The reviewed protocol was developed and approved by all authors prior to initiation. It complies with all recommendations from the systematic review, strictly adhering to the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) checklist and Meta-Analysis of Observational Studies in Epidemiology guidelines [13]. Literature searches were performed across the following databases: PubMed, Web of Science, Embase, CINAHL, Scopus, and Wiley Online Library. The search period spanned from each database’s inception until May 20, 2025. This study was registered with the International Prospective Register of Systematic Reviews (PROSPERO; registration number CRD420251056689).

In most studies on gamification, younger individuals are typically the primary target group for game-based interventions due to their generally higher sense of self-efficacy and greater experience with digital gaming. In this context, the study initially focused on the research subjects and intervention measures but overlooked other retrieval elements when designing the retrieval strategy. However, during the subsequent literature screening, these overlooked elements will be reassessed and evaluated. To maximize the breadth of the search, the study selected the keywords “gamification,” “game elements,” and “elderly” and refined the retrieval strategy by incorporating a combination of controlled vocabulary, natural language terms, and synonyms, in accordance with the requirements of different databases. An additional file shows the search strategy in more detail (Multimedia Appendix 1).

To ensure that the analyzed literature aligns with the objectives and to mitigate the impact of conceptual confusion on the validity of the research results, this review clarifies the distinction between serious games and gamification, incorporating this differentiation throughout the literature selection process. Gamification refers to the integration of game elements, such as points, leaderboards, and progress bars, into nongame environments to enhance user engagement and motivation through game-like experiences, without altering the nature of the tasks. For example, in rehabilitation training for older adults, adding a reward system or virtual progress bar can extend training time [14].

Serious games, on the other hand, represent a comprehensive design approach that provides an immersive gaming experience, with the primary goal of education, training, or therapy, rather than entertainment. They are sometimes referred to as purpose-driven games (eg, virtual reality rehabilitation training games for older adults) [15,16]. However, the distinction between these 2 concepts may be ambiguous and subjective. When these concepts cannot be clearly differentiated, the research team engages in discussion and makes a final decision.

The studies included in this systematic review must meet the following criteria: (1) studies targeting older adults with a mean age ≥60 years and minimum age threshold aligned with World Health Organization functional aging criteria (typically ≥55 y). When the exact minimum age was unreported, statistical estimation (mean − 2SD) confirmed ≥90% of the participants met age criteria. (2) The use of gamification or at least 1 game element is clearly specified. Serious games (eg, video games, immersive virtual reality games, and augmented reality games) or studies that merely mention gamification without analysis are excluded. (3) Gamification is implemented through digital devices such as tablets, smartphones, or wearables. (4) At least 1 change related to physical or social activities is described, which may be presented as subjective self-reports or objective measurement indicators. (5) The study is original empirical research (must be experimental in nature). Review articles, research proposals, guidelines, policy documents, and research programs are excluded. (6) Only peer-reviewed, published papers in English are included, including doctoral dissertations and conference papers.

The research team conducted quality assessments of the studies that met the inclusion criteria, following the quality assessment standards established by the Joanna Briggs Institute [17,18]. There are 13 criteria for evaluating randomized controlled trials (RCTs) and 9 criteria for evaluating quasi-experimental studies. Studies that meet at least 50% of the assessment criteria are accepted for review. For RCTs, the research team decided to exclude the blinding of researchers and participants from the quality assessment, as blinding is considered impractical in the context of gamified interventions. Any disputes are resolved through discussions with a third-party investigator to reach a final consensus.

Two researchers independently extracted data from the selected studies using standardized forms based on the content of the literature. The extracted data included the first author’s name, publication year, country, study design, subject characteristics (sample size, mean age, and gender distribution), intervention characteristics (study setting, type of equipment, and duration), gamification characteristics (game name, game elements, and theories used), primary and secondary outcomes related to physical activity, and a summary of the research conclusions. Any disputes were resolved through discussions with a third-party investigator to reach a final consensus.

Due to substantial methodological and reporting differences that prevented quantitative data synthesis, we conducted a narrative synthesis organized by the PICO framework to systematically summarize the findings of the studies.

To systematically address heterogeneity in gamification elements and technological platforms, we developed a dual-axis classification framework. The first dimension categorizes interventions by gamification complexity, distinguishing basic designs employing 1 or 2 core game elements from intermediate configurations integrating 3 or 4 elements, and advanced systems incorporating 5 or more interactive components. The second dimension classifies technological integration into standalone mobile apps, hybrid systems combining mobile apps with wearable devices, and wearable-centric approaches relying primarily on sensor technology. This framework enables structured cross-study comparisons while acknowledging inherent methodological variations.

A total of 6333 records were identified. After removing 2077 duplicate entries using EndNote, the remaining 4256 records were screened by title and abstract. Following rigorous assessment, a substantial number of records were excluded, primarily due to ineligible age, nonphysical activity focus, or the others (Multimedia Appendix 2). Subsequently, 21 articles were initially included. Reviews, clinical record comparisons, case reports, and other literature that did not meet the inclusion criteria were excluded based on their titles and abstracts. Following a more detailed examination of the full text, 13 articles were further excluded for not meeting the inclusion criteria. An additional file shows this in more detail (Multimedia Appendix 3). Ultimately, 8 studies [14,19-25] were included in this review. The search results are presented in Figure 1.

The studies selected for this review were conducted between 2018 and 2024 in Sweden [14], Portugal [19], Japan [20,21,24], Brazil [21], and the United States [23,25]. At least 5 studies [14,22-25] were randomized controlled trials, while 3 studies [19-21] were quasi-experimental studies. Sample sizes ranged from 12 to 1062 participants, with a total of 1454 participants. The average age of participants ranged from 62 to 81.5 years, with a higher percentage of women. At least 4 studies [20-22,24] included community-dwelling older adults, 1 study [14] included participants with musculoskeletal conditions, 1 study [23] included participants at risk of cardiovascular disease, 1 study [25] included participants at risk of Alzheimer disease, and 1 study [19] did not specify participant characteristics or the gender ratio. A summary of the participants, measures, interventions, controls, and outcomes is presented in Table 1.

The quality assessment scores for the RCT studies ranged from 7 to 10 points (Table 2). All 5 studies employed randomization; however, the concealment of treatment group allocation was unclear in 3 studies. Due to the nature of the intervention, blinding of participants or personnel was not possible. In 3 studies [23-25], outcome assessors were blinded to the treatment allocation. Control of confounding factors was unclear in 1 study [14]. The average score was 9/11, indicating that all the RCT studies demonstrated high methodological quality.

The 3 quasi-experimental studies each received a score of 6 points (Table 3) for methodological quality. In 1 study [21], the age of participants in the 2 groups was not similar, and the characteristics of health status and technical proficiency were not clearly presented. The criteria for this assessment were not explicit. None of the studies reported in detail whether participants in the intervention group received any other care or treatments besides the intervention. At least 2 studies [19,20] employed before-and-after designs without an independent control group. Another 2 studies [20,21] had incomplete follow-ups, and although the reasons for dropout were stated, the impact of dropout on the results was not analyzed. At least 1 study [20] only reported descriptive statistics (means, standard deviations) and did not perform significance testing or calculate effect sizes. The statistical methods used were relatively simple.

Table 1 and Multimedia Appendix 4 display the gamification features of the studies included in our systematic review. The number of game elements used in the gamification interventions for older adults' physical activity ranged from 2 to 6, with the majority incorporating 5 game elements. The most commonly used game elements were goals and rewards, followed by challenges and collaboration, while virtual identity and points were the least used.

Half of the 8 studies [19,23-25] employed theories or principles to design the gamification interventions. At least 2 studies [23,25] conducted in the United States used prospect theory from behavioral economics, while the other 2 studies employed the theory of behavioral change [24] and the emotional circumplex model [19]. All of these studies used a single theory without combining them.

In 1 study, family members or friends were incorporated into the gamified intervention as support partners to provide encouragement and assist participants in achieving their activity goals.

There are 3 digital approaches to gamification for older adults. Most studies [19-22] use mobile apps only (4/8), while some studies [14,24,25] combine mobile apps with activity monitors (3/8), and only 1 study [23] uses activity monitors alone (eg, Stepwatch or Fitbit). The shortest intervention time was 1 week [19], and the longest was 12 months [23]. Only 3 studies [18,23,25] included a follow-up period, highlighting the need for further evaluation of gamification interventions to assess their long-term sustainability. Another 5 studies [14,21,22,24,25] compared a gamification intervention to usual care, while 2 studies [19,20] did not include a control group. The study by Fanaroff et al [23] assigned participants to 4 groups and compared the effects of gamification to financial incentives, a combined intervention of both, and explored the impact of gamification on physical activity in older adults.

The included studies all assessed the effect of gamification interventions on the number of steps taken by older adults in their daily lives, as measured by activity monitors or smartphone sensors such as Google Fit. The findings from both controlled and single-arm studies were largely consistent in confirming the positive impact of gamification interventions on step count, including results observed during the follow-up period. The only exception was the study by Kawaguchi et al [24], which reported a nonsignificant increase in the number of steps following the gamification intervention. However, this study did report that intervention participants had a higher frequency of participation in specific types of activities, particularly hobbies and culture-related activities, compared to the control group.

At least 2 RCTs [23,25] reported changes in the time spent in moderate-to-vigorous physical activity (MVPA) per day, as measured by activity monitors. After adjusting for factors such as baseline activity levels, proficiency with the devices, and genetic risk, between-group differences indicated that gamification can significantly increase the duration of MVPA in older adults, with these effects being sustainable.

Application of the classification framework revealed consistent patterns across the 8 included studies. Advanced gamification designs consistently demonstrated superior outcomes [23,25], particularly when integrated with hybrid technology systems, showing statistically significant improvements in both step counts and MVPA duration [23,25]. Basic gamification approaches proved effective for step count enhancement [20,21] but exhibited limited capacity for tracking complex activity metrics [19]. Standalone mobile apps demonstrated strengths in facilitating social interaction mechanisms [20,22], while wearable-centric implementations may suffer from scalability limitations [14]. Crucially, all advanced interventions incorporated theoretical frameworks [19,23-25], whereas basic and intermediate designs operated predominantly without explicit theoretical foundations [20-22] (Table 4).

Effect strength was evaluated through standardized criteria: strong, >15% improvement from baseline OR Cohen d >0.8 OR η²>0.25 with P<.05; moderate, 5%‐15% improvement OR Cohen d 0.5‐0.8 OR η² 0.10‐0.25 with P<.05; weak, <5% improvement OR Cohen d <0.5 OR η² <0.10 with P<.05, NS, nonsignificant.

The present review included only 8 full-text articles that met stringent inclusion criteria, reflecting both methodological rigor and domain-specific constraints. Strict adherence to gamification standards, mandatory mHealth components, and a focused eligibility criterion of adults aged 60 years and above resulted in the exclusion of a considerable proportion of initially identified records. This selectivity contrasts sharply with the broad application of gamification in general healthcare, which has largely targeted younger populations [11]. Nonetheless, this condensed evidence base captures a developing nexus among gerontechnology, behavioral engagement, and mHealth—an area that remains understudied yet highly relevant for physical activity promotion in older adults. Despite these constraints, the main findings indicate that mHealth-based gamified interventions offer meaningful benefits in improving physical activity levels within this demographic.

A larger sample size and longer intervention duration would enhance the generalizability of the data and provide more opportunities to determine usage patterns and preferences. However, most of the reviewed studies employed shorter intervention durations and did not include follow-up assessments, involving only a small number of participants. This phenomenon is also observed in other areas of gamified interventions for older adults. For example, Savulich et al [29] used a sample size of 21 participants in their intervention group for cognitive training using iPads for older adults diagnosed with amnestic mild cognitive impairment, with an intervention duration of 4 weeks. Costa et al [30] used a sample size of 20 participants in their intervention group for mental health interventions for community-dwelling older adults through a gamified learning platform, with an intervention duration of 6 weeks. Despite the increased opportunities to access new technologies in recent years, the application of these tools among the older adult population still faces significant challenges.

Choosing game elements that cater to the needs of older adults or developing corresponding gamified applications can significantly enhance the effectiveness, practicality, and sustainability of interventions, thereby increasing user participation and reducing sample attrition rates during follow-up processes [31]. The studies included in this review generally adopt goal-setting (eg, daily step counts) and instant rewards (eg, points and badges) as core elements. These elements are widely used due to their strong alignment with the core mechanisms of behavior change [32], while also addressing the research need for quick verification of the immediate effects of gamification through short-term interventions. This approach aligns with the conclusions of 2 previous systematic reviews, which did not specify the research subjects [33,34].

However, the adoption of complex gamification elements, such as virtual identities and social collaboration, requires older adults to construct abstract self-representations or coordinate rules for multilateral interactions. This may lead to cognitive load or operational obstacles for some older adults [29], resulting in fewer studies designed with such elements. Additionally, we note that the included studies all employed interventions with 2 or more combined elements, suggesting that combinations of multiple game elements may be more effective than single elements. However, due to the lack of detailed information in the initial studies, no definitive conclusions can yet be drawn about the optimal number or composition of game elements in these interventions.

One point that needs clarification is that the most frequently used gamification elements do not necessarily indicate the optimal effectiveness of those elements on older adults or physical activity. As Dugas et al [35] pointed out, personality differences can significantly affect how certain gamification characteristics influence older adults’ use of mobile health technologies. When gamification mechanisms are unsuitable, older adults often fail to understand and recognize the relevant benefits of these services. Therefore, we must approach the results of this systematic review with caution, as it could lead to cognitive biases. In the study by Kostopoulos et al [36], the authors established goal-setting and achievement features that allow users to receive different rewards after completing various goals and tasks. The purpose was to further explain how these goals and tasks are generated for each user, which may help in understanding the different motivations and preferences that drive users to perform various tasks. This is an innovative approach, as it provides feedback on user performance while validating the effects of gamification interventions.

We have also found that Greysen et al [25] systematically included primary caregivers (support partners) as a core component of the intervention. And although other studies did not independently analyze or evaluate the involvement of nursing staff, most of them indirectly addressed similar concepts through social functions or external support mechanisms. This suggests that incorporating nursing staff into the network could be a valuable development. Including family members, health care professionals, and social workers in mobile health gamification interventions may enhance older adults’ perceptions of social and emotional support, thereby increasing their acceptance of these interventions [37]. Additionally, from a technical perspective, this model can improve older adults’ understanding of the intervention content and their sense of security in using the interventions [38]. This approach is particularly beneficial for older adults who are interested in gamification but lack independent operational skills, as well as for those who prefer direct care from health care professionals, providing an effective pathway for the broader application of gamification.

Technological platforms that accompany gamification are another important factor attracting the interest of older adults. Currently, validated technological features used among the older adult population include self-management systems [9], physical robots [10], consoles [11], portable devices, and wearable technology [12]. Although studies by Kitakoshi et al [38] and Mocanu et al [39] have confirmed that older adults enjoy using systems and interacting with robots, mobile health offers more diverse gamification opportunities in real-world applications. This aligns with the viewpoint of Steinert et al [40], who pointed out that older adults are more motivated by tangible information provided by activity monitors, sensors, and smartphones, rather than more abstract game feedback. This is also the primary reason why our review is limited to gamification intervention research supported by mobile technology.

In this systematic review, the mHealth technologies used to provide physical activity gamification interventions include 1 activity monitor (1/8), 4 mobile apps (4/8), and 3 combinations of both (3/8). This distribution of technological application patterns reflects a shift in the current research field, from single-device use to multimodal and collaborative intervention plans. Tablet devices offer significant advantages in screen size and clarity, and previous research [30] has confirmed their ability to improve specific cognitive functions in older adults. However, the physical characteristics of the devices, such as weight, button design, and keyboard operation, may affect the user experience of older adults participating in activities. As a result, only 1 study in this review mentioned tablet devices [23].

The theoretical framework also warrants discussion, as it enables a deeper analysis of the behavioral mechanisms behind older adults’ participation in physical activities, accurately addressing their physiological, psychological, and social needs. However, among the reviewed studies, only 4 [19,23-25] explicitly adopted a single theoretical framework or model to design gamified interventions for physical activities targeting older adults. The classification framework revealed that these theory-oriented designs demonstrated more effective outcomes while accommodating greater gamification complexity, suggesting that theoretically integrated gamification may play a positive role in reducing engagement burden among older adults.

At least 2 of these studies [23,25] utilized behavioral economics, incorporating design principles such as loss aversion, precommitment, and the fresh start effect. The significance of this theoretical application lies in viewing participation in physical activities as an investment in future health, guiding older adults’ behavioral choices through subtle environmental adjustments rather than rigid directives. In another study [19], the circumplex model of affect was translated into operational gamification solutions, offering new insights for the design of digital therapeutics in geriatric health.

Self-Determination Theory, a comprehensive motivation theory, has become a key framework for health behavior interventions, but only 1 study [24] claims to have used it, differing from the 2 systematic reviews mentioned earlier [33,34]. Other commonly considered behavior change theories include the Health Belief Model [41], Theory of Planned Behavior [42], Transtheoretical Model [43], and Habit Formation Theory [44], but these theories appear to be rarely applied in gamified interventions for older adults, even in nonphysical activity areas. The reasons for this may partly stem from the relatively novel concept of introducing gamified design to the older adult population, where traditional theories may not align well. Alternatively, it could be that the behavior patterns of older adults are complex and diverse, making the application of theory more challenging. Therefore, a more systematic and appropriate gamification theory system for the older adult population could be developed in the future through qualitative exploration, theoretical integration, and other methods.

Finally, a study published by Bellotti et al [45] in the field of human-computer interaction suggested that game-based interventions could serve as a strategy for patient self-assessment, enriching the typical content of assessments for older adults. Although no authors in the reviewed studies directly made such a claim, this idea may also apply to gamification for older adults. By continuously tracking the activity performance of older adults, research can gather multidimensional data, such as task completion rates, execution times, and failure counts. This information, when combined with commonly used assessment tools for older adults, such as the Activity of Daily Living scale [46] and the Morse fall risk assessment scale [47], could provide a more comprehensive reflection of functionality and psychological state, thereby guiding clinical decision-making and personalized intervention strategies. While this is a reasonable inference, further investigation is needed. Additionally, it is important to consider data privacy and the risk of false positives to avoid excessive monitoring, which could lead to psychological stress or self-report biases due to the Hawthorne effect.

Although this study systematically reviews the effectiveness of mHealth gamification interventions on physical activity in older adults, several limitations should be noted. First, the included studies are relatively few (n=8), with generally small sample sizes, potentially limiting the generalizability and reliability of the findings. Second, intervention durations varied significantly (1 wk to 12 mo), and most studies lacked follow-up periods, complicating the assessment of long-term sustainability. Additionally, considerable heterogeneity existed across studies, including variations in gamification elements, technological platforms, theoretical frameworks, and outcome measurement methods, precluding meta-analysis and unified conclusions. Another limitation is the incomplete reporting of baseline participant characteristics (eg, digital literacy, health status), potentially biasing intervention effect interpretations. Finally, as most studies focused on high-income countries (eg, the United States, Japan, Switzerland), the findings may lack generalizability to older adults in other regions or cultural contexts. Furthermore, due to incomplete reporting of minimum participant ages, this review inferred population homogeneity using a functionally oriented aging definition and statistical estimation, which may overestimate or misrepresent intervention effectiveness.

This systematic review suggests that mHealth-based gamified interventions have potential benefits in increasing the physical activity levels of older adults, particularly by enhancing daily step counts and the duration of MVPA. Common gamification elements, such as goal-setting and immediate rewards, effectively stimulate the participation motivation of older adults, while the combination of mobile apps and wearable devices further enhances the flexibility and accessibility of these interventions. However, existing evidence still has limitations, including a limited number of studies, small sample sizes, and a lack of long-term follow-up data. Future studies should aim to expand sample sizes, extend intervention durations, and include more diverse older adult populations to validate the universality and sustainability of the intervention effects. Additionally, there is a need to further explore the optimal combination of gamification elements, the integration of theoretical frameworks, and the potential impact of caregiver network participation on the effectiveness of these interventions. Overall, mHealth gamified interventions offer an innovative and feasible strategy for promoting physical activity among older adults, but their design and implementation must carefully consider the specific needs and varying abilities of older adults to ensure the practicality and acceptance of these interventions.