, or service quality, but as a higher-level interpretive lens for understanding whether these conditions support direct, secure, inclusive, and independent citizen interaction.

Guided by this lens, the review addresses the following research question:

RQ1. Which determinants enable or hinder ADI with e-government, and how do contextual conditions shape their effects?

2.2. Review Design and Protocol

To address RQ1, this systematic review was reported in accordance with the PRISMA 2020 statement for systematic reviews [21]. A review protocol was prepared prior to data collection to define the research question, databases, search strategy, eligibility criteria, screening procedures, data extraction fields, and synthesis approach. The protocol was not registered in a public repository. The protocol was developed for internal review planning and is not publicly accessible. No formal amendments to the protocol were recorded after the review commenced. Protocol adherence was maintained by applying the predefined criteria consistently across all screening stages, documenting reasons for exclusion during full-text review, and resolving screening disagreements through consensus among the reviewers.

2.3. Review Scope and Timeframe

The review examined peer-reviewed studies published between January 2020 and October 2025 that investigated determinants shaping citizen interaction with e-government and digital government services in ways relevant to ADI. Database searches were conducted between 4 November 2025 and 10 November 2025 and were limited to English-language journal articles and review papers. Figure 1 presents the PRISMA 2020 flow diagram [21], including the record counts and exclusion reasons at each stage. A total of 178 records were identified through the overall search process; after the removal of 8 duplicates, 170 records were screened.

2.4. Information Sources

Electronic searches were conducted in five databases commonly used in e-government, information systems, computer science, and public administration research: Scopus, Web of Science, IEEE Xplore, ACM Digital Library, and ScienceDirect. These sources were selected to capture interdisciplinary evidence across digital government services, citizen interaction and use, and socio-technical determinants of engagement. All databases were searched between 4 November 2025 and 10 November 2025.

2.5. Search Strategy

The search strategy followed a PICO-informed logic to structure key concepts [22]. The population was citizens or public service users. The phenomenon was interaction with e-government or digital government services, including portals, mobile applications, and smart government platforms. The context was public sector service delivery, including rapidly transforming environments such as GCC and developing countries. The outcomes were determinants relevant to the quality of citizen interaction, including constructs frequently examined in digital government research such as usefulness, ease of use, trust, security and privacy, digital literacy, interactivity, transparency, and participation.

Search strings were implemented using Boolean operators (AND, OR) and phrase searching with quotation marks where supported. Database-specific syntax was used to apply filters for publication years (2020–2025), English language, and document type (article or review). Search strings were refined through initial trial runs to balance sensitivity and specificity and then applied consistently across databases. Full search strings for each database are reported below in Appendix A.

2.6. Eligibility Criteria

Studies were included if they:

• Examined citizen-facing interaction with e-government or digital government services (web or mobile);
• Investigated at least one determinant relevant to ADI (for example usefulness, ease of use, trust, security and privacy, usability, service or information quality, digital literacy, accessibility, interactivity, transparency, participation, cultural or demographic influences, value co-creation);
• Were peer-reviewed journal articles or review papers;
• Were published in English between January 2020 and October 2025;
• Had full text available.

Studies were excluded if they focused on non-government or private-sector ICT contexts, were purely technical back-end papers without citizen interaction, were grey literature (for example theses, reports, or news), were non-English publications, were conceptual papers without empirical evidence, or did not address determinants related to the quality of citizen interaction. Included studies were subsequently grouped for synthesis according to the determinant categories identified during data extraction and consolidated during the descriptive mapping stage.

2.7. Study Selection and Screening

All records retrieved from the search process were compiled, and duplicates were removed prior to screening. Screening was conducted independently by three reviewers using the predefined inclusion and exclusion criteria. In the first stage, titles and abstracts were assessed and coded as included, excluded, or uncertain. All records coded as included or uncertain were retrieved for full-text assessment. In the second stage, full texts were assessed using the same eligibility criteria, and reasons for exclusion were recorded. Disagreements at either stage were resolved through discussion until consensus was reached. No automation tools were used in the screening process. The final set of included studies (n = 43) is shown in Figure 1.

2.8. Data Extraction

For each included study, the following variables were extracted: author(s), title, country or region, key factors studied, theoretical framework used, methodology or design, and sample or participants. Data extraction was conducted using a predefined extraction form. One reviewer performed the initial extraction, and the extracted data were checked by the other reviewers for accuracy and consistency. Any discrepancies were resolved through discussion and consensus. No automation tools were used in the data extraction process. The main outcome domains sought from the included studies were reported relationships between determinants and citizen interaction outcomes, particularly intention to use, actual use, satisfaction, and continuance, together with contextual conditions affecting these relationships. Additional study characteristics collected included publication details, study setting, theoretical framing, methodological design, and participant or sample information. Where reporting was unclear, extraction was limited to the information explicitly stated in the source article, and no assumptions were made beyond the published text. These fields were used to describe the characteristics of the included studies and to support synthesis of determinants relevant to ADI.

2.9. Quality Considerations

No formal risk-of-bias or quality appraisal tool was applied to the included studies, and no formal assessment of reporting bias (e.g., publication bias or selective reporting across studies) was undertaken. No formal certainty assessment framework was applied to evaluate confidence in the body of evidence for specific outcomes. This decision was taken because the review objective was to map determinants and synthesize findings across heterogeneous study designs, contexts, and measures rather than to estimate pooled effect sizes. During synthesis, study limitations reported by the original authors were noted, and conclusions are presented cautiously where evidence was limited, inconsistent, or context-dependent.

2.10. Data Synthesis

Synthesis followed a two-stage approach consisting of systematic descriptive mapping and an ADI-guided narrative synthesis. All studies meeting the eligibility criteria were included in the descriptive mapping stage and were then organized into aggregated determinant categories for narrative synthesis according to the factors reported in each article. This synthesis approach was selected because the included studies were heterogeneous in design, measures, and contexts, making a pooled quantitative synthesis inappropriate while still allowing structured comparison through descriptive mapping and narrative interpretation. First, determinants reported in the “key factors studied” field were standardized and consolidated by merging conceptually similar terms into aggregated factor categories. A mapping list was maintained to ensure consistent assignment of study-level terms to aggregated categories. Category assignment and consolidation decisions were reviewed by the research team, and any uncertainties were resolved through discussion and consensus to support consistency in the synthesis process. Frequencies were calculated as the number of included studies examining each aggregated category and were reported as counts and percentages. Because the review did not aim to produce pooled quantitative estimates, results were presented using descriptive summary measures (counts and percentages) and narrative reporting of the direction and contextual conditions of reported associations. Variation across study findings was explored qualitatively by comparing reported associations across different national contexts, study designs, participant groups, and service environments, rather than through formal statistical heterogeneity testing. Study characteristics were tabulated in

Table 1. Characteristics of the included studies.

No.	Author(s)	Title of Study	Country/Region	Key Factors Studied	Theoretical Framework Used	Methodology/Design	Sample/Participants
1	[23]	Factors Affecting E-government Adoption at the Micro Level of Government in Developing Countries: A Systematic Literature Review	Cross-country (Malaysia, Namibia, India, Africa, Tanzania, Peru, Indonesia, Nebraska, China, Pakistan, Afghanistan, Thailand)	-Social influence -Facilitating conditions -ICT skill/capability -Financial -Leadership -Trust -Self-efficacy -Compatibility -Ease of use -Usefulness -Awareness -Image -Trialability -Government support	Technology Acceptance Model (TAM); Unified Theory of Acceptance and Use of Technology (UTAUT); Theory of Diffusion of Innovation (DoI)	Systematic Literature Review (SLR)	Secondary data review (35 relevant papers from Science Direct, Scopus, IEEE Xplore, and ACM)
2	[8]	Challenges, Citizens’ Trust and Privacy Protection Models in e-Government Systems: Security and Privacy Perspective: Student paper	Cross-country (Switzerland, United Kingdom, Oman, India, China, Netherlands, Russian Federation, and others from Africa, South America, Asia, and Europe)	-Security -Privacy -Trust -Digital Divide -Inadequate Services -Limited Technology Access -Robust Security Protocols -Secure Communication -Data Protection Legislation	Barbara Kitchenham’s framework for systematic literature reviews	Systematic Literature Review	Secondary data review
3	[24]	A Review E-Government Implementation: Its Impact on Public Value Creation and Citizen Perspectives	Cross-country (China, India, Nigeria, Pakistan, and others)	-Information Quality -System Quality -Trust in e-government -Behavioral Intention to Use -Citizen Satisfaction -Public Value -Effort Expectancy -Ease of Use -Facilitating Condition -Performance Expectancy -Social Influence -Perceived Usefulness -Transparency -Openness -Responsiveness -Trust of Internet/Social Media/Technology -Citizen Engagement -Performance -Effectiveness -Acceptance -Adoption of e-Government	Unified Theory of Acceptance and Use of Technology (UTAUT); DeLone & McLean Model; SAP-LAP Framework; Conceptual models inspired by UTAUT, TAM, IS Success Model, UMEGA, Public Value theory, and eGOVQUAL	Systematic Literature Review (using PRISMA method)	Secondary data review (systematic literature review of 69 papers)
4	[25]	E-government Adoption: the Role of Perception of Digital Technology in the Public Service of Ghana	Ghana	-Perceived Usefulness -Perceived Ease of Use -Language as a Medium of Communication -Belief in Job Performance Gains -Commitment and Skills of Workmates -Availability of Digital Tools -Willingness to Go Paperless -Availability of Skilled Human Resources -Lack of Clear Plans for Digitization -Lack of Enjoyment in Using Digital Technology -Flexible Access to Information and Services	Technology Acceptance Model (TAM)	Quantitative (online questionnaire-based survey analysis)	61 public servants (Ghana)
5	[26]	Validation of a Measuring Model as a Key Aspect in the Adoption of E-Government in Developing Countries	Iraq	-Political Factors -Economic Factors -Social Factors -Technological Factors -Expected Performance -Expected Effort -Social Influence -Facilitating Conditions -Behavioral Intention	Unified Theory of Acceptance and Use of Technology (UTAUT) model integrated with the PEST framework	Quantitative (survey-based SEM analysis)	560 Iraqi citizens
6	[27]	Investigating the impact of citizen relationship quality and the moderating effects of citizen involvement on E-government adoption	Vietnam	-Relational bonds (financial, social, and structural) -Relationship quality (trust and satisfaction) -Citizen involvement (enduring and situational) -SocioCitizenry theory -Traditional models (TAM, TRA, UTAUT)	SocioCitizenry theory; Technology Acceptance Model (TAM); Theory of Reasoned Action (TRA); Unified Theory of Acceptance and Use of Technology (UTAUT)	Quantitative (survey-based SEM analysis)	595 citizens (Tien Giang Province, Vietnam)
7	[4]	E-government performance in democracies versus autocracies	Egypt	Citizen Participation, Transparency, Efficiency, International Competitiveness, Strategic Management, Policy Objectives	Not mentioned (the paper does not specify a particular theoretical framework or model like TAM, UTAUT, or IS Success Model)	Mixed-method (quantitative analysis using statistical databases and Jupyter software; qualitative analysis using expert interviews)	4 expert interviewees (professors of political science and economics at Cairo University)
8	[2]	E-government impact on developing smart cities initiative in Saudi Arabia: Opportunities & challenges	Saudi Arabia	Trust, Digital Infrastructure, Funding, Confidence, Accessibility, Usability, Participation, Autonomy, Service Quality	Not mentioned (the paper does not specify any theoretical framework or model like TAM, UTAUT, or IS Success Model)	Qualitative (meta-synthesis technique)	Secondary data review
9	[28]	Does e-government help shape citizens’ engagement during the COVID-19 crisis? A study of mediational effects of how citizens perceive the government	China	Trust in Government, Government Transparency, Government Reputation	Not mentioned (the paper does not specify a particular theoretical framework or model)	Quantitative (survey-based SEM analysis)	866 Chinese citizens (Hefei, Shanghai, and Nanjing)
10	[29]	Citizen engagement in co-creation of e-government services: a process theory view from a meta-synthesis approach	Cross-country (United Kingdom, USA, Estonia, Burkina Faso)	-Government’s Role in Fostering Participation -Bottom-Up Approach -Organizational Readiness (Strategies, Leadership, Human Resources, Communication, Collaboration, Policy Formulation, Financial Sustainability, Change Management) -Social Readiness (Government Officials’ Commitment, Societal Awareness, Civic Capacity) -Environmental Readiness (Legal Systems and Regulations) -Stages of Co-creation (Initiation, Open Participation, Open Collaboration, Engagement)	Process Theory View of Enabling Co-creation of E-Government Services; Open Innovation	Qualitative (meta-synthesis of qualitative case studies)	Secondary data review (meta-synthesis of 14 qualitative studies)
11	[5]	Comparative analysis of E-government development status of ASEAN member states: Accomplishments and challenges	Brunei; Cambodia; Indonesia; Laos; Malaysia; Myanmar; Philippines; Singapore; Thailand; Vietnam	-Inadequate internet access -Low digital literacy -Digital divides -Public distrust in digital platforms -Digital infrastructure development -Cybersecurity and data protection laws -Transparency -Participation (EPI) -Digital literacy enhancement -Public trust	Neoinstitutional Theory; Diffusion of Innovation Theory	Mixed-method (qualitative literature review; quantitative analysis using descriptive statistics, Pearson correlation analysis, and cluster analysis)	Secondary data review (United Nations E-government Survey data from 2014 to 2024)
12	[30]	Drivers of citizens E-loyalty in E-government services: E-service quality mediated by E-trust based on moderation role by system anxiety	Jordan	-E-service quality (e-SQ) -E-trust -System anxiety -Transparency -Accountability -Data security -Personalized services -Citizen engagement -Continuous feedback evaluation -User-centric design principles	Not mentioned (the study does not specify a particular theoretical framework like TAM or UTAUT, but rather explores the interconnections between e-SQ, e-trust, e-loyalty, and system anxiety in the context of e-government services)	Quantitative (survey-based SEM analysis)	532 Jordanian citizens using e-government services; 47.4% male, 52.6% female; majority under 24 years old; predominantly high educational background (63.3% with bachelor’s degree or higher)
13	[31]	An Analytical Approach to Evaluating the Usability of E-Government Websites in Bangladesh	Bangladesh	Usability, User Satisfaction, Digital Literacy, Language Flexibility, Symbolic Layout, Operational Status	ISO9241 standards; Usability Heuristics	Mixed-method (systematic literature review and heuristic evaluation)	Secondary data review (190 government websites in Bangladesh)
14	[19]	E-government Service Adoption by Citizens: A Literature Review and a High-level Model of Influential Factors	Not mentioned (the paper does not specify a country or region, and the context is broad and international)	-Citizen’s internal factors -Risk and security -Practicality -Sociodemographic factors -Social factors -Potential benefits -User adoption -Citizen’s trust -Government factors -Perceived ease of use -Perceived usefulness -Trust	The study proposes a new model specific to the e-government context, informed by existing theories such as TRA, TPB, TAM, and UTAUT, but does not primarily apply any of these models as its theoretical framework.	Systematic Literature Review	Secondary data review
15	[7]	E-government Mobile Web Accessibility Challenges facing People with Visual Impairments: A Multi-Method Evaluation	Kuwait	Usability, Accessibility, User Experience, Technical Barriers (e.g., alternative text, color contrast)	Not mentioned (the paper does not specify a theoretical framework like TAM, UTAUT, or IS Success Model)	Mixed-method (usability testing with visually impaired participants, automatic accessibility evaluation tools, and expert evaluations)	12 participants with visual impairments and blindness (Kuwait)-recruited from the Kuwait Blind Association
16	[32]	Evaluation of the Usability and User Experience of Selected Philippine E-Government Service Websites	Philippines	Usability, Perceived Usefulness (PU), Perceived Ease of Use (PEOU), Behavioral Intention to Use (BI), Learnability, Efficiency, Memorability, Errors, Satisfaction	System Usability Scale (SUS); Technology Acceptance Model (TAM)	Quantitative (survey-based SUS and TAM analysis with multivariate correlation and linear regression)	30 citizens/users (Philippines)
17	[33]	Factors Influencing the Adoption of E-Government Services: A Study among University Students	Colombia	-Perceived Usefulness -Subjective Norm -Perceived Ease-of-Use -Cultural Factors -Usability -Data Privacy -Lack of Trust in Governments -Entrenched Mentalities -Inadequate Infrastructure -Digital Literacy Gaps -Resistance to Change	Technology Acceptance Model (TAM); Unified Theory of Acceptance and Use of Technology (UTAUT); Decomposed Theory of Planned Behavior (DTPB); C-TAM-TPB Model	Quantitative (survey-based SEM analysis)	403 university students (Medellín)
18	[34]	Factors influencing indirect adoption of e-Government services: a qualitative study	India	-Access convenience to intermediary -Intermediaries’ service charge -Risk-averse characteristics -Value-added services -Lack of resources -Lack of computer self-efficacy -Perceived difficulty-to-use -Lack of multilingual option -Perceived awareness -Perceived benefits -Compatibility -Trust of intermediary -Social influence	Grounded Theory Approach	Qualitative (grounded theory approach using semi-structured interviews)	47 e-Government users in India (including government servants, working professionals, businesspersons, academicians, unemployed, and retired people)
19	[35]	The Role of Media in the E-Government Adoption in Morocco: A Diffusion of Innovation and Technology Acceptance Model Perspective Using PLS-SEM	Morocco	Trust, Perceived Ease of Use, Satisfaction, Relative Advantage, Complexity, Observability, Compatibility, Media Influence, Digitalization	Technology Acceptance Model (TAM); Diffusion of Innovations (DOI) theory	Quantitative (survey-based SEM analysis)	311 residents of the Rabat-Salé-Kénitra region, Morocco; including civil service employees, students, and private sector employees
20	[36]	Unlocking e-government adoption: Exploring the role of perceived usefulness, ease of use, trust, and social media engagement in Vietnam	Vietnam	-Perceived Usefulness (PUF) -Perceived Ease of Use (PEU) -Trust in E-government (TEG) -Social Media Use -Citizen Satisfaction	Technology Acceptance Model (TAM)	Quantitative (survey-based SEM analysis)	529 citizens (southern Vietnam), primarily heads of households with experience in online public services and in management positions
21	[37]	Measuring user acceptance of e-government adoption in an Indonesian context: a study of the extended technology acceptance model	Indonesia	Trust, Perceived Risk, Attitude towards Use, Perceived Usefulness, Perceived Ease of Use, Computer Self-Efficacy, Subjective Norms, Behavioral Intention	Extended Technology Acceptance Model (TAM)	Quantitative (survey-based SEM analysis)	363 Surabaya residents (citizens/users of KLAMPID)
22	[17]	Citizen Adoption of E-Government Services: A Systematic Literature Review with Weight and Meta-Analysis	context: developing countries in Asia and Africa	-Perceived Trust -Perceived Quality -Performance Expectancy -Effort Expectancy -Social Influence -Self-Efficacy -Facilitating Conditions -Perceived Satisfaction (mediator) -Attitude (mediator) -Age (moderator) -Gender (moderator) -Education (moderator) -Experience with e-government services (moderator)	Unified Theory of Acceptance and Use of Technology (UTAUT); Technology Acceptance Model (TAM); UMEGA model; Government Adoption Model (GAM); Theory of Reasoned Action (TRA); Theory of Planned Behavior (TPB); Information Systems Success Model (DeLone & McLean)	Systematic Literature Review (quantitative meta-analysis)	Secondary data review of 43 quantitative research articles on e-government adoption in developing countries (Asia and Africa)
23	[38]	Factors influencing e-government adoption in indonesia: The importance of perceived risk	Indonesia	Perceived Usefulness, Perceived Ease of Use, Perceived Risk, Trust, Social Influence	Technology Acceptance Model (TAM) with additional factors: Perceived Risk, Social Influence, and Trust	Quantitative (survey-based SEM analysis)	472 students from several universities in Indonesia (Institut Teknologi Sepuluh Nopember, Universitas Indonesia, Universitas Airlangga, Universitas Hasanuddin, Universitas Diponegoro, and Institut Pertanian Bogor)
24	[39]	Determinants of continuous intention to use e-government services: an extension of technology continuance theory	Malaysia	Trust, Transparency, Habit, Perceived Ease of Use (PEOU), Perceived Usefulness (PU), Confirmation, Satisfaction, Attitude	Technology Continuance Theory (TCT), which integrates elements from Technology Acceptance Model (TAM), Expectation-Confirmation Model (ECM), and Cognitive Model (COGM), extended with trust, transparency, and habit constructs.	Quantitative (survey-based SEM analysis)	260 residents (Penang, Malaysia)
25	[13]	Does trust in e-government influence the performance of e-government? An integration of information system success model and public value theory	Nigeria	Trust, Information Quality, Service Quality, System Quality, Public Value, Citizen Participation, Usability, Accessibility	DeLone and Mclean IS Success Model; Public Value Theory	Quantitative (survey-based SEM analysis)	369 e-government users (Nigeria); primarily staff and students from six federal universities
26	[16]	Cultivating the digital citizen: trust, digital literacy and e-government adoption	Nigeria	Trust, Digital Literacy, Perceived Usefulness, Perceived Ease of Use, Reliability, Security, Transparency, Usability, Accessibility	Technology Acceptance Model (TAM)	Quantitative (survey-based SEM analysis)	876 Nigerian citizens who had used e-filing services for tax returns
27	[40]	Factors affecting e-government adoption in the UAE public sector organisations: the knowledge management perspective	United Arab Emirates (UAE)	-Performance Expectancy: Short-term job performance, Long-term job performance, Client impact -Facilitating Conditions: Leadership support, Employee training, Organizational preparedness	Unified Theory of Acceptance and Use of Technology (UTAUT); 10-factor model	Quantitative (survey-based SEM analysis)	172 government employees (Dubai and Sharjah, UAE)
28	[41]	The determinants of smart government systems adoption by public sector organizations in Saudi Arabia	Saudi Arabia	-Security concerns -ICT strategy -Managerial support -Incentives -Trust -Perception	Technology Acceptance Model (TAM); Unified Theory of Acceptance and Use of Technology (UTAUT); Technology, Organization, and Environment (TOE) framework	Quantitative (survey-based SEM analysis)	419 government employees (Saudi Arabia) from the IT division of the Ministry of Health, Ministry of Foreign Affairs, Ministry of Justice, and Ministry of Education
29	[18]	Key factors influencing the e-government adoption: a systematic literature review	Indonesia; Ireland; Saudi Arabia	-Trust in e-government (TEG) -Security perceptions -Non-technical factors (cultural, religious, social influences) -Performance expectancy -Effect expectancy -Perceived system quality (PSQ) -Perceived information quality (PIQ) -User satisfaction (US) -Perceived ease of use (PEOU) -Transparency -Previous experiences -Digital divide	Unified Theory of Acceptance and Use of Technology (UTAUT); Technology Acceptance Model (TAM); Diffusion of Innovation (DOI); Theory of Planned Behaviour (TPB); Social Cognitive Theory (SCT); Model of Personal Computer Utilisation (MPCU); Self-determination Theory (SDT); Uncertainty Reduction Theory (URT); Information Systems [IS] Continuance Model; Quality-value-loyalty Chain Model; Theory of Reasoned Action (TRA); Unified Model of Electronic Government Adoption (UMEGA)	Systematic Literature Review	Secondary data review
30	[42]	Investigating the multifaceted dynamics of cybersecurity practices and their impact on the quality of e-government services: evidence from the KSA	Saudi Arabia	-Organizational culture -Technology infrastructure -Adherence to standards and regulations -Employee training and awareness -Financial investment in cybersecurity	Not mentioned (the paper does not specify a particular theoretical framework or model)	Quantitative (survey-based SEM analysis)	285 employees (Saudi Electronic University) from a sample of 320 participants
31	[6]	Evaluating the Usability and the Accessibility of Saudi E-Government Websites	Saudi Arabia	Usability, Accessibility (WCAG 2.0 principles: perceivable, operable, understandable, robust), Consistency and standards, Visibility of system status, Error prevention and recovery	Nilsson’s 10 heuristics; Web Content Accessibility Guidelines (WCAG 2.0)	Mixed-method (expert reviews for usability, automated tools and manual evaluation for accessibility)	Secondary data review (evaluation of five e-government sub-websites)
32	[15]	Do you see my effort? An investigation of the relationship between e-government service quality and trust in government	Kuwait	-E-government Service Quality -Satisfaction -Perceived Government Effort -Trust in Government	SERVQUAL model; e-SERVQUAL model; Satisfaction framework	Quantitative (survey-based mediated regression analysis)	723 citizens (Kuwait)
33	[1]	A periodical analysis of e-government maturity in Saudi Arabia	Saudi Arabia	-Service Quality -Accessibility -Financial Investment -Institutional Governance -Technical Infrastructure (TII) -Institutional Coordination -Usability	Five-stage model	Quantitative (secondary data analysis using the five-stage maturity model)	Secondary data review of 22 government websites in Saudi Arabia
34	[14]	Technology Acceptance Factors for Implementing the E-Government Systems in Saudi Arabia	Saudi Arabia	-Relative Advantage -Compatibility -Security -Management Support -Performance Expectancy -Perceived Usefulness -Ease of Use -IT Infrastructure -Language -Uncertainty -Financial Resources -Social Influence	Unified Theory of Acceptance and Use of Technology (UTAUT)	Quantitative (survey-based SEM analysis)	200 participants (employees and end-users of E-Government systems) with a 58% response rate
35	[43]	Arab cultural dimensions model for e-government services adoption in public sector organisations: an empirical examination	Saudi Arabia	-Power Distance -Uncertainty Avoidance -Collectivism vs. Individualism -Masculinity vs. Femininity -Nepotism -Face-to-face Interactions	Hofstede’s Cultural Dimensions Model; Nepotism; Face-to-face interactions	Quantitative (survey-based SEM analysis)	137 administrative employees (Saudi Arabia)
36	[20]	Examining the drivers and barriers to adoption of e-government services in Saudi Arabia	Saudi Arabia	-Cultural Factors -Digital Literacy -Government Policy and Interventions -Privacy and Security -Technical Infrastructure -Support Services -Citizen Trust -Citizen Motivation -Perceived Usefulness -Ease of Use -Perceived Risk	-Technology Acceptance Model (TAM) -Unified Theory of Acceptance and Use of Technology (UTAUT) -Institutional Theory -Hofstede Cultural Dimensions Theory	Mixed-method (quantitative survey-based SEM analysis)	487 respondents (Jeddah and Madina)-predominantly aged 21–30, nearly equal gender distribution, diverse education levels, well-distributed income, and various occupations including service, students, and business.
37	[3]	Modeling the Barriers Surrounding Digital Government Implementation: Revealing Prospect Opportunities in Saudi Arabia	Saudi Arabia	-Institutional habits -Political coordination -Ethical concerns -Perceived barriers related to law, organizational practice, finances -Risk aversion -Capacity and skills (project management) -Lack of engagement with and demand from users/citizens -Lack of awareness/strategic thinking -Legal framework issues -Technological resources (software and standards) -Technological infrastructure (computers and networks) -Difficulty articulating benefits to others -Political and management support and leadership	Interpretive Structural Modeling (ISM)	Qualitative (ISM approach with expert opinions collected via questionnaire)	24 DG experts (Saudi Arabia); leaders and decision-makers in Saudi government agencies responsible for digital transformation programs
38	[44]	Determinants of continuance intention of using e-government services in Tanzania: the role of system interactivity as moderating factor	Tanzania	System Interactivity, Computer Self-Efficacy, Management Support, Confirmation, Satisfaction, Perceived Usefulness	Expectancy Confirmation Model (ECM)	Quantitative (survey-based SEM analysis)	213 citizens (Tanzania) who were users of e-government services
39	[45]	Cross-country determinants of citizens’ e-government reuse intention: empirical evidence from Kuwait and Poland	Cross-country (Kuwait and Poland)	-System Quality -Service Quality -Information Quality -Perceived Value -User Satisfaction -Citizen Trust -Overall Risk -Time Risk -Privacy Risk -Psychological Risks -Uncertainty Avoidance -Masculinity-Femininity -Individualism-Collectivism -Cross-Cultural Trust and Risk	-Delone and McLean IS Success Model -Trust and Risk Models -Hofstede’s Cultural Model	Mixed-method (qualitative interviews and quantitative surveys with PLS regression analysis)	-Qualitative interviews: 81 Kuwaiti citizens -Quantitative surveys: 1582 Kuwaiti citizens; 355 Polish citizens
40	[46]	Assessing the factors influencing intention to use e-government in Tanzania: the perspective of trust, participation and transparency	Tanzania	Participation, Trust, Transparency	Technology Acceptance Model (TAM) extended with additional constructs: trust, transparency, and participation	Quantitative (survey-based SEM analysis)	153 respondents (Tanzania)-citizens with experience using e-government sites; 80 males, 119 females
41	[47]	Bridging the gap: assessing disparities in e-Government service offerings and citizen demand	Greece	Usability, Accessibility, Service Quality, Citizen Participation, System Quality	The study uses a task-oriented categorization framework and a people-centric approach, emphasizing resident engagement and participatory design. It does not explicitly mention specific theoretical models like TAM or UTAUT.	Mixed-method (quantitative survey-based analysis and qualitative content analysis)	707 residents from 49 Greek municipalities
42	[48]	E-government quality from the citizen’s perspective: the role of perceived factors, demographic variables and the digital divide	Greece	Perceived Attractiveness (PA), Perceived Usefulness (PU), Perceived Ease of Use (PEOU), Awareness (AWA), Demographic Variables (Age, Education, Economic Activity, Income), Digital Divide	Technology Acceptance Model (TAM); Cognitive Theory	Quantitative (survey-based SEM analysis)	707 respondents (Greek municipalities)
43	[49]	Older adults’ e-government use for bureaucratic and transactional purposes: the role of website-related perceptions and subjective digital skills	Israel	-Perceived Security -Perceived Clarity and Simplicity -Subjective Digital Skills	-Uncertainty Reduction Theory -Technology Acceptance Model (TAM) -Resource Appropriation Theory	Quantitative (survey-based binary logistic regression analysis)	735 Israeli older internet users (aged 60 years and older)

, and synthesis outputs were visually presented using frequency-based figures to display the distribution of study contexts, theoretical frameworks, methodologies, and aggregated determinant categories. No formal sensitivity analyses were conducted, as the review used a descriptive and narrative synthesis approach rather than pooled quantitative estimation. Second, an ADI-guided narrative synthesis summarized how these determinants were reported to relate to outcomes such as intention, use, satisfaction, and continuance, and it captured contextual conditions noted in the literature.

3. Results

The study selection process is shown in Figure 1. A total of 178 records were identified through the search process. After the removal of 8 duplicate records, 170 records were screened by title and abstract, of which 112 were excluded. Fifty-eight full-text reports were sought for retrieval and assessed for eligibility. Of these, 15 reports were excluded for the following reasons: no empirical evaluation (n = 4), conceptual only with no data (n = 2), technical/backend focus (n = 2), not in the e-government context (n = 2), pre-2020 frameworks or data (n = 1), private-sector platforms (n = 2), and incomplete or unavailable full text (n = 2). This resulted in 43 studies being included in the final review. The seven variables extracted from the selected studies are presented in Table 1.

No formal risk-of-bias assessment was conducted for the included studies; accordingly, no study-level risk-of-bias ratings are reported. No formal reporting bias assessment was undertaken; accordingly, no synthesis-level reporting bias findings are reported. No formal certainty assessment was conducted; accordingly, no outcome-level certainty ratings are reported.

3.1. Descriptive Characteristics of the Included Studies

The descriptive mapping provides an overview of the main characteristics of the 43 included studies. This stage of the synthesis summarizes the geographical distribution of the evidence, the theoretical frameworks used, the methodological approaches adopted, the participant groups examined, and the determinants most frequently studied. Together, these descriptive characteristics help contextualize the evidence base and show how current research on authentic digital interaction with e-government is distributed across settings, designs, and conceptual emphases.

3.1.1. Country and Regional Distribution

Figure 2 illustrates that the most frequently examined contexts are Saudi Arabia (20.93%), cross-country studies (16.28%), and Vietnam (16.28%). Conversely, Malaysia, the United Arab Emirates, and developing countries in Asia and Africa, each at 2.33%, are among the least represented. This geographic concentration influences the present review by limiting the breadth of contextual coverage underpinning the synthesis. Because ADI-related determinants particularly institutional trust, governance capacity, privacy/security assurance, and inclusion are sensitive to legal, infrastructural, and socio-cultural conditions, the strength and direction of reported associations should be interpreted as most robust for the most-studied settings. Accordingly, the identified determinant patterns may not transfer uniformly to under-represented developing contexts, reinforcing the need for further empirical evidence across a wider range of regions and service environments.

3.1.2. Theoretical Frameworks Used

In Figure 3, it shows TAM (20.43%) and UTAUT (13.98%) as the prevalent theoretical lenses, with a smaller group of studies not specifying a framework (6.45%). At the other end, Public Value Theory, Uncertainty Reduction Theory, and Institutional Theory (each 2.15%) are rarely applied, indicating limited engagement with broader governance and value-oriented perspectives. This distribution affects the present review in two ways. First, it indicates that much of the evidence base is structured around adoption-centric constructs (e.g., usefulness, ease of use, and behavioral intention), which strengthens synthesis confidence for value/effort pathways but provides thinner theory-driven explanation for ADI-relevant concepts such as empowerment, autonomy, co-creation, and institutional legitimacy. Second, because broader public value and institutional perspectives are under-represented, the review’s ADI lens is used to integrate findings beyond adoption and to highlight where the existing literature provides limited conceptual coverage particularly for citizen agency and governance-related interaction quality.

3.1.3. Research Methodologies and Study Design

Figure 4 indicates that the evidence base informing this review is dominated by quantitative designs (58.2%), with comparatively fewer qualitative (14.6%), mixed method (14.6%), and review studies (12.7%). This distribution influences the present synthesis in three ways. First, because most primary studies rely on cross-sectional surveys, the review can summarize the direction of associations between determinants (e.g., usefulness, trust, perceived risk) and outcomes (e.g., intention, use, continuance), but it cannot draw strong causal inferences about what produces Authentic Digital Interaction. Second, the limited volume of qualitative and mixed-method work means there is less detailed evidence on how and why citizens experience interactions as independent, secure, or confidence-building particularly for ADI-relevant concepts such as agency, empowerment, and reliance on intermediaries. Third, the dominance of survey-based measurement increases reliance on self-reported intention and perception constructs, which may under-capture interaction quality during real service journeys (e.g., task breakdown, accessibility barriers, or support-seeking behavior). Accordingly, this review emphasizes narrative synthesis, explicitly signals evidence strength, and frames future research priorities around longitudinal and mixed-method designs that can better explain ADI mechanisms across contexts.

3.1.4. Reported Determinants Across Included Studies

To identify determinants examined in relation to ADI, we re-analyzed the “Key Factors Studied” field across the 43 included papers (Table 1). Conceptually similar terms were merged into broader categories to support consistent reporting. For example, performance expectancy was merged with perceived usefulness, and closely related terms for trust were grouped under Trust and Confidence. Factors that could not be meaningfully aligned with any shared construct were grouped as “Other” and excluded from the frequency chart. This process resulted in 20 aggregated determinant categories, summarized in Figure 5.

Figure 5 shows that Perceived Usefulness and Performance Expectancy, Policy and Governance, and Trust and Confidence are the most frequently examined determinants in the included literature. In contrast, Awareness and Promotion, Value Co-Creation, and Autonomy and Empowerment are the least examined. This distribution indicates that the evidence base is strongest for performance-related beliefs and institutional conditions, while determinants linked to citizen agency, participation, and empowerment remain comparatively under-investigated. This frequency profile shapes how this review synthesizes evidence, because determinants in the top tier support stronger and more confident synthesis, whereas low-frequency determinants are interpreted cautiously as under-tested and are positioned as research gaps.

3.2. Determinant Evidence Summary by Higher-Order Groupings

To improve clarity and reduce repetition, the 20 determinant categories in Figure 5 were organized into six higher-order groupings. Each grouping consolidates closely related determinants that commonly co-occur in the included studies and that represent complementary aspects of citizen-facing digital government interaction. Grouping 3.2.1 covers perceived usefulness and performance expectancy, perceived ease of use and effort expectancy, intention pathways, and interactivity where it reduces effort during tasks. Grouping 3.2.2 covers trust and confidence, security and privacy, perceived risk and uncertainty, and transparency and accountability where it functions as assurance. Grouping 3.2.3 covers usability, accessibility and UX, service quality, system and information quality, and satisfaction as an experience outcome. Grouping 3.2.4 covers digital literacy and skills, demographic and personal factors, and awareness and promotion as enabling conditions for citizen capability. Grouping 3.2.5 covers policy and governance and technology infrastructure, with transparency and accountability noted where it reflects institutional capacity. Grouping 3.2.6 covers citizen participation and engagement, value co-creation, and autonomy and empowerment.

In the summaries below, “frequently examined”, “moderately examined”, and “sparsely examined” refer to the relative frequency tiers shown in Figure 5. Statements about “positive” or “negative” effects refer to the dominant direction of reported associations across the included studies, rather than causal claims. Where secondary reviews or meta-analyses are included among the 43 papers, their findings are reported as corroborating evidence and are not treated as additional counts of primary-study effects.

3.2.1. Citizen-Perceived Value and Effort

This grouping covers perceived usefulness and performance expectancy, perceived ease of use and effort expectancy, attitude and intention to use, and system interactivity where it supports task completion. Across the included studies, performance-related beliefs and perceived effort are among the most frequently examined determinants, and they are commonly reported to relate positively to intention to use, actual use, and continuance. When citizens expect clear performance benefits such as faster completion, fewer errors, and more reliable outcomes, studies report higher intention and higher use of e-government services [36,37]. When interaction is perceived as low-effort, intention and continued use are also commonly higher, especially when processes are simple and guidance is clear [39,44]. Interactivity that provides timely feedback or confirmations is often reported as reducing uncertainty during tasks and supporting continued use [44].

Included secondary syntheses report similar patterns, suggesting that performance-related beliefs and perceived effort are among the most stable predictors across contexts [17]. However, several studies also indicate that usefulness and ease of use may not translate into sustained engagement when citizens remain uncertain about outcomes, legitimacy, or data handling.

3.2.2. Institutional Assurance and Risk Management

This grouping covers trust and confidence, security and privacy, perceived risk and uncertainty, and transparency and accountability where these provide assurance through clarity of process and outcomes. Trust-related determinants were among the most frequently examined and were commonly associated positively with intention to use and continuance, whereas perceived risk and uncertainty were commonly reported as discouraging factors [36,46]. Studies often report that trust strengthens when services demonstrate reliability in visible ways, including predictable outcomes, clear process status, and responsive support [13,15]. Security and privacy safeguards are frequently reported as trust-supporting conditions, particularly when protections are visible and aligned with citizen concerns about personal data [8,42]. Conversely, perceived risk and uncertainty are reported as discouraging factors, with studies commonly finding that higher perceived risk reduces intention and use even when services are viewed as useful or easy to use [38]. Where reviewed syntheses are included, they also align with the overall pattern of positive associations for trust-related constructs and negative associations for risk [17].

3.2.3. Service Experience and Interaction Design

This grouping includes Usability, Accessibility and UX, Service Quality, System and Information Quality, and Satisfaction as an experience outcome. These determinants appear in the middle-to-upper tier of Figure 5 and are frequently reported as shaping citizens’ ability to complete tasks and their willingness to return. Studies commonly report that clearer navigation, consistent layouts, and understandable content relate to improved usage outcomes and higher satisfaction [6,31]. Evidence focused on accessibility reports recurring barriers for people with disabilities and older adults, and indicates that accessibility improvements support independent task completion [7,49]. Service quality and system or information quality are frequently reported as enabling conditions that shape whether citizens experience services as reliable, predictable, and worth using again [15,48]. Across studies, these determinants often appear alongside trust and satisfaction, indicating that interaction design and service delivery quality are closely connected in practice.

3.2.4. Capability, Inclusion, and User Characteristics

This grouping includes Digital Literacy and Skills, Demographics and Personal Factors, and Awareness and Promotion as enabling conditions for citizens to locate, understand, and complete e-government tasks. Studies generally report that higher digital literacy relates to stronger intention and smoother completion, partly by reducing perceived difficulty and uncertainty [16]. This relationship is especially evident in studies focused on older adults and users with lower digital confidence, where clearer structure and simpler flows reduce skill demands and support independent use [49]. Awareness and promotion are less frequently examined than performance or trust determinants, but the included studies commonly report that clear communication and guidance support first-time use and reduce confusion about where to start or what to prepare [33,36].

Demographic effects are more context dependent. Several studies report that age, education, and experience relate to adoption and continuation patterns and may reflect digital divide conditions that influence who can benefit from digital government and how easily they can do so [33,48]. In the included evidence, capability-related determinants often interact with design and support conditions rather than operating as standalone explanations.

3.2.5. Governance and Service Capacity

This grouping includes Policy and Governance and Technology Infrastructure, and it notes Transparency and Accountability where it reflects institutional capacity rather than citizen assurance. Policy and governance is among the most frequently examined determinants and is commonly reported as shaping service consistency, coordination, and accountability across agencies, particularly where multiple departments contribute to end-to-end completion [3,5]. Where roles are unclear or processes are fragmented, studies report that citizens experience confusion, delays, or rework, which reduces confidence in digital channels [2]. Technology infrastructure is reported as a foundational enabler in contexts where connectivity, integration, and platform reliability vary. Where infrastructure is limited, studies report more interruptions, incomplete completion paths, and higher reliance on offline channels [25,47]. Overall, the findings indicate that governance and infrastructure conditions shape whether citizens can complete tasks independently and reliably, especially in rapidly transforming or resource-constrained settings.

3.2.6. Participation, Co-Creation, Autonomy, and Empowerment

This grouping includes Citizen Participation and Engagement, Value Co-Creation, and Autonomy and Empowerment. These determinants are the least frequently examined in Figure 5. Where they are studied, papers commonly report positive associations between meaningful feedback mechanisms and trust, perceived fairness, and perceived alignment between services and citizen needs [27,29]. Some studies and included reviews also connect participation-oriented determinants to public value and satisfaction outcomes, which may support continued digital use [24]. However, because these determinants appear in relatively few studies, the evidence base remains thinner and more context-specific than the evidence for performance beliefs, trust, and governance-related determinants. Accordingly, these areas are best interpreted as emerging priorities in the literature and provide a clear basis for future empirical work on citizen agency, legitimacy, and interaction quality in digital government.

4. Discussion

This review synthesized evidence on determinants associated with ADI in e-government and organized findings into higher-order groupings spanning citizen-perceived value and effort, institutional assurance, service experience, citizen capability, and governance capacity. Across the included literature, three patterns are most consistently reported. First, perceived usefulness and performance expectancy are among the most frequently examined determinants and show largely consistent positive associations with intention, use, and continuance; especially in transactional services where time savings, error reduction, and outcome certainty are salient. These associations appear strongest when benefits are made visible during the task journey (e.g., eligibility guidance, time/fee expectations, progress visibility, and clear completion confirmations). Second, trust and confidence, supported by security and privacy assurances, frequently condition whether perceived benefits translate into actual behavior. Engagement is typically stronger where citizens can observe integrity cues, understand data-handling practices, and access clear pathways for assistance, complaints handling, or appeal when problems occur. Third, policy and governance capacity shapes the service environment within which other determinants operate. Evidence indicates that coordinated standards, stable service levels, interoperable back-end processes, and accessibility-by-default delivery policies help determine whether usefulness, assurance, and good service experience can be delivered reliably at scale.

Taken together, the findings support an interpretation of ADI as the outcome of interacting socio-technical conditions, rather than a single predictor chain. Service experience determinants (e.g., usability, accessibility, responsiveness, quality) reduce friction and enable citizens to realize expected value in real transactions. Capability and inclusion determinants (e.g., digital literacy, confidence, awareness) influence who can successfully navigate services without informal assistance. Contextual influences (e.g., social norms, cultural expectations, prior experience with government) shape baseline trust thresholds and perceived risk. Across multiple studies, satisfaction and perceived value are reported as post-use mechanisms associated with continuance, consistent with reinforcement dynamics in which successful completion strengthens expectations and reduces future effort. The evidence also suggests that purely informational portals tend to show weaker behavioral effects unless content supports a clear next step (e.g., initiating a service, completing a transaction, or resolving a specific user problem). In ADI terms, designing for action and completion appears more consequential than presentation alone.

• What ADI contributes beyond adoption and IS success models?

An ADI lens contributes something different from adoption-centric and IS success perspectives because it focuses not only on whether citizens use digital services, but on whether they can complete interactions directly, securely, clearly, and with confidence in both the process and the outcome. First, ADI helps explain why citizens may use e-government services without feeling confident, independent, or empowered—for example, where use is mandatory or alternatives are limited, but interaction remains opaque or dependent on intermediaries. Second, ADI foregrounds interaction qualities that intention and usage metrics can obscure, such as process clarity, visible safeguards, accessibility, and the credibility of completion signal features that determine whether citizens can reliably self-serve, not merely transact once. Third, ADI links micro-level user experience to macro-level governance and delivery capacity, showing how front-end improvements can be undermined by fragmented back-end processes, inconsistent service standards, or unclear accountability and remedy pathways. In this way, ADI reframes existing determinants as a configuration of conditions required for secure, self-directed citizen–state digital interaction.

• Strength of evidence and remaining gaps

The review also highlights uneven strength across determinant categories. The evidence base is strongest for citizen-perceived value and effort determinants (usefulness/performance expectancy; ease of use/effort expectancy) and for institutional assurance and risk determinants (trust/confidence; security/privacy; perceived risk/uncertainty). These constructs appear frequently and exhibit broadly consistent directions of association across contexts. Evidence is also substantial for service experience determinants (usability, accessibility, system/information quality, service quality, satisfaction), although reported effects vary by service type and user group. By contrast, determinants linked to citizen agency and interaction legitimacy—including participation, value co-creation, autonomy, and empowerment—are sparsely examined. Where they are studied, associations are generally positive, but the small number of studies limits confidence in stable cross-context conclusions. This imbalance suggests that current evidence most strongly supports adoption-adjacent beliefs and assurance conditions, while agency-related mechanisms remain under-tested empirically.

4.1. Implications for Practice and Policy

This paper shows that the use of a digital government service does not necessarily mean that the interaction itself is meaningful, citizen-driven, or authentic. Citizens may complete an online transaction yet remain unsure about what occurred, what the outcome means, or whether the process was completed correctly, and they may still rely on others for assistance. By focusing on the quality of interaction with digital services, this study moves beyond simple measures of service uptake and highlights the importance of direct, secure, transparent, and independent citizen interaction. This has important implications for research. Much of the e-government literature treats adoption, intention to use, or satisfaction as indicators of success. However, the findings of this review suggest that such measures may be misleading if they do not account for whether citizens can manage the interaction independently, understand the process, and feel confident about the outcome. An ADI perspective helps explain why high usage rates do not always lead to trust, confidence, or empowerment. Future research should therefore move beyond access and usability alone and examine the quality of interaction itself, particularly in relation to citizen autonomy, identity assurance, and process clarity.

From a practical perspective, the findings suggest that the effectiveness of digital services cannot be judged only by speed, convenience, or visual design. Citizens interact with greater confidence when systems are designed to support successful task completion on the first attempt and to reduce uncertainty throughout the transaction. This includes guided processes, fewer and simpler steps, clear instructions, error prevention, visible progress tracking, and explicit confirmation of outcomes. Usability and accessibility should be strengthened, especially for older adults and those with lower digital confidence, so that they can complete tasks independently and with less reliance on intermediaries.

At the governance and service-delivery level, the evidence indicates a need for stronger institutional assurance. Agencies should embed security-by-design practices, communicate clearly about how citizen data are used, and provide accessible complaint and appeal mechanisms. Service standards should incorporate accessibility requirements as core design principles rather than optional additions. Clear and enforced service-level targets can also strengthen predictability and trust. In addition, cross-agency coordination is essential when services span multiple departments, as fragmented processes can directly undermine citizen confidence and reduce perceptions of control.

Equity considerations are equally important. Without deliberate attention to digital literacy, language needs, disability access, and low-connectivity environments, digital services may primarily benefit those who are already digitally capable. To avoid this, digital-first approaches should be complemented by inclusive design and supportive pathways that enable, rather than replace, citizen agency. In this way, authentic digital interaction can become achievable across diverse population groups rather than only among highly digital users. Overall, the findings suggest that the success of digital government should not be evaluated solely by usage or satisfaction rates. Instead, it should be assessed by the extent to which it enables citizens to interact directly, understand what is happening, and have confidence in both the process and the outcome of their digital interactions.

4.2. Challenges and Future Directions

This section outlines two types of issues identified in the reviewed literature: (i) current challenges that limit how well ADI can be studied and compared across contexts, and (ii) future research directions required to strengthen evidence on authentic, inclusive, and independent citizen interaction with e-government.

This review also has several process-related limitations. The search was limited to English-language publications indexed in five databases and to studies published between January 2020 and October 2025, which may have excluded relevant evidence published in other languages, sources, or earlier periods. No formal risk-of-bias, reporting-bias, or certainty assessment framework was applied, and the synthesis was descriptive and narrative rather than meta-analytic because of heterogeneity in study designs, measures, and contexts. In addition, full-text exclusion reasons were retained at the category level rather than as a separate study-by-study appendix. These review-process decisions should be considered when interpreting the scope and confidence of the conclusions.

With these review-process limitations in mind, the main challenges identified in the reviewed literature are outlined below.

I.

Conceptual and measurement challenges: Analytically distinct constructs are sometimes operationalized in overlapping ways. Perceived usefulness or performance expectancy can drift toward service quality, while effort expectancy can be conflated with usability. Future work should use validated scales, report reliability and model fit, test for multicollinearity when co-modeled with quality constructs, and complement self-reports with behavioral indicators such as completion, abandonment, and reuse.

II.

Methodological challenges: Much of the evidence is cross-sectional and based on convenience samples, which limits causal inference and generalizability. There is limited reporting of effect sizes, moderator tests, and robustness checks. Longitudinal designs, natural experiments, and field trials are needed to trace intention into continued use and e-loyalty, and to isolate the effects of transparency, participation features, or accessibility improvements on behavior.

III.

Data and reporting challenges: Studies rarely report accessibility conformance or detailed user characteristics related to disability, language, or connectivity constraints. Participation and value co-creation are under-examined despite their conceptual relevance to authenticity. Future research should adopt reporting standards that include accessibility, assisted-channel use, device patterns, and socio-demographic moderators, and should elevate participation from rhetoric to measurable features in the service journey.

IV.

Context and equity challenges: Evidence is concentrated in specific regions and service types, while rural settings, low-resource contexts, and complex multi-agency transactions are less visible. Cultural norms and legal safeguards appear to moderate trust and value perceptions, yet are not consistently modeled. Comparative, multi-country studies and multi-level models that incorporate institutional variables would strengthen generalizability and external validity.

V.

Implementation challenges for agencies: Delivering ADI at scale requires balancing security and privacy with ease of use, modernizing legacy systems without disrupting essential services, coordinating across agencies, and sustaining accessible-by-default practices under tight budgets. Practical strategies include progressive enhancement for low-bandwidth users, mobile-first patterns that reduce data entry, clear consent and data-use notices at the point of capture, and transparent service level indicators that set expectations and build trust over time.

Research agenda for future work: Based on the gaps above, priority directions for future empirical research include:

• Rigorous tests of transparency, status tracking, and participation features on intention, use, and continuance.
• Longitudinal studies linking early experience, satisfaction, and perceived value to e-loyalty.
• Inclusion of accessibility and literacy measures as first-class variables.
• Mixed methods design that explains why determinants succeed or fail in particular contexts; and
• Development of a concise ADI measurement toolkit that pairs validated scales with behavioral telemetry for replication across services.

Together, these challenges suggest that the field is increasingly positioned to complement adoption research with stronger evidence on authentic, equitable, and sustained digital interaction, provided that conceptual clarity, methodological rigor, and inclusive implementation are prioritized.

5. Conclusions

Across 43 studies, ADI arises when citizens perceive clear performance benefits and can realize them through usable, accessible and responsive services within trustworthy, legally safeguarded and well-governed environments. Perceived usefulness and performance expectancy are consistent precursors of intention, use and continuance, but their effects depend on visible security and privacy practices, predictable service quality, inclusive design and genuine participation. However, evidence on participation, value co-creation, autonomy and empowerment remains comparatively limited in the included literature, so these elements should be interpreted as emerging priorities rather than well-established determinants. In practice, agencies should design for first-time task completion, make value visible up front, reduce friction in the flow, implement security by design, meet WCAG standards, ensure interoperability, and provide status tracking. The evidence base remains uneven, so future work should use longitudinal and experimental designs, including behavioral indicators, and treat accessibility, literacy and participation as primary variables, supported by a concise ADI measurement toolkit. Prioritizing these moves shifts e-government beyond adoption towards authentic, equitable and sustained digital interaction.