Gambling disorder (GD) is the first behavioral addiction to be approved in the DSM-5 [Diagnostic and Statistical Manual of Mental Disorders, 5th ed. (1)], in which it is included in addictive disorders rather than in impulse-control disorders. Indeed, GD presents more similarities with substance-use disorders than with impulse-control disorders (2, 3), as the persistent and current pattern of gambling associated with GD presents symptoms such as tolerance and withdrawal that are typical of addiction (1). In Europe, GD prevalence rates range between 0.12 and 5.8% (1), but it often goes undiagnosed and untreated, and it is associated with negative health measures (4, 5). GD is one of the new addictions that affect people in terms of substantial clinical distress in terms of their social life, work, training, and education (1), as well as from a psychosomatic point of view. Thus, focusing on the efficacy of diagnostic and therapeutic measures for this disorder with a scientific approach is especially relevant to deliver patient-tailored treatments (6). Further studies are needed to compare the efficacy and acceptability of individual and combined psychosocial and pharmacological interventions to deliver patient-tailored treatments.

To address this issue, the current study protocol—Alter Game- is aimed at (1) developing an innovative psychological treatment for GD by using a virtual reality (VR) approach, and (2) verifying its efficacy. The rationale for using this approach in GD treatment lies in the fact that cognitive behavioral therapy (CBT) is the most grounded and effective theoretical and applied approach to treating addiction (7). Moreover, a review of evidence-based best practices in GD treatment demonstrated that psychological interventions are the most successful protocols, especially concerning motivational and CBT therapies (6, 8), confirming the findings of the Cochrane review (9). In GD treatment, CBT considers cue exposure therapy (10), cognitive restructuring, problem solving techniques, social skills training, and relapse prevention (11) as fundamental steps. Among them, relapse prevention (RP) is one of the most important steps for GD. Based on the RP model proposed by Marlatt and Gordon (12) for substance abuse, the RP model applied in CBT is focused on a set of cognitive and behavioral strategies to prevent or limit relapse episodes. In general, the RP model refers to immediate determinants (e.g., high-risk situations, coping skills, outcome expectancies, the abstinence violation effect) and covert antecedents (e.g., craving, lifestyle imbalances, and urges) (13). RP therapy aims to find, along with the patients, the factors or situations that can precipitate or contribute to relapse episodes and teaches patients how to manage them, increasing their perceived self-efficacy. Self-efficacy is the degree to which an individual feels confident and capable of performing a certain behavior in a specific situational context (14). The presence of low perceived self-efficacy may contribute to maintaining pathological gambling behavior (15). CBT presents different advantages in pathological gambling treatment; there are lower risks of relapse during the follow-up period, and it has a limited duration and involves structured interventions with long-term positive effects (16, 17). On the other hand, CBT presents some limits. Indeed, difficulties were detected in the treatment of pathological gamblers with high levels of sensation seeking and problematic emotional regulation, in which a higher drop-out rate and higher frequency of relapse emerged (18, 19). Thus, the improvement of CBT efficacy for GD is necessary, especially concerning RP. For this reason, there is a need for innovative methodologies to support traditional intervention techniques.

Encouraging results in this regard have come from the blending of psychotherapy and VR, an immersive computer-generated three-dimensional virtual environment, which is viewed through a head-mounted display (HMD) and interacted with by using controllers (20, 21). VR has been the focus of international interest for the last 20 years or so. Virtual applications are used, in particular, in psychotherapeutic interventions for the treatment of post-traumatic stress disorder (22), social anxiety (23, 24), paranoia (25), specific phobias, such as acrophobia (26), aerophobia (27, 28) and arachnophobia (29, 30). Recent literature has also recently pointed to the usefulness of VR in the treatment of GD (31–34), since VR is a methodological approach that allows to recreate a realistic ecological representation of craving situations (35–37). It has been shown that virtual environments containing gambling cues elicit cravings of the same intensity as those induced by real gaming-related stimuli (e.g., real and virtual slot machines) (31). Craving is defined by the DSM-5 as a persistent, intense or irresistible desire for a specific substance (1) or, in the case of gambling, for an addictive behavior. Craving presents cognitive, emotional, and behavioral characteristics, and it is classified in three different types: reward, relief, and obsessive craving (38); therefore, RP in virtual cue-exposure therapy allows pathological gamblers to manage their urge to gamble and to avoid relapse, given that they become aware of which internal and external triggers are related to craving. The idea of targeting VR for the implementation of RP interventions is supported by other advantages of its application in psychotherapy. Indeed, a patient can experience himself/herself within a protected environment, taking advantage of the possibility of interrupting the virtual experience when he/she feels discomfort or difficulties. The virtual exposure is constantly supervised by the therapist through a monitor: what gamblers see through the HMD is, at the same time, watched by the psychotherapist as well. Furthermore, the ecological validity of the environments guarantees greater generalizability of the skills acquired in VR to daily life environments and situations (39).

However, studies on the use of VR with gamblers present some limits. Some studies were conducted on small samples that, sometimes, did not include pathological gamblers (10, 31, 34). Other important weaknesses that could be improved are the following: (a) distinguishing the physical activation due to gambling craving from the one due to possible cybersickness, i.e., discomfort, apathy, nausea, drowsiness, disorientation, eyestrain, and fatigue that may be elicited by an experience in immersive VR (40, 41); (b) using virtual gambling stimuli in association with other virtual addiction triggers (such as alcohol), (c) using VR with other psychotherapeutic techniques, like mindfulness (31); (d) exploring the role of sense of presence in virtual experiences. Sense of presence is a mental manifestation (42), a cognitive state which is consistent with a subjective feeling of “being there,” in a given virtual environment, which is different from the physical one and whose fruition is mediated by a device (43).

Keeping this background into account, the present Alter Game study was designed. Alter Game aims at developing an innovative psychological treatment for GD by using VR and verifying the efficacy of an innovative psychological treatment for GD based on the integration of traditional CBT therapy and VR approach. To that goal, we aimed at creating a virtual serious game to prevent relapse in gambling behavior by increasing self-efficacy in managing craving and emotional states and by reducing perceived gambling craving. Indeed, virtual environments are made ad hoc for the exposure of patients to specific gambling cues, directed at the detection and management of craving. We chose self-efficacy and craving as dependent variables because they are key aspects in maintaining abstention from gambling. Gamblers that perceive good self-efficacy in managing cravings present a lower probability to relapse, on the basis of the Marlatt RP model (12); thus, craving itself is an important relapse risk factor to be managed and recognized. On the other hand, perceived self-efficacy is a protective factor against relapse (12). Moreover, in response to the above-described limitations in previous studies applying VR approach to gambling treatment, Alter Game aims at evaluating the sense of presence through multiple means: the Independent Television Commission's Sense of Presence Inventory [ITC-SOPI (44)]; the use of diaphragmatic breathing to manage craving; the presence of cigarette packs, lit cigarettes, snacks, alcoholic beverages (wine, beer), and wine glasses in the specific virtual environments; the use of biofeedback to measure the psychophysiological activation. Cue reactivity, indeed, is measured across several domains of human functioning. The most commonly collected measures are self-reports that assess craving or the desire for a particular substance and physiological responses, which usually include those controlled by the autonomic nervous system, such as heart rate, skin conductance, and skin temperature (45). Psychophysiological arousal is a potential proxy indicator of craving as a reaction to gambling-related stimuli. Conditioned stimuli associated with gambling could simultaneously elicit both subjective craving and psychophysiological arousal (46).

Our hypothesis is that an integrated intervention with VR will be more effective than simple CBT for improving self-efficacy and lowering cravings. Furthermore, in line with CBT, we expect gambling-related cognitive distortions to decrease. This goal is very important because gambling-related cognitive distortions mediate the relationship between depressive symptoms and gambling severity (47) and contribute to the maintenance of maladaptive emotional regulation styles to reinforce and support biased beliefs about gambling outcomes and controllability (48).

Alter Game was conceptualized by the Addiction Medicine Unit's interdisciplinary team at the G.B. Rossi Hospital in Verona (Italy). The hospital has signed a collaboration agreement with the Department of Neuroscience, Psychology, Drug, and Child's Health, Section of Psychology in University of Florence (NEUROFARBA—Italy) for methodological and psychometric consultancy related to the study protocol.

In general, all the patients' demographic and clinical characteristics will be summarized using descriptive statistical methods. In particular, mean, median, and standard deviations will be used for symmetrically distributed continuous variables, median and interquartile distance for asymmetrically distributed continuous variables, and frequency distributions for categorical variables. Multi-method analysis of craving measurement in order to draw up a physiological and psychometric profile of craving (using biofeedback) will be performed. This analysis will be conducted through linear correlation measures. For the analysis of the efficacy of the training course for the treatment providers, Student's t-tests for dependent samples or Wilcoxon's non-parametric test for paired samples will be used. Contingency tables and McNemar tests may also be used. For the analysis of the efficacy of the CBT + VR group compared to the CBT group, a mixed ANOVA will be performed for repeated measures with Time as within factor (pre-test/post-test/follow-up) and Group as between factor (the CBT + VR group/the CBT group).

Moreover, we will pay attention to the issue of an attrition rate. We are expecting to find a retention rate of at least 80% (78, 79) to avoid that participant loss will bias research outcomes. After that, we will check for any significant difference between the two treatment groups (CBT and CBT + VR) to verify if there is a differential attrition rate. In particular, we will compare the average number of meetings attended by the participants between the two treatment groups through t-tests, and we will compare the presence of drop-outs across the two groups through Chi-square tests. Moreover, we will determine if there are any significant differences (at the baseline) between participants who will complete the post-test session and those who will not in order to verify if specific subgroups of participants have a higher drop-out risk. This analysis will be conducted separately for each group. If any significant differences between participants of the two treatment groups will be found, we will verify that systematic attrition does not occur and we will conduct the analyses to verify the effects of the two interventions (CBT and CBT + VR) with the participants who completed the baseline and post-test assessments. Those lost to post tests will be excluded. If significant differences between participants of the two treatment groups will be found for the average number of meetings attended and the treatment drop-out, and participants who will complete the post-test session will be significantly different from those who will not, subsequent analyses will be conducted by using the last observation carried forward (LOCF) to assess the effects of attrition. Furthermore, we will examine within-group changes in outcomes as a function of session attendance to assess the effects of intervention non-compliance on outcomes. These subsequent analyses will be conducted, not excluding participants who will be lost to post-test and by imputing values for the follow-up variables for the participants who will not complete the post-test assessment.

A recent review has confirmed the efficacy of CBT to treat various addiction problems (7) and, especially, GD (9). Several published clinical studies have used VR to treat different psychopathological problems (anxiety, PTSD, depression, etc.), with a lot of them underlining that CBT and VR combined could improve the efficacy of social anxiety disorder and paranoia treatments (22, 24). Indeed, VR is a methodological approach that allows to recreate a realistic ecological representation of craving-eliciting situations (34–36). However, as of yet, no clinical study of CBT and VR in GD treatment has been conducted. Keeping this background into account, the necessity to develop a new study protocol—Alter Game– emerged. In detail, Alter Game aims at developing an innovative psychological treatment for GD by using VR and verifying the efficacy of an innovative psychological treatment for GD based on the integration of traditional CBT, coupled with a novel VR approach.

Based on previous studies, suggesting that CBT and VR combined are more efficient than CBT or VR only (23, 25), we aim at confirming this evidence in gambling addiction by taking specific target variables into account, such as self-efficacy, craving, and cognitive distortions. We can hypothesize that CBT and VR combined will be more effective than CBT only. In particular, at the conclusion of the psychotherapy sessions, we expect to find a higher increase of self-efficacy and a greater reduction of cravings and cognitive distortions—in terms of effect size—in the case of the integration of CBT and VR. To better evaluate these expected results, we aim at using biofeedback in order to observe and measure the psychophysiological activation associated with cue exposure, which would allow us to overcome the limits of self-report craving questionnaires or craving Likert scales. Moreover, we already implemented a training course for the treatment providers to guarantee a correct administration of the psychotherapy protocol with the patients before the study will start.

By increasing self-efficacy and reducing gambling-related craving and cognitive distortions, we aim at reducing the risk for relapse, which represents one of the most important psychotherapeutic steps for GD. Using VR cue-exposure therapy, pathological gamblers may increase their skills to prevent a relapse, or they could improve the management of their gambling craving and identify the triggers associated with it by pinpointing its cognitive, emotional, and bodily characteristics. These skills are important to prevent relapses in gambling behavior and to increase the self-efficacy of gamblers. By decreasing gambling-related distortions, which contribute to maintaining a maladaptive emotional regulation and reinforcing the irrational beliefs about gambling outcomes and controllability (48), we aim at promoting higher levels of management regarding thoughts about gambling in general. Another important aspect to be considered is that VR will provide contexts that are typical of everyday life, contributing to the generalizability of the craving management. Several studies using VR and cue reactivity to evaluate craving and other relapse risk factors in addictions (alcohol, tobacco, drugs, etc.) underlined the influence of life context in maintaining dependence or relapsing (82–84).

Despite these strengths, some limitations must be underlined. As a limitation of the study protocol, the use of self-report measures may be affected by the lack of insight/awareness, inability to identify or articulate beliefs/thoughts, and intentional dissimulation in pathological gamblers. However, the use of self-report measures is inevitable, so we chose tests that present good psychometrics properties, and we will consider the MCMI-III validity scales to control the social desirability index and the assessment protocol validity. In order to better control these biases, future studies could consider the introduction of behavioral tasks that in the Alter Game study protocol are not included except in the therapeutic intervention section. Also, the absence of a control group (non-treatment), which could serve to better analyze the efficacy of the integrated intervention, could be a further point to be addressed in the future. A public clinical service cannot omit to offer clinical treatment to a patient with GD for ethical reasons. Moreover, there may be potential threats to the internal validity of the research design due to its longitudinality, with a possible consequent loss of patients. In line with this, we chose a 1-month follow-up to keep the patients as engaged with the study as possible, since the dropout rate in pathological gamblers is high (18, 19). We are aware, however, that this time may not permit the detection of spontaneous recovery. Furthermore, studies could implement a strategy to better control susceptibility to conditioning. Another important consideration deals with the changes due to the COVID-19 pandemic. Indeed, GD is changing: a lot of patients are addicted to online gambling, so some patients that request our treatment may not respond to our inclusion criteria, in particular regarding the type of gambling activities considered in the Alter Game study before the beginning of the pandemic. In the future, the development and creation of new virtual environments on online gambling may be needed. The COVID-19 pandemic also brings additional complications: wearing the surgical mask can cause fogging problems to the HMD that compromise the correct use and the ideal fruition of the immersive virtual experience. In addition, the presence of restrictions for the containment of COVID-19 (lockdown, isolation, quarantines) may hamper the possibility of maintaining the 1 session per week that would be required by the study protocol. Another limitation regards the generalizability of the integrated protocol due to the possible presence of cybersickness and specific medical problems. Indeed, patients that present cybersickness during the virtual treatment or clinical problems like epilepsy, visual or hearing impairments, vestibular or cardiovascular or neurological diseases, and schizophrenia or psychotic traits will not be granted access to CBT + VR psychotherapy. Finally, future developments could improve our virtual environments through the insertion of avatars that interact with gamblers, thanks to artificial intelligence. This will be more expensive, but it may be useful to work on the impact of social interaction in the maintenance of GD and to manage situations in which gamblers are invited to eat snacks and drink alcoholic beverages. Indeed, it is known that social isolation and a sense of loneliness are related to problematic gambling (85, 86). Therefore, inserting social interaction in virtual gambling-related environments could be an important development to treat these daily experiences that may be relapse risk factors. Subjects that are socially excluded tend to self-medicate their unpleasant emotional states [e.g., anger, sadness, and anxiety; (87)] through emotion regulation processes (88) that can incorporate gambling too (89). Moreover, some studies highlighted that there is an association between gambling and alcohol use: those who drink alcohol more frequently are more inclined to gamble and to experience negative consequences due to their gambling behavior (90–93), as well as increased speed and duration of the game session (94, 95), increased risky wagering (94, 96, 97), and more rapid depletion of available funds (94–96).

To conclude, if our expected results will be confirmed, Alter Game could promote new possible tools for RP therapy that may be applicable within specialist-care services for GD and within therapeutic communities. These may also be suitable for poly-addicted gamblers, given the known high comorbidity between gambling and substance use disorders (98). Indeed, if the integrated virtual therapeutic protocol proposed by Alter Game will be effective, it could be extended to the treatment of other addiction disorders, reinforcing the traditional RP program.

Approval for the research was obtained from the Ethics Committee for Clinical Trials (CESC) of the Provinces of Verona and Rovigo based at the Integrated University Hospital of Verona, Italy (approval code: 3004CESC with Protocol No. 69346 of 21/12/2020). The latest revision of the Declaration of Helsinki as well as the Oviedo Declaration is the basis for the ethical conduct of the study. The study protocol is designed and will be conducted to ensure adherence to the principles and procedures of Good Clinical Practice and to comply with Italian law, as described in the following documents and accepted, by signature, by the study investigators: ICH Harmonized Tripartite Guidelines for Good Clinical Practice 1996. Directive 91/507/EEC, The Rules Governing Medicinal Products in the European Community. D. L.vo n. 211 of June 24, 2003. D. L.vo n. 200, November 6 2007. Ministerial Decree of December 21, 2007. AIFA Determination, March 20, 2008. All essential clinical records will be retained to demonstrate the validity of the study and the integrity of the data collected. The promoter of this study, in accordance with the responsibilities required by the rules of good clinical practice (Legislative Decree 211/2003) and in accordance with the laws and regulations regarding data protection (including the European Regulation on the protection of personal data 2016/679), will process the personal data that will be collected exclusively for the implementation of the study and for the purpose of device surveillance.

FL and RG: conceptualization. RG: data curation and investigation. MD: statistical analysis. FL, RG, and MD: methodology. FL and MD: supervision. RG and LZ: writing—original draft. MD, FF, CP, and FL: writing—review and editing. All authors have read and agreed to the published version of the manuscript.

The study described in this manuscript was funded by the Veneto Region (Italy) connected to the funding assigned by DGR n. 2154/2015, with the deliberation of the General Manager 1216 of 30/12/2016 to the Unit of Addiction Medicine in the Department of Internal Medicine in G.B. Rossi Hospital, Verona, Italy. The regional funding concerns the prevention and contrast of pathological gambling. The funding agency had no role in the design of the study or in writing this manuscript and will not influence the collection, analysis, and interpretation of data, apart from their financial contribution. All materials expected in the Alter Game study and all psychologists and psychotherapists involved were paid through Veneto Region's funding mentioned above.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.