Sage Journals: Discover world-class research

Abstract

Objective

Intervention research using digital games to promote physical activity has proliferated. Yet few studies have attempted to systematically catalog features that characterize this research. To address this gap, we undertook a systematic review and content analysis of active video game interventions, examining only published longitudinal interventions that prominently featured active video game technology (≥50% of the intervention).

Methods

Our protocol was registered in the International Prospective Register of Systematic Reviews (CRD42020204191). For inclusion, an active video game intervention had to require gross movement beyond finger movement, and target improvement, maintenance, or recovery of health. The intervention design had to include at least two conditions, within- or between-subjects, with ≥10 participants per condition to examine the chronic effects of active video game exposure.

Results

The search resulted in 232 studies published in English between 1996 and 2020. The majority of active video game interventions (69.8%) targeted physical fitness (physiological functioning as a consequence of physical activity), followed by cognitive performance (11.3%), physical activity (5.5%), or a mixture of those outcomes (13.4%). Total enrollment across all studies was 14,849 participants (M_Participants = 62, SD = 106; M_Age = 50.2, SD = 25.2 years; 47.3% men). A strong majority of the samples (69.8%) were recruited from medical subpopulations, and only 30.2% were recruited from the general (healthy) population. A strong majority of active video games (72.0%) were developed by industry for the commercial market, and only 13.3% were funded by government or foundation grants.

Conclusions

Suggested directions for improving future active video game development and intervention research include greater consideration of promising features (social connectedness, novelty, narrative, rhythmic movement to music) and new models for productive collaboration between industry and academia.

Keywords

Systematic review content analysis games for health active video games exergames industry-academia collaboration

The diverse category of games collectively referred to as “active video games” (AVGs) represents an evolving term of art, now widely used in multiple academic research literatures, in popular media, in the game industry, and with increasing frequency. These are electronic games that allow players to interact physically (using sensor-detected arm, leg, or whole-body movement) with images on a screen (head-mounted displays and/or an external screen; synchronously and/or asynchronously) while performing a variety of physical activities, including a variety of simulated sports (e.g., football, tennis, martial arts) and other activities (e.g., recreational dancing). However, many individuals still speak and write about AVGs in a relatively limited way, using narrow definitions. This project aims to provide an updated understanding of the full breadth that characterizes AVGs and interventions that employ them today, mainly to help facilitate critical dialogue and inform the development of future AVGs and AVG interventions to improve public health and well-being. We approached this challenge by reviewing the historical context from which AVGs emerged, and through a descriptive research method called comprehensive systematic review and content analysis. After reporting the results and highlighting important features of the latest AVG intervention studies, we point to some key limitations and promising directions for future applications and research.

A brief history of games, games for health, and AVGs

Tekinbas, Salen, and Zimmerman¹ define a game as “a system in which players engage in an artificial conflict, defined by rules, that results in a quantifiable outcome” (p. 80). Intersecting with this definition is the concept of intrinsic motivation, a form of motivation associated with psychological need satisfaction and positive emotions.² Although games can provoke a wide range of emotions with both positive and negative valences, the vast majority of games are designed to promote positive emotions, especially interest and enjoyment.

Games for Health. A “game for health” (G4H) is a broad category that includes any game designed to promote mental, behavioral, or physical health. Seminal work by Debra Lieberman was among the first to focus on patient health promotion with video games. These early G4H were directly inspired by educational research using video games to promote learning.³ The scope of research on G4H has expanded dramatically since the late 1990s, spawning multiple academic conferences and dedicated scholarly journals. An appropriately expansive definition of G4H offered by the Robert Wood Johnson Foundation's Games for Health Project in 2017 was: “games and game technologies [intended] to improve health and health care.” This definition includes games for prevention, treatment, patient education, medical education, continued education, and training.

AVGs. A major development in the history of G4H involved the integration of technologies capable of detecting range and intensity of body movement and the extension of G4H to changing behavior by integrating detected body movements into the games’ interactive systems themselves. Collectively, this subcategory of G4H is now referred to as AVGs. AVGs now represent one of the largest and fastest-growing categories of G4H in terms of industry investment, consumer interest, and academic research (e.g. relative to G4Hs dedicated to promoting mental health or to medical education for patients or providers). For example, one recent systematic review of G4H for mental health found nine RCTs published between 2007 and 2015.⁴ Another recent systematic review of G4H for supporting engagement and learning among healthcare professionals found just 37 RCTs published between January 2005 and April 2019.⁵ Both of those previously published reviews of G4H RCTs included considerably fewer RCTs than the 232 RCTs on AVGs published between January 1996 and October 2020 and reviewed here. During the COVID-19 pandemic, in particular, demand for AVGs soared, resulting in shortages of games like Nintendo's Ring Fit Adventure.⁶ As the sophistication of motion-detecting and computing and graphic technologies has advanced, AVG's breadth has also expanded significantly. The target audiences cover nearly the entire human life course, from early childhood to geriatric populations, including those from the general population (prevention) and those with acute or chronic injuries or conditions (treatment/intervention). The definition of “active” in AVG includes a wide range of physical activity intensity, including subtle movements (as in some rehabilitation contexts) and moderate-to-vigorous exercise. AVGs that promote exercise by requiring movements mimicking “real-life” physical activities are also known as exergames.⁷

Past systematic reviews of AVG intervention research

Over the past decade, a number of outcome-focused systematic reviews and meta-analyses related to AVG research have been published covering a wide spectrum of target populations and health conditions. For example, outcome-focused systematic reviews have focused on children targeting weight loss,^8–10 adults targeting energy expenditure,¹¹ children with neuromotor dysfunction,^12,13 adults with neuromotor dysfunction,¹⁴ older adults targeting physical performance,¹⁵ and older adults targeting rehabilitative outcomes.¹⁶ In terms of audience or target population, past systematic reviews of AVGs have often focused on a limited age range, most often children or young adults, followed by elderly adults, followed by middle-aged adults. A 2022 protocol for an AVG review by Hoffmann and Wiemeyer pointed out that “most studies focus on specific training effects or specific target groups” and that “a comprehensive summary of … effects with exergames in healthy adults is still missing.” Primary outcomes of interest in past systematic reviews of AVGs include physical activity, weight loss, motor skills, rehabilitation, and physical education.¹⁷

Importantly, the focus of these past systematic reviews has been AVG interventions’ efficacy and/or effectiveness. Specifically, many meta-analyses of AVG interventions have assessed AVGs based on changing targeted health outcomes, either within relatively tightly controlled research settings (efficacy) or, once disseminated on a large scale, in less tightly controlled field settings (effectiveness). These are important questions, as evidenced by the volume of past publications investigating them, but they are not the focal questions of the present research. Our current study fills a significant gap in the AVG systematic review literature by undertaking a systematic review and content analysis, a descriptive approach designed to identify the qualities and characteristics of AVGs and AVG interventions, as opposed to their efficacy or effectiveness for changing health outcomes.

Systematic content analysis of AVG intervention research

As the number of studies on AVGs has grown exponentially over the last three decades, regularly updated descriptive analysis studies can help inform our understanding of important thematic characteristics, identifying game and study design patterns, including over- and underrepresented features. To date, this kind of descriptive analysis of the AVG research literature has been rather limited, since most of the systematic review or meta-analysis studies have been relatively narrowly focused on specific health outcomes or subpopulation.

Objectives. Many basic questions about AVG intervention research have not been fully explored, especially with respect to characterizing this research space and where researchers have focused their attention. For example, how many AVG intervention studies have been published? What were the characteristics of the samples and settings for those studies (i.e. which nations were represented; how large were the samples; how old were the participants; what were the gender distributions; were interventions delivered in lab settings or in the field; what medical subpopulations were targeted)? What were the primary outcomes targeted? What were the defining design characteristics of these AVG intervention studies (e.g. time of exposure to AVGs, ideally vs. actually)? What were the defining characteristics of the AVGs themselves (e.g. what AVG platforms were used; how immersive was the gaming experience; who designed and funded the development of these AVGs; and what features were most frequently employed)? Finally, how can we characterize the methodological rigor or quality of AVG intervention studies to this point?

Filling these gaps and providing a full spectrum snapshot of the field will help a wide range of AVG stakeholders, including industry and academic AVG designers, researchers, and funders. Specifically, AVG designers may find it useful to know that AVGs designed for one population are frequently being used in interventions targeting different subpopulations; researchers and funders may benefit from greater awareness of which types of AVG interventions have been more or less studied, and how are these AVG interventions deployed, when setting research priorities and funding strategies as well as finetuning intervention strategies.

Methods

To address these gaps, we undertook a comprehensive systematic review and content analysis of AVG interventions, examining published interventions that prominently featured AVG technology (at least 50% of an intervention needs to involve one or more AVGs). To assemble a broad scope of AVG research interventions, we undertook an extensive process to identify as many eligible studies as possible. After consulting with experts in the field of G4H and AVG research, we decided to use multiple existing international academic databases of games for health research publications instead of simply soliciting game entries from health game databases given the lack of overlap between the two¹⁸ and the lack of peer-review for quality control. This review was performed in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines.¹⁹ The protocol of this review was registered in the International Prospective Register of Systematic Reviews (PROSPERO: CRD42020204191).

Inclusion and exclusion criteria

For inclusion, the studies must have appeared in peer-reviewed English language journal articles or full-length conference papers, as these are the dominant outlets for sharing findings from rigorously conducted AVG intervention research. The AVGs involved must have been interactive, powered by electricity, and require gross motor movement beyond mere finger movement with the goal to improve, maintain, or regain health. The intervention design must have included at least two conditions, within- or between-subjects, with ≥10 participants per condition to examine the chronic effects of AVG exposure instead of a one-time exposure. Studies that included one group pre-test post-test or one group post-test-only design were excluded, as were synthesis studies (e.g. past systematic reviews, meta-analyses). Synthesis studies were used only to help identify individual studies.

Primary outcomes

The three primary outcomes considered were physical activity, physical fitness, and cognitive performance. Physical activity refers to all bodily movement produced by contracting skeletal muscles that substantially increases energy expenditure, and can be represented by direct assessments of physical movement (e.g. using one or more accelerometers, GPS technology, or validated questionnaires).²⁰ Physical fitness is defined as a set of attributes that people possess or achieve that relates to the ability to perform physical activity and is comprised of skill-related, health-related, and physiological components, including cardiorespiratory (e.g. resting heart rate, VO₂ max) and muscular fitness (e.g. muscular strength, power, balance, flexibility, mobility).²⁰ It can be understood as a consequence of physical activity. Cognitive performance includes performance on a myriad set of tasks to measure selective aspects of cognition including, but not limited to, perception, attention, memory, reasoning and problem solving, and executive function.²¹

Search strategy

Between March 2020 and December 2020, we searched electronic databases including PubMed, EBSCO (PsycINFO, Sport Discus, MEDLINE), Web of Science, and Google Scholar for relevant studies. The search process was divided into two phases. In Phase I, we first searched for all synthesis articles (review, narrative review, systematic review, meta-analysis, and synthesis of any previously mentioned article type) published by 30 April 2020 in Google Scholar and PubMed databases. Remarkably, a search for synthesis articles, including AVG, exergames, virtual reality (VR) health, and VR rehabilitation, uncovered 201 publications. For the synthesis of synthesis articles, we first extracted the separate synthesis articles and then examined the individual original articles from each of the separate synthesis articles. We extracted 3724 individual original articles from these synthesis articles. In Phase II, we searched for individual original studies published between 1 January 2016 and 31 December 2020, to ensure more recent studies not included in the synthesis articles were included. The same set of search engines as used in Phase 1 were used in our Phase 2 individual article search. We found 1866 individual articles using the keyword search. We then merged the search results in both Phases and identified 630 unique articles by abstract and continued to read the full text of these articles. This full-text reading resulted in 232 unique publications published between January 1996 and October 2020 that fully met the inclusion criteria. This two-phase process proved more comprehensive than Phase II alone, even without limiting the original publication's time range. A summary of this two-phase search process is illustrated in Figure 1, a PRISMA Flow Diagram (https://prisma-statement.org). A table listing all 232 studies’ author(s), year of publication, the interventions’ primary outcome, study design, type of AVG, type of AVG platform, and bias scores is included as Electronic Supplemental Materials, along with the Boolean search phrases that were used for both Phases, a figure illustrating the frequency of studies published by year, and a PRISMA checklist.

Figure 1.
Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) flow diagram.

Data Extraction and Content Coding. Three independent reviewers with different academic backgrounds (Biology, Health, and Physical Education) participated in article selection and data extraction to ensure fair and comprehensive coverage. They received two training sessions per week throughout the 4-month project preparation period from May to September of 2020. The inter-rater reliability was assessed every week to ensure that it was consistently higher than 85% during the training and was maintained at over 93% for the rest of the coding process. Differences were solved through discussion until all coders agree on how to proceed.

Study quality. To assess the quality of the articles included, the Grading of Recommendations Assessment, Development and Evaluation (GRADE) quality assessment tool was implemented.²² It contains six components [random assignment (to avoid selection bias), allocation concealment (to avoid selection bias), blinding of participants and personnel (to avoid performance bias), blinding of outcome assessment (to avoid performance bias), completeness of reporting of some outcome data (to avoid attrition bias), and selectivity in reporting (to avoid attribution bias)]. The research team added a seventh component, other bias, which assessed whether there was any potential adjustment for additional confounding variables (confounder adjustment). Each category is given a score of −1 to +1 (−1 = low quality or high risk of bias; 0 = unclear; +1 = high quality or low risk of bias). The overall score was an average of each individual component on the −1 to +1 scale. No articles were excluded on the basis of their quality assessment.

Results

The total number of unique individual studies included was 232. Six of these studies included multiple AVG conditions, i.e. comparing exposure to two AVGs or sets of AVGs; often a single experimental group was given multiple (up to 12) AVGs to choose from. Thus, when reporting descriptions of game characteristics, the sample size increased from 232 (studies) to 238 (games or sets of games).

Samples, settings, and subpopulations

The interventions were internationally distributed, and the top five nations represented were the US (16.0%), Taiwan (10.1%), South Korea (8.4%), Turkey (7.1%), and Brazil (6.3%), The total enrollment across all 232 intervention studies was 14,849 (M_Participants = 62, SD = 106, range = 20–1112; M_{Experimental Group} = 28, SD = 54, range = 10–555; M_{Control Group} = 34, SD = 51; range = 10–557). Although the participants’ mean age was 50.2 years, the studies involved participants from a wide range of ages (Range: 3–99 years, SD = 25.2): 22.8% studies involved children (0–18 years), 14.2% of studies involved adults but not seniors (18–64 years), 32.8% studies focused on seniors (>64 years), 1.3% studies included children and adults and not seniors (0–64 years), and 27.2% studies included adults as well as seniors (>18 years). Of all participants, 47.3% identified as men. In terms of the research settings, 63.9% of the studies were conducted in the field, with 6.3% in a lab setting (29.8% did not specify the setting). Among field studies, most were conducted in physical therapy offices (30.9% of field studies), followed by hospitals (26.3%), schools (17.8%), at home (16.4%), and group living facilities (8.6%). A minority, 30.2%, of the samples were recruited from the general (healthy) population, while 69.8% were recruited from medical subpopulations. The most common medical conditions targeted included stroke (15.6%), Parkinson's disease (8.0%), cerebral palsy (5.5%), overweight/obesity (3.9%), and multiple sclerosis (3.2%).

Primary outcomes

Primary outcomes for each study were coded as one of the four categories: physical activity, physical fitness, cognitive performance, and mixed (more than one of the three outcomes). The majority of published AVG interventions targeted physical fitness (69.8%), followed by cognitive performance (11.3%), physical activity (5.5%), or a mixture of those outcomes (13.4%).

AVG intervention characteristics

The duration and frequency of AVG intervention exposure in different trials varied considerably, from 8.6 to 100 min/session, and from 1 to 10 sessions/week. The total mean ideal vs. actual session time (min/session), and frequency (sessions/week) were similar: 38.81 vs. 38.75 min/session and 3.18 vs. 3.16 sessions/week, respectively. Weeks of exposure ranged from 1 to 52 (M_Weeks = 9.37). The total idealized duration of AVG exposure for an average intervention was 21.06 h vs. 20.30 h of actual AVG exposure. The high concordance between the ideal (prescribed) and actual AVG exposure reflects high levels of compliance with the study protocol. However, we note that some studies failed to report some aspects of ideal and/or actual AVG exposure. With respect to estimating total duration of AVG exposure, 78.4% of papers reported total ideal exposure time and 79.7% reported total actual exposure.

We then carried an additional analysis among studies of different primary outcomes. More specifically, in terms of intervention session duration, studies that focused on a mixture of the outcomes (M_Idealized _min/session = 45.5, SD_Idealized = 21.6; M_Actual _min/session = 45.3, SD_Actual = 21.5) tended to be longer (ps < 0.02) than studies that focused on physical fitness (M_Idealized _min/session = 36.0, SD_Idealized = 14.3; M_Actual _min/session = 36.0, SD_Actual = 14.4). While there was no difference among idealized and actual intervention session frequency (ps > 0.27), studies that focused on physical activity (M_Idealized _weeks = 14.3, SD_Idealized = 10.9; M_Actual _week_s = 14.3, SD_Actual = 10.9) tended to be scheduled to last more weeks (ps < 0.02) than studies that focused on physical fitness (M_{Idealized weeks}= 8.6, SD_Idealized = 6.2; M_Actual _weeks = 8.6, SD_Actual = 6.2). There was no difference between the idealized or actual overall intervention duration in minutes across the studies of four primary outcomes (ps > 0.61).

AVG characteristics

In terms of AVG platforms, 13.4% were PC-based games and the remaining 86.6% required a dedicated gaming console. Among consoles, Nintendo Wii or Wii U was by far the most popular line of consoles (44.5%), followed by the Microsoft Xbox Kinect (18.9%), and Sony PlayStation (3.8%). Another 8.0% required integrated AVG-specific equipment (e.g. an exercise bike or step-detecting floor mat), and just one study (<1%) used a mobile smartphone platform (Zombies, Run! 5 K Training). In total, 10.5% used a unique combination of the above platforms or an otherwise unique platform. In terms of immersive and non-immersive VR, just 3.8% of AVGs use in interventions published pre-2021 involved VR that required a head-mounted display, another 18.1% were identified as including VR without requiring a head-mounted display.

A large majority of AVGs involved in published intervention research (72.0%) were developed by private companies for the commercial market. Only 13.3% were funded by government or foundation grants specifically for research and development, and the remaining 14.7% of studies did not report sufficient information to allow us to determine the source of funding for AVG development. There were no significant differences in terms of AVG platform usage or funding support across studies of four primary outcomes (ps > 0.10).

In total, 93.6% of AVGs did not require an Internet connection during gameplay. Only 2 (0.9%) did, and for 5.2% it was not clearly reported. AVGs were also coded for the presence of narrative. Surprisingly, less than 1% of AVGs (1 of 238) included this feature. By contrast, we observed that 10.5% of AVG conditions included rhythmic movement to music as a central game feature (5.0% of AVG conditions included several AVGs with at least one AVG with this feature as an option; 5.5% of AVG conditions included only AVGs with rhythmic movement to music as a central game feature).

Study quality

As mentioned previously, the overall study quality GRADE was calculated by averaging the scores for seven components of study quality on a −1 to +1 scale (−1 = low quality or high risk of bias; +1 = high quality or low risk of bias). The distribution of GRADE scores on all seven components is summarized in Figure 2. Overall, the mean total GRADE score tended to be relatively low, −0.38 (Range: −1.0 to +0.43; SD = −0.31). In terms of the studies of four primary outcomes, significant differences were found among their GRADE scores (p < 0.03). More specifically, studies that focused on physical fitness (M = −0.40, SD = 0.29) tend to have worse (p = 0.05) GRADE score ratings compared to studies that focused on multiple outcomes (M = −0.24, SD = 0.37).

Figure 2.
Frequency of grading of recommendations assessment, development, and evaluation (GRADE) research quality assessment scores.

The mean total GRADE score was then correlated with the year of publication and other study descriptors such as the continent where the study was conducted, participant age, and idealized and actual intervention duration. The total GRADE score was negatively correlated with studies mean participant age (r = −0.16, CI: −0.29 to −0.02, p = 0.02, N = 213) and with idealized intervention duration (r = −0.17, CI: −0.31 to −0.02, p = 0.02, N = 182), meaning that interventions targeting younger participants and with shorter intervention durations were associated with higher publication quality. All other correlations were non-significant (ps. = 0.06–0.16).

Discussion

We attempt to synthesize patterns of content, identify key themes, and consider why the field of AVG intervention research may have evolved as it has to this point. Finally, we offer some recommended directions for future AVG development and intervention research.

Setting, targeted subpopulations and outcomes

Through this systematic review and content analysis, we have identified several characteristics of AVGs and study designs that were highly represented (i.e. characteristics of ≥60% of AVGs or AVG intervention studies). In particular, the majority of AVG intervention studies were field studies (63.9%), targeting physical fitness (69.8%), and targeting participants with underlying medical conditions (69.8%). The high prevalence of these three features in AVG intervention research may be interrelated. First, subpopulations with medical conditions are frequently prioritized in public health research²³; reasons for this include higher experienced distress, more costly healthcare, and thus larger potential return-on-investment. Given that researchers are more likely to study the influence of AVGs on subpopulations with medical conditions, field studies targeting more clinically relevant primary outcomes (e.g. physical fitness, which includes the clinically relevant consequences of physical activity) are likely prioritized. It's also noteworthy that we included “virtual reality rehabilitation” (italics added) as one of the search terms used for our review, as this may have helped us identify AVG interventions targeting participants with underlying medical conditions.

Interestingly, the average age of study samples was relatively high (M = 50.2 years, SD = 25.2), and high variability was observed in terms of age-related eligibility requirements. Specifically, the two most common categories of age-eligibility were seniors, >64 years (32.8%), followed by children, 0–18 years (27.2%). This speaks to the wide appeal and applicability of AVGs with respect to age. Future research could investigate with more granularity how qualities of AVGs and the interventions using them may differ based on the age of participants targeted.

Duration and frequency of AVG exposure

One frequent criticism of AVGs concerns questions about potentially limited duration and frequency of AVG exposure. Specifically, while lab studies involving brief exposure to AVGs during a single session may help address some basic research questions, these studies offer limited insight into important applied questions, especially those concerning maintenance of PA as a determinant of potential public health impact. To help address this concern, our systematic review and content analysis screened out studies with fewer than two sessions of AVG exposure and only looked at chronic intervention effects. As a result, our sample of intervention studies involved relatively long-term engagement: average 38.75 min/session, for 3.16 sessions/week, and for 9.37 weeks, for a total of 20.30 h of actual AVG exposure. This shows that AVGs are capable of sustaining relatively long-term engagement during a longitudinal intervention. However, less is known about the long-term maintenance of AVG engagement or PA behaviors post-intervention; few studies assess or report those outcomes. Additional analysis of studies focusing on different primary outcomes indicated that physical fitness studies tended to have shorter session duration than mixed outcome studies and tended to monitor engagement for fewer weeks than studies targeting PA. While exploring the reasons for those differences is beyond the scope of this project, this may warrant future exploration. Additionally, more research is needed to address follow-up questions concerning what types of AVGs, and which AVG design features, are associated with longer user engagement and PA maintenance.

Researchers have proposed general models guiding intervention design that delineate different psychological mechanisms driving initiation versus maintenance of health behavior change.^24–26 For example, Rothman et al.’s²⁴ model proposes four distinct phases in the behavior change process and identifies different determinants driving shifts from one phase to the next (from initiation to maintenance). AVG designers and researchers interested in promoting long-term engagement with AVGs and PA can find inspiration from this and other multiphase models of health behavior change, intervention design, and implementation (e.g. intervention mapping).

Based on our review of AVG interventions and the success of commercial AVGs, several factors that may help drive longer engagement are (a) support for social connectedness in the context of enduring interpersonal relationships, (b) regular updates that evoke persistent novelty experiences, (c) incorporation of narrative, and (d) incorporation of music and rhythmic movement. Each of these factors have been associated with sustaining intrinsic motivation, one mechanism for supporting long-term engagement, in other contexts: social connectedness,²⁷ novelty,^28,29 narrative,³⁰ and rhythmic movement to music.³¹ Their potential for sustaining intrinsic motivation and long-term engagement with AVGs, however, are speculative hypotheses that we believe warrant future investigation.

Narrative

Narratives, or stories, typically consist of characters and plot. Narratives have a strong potential for behavior change through mechanisms such as narrative transportation, engagement, character identification, suspension of disbelief, and reduction of counterarguments.³² Narratives have been identified as a crucial layer of player motivation and gaming experience.³³ Although narratives were previously found in around one in five of all G4H,¹⁸ only one of the 238 AVGs included in our content analysis had an identifiable narrative. This was surprising given that narratives, when added to existing AVGs, have been found to induce significantly more physical activity behaviors across different age groups and game platforms.^30,34–36

Rhythmic movement to music

Our content analysis revealed that 10.5% of AVG interventions include integration of rhythmic movement to music as a feature of game play. We also noted the popularity of this feature in many commercially successful AVGs, especially immersive AVGs (e.g. Beat Saber, Supernatural, Peloton Lanebreak). One form of secondary evidence for the importance of music in AVGs comes from public disclosures of financial investments in copyrighted music made by companies making AVGs.³⁷ Yet surprisingly few studies have explored whether rhythmic movement to music uniquely enhances AVG engagement.³⁸

AVG platforms and funding

Our project revealed a noteworthy discrepancy between the target populations prioritized by designers of AVGs used in AVG intervention studies versus the subpopulations frequently targeted by behavioral health interventions. Specifically, AVG intervention research has tended to focus on subpopulations with specific medical conditions,¹⁷ but the AVG platforms and games used in those interventions tend to be designed primarily as entertainment devices for the general population. Two AVG platforms designed for the general population, the Nintendo Wii or Wii U (44.5%) and Microsoft Xbox Kinect (18.9%), supported 63.4% of the platforms used in the AVG intervention studies we reviewed. Similarly, the development of the AVGs themselves was overwhelmingly funded by the for-profit entertainment gaming industry rather than non-profit government or foundation sources (72.0% vs. 13.3%, respectively). Although using AVGs originally designed for entertaining the general population in more targeted health interventions can help researchers reduce the cost of intervention development, this approach could also inhibit the clinical effectiveness of AVG interventions targeting medical subpopulations. Better design practices would include user-centered, participatory research that involves members of medical subpopulations from the earliest stages of AVG design planning.^26,39–41 Furthermore, although all the major AVG platforms are open to third-party game developers, the AVGs themselves tend to be relatively closed, lacking options for modifications and customizability that would allow researchers to tailor AVGs according to the needs of targeted clinical populations.

Given that behavioral health researchers are already embracing the use of commercial AVGs, we believe new models for enhanced industry-academia collaboration could improve future AVG intervention research. As mentioned above, collaborations between industry and academia that begin as early in the AVG design process as possible are likely to produce a larger public health impact, as evidenced by related work demonstrating the advantages of user centered and community-based participatory design.⁴² How can earlier-stage industry-academia collaborations on AVG design be encouraged? Below we outline three themes for AVG stakeholders' consideration: navigating profit vs. non-profit motives, making AVGs that are customizable and modifiable, and establishing models for ethical AVG data sharing.

Navigating for-profit versus non-profit motives. One potential barrier to AVG collaborations involving industry and academia is the tension between for-profit and non-profit motives. However, there are examples of models developed to mitigate this tension.^43,44 Some for-profit game studios have created relatively independent philanthropic divisions that make public health impact their first priority, and/or have endowed faculty lines or research centers at universities with missions prioritizing public health impact over profit. For example, Microsoft's Research Division collaborated with university researchers on the development of AVGs using their hardware (Xbox Kinect), such as an AVG called Recovery Rapids designed to support stroke recovery.⁴⁵ Another instructive example from the video game industry involves a coordinated effort across multiple companies dedicated to a different public health issue: climate change and environmental sustainability. The Playing for The Planet Alliance, formed in 2019, includes over 40 game studios and industry groups, and sets standards to reduce greenhouse gas emissions and plastic pollution using initiatives like the Green Game Jam developer competition to incorporate environmental messages into video games, including big-name franchises. Given the far-reaching significance of physical activity/inactivity for both mental and physical health, we urge game studios to consider expanding their coordinated corporate responsibility efforts to include more support for AVGs, e.g. a Playing for Global Health Alliance.

Designing customizable and modifiable AVGs for research. One common feature of AVGs developed by university researchers that differentiates them from most AVGs developed by large industry game studios is the degree of customizability and modifiability. AVGs designed by researchers often include an administrator dashboard or tool for customizing different features and affordances available to players, and sometimes access to source code with permission to engage in modifications (i.e. add-ons called “mods,” or even “total conversion mods”). This support for customizability and modifications facilitates research by allowing randomization to many different intervention conditions, not only for those researchers involved in the original AVG design but for any researcher given access to these tools. Large game studios could help advance public health research using AVGs by providing researchers with more customization and modification tools like these, as well as tools for accessing passively collected player engagement data. Successful models encouraging sedentary video game customization via dashboards or source code exist, typically for fans and third-party developers,⁴⁶ but few companies have extended this to popular AVGs or prioritized the needs of behavioral health researchers.

Models for ethical AVG data sharing and oversight. If the frequency and intensity of industry-academia AVG research collaborations do increase in the future, we note that a challenge to ensuring a net benefit to society will require careful consideration of data storage and respect for AVG players’ privacy. This includes developing processes for collecting meaningful, affirmative consent for data sharing and ethical oversight. Having ethical oversight is often critical, as ethical standards frequently exceed legally required standards. Recent examples of successful data sharing between a single university lab and large video game studios exist,^47,48 though these were limited to sedentary video games. The ReCODE Health project offers resources to inform ethical digital health research practices, broadly defined,⁴⁹ which could be extended and tailored specifically for AVGs and other G4H projects.

More AVG intervention research targeting the general population

Although health researchers often prioritize subpopulations with medical conditions to maximize intervention effect size, public health impact is understood as a product of both an intervention's effect size and scalability. In other words, a smaller benefit experienced by tens of millions of people may have a larger public health impact than a larger benefit experienced by thousands of people. With this in mind, it is noteworthy that relatively few AVG intervention studies (30.2%) recruited samples from the general (healthy) population. Given that for-profit companies currently dominate AVG development, and tend to design AVGs for the general population, researchers may consider the value of running more rigorous intervention studies that assess the potential benefits of AVGs for the general population (e.g. see Baranowski and Lyons⁵⁰) to help assess and maximize their preventive, rather than treatment, potential.

Future systematic content analyses and umbrella reviews

To the best of our knowledge, the present research is the first comprehensive systematic review and content analysis of AVG intervention studies. It offered insights into underexplored topics for future AVG intervention research, as well as suggestions for more successful collaborations between AVG stakeholders. However, it is worth mentioning that the current project was limited to AVGs studied chronically (≥2 sessions) using rigorous intervention designs (experimental designs with ≥2 conditions). Systematic content analyses of AVG studies using alternative study designs (e.g. one-time exposure study, cross-sectional surveys, and focus groups) would add value by providing a more comprehensive picture of the field. Despite our focus on studies that used relatively rigorous experimental intervention designs, we found that the average study quality based on the GRADE rating scale was still relatively low, especially studies focusing on physical fitness outcomes, which had a much lower GRADE score than studies focusing on mixed outcomes. Thus, we encourage researchers designing and developing future AVG interventions to review the seven GRADE rating scale components and strive for higher quality and study rigor. When the number of AVG intervention studies grows sufficiently, future content analyses might focus exclusively on high-quality studies (e.g. those meeting a certain GRADE rating threshold).

Given the large number of published outcome-focused meta-analyses on AVGs, we believe one or more subgroup or umbrella meta-analyses (i.e. a meta-analysis of meta-analyses) is now warranted. Based on the exponential trajectory of AVG growth, we anticipate more systematic reviews and content analyses of AVGs will be published in the future, justifying the first umbrella content analysis (a content analysis of content analyses) of AVGs within the next decade.

Supplemental Material

sj-docx-1-dhj-10.1177_20552076231171232 - Supplemental material for A comprehensive systematic review and content analysis of active video game intervention research

Supplemental material, sj-docx-1-dhj-10.1177_20552076231171232 for A comprehensive systematic review and content analysis of active video game intervention research by Arlen C Moller, Caio V Sousa, Kelly J Lee, Dar Alon and Amy S Lu in DIGITAL HEALTH

Footnotes

Acknowledgments

The authors thank Severn Ringland for his help in proofreading an earlier version of this manuscript.

Contributorship

Conceptualization (ASL); Data Curation (CVS, KJL, and DA); Formal Analysis (ACM); Funding acquisition (ASL); Investigation (ACM, CVS, KJL, DA, and ASL); Methodology (ACM, CVS, and ASL); Project Administration (ACM and ASL); Resources (ASL); Software (ASL); Supervision (ACM, ASL, and CVS); Validation (ACM, CVS, and ASL); Visualization (ACM and ASL); Writing – original draft (ACM); Writing – review and editing (ACM, CVS, KJL, DA, and ASL).

Declaration of conflicting interests

The author(s) declared the following potential conflicts of interest with respect to the research, authorship, and/or publication of this article: The first author, Arlen C. Moller, holds a U.S. patent related to an AVG (US20140163705A1); however, no studies related to that AVG were eligible for inclusion in this systematic content analysis and no income has been generated from that patent. The other co-authors declare that they have no conflicts of interest.

Ethical approval

The research methods used for this systematic content analysis involved only aggregated, deidentified data reported in published studies.

Funding

The author(s) received the following financial support for the research, authorship, and/or publication of this article: This project was supported in part by a grant from the National Institute of Diabetes and Digestive and Kidney Diseases (R01DK109316) and a grant from the Northeastern University Institute for Health Equity and Social Justice Research (IHESJR).

Guarantor

ACM

ORCID iD

Arlen C Moller

Supplemental material

Supplemental material for this article is available online.

Notes

References

Tekinbas

Zimmerman

. Rules of play: game design fundamentals. Cambridge, MA, USA: MIT Press, 2003.

Moller

Deci

. Intrinsic motivation. In Maggino F (ed) Encyclopedia of quality of life and well-being research. New York: Springer Nature Press, 2020. DOI: 10.1007/978-3-319-69909-7.

Lieberman

. Health education video games for children and adolescents: Theory, design, and research findings, https://eric.ed.gov/?id=ED422312 (1998, accessed 11 January 2022).

Lau

Smit

Fleming

, et al. Serious games for mental health: Are they accessible, feasible, and effective? A systematic review and meta-analysis. Front Psychiatry; 7, https://www.frontiersin.org/articles/10.3389/fpsyt.2016.00209 (2017, accessed 8 July 2022).

Maheu-Cadotte

M-A

Cossette

Dubé

, et al. Efficacy of serious games in healthcare professions education: A systematic review and meta-analysis. Simul Healthc 2021; 16: 199–212.

D’Anastasio

. Coronavirus fears spark a run on Nintendo’s Ring Fit Adventure. Wired, https://www.wired.com/story/ring-fit-shortage-coronavirus-covid-19/ (accessed 8 July 2022).

Bailey

McInnis

. Energy cost of exergaming: A comparison of the energy cost of 6 forms of exergaming. Arch Pediatr Adolesc Med 2011; 165: 597–602.

Bochner

Sorensen

Belamarich

. The impact of active video gaming on weight in youth: A meta-analysis. Clin Pediatr (Phila) 2015; 54: 620–628.

Gao

Chen

Pasco

, et al. A meta-analysis of active video games on health outcomes among children and adolescents. Obes Rev 2015; 16: 783–794.

10.

Mack

Bayer

Schaeffeler

, et al. Chances and limitations of video games in the fight against childhood obesity—A systematic review. Eur Eat Disord Rev 2017; 25: 237–267.

11.

Dutta

Pereira

. Effects of active video games on energy expenditure in adults: A systematic literature review. J Phys Act Health 2015; 12: 890–899.

12.

Hickman

Popescu

Manzanares

, et al. Use of active video gaming in children with neuromotor dysfunction: A systematic review. Dev Med Child Neurol 2017; 59: 903–911.

13.

Page

Barrington

Edwards

, et al. Do active video games benefit the motor skill development of non-typically developing children and adolescents: A systematic review. J Sci Med Sport 2017; 20: 1087–1100.

14.

Mura

Carta

Sancassiani

, et al. Active exergames to improve cognitive functioning in neurological disabilities: A systematic review and meta-analysis. Eur J Phys Rehabil Med 2017; 54: 450–462.

15.

Taylor

Kerse

Frakking

, et al. Active video games for improving physical performance measures in older people: A meta-analysis. J Geriatr Phys Ther 2018; 41: 108–123.

16.

Zeng

Pope

Lee

, et al. A systematic review of active video games on rehabilitative outcomes among older patients. J Sport Health Sci 2017; 6: 33–43.

17.

Hoffmann

Wiemeyer

. Physical and motivational effects of exergames in healthy adults—protocol for a systematic review and meta-analysis. PLoS One 2022; 17: e0266913.

18.

Kharrazi

. A state-of-the-art systematic content analysis of games for health. Games Health J 2018; 7: 1–15.

19.

Moher

Liberati

Tetzlaff

, et al. Preferred reporting items for systematic 658 reviews and meta-analyses: the PRISMA statement. PLoS Med 2009; 6: e1000097.

20.

National Center for Health Statistics—Adult Physical Activity—Glossary, https://www.cdc.gov/nchs/nhis/physical_activity/pa_glossary.htm (2019, accessed 8 July 2022).

21.

Colcombe

Kramer

. Fitness effects on the cognitive function of older adults: a meta-analytic study. Psychol Sci 2003; 14: 125–130.

22.

Guyatt

Oxman

Vist

, et al. GRADE: An emerging consensus on rating quality of evidence and strength of recommendations. Br Med J 2008; 336: 924–926.

23.

El-Harakeh

Morsi

Fadlallah

, et al. Prioritization approaches in the development of health practice guidelines: a systematic review. BMC Health Serv Res 2019; 19: 692.

24.

Rothman

Baldwin

Hertel

, et al. Self-regulation and behavior change: disentangling behavioral initiation and behavioral maintenance. In: Handbook of self-regulation: research, theory, and applications. 2nd ed. New York, NY, USA: Guilford Press, 2011, pp.106–122.

25.

Schwarzer

. Modeling health behavior change: How to predict and modify the adoption and maintenance of health behaviors. Appl Psychol Int Rev 2008; 57: 1–29.

26.

Eldredge

LKB

Markham

Ruiter

RAC

, et al. Planning health promotion programs: an intervention mapping approach. San Francisco, CA, USA: John Wiley & Sons, 2016.

27.

Stevens

Rees

Coffee

, et al. A social identity approach to understanding and promoting physical activity. Sports Med 2017; 47: 1911–1918.

28.

González-Cutre

Sicilia

. The importance of novelty satisfaction for multiple positive outcomes in physical education. Eur Phys Educ Rev 2019; 25: 859–875.

29.

Lakicevic

Gentile

Mehrabi

, et al.

Make fitness fun: Could novelty be the key determinant for physical activity adherence?

Front Psychol 2020; 11. Epub ahead of print 15 October 2020. DOI: https://doi.org/10.3389/fpsyg.2020.577522.

30.

Baranowski

Hong

, et al. The narrative impact of active video games on physical activity among children: a feasibility study. J Med Internet Res 2016; 18: e6538.

31.

Fortier

Duda

Guerin

, et al. Promoting physical activity: development and testing of self-determination theory-based interventions. Int J Behav Nutr Phys Act 2012; 9: 20.

32.

. Narrative appeals. In: Ho

Bylund

van Weert

(eds) The international encyclopedia of health communication. Hoboken, NJ, USA: Wiley, 2023, pp.1–8.

33.

Yee

. The gamer motivation profile: What we learned from 250,000 gamers. In: Proceedings of the 2016 annual symposium on computer-human interaction in play, p.2. New York, NY, USA: Association for Computing Machinery.

34.

Hwang

. Narrative and active video game in separate and additive effects of physical activity and cognitive function among young adults. Sci Rep 2018; 8: 1–12.

35.

Sousa

Fernandez

Hwang

, et al. The effect of narrative on physical activity via immersion during active video game play in children: Mediation analysis. J Med Internet Res 2020; 22: e17994.

36.

Pelarski

Alon

, et al. The effect of narrative element incorporation on physical activity and game experience in active and sedentary virtual reality games. Virtual Real 2023. Epub ahead of print 31 January 2023. DOI: https://doi.org/10.1007/s10055-023-00754-7.

37.

Pahwa

. How the heck is peloton the best-paying music streaming service? Slate, https://slate.com/culture/2021/07/peloton-music-royalties-spotify-apple-music.html (2021, accessed 9 June 2022).

38.

Ganiti

Politopoulos

Tsiatsos

. The impact of background music on an active video game. In: Auer

Tsiatsos

(eds) Interactive mobile communication technologies and learning. Cham: Springer International Publishing, 2018, pp.18–28.

39.

Borda

Molnar

Kostkova

. Serious games and participatory research in public health. In: Proceedings of the 9th international conference on digital public health, pp.133–133. Marseille, France: ACM.

40.

Morton

Barnes

Michalak

. Participatory digital health research: a new paradigm for mHealth tool development. Gen Hosp Psychiatry 2020; 66: 67–69.

41.

Straker

Fenner

Howie

, et al. Efficient and effective change principles in active videogames. Games Health J 2015; 4: 43–52.

42.

Orlowski

Lawn

Venning

, et al. Participatory research as one piece of the puzzle: a systematic review of consumer involvement in design of technology-based youth mental health and well-being interventions. JMIR Hum Factors 2015; 2: e12.

43.

Figueroa

Murayama

Amorim

, et al. Applying the digital health social justice guide. Front Digit Health; 4, https://www.frontiersin.org/article/10.3389/fdgth.2022.807886 (2022, accessed 23 March 2022).

44.

Jazwinska

. The tech industry controls CS conference funding. What are the dangers? Freedom to Tinker, https://freedom-to-tinker.com/2022/03/11/the-tech-industry-controls-cs-conference-funding-what-are-the-dangers/ (2022).

45.

Maung

Crawfis

Gauthier

, et al. Development of recovery rapids—a game for cost effective stroke therapy. 2014. Epub ahead of print 3 April 2014. DOI: https://doi.org/10.13140/2.1.3968.8643.

46.

Poretski

Arazy

Placing value on community co-creations: a study of a video game ‘modding’ community. In: Proceedings of the 2017 ACM conference on computer supported cooperative work and social computing, pp.480–491. New York, NY, USA: Association for Computing Machinery.

47.

Johannes

Vuorre

Przybylski

. Video game play is positively correlated with well-being. R Soc Open Sci 2021; 8: 202049.

48.

Vuorre

Johannes

Magnusson

, et al. Time spent playing video games is unlikely to impact well-being. R Soc Open Sci 2022; 9: 220411.

49.

ReCODE Health. ReCODE Health , https://recode.health/about/ (accessed 10 August 2022).

50.

Baranowski

Lyons

. Scoping review of Pokémon Go: comprehensive assessment of augmented reality for physical activity change. Games Health J 2020; 9: 71–84.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.11 MB