The Take Control recovery support game uses several familiar and well-researched therapies to improve player treatment outcomes for addiction. Cue exposure therapy (CET) is a commonly used method in substance abuse treatment [1-6]. Traditionally, CET is performed with pictures of a substance, the actual substance itself, or even its scent [2]. The patient is repeatedly exposed to cues and stimuli and encouraged to ignore the craving response or use a coping response [7-10]. In keeping with the theory that the treatment effect is due to practicing a healthier response to a cue, the player in our game is repeatedly exposed to an image of a substance, and rather than responding with use, the player is trained to substitute a more dynamic, adaptive response. They must react appropriately (destroy the substance) in order to advance in the game.

Another way to explain the effect of this game is counter-conditioning. The unwanted behavior of responding to the cue to use a substance is being replaced with a positive action, and the new behavior is rewarded [11]. The game supports the rehearsal of the positive action of actively refusing a substance when it is presented and offers a reward to reinforce the more positive response in the form of success in the game.

The effectiveness of this game’s approach to substance abuse treatment also might be partially explained by the extinction response. Extinction therapy aims to reduce a patient’s conditioned response to a substance by repeated exposure without reinforcers in order to dull the craving response over time [12,13]. In our game, the patient is repeatedly exposed visually to the substance without receiving the reinforcing effect of the substance, which may produce some extinction effects. Creating new memories will overlap former memories, thus extinguishing old habits and responses [13]. Our game will allow players to use movement to form new, more adaptive associations with the substances.

Virtual reality therapy (VRT) uses virtual environments to expose patients to stimuli in a safe and controlled manner, such as with phobias or posttraumatic stress disorder [14]. In VRT for addiction treatment, the stimulus is the substance of the patient’s addiction [5,7,9,10,15,16]. A VR video game has distinct advantages over other exposure methods (eg, pictures produced by a counselor). Being in a VR environment allows the player to feel more immersed in the game, resulting in greater involvement and translation into real-life actions [5]. The game also addresses the need for a safe environment to practice refusal skills, as seen in coping skills training [5,17].

Exercise, which has been shown to aid in recovery from substance use disorders [18] and reduce comorbid factors that hinder overall health and wellbeing [19,20], is another factor that may mediate the game’s effect. Free movement is possible with our game because Kinect (the system) is not hindered by a controller, cords, or bulky head gear.

Cognitive behavioral therapy (CBT) has extensive research backing its effectiveness in addiction treatment [21,22]. One important interpersonal component of CBT is refusal skill practice. Patients learn how to respond rapidly, maintain eye contact, and give a clear “no” when offered drugs [23]. In our proposed game, players will practice refusal skills (such as verbally saying “no” and physically turning one’s back) when offered a substance by a character.

In the specific field of technology-based interventions for substance abuse, we did not identify any similar games to Take Control. However, there was a related study conducted by Girard et al [24] showing that 4 sessions performing behaviors incompatible with smoking cigarettes (crushing virtual cigarettes) within a virtual environment were more efficacious for smoking cessation than a similar game in which patients found and crushed virtual balls. The mechanism of this treatment in the study was not well understood, but we surmised that such virtual practice in a game environment may be uniquely helpful because it can deliver a large dose of alternative practice in a manner that people not only tolerate but enjoy. The fact that a short duration gaming experience in Girard et al [24] could improve outcomes in comparison with a placebo control suggests that games that involve the body in alternative practice may hold promise for treating addiction.

The Take Control recovery support game was developed for use with the Kinect motion sensor camera and device available with Xbox One and Windows operating systems.

Users hit or kick away cue images as they fly toward the user, as seen in Figure 1. If a user successfully hits the image, it explodes and the user gains in-game points. If an image hits the user or flies off the screen without being exploded, there is no negative consequence to the user’s score.

Users choose a background image, like the one in Figure 2, from a menu using voice or mouse controls and then choose 1 cue item to reject per round. Users were encouraged to focus on 1 substance but could change items between rounds. Users could replay the game using different backgrounds or cues as often as desired.

The game includes photo realistic backgrounds, seen in Figure 3, but drawn substance images, seen in Figure 4. During formative studies with target audience members, it was determined that photo realistic images of cue items (cigarettes, beer bottles, etc) were not preferred since users felt that such images were too specific, and thus made the experience less relevant to them individually. There were also reports from formative testers that they believed realistic looking cues might trigger cravings, while illustrations would be less likely to do so.

This study considers how a lower cost, easily accessible video game could be used to support recovery treatment or individual self-efficacy, attitude, and behavior. The goal is to support users in practicing refusal skills and increase self-efficacy by denying trigger or cue items in the nonthreatening environment of the video game. Using realistic backgrounds, users can hit or kick trigger items that fly toward them. Hit items explode, and users gain points.

Primary outcome measures were an increase in reported self-efficacy, attitude, and behavior, a decrease or lack of increase in craving after having seen the trigger items, and satisfaction with the game experience. Self-reported data on continued recovery status were also assessed.

The primary goal of the study was to increase user self-efficacy, attitude, and behavior by allowing the user to practice refusing trigger items in the game context.

The study was reviewed and approved by the Clinical Tools Inc Institutional Review Board. Participants were healthy adult volunteers who self-identified as having recently quit using cigarettes, tobacco, or alcohol. We did not collect data on coaddictions. Participants were recruited through advertisements in a local weekly news circular, on the Internet (Raleigh Craigslist), flyers placed in local public places (community centers, outside grocery stores), and via word of mouth.

Interested volunteers completed a short, open online survey that reviewed eligibility requirements (aged over 18 years, recent quitting, lack of mobility issues that would prohibit game use, fluent in English). The survey contained an informed consent section with study purpose, methods and procedures, confidentiality, benefits and inconveniences, precautions and risks, and survey submissions were limited via IP addresses. Participants also had to be able to travel to a game setup location. For most participants, this was a small office near a local church building, on the local free bus line, with free parking available. Participants received a gift card to a national store with multiple locations in the area at each of the 4 possible sessions. Sessions typically lasted more than 15 minutes but less than 30 minutes. Participants were asked to play 8 rounds (60 seconds per round) of the game.

Use of the video game was private; research staff were available to assist with any technical difficulties or usability question outside of the testing room, but staff did observe use of the game through a window (where players could not see them) to allow users to behave naturally and not feel judged for their ability to play the game. A computer log documented use time and score, and this was associated with the participant number. The study version incorporated the Kinect for Windows software and ran on a personal computer with a large monitor to facilitate viewing of the game.

The study was a quasi-experimental, stratified, wait-list control trial using a convenience sample due to time limitations.

Participants in each group had the opportunity to play the game at the Clinical Tools office 3 times, with 7 to 12 days in between uses. Case group participants were asked to complete a follow-up set of measures 1 to 2 weeks after the final game use. Control group users completed a baseline set of instruments and then after at least a 2-week wait period repeated the baseline measures and then used the game 3 times.

Thus, each group had a total of 4 assessment interactions and 3 game play interactions (see Textbox 1). Assessments were mostly Likert-style questions except for the 7-Day Timeline (fill-in-the-blank), Side Effects (multiple choice), and Stages of Change (multiple choice). A random number was assigned to participants and the information kept in a locked location. Researchers used this number to log participants into the surveys associated with specific session numbers to keep the participant’s multiple sessions linked.

Data collected for this study were sent to University of North Carolina at Chapel Hill, where a doctoral student in statistics analyzed the data.

A total of 76 participants were enrolled in the summative study. A total of 7 case participants were withdrawn from the analysis: 2 case participants were withdrawn due to inconsistent data and 5 were withdrawn due to not completing the study. A total of 8 control participants were withdrawn from the study, all due to not completing the study. A total of 61 participants were included in the analysis. There were 28 females (1 Asian, 10 African American, 13 white, 1 other, 2 multiracial, 1 prefer not to answer) and 32 males (1 Asian, 14 African American, 15 white, 1 other, 1 prefer not to answer); 1 participant was unknown (chose prefer not to answer for both gender and race categories).

There are several important results of this brief study. First, that participation in the study and use of the game seem to support abstinence from substance use based on the 7-Day Timeline reports when game users are compared to participants in the control group. The 7-Day Timeline instrument reported decreased or maintenance of no substance use at a poststudy follow up for most of the 29 participants in the game use group. That is, substance use decreased or remained the same for most users, although more for alcohol users than for tobacco users.

This suggests that participation in the study did support abstinence from substance use and that the effect might be stronger for alcohol use than tobacco use. Additional research is needed to determine if this effect is due to participation in a study or the intervention.

A second finding is that there were fewer positive results seen in substance use during the wait-list control period for those participants. In other words, there was a general increase in use, not decrease while controls waited. This suggests that being in a study and knowing that they would have to report on their substance use did not change their baseline behavior. This result strengthens the argument that it was use of the game that decreased substance use for participants, rather than participation in the study. Future research could examine this finding further.

A third finding is that scores for self-efficacy, attitude, and intended behavior went up significantly from baseline to week 3 (P=.03) for patients still using alcohol at baseline. This, together with the greater decrease in substance use for the group that selected alcohol as the substance to work on, suggests that the game benefits may be greater with respect to alcohol use.

Finally, it is noteworthy that few participants reported an increase in use of their substance. This was a concern due to the possibility, as seen in CET modalities, that exposing users to cues for substance use can have a triggering effect and thus have the potential to undermine recovery. This did not happen often, which is encouraging and a necessary result for further research into the use of game- or electronic-based CET adjunctive technologies.

Secondary findings revolve around participant enthusiasm for the game experience and impact the potential of the game as a supportive product in the future. Participants found the game engaging and fun.

Additionally, participants felt playing the game during recovery would help with relapse prevention and related behaviors. Agreement with this was highest after session 3.

There was a slight but intriguing difference in results for those who reported recovery for alcohol use as their primary goal versus those who chose recovery from tobacco use. Specifically, self-efficacy increased for those in recovery for alcohol use, but there was a minor, not statistically significant decrease in self-efficacy for former tobacco users. Differences between these 2 groups could be an area of further research, as misuse of both substances is harmful to health in the US population. Another improvement seen from pre- to postintervention follow-up was an increased intention to use health care, resources, and support groups.

There are limitations worth noting in this study. The study population was small and was a convenience sample of participants who were interested in maintaining their status in recovery, and thus they may not be typical of all individuals having substance use problems. Second, time was constrained, which impacted how often participants could use the game. In an ideal setting, the game experience would be available to participants more frequently, and a dose-response investigation could be conducted.

Kinect is a kinesthetic game and requires a minimum level of physical ability to move arms or legs, and this limits the reach of the game. A participant was able to use the game from a wheelchair in early testing, but more modifications are needed to effectively reach a mobility-limited population. Also, in terms of the physicality of the game, we believe that the possible aggressiveness of the game is balanced by the positive interactions of taking control of one’s environment; however, a psychological professional would need to evaluate whether this game is appropriate for individuals with aggressive tendencies.

Given this was a short-term feasibility study, long-term studies would need to be conducted to address the complexities of rehabilitation from various addictions.

Finally, Microsoft is no longer actively developing Kinect applications; although current versions of the Xbox One continue to support use (as of August 2017). Thus, future versions of the game should explore additional platforms while maintaining the kinesthetic element of game play and explore how this impacts results.

This study indicates that a serious game–based intervention has potential to be a useful part of recovery efforts for individuals seeking to maintain abstinence from alcohol or tobacco misuse or use. The use of a kinesthetic game based in a cue refusal theory framework-based intervention could prove a valuable adjunct to therapy in the future. Games have the ability to reach and engage a significant audience segment, and the use of an individually tailored game could expand potential treatment experiences.