This review is registered within the Open Science Framework (OSF) [18], and adheres to the PRISMA-ScR (Preferred Reporting Items for Systematic reviews and Meta-Analyses extension for Scoping Reviews) [19]. Since we aimed to map all existing evidence on eHealth support regardless of methodological quality and risk of bias, critical appraisal—an optional component of the PRISMA-ScR—was not conducted. Our review follows the step guide of Arksey and O’Malley [20], comprising five stages: (1) identifying the research question, (2) identifying relevant studies, (3) performing the study selection, (4) charting the data, and (5) summarizing and reporting the results [20].

We have chosen to conduct a scoping review to map the diversity of research available on eHealth support. Following Arksey and O’Malley’s framework [20], scoping reviews can be performed for several reasons. Our review addresses two: (1) providing an overview of the extent, range, and nature of research activity on a broad topic, and (2) identifying research gaps. This approach is particularly suited to eHealth support, as the field is broad and initiatives are scattered across domains.

Throughout the review, two types of studies are considered:

Eligible studies were empirical, written in English, published in peer-reviewed journals, from 2014 onwards, meeting the population, intervention, and outcome (PIO) criteria. Studies included those targeting (potential) patients (population) receiving help or support (intervention), with reported outcomes related to eHealth inclusion (outcome). We defined eHealth inclusion as accessibility, (digital or eHealth) literacy, and overall inclusion. In the included studies, related outcomes such as adoption, frequency of use, user satisfaction, and digital skills were also reported. These were considered part of the broader conceptualization of eHealth inclusion and were included in our outcome categorization.

Our review focused exclusively on peer-reviewed studies, as our research questions concerned both proposed strategies (RQ1) and evidence of the effectiveness of support (RQ2). Peer-reviewed publications provide a transparent and reproducible source of such evidence. Gray literature was excluded to maintain methodological transparency and reproducibility.

After consultation with a librarian, search terms were developed based on the PIO framework, resulting in 3 search strings. The population terms, linked using the OR operator, included: “patient*,” “people,” “client*,” “care consumer*,” “care receiver*,” and “older adult*.” The intervention string consisted of terms related to support, combined with “intervention*,” “network,” and “system.” The outcome string included synonyms for “telemedicine,” “eHealth,” and “mobile health,” as well as terms related to “health literacy,” “inclusion,” and “accessibility.” These strings were combined using an “AND” operator. The complete search strategy can be found in Multimedia Appendix 1.

Data were extracted from the included records based on a personalized configuration of the NASSS and PIO framework. Figure 1 shows the NASSS framework [17]. Table 1 shows how the NASSS dimensions were combined and interpreted in this review. In addition, Multimedia Appendix 2 shows the items for which data were extracted. For strategy studies, data were extracted on the population and intervention only, as these studies typically proposed forms of support without reporting outcomes. For intervention studies, data were extracted based on the full PIO criteria, since these studies reported outcomes related to eHealth inclusion, either descriptively or through statistical testing. We also classified countries of the studies by region according to the United Nations geoscheme. To gain insight into health system characteristics, we drew on the Commonwealth Fund’s International Profiles of Health Care Systems of 2015 and 2020 [21,22]. Multicountry studies were counted once overall. In Multimedia Appendix 3, we present the context of the different countries within the study row. Data extraction was primarily performed by WEB, in consultation with authors EH, EIM, and NDS.

While the NASSS framework does not explicitly categorize support in the context of eHealth, its dimensions do address micro- (individual adoption), meso- (organizational assimilation), and macrolevel challenges (policy and regulatory environment) [17]. These levels are adopted to differentiate actors across levels for providing eHealth support.

The PRISMA-ScR flowchart in Figure 2 illustrates the screening process. Deduplication was performed in EndNote (Clarivate) following the Bramer method [23] to remove duplicates. After deduplication, 4352 articles remained for title and abstract screening. Authors WEB and LART used Rayyan.ai, a web-based and mobile app developed to facilitate the screening process for reviews, to independently screen all titles and abstracts for inclusion based on the predefined criteria. Rayyan enables blinded screening by allowing reviewers to hide each other’s decisions during the screening process and allows reviewers to define and apply their own exclusion criteria [24].

Before screening, authors performed a pilot test on 20 random articles, yielding a 75% (15/20) agreement with a Cohen kappa of 0.42, indicating moderate agreement [25]. After discussing and refining the inclusion criteria, the authors independently screened all titles and abstracts, resulting in discrepancies for 28 cases, yielding a 99.4% (4324/4352 articles) agreement and a Cohen kappa of 0.99. The 28 discrepancies were resolved collaboratively, and 2 non-English articles were excluded, leaving 108 articles for full-text screening.

For the full-text articles, the authors again assessed agreement on a sample of 20 random articles, for which full consensus was achieved. After screening all 108 articles, 11 conflicts were found. A total of 10 of these were resolved by authors LART and WEB through discussion, and EIM made the final decision on the remaining articles. After this process, 36 articles remained, and backward reference checking resulted in 40 included studies [26-65].

In Table 2, the included studies are categorized by population, intervention, and outcome. The second column summarizes the populations and interventions of the strategy studies, while the third column summarizes the populations, interventions, and outcomes of the intervention studies. Detailed descriptions of both study types are provided in Multimedia Appendices 4 and 5 [26-65].

Next to microsupport, studies also proposed to include support actors at the mesolevel, by focusing on various actor types described in this section.

On top of identifying actors to provide eHealth support, broader recommendations exist, such as the stressed need for in-person support to overcome barriers for eHealth inclusion [38,39]. Also proposed are allowing blended models of care, where eHealth and physical care are combined to accommodate diverse patient needs [29,34,39,40]. Given differences in patient preferences and learning styles, studies emphasize the importance of providing materials in personalized, appropriate formats ranging from printed information booklets to spoken tutorials, and in-person at-home or clinic-based visits [29,34,39,40]. In the “Findings Part II: Intervention Studies” section, we report findings from intervention studies, which examined the outcomes of specific support strategies.

This section reports findings from intervention studies, for which we examined reported outcomes related to eHealth inclusion.

We analyzed studies by country and health system (see Multimedia Appendix 3 [26-62] for the overview of included studies by country and health system characteristics). Most studies were conducted in the United States (21/40), followed by Asia (8/40; South Korea, China, Israel [Western Asia], and Singapore), Europe (5/40; Italy, Germany, the Netherlands, and the United Kingdom), Africa (1/40; Malawi), Canada (2/40), and Oceania (1/40; Australia). A total of 2 studies had a multicountry design: one in Canada and the United Kingdom, and another spanning India, Kenya, several West African countries, and Tanzania.

Across regions, some patterns emerged. US studies reported health care–provided support in 16/21 cases [33,38,42,45,48-54,58,60-62,65]. Among Asian studies, 5/8 emphasized interpersonal support [29,30,34,35,43], 2/8 emphasized community support [36,57], and 2/8 focused on policy-level support [44,57]. European studies showed a mix of support actors, including health care provider support as well as educational and community organization support [26,32,41,47,55]. The single Australian study highlighted family and friends as central intermediaries [31]. In Africa, the Malawi study described a community-driven intervention [59], while the multicountry African study examined public initiatives aimed at addressing community needs across India, Kenya, West Africa, and Tanzania [39].

A total of 2 discrepancies were identified between strategy and intervention studies regarding actor involvement. Interpersonal support appeared in only 14.3% (3/21) of interventions, though it was proposed in 57.9% (11/19) of strategies. While interpersonal support can be beneficial, authors indicated it could lead to dependency rather than fostering autonomous eHealth use. A study incorporating health care support, followed by interpersonal support, if necessary, did not find significant improvements in digital literacy outcomes [53].

Health care providers engaged in 71.4% (15/21) of intervention studies but only 36.8% (7/19) of strategy studies. Interventions involving solely healthcare support showed no statistically demonstrated effects on adoption outcomes. Some of these interventions were prompted by the need to transition to eHealth during the COVID-19 pandemic [50]. Several studies also emphasized the need for sustainable funding beyond the project’s duration [60,62].

For other actors, such as community and policy-level participants, the differences between strategies and interventions were less pronounced.

This scoping review mapped evidence on support strategies and interventions to enhance eHealth inclusion. Across 19 strategy studies, 6 support actor types were identified: interpersonal contacts, technology suppliers, health care organizations, educational institutions, community organizations, and policymakers. Interpersonal contacts were the most frequently proposed actors, followed by community and health care organizations. The 21 intervention studies targeted 4 primary outcome measures: adoption, use, skills, and attitudes. Interventions solely involving health care organizations showed no statistically demonstrated effects on adoption, while use-focused interventions led by health care yielded mixed results, involving positive outcomes for 4 interventions. Attitude-focused interventions yielded mixed results. Multiactor collaborations showed statistically significant improvements in skills.

This review makes 5 research contributions. First, it identified a discrepancy: while interpersonal contacts were proposed in 58% (11/19) of strategy studies, only 14% (3/21) of intervention studies involved them. Strong ties with family and friends may appear to be a logical choice for support provision, given that they are often associated with high levels of trust [66]. However, research on self-management indicates that weak ties—such as those with organizations—tend to provide greater long-term durability, while focusing on strong ties could result in dependency [67]. Future research could explore the effectiveness of organizational support in providing durable, independence-oriented interventions.

Second, health care actors were advised to offer support in 37% (7/19) of strategy studies but engaged in 71% (15/21) of intervention studies. This might reflect the urgency that health care actors perceived in adopting eHealth and the immediate need for support. Most of these interventions were in pilot phases, with 14 out of 21 occurring in response to or during COVID-19 (see Multimedia Appendix 7 [45-65]) when physical consultations were limited and remote care became essential. Sustaining health care–led support beyond the pilot phase may be challenging, which might partly be rooted in health care providers’ high workloads. Addressing workforce shortages is important [1], although adding eHealth support tasks could further increase workloads. Therefore, multiactor interventions, such as the multiactor DOORs program between health care and community organizations—implemented in 2 included studies [45,48]—might offer more sustainable and transferable alternatives. However, while several multiactor interventions show positive effects, results are mixed, and further research is needed to explore factors influencing long-term effects.

Third, our review showed intervention heterogeneity in terms of the types of eHealth addressed, the support actors involved, and the outcome measures. This high variability presents challenges for comparing studies and for developing scalable and generalizable support solutions, as the inconsistency in outcome measures complicates comparison across studies. This aligns with research on eHealth, where heterogeneity across study approaches also makes comparisons difficult [68,69]. Even though researchers like Eysenbach have advocated for standardized ways to evaluate eHealth [70], actual evaluations are often not based on these guidelines. Future research is needed into guidelines allowing comparisons while considering the flexibility of the field of eHealth, experiencing fast developments.

Fourth, individuals may respond differently to distinct types of support or learning formats, resulting in varying outcomes depending on the intervention setting and the target group. As past research has emphasized, there is no one-size-fits-all approach to eHealth support, contributing to the wide range of strategies and outcomes observed in literature [71]. Future research should explore possibilities for the development of sustainable eHealth support interventions that incorporate personalization to accommodate diverse learning styles.

Fifth, our findings suggest regional differences in how eHealth support is organized, shaped by contextual factors. For example, US studies more often reflected health care–based interventions, whereas European studies reported a broader mix of actors. Asian studies highlighted interpersonal and community involvement, while the 2 studies conducted in African settings described grassroots or public initiatives designed to meet community needs. These contextual findings underline the importance of examining how health system design and cultural context influence effective eHealth support. At the same time, interpretation must remain cautious: the evidence base is dominated by US studies, and observed regional patterns may partly reflect study design rather than systemic differences. Future research could focus more specifically on linking health system characteristics, economic and cultural factors to the types of support that are most effective in different contexts, and on developing strategies to translate these insights into practice.

This review has several limitations. The heterogeneity in study aims, methods, and outcomes complicates comparisons. A scoping review was chosen to account for this variability, enabling a broad mapping of evidence. Consequently, no quality assessment was performed. We excluded gray literature to ensure a reproducible focus on peer-reviewed evidence, thereby providing a solid basis for future research, although this may have led to the omission of relevant practice-based initiatives. None of the reported intervention effects were negative. While this may suggest eHealth support is beneficial, it could also reflect publication bias, as studies with negative results are less often reported. The Hawthorne effect may also play a role, where participants alter behavior simply due to being observed [72]. Both factors could contribute to overestimating the appearance of positive results. Furthermore, many interventions, especially those involving health care providers, are resource-intensive or dependent on short-term funding, raising concerns about long-term sustainability. Also, studies included in this review often lack reporting information on the actual reasoning behind the design of the support and actors involved, stressing a need for future research.

While this review maps different strategies and interventions for support in eHealth use, the actual reported outcomes of interventions remain mixed, especially for adoption and attitudes. However, this review shows potential for health care organizations to establish support, as part of the interventions involving health care actors did show positive effects on eHealth inclusion. Health care organizations could benefit from external support networks or by integrating existing initiatives into healthcare practice. An example of this could be to explore opportunities for collaborations with community organizations or technology suppliers. However, sustaining positive effects requires both long-term funding and shared responsibility across actors. Further research is needed to determine which approaches yield sustainable, long-term improvements in eHealth inclusion.

This scoping review aimed to identify strategies and interventions to enhance eHealth inclusion. Using a novel approach that combined the PIO and NASSS frameworks, we extracted and analyzed data across 40 studies. Strategy studies highlighted interpersonal contacts, technology suppliers, health care organizations, educational organizations, community organizations, and policymakers as key support actors. Although studies had a wide variety of methods used, primary outcomes targeted, and results found, some conclusions can be drawn to inform eHealth support. While many strategies focused on interpersonal contacts at the microlevel—such as those of family and friends—health care providers emerged as primary actors in targeted interventions. Their sole support was offered in interventions targeting adoption and use, yet research yielded mixed results for adoption. Support for improving eHealth skills often benefited from collaborative efforts involving educational organizations, community networks, and interpersonal support.