Worldwide, cigarette-smoking–related death and illness are leading public health concerns, resulting in an estimated seven million deaths each year [1]. There are almost one billion smokers worldwide [1]. Thus, smoking cessation is a critical, worldwide public health concern [2].

Many efficacious smoking cessation interventions have been developed, yet ample room remains for improvement. Each year, one-third to one-half of all smokers attempt to quit at least once [3], but the annual incidence of successful quitting is less than 5% [4]. Media campaigns and cessation programs can increase quit attempts and successful cessation rates, respectively, by a margin of nearly 50% [3,5]. Smoking cessation rates that are better than those obtained with a placebo are achieved with: physician advice; counseling by health professionals; a variety of cognitive-behavioral, social-influence, and motivation-enhancement cessation programs; and drug treatments, including nicotine replacement therapies (gum, patch, spray, lozenge, and inhaler), selected antidepressant therapies (eg, bupropion), and nicotinic receptor agonist therapy (varenicline) [5,6]. Nevertheless, nearly 80% of smokers who attempt to quit do so without the assistance of any of these approaches [7]. Half or more consider counseling and cessation programs ineffective, and over a third consider pharmacotherapy ineffective [8]. Most young smokers report they would never use any of these methods, other than the nicotine patch (which only 50% would use) [9]. Thus, three major weaknesses of current approaches to smoking cessation are: underutilization, lack of appeal to smokers who wish to quit, and in general, modest efficacy in supporting smoking cessation. Clearly, more appealing and more efficacious smoking cessation interventions are needed.

Contingency management (CM) is one of the most efficacious aids for initiating smoking abstinence [10-12]. Contingency management for smoking cessation consists of delivering rewards (typically financial) contingent on objective evidence of smoking abstinence (eg, low levels of the combustion product carbon monoxide (CO) in the exhaled breath). Unfortunately, the effectiveness of CM on a population level has been limited by several constraints leading to low adoption rates and shorter than optimal treatment durations. These factors include the cost of providing the cash or cash-equivalent rewards [13-16], the distances that must be traveled, and the time required for participation in supervised monitoring procedures at a clinic [13,15]. Monetary rewards for smoking cessation can range from $100-$500 per person for approximately two 12-week interventions [11,17,18]. The cost of these payments limits the feasibility of widespread CM adoption. Furthermore, the ongoing nature of these costs limits the acceptability of longer-term treatment or booster sessions that could otherwise extend program effects (by reducing relapse).

CM procedures require biochemical verification of abstinence because participants are much more likely to falsify self-reports when rewards are delivered contingent on abstinence [19]. Carbon monoxide (CO) is one method to biochemically verify abstinence; however, the half-life of CO is short (approximately 3-6 hours), requiring at least twice-daily check-ins to verify abstinence [20]. To address this barrier, Dallery and colleagues developed an efficacious CM intervention that is delivered over the internet [11,17,21,22], in which participants are provided with a breath CO monitor, and remotely record and submit video clips of themselves providing their breath CO samples. More recently, mobile CM for smoking cessation (where participants use the camera on their smartphone to record and submit the video clips) has been shown to be feasible, acceptable, and efficacious in supporting smoking abstinence [23-25]. However, even these internet and mobile CM interventions rely on monetary incentives to support abstinence; thus, cost remains a barrier to widespread dissemination.

To directly address the remaining barriers for widely disseminating CM (ie, cost and sustainability), we proposed to develop a mobile game–based CM intervention for smoking cessation. As with existing mobile CM for smoking cessation, participants would be provided with a breath CO monitor and required to remotely record and submit video clips of themselves providing their breath CO samples twice daily. In our proposed mobile app, the monetary rewards typically used to incentivize bio-verified abstinence will be replaced with in‑game virtual-good rewards that can immediately be used to help players meet game objectives. Virtual goods can be provided by software at essentially no cost, yet they can have significant economic and monetary value (as evidenced by the multi-billion-dollar market for game-based virtual items) [26,27]. This suggests that rewards in the form of in-game content may readily substitute for monetary rewards in a CM procedure, drastically reducing cost while maintaining efficacy. The game will have the benefit of maximizing reward potency by minimizing delays to the receipt of rewards for abstinence once they are earned, as participants can immediately “consume” the rewards in the game. The proposed game design operationally encourages social support for smoking abstinence by imposing group contingencies [28], such as assembling players into teams and providing a reward that is only obtainable if all, or a majority of members reach a specified smoking cessation milestone (eg, no smoking for 24 hours). Operationally this design leverages self-interest (in obtaining access to the team reward) to incentivize social support (through interteam messaging) for others’ smoking cessation. The app will also enable standard, nonincentivized social support (eg, “click here to send congratulations to player C for <meeting a cessation milestone>”).

We previously published the results of an online survey of smokers to assess the social validity of a mobile game–based CM intervention for smoking cessation which uses virtual goods, instead of money, as rewards [29]. From a sample of 235 smokers recruited through Craigslist (ranging in age from 18-64 years), 75% reported playing video games. Among most smokers, 78% reported playing social games (ie, played casual games online). This rate is slightly higher than the population at large. Approximately 73% of all smokers and 70% of all video game players reported that contingent access to virtual rewards in the place of money would motivate smokers to abstain. Additionally, 75% of those surveyed would recommend or use a treatment such as this if they knew someone who wanted to quit, or if they were trying to quit themselves [29].

With the support of these promising outcomes, we developed and evaluated a prototype of Inspired (working title), a mobile game–based CM intervention to promote smoking cessation. The goal of the current project is to assess the feasibility and promise of Inspired by having treatment-seeking smokers play several levels of a prototype of the game and then ask them for qualitative feedback through a survey. We also indirectly observed feasibility and promise by recording users’ decisions to play an optional, extra level of the prototype.

Twenty-eight individuals participated in the prototype evaluation study (see Table 1 for demographic information). To analyze the VAS scores, prototype evaluations and their median (interquartile range, IQR) scores are presented in Table 2.

When an endorsement was defined by a ranking of 51 or higher (binary), at least 71% of participants endorsed statements that they would use Inspired (Q1), recommend Inspired to a friend (Q2), expect Inspired would help themselves (Q3) and others (Q4) quit smoking, consider both the game (Q6) and the intervention as a whole (Q5) to be “fun,” and that they would be more likely to use Inspired than any other smoking cessation aid they were familiar with (Q10a-e), including the nicotine patch (Q10b_, range 71%-100% , n=20-28, depending on the item; see Table 2).

The items with the lowest percentages of endorsement were Q3 (would help me), Q8 (add health benefit tips), and 10a (1st choice, 21/28, at least 75%), and the highest percentages of endorsement were Q6 (fun), and Q9 (add tips about cravings: 28/28, 100%). Alternatively, if endorsement was defined as a ranking of 67 or higher (trinary), at least 29% of participants (8/ 28) endorsed every item, at least 50% (14/28) endorsed all but two items (Q3 and Q7), and at least 75% (21/28) endorsed all but six items (Q3, Q7, Q8, Q10a, Q10c, and Q10e), with a range of at least 29%-89% (n=8-25) across all the items. The item that received the lowest percentage of endorsements was Q7 (fun relative to other games: 8/28, 29%), whereas the item that received the highest percentage of endorsements was Q2 (would recommend to a friend: 25/28, 89%). Additionally, when asked a multiple-choice question whether the game was fun, 63% (17/27) said “Yes,” 22% (6/27) said “Maybe, it has the potential to be fun,” and 15% (4/27) said, “No.” When asked the question, “If the Inspired program had been demonstrated to be just as effective as other smoking cessation aids (such as the nicotine patch), and included ongoing access to the game, the monitoring program, and a CO monitor that was yours to keep, how much would you be willing to pay for the program?”, the mean (SD) responses were $14.40 per month (SD $16.20) or $133 for a one-time purchase (SD $186).

In the free-response portion of the survey, when participants were asked what they liked best about the Inspired game, the most frequent response was that they liked the game itself, either because it was fun, creative, challenging, or because they liked the puzzle style of the game (12/28, 43%). Participants also reported that they liked the rewards delivered for abstinence (8/28, 29%), the community and social support aspects of the game (7/28, 24%), the simple instructions (5/28, 19%), the graphics (4/28, 14%), and the ability of the game to serve as a distraction from smoking (4/28, 14%). A couple of participants also mentioned that they liked the CO monitor (3/28, 10%). When asked what they liked least, the most frequent response was that they thought the game lacked variety (8/28, 29%). Participants also noted that glitches with the game needed to be resolved (eg, swiping gems to the correct location, screen loading, and more; 4/28, 14%), the graphics could be improved (4/28, 14%), and that the game was too easy (3/28, 10%). Other comments included: the CO meter was too bulky, there were too many screens, the game was too challenging, they did not like the idea of social support, and that the game felt disconnected from the rewards. Two participants did not indicate any weaknesses with the proposed intervention (2/28, 7%). Participants were also asked to give suggestions about how to move forward with the game, and the only response that appeared more than once was to improve variety in the game (9/28, 33%). Other suggestions were to provide real rewards for abstinence, resolve glitches, improve graphics, and to explore having insurance companies cover the cost of the game.

Fifty-eight percent of participants (15/26) given the option to play an extra level of the game chose to do so, and 86% (23/28) signed up to be notified when the full version of the game was released.

This was the first evaluation of a mobile game–based CM intervention for smoking cessation. The prototype of Inspired was endorsed on multiple dimensions by a group of treatment-seeking smokers. Most participants reported that they felt the game would help them, or a friend, quit smoking. For Inspired to be effective at motivating smokers to quit using game-based rewards as incentives for abstinence, it is critically important that the game be fun. If the game is not fun, the virtual rewards will not be effective at reinforcing abstinence. In the current study, the extent to which the prototype of Inspired was fun was evaluated in multiple ways. To begin, VAS responses in the prototype evaluation survey (see Table 2), addressed whether participants thought the Inspired intervention program was fun (Q5), and 71%-93% (n=20-26) of participants agreed, depending on how an “endorsement” was defined. Additionally, 82% (23/28) of participants reported that the game was either already fun or had the potential to become fun with further development. Finally, probably one of the strongest indicators that the game has potential to be both fun and effective at supporting smoking cessation was the behavior of participants when they were given the option to play an extra level. More than half of participants (16/ 28, 58%) decided to play the extra, optional level, which meant they may have delayed smoking their next cigarette (following about one hour of abstinence; ie, the duration of the study), as well as getting paid, by at least five additional minutes. The goal of the game is to decrease smoking, and the fact that the prototype for Inspired may have been capable of displacing smoking for even a brief period is encouraging. Between 57%-76% of participants (n=16-21) said they were more likely to use Inspired than any other smoking cessation aid, including evidence-based pharmacological interventions such as varenicline and the nicotine patch, which are endorsed in the Clinical Practice Guidelines for smoking cessation [6]. This finding supports previous research suggesting that the use of pharmacological or other evidence-based interventions may be not be preferred among individuals attempting to quit [8,9]. It should also be noted that mobile game–based smoking cessation is not incompatible with these other interventions but could instead serve as a fun, alternative, and yet evidence-based complement to these existing interventions.

Item Q7 on the prototype evaluation survey was included to help inform the game design team of how they were doing, at this early stage of development. Participants were asked to rank the intervention game relative to every other game they had ever played on their smartphone. Because participants were asked to compare this prototype of the game, which was less than 10% developed at the time of testing, to existing and fully developed mobile games, a median VAS of 64 was a promising outcome. Furthermore, the question asking how much participants would be willing to pay for the intervention suggests that there may be commercial viability of the proposed game, with answers ranging from $1.99-$60 per month, or $9.99-$700 as a one-time fee. Finally, 85% (23/27) of study participants signed up to be notified when Inspired becomes available so they can use it if they have not yet successfully quit smoking by that time, further supporting the potential commercial viability of the game.

Participants provided useful feedback for moving forward with game development, probably the most consistent of which involved adding variation to the game to keep it interesting and engaging. Overall, feedback about the type of game, the game graphics, as well as the social elements and CO monitoring in the game, were viewed favorably. Smaller, portable versions of the CO monitor have come to market since this prototype evaluation was conducted, thereby addressing concerns about the meter being bulky (eg, CO by Bedfont).

It should be noted that Inspired was designed to address smoking cessation specifically; therefore, multiple game design decisions were made to address the unique needs of individuals trying to quit. First, the core game mechanic, which required players to hold the device with both hands in landscape view and swipe pollen-gems into locations on various lotus flowers (see Figure 1), was chosen to make it difficult to simultaneously smoke while playing the game. Second, each level of the game was designed for casual gameplay, lasting approximately five-minutes, to reflect how long it might normally take to smoke a cigarette [31]. Third, visual elements of the game motif were associated with wellbeing and growth. The design intentionally avoided anything that might serve as a cue for smoking (eg, smoke, certain words, and more). This was done to avoid having the game elicit cue-induced cravings and subsequent smoking, and to enhance the ability of the game to displace smoking. Although not asked to comment on this directly, participants were asked to indirectly address this decision with Q8 and Q9 on the survey (see Table 2), where they reported the mobile game would be stronger if it incorporated messages about the health benefits of smoking cessation and tips about avoiding cravings. High endorsements on these two items suggested that future iterations of the game should explore incorporating this information in to the game, but in a way that does not also elicit cravings or trigger smoking.

The current study has a few limitations worth noting. First, because of the small sample size it was not possible to determine whether there were differences in endorsements, or other measures, between high and low nicotine dependent participants. Some items seemed to suggest differences, but it was not possible to evaluate the potential differences using inferential statistics. Second, information about the participants' individual histories with playing games, particularly mobile games, was not collected. Anecdotally, it was made clear that there was a range of past experiences. However, because the game is being designed as a smoking cessation aid, a level of heterogeneity of past experiences with video games is expected among the target population of treatment-seeking smokers as well. A third limitation is the possibility that participants rated the game favorably to avoid offending the experimenters and game designers (ie, demand characteristics). To mitigate this concern, we made it clear to participants at the beginning of the group sessions that they were being asked to give an honest evaluation of a very early version of the game, and that their feedback could help shape the future development of the game. Although it is impossible to rule out potential bias, participants felt comfortable giving the game a low rank-order relative to other, commercially available games (Q7 received the lowest scores), as well as other available smoking cessation aids. In the free response section participants provided useful feedback for improving the program. Finally, the fact that over half of the participants voluntarily played an extra, unrequired level of the game alleviates some concerns about bias; however, it cannot be ruled out.

Although there are other digital games that have been evaluated for smoking cessation, none are based on the empirically supported procedures and theoretical foundations of contingency management and behavior analysis [32-34]. This study is the first to show that a mobile game–based CM intervention has potential to be both helpful and fun to smokers who wish to quit. The prototype evaluation suggests that the proposed game would not only reduce the cost of delivering CM for smoking cessation and enable extended program reach and duration, it might also be preferred over currently extant smoking cessation aids and interventions.