Serious Games, Gamification, and Serious Mental Illness: A Scoping Review

The role of gaming and the use of gamification in promoting treatment in health care settings is an emerging area of practice (1, 2). However, there is currently little evidence to support the use of gaming in mental health settings.

The purpose of this research was to map key concepts in the use of serious games and gamification to promote treatment of serious mental illness (schizophrenia, bipolar disorder, and schizophreniform disorder). Specifically, this study identified the common characteristics of gamified interventions, the effectiveness of the interventions in promoting treatment outcomes, and the acceptability and feasibility of their use in clinical settings. Findings are discussed in relation to the research question and its objectives. The article also includes recommendations to support future design and development of serious games.

Basic Concepts and Theories

Scoping reviews clarify, define, and develop conceptual boundaries within a topic area (3). They have particular relevance when emerging evidence is being reviewed, when there is a paucity of high-quality literature, and when a range of study designs need to be considered (4–8). Scoping reviews can have a number of aims, including to identify and map key concepts underpinning research within a topic area, to identify research gaps, and to inform future research and practice (4, 6, 9).

Persons who experience serious mental illness can present with an increased risk of re-emergence of elevated symptoms and disengagement from treatment (10–12). Disengagement presents with substantial cost—treatment may cost up to four times more for patients who disengage than for those who do not relapse (13, 14). Disengagement can reflect a lack of perceived fit between service users and the treatment they receive. More often than not, traditional treatment has been restricted to symptom reduction, lifelong adherence to antipsychotic medication, and guarded expectations of a return to premorbid functioning (15–17). For service users, these approaches have fostered a clinical environment that has nurtured learned helplessness; institutionalization (18, 19); and a deep sense of hopelessness, loss, and despair that continues to impede recovery today (12, 20, 21). Service users often reject treatment because they do not value it or do not believe that they can benefit from it, and new approaches are needed to engage this population (22, 23).

Serious games are constructed to engage the player and use pedagogy to impart instruction into the game experience (2, 24). As such, they pose a novel opportunity to introduce and engage service users in treatment, often when engagement has otherwise been unsuccessful (24–26).

Serious games are a subcategory of games in which the goal of the game is positioned outside the game itself (27–30). Serious games convey ideas and values that persuade players, influence their real-life thoughts and actions, and have an explicit, thought-out educational purpose (31, 32). Serious games have been collectively described in the literature as applied games, pervasive games, games for health, and alternate-reality games (33–35). For the purpose of clarity and simplification of terminology, the terms serious games and gamification will henceforth be used. Often serious games are digital, but they are not limited to that format (30). Serious games are an effective medium for sustaining idea generation, improving communication, the resetting of power relations within a group, and creating a nonthreatening and engaging learning environment (36). The gamification of treatment involves the utilization of these characteristics to engage the service user in an activity that promotes treatment.

The term gamification was coined at the beginning of the 21st century (37–40) and involves the redesign of one’s life activities through the structure, design, and methodology of games (34, 40–42). Gamification provides a safe and controlled environment in which to practice and learn new behaviors in response to real-life challenges. Once learned, these lessons can then be applied in uncontrolled, real-life situations (39, 43). Most games rely on “pointsification”—the use of points, timed performance, badges, or leader-board systems—to reward players for completing an activity (37, 40, 43). The success of gamification, however, does not depend exclusively on pointsification, nor does it lie in the use of technology to motivate and engage players (42). Rather, attainment of the game’s goals is supported by the application of three separate but overlapping elements—game mechanics (rules and procedures), dynamics (interactions, behaviors, and player experience), and instructional design (pedagogy) (37). The goals are not achieved within the game, such as winning a race; instead, players are rewarded for meeting goals positioned outside the game, such as the acquisition of skills, knowledge, and experiences (37, 39, 43).

The use of gamification to structure learning environments that have treatment utility offers a new approach to engaging mental health service users (2). However, current knowledge and understanding of the use of gamification with those who experience serious mental illness remain nascent.

Practitioners who wish to explore the use of serious games and gamification in their clinical settings as a way to improve service delivery, user experience, and engagement could benefit from the findings and recommendations of this review. The findings could also benefit researchers developing serious games or gamified interventions for the treatment of serious mental illness.

Methods

To answer the research question, this review follows the five stages of a scoping review: identifying the initial review question; identifying relevant studies; study selection; charting the data; and collating, summarizing, and reporting the results (4, 44).

Stage 1: Identifying the Initial Review Questions

The review question was, What is the current state of knowledge about how games and gamification are used to promote treatment of service users who experience serious mental illness? The process of identifying the review questions followed an iterative, back-and-forth analysis of earlier findings and new insights, new search terms, and changing review questions (4, 45).

Stage 2: Identifying Relevant Studies

AMED, ASSIA, CINAHL, Cochrane, EMBASE, Medline, and PsychINFO databases from January 2000 to January 2019 were searched. Because the use of gamification in treatment is a new phenomenon, it was considered unlikely that relevant articles from before 2000 would be found. Prieto de Lope and Medina-Medina’s (46) search from 1990 to 2012, for example, found that 54% of articles on the subject of serious games were published after 2007. Included articles focused on the use of off-the-shelf or bespoke games (unique games designed to meet the needs of a specific population or condition) and gamification to promote treatment or treatment outcomes for people who experience serious mental illness.

Key words were established by following an initial broad search of the literature and consultation with a university librarian who is a specialist on literature reviews. The librarian’s input was valuable for the refinement of search terms and identification of relevant data bases. The search terms included mental illness, serious mental illness, psychosis schizophrenia, schizophreniform and bipolar disorder and key words such as games, gaming, serious games, and games for health, engagement and treatment.

The purpose of a scoping review is to comprehensively address a broad review question, but parameters are required to guide the search strategy and the formation of the research question (7, 44). Person, concept, and context (PCC), as recommended by Peters et al. (9), was used to structure the inclusion criteria as follows: person—mental illness, serious mental illness, psychosis (schizophrenia, schizophreniform disorder, and bipolar disorder); concept—games, gaming, serious games, pervasive games, alternate-reality games, applied games and games for health and promotion of treatment, interventions, engagement and reduction in illness and symptomology and behavior change and health behaviors and acceptability and feasibility of game and technology; and context—adult (age >18), English language and service users receiving treatment for serious mental illness and community, inpatients, secure setting and rehabilitation services.

The inclusion criteria and exclusion criteria were used to refine the search strategies and provided a framework for searching the literature (Table 1). Continual refinement of search strategy in response to the iterative process of investigation was ongoing and helped create greater methodological clarity and rigor (5).

Stage 3: Study Selection

Through use of the search terms, 386 articles were identified. Twenty-two additional articles were identified through skimming of reference lists and review of Web sites of various organizations, such as the Serious Games Institute and the Serious Games Association. Forty-seven duplicates were removed, leaving a final total of 339 identified articles. We applied the inclusion and exclusion criteria to the abstracts of these articles. Abstracts that met the inclusion and exclusion criteria progressed to the next stage, and full articles were obtained. If uncertainty existed as to the appropriateness of an article, the full article was obtained. At the end of this process, 264 articles were rejected and 75 articles were identified for full-article review. We then reviewed these articles. Twenty articles were confirmed to be appropriate for inclusion. Articles were excluded because they did not have game structure or gamification as their primary focus; were not focused on treatment and treatment outcomes; had an emphasis on education; were concerned with an excluded population, such as adolescence and stroke patients; or used games as a control condition. This process followed the Preferred Reporting of Items for Systematic Review and Meta-Analysis (47) (see diagram in the online supplement).

Stage 4: Charting the Data

Articles were summarized and recorded according to study/author/year, location, intervention, study design/sample, and outcome.

Stage 5: Collating, Summarizing, and Reporting Results

Through use of thematic content analysis, six themes were identified. Thematic content aims to provide a map of content and topics across a wide data set (5, 48, 49). Following recognized stages of thematic analysis (49, 50), two researchers coded each article individually and then collaborated to review their findings until a consensus was achieved. Where differences occurred, both researchers returned to the codes and worked together to find agreement. During this process, proposed themes were often discarded, and new themes were created from the codes. This was done until a consensus on theme content and title was achieved.

Results

This scoping review generated 20 articles from 12 countries (51–70) (Table 2). Some studies involved participants from more than one country. Six articles were from the United States, five from France, three each from Switzerland and the United Kingdom, two each from Belgium and the Netherlands, and one each from Australia, Italy, Japan, Korea, and Portugal. Thirteen studies were of quantitative design, five of qualitative design, and two of mixed design. All games involved service users who experienced schizophrenia, schizoaffective disorder, and delusional disorders. The number of study participants ranged from four to 172, but most studies had fewer than 55 participants, although two had 135 and 172 participants, respectively.

Seven articles involved games of physical activity; six were original, bespoke games designed for a specific purpose and population; four involved the use of card games; two were online games; and one involved the use of a game to evaluate an intervention. The seven physical activity games involved Xbox (N=6) and Nintendo Wii (N=1) games and investigated the effect of game play on physical performance such as balance, hand grip, peak VO₂, walking speed, and clinical symptoms of schizophrenia. The six bespoke video games included a computer-based, role-playing game; a prototype discharge planning gaming; a virtual game in an imaginary town to improve cognitive abilities; a cognitive-bias game that requires players to appraise social situations; a game designed to consult service users in the redesign of their environment; and a medieval fantasy game that provides psychoeducation for psychosis.

As study outcomes, Olivet et al. (51) investigated hope, recovery, empowerment, and stigma; Amado et al. (52) investigated the clinical symptoms of schizophrenia and cognitive skills of visual scanning, verbal and visual working memory, and visuospatial abilities; and Shrimpton and Hurworth (53), van der Krieke et al. (54), Reynolds et al. (55), and Fitzgerald et al. (56) investigated adherence, feasibility, and acceptability. Three of the four card games investigated Michael’s Game, an 80-card game designed to familiarize professionals and service users with cognitive-behavioral therapy (CBT) for psychosis; the fourth game encompasses a virtual reality (VR) card game that helps service users identify emotions from facial expressions. Outcomes for these studies focused on clinical symptoms of schizophrenia (57–59) and adherence, feasibility, and acceptability (60). Online games included two studies. One involved a cognitive-remediation game that assesses performance in memory, attention, working memory, psychomotor speed, and executive function as outcome measures. The second compared the playing of video games with watching films and assessed clinical symptoms of schizophrenia. Games used to evaluate interventions involved one game that utilized a collaborative group game designed to explore patient and practitioner experience of receiving C-Co CBTp (a variant of CBT for high-security settings) and assessed user experience of receiving therapy as the outcome measure.

Nearly all games were viewed as acceptable formats for engaging service users and in promoting treatment outcomes. Thirteen games were associated with improvement in all measured outcomes, four showed mixed results, and three yielded no differences. Active, off-the-shelf games such as Xbox games were successful in promoting physical exercise (61–63). Other studies found that active games were associated with a significant increase in frontal lobe blood flow (64) and a clinically important difference in walking speed (57). Leutwyler et al. (61, 62) found that these games were associated with high adherence and acceptability, but Leutwyler et al. (60) and Campos et al. (67) found that active games produced no significant difference in objectively measured physical activity.

Games utilized to address the symptoms of schizophrenia were also successful, for example, in the reduction of positive symptoms (52, 52, 57–59, 63) and improvement in quality of life (52, 64). Nieman et al. (65) and Amado et al. (52) investigated cognition and found significant improvement in verbal, working, prospective, and retrospective memory. Olivet et al. (51) measured changes in hope, recovery, stigma, and empowerment associated with use of their computer-based game and found significant improvement in service users’ attitude toward recovery. Service users enjoyed playing van der Krieke et al.’s game (54) but found it not challenging enough. and Shrimpton and Hurworth’s (53) game was rejected outright by players because it was seen as too challenging and complicated and because there was no clear and easily discernible purpose.

Themes in Use of Serious Games

This scoping review identified a number of advantages and opportunities in the use of gamification to promote treatment of people with serious mental illness. These findings are grouped in six themes: game design, ease of use, adjuncts and complements to treatment, problem solving and learning, collaboration, and goal orientation. Table 3 displays the frequency of these themes within the 20 articles.

Game design.

This theme was divided into two interrelated parts—game mechanics (features such as character, plot, points, control stick, or graphics that a player utilizes to engage game content) and service user involvement in the design and development of game mechanics. Fifteen of the 20 articles reviewed referred to the importance of design to the game’s ability to engage players and achieve intervention outcomes.

Seven studies used commercially available video games to deliver exercise programs. Five used originally designed role-playing game or VR formats, and one used a VR agent. All demonstrated high user acceptability, and players considered these formats to be authentic and realistic. Not all studies, however, used modern technology; three investigations involved Michael’s Game, and one used the traditional format of a board game. These games also demonstrated good retention rates and were effective in delivering their intervention outcomes. Having a choice and variety of options, appropriate levels of challenge, easy-to-follow instructions, and familiar, intuitive control devices corresponded to player enjoyment, confidence that they can play the game, and successful delivery of intervention outcomes.

Shrimpton and Hurworth (53), however, found that the complexity of their game’s design had a negative impact on the integration of player engagement with game content. Players disliked the game’s graphics and how the technology mediated their interaction with the game’s character, and they struggled to see the purpose of playing the role of their character within game play. Players in van der Krieke’s (54) cognitive-bias game and Nieman et al.’s (65) cognitive-training game found the games’ content unchallenging, became demotivated, and eventually stopped playing.

Games that had service user involvement early in the design process produced better game mechanics and, therefore, were better able to engage service users in game play and the delivery of intervention outcomes. Slater and Painter’s (66) participatory action study and Reynolds et al.’s (55) user-centered design involved service users at the beginning of the design stage and, as such, successfully delivered their intervention outcomes. The games studied by Fitzgerald et al. (56), Nieman et al. (65), and Olivet et al. (51) all showed promise, but the authors acknowledged a lack of service user involvement in the design process and recommended further game development with service users. Shrimpton and Hurworth (51) and van der Krieke et al. (54) struggled to deliver their intervention outcomes because of irreparable design flaws; in both cases, users had little to no involvement in the design process.

Ease of use.

Ease of use was cited by 13 of the 20 studies. The majority of games required no additional training and could therefore be easily administered by any member of the clinical team (51, 54, 56, 58, 61, 62, 67, 68). Michael’s Game (58, 59, 69) was the only game that required some facilitator training. However, this training was relatively straightforward, involving 2 hours of staff time and required no previous skills or understanding in cognitive therapy. Games that were straightforward, easy to use, and fun and that required no special skills were acceptable to both service users and staff (51, 54–57, 61, 64, 67, 68). For games where ease of use was a challenge, there was also not enough game narrative to contextualize the player’s activity and a lack of clear instructions to inform the player as to what to do (53, 54). Easy-to-use games were also enjoyable to play (52, 55, 57). Participants who enjoyed playing games also reported an improved quality of life and said that they would recommend the intervention to others (61, 62, 68).

Adjuncts and complements to treatment.

Eleven articles identified the opportunities for games and gamification to be an adjunct or complement to existing treatment programs. Seven studies used commercially available Xbox or Nintendo Wii games to increase physical activity among service users. The authors suggested that these games be considered part of an overall health program for this population, providing a gateway for a more active lifestyle and a more enjoyable and accessible alternative to traditional forms of exercise (57, 61, 62, 67, 68). Shimizu et al.’s (64) investigation of the association between active games and frontal lobe function also found improvement in health-related quality of life measures.

Problem solving and learning.

Eleven articles identified the inclusion of problem solving as important. Problem solving was enabled during game play by the use of characters (53, 70), the navigation of complex virtual environments (51, 52, 55), or scenarios (54, 59). The gaming format enabled the opportunity for meaningful discussion, provided a safe way to engage users in challenging situations and their solutions, and served as a way of identifying and practicing alternative attributions. Participants reported valuing the opportunity to practice real-life scenarios, to employ autonomy in decision making, and to reflect, with peers, on the outcomes of their actions (51, 52, 55). However, authors noted that learning can successfully occur within a game context only if players understand that learning is one of the objectives of the game (53, 54).

Collaboration.

Having an opportunity for collaboration was identified in eight articles as helpful for enhancing participation. Khazaal et al. (69), Khazaal et al. (58), Khazaal et al. (59), and Fitzgerald et al. (56) found that the collaborative nature of their games promoted discussions between patients and staff, encouraged active peer-to-peer interactions, and increased service-user participation. Leutwyler et al. (61) and Leutwyler et al. (62) found that the group play options in commercially available games enhanced enjoyment and increased participation. Participants liked being able to cheer on one another and coach one another to improve goal achievement (61, 62). Collaboration facilitated the sharing of experiences, which enhanced learning from others and created a strong sense of hope among participants (66). Collaboration in the design process created a more realistic and enjoyable game experience.

Goal orientation.

In the articles reviewed, it did not matter whether the game was delivered via the use of complex technology or as a simple card game. If players have a clear purpose and the goals that indicate progress and success are clearly defined, the game is better accepted and engagement is enhanced.

Informed Consent and Institutional Ethics Approval

Of the studies reviewed, informed consent and institutional ethics approval was obtained in studies by Khazaal et al. (69), Han et al. (63), Khazaal et al. (58), Leutwyler et al. (61), Leutwyler et al. (62), Campos et al. (67), Khazaal et al. (59), Kimhy et al. (68), Leutwyler et al. (60), Amado et al. (52), Reynolds et al. (55), Shimizu et al. (64), Leutwyler et al. (57), and Olivet et al. (51). Without giving details, Slater and Painter (66) made reference to an institutional review process, and the study by Fitzgerald et al. (56) was a service evaluation. Studies by Shrimpton and Hurworth (53), van der Krieke et al. (54), Nieman et al. (65), and Oker et al. (70) made no reference to informed consent and institutional ethics approval.

Discussion

Games that were well-designed and that integrated seamlessly with the game’s technology had higher levels of feasibility and acceptability and were more successful in engaging users in game content and in achieving treatment outcomes. The success of the game was not determined per se by the technology used. It did not matter whether the game used system-based technology (such as Xbox), bespoke technology, or a traditional game format. What mediated success was the seamless integration of the game’s structure, content, and purpose with the technology used for delivery. If the structure and purpose of the game made sense to the player and the technology easily facilitated game play and engagement with game content, the player was more likely to feel confident regarding their skills to play the game, be motivated to play, and continue playing or return to play at another time.

User involvement early in game design improved the design of a game and its capacity to successfully deliver intervention outcomes. Of the games reviewed, those that facilitated user involvement were better able to identify and eliminate design problems and barriers at an early stage, which had a positive impact on uptake and sustained use (51, 55, 60, 61, 62, 66–70). Games that contained real-life scenarios requiring problem solving and collaboration between service users and staff provided a clear purpose for playing and an opportunity for collaboration. Third-person agents—such as an avatar (70) or the fictional character of “Michael” in Michael’s Game (65)—or the use of electronic environments (51, 54, 55) helped to make the learning activity more realistic and the game more interesting to play.

The literature suggests that disengagement may result from a poor fit between the treatment provided and how meaningful it is to the service user. For most of the service users who participated in the reviewed studies, there seemed to be an acceptable fit between the demands of the intervention and the value they placed upon it as something worth engaging in. The key to improving fit between user and game was user involvement in game design and the ease by which game mechanics facilitated game play. When this goal was achieved, service users were better able to engage in the game and the gamified intervention it was designed to deliver. Because the intervention was gamified, it remained familiar, easy to use, enjoyable, and nonthreatening to the user. In this way gamification has the potential to overcome the barriers to engagement that have so often confounded traditional treatment approaches for serious mental illness.

Gamification can take many different forms. Studies reviewed in this article utilized gaming platforms that ranged from off-the-shelf commercial games and Internet technology to bespoke electronic games and traditional card and board game formats. Success seemed to be related to user involvement early in design, seamless integration with technology so as to enable ease of use, clear game purpose, and realistic learning scenarios. Those wishing to design games and to gamify interventions to engage this population will therefore need to be mindful of these important determinants.

All reviewed studies involved service users who experience serious mental illness and receive treatment in real-world settings, such as inpatient, outpatient, and secure services and early intervention in psychosis. One study was an expert review, and another an action participation study; both studies viewed service users as experts by experience. As such, the transferability of findings from study settings to real-world clinical practice is possible.

The studies investigated ranged from randomized controlled trials (RCTs) (59, 68) and studies with a quasi-experimental design to qualitative expert reviews (51) and participation action research design (66). Of the qualitative investigations, six studies had no control condition or only a limited description of their use of a control group, others had low to modest sample sizes, and two used self-report questionnaires. A number of studies did not collect pretest data or conduct follow-up. Qualitative studies also varied in their data collection methods. As such, the findings of this review should be viewed within this disparate and limited context.

However, the themes identified were consistent across different study designs and methods. Four of the six themes (game design, ease of use, adjunct to treatment, and problem solving) were present in over half of the studies reviewed, and no theme was present in fewer than six studies. The use of serious games and gamification to promote treatment had high feasibility and acceptability with service users who experience serious mental illness and with staff who work with them. When service users rejected games, they did so because of poor design—particularly if the game’s mechanics failed to create a game play experience that was meaningful and therefore motivating to play. When players were motivated to play, engagement in game content was possible.

User involvement early in game design may be the biggest challenge for those wishing to utilize serious games and gamification in clinical settings. It can be tempting for the clinician to provide or design a game with content that they think the service user needs and will benefit from. But enabling a perceived fit between service user and treatment is important for engagement. Despite the diverse range of study designs and methods presented in this review, games that were able to facilitate a better fit were more successful in engagement than those that did not. The six themes described in this review provide the reader with insight into how this perceived fit can be supported across a range of game and study designs.

Conclusions

People who experience serious mental illness and the services that provide care and treatment for them face very real challenges, not least of which include relapse, disengagement, and cost of care. Disengagement can result from a poor fit between the treatment offered and the service user; therefore new approaches that successfully engage the patient in care are desired. Serious games and gamification have a long history of use in management and health education. More recently, they have been used as novel approaches to the treatment of hard-to-engage populations, such as persons with serious mental illness.

This scoping review set out to investigate the current state of knowledge about how games and gamification are used to promote treatment of serious mental illness. Our conclusion is that such approaches have high user and provider feasibility and acceptability in promoting user engagement in treatment. Success was mediated by user involvement in the game design process.

Many of the studies reviewed were of low design quality, although three RCTs were promising in their design and findings. That is not surprising considering that the use of gamification in treatment of serious mental illness is an emerging area of practice. Scoping reviews have particular relevance when emerging evidence is being considered or a range of study designs need to be considered (4–8). Better-quality investigations that include controlled designs and postintervention follow-up are needed to improve our understanding of game design the use of gamification, and how that knowledge can be transferred to practice.