This scoping review used the guidelines of Joanna Briggs Institute’s (JBI’s) Methodology for Scoping Reviews [26] and follows the methodological framework proposed by Arksey and O’Malley [27] which consists of the following stages: (1) identifying the research question; (2) identifying relevant studies; (3) selecting studies; (4) charting the data; and (5) collating, summarizing, and reporting the results. Our review also complies with the PRISMA-ScR (Preferred Reporting Items for Systematic Reviews and Meta-Analyses extension for Scoping Reviews) checklist (Multimedia Appendix 1) [25]. We first developed a scoping review protocol, including a rationale for conducting the review, the main objectives, search strategy, inclusion and exclusion criteria, and methods for screening and data extraction, that was then piloted and discussed by the research team before finalizing. The final protocol was registered retrospectively in Open Science Framework on May 15, 2023.

A systematic search strategy was developed in consultation with an expert librarian (RS). The search strategy also adhered to the Peer Review of Electronic Search Strategies (PRESS) guidelines [28]. We systematically searched 15 electronic databases from inception up to 28 July 2023: MEDLINE, Global Health, Cumulated Index to Nursing and Allied Health Literature (CINHAL), SCOPUS, APA PsycInfo, Embase, Academic Search Premier, International Bibliography of the Social Sciences (IBSS), Applied Social Sciences Index & Abstracts (ASSIA), Association for Computing Machinery Digital Library (ACMDL), Institute of Electrical and Electronics Engineers (IEEE), Web of Science, Educational Resources Information Centre (ERIC), and Cochrane library. Policy Commons was used to search for gray literature.

The systematic and comprehensive search strategy consisted of key search terms derived from existing search strings and bespoke for each electronic database. The search terms were as follows: homeless* OR temporary accommodation OR roofless OR unfit hous* OR inadequate hous* OR night shelter OR shelter* OR sofa surf* OR rough sleep* AND information communication technolog* OR cell phone* OR mobile app* OR mobile technolog* OR mobile health OR (m health OR e health OR mhealth or ehealth) OR online OR digital OR (telehealth OR tele health OR telemedicine OR tele medicine OR telecare OR tele care) OR social media OR internet OR (web based OR web-based) OR wearable* OR (Smartphone OR smart phone) OR Mobile phone OR Instant messag* OR (Email or electronic mail or e mail) OR (Smartwatch OR smart watch) OR (WhatsApp OR Instagram OR Facebook OR Telegram OR Signal OR Viber) (note: * indicates a wildcard). The results were combined using Boolean operations and adapted for each database (Multimedia Appendix 2). We also scanned references of the included articles for any relevant studies.

After the publications were retrieved and duplicates removed using Endnote (Clarivate), search results were imported into the Covidence software management system (Veritas Health Innovation) for additional deduplication and screening by multiple reviewers.

As advised by JBI guidelines for conducting scoping reviews [26], the population, concept, and context framework was used to define eligibility. Textbox 1 shows the inclusion and exclusion criteria in line with the population, concept, and context framework and contains additional study elements relevant to the eligibility criteria. The ETHOS definition of homelessness framed the inclusion of participants experiencing homelessness [29]. Organizations supporting older people experiencing homelessness were considered to be any health or social care services or third-sector organizations providing a service to people experiencing homelessness. Due to accelerated aging, those who are 50 and experiencing homelessness are defined as “older” in contemporary research [2,30]. Therefore, only studies that included participants with a mean age of 50 years and older or studies with a clearly defined group of participants older than 50 years were included. The nature and causal pathways of homelessness vary globally [31]. To acknowledge that interventions for people experiencing homelessness will diverge due to social and cultural structures, health systems, and the provision of emergency accommodation [32], the scope of this review will be refined to solely high-income countries as defined by the Organization for Economic Co-operation and Development. This allowed for interventions to be more appropriately compared in the analysis [2]. So as not to exclude any innovative interventions, the definition of “digital intervention” was kept intentionally broad to include digital, web-based, or mobile interventions used by people experiencing homelessness to improve social, health, or prospective outcomes.

Further to the inclusion and exclusion criteria outlined in Textbox 1, throughout the review, publications were excluded for the following: (1) no full text available and (2) no English language version available.

Level 1 screening focused on inclusion criteria based on titles and abstracts, while level 2 screening involved reviewing full-text articles. Four reviewers (EA, LN, CS, and RG) independently screened all titles and abstracts. Reviewers met throughout the screening process to discuss queries and reduce uncertainties. Two reviewers (EA and LN) completed the full-text screening independently, with disagreements resolved by discussion with the reviewers and SWM.

The selected publications were read, annotated, and entered into a Microsoft Excel spreadsheet. EA and LN piloted the data extraction sheet with 2 of the included studies and then revised it in consultation with ED and SWM. We did not critically appraise the included studies, given that this is not typically an objective of a scoping review and the large research design heterogeneity of the studies reviewed [33].

To summarize the data, where applicable, we conducted a basic numerical analysis, for example, the proportion of participants older than 50 years in each study. Meta-analysis was not feasible due to the necessary inclusion of heterogeneous studies in answering the research question. We used a narrative synthesis approach to organize and present relevant findings. Qualitative data (eg, perceptions of digital interventions, barriers to digital interventions for older people experiencing homelessness, and facilitators to digital interventions for older people experiencing homelessness) were imported to NVivo 12 software (Lumivero) for analysis by EA. This approach is characterized by textual summaries and explanations of findings, which are first synthesized by thematic analysis to explore relationships among studies. Thematic analysis in this context involved iteratively identifying, classifying, and sorting the most important themes and concepts across studies [34]. The core research team (EA, LN, SWM, ED, MW, and CS) applied Braun and Clarke’s [35,36] reflexive thematic analysis approach, which involved familiarization with the data, generating initial codes, identifying and refining preliminary themes, reviewing themes collaboratively, and ultimately constructing and defining final themes. This iterative, reflexive process—guided by our own perspectives and the review’s research questions—enabled a deeper interpretation of the data, resulting in us developing the reported themes.

Our searches yielded 19,915 records. After removing duplicates 18,728 records were title and abstract screened (Multimedia Appendix 3). Ten articles met the inclusion criteria [37-46]. All records included in our review were peer-reviewed studies reporting on digital interventions for older people experiencing homelessness that had a clearly defined age group of participants older than 50 years or the mean participant age was older than 50 years (Figure 1 portrays the adapted PRISMA-ScR flow diagram). This screening criteria resulted in 90.5% (n=5902) of the total participants (N=6557) being older people experiencing homelessness (older than 50 years). In 6 of the studies, participants were veterans experiencing homelessness [38,39,41,42,45,46]. A total of 9 of the 10 studies were conducted in the United States [38-46], and the other was conducted in Hungary [37]. Studies were published between 2013 and 2023, with 4 published after the 2020 lockdowns of the COVID-19 pandemic [37,41,45,46]. The study design included descriptive studies [39,43], uncontrolled pilot studies [37,38,42,44,46], and pilot randomized controlled trials [40,41,45].

The included studies were heterogeneous concerning intervention content, delivery, and reported outcomes. Tables 1 and 2 show the intervention delivery and implementation. Interventions could be categorized into implementing telecare for people experiencing homelessness [37,38], distributing technology to enable digital inclusion [39,46], text message reminder interventions [40,42], and 4 interventions delivered digitally [41,43-45] (summarized in Figure 2). Interventions delivered digitally ranged from smoking cessation support [43], vocational training [41], physical activity promotion [44], and cognitive behavioral therapy [45]. We used the broad definition of telecare used by Barlow et al [47], “the use of communications technology to provide health and social care directly to the user (‘patient’)” where interventions self-described as “telemedicine” [38] and “telehealth” and terms were used interchangeably.

No papers included aimed to develop or deliver interventions specifically for older people experiencing homelessness participants. Additionally, no interventions were found specifically for organizations supporting people experiencing homelessness. Thus, all included studies were developed to be used by individuals experiencing homelessness without specific considerations for older age.

Three included studies concluded that the intervention demonstrated the viability of digital delivery for older people experiencing homelessness [37,39,42]. There was evidence of a feasible telecare setup in shelters offering accommodation to people experiencing homelessness that might support the planning of future telecare services for vulnerable populations [37]. McInnes et al [42] proposed that text messages are a feasible digital intervention as they are an unobtrusive connection to patients, and mobile phones are one of the few communication tools that people experiencing homelessness can attain. Further, they concluded that appointment reminders are greatly needed for this population because they frequently lack the tools that nonhomeless take for granted: reliable mailing addresses, landline phones, wall or computerized calendars, and social supports to remind them of appointments [42].

The scoping review examined the range, nature, and use of digital interventions available to older people experiencing homelessness and organizations that support older people experiencing homelessness. We identified only concerned studies within an Organization for Economic Co-operation and Development context detailing digital interventions with participants who were older and experiencing homelessness. The nature of interventions included digitally delivered primary health care (telecare) [37,39], appointment reminders [40,42], technology distribution [39,46], and well-being interventions delivered in a digital format [41,43-45]. Included interventions found that common barriers were existing digital exclusion factors for older people experiencing homelessness, such as digital literacy, absence of safe storage for technology, and inconsistent internet connectivity. Suggested facilitators for older people experiencing homelessness in digital interventions were organizational and peer support. The searches found no interventions designed for adoption by support services (as opposed to the older people experiencing homelessness user). Further, it should be noted there was a lack of studies reporting on other dimensions of exclusion (ethnicity, gender, etc) for older people experiencing homelessness. Finally, no included interventions were intended for sole use by a cohort of older homeless adults; subsequently, there were no specific considerations for older people in the intervention design. Therefore, this review highlights the paucity of digital interventions designed for and delivered to older people experiencing homelessness.

This review demonstrates evidence that digital interventions could benefit older people experiencing homelessness. Consequently, it is crucial to understand the prevalence and use of technology among older people experiencing homelessness to implement digital interventions effectively. People experiencing homelessness access to mobile devices varies greatly. One study showed that as many as 95% had a mobile phone, and 77% reported having a smartphone [48]. However, participants from an ongoing clinical trial at a homeless shelter in Texas reported lower (28.4%) access to an active cell phone. However, 88.6% of participants reported at least weekly internet use, and 77.2% used email [49]. It is well established that older adults in the general population use technological solutions at lower rates than younger adults [50]. In addition, little is known about access to and use of the internet and mobile phones by older homeless adults. Raven et al [20] make one of the only attempts to describe the access to and use of mobile phones, computers, and the internet among 350 homeless adults older than 50 years. They found that most (72.3%) participants owned or had mobile phone access. Participants used phones and the internet to communicate with medical personnel (64.6%), search for housing and employment (30.7%), and to contact their families (82.3%). They concluded that participants had a lower prevalence of smartphone and internet access than adults aged older than 65 years in the general public or low-income adults. Participants faced barriers to mobile phone and internet use, including financial barriers and functional and cognitive impairments [20].

Only one included intervention analyzed age as a variable for use, where tablet use was less likely for older participants [39]. Similarly, in a sample of homeless-experienced adults aged 50 years and older, Raven et al [20] found that almost 3-quarters of participants had current access to a mobile phone. However, participants had a lower prevalence of smartphone and internet access than adults older than 65 years in the general public or low-income adults [20]. This demonstrates that premature aging and complex social challenges that are attributed to homelessness are significant factors in digital exclusion. Concurrently, in the literature, there are significantly more evaluated digital interventions for youth experiencing homelessness (YEH) than those retrieved for this review [49,51-61]. The disparity in tailored interventions for YEH and older people experiencing homelessness further illustrates the widening digital divide for older people experiencing homelessness.

Three included studies highlighted exclusionary factors associated with homelessness, causing barriers for digital interventions [40,42,46]. Technological competency, limited safe storage, and lack of internet connectivity were all referenced in this review as barriers to digital interventions for individuals experiencing homelessness [40,46]. Sieck et al [62] state that digital literacies and internet connectivity have been called the “super social determinants of health” because they address all other social determinants of health. For example, applications for employment, housing, and other assistance programs, are increasingly, and sometimes exclusively, accessible via the web [62]. In their systematic review of technology for people experiencing homelessness, Heaslip et al [19] found that older people experiencing homelessness felt further marginalized by the modern benefits system that “assumes” digital competence and confidence. As participation in most included studies was voluntary, older people experiencing homelessness with a more positive attitude and openness toward telecare might have been overrepresented in the sample, skewing the sample to those more digitally literate [37]. “Access instability” is a term used by Galperin et al [63] to describe their findings that reliable access to electrical power represents a fundamental yet understudied barrier to mobile use. Lacking a safe and reliable place to charge devices, the unstably housed must activate coping strategies that limit digital engagement and constrain use [63]. Overall, this suggests that while digital interventions have the potential to expand inclusion, existing literacy and connectivity barriers must be addressed in tandem with implementation.

Despite those barriers, digital interventions can also facilitate access to care. In this review, Garvin et al [39] suggest older veterans would benefit from simplified computer app designs and digital literacy training to increase comfort, confidence, and willingness to engage in their tablet adoption intervention. Concurrently, Sieck et al [62] posit that improving digital literacy skills to access valuable digital tools is key to reducing disparities. In a digital access survey for people experiencing homelessness, Sturman et al [64] found that digital literacy was positively associated with uptake in digital health interventions. Relatedly, Van den Berk-Clark and McGuire [65] argue that the issue of trust between people experiencing homelessness and support services is multifaceted and is influenced by technical competence and the degree to which the individual is made to feel welcome. A study of video consultation suggested that clinicians interacting with homeless-experienced older adults should prioritize addressing the potential skepticism of video calls. Further, it is proposed that clinicians should assess their access to and knowledge of video conferencing technology [66].

Similarly, 4 studies found that organizational or research team support facilitated the engagement of people experiencing homelessness with digital interventions. Furthermore, the authors identified peer support as a key contributor to participants’ comfort with digital interventions [37,40,44,46]. However, Gabrielian et al [38] stated that most participants did not use peer support and highlighted that fostering trust with those providing technology assistance was of primary importance. Similarly, Glover et al [54] found that YEH preferred automated mobile phone functions that avoided interaction with professionals and peers.

This review included 6 studies with interventions conducted with veterans in the United States [38,39,41,42,45,46]. The interventions benefited from the support and infrastructure of the US Department of Veterans Affairs (VA) [67]. It can be assumed that the integration of VA into the health system facilitated the implementation of the interventions. However, additional evaluative research on the context of the implementation is required to determine if internal process barriers nullify any potential facilitatory effects [68].

This scoping review highlights several research gaps, upon which we base the following recommendations:

The use of digital interventions is a rapidly developing area of practice for professionals with several elements to consider, including:

Due to longstanding digital barriers, implementing digital interventions without addressing older people experiencing homelessness’s digital exclusion will likely further damage trust and perpetuate existing poor support service access. As such, services should:

5.     systematically assess individual patients’ digital literacies,

6.     learn about their internet access, and

7.     work to address their needs.

8.     Partner with community-based organizations with expertise in digital literacy training to address comfortability.

Policy for digital inclusion of older people experiencing homelessness should prioritize free and accessible technology in public settings (eg, shelters, community centers, libraries, and harm reduction centers) and free access to mobile devices. Action is needed across government, public, private, and third-sector organizations to ensure capitalization on the potential for digital interventions to address health and well-being while minimizing the risk of exacerbating existing health inequalities.

This scoping review has several strengths. To the best of our knowledge, it is the first scoping review to describe the digital interventions for older people experiencing homelessness. We followed the JBI guidance for scoping reviews [33] and our database searching, handling of data, and reporting adhered to published guidelines for undertaking a robust standard scoping review [25,27].

Several limitations should be highlighted. Given the heterogeneity of study methods, we did not systematically assess the quality of studies. Similarly, only English language publications were included due to time and human resources. As there was inconsistency in how papers reported participant age, either a subgroup of participants was clearly defined as older than 50 years or the entire participant group’s mean age was older than 50 years. We acknowledge that this inclusion method may generate an incomplete picture of available interventions for older people experiencing homelessness. Most participants in the included studies are veterans based in the United States; therefore, any attempt to generalize these results should be undertaken with caution.

Finally, as this study is not a formal meta-analysis, we did not use more complex statistical pooling methods or analyze the heterogeneity in outcomes reported; as such, our results should be interpreted with these considerations in mind.

Our findings demonstrate the paucity of bespoke digital interventions for older people experiencing homelessness. However, the included studies demonstrate some evidence for the acceptability and feasibility of digital interventions for older people experiencing homelessness. To leverage the potential benefits of digital interventions for older people experiencing homelessness, implementing such interventions will require additional consideration of the multiple exclusionary factors experienced by older people experiencing homelessness. The anticipated increase in the number of older people experiencing homelessness warrants further research on the impact of digital interventions for this vulnerable population.