The global population is aging [1,2]. Care needs for older adults, including maintaining and extending older adult independent living, have likewise grown as an area of research focus [1,2]. An approach toward building and sustaining quality of life is to develop digital technologies to support fitness practices such as mobility, strength building, and aerobic health, defined as fitness technologies. The rationale for this technology-supporting fitness approach notes that the benefit for older adult independence in improving overall fitness in general [3,4] and strength building in particular [5] is well-established; and that, likewise, the scalability and affordability of fitness technologies relative to in-person specialist care [6] would enable such fitness support to be more accessible to more older adults.

In the last 7 years, there have been no less than 8 related review papers around computational health interventions for older adults including: wearable devices for increasing physical activity [7], mHealth and eHealth solutions for increasing physical activity [8], smart technology interventions for increasing physical activity in comparison to face-to-face interventions [9], eHealth interventions for increasing balance and preventing falls [10], virtual reality (VR) based interventions for improving functional mobility in older adults [11], adherence of technology-based exercise interventions [12], and older adult experiences using wearable-based interventions [13,14].

Crucially, the conclusions of these reviews, summarized in Table 1 and discussed in the Related Work section, indicate that successes in this area seem to be relatively mixed. Findings from the review are characterized as very low to moderate evidence of effectiveness [8,9], high risk of bias [7], and low methodological quality [10-12].

Given both the increasing need for such work and the growing interest across disciplines, it is concerning that these reviews, summarizing work across a range of technologies from accelerometers to VR, do not evidence stronger successes and impact. The focus of this scoping review, therefore, has been to determine what the factors may be in common across this research that may be hindering better outcomes. To interrogate this problem, we took a dual approach to our review and its analysis:

First, we reviewed research from across disciplines that used technology to help older adults build strength and maintain their independence. As such, the main criteria for inclusion are that a novel technology-based intervention was intended to support or improve fitness for older adults. From this collection, we then considered a range of shared quantitative features such as: technologies used, types of evaluations, whether co-design is part of the process, and duration of evaluations. Each of the metrics is described in the Methods section below.

Second, we developed an assessment tool drawing from research that identifies the needs and preferences of older adults with respect to technology and physical fitness to assess current older adult fitness technologies’ design and research processes. The assessment tool was developed through an analysis and synthesis of extant research addressing 3 issues: needs and preferences for older adult technology use for strength building and other exercise [15-18], older adult–specific factors that influence technology acceptance in general [19,20], and barriers and enablers for older adults’ physical activity [16,21,22]. This tool, which we have called the Older Adult Fitness Technology Translation Assessment (discussed in the Methods section), gives us a way to compare our heterogeneous collection of papers against a shared set of older adult fitness technology preference criteria. The use of the word translation emphasizes our approach of focusing on how these technologies will be translated from research into use in practice. We also used this assessment tool to assess gaps in the information collected from the assessment processes to inform best practices.

To our knowledge, no review has compared the design and assessment of existing research on older adult–oriented fitness technology interventions with the known needs and preferences of older adults in terms of technology acceptance and barriers and enablers to physical activity. In undertaking this scoping review, we hypothesized that by reviewing these papers against both their mechanisms of study and these factors for older adult engagement, we would get a clearer picture of the factors that may be leading to the kinds of results the reviews capture.

In the following sections, we first present the Related Work section to situate our work and our main research questions motivating this study. We then present our methodology, including the process of paper selection and assessment criteria. Results, their analysis, and discussion follow.

Building on the above related work, we review the literature by assessing health interventions across what is already known about older adult needs and concerns about fitness technologies and physical activity interventions. We specifically examine health technology interventions intended to help older adults build strength, improve physical fitness, and maintain independence. To assess how well these interventions align with the needs and preferences identified in older adult–specific technology acceptance studies, we have structured a review to answer the following questions:

Our review was conducted in compliance with the PRISMA-ScR (Preferred Reporting Items for Systematic Reviews and Meta-Analyses Extension for Scoping Reviews); see Multimedia Appendix 2 for the PRISMA-ScR checklist. The review approach consisted of, first, a search of older adult fitness technology research across the relevant disciplines, second, the selection of papers to be reviewed, third, the summarization of major technology types, intervention types, research goals, and major outcomes of the reviewed papers, fourth, the synthesis of the Older Adult Fitness Technology Translation Assessment tool from older adult’s needs and preferences from older adult–specific technology acceptance, barriers and enablers to older adult physical activity, and qualitative feedback from older adults on fitness technologies research, and, fifth, the comparison of both the design and assessment of the reviewed research papers against the synthesized assessment tool. The search methodology, selection process, summarization of technology types, research goals, and outcomes, synthesis of the Older Adult Fitness Technology Translation Assessment tool, and evaluation process are described in this section.

Work related to health technologies for older adults is present in a variety of different fields, including medicine, engineering, human-computer interaction (HCI), and psychology. Thus, the search strategy targeted the following databases that target these domains to obtain a full interdisciplinary establishment of the evidence in this research topic: (1) ACM Digital Library, for health technology intervention in the field of HCI; IEEE Xplore, for interventions in the field of engineering and computer science; Medline, for interventions in the medical field; and PsycINFO, for interventions in the field of psychology and public health.

The specific search terms in the review were chosen to collect research that related to older adults, used any form of digital health-related technologies, and aimed at building physical fitness or maintaining independence. The following terms and Boolean structure were used to account for the different vocabulary used across fields in this type of work and ensure that selected papers met the 3 requirements of a technology, intended for older adults, and about exercise or strength building: (1) “Older Adults” OR “Elderly”, OR “Elder Adults” OR “Geriatric” OR “Senior,” AND (2) “Digital Health” OR “Technology” OR “Interactive Systems” OR “eHealth” OR “mHealth” AND (3) “Strength Building” OR “Balance Training” OR “Physical Activity” OR “Exercise” OR “Exercise Programs” OR “training program.” Searches were also filtered to only include results from the last 10 years, research papers published in journals, and participants aged 60+ years in Medline. Search terms were adjusted to account for field-specific differences in databases. A full breakdown of the search strategy is included in Multimedia Appendix 3.

From the initial searches, an evaluation on the applicability of the papers was conducted by the research team against the following inclusion criteria:

Studies were included if they:

A detailed review of the full text of each paper meeting the initial inclusion criteria was conducted; aspects such as study design, sample size, intervention methods, and outcomes were evaluated. Studies were excluded if they lacked a clear connection between the technology used and the health intervention outcomes, or if the sample did not include older adults as defined by our criteria. Following this detailed review, a final list of studies meeting all the criteria was compiled.

Included in this review are (1) overall themes and categories of research in this space; (2) a summary of the common technologies, interaction types, and goals; (3) overall takeaways from intervention types and research methodologies; and (4) a comparison of the health interventions with known preferences of older adults from older adult–specific technology adoption work.

The 43 research studies were summarized by the journal they were published in, technological system, health goal, research goal, study target, number of participants, study duration, whether there was a control condition, and finally, the major outcomes presented in the work. A flowchart of the full selection process is shown in Figure 1. Results from this process are shown in Figure 2.

Overall, 16 of the review papers were submitted to medical journals, 15 were submitted to engineering journals, 7 were submitted to HCI or computing journals, 3 were submitted to public health journals, and 2 were submitted to a cross publication between computing and engineering journals. The research focus of the papers was usability, feasibility, and acceptability for 27 of the papers, studies of effects for 14, and design for 2 of the articles. A total of 13 of the 43 research papers were controlled studies. Further, 22 of the research papers were of studies longer than 4 weeks, 4 lasted between 1 and 4 weeks, 13 lasted only a single day, and 4 did not specify a study duration.

Interventions were also assessed for the number of technological components necessary for older adults to interact with the fitness interventions, an approximate measure of complexity, and a portion of similarity and compatibility factors. Roughly half, 21 of 43, of the interventions required older adults to interact with 1 or 2 technological components, and 38 of the 43 interventions required older adults to interact with 4 or fewer technological components, as assessed by reviewers. One health intervention required older adults with 10 different identified technologies, and another with 12 [66,71].

An additional investigation into the presence of co-design, a design methodology shown to reduce assumptions and better align design choices with the preferences of potential users [87], found that only 13 of the 43 papers mentioned any use of co-design. The types of co-design found include focus groups, user experience interviews, and specified user-centric design processes.

Overall, the results of this scoping review illuminate 3 main opportunities for growth in the field, better alignment between the studied needs of older adults around technology interventions and the current technology interventions for research around technology for older adult fitness, better translations of the findings of older adult design work to the designs in practice, and more explicit usage of older adult–specific considerations in the research methodologies for this growing area of research.

From the Older Adult Fitness Technology Translation Assessment, none of the interventions scored a 3 or above on all 8 factors as assessed by our reviewers, suggesting that there was a least 1 easily identified older adult–specific translation problem for every single one of the included interventions as compared to the understood needs and preferences of older adults. The overall middling averages across all the papers considered, and the lack of a consistent pattern across technologies, interactions, and goals between interventions, point to differing strengths and widespread weaknesses. The results from our assessment also point to a need to learn from other intervention designs and interdisciplinary works to better incorporate older adult–specific needs into the designs of older adult fitness technology interventions. The alignment of new designs with the identified factors around older adult preferences has the potential to afford better effects and increase potential for uptake with minimal changes to designs. For example, concerns around compatibility, similarity, and privacy include the space and cost requirements of technologies such as Microsoft Kinect, the potential daily life disruption of strictly regimented exercise plans, the uptake concerns around complex and unapproachable technology systems, and the unease with systems that require 24/7 monitoring, such as fitness wearables. The lack of alignment found in this review was more tied to the contextual details of the overall fitness technology system design rather than the specific technology types themselves. The research team recommends caution when adding additional secondary features, technologies, or data collection that is not directly tied to the specific older adult health goal, as it can run the risk of adding complexity of use for older adults and may introduce resistance based on privacy concerns. Additionally, we recommend a focus on minimal disruption and flexibility of use, including independent and open intervention plans that are more easily incorporated into routines and simple technology interactions with minimal space requirements and easy uptake. Small changes to intervention design could make a significant difference in the experience for older adults. For example, a wearable and app-based intervention to increase step count could, instead of requiring 24/7 monitoring, tie daily goals to only additional “exercise steps” and emphasize to older adults that they only need to wear it when they feel comfortable doing so. Additionally, a Microsoft Kinect-based exercise program could instead opt for a follow-along video-based intervention. What may be lost in interaction and progress monitoring may be gained through the reduction in space requirements and the overall simplicity. In summary, almost any technology or approach can work if it fits with and respects the older adult preference criteria. Drawing on awareness of older adult technology concerns, potential translation problems in the future of even the most promising interventions could have been identified and adjusted for early in the design process.

To better understand the reason for the middling results, reviewers looked to the presence of referenced older adult–specific design work in the reviewed research papers. If very few of the research papers referenced older adult–specific design work, then it could point to a lack of awareness in the field as the reason for the mixed results. Of the 43 research papers, only 16 explicitly cited older adult–specific design works, and only 11 directly referenced those works as part of the fitness technology design. Furthermore, in all the referenced older adult–specific design works, none of them encompassed all the considerations outlined in the Older Adult–Specific Technology Translation Assessment, with 5 of the research only citing works relating to the physical limitations of older adults. This points to inconsistent consideration of the full scope of older adults’ needs and preferences related to fitness technologies.

However, for the research papers that did cite older adult–specific design works, reviewers also found potential evidence of a lack of translation of the outcomes from the cited works and the design choices of the reviewed papers. Only 6 of the reviewed fitness technology interventions fully aligned with the stated outcomes in the older adult design works. An example of this disconnect between the use of older adult–specific design research and the proposed interventions is illustrated in the work of Ogonowski et al [43], chosen as an exemplar because it follows the design approach mostly closely related to what is being proposed as effective in this review. That is, it cited older adult–specific design considerations in the background, design, and methodology, and used participatory research or co-design with older adults. In this example, viewing only the older adult design work cited as forming the system requirements for their fitness technology intervention, a disconnect was found. The cited older adult–specific design work by Meurer and Wieching [100] found a preference against the online social networks and menus with “lots of other accessories and other extras,” and a preference for enabling in-person social connections and simple, easy-to-use technology interactions. Despite those identified preferences, the resulting technology included a social media platform for participants in which participants could meet other users virtually and post training results, news, questions, and suggestions [43]. When comparing it to the known research surrounding the needs and preferences of older adults using the Older Adult Fitness Technology Translation Assessment tool, compatibility concerns were raised over the space requirements of the technology, and privacy concerns were raised over the continuous monitoring of older adults through the senior mobility monitor.

In conducting the scoping review, many of the questions from the Older Adult Fitness Technology Translation Assessment remained unanswered from the published outcomes of the reviewed studies, such as how well the technologies fit into older adults’ routines, how effective the interventions were over the long term, and how effective the interventions were outside of a research setting. The immediate implication of this is that it was hard to predict how well the technology interventions would translate to older adult use in practice. For technology interventions aimed to be used in older adults’ homes, in community centers, as part of public health initiatives, and in medical interventions, ethically, the implications of use in practice should be considered, including older adult–specific needs and preferences to avoid wasting funding and increase the likelihood of success beyond research environments. The reported outcomes from the reviewed studies include strength gains, balance improvements, increased minutes of physical activity or daily steps, perceived usefulness and utility, usability, functionality, and general qualitative feedback from older adults. However practical considerations for older adults remain unanswered, such as how well does the proposed technology intervention fit into older adults’ daily lives, routines, and space at home, do older adults have any privacy concerns when using the intervention, how similar is the interaction to something they have performed before, and does the intervention promote a feeling of dignity and independence for older adults.

Asking these questions can help progress the field by building insights for how older adults feel about different fitness technologies and intervention types, and supporting the overall older adult–specific fitness technology factors with direct outcomes from the field. For example, which types of social engagement systems work best, which intervention types are more likely to fit into routines, and which technologies are older adults more willing to engage with. Implementing explicit inclusion of older adult–specific factors into the assessment methodologies of interventions in the field could help to address this gap and avoid potential pitfalls that may otherwise only be found once used in practice.

This scoping review has considered papers across a range of fields that all have in common presenting novel designs of or novel use of digital technologies to support building or sustaining fitness for older adults. To quantify factors across these papers that may be affecting the efficacy of this research as modeled across the other cited reviews in this domain, we have organized the included papers against several features, such as types of technology used, types of evaluation carried out. We also developed novel assessment criteria based on research literature related to older adult technology acceptance and older adult physical activity uptake. We used the resulting 6 factors as a further way to assess design efficacy.

Overall, as highlighted in the analysis of the results above, while there is awareness of design-specific guidance for older adult technology uptake, there is strong evidence of a disconnect between known older adult preferences and the design and assessment of proposed older adult fitness technologies in the field. This evidence was present across all of the reviewed research domains, including medical, engineering, and HCI.

As we underscore in point 3 in the discussion, this disconnection has implications for the development and delivery of ethical research. It means that participants’ time may be wasted, not least among potentially more vulnerable participant populations of older adults. Considering the needs or preferences of a particular demographic is standard practice to inform design. That there is both a rich literature to draw on around older adult technology adoption and around barriers to fitness uptake, and that this work is largely absent from work around older adult technology adoption for fitness, has implications for the appropriate use of research funding, when known affordances and constraints are not explicitly considered.

Overall, this review contributes the following, an analysis of the common strategies of older adult fitness technology interventions and the strengths and weaknesses in the field and a synthesis of the needs and preferences of older adults from older adult technology acceptance research, barriers and enablers to older adult physical activity, and qualitative data from older adult fitness technology research into an assessment tool called the Older Adult Fitness Technology Translation Assessment Tool.

We proposed a novel rubric to assess across the 6 older adult–specific criteria how well the design and assessment of the presented older adult fitness technology systems align with the known needs and preferences of older adults in engaging with fitness technologies. Our goal has been to create a mechanism that enables very different technologies and approaches to still be compared against critical, meaningful criteria. This analysis does show an overall inconsistent consideration of any of our 6 older adult–sensitive criteria. This finding suggests that if not causation, then a likely strong correlation exists between the types of middling effects captured in the existing review papers and systemic limited attention to older adult–sensitive design and evaluation criteria. Considering the topic area is to support older adult adoption, via technology, of sustainable fitness practices, this lack of attention to well-established older adult technology dynamics is surprising. In general, the results of this review suggest that there is a significant opportunity for improving and revisiting how to best use the potential of these technologies to impact older adults’ daily lives and well-being positively by starting the design process with the known preferences and constraints of older adults and adjusting our assessment processes to explicitly consider older adult–specific factors beyond usefulness and usability.

A limitation for our analysis is the lack of detailed information on the older adult interaction with the technologies and the lack of outcome metrics that are directly related to the factors included in the Older Adult Fitness Technology Translation Assessment tool. Thus, reviewers needed to use their best judgment from the information provided in the published research papers. Second, due to the cross-field nature of these technologies, the keywords used in the search strategy may not have accounted for all the possible older adult fitness technologies across all the relevant fields. Future work is planned to validate the Older Adult Fitness Technology Translation Assessment tool for novel older adult fitness technology design and assessment, including developing and testing a design framework based on the Older Adult Fitness Technology Translation Assessment tool with a novel fitness technology prototype. The assessment will include an initial multistage iterative trial and a follow-up 6-month randomized controlled trial to assess the effects of using the framework, measuring adherence, health outcomes, and older adult–specific qualitative feedback. We hypothesize that more explicit consideration of older adult–sensitive criteria using the Older Adult Fitness Technology Translation Assessment tool, technologies can reduce caregiver burden, lower health care costs by delaying functional decline, and enhance quality of life. Future work is necessary for informing best practices for both the design and assessment of fitness technologies for older adults. An investigation into the role of co-design in aligning fitness technologies with older adult preferences, the best use of older adult fitness technology preference research in design, and a protocol for building toward technology readiness is planned.

A straightforward takeaway from this review is that more explicit inclusion of older adult–specific factors in design and assessment is worth testing to evaluate how these established factors’ integration can support the development of more inclusive and impactful older adult fitness technologies. By integrating the older adult–specific factors summarized in the Older Adult Fitness Technology Translation Assessment into design and assessment methodologies, we hypothesize that we will be much better able to assess the efficacy of the interventions in practice. Furthermore, we hypothesize that it can help bridge the gap between technological capability and real-world applicability, ultimately fostering greater adoption, respect for participants, and long-term success in supporting older adults’ health and well-being.