The primary prevention of HIV infection remains a crucial priority. In 2011, there were 2.5 million new HIV infections worldwide [1]. In the United States, young minority men who have sex with men (MSM) are the most likely to become infected with HIV. Antiretroviral (ARV) medications that reduce the risk of acquiring HIV infection, namely pre-exposure prophylaxis (PrEP), are an efficacious and promising prevention strategy [1,2]. There have been significant advances regarding PrEP including the definitive demonstration that PrEP reduces HIV acquisition, regulatory approval of Truvada (tenofovir/emtricitabine or TDF/FTC) with an indication for sexual HIV prevention, and development of clinical prescribing guidelines. Despite these promising events, the practical implementation of PrEP is challenging [3-10]. Data show that PrEP’s safety and effectiveness could be greatly compromised by suboptimal adherence to treatment, and there is concern about the potential for an increase in HIV risk behavior among PrEP users [11-15]. Due to these challenges, the prescribing of PrEP should be accompanied by behavioral interventions.

Supporting the idea that adherence to medical care is an integral factor in PrEP’s effectiveness, the iPrEx, TDF2, and Partners PrEP clinical trials all conclusively showed that the level of protection from HIV infection depended on how consistently participants took prescribed medication [16-18]. Significantly greater levels of protection occurred among participants with detectable levels of ARVs in these trials. The impact of adherence was also underscored in the FEM-PrEP and VOICE studies, both of which found that few women had detectable levels of ARVs; thus, the studies were unable to demonstrate the efficacy of PrEP [15,19,20]. In 2013, the nested substudy of the Partners trial found that high (>80%) PrEP adherence was associated with 100% efficacy (95% CI 83.7%-100%) [17]. It is clear that the success of PrEP interventions highly depends on good adherence [14-17]. However, concerningly low rates of adherence to PrEP are seen in diverse samples of young Black MSM at high risk of acquiring HIV. Among a sample of 178 young MSM on PrEP (86% identified as Black/African American; median age 26 years; from North Carolina, Washington DC, and California), 36% did not meet the laboratory sensitivity threshold for ≥4 doses/week as measured by FTC/TDF levels in plasma and peripheral blood mononuclear cells at 26 weeks after PrEP initiation [21].

Adherence to PrEP medication is critical; however, engaging patients in comprehensive follow-up care is also imperative [22]. Treatment with PrEP requires consistent contact between patients and clinical providers and includes appointments with clinicians every 3 months for laboratory monitoring, HIV testing, and the detection and treatment of associated side effects [5,9,11-15]. Individuals who are at highest risk of HIV infection often come from populations that historically have been underserved by the health care system [23,24]. Therefore, engaging patients in care could be challenging and will require reinforcement and support for doctors and patients [13,23]. Consequently, behavioral interventions that promote adherence to comprehensive PrEP treatment will need to be tailored to underserved and at-risk populations and will need to reinforce the clinician-patient relationship.

Although there is great optimism about the use of PrEP for HIV prevention, there is concern that PrEP users may take more sexual risks or underutilize traditional risk reduction strategies, such as condom use and HIV and sexually transmitted infection (STI) testing of partners. Behavioral models (behavioral disinhibition and risk compensation) suggest that risk could increase with the reduction of self-imposed constraints or by decreasing individuals’ perceptions of HIV risk. This, in turn, could lead to increased incidence of HIV and other STIs [14,25-28]. Some mathematical and cost-effectiveness models have suggested that even small increases in risk behavior could offset or reverse PrEP’s protective benefits at the population level [13,14]. Data from the iPREX study showed only indirect evidence of increased risk behaviors, as those participants who engaged in unprotected anal sex were more likely to have detectable tenofovir levels than those with less sexual risk taking [17,29]. In real-world clinical settings in which people know that they are actively on a prophylactic medication, behavioral risk taking could measurably increase. Increases in risk behaviors have been documented in the context of microbicide trials, in vaccine trials, and among patients living with HIV on ARV therapy (ART) [30]. Therefore, HIV prevention counseling remains clinically relevant and prudent when prescribing PrEP. This practice is consistent with good clinical care and is recommended in interim guidelines for prescribing PrEP, which state that PrEP has the potential to contribute to effective and safe HIV prevention for MSM if “it is delivered as part of a comprehensive set of prevention services, including risk-reduction and PrEP medication adherence counseling” [22]. Therefore, in order to maximize the positive impact of PrEP, it is necessary to combine the prescription of PrEP with behavioral interventions that promote both adherence and the reduction of HIV risk behaviors [2]. However, these accompanying behavioral interventions need to be cost-effective and easily integrated into clinical settings in which PrEP is prescribed [15]. Additionally, interventions should be enjoyable and tailored to populations targeted for prevention with PrEP. Without these necessary components, integration of behavioral interventions into clinical settings cannot be realistically sustained.

There are many advantages to using newer interactive technology and gaming to promote adherence, rather than traditional face-to-face counseling, including scalability, efficiency, and cost-effectiveness. The use of an intervention that utilizes gaming technology is particularly compelling for use with younger adults who are heavy utilizers of mobile technology and who are most at risk for acquiring HIV. Games can attract and maintain attention, which is a key component for effective behavior change. Compelling, interactive, phone-based games can expose players to essential health-related content countless times and give players unlimited opportunities to rehearse new skills and receive personalized feedback on health choices made within the game [31-33]. Games have been shown to be efficacious in promoting fitness, improving weight management, and improving safer sex skills [31,33-35]. For example, a HIV/AIDS-prevention computer game called Life Challenge was developed by the New York State Department of Health to enhance safer sex negotiation by adolescents and young adults. The game showed significant improvement in self-efficacy for partner negotiation and condom skills for those who started with the least self-efficacy [33]. Two pregnancy prevention games, The Baby Game and Romance, designed for sexually active young adults showed trends in improving knowledge and attitudes about parenting and unprotected sexual behaviors [32].

Video games have also been applied to improve self-management skills and healthy behaviors in those living with asthma, diabetes, and cancer [36-40]. A video game named Re-Mission, designed for a wide age range of patients, namely 13-29 years old, with acute leukemia, lymphoma, and soft-tissue sarcoma, showed promising effects as well. Re-Mission was designed as an action-adventure game with the main character or protagonist shooting cancer-causing agents in the bloodstream. Players gained points and strength by adhering to medications in the game fantasy world. In a randomized control trial (RCT) with a 3-month follow-up, 375 male and female participants who played Re-Mission had significantly improved adherence to trimethoprim-sulfamethoxazole (P=.012) and 6-mercaptopurine (P=.002) compared to controls after an average of only 10.7 hours of play. Adherence to trimethoprim-sulfamethoxazole was tracked by electronic pill monitoring devices (n=200). The proportion of doses taken correctly by those playing Re-Mission was 19% greater than those in the control group. Self-efficacy (P=.011) and knowledge (P=.035) also increased significantly compared with the control group. Interestingly, the intervention did not affect subjective self-report measures of adherence but did affect the aforementioned objective measures [39,40]. Thus, appealing interactive games can target information, motivation, and skills for medical care and lead to a broad spectrum of desirable outcomes including increases in knowledge, attitude changes, and increased medication adherence [34,35,41-43].

Very few studies describe outcomes of theory-driven PrEP adherence interventions, and there are no other publications on gaming interventions to improve PrEP adherence. Fuchs et al [44] showed that a mobile health intervention using weekly bidirectional messaging (iText) was acceptable and demonstrated promising effects on PrEP adherence in a within-subjects trial design. This sample included MSM from San Francisco and Chicago (one-quarter were under 30 years of age, 13% were Black/African American, 11% were Latino, and 88% completed some college). A pre-post intervention regression discontinuity analysis using clinic-based pill counts showed a 50% reduction in missed doses (95% CI 16%-71%; P=.008) and a 77% improvement (95% CI 33%-92%; P=.007) when comparing pill counts at quarterly visits before and after iText enrollment. Liu et al [45] reported on the outcome of PrEPmate, a bidirectional text-messaging intervention, on study visit completion and tenofovir diphosphate (TFV-DP) concentrations assessed at 4, 12, 24, and 36 weeks among 121 participants (mean age 24 years; 27% Black, 36% Latino) living in Chicago. Participants who received PrEPmate were more likely to attend study visits (86% PrEPmate vs. 71% non-PrEPmate, OR 2.62, 95% CI 1.24-5.54) and have TFV-DP levels consistent with ≥4 doses/week (72% PrEPmate vs. 57% non-PrEPmate, OR 2.05, 95% CI 1.06–3.94). Although these studies showed promising preliminary effects and feasibility, there are no known definitive, scalable, evidence-based gaming interventions to improve adherence to PrEP.

The Information-Motivation-Behavioral Skills (IMB) model is a well-established conceptualization for improving adherence to treatment as well as decreasing HIV risk behaviors. HIV prevention and ART adherence interventions based on IMB have demonstrated efficacy [46-48], and reviews have suggested that interventions guided by accepted theories of change are more efficacious than those not driven by theory [49]. According to the IMB model, health information, motivation, and behavioral skills are the fundamental determinants of health behavior. In order for a PrEP-related intervention to be successful, the PrEP user must learn information that is directly relevant to PrEP adherence and HIV transmission. Knowledge is a necessary but not sufficient condition for change. Personal motivation to engage in HIV preventative behavior or adhere to treatment regimens (attitudes about health) and social motivation (perceived social and cultural support for performing these acts) are essential for change. Finally, skills for performing adherence behaviors and a sense of self-efficacy must be easily applied to an individual’s cultural and social setting. The most recent review of factors associated with PrEP adherence suggests that adherence can be facilitated by “accurate knowledge of medication benefits,” “medication optimism and self-efficacy for adherence,” and “support provided by peers and providers” [50]. Our intervention addresses these factors within the context of the IMB model. This model, consistent with Social Learning Theory, is broadly applicable and can be used to create theoretically consistent intervention content [51].

The aims of this project were to develop and test a cutting-edge, engaging, and entertaining app/gaming intervention to improve adherence to PrEP and build HIV prevention knowledge, skills, and behavior. The effectiveness of a novel, scalable, technology driven intervention to improve PrEP adherence is understudied. Consequently, through this RCT, we aimed to leverage technology to engage patients in learning skills, practicing behaviors, and improving motivation for adherence and healthy behaviors. We are not aware of any PrEP-related intervention that utilizes a theoretically informed game. An engaging, technology-based intervention may empower and engage PrEP users, aid over-burdened clinics, and result in improvements in health.

This research involving human subjects took place at collaborating sites. The protocol was reviewed by the Institutional Review Boards of Brown University, Lifespan/ Miriam/Rhode Island Hospital (RIH), and the University of Mississippi Medical Center (UMMC). This study is also registered on ClinicalTrails.gov (RCT02611362).

This research was divided into two major phases. First, a developmental phase (n=20, 18-35 years old, receiving PrEP) consisted of formative research to guide the development of the app/gaming intervention. Second, in an RCT phase, we evaluated the preliminary efficacy of the PrEP-related IMB app/gaming intervention compared to a comparison condition — a non-PrEP–related, non-IMB–informed game (n=60, 18-35 years old, receiving PrEP). Subjects were recruited from and consented in person at the UMMC.

All MSM initiating PrEP between the ages of 18 and 35 years were eligible for enrollment in each phase of study according to the following criteria: (1) English speaking, (2) initiating PrEP, (3) not enrolled in another PrEP-related study, (4) able to give consent or assent, and (5) not impaired by cognitive or medical limitations as per clinical assessment. Clinical assessment occurred by members of the research team with substantial prior clinical (medical and psychiatric) and research experience in care of young adults and adults. There was no overlap between subjects in the developmental and RCT phases. We enrolled only MSM because they are the group most at risk for acquiring HIV. Limiting the study to MSM allowed for the development of an iPhone app/game that was targeted, acceptable, and engaging for this specific at-risk population. Based on previously acquired data, it was estimated that 65% of participants would be Black/African American, 30% would be Hispanic, and the mean age would be 27 years. A total of 20 participants were recruited for the developmental phase and 60 for the RCT phase over a period of 24 months.

The gaming intervention developed for this project by Mission Critical Studios was titled Viral Combat. Viral Combat was consistent with the IMB Model of Health [52] and promoted increased self-efficacy for adherence to PrEP and HIV preventative behaviors by improving motivation and knowledge about health and prevention and by increasing mastery of skills. Successful games are intuitive, engaging, and inherently rewarding through action and feedback. Many of the attributes of a successful game are a natural fit for a successful IMB-informed intervention.

Participants playing Viral Combat fight off HIV and keep it from entering a virtual body. The game takes place on the surface of the skin, in the arterial system, and in the penile and anal canals. HIV-related educational material (eg, puzzles and quizzes) were tailored to prevent HIV (ie, importance of testing and using condoms) and not tailored to persons living with HIV (ie, how viral load affects transmission and health). Adherence messages were relevant to participants using PrEP for HIV prevention (ie, taking PrEP doses as prescribed corresponds to the level of protection). Viral Combat employs graphics, characters, and action content specifically chosen to appeal to our target population. All organ systems are vibrant and distinctive. Answering questions from the doctor or clinician and building adherence and HIV prevention knowledge allows each player to earn strength and points throughout Viral Combat. During play, if an answer to a question or skill-building exercise is wrong, the player is alerted, and the correct answer is explained. “Health Facts” reinforce HIV prevention information during scene changes. HIV prevention skills are built with a condom “puzzle” and the “condom use challenge” that promotes continued condom use with all partners. Each level in Viral Combat provides new challenges and unique and colorful environments. However, the mission stays the same: Kill the virus and build strength through taking medicine, learn HIV prevention information, improve motivation, and engage with healthy charcters in order to build skills. All character controls and gaming are achieved using touch screen technology on the phone with no additional accessories needed. Throughout the game, the terms “HIV,” “AIDS,” “antiretroviral,” and other identifying verbiage are avoided to protect the players’ medical status and to avoid possible stigmatization that could occur from someone seeing a participant playing the game (See Figure 1).

Subjects were reimbursed US $50 for participating in a qualitative interview during the developmental phase. During the qualitative interview phase, participants were also reimbursed US $5 for every level of the game they completed, up to US $30. During the RCT phase, subjects were reimbursed US $50 for baseline, 12-week, and 24-week assessments. Each participant in the RCT received an iPhone with a paid data service plan for the study duration (iPhone 4S; US $100; US $60/month data package through Lifespan/AT&T). Participants in both arms were given phones, because at the time of study commencement, iPhone ownership was not ubiquitous in this clinical population.

To protect participants’ confidentiality, the following measures were taken. Participants’ research data were identified using a numeric ID only. Any records containing potentially identifying information were kept secure and separate from other research data. All physical records were kept in a locked file, and electronic data were password-protected. Study materials were only accessible to research staff. Data collection took place in a secure and supervised clinical setting or with Health Insurance Portability and Accountability Act-compliant software (REDCap). All study personnel completed training in Human Subjects Research Protection and Health Insurance Portability and Accountability Act regulations (Collaborative Institutional Training Initiative Program) and received certification. These certifications were kept current in compliance with hospital policies.

Engagement in game analyses, provided by Mission Critical Studios, was participant nonspecific and generalized. For example, Mission Critical Studios collected data on game usage like any other app or website owner that shows percentage of times players of the game complete levels (eg, 50% of players completed level 5). These data were not linked to a particular participant or phone in use. The phone was designated as owned by Lifespan (the organization affiliated with the study) as opposed to the study participant. The gaming data were not connected nor linked to any participant’s name, demographic information, or health status.

To further protect the privacy of the study participants, we obtained a Certificate of Confidentiality from the US Department of Health and Human Services. With this certificate in place, the researchers cannot be forced to turn over identifying information about a study participant in any federal, state, or local criminal, administrative, legislative, or other proceedings. This certificate does not prevent a study participant from volunteering to turn over their research information nor does it prevent researchers from providing research-related information to others when requested by the study participant.

The use of ARV medications, such as PrEP, to reduce the risk of acquiring HIV infection is a promising prevention strategy. Optimal PrEP treatment requires simultaneous medical care and behavioral interventions to promote adherence and safer sexual behaviors. For interventions to be successful, they must be targeted to populations most at risk, as well as engaging and theory-driven. If Viral Combat proves to be effective in future trials, the intervention would be easy to scale up to larger clinical settings.

This study had limitations. The trial was unblinded; participants were aware of their study arm assignment, which could influence how they reported on outcomes of interest. Time spent gaming was collected by self-report, and individualized paradata were not collected. Therefore, individual participant’s playing time and engagement with the game were not verified. Also, the sample size was small and recruited from the same urban area of the South; therefore, findings may not be generalizable to MSM in other locations.

There are many advantages to using newer interactive technology, such as mobile gaming and text messages, rather than traditional face-to-face counseling, including scalability, efficiency, cost-effectiveness, and appeal. Given the significant health sequelae associated with HIV infections and the paucity of data on PrEP-related adherence and behavioral intervention programs, the knowledge to be gained from this research is significant.