Globally, remarkable progress has been made in medical interventions, health care and technological advancement, contributing to unprecedented decline in mortality rate and increase in life expectancy [1]. There are currently 703 million older adults (≥65 year), and this number is projected to double by 2050 [2]. Harnessing the numerous potentials of rapidly developing digital technology plays an important role in ensuring a better and more inclusive society, better health and social care, and economic support for the older population. However, recent surveys have indicated that a significant proportion of this age group has limited or no access to a range of digital technologies [3-6]. In addition, the diversity and quality of technology use are limited to fewer and familiar functionalities such as communication. For example, using a smartphone as a classic phone or for simply obtaining information [7].

Nowadays, an increasing number of older adults are digitally engaging and becoming competent technology users through improved accessibility features, user-centered and experience-based designs, and further education that equips older adults with essential digital skills. However, there is a long way to closing the digital divide between the ages, and the primary technological design ethos continues to be the supply side (digital developers’) presupposition that one size does fit all which fails to account for older adult’s physical and mental capability, accessibility needs, age-related changes, and lack of skill and support [8,9].

Recently, the SARS-CoV-2 pandemic has further increased the reliance on digital technology for everyday living, working from home, shopping, financial transactions, e-learning, communication, entertainment, and health service delivery (eg, remote consultation through e-consult and e-pharmacy) [10,11]. However, lack of access, awareness, and skills exacerbated existing digital inequality [12]. Beyond mere accessibility and use issues, older adults' digital experience constitutes pragmatic versus hedonic aspects, motivation based on functional, usability and aesthetic dimensions, and emotional ambivalence [13]. The perceived benefits of technologies in restoring autonomy, a sense of independence, improving the quality of life [13], decision-making [14], mobility, and social connectedness [15] constitute a positive experience. Intrusiveness, privacy and safety concerns, nonease of use, vulnerability, and social stigma can be sources of mixed feelings [13].

A scoping review that captures the nature and breadth of literature and older adults’ experiences and factors influencing their digital engagement is pertinent and timely, given the fast-paced nature of this discipline. A recent review to develop a system-level framework for health technology adoption and scale-up highlighted the importance of investigating nonadoption and sustainability, the shortage of studies in this area, and the role of barrier and facilitator research as an input for organizational-level adoption [16]. Similar indications have been made in a recent scoping review that summarized the definition and models of technological adoption, which underscored the importance of research that captures the entire spectrum of the digital technology acceptance cycle, including the continued use of technology with all its temporal aspects of engagement and the quality of technology users’ experiences over a long period [17,18]. This review will summarize studies that investigated older adults’ digital engagement, including nonuse, initial adoption, and sustained digital engagement and the driving factors (see Textbox 1 for key concept definitions).

We used the capability, opportunity, and motivation behavior change model (COM-B) and theoretical domain framework (TDF) models at the hub of the behavioral change wheel to facilitate the synthesis of the barriers to and facilitators of digital engagement among older adults. These frameworks are widely used to identify salient determinants of behavior and develop specific intervention recommendations, particularly in health system research, health care providers and service users’ behavior [21]. Furthermore, the application in synthesizing evidence generated using quantitative and qualitative methodologies has increased owing to robust, structured, and replicable nature of the models [21,22].

The COM-B and TDF models are organized into 14 constructs and 3 main components. The physical and psychological capability domain (skills, knowledge, memory, attention, and decision process), the automatic and reflective motivation domain referring to the intrinsic processes for behavior and decision-making based on whether conscious and unconscious cognitive processes that influence older adults’ behavior and decisions to engage digitally were included (beliefs about capabilities, optimism, consequences, intention, goals, reinforcement, and emotion), and opportunity domain (environmental context and social influence) [23]; see Figure 1 depicting COM-B and TDF behavioral change wheel.

The framework was adopted and customized to fit the purpose of this review and map the factors influencing older adults’ digital engagement. For example, in the capability domain, physical and psychological changes attributed to age and aging-related processes had been included as identities that influence digital engagement. The environmental context, which reflects factors that are physically external to the individual, for example, the technology-related features, was grounded in this domain.

Although our preliminary assessment indicates that there are reviews on older adults’ digital engagement, previous reviews have focused on the effect of technologies on specific health or social outcomes, and there is little evidence showing the whole spectrum of users’ experience journey throughout the technological engagement life cycle, especially on nonuse and sustained digital engagement [18]. Details of this engagement typology, nonuse, initial adoption, and sustained use have been published elsewhere [19]. In line with the mainstream technological models, most studies and reviews have focused on the individual motivation aspect of behavior. A comprehensive, systematic, and robust theoretical framework that helps understand individual motivations, abilities, and external social, environmental, and technological factors is required.

Therefore, in this scoping review, our aim was to map the existing literature on older adults’ digital engagement, including technological nonuse, initial adoption, and sustained use using COM-B and TDF models to answer the following questions:

We conducted a systematic scoping review of the literature guided by the Arksey and O’Malley framework and recent methodological developments to conceptually map the nature and extent of the literature and factors influencing older adults’ digital engagement [24,25]. An extension of the PRISMA-ScR (Preferred Reporting Items for Systematic Reviews and Meta-Analyses extension for Scoping Reviews) was used to present the result of the final review [26].

The inclusion and exclusion criteria of our scoping review were developed based on the participants, concept, and context guidelines of the Joanna Briggs Institute (see Textbox 2 for summary inclusion and exclusion criteria). Primary studies including participants with a mean age of ≥65 years that investigated everyday technologies and remote care technologies on nonuse, initial adoption, and sustained use of technology were included. Peer-reviewed studies published in English and from a global context were included in this review, whereas anecdotal evidence, reviews, and unpublished works were excluded.

A comprehensive search strategy of major electronic databases such as MEDLINE, PsycINFO, CINAHL, Web of Science, Association of Computing Machinery Digital Library, Google Scholar, and Library and Information Science and Technology Abstracts was conducted to locate relevant studies (see Multimedia Appendix 1 for a detailed search strategy). We developed a comprehensive search strategy combining major subject headings and free texts, and their thesaurus, plural forms, and spellings in collaboration with an experienced university research librarian. Other relevant studies were also identified and included through reference checking and citation tracking.

All relevant articles identified in our search strategy underwent 2-stage screening process: title and abstract screening and full-text screening. The Evidence for Policy and Practice Information reviewer software (version 4; Evidence for Policy and Practice Information and Co-ordinating Centre) was used to facilitate the screening process. The articles were screened against the inclusion and exclusion criteria developed by the authors (ASK, LO, HH, and KG).

We reported the study characteristics, types of digital technologies investigated, and level of digital engagement under investigation with numbers and percentages using frequency tables and charts. The factors influencing older adults’ digital engagement reported in the primary studies were extracted and charted. The summary findings, barriers, and facilitators identified from each included study were uploaded to the NVivo (version 12; QSR International). Conventional content analysis was used to determine the presence of certain sentence fragments, words, themes, or concepts in the text and coded into conceptually congruent categories [27]. This was done by reading the charts from individual primary studies and coding them line-by-line into tentative themes. We followed an iterative process (reading, coding, and revisiting the codes) to establish interconnections among the resultant elements and categorized them into COM-B and TDF constructs [28].

Of the total 11,412 articles identified from our search results, 1856 (16.26%) duplicates were removed. In total, 1141 (11.94%) full-text articles were obtained by screening the title and abstracts of 9556 records. Finally, 8.41% (96/1141) of articles were included in the review by assessing 1141 full-text articles against eligibility criteria. The main reasons for exclusion were non-English studies, published before 2005, mean age of the study participants <65 years, and studies with insufficient information on older adults’ digital engagement perspective (see Figure 2 that shows the PRISMA-ScR flow diagram for details of screening and eligible articles).

Most (61/96, 64%) of the studies were published in the last 5 years (between 2016 and 2021), and 28% (27/96) were published between 2010 and 2015. Geographically, most of the literature was from North America (43/96, 45%), followed by Europe (32/96, 33%), Asia (13/96, 14%), and Australia (8/96, 8%). Methodologically, 47% (45/96) of studies used a qualitative method, 36% (35/96) quantitative methods, and 17% (16/96) mixed methods. Table 1 summarizes the characteristics of the articles included in this review (see Multimedia Appendix 2 [29-124] for details on the characteristics of the extracted studies).

Table 2 presents the specific types of digital technologies that were investigated. Most (54/96, 56%) of the studies investigated everyday digital technologies. Among these, 26% (14/54) of studies investigated multiple technologies, followed by the internet (10/54, 19%), information communication technologies (10/54, 19%), social networking sites (7/54, 13%), and computers (6/54, 11%). The rest 44% (42/96) of the studies, were on remote care or assistive technologies. In this category, telehealth or telecare and robots were investigated in 24% (10/42) studies. Furthermore, remote monitoring technologies, tracking technologies, mobile health, and eHealth were investigated in 7% (3/42) of studies.

Most (57/96, 59%) of the articles investigated the initial adoption stage of the digital engagement, followed by sustained digital engagement (13/96, 14%). Only 2% (2/96) of the articles studied digital technology nonuse. A significant number of studies (24/96, 25%) investigated >1 or all engagement levels (Table 3).

This scoping review provides a synthesis of the literature on older adults’ experiences and facilitators of and barriers to digital engagement. We conceptualized digital technology engagement as a three-stage continuum (nonuse, initial adoption, and sustained use) to capture the entire range of individuals’ experiences from technology abandonment and acceptance to actual and continued use. A process predicated on ongoing negotiations or renegotiations and nonlinear progression between stages. Our review included 96 primary studies exploring a range of everyday and remote care technologies and demonstrated the complexity and multiple intertwined factors at personal, sociocultural, and environmental levels influencing digital engagement among older adults. We mapped these factors over the COM-B and TDF behavioral change models to facilitate articulation and provide a basis for future interventions that improve digital engagement among older adults. Environmental context and resources, beliefs about capabilities, and physical and cognitive capabilities were the most cited factors across the engagement stages. There is little research on the nonuse and sustained-use stages, as most studies in our review investigated the initial adoption stage of digital engagement.

One central theme across engagement stages was older adults’ digital knowledge and skill capabilities [29,30]. Over the past years, older adults’ digital competence, access to digital technology, and interest in further education have significantly improved [5,125]. However, a significant proportion of older adults have insufficient or lack the required digital skills. For example, only 1 in 4 European older adults have basic digital skills [126]. According to the European Union (EU) digital competence framework, digital literacy comprises 5 indicators: information and data literacy, communication and collaboration, digital content creation, safety, and problem-solving [127]. Such guidelines with broader definitions and detailed outlines of digital skills could help guide the development of curricula to equip older adults with essential basic digital skills. Innovative and interactive practical learning delivery modalities, for example, web-based learning and digital games, could be used [121]. Although older adults’ digital skills reflect familiarity and varying levels of exposure through education or work contexts in the past, the changing requirements related to capability, as well as rapid technological development, necessitate continued training and support.

Not surprisingly, the costs of procuring and maintaining technology and indirect costs (eg, electricity consumption) were cited several times as barriers to older adults’ digital engagement [36,44,67,73,84,101]. This aligns with previous reviews that low income predicts low technology ownership and low access to quality support and digital engagement in general [128,129]. For example, Choi et al [130] reported a strong correlation between discontinuing internet use and low income among homebound older adults. “Digital poverty,” that is, inability to fully use available digital platforms owing to lack of finance, access (eg, geographic exclusion) and lack of skill, is a growing practical and policy concerns even among economically developed countries. According to the recent report from the United Kingdom House of Commons, a significantly lower proportion of households with income between £6000 and £10,000 “GBP £1 (US $1.42) have home internet access compared with those households who earn £40,000 and above (51% vs 99%); this divide has even worsened during the COVID-19 pandemic with the increasing hybrid ways of coping [131].

Our review demonstrated that the usability of technology is highly dependent on its material features (physical property, functionality, and interoperability). Previous studies have reported that older adults find it cumbersome when technologies have multiple buttons, multi-factor authentications, poor quality user interfaces, and outputs [30,41,44,45,85]. These difficulties could emanate from the inherent complexity of technologies, design failures, or lack of necessary training and skill sets to operate technology. However, it is noteworthy to understand the extreme heterogeneity in older users’ experience, background, and diversity of applications and to take a precautionary approach when making a technological design recommendation based on barriers and facilitator studies. Continued efforts to strike a balance between usable, enjoyable, and secure technologies through value-based design ethos that considers older adults’ physical, psychological, and contextual needs must be promoted. This includes accessibility features that allow older adults to customize technology according to their needs.

We found that fear of safety and invasion of privacy were barriers to digital engagement and a growing concern among older adults, regulatory bodies, and researchers [52]. This was in line with previous findings on the growing digital distrust and apprehension among users owing to technology intrusiveness; increased web-based activities; use of personal data for health and financial reasons; increasing number of data breaches; data monetization; and lack of transparency on why, how, and by whom data will be handled [52,111]. Privacy regulations such as the EU General Data Protection Regulation have improved the privacy accountability of suppliers and raised users’ awareness of their privacy rights [132]. However, older adults’ awareness and proactive prevention of personal data are significantly lower than those of their younger counterparts [133]. Addressing these serious concerns requires cross-cutting interventions that ensures older adults’ empowerment, simultaneously strengthening legal frameworks and institutions and cross-sectoral partnerships. For example, regulatory bodies need to capitalize on and constantly update existing privacy regulations to enforce and protect individuals. In addition, concerned stakeholders need to provide continuous education on safety, privacy rights, and regulations that will improve the older adults’ privacy efficacy, privacy concerns, and trust of older adults. Businesses and service providers also need to play their part in implementing privacy regulations, establishing a clear communication protocol and transparency. Although this will primarily benefit users, recent reports have indicated that firms with effective privacy protection systems have a significantly higher return on investment; “beyond meeting compliance requirement-good privacy is indeed good for business and individuals” [134].

Digital technology takes on multiple explicit and implicit meanings for its users. In line with previous studies, our review demonstrated the role of technology in promoting active and independent living and enhanced personal autonomy, power, and control [96,113,119]. However, technology could also imply a sense of dependency and decline contrary to the primary purpose of promoting independence [28]. For example, studies have reported assistive technologies symbolizing an image of “being old” opposite to the desired or ideal self-image perceived by older adults and could be associated with agist stigma and discrimination. This apparent latent tension between the individual’s identity and perception of society aligns with the mainstream identity theory that describes the role of self-image and the perception of others in individuals’ decisions [135]. These symbolic properties and their influence on the adoption or rejection of technology need further research.

The COM-B and TDF mapping in our review ensured that a wide range of emergent determinants were explored. These comprehensive frameworks cover intrinsic factors pertaining to individuals’ abilities and motivations and extrinsic factors related to social, technological, and environmental factors. These factors can be used by researchers, technology developers, caregivers, and program implementors to inform the development of implementation models for optimal digital engagement among older adults. Previous studies have given a tremendous emphasis on the individual motivational aspect of behavior, including beliefs about consequences and beliefs about capabilities, such as perceived usefulness and ease of use [42,136]. These themes have been widely explored in previous technological acceptance models and theories and have attracted considerable interest for research [44,80,84,92]. However, looking beyond motivation and addressing all other moderating factors are required to close the digital divide between age groups.

This review highlights several areas that require further research. First, research needs to move beyond the prevailing focus of the classic technological acceptance models and theories on initial adoption and individual motivational factors. Accordingly, conceptualizing digital engagement as a continuum instead of a one-time decision could help understand individuals’ journeys holistically, the impacts of disengagement on well-being, and how it marginalizes older adults. Second, there is a need for a standard definition and validated measuring tools for the nonuse and sustained-use stages of digital engagement. Third, theorizing older adults’ digital technology nonadoption, uptake, and continued engagement using in-depth and contextually situated methodologies is needed. Such theorizing from older adults’ experiential accounts could help illuminate the meaning of technology; the interaction between the material and symbolic properties of technology; digital engagement meanings on identity, interpersonal relations, capabilities, motivations, affect, and emotions; and how all these influence adoption, decisions, dignity, and well-being.

This review followed a systematic approach to review evidence on digital technology engagement of older adults, which included identifying review questions, comprehensive search across all major databases for technology and health, application of explicit inclusion and exclusion criteria, use of a systematic and structured theoretical framework to map out the evidence, and synthesis of the findings. However, this review has several limitations. First, only studies published in English were included, and most of the studies were from North America (43/96, 45%) and Europe (32/96, 33%), which might affect the transferability and coverage of the identified literature. Second, in this review, we sought to understand the barriers to and facilitators of overall digital engagement instead of technology-specific engagements and might have missed the nuanced variations. Third, the sustained digital engagement stage is not a well-defined research outcome and a consensus on its definition has yet to be reached. To ascertain what studies could fall under this category, we followed either the description in the primary studies if they explicitly declared that they were investigating the sustainability of digital use, or we ascertained through careful reading of the description of the study to see whether sustained use over a significant time was one of the objectives or implicated in the study. Finally, we did not hold a formal stakeholder consultation because of time and resource constraints. Instead, the findings of this review have been widely discussed in a consortium and many conferences and other informal gatherings involving older adults.

The digital engagement of older adults can be conceptualized as a three-stage continuum (nonuse, initial adoption, and sustained use) and a negotiated process possessing acceptance, rejection, and temporal characteristics. Little research has been conducted on nonuse and sustained engagement stages. Most studies in this review investigated the initial adoption stage. Using the COM-B and TDF models enabled us to identify a wide range of salient intrinsic and extrinsic determinants across engagement stages. Considering the barriers identified, including but not limited to the changing capability requirements, cost, access to technology, safety and privacy concerns, and design stereotypes and assumptions, could improve older adults’ digital experiences, facilitate better digital engagement, and optimize future digital interventions and scale-up. Furthermore, empowering older adults with digital skills through a learner-centered approach on a need-to-know basis should be promoted. Future research aimed at understanding older adults’ everyday world of experience, the meaning of digital technologies, and how they cope with this fast-paced digital development is critical for promoting meaningful digital engagement. The range of contexts and values, which older adults avoid, adopt, or continue to use, in digital technology and standardized tools that measure these outcomes require further research.