What Digital Inclusion Means for Older Adults
Digital inclusion for older adults is the condition in which people in later life have a meaningful, affordable, accessible, secure, and self-directed opportunity to use digital technologies for purposes they value. It includes the ability to obtain suitable connectivity and devices, understand and operate interfaces, judge the credibility and safety of digital information, obtain assistance, and decide when a digital service is worth using. Inclusion is ultimately assessed through effective participation and choice: being able to make a health appointment, communicate with relatives, access a public benefit, manage finances, learn, or enjoy a cultural activity without being excluded by the technology through which it is offered. Access alone provides only the beginning of this opportunity. The interdisciplinary literature on internet use in later life and the World Health Organization's understanding of functional ability support this environmental and person-centered perspective (Hunsaker and Hargittai, 2018; WHO, Healthy Ageing).
The defining psychological question is how technology interacts with an older person's goals, capacities, experience, resources, autonomy, and social relationships. Some adults have used computers for decades and navigate complex online systems with ease. Others enter digital services after retirement, use a smartphone only for specific activities, share a device, or avoid a service because its risks exceed its expected value. Age cannot establish which pattern applies. Research on technology adoption identifies variation in experience, capabilities, anxiety, beliefs, perceived usefulness, and access within every age group; it also shows that the design of the service influences the behavior attributed to the user (Czaja et al., 2006).
Digital exclusion occurs when connectivity, cost, design, institutional rules, inaccessible information, safety concerns, social arrangements, or lack of practical support prevent a person from using digital opportunities that matter to them. Voluntary nonuse deserves separate analysis. Someone who chooses a telephone or in-person conversation after considering alternatives is exercising preference; someone compelled to travel a long distance because an essential service accepts only an inaccessible app encounters exclusion. The boundary is determined by available, effective options and the distribution of consequences, rather than by whether a person owns the newest device. Contemporary systematic reviews describe technical, social, instructional, and psychological barriers that commonly interact (Xu et al., 2025; Wale et al., 2025).
The Digital Divide Is a Multidimensional Developmental and Social Process
Access, skills, outcomes, and the quality of participation
A useful account distinguishes physical access, usable access, digital skills, types of activity, and actual outcomes. Physical access concerns an internet connection or device. Usable access concerns screen visibility, reliable connectivity, an affordable data plan, working authentication, and appropriate assistive technologies. Digital skills encompass practical navigation, information and media literacy, privacy judgment, communication, troubleshooting, and learning how to transfer familiar skills to a redesigned interface. Activities may range from video calls and entertainment to telehealth, employment, banking, civic participation, and public benefits. Outcomes concern whether these activities produce the benefit the person sought, at an acceptable burden or risk. The same household may appear connected in a survey while one member lacks an independent device, a private setting for health consultations, or the confidence to complete a consequential transaction (Hunsaker and Hargittai, 2018).
The stages are related but cannot be collapsed into a single measure. A person may own a smartphone yet lack dependable cellular coverage. Another may navigate entertainment comfortably while feeling uncertain about an unfamiliar bank's identity verification. A third may possess excellent digital skills but reject an app that tracks location unnecessarily. Binary measures such as internet user versus nonuser are useful for monitoring broad trends, but they conceal frequency, autonomy, sophistication, vulnerability, and unequal return on participation. Reviews also warn that exposure to a new device during an intervention does not necessarily translate into sustained beneficial use after formal instruction ends (Xu et al., 2025).
The changing United States context
The American digital landscape illustrates why a fixed image of an offline older generation is misleading. Pew Research Center's 2026 reporting on its 2025 surveys estimates that roughly nine in ten U.S. adults aged 65 and older use the internet and that smartphone ownership has become widespread, while meaningful age-related differences in some usage measures remain (Pew Research Center, 2026). These estimates describe a population at a particular historical moment. They are not evidence that all older people use online services independently, that access is equally affordable, or that a cohort will retain the same capabilities under every new interface. Statistical comparisons also change as younger, more digitally experienced cohorts enter later life.
The precise denominator matters. Studies may define older adulthood as 60 and above, 65 and above, 70 and above, or 80 and above. Some recruit community-dwelling users, exclude long-term-care residents, or require existing internet access. National survey results cannot be generalized automatically to people in institutional settings, those experiencing housing insecurity, or countries with different infrastructure and prices. Online-only surveys are particularly likely to miss people with the greatest access barriers. Good measurement reports age bands, sampling frames, recruitment mode, and the differences between ownership, use, and effective outcomes.
An ecological account: person, task, technology, and environment
Digital inclusion emerges from interactions between the person, the task, the interface, and the support environment. The same task may be effortless on a familiar laptop but difficult on a small phone. A public-service website may function well with a keyboard yet become unusable when an authentication code expires before a user with limited dexterity can enter it. A resident with low vision may benefit from screen magnification, while another prefers voice input or a trusted helper. The World Health Organization emphasizes that functional ability reflects the interaction of an individual's capacities and the environment, including products and services (WHO, Healthy Ageing). The design implication is direct: improving the environment may expand participation without requiring a person to change their body, identity, or preferred pace.
The ecology includes household finances, accessible transportation to training, libraries, community organizations, family relationships, institutional rules, commercial practices, language, and public policy. Accessibility tools, secure human assistance, and non-digital service channels are therefore components of digital inclusion. A service that assumes universal broadband access can produce exclusion even when its visual interface meets technical accessibility standards. Conversely, an organization that supplies access, respectful instruction, and a staffed alternative can preserve autonomy across a broad range of personal circumstances.
Why Older Adults Adopt, Adapt, Avoid, or Abandon Technology
Perceived usefulness, usability, and the limits of adoption models
The Technology Acceptance Model, originally developed by Fred Davis, identifies perceived usefulness and perceived ease of use as important predictors of technology acceptance. These beliefs are relevant to later life, especially when a system has a clear practical purpose: a video call connecting distant family members or an online service avoiding a long trip may appear valuable. Yet usefulness and ease cannot explain all decisions. Costs, privacy, accessibility, past experiences, cultural expectations, institutional pressure, and whether a trusted alternative exists also matter (Davis, 1989; Czaja et al., 2006).
Evidence from health-technology research supports the relevance of usefulness, ease, and social influence while illustrating the limits of correlational models. A 2025 meta-analysis of technology acceptance among older adults found substantial associations with acceptance-related outcomes, but included studies varied in devices, settings, measurements, and research quality. Association does not establish that increasing a survey rating will cause lasting usage or improved health (Yang et al., 2025). Health-technology evidence is informative for mechanisms, although banking, creative applications, social media, and public services have different risks and motivations.
Adoption models are strongest when they are treated as maps of plausible influences rather than universal formulas. A person's refusal may reflect sound risk assessment or a conflict with personal values. Evaluating inclusion by the number of applications installed can mistake institutional compliance for well-being. The relevant outcome is whether the person has genuine agency over a service's use and receives a suitable benefit if they choose it.
Self-efficacy, confidence, and emotional learning history
Technology self-efficacy describes a person's belief that they can perform relevant digital tasks. Confidence is shaped by successful practice, clear feedback, previous failures, how much error costs, and the way helpers respond. A confusing security warning or repeated password failure can lead to uncertainty; an accessible explanation and a chance to retry may restore control. A learner who has experienced ridicule may anticipate embarrassment and avoid requesting help. Psychological mechanisms therefore operate within instructional and social relationships, rather than existing as immutable properties of a generation. The large multiage technology-adoption study by Czaja and colleagues documented the relevance of computer experience and attitudes alongside ability measures, while recent instructional reviews support paced, needs-based learning environments (Czaja et al., 2006; Xu et al., 2025).
Technology anxiety is task-specific. A person may be relaxed about a weather app and fearful about transferring savings. Uncertainty tends to rise when an action seems irreversible, instructions are opaque, or recovery is unclear. Repeated exposure can foster competence if tasks are meaningful and safe; repeated exposure to a defective interface can instead teach avoidance. Designers and instructors should distinguish learning difficulty from rational dissatisfaction with an unnecessarily difficult product.
Trust, perceived risk, and the right to skeptical judgment
Trust is often reduced to a single positive attitude, yet an older adult must evaluate several different objects: the device, the software provider, the information source, the institution, the security of a transaction, and any person offering help. Privacy concerns include unauthorized access, excessive data collection, profiling, identity theft, and disclosure of sensitive health or financial information. Risk also includes scams, misleading interfaces, recurring charges, mistaken appointments, and missed notifications. The appropriate objective is informed, calibrated trust: relying on a service for a suitable task when its safeguards are understood, and reserving the option to verify or decline when consequences are high.
A 2026 scoping review of older adults' resistance to digital health technology found barriers involving perceived functional value, risk, self-efficacy, technology anxiety, trust, preferences for relational care, and concerns about the legitimacy of digital substitutes. The review covered a specific health care context, drew substantially on Western settings, and contained limited longitudinal evidence; its findings should therefore inform rather than dictate explanations of ordinary technology use (Birati and Tzemah-Shahar, 2026).
Trust grows through honest information about what a system does, visible control over permissions, clear records, reliable correction procedures, and access to a human being who can resolve disputes. Trust falls when organizations compel adoption while transferring the consequences of errors to the user. A digital-service operator should regard a request for privacy information, a printed receipt, or a telephone alternative as an appropriate form of participation.
Social influence, independence, and identity
Family members, friends, colleagues, neighbors, and community instructors influence both the opportunity and motivation to adopt technology. A partner who uses video calls may offer a reason to learn; a family member who repeatedly takes the phone away and completes tasks can unintentionally prevent learning. A supportive helper demonstrates a task, allows the learner to make decisions, and gradually transfers responsibility. For some people, online communication protects connection across geographical distance; for others, it is an additional channel rather than a preferred replacement for face-to-face contact. Research on later-life internet use demonstrates this diversity of activities and preferences (Hunsaker and Hargittai, 2018).
Identity matters when technology has become a symbol of competence, independence, adulthood, or generational belonging. A person who believes that technological skill is 'for the young' may approach an interface with lower expectations. Conversely, portraying everyone in later life as an eager digital consumer can erase those who value different forms of participation. Digital inclusion respects the self-definition of older adults and avoids requiring technology to become a marker of personal worth.
Lifespan Development, Cohorts, and Individual Differences
Cognitive change and experience
Normal aging involves heterogeneous trajectories across cognitive domains. Prior experience and accumulated knowledge can support familiar tasks; unfamiliar interfaces may place greater demands on working memory, attention switching, processing speed, and the organization of multistep actions. These patterns interact with education, practice, sensory demands, anxiety, device design, and task complexity. Consequently, a difficult authentication procedure is not a cognitive assessment, and slow navigation does not diagnose impairment. Studies comparing adults of different ages cannot automatically identify the causal effect of aging because participants also differ in schooling, prior exposure, employment technology, socioeconomic history, and the technologies available during their lives (Czaja et al., 2006; Hunsaker and Hargittai, 2018).
Effective design can reduce unnecessary memory burden: show the current step, preserve progress, provide consistent navigation, keep crucial instructions visible, offer undo and recovery, and avoid forcing a user to remember several temporary codes across multiple screens. Assistive tools can support a user's goals regardless of the presence or absence of a medical diagnosis. Individuals with persistent new cognitive concerns may benefit from appropriate professional evaluation, which is conceptually separate from judging the quality of a digital experience.
Vision, hearing, motor function, and disability
Accessible technology recognizes that visual, auditory, motor, cognitive, and situational needs vary continuously across people and settings. Adjustable text size, sufficient contrast, scalable layouts, readable labels, keyboard access, visible focus indicators, captions, compatibility with assistive technologies, and generous touch targets improve usability. Tasks should allow sufficient time, and time limits should be extendable where feasible. The World Wide Web Consortium's guidance on older web users identifies substantial overlap between age-related functional needs and established web accessibility practices (W3C, Older Users and Web Accessibility; W3C, WCAG 2.2).
Accessibility should also account for fluctuating conditions. Fatigue, temporary injury, glare, noisy surroundings, stress, limited bandwidth, and an unfamiliar borrowed device can impede a task without indicating a stable disability. Organizations gain more reliable systems when accessibility is treated as normal quality engineering, evaluated with real users and assistive tools rather than as a specialized exception.
Age, cohort, and period effects
Cross-sectional research compares different age groups at one time; longitudinal research follows participants across time. Both contribute to understanding digital behavior, but each has limitations. Apparent differences associated with age may reflect birth cohort, educational opportunities, occupations, public infrastructure, or changes in device design. Longitudinal studies can examine change within persons but must address selective attrition, changing technology platforms, and the tendency of participants to improve at a repeated test. Research that follows users of a single app can become obsolete when the service is redesigned. The outcome must therefore be described precisely, and the evidence should not be converted into a claim of inevitable technological decline with age.
A generation whose work involved office software may have excellent desktop skills and limited experience with touch gestures; a retired engineer may understand system settings but avoid social media on privacy grounds. A person who did not attend school in the language of the application may depend on a family translator. Those differences are grounded in lived experience, institutional opportunity, and design choices. They are more useful for intervention planning than an assumption about chronological age.
Unequal resources and intersecting barriers
Digital disadvantage can accumulate when limited income, unstable housing, disability, rural infrastructure, language access, low formal education, migration histories, or social isolation overlap. Shared devices may prevent confidential consultations; expensive data plans may limit video calls; a public library may close before a person can attend; a government portal may demand documentation that is difficult to upload. Cultural context changes both the meaning of independence and the available forms of family support. The evidence base for older-adult digital interventions is geographically uneven, making local evaluation essential (Birati and Tzemah-Shahar, 2026; Wale et al., 2025).
A person with an intellectual or physical disability has the same entitlement to agency, accessibility, security, and participation. Accommodations should be selected with the individual, and access to a trusted representative must be governed by consent and relevant legal safeguards. The presence of assistance need not remove independence in decision-making; the important distinction is between chosen support and imposed control.
Digital Literacy: What Competence Actually Involves
Operational, informational, communicative, and safety skills
Operational skills involve turning on devices, connecting, navigating menus, using passwords or accessible alternatives, managing files, adjusting settings, and recovering from common errors. Information literacy involves searching, recognizing the source and date of material, distinguishing advertising from evidence, comparing claims, and deciding when a specialist is needed. Communication skills include managing contacts, messages, video calls, groups, and settings that determine who can see a post. Safety literacy concerns authentication, privacy permissions, phishing, suspicious links, fraudulent payment requests, and the consequences of sharing personal information. These abilities are related but distinct; proficiency in one should not be assumed to transfer automatically to another (Xu et al., 2025).
A practical training assessment asks someone to complete meaningful tasks under realistic conditions. Can the learner check the sender before opening a financial message? Can they locate a health service's authentic telephone number, understand when a digital form is submitted, and recover after an error? Can they explain why a requested permission is necessary? Such tasks reveal barriers more directly than whether a participant can repeat technical vocabulary. Safety competence includes the judgment to stop and request help when something is unclear.
Digital health literacy and clinical boundaries
Digital health literacy adds the ability to locate, understand, evaluate, and apply electronically delivered health information and services. A person may find a telehealth platform easy to enter but have difficulty locating the relevant clinical instructions, verifying a physician's identity, interpreting an automated result, or distinguishing general educational material from personalized care. Providers should therefore assess the whole service journey, including booking, consent, communication, follow-up, and alternatives. The literature indicates relevant uptake barriers but cannot establish that telehealth works equally well for all diagnoses, accessibility needs, and clinical circumstances (Yang et al., 2025; Birati and Tzemah-Shahar, 2026).
Digital literacy training is educational support, not a substitute for medical diagnosis or emergency care. Difficulty using an app should not be treated as a clinical sign of dementia; an online symptom checker cannot reliably settle a serious new health concern. When communication is clinically consequential, accessible written instructions, direct contact with a qualified professional, and a dependable human fallback are essential.
Transfer, updating, and the problem of constant redesign
Learning a particular interface is only part of digital literacy. Commercial applications frequently change icons, authentication procedures, payment terms, and menu structures. Instruction that teaches recognition of common patterns, safe experimentation, error recovery, and help-seeking can be more transferable than memorizing a single sequence of button presses. Yet constant redesign is itself an environmental burden, particularly when an essential service provides no stable alternative. Effective inclusion therefore requires organizations to preserve continuity where possible, communicate changes clearly, and offer ongoing assistance instead of assuming one-time training resolves future barriers.
The Psychological Effects of Digital Participation
Autonomy, self-direction, and daily life
A digital tool can increase practical independence by reducing the time and travel needed to obtain information, communicate, complete a transaction, or manage a household. However, independence in the practical performance of a task is only one dimension of autonomy. A family member may complete an online application for an older relative while excluding that person's preferences; the task is finished, but agency has been compromised. A more respectful arrangement makes the procedure understandable, seeks the person's decisions, and provides assistance only to the extent requested. This perspective aligns with the WHO emphasis on a person's ability to do and be what they value (WHO, Healthy Ageing).
Benefits must be assessed against the person's actual needs. An online banking application may be liberating for someone who previously relied on transport, yet an exclusively digital bank can diminish choice for people who cannot use the system. In both situations the service's design, error correction, and availability of alternatives determine whether technology expands capability.
Connection, loneliness, and social participation
Online groups, messaging, video calls, community platforms, and event calendars can maintain relationships across distance and create new paths to participation. They are especially valuable when transportation, mobility, or geography limit certain activities. At the same time, social connection is more complex than the number of messages exchanged or hours online. Meaningful relationships depend on reciprocity, perceived responsiveness, shared activities, and individual preferences. Training alone does not ensure these social outcomes. A review of interventions for loneliness among community-living older adults found heterogeneous effects and limitations in the certainty of evidence, supporting tailored support rather than universal claims that internet adoption reduces loneliness (Shekelle et al., 2024).
For a person who values in-person relationships, digital participation may complement, organize, or occasionally replace difficult travel. For someone who already uses online networks extensively, additional devices may bring limited benefits. Institutions should evaluate whether digital channels expand access to desired social activities, and retain community and face-to-face options. The specific psychology of perceived loneliness and objective social isolation in later life is examined in the related published research on loneliness and social isolation in older adults.
Well-being and the limits of causal claims
Observational studies sometimes associate digital engagement with life satisfaction, social connection, lower loneliness, or other indicators of well-being. The direction of influence is difficult to isolate: healthier, wealthier, more socially connected, or more motivated adults may be more likely to use technology in the first place. Benefits may also depend on the particular activity. A daily call with a close friend, reading distressing news late at night, and struggling with a mandatory online benefits portal have different psychological meanings. Interventions that increase an individual's confidence or test performance do not automatically demonstrate lasting improvements in mental health or quality of life (Hunsaker and Hargittai, 2018; Wale et al., 2025).
A responsible evaluation distinguishes proximal outcomes, such as completing a task or gaining confidence, from distal outcomes, such as durable social participation or health-related quality of life. Randomized comparisons, repeated follow-up, and valid outcome measures can strengthen causal inference, although real-world effectiveness still depends on the service environment. Claims that technology use prevents cognitive decline, cures loneliness, or ensures 'successful aging' require direct evidence from appropriately designed studies and should not follow from correlations alone.
Dignity, privacy, and meaningful consent
Digital services frequently ask users to surrender data or delegate actions. When a family member, caregiver, or institution controls passwords and devices, the arrangement can shift the balance between help and oversight. Privacy includes the capacity to communicate without unwanted observation, to understand who sees information, and to revoke permissions. The ethical quality of support can be assessed by whether the user understands the arrangement, can express preferences, and retains an accessible way to dispute mistakes. These concerns apply broadly to adult autonomy and become especially salient in care arrangements or essential services. They are not uniquely caused by older age.
Ageism in Digital Design, Research, and Instruction
Ageism shapes digital inclusion when designers assume older people will never learn, when instructors use infantilizing language, or when institutions treat the refusal of a difficult app as evidence of incapacity. It also operates when research excludes older participants from usability trials or uses a small convenience sample to represent an entire age group. The World Health Organization defines ageism through stereotypes, prejudice, and discrimination, including institutional and self-directed forms (WHO, Ageism). The published article Ageism Against Older Adults: Stereotypes, Discrimination, and Prevention examines those mechanisms across the broader social environment.
Accessible design and respectful teaching offer an alternative. Ask participants which activities matter to them, observe real interactions, invite critique, and compensate older users for design participation. Provide advanced options for experienced users and clear onboarding for new users without assigning these needs by age. A participant may want to use a complex financial tool, decline social media, or avoid a health app for legitimate reasons; each decision deserves evaluation in its own context. Digital inclusion can be measured by accessible opportunity and self-determined outcomes, not conformity to a preferred technological lifestyle.
What the Evidence Says About Digital Training and Support
Systematic reviews of education and inclusion programs
The strongest general conclusion is that targeted, practical, supportive training can improve selected digital skills, confidence, and task-related outcomes for some older adults. Xu and colleagues' 2025 systematic review synthesizes digital literacy education in lifelong learning, discussing barriers, instructional strategies, and reported learning outcomes. Successful approaches commonly involve relevant tasks, manageable pacing, repeated practice, and opportunities to ask questions. The studies differ in intervention design, recruitment, technologies, duration, and outcome assessment; the review does not establish one universally superior curriculum (Xu et al., 2025).
Wale and colleagues' 2025 rapid review of interventions addressing digital exclusion in older adults in social care identified a diverse evidence base but emphasized methodological limitations. Some programs combine equipment, connectivity, mentoring, and training, making it difficult to identify the independent causal contribution of a single element. Short follow-up, selection of volunteers, inconsistent measurement, and contextual differences restrict confident conclusions about durability and transfer. The most credible recommendation is to adapt a promising approach to local goals, measure outcomes, and revise the program in response to the users' experiences (Wale et al., 2025).
Evidence certainty should be named precisely. Established evidence supports the importance of access, usability, accessibility, support, and heterogeneous individual preferences. Emerging but methodologically mixed evidence suggests that practical training can increase specified competencies and use. Evidence for broad mental-health benefits, long-term autonomy gains across all settings, or a single best teaching method remains limited. Claims of inevitable digital exclusion in later life, or of technology as a universal cure for isolation, are contested by differences in samples, outcome definitions, and observed trajectories.
Design principles for effective instruction
A well-designed course starts with the learner's purpose rather than a predetermined app list. An instructor can ask whether the person wants to contact someone, navigate a health portal, locate trustworthy information, obtain an accessibility setting, or complete a government form. The task should be broken into meaningful steps with visible goals, demonstrations, guided practice, independent repetition, and an explicit recovery route. Instruction becomes more sustainable when the learner uses their own device, or a realistically similar one, and receives a printed or accessible digital guide they can consult after the class. The review of lifelong digital literacy programs supports personalization and practical, supported learning, while comparative evidence for precise instructional components is still evolving (Xu et al., 2025).
Teaching safety should be integrated with everyday tasks. Learners can practice identifying a trustworthy URL, questioning an unexpected payment demand, recognizing urgency tactics, checking a site's contact information, and requesting help without sharing passwords. Security recommendations must reflect current systems, be tested for accessibility, and avoid making authentic services impossible to use. A training program that builds confidence without addressing fraud, privacy, and account recovery leaves an important part of digital competence unfinished.
Peer learning, intergenerational support, and patient coaching
Intergenerational programs and peer-led instruction can supply social motivation, opportunities to practice, and a relationship in which questions feel welcome. Their success depends on the helper's behavior, language, technical competence, and responsiveness; the helper's age does not guarantee good instruction. An effective coach invites the learner to control the device, demonstrates only when useful, explains the reason for actions, and lets the learner repeat the task. A person who needs continuing assistance can still participate actively through shared decisions. Such learning arrangements fit the social and motivational dimensions identified in evidence reviews (Xu et al., 2025).
The practical choice between a group workshop, home visit, telephone support, library program, accessible online module, or one-to-one coaching depends on transport, privacy, disability accommodation, cost, and the task. Online-only training assumes the access it is intended to improve. Program designers should offer an initial non-digital enrollment method and ensure that support remains available when a device is lost, an account is locked, or an interface changes.
Access and service redesign are interventions too
A class cannot compensate for unaffordable broadband, a noncompliant website, or an essential service that refuses telephone assistance. Digital inclusion programs therefore combine person-level learning with environmental changes. Institutions can lend devices, offer affordable connectivity, supply accessible hardware, simplify consent forms, maintain human assistance, and test online processes with diverse users. Universal and inclusive design standards provide a more durable foundation than expecting everyone to learn around avoidable barriers (W3C, WCAG 2.2; WHO, Healthy Ageing).
Domain-Specific Applications and Risks
Health care and telehealth
Telehealth can make care accessible when travel is difficult or specialists are remote, but the entire pathway must be workable. Scheduling, identity checks, audio-video setup, finding a private space, communicating with the clinician, obtaining records, and following a care plan can each become obstacles. The 2026 scoping review of digital health resistance emphasizes perceived value, risks, trust, personal contact, and self-efficacy; adoption-focused meta-analysis identifies associations with usefulness and ease of use. These findings justify user-centered design and clinical fallback, but they cannot prove that a particular remote intervention is clinically equivalent to in-person care for every condition (Birati and Tzemah-Shahar, 2026; Yang et al., 2025).
Healthcare teams should offer an accessible preparation pathway, verify communication preferences, and accommodate assistive devices, interpreters, or a consented support person. Emergency care and serious new symptoms require established clinical procedures. The aim is access to suitable care, with technology one available pathway.
Banking, commerce, and public services
Online finance and public administration may reduce travel and waiting, yet these tasks combine sensitive personal data, irreversible actions, opaque terminology, and fraud exposure. A secure workflow should explain the transaction, disclose fees and recurring obligations, verify the recipient, and provide confirmation and recovery instructions in a readable format. Security must be usable: an account system that demands repeated, inaccessible authentication creates pressure to share credentials or abandon the service. Staffed support and alternative channels are legitimate safeguards against exclusion.
Financial vulnerability cannot be inferred from age. Scams exploit many groups through deception and urgency; risk depends on opportunity, exposure, safeguards, and context. In instruction, practice with fictional information and demonstration accounts can protect real assets while still teaching verification. A program should never require a participant to disclose banking passwords or recovery codes to an instructor.
Social media, information quality, and civic participation
Digital public spaces allow people to join organizations, access local announcements, volunteer, publish creative work, and participate in civic life. They also expose users to misinformation, persuasive advertising, scams, unwanted contact, and platform design that may not clearly distinguish paid promotion from ordinary communication. Media literacy should focus on examining provenance, evidence, dates, financial incentives, and corroboration, rather than imposing a simplistic rule that information online is uniformly dangerous. The meaningful outcome is competent, voluntary participation in the digital public sphere.
Housing, mobility, and everyday services
Navigation, shopping, transportation, and smart-home functions can extend options for people managing changing mobility, vision, time, or geographic constraints. These benefits depend on whether the devices work reliably and whether support is available after setup. Requiring an app for a bus ticket can exclude someone who otherwise travels independently; providing an accessible alternative can preserve that independence. Broader housing and environmental adaptation is addressed by the published Healthy Aging and Psychological Well-Being article, which centers functional ability rather than device adoption.
AI-specific systems and a distinct canonical research domain
Search engines, messaging, online forms, video calls, and ordinary digital-service design supply this article's main domain. AI-powered companions, generative chatbots, automated decision support, algorithmic monitoring, and delegated cognitive work introduce additional issues of artificial-system competence, transparency, attachment, algorithmic bias, and appropriate reliance. Their age-specific psychological questions are examined in the already published Aging in the Age of AI: Older Adults, Autonomy, Trust, Companionship, and Cognitive Support. Distinguishing these domains prevents evidence from ordinary digital training or telehealth from being improperly generalized to every AI system.
Designing and Evaluating Digital Inclusion Programs
Start with needs, choices, and a community assessment
A useful program begins by asking which activities people want to perform, what they currently do, what prevents access, and what existing resources could help. Assessment should include nonusers, users with limited tasks, skilled users, family caregivers, and those who have discontinued a service. Recruitment must reach beyond digital surveys: telephone calls, community organizations, accessible print materials, care services, and direct invitations may be necessary. Programs should invite older adults as collaborators in defining problems, interpreting results, and judging whether the outcomes were worthwhile.
Provide an integrated pathway
The pathway includes a suitable device, affordable connectivity, accessible accounts, understandable privacy settings, task-specific teaching, safe opportunities for practice, realistic support after instruction, and a human route for unresolved issues. Each component addresses a distinct mechanism. Equipment without support may go unused; training without reliable connectivity can lose its value; a secure application without readable error messages may produce account lockouts. Reviews of inclusion interventions underscore the importance of describing the intervention's components and setting before attributing changes to an educational method (Wale et al., 2025).
Measure meaningful outcomes and possible harms
Programs should assess task completion, accessibility, confidence, privacy knowledge, independence in decisions, sustained use when desired, satisfaction, service access, and participation in valued activities. Baseline measurement and follow-up make it possible to identify improvement or abandonment. Data should be disaggregated thoughtfully by relevant access needs, language, income, disability, and starting experience without turning these factors into stigmatizing labels. The program should record burdens too: costs, frustration, exposure to fraud, unwanted information sharing, reliance on helpers, and displacement of preferred face-to-face services.
A controlled study of digital learning ideally specifies an active comparison, validated measures, attrition, and the durability of effects after assistance stops. Qualitative interviews can reveal whether a participant is exercising choice or merely complying with an institutional requirement. It is methodologically inadequate to infer broad psychological transformation from a rise in the number of apps opened.
Preserve an effective non-digital route
A well-governed digital institution maintains proportionate, usable alternatives for essential services. These may include a staffed telephone line, face-to-face support, accessible paper processes, or an authorized representative who acts according to the person's decisions. A person who prefers non-digital participation can remain fully included when these alternatives deliver equivalent core rights and benefits. Technology should enlarge the number of viable pathways, not determine who qualifies for public life.
Evidence Strength, Uncertainties, and Research Priorities
Established findings include the heterogeneous distribution of digital access and experience, the importance of perceived value and practical usability, and the relevance of age-inclusive accessibility design. Numerous observational studies connect technology behavior with personal, relational, and environmental factors, and qualitative investigations document the consequences of distrust, anxiety, and institutional friction (Czaja et al., 2006; Hunsaker and Hargittai, 2018; Birati and Tzemah-Shahar, 2026). The exact effect of any one factor depends on setting, measurement, and research design.
Preliminary or limited evidence supports several kinds of digital literacy intervention, particularly for proximal skills and confidence. The evidence is less definitive about long-term persistence, generalization across rapidly changing services, or broad improvements in loneliness, cognitive health, or life satisfaction. Reviews note small or heterogeneous studies, incomplete controls, limited follow-up, and variation in outcomes (Xu et al., 2025; Wale et al., 2025). Promising findings should be paired with realistic claims and local quality improvement.
Several debates remain contested. Should inclusion be evaluated by frequency of internet use or by self-determined access to valued opportunities? How should institutions weigh privacy and verification against simplicity? When does family assistance protect independence, and when does it become an unwanted transfer of control? Which instructional features produce durable skill transfer rather than short-lived performance gains? The empirical answer will vary with tasks, infrastructure, social circumstances, and preferences. Research priorities include larger longitudinal studies, inclusive recruitment of people with disabilities and very old adults, non-Western and multilingual settings, more transparent intervention descriptions, economic evaluations, and measures centered on rights and meaningful participation.
Frequently Asked Questions
Does digital exclusion affect all older adults?
No. Digital practices vary greatly within older populations, and many older adults use smartphones, the internet, and sophisticated services routinely. Exclusion refers to obstacles to meaningful participation, not a universal attribute of age. In the United States, recent Pew data document widespread internet and smartphone use among adults aged 65 and above while also revealing continuing divides in patterns of participation (Pew Research Center, 2026).
What are the most important barriers to digital inclusion?
Commonly identified barriers include unaffordable or unreliable connectivity, inaccessible hardware and interfaces, unfamiliar procedures, limited practical support, privacy and fraud concerns, language barriers, and a poor match between the service and the user's priorities. They interact; solving only one may leave the person excluded (Wale et al., 2025; Birati and Tzemah-Shahar, 2026).
Is distrust of technology a sign of poor digital literacy?
Distrust can reflect unfamiliarity, but it can also be an informed response to privacy practices, past fraud, opaque errors, or an unwanted service. Good digital literacy includes assessing credible risks, verifying consequential information, choosing suitable safeguards, and declining a system when its benefits do not justify its costs.
Can an older adult learn to use a smartphone or new digital service?
Yes. Adults can acquire new skills throughout life, although the pace, learning environment, prior experience, and accessibility needs differ. Structured and relevant practice, accessible teaching, repeated opportunities for mastery, and ongoing assistance can improve defined competencies. A training program should assess the learner's own goals rather than impose a uniform proficiency benchmark (Xu et al., 2025).
Does teaching technology reduce loneliness?
Technology can facilitate communication, but the relationship between training and loneliness is not uniform. Intervention findings are heterogeneous and depend on whether the new technology actually supports valued relationships. It should complement accessible community and in-person opportunities, with outcomes evaluated rather than assumed (Shekelle et al., 2024).
How can families help without taking away independence?
Ask what the person wishes to achieve, demonstrate the relevant steps, let them control their own device, and respect privacy and consent. Provide reminders or a trusted recovery route when requested, and avoid collecting credentials or making decisions on the person's behalf. The goal is support that enlarges agency.
Are older adults' digital difficulties evidence of dementia?
No. Interface difficulty can result from unfamiliar design, low accessibility, stress, sensory barriers, cost, or limited practice. A digital task does not provide a clinical diagnosis. New, persistent concerns affecting everyday cognition can be assessed separately by qualified professionals using appropriate clinical methods.
Should essential public and health services require an app?
An app can be a useful option, but essential services should be reachable by people with different devices, abilities, languages, access arrangements, and preferences. A dependable human or non-digital alternative can prevent technology from becoming an exclusionary gatekeeper.
Is ordinary digital inclusion the same subject as AI companionship for older adults?
They overlap in access and autonomy but have different scientific questions. Digital inclusion concerns the conditions for effective use of everyday technologies and services. AI companionship, generative assistants, and automated cognitive support require additional evidence about system behavior, appropriate reliance, privacy, and psychological relationships. Their distinct aging-specific analysis belongs to Aging in the Age of AI.
Conclusion
Digital inclusion for older adults is a measurable condition of opportunity, capability, safety, and choice. The evidence directs attention to person–environment fit: meaningful tasks, affordable and accessible systems, appropriate trust, reliable teaching, responsive assistance, and rights-preserving alternatives. Later life contains diverse histories and ambitions, and digital services produce benefit when they adapt to that diversity. The most defensible objective is that every person can participate in valued activities through a suitable digital or non-digital pathway, with dignity and control over the decisions that shape everyday life.
Related Articles
Cluster Adult Development, Aging and Later Life — The central lifespan and aging research chapter situates digital participation within autonomy, functional ability, and age-related social context.
Ageism Against Older Adults: Stereotypes, Discrimination, and Prevention — Explains institutional and interpersonal ageism that may shape digital design and service provision.
Healthy Aging and Psychological Well-Being: Functional Ability and Adaptation — Examines how people and environments combine to support self-directed daily life.
Loneliness and Social Isolation in Older Adults: Differences, Risks, and Support — Distinguishes social participation from measured loneliness and isolation.
Aging in the Age of AI: Older Adults, Autonomy, Trust, Companionship, and Cognitive Support — Addresses specifically AI-mediated interactions, assistance, autonomy, and reliance.
Aging in Place: Autonomy, Home, and Psychological Well-Being
Independence and Decision-Making in Older Adults: Autonomy and Supported Choices
Moving to Assisted Living: Psychological Adjustment and Family Support
Age-Friendly Communities: Psychological Needs and Quality of Life
References
Birati, Y., and Tzemah-Shahar, R. (2026). Barriers to digital health adoption in older adults: Scoping review informed by innovation resistance theory. Journal of Medical Internet Research, 28, e75591. https://doi.org/10.2196/75591.
Czaja, S. J., Charness, N., Fisk, A. D., Hertzog, C., Nair, S. N., Rogers, W. A., and Sharit, J. (2006). Factors predicting the use of technology: Findings from the Center for Research and Education on Aging and Technology Enhancement (CREATE). Psychology and Aging, 21(2), 333–352. https://doi.org/10.1037/0882-7974.21.2.333.
Davis, F. D. (1989). Perceived usefulness, perceived ease of use, and user acceptance of information technology. MIS Quarterly, 13(3), 319–340. https://doi.org/10.2307/249008.
Hunsaker, A., and Hargittai, E. (2018). A review of Internet use among older adults. New Media and Society, 20(10), 3937–3954. https://doi.org/10.1177/1461444818787348.
Pew Research Center. (2026, January 8). Internet use, smartphone ownership, and digital divides in the U.S. https://www.pewresearch.org/short-reads/2026/01/08/internet-use-smartphone-ownership-digital-divides-in-u-s/.
Shekelle, P. G., et al. (2024). Interventions to reduce loneliness in community-living older adults: A systematic review and meta-analysis. Journal of General Internal Medicine. https://doi.org/10.1007/s11606-023-08517-5.
Wale, A., Everitt, J., Ayres, T., Okolie, C., Morgan, H. E., Shaw, H., Cooper, A., Edwards, A., and Lewis, R. (2025). Effectiveness of interventions for addressing digital exclusion in older adults in the social care domain: Rapid review. JMIR Aging, 8, e70377. https://doi.org/10.2196/70377.
World Health Organization. (n.d.). Ageism. https://www.who.int/health-topics/ageism.
World Health Organization. (n.d.). Healthy ageing and functional ability. https://www.who.int/news-room/questions-and-answers/item/healthy-ageing-and-functional-ability.
World Wide Web Consortium. (n.d.). Older users and web accessibility: Meeting the needs of ageing web users. https://www.w3.org/WAI/older-users/.
World Wide Web Consortium. (2023). Web Content Accessibility Guidelines (WCAG) 2.2. https://www.w3.org/TR/WCAG22/.
Xu, L., Ng, D. T. K., Lee, C. S. Y., Peng, P., and Chu, S. K. W. (2025). A systematic review of digital literacy in lifelong learning for older adults: Challenges, strategies, and learning outcomes. Educational Technology Research and Development, 73, 3627–3674. https://doi.org/10.1007/s11423-025-10530-w.
Yang, H. J., Lee, J.-H., and Lee, W. (2025). Factors influencing health care technology acceptance in older adults based on the Technology Acceptance Model and the Unified Theory of Acceptance and Use of Technology: Meta-analysis. Journal of Medical Internet Research, 27, e65269. https://doi.org/10.2196/65269.