Context matters for the effectiveness of video games for young people’s mental health Aniek Wols
Context matters for the effectiveness of video games for young people’s mental health Aniek Wols
Funding This work was supported by the Dutch Research Council (NWO Talent grant, 406-16-524), and the Behavioural Science Institute, Radboud University Nijmegen, the Netherlands. Colophon Cover design: Arina van Londen Lay-out: Midas Boering - Ridderprint, the Netherlands. Printing: Ridderprint, the Netherlands. 978-94-6506-240-2 Copyright © 2024 by A. C. Wols All rights reserved. No part of this publication may be reproduced, stored in a retrieval system of any nature, or transmitted in any form by any means, electronic, mechanical, photocopying, recording or otherwise, without prior written permission of the author.
Context matters for the effectiveness of video games for young people’s mental health Proefschrift ter verkrijging van de graad van doctor aan de Radboud Universiteit Nijmegen op gezag van de rector magnificus prof. dr. J.M. Sanders, volgens besluit van het college voor promoties in het openbaar te verdedigen op woensdag 22 januari 2025 om 16.30 uur precies door Antje Cornelia (Aniek) Wols
Promotoren Prof. dr. Anna Lichtwarck-Aschoff (Rijksuniversiteit Groningen) Prof. dr. Isabela Granic Manuscriptcommissie Prof. dr. Tibor Bosse Prof. dr. Maaike H. Nauta (Rijksuniversiteit Groningen) Prof. dr. Rutger C.M.E. Engels (Erasmus Universiteit Rotterdam)
CONTENTS Chapter 1 General introduction 9 Chapter 2 Effectiveness of applied and casual games for young people’s mental health: A systematic review of randomised controlled trials 29 Chapter 3 In-game play behaviours during an applied video game for anxiety prevention predict successful intervention outcomes 171 Chapter 4 Mental health outcomes of an applied game for children with elevated anxiety symptoms: A randomised controlled noninferiority trial 197 Chapter 5 The effect of expectations on experiences and engagement with an applied game for mental health 233 Chapter 6 Explicit mental health messaging promotes serious video game selection in youth with elevated mental health symptoms 259 Chapter 7 The role of motivation to change and mindsets in a game promoted for mental health 297 Chapter 8 General discussion 325 Appendices 345 References 346 Dutch summary (Nederlandse samenvatting) 404 Research data management statement 410 Acknowledgements (Dankwoord) 414 Publication list 420 About the author 422
Chapter 1 General introduction Chapter 1 General introduction
10 Chapter 1 Anxiety and depression are the most prevalent mental health problems in youth. Their associated significant negative outcomes and economic consequences for society, call for effective prevention programs. Video games for mental health have been proposed as an alternative delivery approach and potential solution to tackle social and practical access barriers as well as limitations regarding engagement related to conventional prevention programs. The field of games for mental health has been rapidly developing, and research on both applied and casual games has grown extensively; however, a comprehensive overview of the field is lacking. Therefore, Part 1 of the current thesis aims to provide an overview of the field. Furthermore, research on games for mental health has largely overlooked the influence of nonspecific factors such as expectations, motivation to change, and mindsets on game uptake, engagement and mental health outcomes. The studies described in Part 2 and 3 of this dissertation aim to address this significant gap in the literature. MENTAL HEALTH PROBLEMS IN YOUTH Anxiety and depression rank among the most common mental health issues experienced by youth (Costello et al., 2003; Global Burden of Disease Collaborative Network, 2020; D. Knopf et al., 2008; Merikangas et al., 2010; Polanczyk et al., 2015). Although strongly related to each other and showing high comorbidity rates (Axelson & Birmaher, 2001; J. R. Cohen et al., 2014; Cummings et al., 2014), symptoms of anxiety and depression pertain to different constructs and represent different disorders with distinct features (Hale III et al., 2009; Seligman & Ollendick, 1998). Anxiety is a natural, common feeling of nervousness in response to dangerous, stressful or unfamiliar situations (American Psychiatric Association, 2013). One’s reaction to these situations may be physiological (e.g., palpitations, sweating, tense muscles), behavioural (e.g., avoidance of feared objects, stressful situations or people), and cognitive (e.g., increased attention, worrying) (American Psychiatric Association, 2013). While anxiety is a useful and adaptive emotional response to stress, facilitating avoidance of danger or coping with challenging situations, it becomes maladaptive when intense levels of anxiety persist and interfere with social or occupational functioning (Beesdo et al., 2009; Davis III, 2009; National Library of Medicine). Subsequently, an anxiety disorder may develop, which is characterised by intense feelings of anxiety, excessive worry about future events or activities, apprehensive expectations, increased attentional bias towards threat-related information, selective memory processing, and
11 1 GENERAL INTRODUCTION distorted judgements of risk (American Psychiatric Association, 2013; Gelder et al., 2005; National Library of Medicine, 2016; World Health Organization, 2010). Core symptoms of depression, on the other hand, include a depressed mood (e.g., feeling empty or hopeless, persistent unhappiness) and loss of interest or pleasure in most activities (American Psychiatric Association, 2013; Thapar et al., 2012; World Health Organization, 2010). Additional symptoms may include sleep disturbances, change in eating patterns and subsequent change in weight, loss of energy, feeling worthless or guilty, diminished ability to concentrate or make decisions, thinking or moving slower or showing restless behaviour, and suicidal ideation (American Psychiatric Association, 2013). People with depression experience significant distress and impairment in social or occupational functioning (American Psychiatric Association, 2013; Jaycox et al., 2009; Verboom et al., 2014). Unlike depression, anxiety disorders usually have their onset in childhood. They are one of the earliest form of mental health problems to emerge in childhood and the most prevalent disorders in children (Bandelow & Michaelis, 2015; Palitz & Kendall, 2020). Already at a young age, approximately 20% of children are diagnosed with an anxiety disorder (Beesdo et al., 2009; Chavira et al., 2004; Kroes et al., 2001). Moreover, up to 49% of children are impaired by subclinical levels of anxiety symptoms (Muris et al., 2000a). During adolescence, anxiety symptoms continue to rise (Grant, 2013; Roza et al., 2003), with 25% of adolescents being diagnosed with an anxiety disorder in the past 12 months (Kessler et al., 2012) and up to 32% experiencing subclinical anxiety symptoms (Balázs et al., 2013). Depression onset sharply peaks during adolescence (Kessler et al., 2012; Merikangas et al., 2010; Roza et al., 2003). Prevalence rates of 10% of adolescents being depressed have been reported for large international and US samples (Balázs et al., 2013; Kessler et al., 2012). Furthermore, almost a third of young adolescents report subclinical depressive symptoms (Balázs et al., 2013). Moreover, remarkable differences between boys and girls emerge during adolescence and throughout (young) adulthood. Females are approximately twice as likely as males to experience subclinical levels of anxiety and depressive symptoms, and the same holds for clinical diagnoses of anxiety and depression (Balázs et al., 2013; Hankin et al., 2007; Kessler et al., 2005; Twenge & Nolen-Hoeksema, 2002). Even into young adulthood, anxiety and depression remain the most prevalent mental health problems. Up to 11.7% and 15.6% of young adults, respectively, report to have been diagnosed with an anxiety disorder or major depressive episode in the past year (De Graaf et al., 2012; Kessler & Walters, 1998). By the time they reach early adulthood, a quarter of young adults have
12 Chapter 1 experienced anxiety and/or depression throughout their lives (Copeland et al., 2014; Kessler et al., 2005; Kessler & Walters, 1998). Most concerning is that mental health problems among older adolescents and young adults have been increasing over the past decades (Centraal Bureau voor de Statistiek, 2018; Collishaw et al., 2010; Schoemaker et al., 2019), as well as more recently due to the Covid-19 pandemic (Al Omari et al., 2020; A. Knopf, 2020; Rauschenberg et al., 2021; Salari et al., 2020). Anxiety and depression have been associated with detrimental short- and long-term consequences for the individual (Balázs et al., 2013; World Health Organization, 2001), such as academic underachievement (Owens et al., 2012; Woodward & Fergusson, 2001), problems in social and family functioning (Hoglund & Chisholm, 2014; Woodward & Fergusson, 2001), increased risk for substance (ab)use and dependence (Merikangas et al., 1998; Woodward & Fergusson, 2001), suicidal behaviour (Bolton et al., 2008; Glied & Pine, 2002), other psychopathology (Lavigne et al., 2015; Priddis et al., 2014), unemployment (Fergusson et al., 2001), and early parenthood (Fergusson & Woodward, 2002). These negative outcomes have not only been associated with clinical anxiety and depressive disorders, but also subclinical levels of anxiety and depressive symptoms cause significant distress and impair youth’s functioning in multiple domains (Balázs et al., 2013; Roza et al., 2003). Moreover, subclinical anxiety and depressive symptoms put individuals at significant risk for later development of full-blown anxiety and depressive disorders (Aalto-Setälä et al., 2002; Copeland et al., 2014; Lewinsohn et al., 2000; D. S. Pine et al., 1999). Besides individual consequences, (subclinical) anxiety and depression also have societal consequences. They are the leading cause of disability among youth (GBD Mental Disorders Collaborators, 2022; Gore et al., 2011; World Health Organization, 2021) and impose an enormous burden on health care costs (Bodden et al., 2008) and other (indirect) societal costs, such as absenteeism at work, sick leave and incapacity for work (Meijer et al., 2006; Smit et al., 2006). It is known that untreated symptoms persist into adulthood, and that recurrence and chronic courses of anxiety and depression are common (Asselmann & Beesdo-Baum, 2015; Kovacs et al., 2016; Reef et al., 2009), continuing to affect youth’s daily life and increasing risk for further health issues (Balázs et al., 2013; Essau et al., 2002). Given the high rates of anxiety and depressive symptoms in youth, as well as their severe impact on individuals’ daily and future functioning and economic consequences, it is of critical importance to prevent these symptoms from exacerbating in to full-blown disorders.
13 1 GENERAL INTRODUCTION CURRENT PREVENTION PROGRAMS Prevention programs can be classified into three types, namely universal, selective, and indicated prevention (Garber & Weersing, 2010; Munoz et al., 2010). All three levels of prevention aim to decrease symptoms and/or the likelihood that disorders develop by reducing risk factors and increasing resilience (McGorry et al., 2011). Universal prevention programs target the entire population regardless of individuals’ risk status or symptoms, but some broader target groups may be distinguished. For anxiety and depression specifically, these broader target groups may be: women, young people, the elderly, ethnic minorities, people with low socio-economic status or little social support (Meijer et al., 2006). Selective prevention programs target specific subgroups of individuals who are known to have an increased risk of developing a disorder. Risk factors for anxiety and depression include having a parent with psychopathology, having a chronic disease, problems in social contacts (Meijer et al., 2006) or an avoidant coping style, or showing behavioural inhibition (Rapee, 2002). Indicated prevention programs are delivered to individuals who already show symptoms of anxiety or depression, which do not (yet) meet the diagnostic criteria of the disorder. This type of prevention is about early detection and intervention to prevent symptoms from getting worse (Meijer et al., 2006). Research indicates that selective and indicated (i.e., targeted) prevention programs generally have larger effect sizes and sustained effects over longer time periods compared to universal approaches in preventing and reducing symptoms of anxiety and depression in youth (Horowitz & Garber, 2006; Merry, 2007; Stice et al., 2009; Stockings et al., 2016; Teubert & Pinquart, 2011). Prevention programs for anxiety and depression often target transdiagnostic mechanisms related to both disorders (Butler et al., 2006; Schaeuffele et al., 2021). Anxiety and depression tend to have a high rate of comorbidity; not only the disorders often coexist, but also symptoms of anxiety and depression are highly correlated (Axelson & Birmaher, 2001; Balázs et al., 2013; J. R. Cohen et al., 2014; Cummings et al., 2014). Furthermore, overlapping risk factors and similar underlying mechanisms play a role in the onset, exacerbation and maintenance of the disorders. For example, it is known that parental psychopathology, negative parenting behaviour and parenting stress increase the risk for youth to develop mental health problems (Bijl et al., 2002; Connell & Goodman, 2002; Fox et al., 2010; Kane & Garber, 2004; Knappe et al., 2009; Lieb et al., 2002; Merikangas et al., 1998; Needham, 2008). Additionally, females, youth with (chronic) stress or those who have been through traumatic events have a higher risk as well (Chaplin et al., 2009;
14 Chapter 1 De Graaf et al., 2010; Garber et al., 2002; Ge et al., 1994; Larson & Ham, 1993; Meijer et al., 2006). Regarding underlying mechanisms, both anxiety and depression are characterised by cognitive biases such as negative automatic thoughts, maladaptive beliefs and behaviours, negative affect, elevated distress and dysfunctional emotions (Calvete et al., 2013; Chu & Harrison, 2007; Cole et al., 2008; Dozois et al., 2009; Ehring & Watkins, 2008; Farchione et al., 2012; Harvey et al., 2004; McEvoy et al., 2013; Muris et al., 2005; Trosper et al., 2012). Moreover, clinical reliability in distinguishing anxiety and depression from one another has proven to be low (T. A. Brown et al., 2001). Taking this into account, as well as the fact that anxiety and depression share similar underlying mechanisms and cognitive processes, prevention efforts often focus on transdiagnostic mechanisms underlying both anxiety and depression, which potentially also lead to a larger benefit of prevention programs (Chu et al., 2015; Dozois et al., 2009). Most prevention programs for anxiety and depression include elements of cognitive behavioural therapy (CBT) (Butler et al., 2006; Kendall, 2011; Mychailyszyn et al., 2012), which aims to address and target the underlying mechanisms associated with anxiety and depression (Calvete et al., 2013; Chu & Harrison, 2007; Cole et al., 2008). CBT is designed to focus on the interplay between thoughts, feelings, and behaviours, and to practice adaptive responses to difficult events (Compton et al., 2004). Individuals learn to identify, challenge and replace dysfunctional or negative thoughts with more adaptive thoughts and behaviours (Beck, 2005; Nolen-Hoeksema, 2001). CBT-based prevention programs typically include various therapeutic techniques, such as problemsolving skills, cognitive restructuring, family communication skills training, exposure therapy, pleasant activity scheduling, and behavioural activation (Compton et al., 2004; Kendall, 2011). These techniques help individuals develop effective coping skills and challenge negative thinking patterns, with the goal to improve negative affect, decrease distress and increase cognitive coping, subsequently reducing symptoms and the risk of developing an anxiety or depressive disorder (Calvete et al., 2013; Chu & Harrison, 2007; Cole et al., 2008; Sander & McCarty, 2005). LIMITATIONS AND BARRIERS OF CONVENTIONAL PREVENTION PROGRAMS Although CBT-based prevention programs are widely used and several meta-analyses have found them to be effective in reducing symptoms and
15 1 GENERAL INTRODUCTION preventing the onset of disorders in some cases, the overall effectiveness of these programs is generally considered to be small to moderate (Beardslee et al., 2013; Christensen et al., 2010; Fisak et al., 2011; Merry, Hetrick et al., 2012; Mychailyszyn et al., 2012; Rasing et al., 2017; Stockings et al., 2016). Additionally, conventional prevention programs face several social and practical access barriers that hinder the implementation, accessibility, and effectiveness of these programs. First, perceived stigma around mental health can discourage youth from seeking or participating in prevention programs (Clement et al., 2015; Gulliver et al., 2010; Mukolo & Heflinger, 2011; Salloum et al., 2016). Furthermore, access and resources may be limited depending on, for instance, geographical location, socioeconomic status, healthcare and ethnic disparities, and shortage of mental health professionals and long waiting lists (Bijl et al., 2003; Collins et al., 2004; Kataoka et al., 2002; Wells et al., 2001). Additionally, high costs are an important barrier that impedes youth from seeking help, especially those who need it the most (Collins et al., 2004; Salloum et al., 2016). Moreover, limitations and challenges have been identified in the delivery of CBT-based prevention programs. It is important to note that these limitations specifically relate to how CBT is delivered, rather than the underlying therapeutic principles of CBT itself (Granic et al., 2014; Kazdin, 2011). CBT programs largely rely on a didactic approach, which might not be appealing, engaging and motivating for youth (Crenshaw, 2008; De Haan et al., 2013; Weisz & Kazdin, 2010; World Health Organization, 2012). Another limitation in many CBT approaches is the considerable disparity between what youth learn and know, and what they actually do in their daily lives (Eichstedt et al., 2014). Recognising this gap between knowledge and behaviour, CBT programs often incorporate activities such as role-playing, problem-solving exercises, and homework assignments (Kendall, 2011), but these activities are often disconnected from authentic emotional experiences and lack real-life context hampering the transfer of learned skills to other contexts and youths’ everyday lives (Granic et al., 2014). Finally, the aforementioned social barriers (e.g., stigma), practical barriers (e.g., high costs) and the unappealing didactic approach may contribute to the large percentage of premature dropout of mental healthcare that has been reported (De Haan et al., 2013; Salloum et al., 2016). Taken together, these limitations and barriers of conventional prevention programs highlight the need for alternative delivery approaches that are engaging, motivating, accessible, and cost-effective (P. J. Jones et al., 2019; Kazdin, 2019; Liverpool et al., 2020; Weisz et al., 2019).
16 Chapter 1 VIDEO GAMES AS AN ALTERNATIVE TO CONVENTIONAL PREVENTION PROGRAMS Video games have shown promise in serving as a potential alternative approach, and present a potential solution to tackle the aforementioned limitations (Granic et al., 2014; Kazdin, 2015). Video games offer various advantages when compared to traditional mental health interventions, including prevention programs for anxiety and depression. First, reflected in their ubiquitous use among youth, video games provide a non-threatening and non-stigmatising medium for youth to engage with mental health interventions (Lenhart et al., 2008; McFarlane et al., 2002; McGonigal, 2011; NPD Group, 2011). This can help reduce the perceived stigma associated with seeking help or participating in traditional forms of interventions (Granic et al., 2014). Additionally, the interactive and immersive nature of games is appealing and can enhance engagement and motivation, making it more likely that youth will actively (continue to) participate in the intervention (Fleming et al., 2017; Granic et al., 2014; Kazdin, 2015). Secondly, games allow youth to engage in simulated (emotion eliciting) scenarios: through gameplay youth can practice newly learned skills or strategies repeatedly in a safe and controlled environment, promoting learning and skill transfer to real-life situations (Buday, 2015; Fleming et al., 2017; Granic et al., 2014). Furthermore, games can be designed to be adaptable and personalised, allowing for individualised learning experiences based on users’ needs, preferences, and skills levels. This customisation can enhance the relevance and effectiveness of the intervention (Bakkes et al., 2012; Bakkes et al., 2014; Barnes & Prescott, 2018). Another benefit of video games is that they can easily be distributed and used on various ubiquitous devices such as smartphones, tablets and gaming consoles, making them accessible to a larger population including individuals in remote or underserved areas (Entertainment Software Association, 2017; Fleming et al., 2023; Granic et al., 2014). This potential for scalability offers a cost-effective alternative to traditional approaches, such as in-person therapy or training sessions. Once developed, games can be distributed with relatively low cost and effort (Eichenberg & Schott, 2017; Granic et al., 2014; Kazdin, 2015; Lau et al., 2017). In summary, video games could potentially overcome the shortcomings of conventional programs, improving appeal, reach, uptake and accessibility, and ultimately effectiveness.
17 1 GENERAL INTRODUCTION VIDEO GAMES FOR MENTAL HEALTH Given the aforementioned benefits of video games in comparison to traditional prevention programs, it is unsurprising that in recent years there has been a growing interest among mental health practitioners and researchers in utilising applied video games as a means of promoting mental health (Fleming et al., 2017; Fleming et al., 2023; Granic et al., 2014; Kazdin, 2015; Lau et al., 2017; Townsend et al., 2022). Furthermore, the effects of casual games (i.e., entertainment or commercially available, off-the-shelf games) have been increasingly examined in the field of mental health, because they are affordable and already widely available for the general population and provide repeated training for target behaviours (Ceranoglu, 2010; Colder Carras et al., 2018; Granic et al., 2014; Steadman et al., 2014). Applied games are digital interventions that employ game design elements in an effort to making interventions more enjoyable, motivating and engaging (Schmidt et al., 2015). Their primary aim is to educate or motivate users, and to train or promote behaviour change, other than pure entertainment purposes (Michael & Chen, 2005; Stapleton, 2004; Stokes, 2005; Vajawat et al., 2021). Two categories of applied games can be distinguished, namely ‘gamified’ interventions and ‘serious games’ (Fleming et al., 2017; Schmidt et al., 2015; Vajawat et al., 2021). Gamification refers to adding game elements such as points and rewards, increasing difficulty levels and narratives to interventions (M. Brown et al., 2016; Cheng et al., 2019; Cugelman, 2013; H. J. Park & Bae, 2014; Seaborn & Fels, 2015), without necessarily focussing on playfulness and fun (Fleming et al., 2017). Serious games, on the other hand, are full game experiences: they are designed to be immersive and entertaining as well as incorporating evidence-based therapeutic techniques (Eichenberg & Schott, 2017; Fleming et al., 2014; Zayeni et al., 2020). Concerning mental health, the potential and promising effects of applied games have been shown in several reviews and meta-analyses focussing on a variety of mental health problems (David et al., 2020; Dewhirst et al., 2022; Eichenberg & Schott, 2017; Fleming et al., 2017; Halldorsson et al., 2021; Johnson et al., 2016; Lau et al., 2017; Shah et al., 2018), but also specifically for anxiety (Barnes & Prescott, 2018) and depression (Dias et al., 2018; Fleming et al., 2014; Li et al., 2014; Rasing, Stikkelbroek, & Bodden, 2020). Casual games, on the other hand, are built for player enjoyment and recreational purposes, with no consideration of their therapeutic potential (Deterding, 2015; Fullerton, 2014). While casual games are not explicitly designed as such, they may have a positive impact on mental health. Players
18 Chapter 1 may feel better after playing casual games and skills that may relate to improved mental health can also be learned (Ferguson & Olson, 2013; Olson, 2010; R. Pine, Sutcliffe et al., 2020). Casual games can also trigger positive emotions (McGonigal, 2011; Osmanovic & Pecchioni, 2016; Ryan et al., 2006) which, in turn, aid individuals in expanding their momentary thought-action repertoires and personal resources (Fredrickson, 2001; Quinn et al., 2012). There may also be benefits from games that trigger intense (negative) emotions, which may allow the player to practice coping strategies in a safe environment (Granic et al., 2014). In addition, playing casual games may provide distraction from problems and worries, showing beneficial effects on mood (Bowman & Tamborini, 2012; Colder Carras et al., 2018; Pallavicini et al., 2021). Finally, casual games may evoke the experience of flow, intrinsic motivation and basic need satisfaction, which have been linked to mental health benefits as well (Nakamura & Csikszentmihalyi, 2009; Ryan et al., 2008; Ryan et al., 2006; Sherry, 2004). Indeed, there is growing evidence for the potential of a wide range of casual games to improve emotion regulation (Pallavicini et al., 2018; Villani et al., 2018), general well-being (Halbrook et al., 2019; C. Jones et al., 2014), and alleviate mental health problems, such as depression, anxiety, and stress (Kowal et al., 2021; Lee et al., 2021; Li et al., 2016; Pallavicini et al., 2021; R. Pine et al., 2020; Russoniello et al., 2009) and posttraumatic stress symptoms (Holmes et al., 2009; E.L. James et al., 2015). In conclusion, both applied and casual games have been researched as a means to address the limitations of traditional intervention programs and to increase reach, uptake and accessibility, appeal and effectiveness of prevention and intervention efforts for mental health. In the past decades, research on games for mental health has grown extensively and the field has been rapidly developing. A comprehensive overview of the field, however, is lacking. Therefore, in Part 1 of this dissertation, we performed a systematic review to provide a state-of-the-art overview of the field. PREDICTORS OF EFFECTIVENESS: NONSPECIFIC FACTORS IN APPLIED GAMES As previously discussed, traditional evidence-based programs in the field of developmental psychopathology are mostly based on CBT, which includes evidence-based therapeutic techniques (Kendall, 2011). These ‘specific’ therapeutic techniques are derived from theories that explain the underlying processes involved in the development and persistence of mental health
19 1 GENERAL INTRODUCTION disorders (e.g., relaxation and exposure training for anxiety; Kendall, 2011). In the development of applied games for mental health, the primary focus is on translating these specific therapeutic techniques into game elements and mechanisms (Eichenberg & Schott, 2017; Fleming et al., 2017). In research on applied games, the main objective is to examine the game’s overall effectiveness, and to a lesser extent, how specific techniques incorporated contribute to changes in mental health (for exceptions, see De Vries et al., 2015; Dovis et al., 2015; Van Houdt et al., 2019). The assumption underlying this research is, however, that the observed improvements in mental health can be attributed to the specific CBT techniques incorporated in the game. A substantial body of evidence, however, consistently indicates that nonspecific factors – factors not specific to any particular psychotherapeutic approach – significantly contribute to positive intervention outcomes (Colloca, 2018a, 2018b; Grencavage & Norcross, 1990; Ilardi & Craighead, 1994; Thiruchselvam et al., 2019; Wampold, 2015) as well as process-related variables such as engagement, adherence, alliance quality, effort and invested time in the treatment (Boettcher et al., 2013; Boot et al., 2013; Constantino et al., 2011; Greenberg et al., 2006; Meyer et al., 2002; Westra et al., 2007). In fact, specific factors explain relatively little and account for only a small percentage of the variance in outcome measures, whereas nonspecific factors play a more important role in treatment efficacy and improving mental health outcomes (Ahn & Wampold, 2001; Lambert, 2005, 2011). Examples of the most important and most researched nonspecific factors include the client-therapist relationship (Krupnick et al., 2006; Norcross, 2002), expectations for improvement and placebo effects (Asay & Lambert, 1999; Constantino et al., 2018; Crum & Phillips, 2015; Greenberg et al., 2006; Kazdin, 1979; Thiruchselvam et al., 2019), and client-related variables such as hope (Ilardi & Craighead, 1994), mindset or implicit theories of beliefs about the malleability of personal attributes (Crum & Phillips, 2015; Tamir et al., 2007), and motivation or readiness to change (Dozois et al., 2004; Norcross et al., 2011; Prochaska & Norcross, 2001; Taylor et al., 2012). Given the central role of nonspecific factors in nearly all psychological interventions (Lambert, 2005), it is likely that nonspecific factors, at least in part, drive mental health improvements in well-designed applied games as well. After all, applied games are often part of a treatment context, are delivered in a specific way, and individuals may have different expectations and beliefs about the effectiveness of applied games and different motivations to install or follow an applied game program. Findings from two randomised controlled trials (RCTs) performed by our own Games for Emotional and Mental Health (GEMH) lab further highlight the need to examine nonspecific
20 Chapter 1 factors in applied games. These two RCTs investigated the effectiveness of two applied games, MindLight and Dojo, specifically designed to reduce anxiety symptoms, and compared the effects to commercial control games (Scholten et al., 2016; Schoneveld et al., 2016). Results from both trials showed equal improvements in anxiety symptoms for both the intervention and the control group. Although the applied games explicitly incorporated evidence-based techniques for anxiety and the two control games did not, equal reductions in anxiety symptoms were found. This suggests that nonspecific factors such as expectations and motivation to change may have contributed to these findings. To date, limited attention has been given to nonspecific factors in the literature on applied games and their potential effects remain largely unknown (Enck et al., 2017; Torous & Firth, 2016). In order to optimise the effectiveness of applied games to their full potential, it is crucial to examine and harness the benefits of nonspecific factors (Enck et al., 2013). It is hypothesised that nonspecific factors positively influence mental health outcomes directly, as well as through engagement. Previous research on conventional therapy has shown that nonspecific factors contribute to both positive intervention outcomes and process-related variables such as engagement, adherence, effort and invested time in the treatment (Boettcher et al., 2013; Boot et al., 2013; Colloca, 2018b; Constantino et al., 2018; Greenberg et al., 2006; Wampold, 2015; Westra et al., 2007). In turn, research has found that these variables related to engagement predict positive intervention outcomes (Becker et al., 2015; Lindsey et al., 2019; C. M. Yeager & Benight, 2022). The variables studied in the current dissertation are explained in more detail below. NONSPECIFIC FACTORS, ENGAGEMENT AND POSITIVE OUTCOMES As briefly mentioned earlier, one of the most significant nonspecific factors to consider in applied games is individuals’ expectations for improvement (Asay & Lambert, 1999; Lambert, 2005). Applied games are usually introduced with a clear aim to promote (mental) health, which naturally induces expectations for improvement. Previous research has demonstrated that expectations drive a large majority of (conventional) intervention effects (Greenberg et al., 2006), but also particularly so in experimental game design studies (Boot et al., 2013). Research on conventional programmes has also shown that individuals with higher expectations for improvement invested more time and effort in, for example, an unguided internet-based self-help programme for social
21 1 GENERAL INTRODUCTION anxiety (Boettcher et al., 2013). Thus, expectations for improvement influence intervention outcomes as well as engagement with conventional programs. It is unknown whether similar processes may be at hand in applied games, but the findings of the two earlier mentioned RCTs performed by our lab suggest so. More specifically, expectations for improvement may explain the equal improvements in anxiety that were found for both the intervention and control games (Scholten et al., 2016; Schoneveld et al., 2016). In our RCTs, we made sure that youth’s expectations for both the applied and (casual) control game were equal, prior to the random assignment. However, in doing so, youth in both groups had similar expectations of effective anxiety reduction, which might have resulted in the equal improvements we observed in both the intervention and control group. Therefore, it is important to examine the effect of expectations on engagement and positive intervention outcomes in applied games, such that we understand to what extent specific therapeutic techniques and nonspecific factors each explain positive outcomes and whether enhancing expectations prior to gameplay would be fruitful. Motivation to change may be a second variable that could explain the equal improvements in anxiety we found in our RCTs , and a nonspecific factor that generally may be relevant to take into account in research on applied games. Motivation to change refers to an individual’s readiness and willingness to change the symptoms or challenges they are facing (Prochaska & DiClemente, 1982). It has been recognised as one of the key predictors of (conventional) treatment outcomes, as well as a significant factor influencing variables associated with engagement (e.g., treatment engagement, adherence, and dropout; Brogan et al., 1999; Derisley & Reynolds, 2000; Lewis et al., 2012; Lewis et al., 2009; Norcross et al., 2011; Taylor et al., 2012). Individuals lacking motivation or not yet prepared to address their symptoms are less likely to show improvement post-treatment, and tend to invest less time or engage less actively in the treatment process. Although games are naturally appealing and considered intrinsically motivating and engaging (e.g., Granic et al., 2014), based on previous research we expect that individuals’ motivation to change may still influence actual engagement and their perseverance in the face of failure, and subsequent mental health improvements. Therefore, it is important to examine motivation to change in applied games. Symptom severity may be another important nonspecific factor to consider. Although previous research on conventional programs has not yielded conclusive results, there is research suggesting that higher symptom severity or symptoms at baseline are related to more professional help-seeking and greater symptom decreases after intervention (Merikangas et al., 2011; M.
22 Chapter 1 I. Oliver et al., 2005; Sawyer et al., 2012; Van Starrenburg et al., 2017). Yet, little research has examined the influence of symptom severity on the engagement and positive outcomes of applied games. While certain nonspecific factors may have independent effects in applied games, they may also interact with one another. For instance, individuals’ motivation to change may be related to the severity of symptoms experienced (Dozois et al., 2004). Additionally, the influence of expectations on engagement and positive intervention outcomes may be moderated by motivation to change and symptoms severity (Buday, 2015; M. B. Oliver & Krakowiak, 2009). Furthermore, it has been hypothesised that nonspecific factors interact with specific factors in the prediction of positive intervention outcomes (Boot et al., 2013; Greenberg et al., 2006; Kazdin, 2005; Messer & Wampold, 2002). For example, an individual more motivated to change or experiencing more severe symptoms may engage more with the (specific) therapeutic techniques in the game, and subsequently show larger improvements in mental health (Buday, 2015; Dean et al., 2016; M. B. Oliver & Krakowiak, 2009). Conversely, it is possible that individuals experiencing more severe symptoms exhibit lower levels of engagement with the therapeutic techniques embedded in the game, as it confronts them with their problems (Poppelaars, Lichtwarck-Aschoff et al., 2018). It is important to devote attention to these interaction effects as insights into the complexity of factors at work may give critical information about individual differences in intervention outcomes. By utilising knowledge about both specific and nonspecific factors, applied games for mental health can be optimised. PROMOTING AND DELIVERING GAMES FOR MENTAL HEALTH: INTERVENTION REACH AND UPTAKE As outlined earlier, video games offer a significant advantage when it comes to their implementation potential (Granic et al., 2014; Kazdin, 2015; Lau et al., 2017). On the one hand, effective applied and casual games can be used in a clinical setting to complement traditional intervention approaches, reinforcing therapeutic techniques, providing additional support or enhancing engagement between sessions (e.g., Beaumont et al., 2021; Ducharme et al., 2021; Schuurmans et al., 2018). They can also be utilised as replacements for (school-based) prevention programs, offering interactive and engaging interventions to address mental health concerns (e.g., Schoneveld et al., 2018). On the other hand, video games also have the capacity to be distributed and
23 1 GENERAL INTRODUCTION utilised outside the clinical context; they may be offered as (applied) standalone and/or a freely accessible (commercial) game. For individuals with mild symptoms, who may not require intensive therapy but could benefit from self-guided interventions, video games may provide support, education, and self-help resources. That way, mental health games can help individuals manage and improve their mental health in a convenient manner. Outside the traditional healthcare setting, however, individuals’ motivation to pick and play a video game to improve their mental health becomes an important condition for the exploit of the implementation potential of video games. Therefore, it is essential to consider and address the factors that enhance motivation to engage with games for mental health. One factor potentially influencing the reach and impact of video games is the way in which games for mental health are presented and promoted. How these games are marketed, framed, and communicated can influence their acceptance and engagement. One approach may be promoting mental health games as proven tool for enhancing mental health. This approach involves highlighting the evidence-based benefits of the games. By emphasizing the positive outcomes and scientific validity of mental health games, individuals may be more inclined to engage with them, recognising the potential value they can offer to mental health (M. B. Oliver & Krakowiak, 2009). While some individuals may be motivated by the potential health benefits, others may experience limited autonomy and feelings of resistance in response to an explicitly promoted mental health aim (Brehm, 1966; Dillard & Shen, 2005; A. S. Richards & Banas, 2015). Another approach would therefore be to position mental health games as regular entertainment games rather than explicitly highlighting their therapeutic nature. This ‘stealth’ approach involves presenting the mental health game as enjoyable and engaging entertainment experiences and may motivate individuals to engage with the game without feeling stigmatised or triggering resistance. Understanding messaging effects can inform the design and promotion of mental health games to maximise engagement and motivation. Moreover, personal motivational traits of youth are likely to influence their receptiveness and the degree to which explicit messages about mental health resonate with them. For example, the desire to improve one’s mental health may motivate individuals to pick and play games explicitly promoted for their mental health benefits. As outlined earlier, motivation to change has been found to be an important predictor of help-seeking, adherence, drop-out and treatment engagement (Brogan et al., 1999; Derisley & Reynolds, 2000; Norcross et al., 2011; Taylor et al., 2012). Likewise, symptom severity may be
24 Chapter 1 another factor influencing game choice and gameplay. Individuals experiencing more (severe) symptoms may feel motivated by the personal relevance of a game promoted for its mental health benefits, such that they want to play the game compared to individuals with less (severe) symptoms. On the other hand, however, these individuals may avoid such games as they may think that it confronts them with their problems. In research on conventional programs, symptom severity has been associated with more help-seeking (Merikangas et al., 2011; M. I. Oliver et al., 2005; Sawyer et al., 2012; Van Starrenburg et al., 2017) as well as lower help-seeking tendencies (Chin et al., 2015; Sawyer et al., 2012), suggesting that evidence is inconclusive at this moment, and it remains unknown how symptom severity may impact game selection and engagement with mental health games. Additionally, individuals’ implicit theory or mindsets may play a role in the likelihood of selecting and playing a mental health game. Individuals can have mindsets about every personal attribute such as intelligence (Dweck, 2017c), personality (D. S. Yeager & Dweck, 2012), emotions (Tamir et al., 2007), and the nature of stress (Crum et al., 2013). In the literature, a distinction is made between two types of mindsets: a growth mindset (incremental theory) and a fixed mindset (entity theory). When adopting a growth mindset regarding a particular attribute, one believes that the attribute can be altered through dedicated effort, experience, and assistance from others (Dweck, 2013; Tamir et al., 2007). Individuals with a fixed mindset believe that a specific attribute is not (or less) malleable and cannot be controlled. The concept of mindsets, as proposed by Dweck (2013), revolves around the notion that mindsets shape individuals’ goals, action tendencies, beliefs about effort, and responses to setbacks (Dweck, 2017a, 2017c). As a result, mindsets emerge as significant motivational factors influencing behaviour (Burnette et al., 2013; Dweck, 2017a), potentially impacting help-seeking tendencies and treatment engagement (Burnette et al., 2019; Schroder et al., 2015). Furthermore, having a growth mindset has been linked to better mental health and well-being (Miu & Yeager, 2015; Romero et al., 2014; Schleider et al., 2015; Schroder et al., 2015; Zeng et al., 2016). Thus, given their relevance for behaviour and motivation, mindsets may influence the likelihood of selecting and playing a mental health game. Mindsets themselves, however, are malleable as well. Past studies have demonstrated that mindsets can change as a result of relatively brief interventions, such as watching short video clips, reading an article or performing a short writing exercise (e.g., Crum et al., 2013; Dweck, 2008; Jamieson et al., 2018; Miu & Yeager, 2015; D. S. Yeager et al., 2014). Of relevance in the current context is that games are known to be able to effectively
25 1 GENERAL INTRODUCTION motivate players to persist in playing, even in the face of failures. Games serve as a platform for cultivating perseverance, which involves exerting continuous effort towards achieving a goal despite encountering difficulties or setbacks (Malone & Lepper, 2021). Individuals persevering in the face of failure are more likely to have a growth mindset (Dweck, 2017b), suggesting that games may boost one’s growth mindset (Weerdmeester et al., 2020). Therefore, in this dissertation, we also examined the influence of gameplay on changes in one’s mindset. THE CURRENT THESIS Overall, this dissertation aims to examine the potential of applied and casual games as a possible alternative delivery approach of traditional intervention programs for mental health, and to investigate the effects of nonspecific (motivational) factors on game selection, game experiences, engagement, and mental health outcomes. This dissertation is divided into three parts. Part 1 of my dissertation aims to provide a state-of-the-art overview of the field. In Chapter 2, we performed a systematic review of randomised controlled studies that have assessed digital games for improving mental health in children, adolescents and young adults and examined the effectiveness of both applied and casual games. Thus far, previous reviews and meta-analyses focused on either applied or casual games exclusively, on very specific mental health domains, on clinical populations only, and/or they did not include multimodal interventions (i.e., using a digital game in addition to other therapy components). In order to gain a more comprehensive understanding of the impact of games on mental health outcomes and to examine when games are most effective, we evaluated applied and casual games simultaneously, included and distinguished between both clinical and healthy populations, and compared the effectiveness of these games across a variety of mental health domains, including internalising, externalising, neurodevelopmental, psychotic and personality-related outcomes. Moreover, we examined methodological characteristics to explore current research trends and whether nonspecific factors have been taken into account. To maximise the effectiveness of applied games to their fullest potential, it is essential to explore and leverage the advantages offered by nonspecific factors. Literature on applied games has largely overlooked nonspecific factors, and their impacts on engagement and mental health outcomes in this context are unknown. To address this significant gap in the literature, Part 2 of my
26 Chapter 1 dissertation aims to contribute to our understanding of the effect of specific and nonspecific factors on engagement and mental health outcomes in an applied game. We start off with Chapter 3, in which we examined to what extent engagement with the therapeutic techniques in an applied game for anxiety prevention (i.e., MindLight) predicts mental health improvements. More specifically, we examined whether baseline anxiety symptoms were related to in-game play behaviours as well as whether changes in in-game play behaviours predicted changes in anxiety symptoms three months after playing the game. The study in Chapter 4 aimed to further unravel whether MindLight also has beneficial effects on other mental health outcomes associated with anxiety symptoms (i.e., internalising problems, externalising problems and selfefficacy), given their overlap in symptoms and transdiagnostic mechanisms. Additionally, we examined who benefitted most from the game by assessing several possible nonspecific variables as predictors of changes in anxiety symptoms, namely baseline anxiety symptoms, maternal mental health problems, and self-efficacy. In Chapter 5, we experimentally manipulated expectations for improvement and examined the effect on engagement while playing MindLight, as well as the moderating role of symptom severity and motivation to change. Additionally, changes in state anxiety and arousal were examined. Despite the notable implementation benefits of games, there is a lack of research examining the uptake of and engagement with mental health games among individuals with mental health symptoms (Fleming et al., 2018; Fleming et al., 2016). Understanding how motivational factors influence selection and engagement with mental health games is crucial as it will enable us to customise and tailor the promotion of mental health games, maximising their uptake and sustained use. Therefore, in Part 3 of my dissertation, we aimed to examine how messaging and several (nonspecific) motivational factors influence choice for a mental health game, and subsequent game experiences and engagement. In Chapter 6, we examined how messaging affects game choice, perceived attractiveness and fun of the game, as well as subsequent engagement and experiences of gameplay. We further explored how severity and type of mental health symptoms influence game choice, game experiences and engagement. To better understand the influence of personal motivational traits, we examined how motivation to change and one’s mindsets influence game choice and engagement in Chapter 7. We also examined whether mindsets change after playing a game promoted as mental health game. To conclude my dissertation, Chapter 8 presents a summary and general discussion of the
27 1 GENERAL INTRODUCTION main findings, including limitations and implications for future research and implementation.
Chapter 2 Effectiveness of applied and casual games for young people’s mental health: A systematic review of randomised controlled trials Based on: Wols, A., Pingel, M., Lichtwarck-Aschoff, A., & Granic, I. (2024). Effectiveness of applied and casual games for young people’s mental health: A systematic review of randomised controlled studies. Clinical Psychology Review, 108, 102396. https://doi.org/10.1016/j.cpr.2024.102396 Based on: Wols, A., Pingel, M., Lichtwarck-Aschoff, A., & Granic, I. (2024). Effectiveness of applied and casual games for young people’s mental health: A systematic review of randomised controlled studies. Clinical Psychology Review, 108, 102396. https://doi.org/10.1016/j.cpr.2024.102396 Chapter 2 Effectiveness of applied and casual games for young people’s mental health: A systematic review of randomised controlled trials
30 Chapter 2 ABSTRACT Many youth experience mental health problems and digital games hold potential as mental health interventions. This systematic review provides an overview of randomised controlled studies assessing the effectiveness of digital applied and casual games for improving mental health in youth aged 6–24 years. A systematic search of PsycINFO, Web of Science and Pubmed yielded 145 eligible studies. Studies on (sub)clinical participant samples (n = 75) most often focused on attention-deficit/hyperactivity disorder (ADHD), autism and anxiety. Applied games were found most effective for improving social skills, verbal memory and anxiety, whereas casual games were found most effective for improving depression, anxiety and ADHD. Studies involving healthy youth (n = 70) were grouped into papers examining anxiety in medical settings, momentary effects on positive and negative affect, and papers employing a longitudinal design measuring mental health trait outcomes. Promising results were found for the use of games as distraction tools in medical settings, and for applied and casual games for improving momentary affect. Overall, our findings demonstrate the potential of digital games for improving mental health. Implications and recommendations for future research are discussed, such as developing evaluation guidelines, clearly defining applied games, harmonising outcome measures, including positive outcomes, and examining nonspecific factors that may influence symptom improvement as well.
www.ridderprint.nlRkJQdWJsaXNoZXIy MTk4NDMw