Physical activity (PA) in older adults is crucial for the prevention of major chronic noncommunicable diseases and for the improvement or maintenance of mobility, independence, and quality of life [1-3]. Although the health benefits of PA are well established, the prevalence of insufficient PA among older adults is high and points to a considerable scope and need for improvement [4-7]. Therefore, interventions are needed to encourage older adults to initiate and maintain regular PA [8].

In the past 2 decades, PA interventions have increasingly been incorporating technological applications because these may help increase motivation and adherence among participants, (remotely) measure and monitor (changes in) intervention outcomes, and provide feedback about this to participants [9-11]. Previous reviews and meta-analyses have shown that technological applications such as websites, mobile or wearable devices, smartphone apps, and virtual reality have been reported to support PA in the older adult population [9,11-16].

In parallel, the importance of applying socioecological approaches to advancing the understanding of PA determinants, including those from different environmental domains, has increasingly been acknowledged [8,17]. Previous research has emphasized that a variety of physical, social, socioeconomic, and systemic environmental factors play a crucial role in facilitating or hindering PA in older adults [18-20]. Therefore, environmental factors have also been considered increasingly in PA interventions among older adults [8,17]. For instance, one study indicated that a PA intervention was more effective in maintaining or increasing older adults’ PA when implemented in more walkable neighborhoods that are characterized by higher levels of residential density, land use mix, and intersection density [21]. In addition to characteristics of the physical environment (eg, walkability) [8,21], previous studies have also indicated that aspects of the social (eg, receiving social support) [22], socioeconomic (eg, area-level income) [23], and systemic (eg, ethnicity and climate) [24,25] environment are important to consider when implementing PA interventions among (older) adults. However, environmental factors have rarely been investigated regarding their impact on adherence and effectiveness of technology-assisted PA interventions [10,26]. The rapidly emerging integration of both technological applications and environmental factors in PA interventions in the recent past justifies and at the same time highlights the need for an overarching investigation.

Against this background, this scoping review compiled the current literature and provided an overview of the role of physical, social, socioeconomic, and systemic environmental factors in technology-assisted PA interventions for older adults. In this way, we provided new knowledge on the specific environmental factors that have been considered in such previous interventions and investigated whether these factors are associated with adherence to and outcomes of interventions. Furthermore, we aimed to increase insights into how environmental factors may modify outcomes, how they are affected by technology-assisted PA interventions, or how they might be part of underlying mechanisms of this type of interventions. As a consequence of the work at hand, health care professionals, policy makers, and researchers may be enabled to better design effective technology-assisted PA interventions for the target group of older adults.

A search query was carried out on March 16, 2023, and updated on September 16, 2024. It was run in 6 bibliographic databases: CINAHL, Embase, MEDLINE, PsycINFO, Scopus, and Web of Science. The search algorithm was built using search terms based on definitions and synonyms of intervention types; technologies; the target group (ie, older adults); and the physical, social, socioeconomic, and systemic environment and its attributes. The search was not limited to a specific time frame. A detailed search strategy for each bibliographic database can be found in Multimedia Appendix 2. The reference lists of the included articles were screened to identify additional eligible papers.

In this scoping review, only published articles in the form of original, longitudinal intervention studies with at least one preintervention measurement and one postintervention measurement and with any health-, function-, or behavior-related outcome were included. Of those articles, only papers were selected that included (1) independently living, community-dwelling older adults aged ≥60 years (ie, study sample average age of ≥60 years with a minimum individual age of 50 years) regardless of their health status; (2) an intervention period involving PA (components); (3) a PA intervention with or without a control setting that was based on or aided or strengthened by technology applications or technological components; (4) an assessment of physical, social, socioeconomic, or systemic environmental factors or a specific comparison of groups with different environmental conditions; and (5) a report of associations of environmental factors with adherence to or outcomes of the PA intervention or whether environmental factors were included as an outcome. The language restriction was set to English, Dutch, German, French, Finnish, Latvian, Norwegian, Swedish, and Turkish because these languages were spoken by the review team. Articles that were literature reviews, study protocols, conference proceedings, or abstracts only were excluded.

After removing duplicate records in EndNote (version 20.0; Clarivate Analytics), the title and abstract of each record were independently screened by 2 reviewers out of the pool of reviewers (ie, all authors of this scoping review). To increase consistency between reviewers, the procedure was discussed within the group of reviewers, who first screened a set of 30 records as a pilot test and discussed results before initiating the full-text screening. Subsequently, the full text of each potentially relevant paper was independently assessed for eligibility by 2 reviewers. Any disagreement between reviewers was resolved through discussion or, if no consensus could be reached, through discussion with a third reviewer. The screening of titles, abstracts, and full texts was conducted using the Rayyan software (Qatar Computing Research Institute), a noncommercial web-based application [29].

The data extraction from each eligible paper was performed by 2 reviewers per manuscript using a predefined standardized extraction form in Microsoft Excel (Microsoft Corp). Both reviewers then compared their results and harmonized their findings. Any data extraction issues identified by the reviewers were resolved through group discussion. For each eligible paper, information was extracted on (1) study characteristics (ie, full reference, country in which the study was conducted, and main study objectives); (2) methodological aspects of the study (ie, sample size, study sample characteristics, and study design); (3) intervention components (eg, PA component, duration and frequency of PA or exercise sessions, and supervision of sessions); (4) technology components (ie, technology devices used); (5) environmental factors (ie, type of environmental factor included, such as physical, social, socioeconomic, or systemic), an assessment tool for environmental factors, specific environmental characteristics assessed, and the role of an environmental factor (eg, outcome, a feature of the study design, or a factor with a modifying effect on the intervention outcome); (6) study outcomes (ie, primary and secondary outcomes, outcome measurements, and main study findings); and (7) main study findings in relation to the environmental factors assessed. The reviewers did not contact the authors of the eligible articles to collect unreported data or additional details.

In this scoping review, data from the included studies were analyzed through a stepwise process. First, study characteristics were described using descriptive statistics. Second, the environmental factors and their role in technology-assisted PA interventions were described per environmental domain. Third, intervention outcomes were described by each type of environmental factor, and the actual effects of or on environmental factors were summarized.

A total of 12,228 articles were identified from CINAHL (n=703, 5.75%), Embase (n=2322, 18.99%), MEDLINE (n=1497, 12.24%), PsycINFO (n=679, 5.55%), Scopus (n=3951, 32.31%), and Web of Science (n=3076, 25.16%). After removing 34.41% (4208/12,228) of duplicates, 65.59% (8020/12,228) of the articles were included in the title and abstract screening phase. After this phase, of the 8020 included articles, 266 (3.32%) were screened for eligibility in the full-text screening phase. In total, 25 articles met the inclusion criteria and were included in this scoping review (Figure 1) [30-54].

An overview of the characteristics of all 25 included studies is presented in Table 1. The papers were published between 2011 and 2024, and most articles (15/25, 60%) were published between 2019 and 2024. Of the included studies, most (11/25, 44%) were conducted in Europe [33,36,37,39,41,44,45, 47,50-52], followed by North America (5/25, 20%) [32,42,46,49,53], Asia (5/25, 20%) [30,34,35,43,48], and Oceania (4/25, 16%) [31,38,40,54]. The 2 study designs that were most often used across the studies were pretest-posttest (12/25, 48%) [30,32,33,36,41,43,44,46,48,50,52,54] and randomized (controlled) trials (8/25, 32%) [31,34,36-39,41,46]. The sample sizes in the included studies ranged from 1 to 409 (mean 85.7, SD 98.9 participants), and the percentage of women in the study samples ranged from 0% to 100%. Across all the included studies, the lowest and highest reported age of an individual was 50 and 99 years, respectively. The PA components that were most often considered in the technology-assisted PA interventions were (treadmill) walking (8/25, 32%) [30,36,40,42,45,49,53,54] and balance and coordination exercises (8/25, 32%) [34,35,38,44,48,51-53], followed by stretching and flexibility (7/25, 28%) [31,33,34,40,43,50,51] and functional training (5/25, 20%) [33,34,36,51,54]. Of the 25 included studies, 18 (72%) reported supervised PA interventions [34-45,47-49,51,53,54], 4 (16%) included a nonsupervised PA intervention [31,48,52,54], and 3 (12%) did not provide information regarding the supervision of the technology-assisted PA intervention [30,32,33]. The duration of these PA interventions, the frequency and duration of individual intervention sessions, and the technology device used varied across the studies (Table 1).

An overview of the environmental factors and their role in technology-assisted PA interventions for older adults is presented in Table 2. Of the 25 included studies, 3 (12%) included aspects of multiple environmental domains [41,47,51]. The environmental domain that was most often considered in the included studies was the social environment (19/25, 76%) [30-34,36,40-48,50-53], followed by the physical environment (8/25, 32%) [35,37-39,47,49,51,54], the socioeconomic environment (1/25, 4%) [41], and the systemic environment (1/25, 4%) [41].

Several specific social environmental factors were considered in the various technology-assisted PA interventions across the studies, including social connectedness [30,48,51] (including loneliness [30,33,46]), social interaction [34,43,50,52,53], social support [31,33,36,44,45,47], and delivery aspects (eg, home environment vs gym environment or the involvement of virtual advisors vs human advisors) [32,40,42]. Furthermore, a variety of specific physical environmental factors were also considered, including aspects of the street-crossing environment (eg, traffic speed) [37,49] and the neighborhood built environment (eg, walking paths; presence of benches; and hot spots, ie, highly frequented and meaningful nearby places, such as weekly markets and parks) [38,39,47,49,51,54]. A range of social (eg, household composition), socioeconomic (eg, costs of treatments), and systemic (eg, local preferences on the quality of patient care) environmental factors were considered in one study [41]. In total, 16% (4/25) of the studies were conducted in a simulated or virtual reality environment and included aspects of the physical environment [35,37,38,49]. The tools used to assess the environmental factors differed across the studies (Table 2).

The specific environmental factors fulfilled various roles in the technology-assisted PA interventions in the included studies. A total of 32% (8/25) of the studies included an environmental factor as an outcome. Of these 8 studies, 7 (88%) focused on a social environmental factor [30,34,36,43,45,46,48] and 1 (12%) focused on a physical environmental factor [39]. In 28% (7/25) of the studies, social [40,42,48,51] and physical [35,37,38,47] environmental factors were used as a comparator of the intervention; that is, different environmental backgrounds were compared, such as a virtual program and a face-to-face program, without direct measurement of environmental factors. In another 32% (8/25) of the studies, environmental factors were used as study results or factors influencing the intervention [31,42,44,46,50,51,53,54]. In 12% (3/25) of the studies, social [32,33] and physical [49] environmental factors were used as components of the intervention.

As described in Table 2, the technology-assisted PA interventions in most studies (19/25, 76%) were evaluated positively [30,32-38,41-43,45,48-54]. The included environmental factors played a supportive role in achieving this effect in several of these studies in which a between-group comparison was made [33-38,42]. In some studies, no effects (3/25, 12%) [31,39,40], mixed effects (2/25, 8%) [44,47], or adverse effects (1/25, 4%) of the interventions or environmental factors (eg, moderators) were reported [46].

The studies that considered social environmental factors in the technology-assisted PA interventions (12/25, 48%) focused on a variety of outcomes, including PA and exercise [30-32,46]; mental health outcomes [30,32,46]; quality of life [34,40]; physical performance [52,53]; physical health outcomes [42]; social interaction network [43]; and aspects such as delivery, acceptability, usability, adherence, attrition, satisfaction, experiences, and motivation [31,33,36,44,45,48,50,52]. The studies that considered physical environmental factors (7/25, 28%) also focused on various outcomes, including cognition [35,39], physical performance [35,38,39,49,51,54], adherence [38], and street-crossing decision-making [37]. The 12% (3/25) of the studies that considered aspects of multiple environmental domains also focused on various outcomes, including PA [47,51]; mental health [41,47]; cognition [41]; physical performance [51]; costs [41]; and aspects such as satisfaction, commitment, experience, and acceptability [47].

In several studies (5/25, 20%), the technology-assisted PA interventions improved social environmental factors, such as social connectedness [30], social interaction networks [43], and other societal factors [34]. In one study, the technology-assisted PA intervention did not change social isolation in older adults [40]. In the study by Pauly et al [46], more frequent use of the social component of an app was associated with more loneliness in older adults. In various studies (7/25, 28%), social interaction and support were found to increase the exercise effort in more competitive but not in less competitive older adults [32], as well as adherence [36,50,52,53] and the positive experiences [48] of older adults in technology-assisted PA interventions. In contrast, in the study by Alley et al [31], older adults with lower baseline levels of social support increased their moderate to vigorous PA more than those with higher levels of social support.

In various studies (5/25, 20%), physical environmental factors were considered in a virtual or simulated environment [35,37,38,49]. Compared to usual home exercises, virtual kayak paddling was beneficial for balance, muscle performance, and cognition [35]. In a simulated street-crossing situation, the intervention group improved street-crossing decisions compared to controls, but these differences disappeared 6 months after training [37]. On postintervention tests, the intervention group and the control group both made more unsafe decisions when a car approached at a high speed and missed more crossing opportunities when a car was approaching at a low speed compared to the preintervention test [37]. In the study by Duque et al [38], there were significantly fewer falls and lower levels of fear of falling in older adults who received balance training in a virtual environment compared to a control group that received usual care. The study by Richards et al [49] showed that virtual reality training was superior to the control protocol training for improving motor capacity, balance self-efficacy, and anticipatory locomotor adjustments.

One study showed that a technology-assisted PA intervention that made use of features of the outdoor built environment (community hot spots such as parks and markets) improved physical function in older adults [51]. Through the definition of hot spots in the neighborhood, older adults were encouraged to carry out certain activities with their peers when meeting there. However, other technology-assisted PA interventions that considered such physical environmental factors did not significantly affect functional mobility [39] or only showed limited effectiveness on PA outcomes [47]. The study by Jurkeviciute et al [41] indicated that several social, socioeconomic, and systemic environmental factors influenced the effects of an eHealth intervention focusing on cognitive performance, mental health outcomes, satisfaction, and costs.

On the basis of our findings, it becomes apparent that there is a significant potential for a better understanding of intervention results and better tailoring of intervention programs when including environmental aspects in research endeavors. Environmental factors demonstrated multifaceted effects on intervention outcomes, albeit sometimes contradictory. A thorough understanding of the underlying mechanisms is still lacking as the use and applications of these environmental factors remain scarce. Future research should elucidate the causal pathways through which environmental factors exert their effects, considering potential mediators and moderators that may influence intervention outcomes. In particular, social aspects related to intervention delivery and group dynamics seem to have the potential to reveal important mechanisms that could positively enhance technology-assisted PA interventions for older adults. The same applies to the physical environment. We found that physical and even cognitive capacity may benefit largely from exercise in virtual or simulated environments, showing the large potential for augmented and virtual reality exercises. As this is a field growing at high speed, further work should be carried out to define more specific methodologies and target outcomes especially for the population of older adults. Considering that young and middle-aged adults are using these kinds of technologies more and more, the potential for interventions targeted at PA will be enormous. As such technologies become increasingly accessible and user-friendly, their integration into PA interventions holds promise for enhancing engagement and adherence and obtaining positive outcomes among older adults. As older adults are not left out of these developments and show more acceptance of such technologies [62], intervention development can be expected to be more inclusive of virtual environments in the future.

We followed a well-defined methodology for this scoping review. Although we conducted a detailed search in various established databases, we might have missed potentially relevant papers (eg, from gray literature or conference proceedings not yet published as full papers). It also has to be acknowledged that there is no clear framework for categorizing environmental factors into specific domains (ie, social, physical, socioeconomic, and systemic). The framework we decided to use was based on thorough discussions within the group of researchers involved. We chose not to evaluate the quality of the studies included in this scoping review as the objective of this work was to explore the role of environmental factors in technology-assisted PA interventions among older adults and not to assess the quality of the studies. As a result, the study findings need to be interpreted with caution, especially those of studies without randomization procedures or controls. The comprehensive summary of methods in Table 1 provides all the information in this regard. As technology-assisted PA interventions can be expected to grow in number and application in the following years, evaluating the quality of the evidence is undoubtedly warranted at some point in the future.

In conclusion, the results of this scoping review addressing the fast-growing domain of technology-assisted PA interventions for older adults show that the role of environmental factors is still emerging in this field. The studies predominantly focused on social environmental factors, followed by physical environmental factors. Studies that integrate socioeconomic and systemic environmental factors in technology-assisted PA interventions were scarce. Important findings were that the included environmental factors played a supportive role in achieving beneficial effects of technology-assisted PA interventions. The studies reviewed exhibited heterogeneity in how environmental factors were incorporated—some studies incorporated them as integral components of the experimental design, whereas in other studies, they served as effect modifiers or outcomes. Drawing from the results, there is a significant potential for a better understanding of intervention outcomes and better tailoring of intervention programs when systematically including environmental aspects in technology-assisted PA interventions for older adults.