This is an open-access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work, first published in JMIR Serious Games, is properly cited. The complete bibliographic information, a link to the original publication on http://games.jmir.org, as well as this copyright and license information must be included.
Improving the reach of behavioral health services to young adult veterans is a policy priority.
The objective of our study was to explore differences in video game playing by behavioral health need for young adult veterans to identify potential conditions for which video games could be used as a modality for behavioral health services.
We replicated analyses from two cross-sectional, community-based surveys of young adult veterans in the United States and examined the differences in time spent playing video games by whether participants screened positive for behavioral health issues and received the required behavioral health services.
Pooling data across studies, participants with a positive mental health screen for depression or posttraumatic stress disorder (PTSD) spent 4.74 more hours per week (95% CI 2.54-6.94) playing video games. Among participants with a positive screen for a substance use disorder, those who had received substance use services since discharge spent 0.75 more days per week (95% CI 0.28-1.21) playing video games than participants who had not received any substance use services since discharge.
We identified the strongest evidence that participants with a positive PTSD or depression screen and participants with a positive screen for a substance use disorder who also received substance use services since their discharge from active duty spent more time playing video games. Future development and evaluation of video games as modalities for enhancing and increasing access to behavioral health services should be explored for this population.
Behavioral health issues such as posttraumatic stress disorder (PTSD), depressive disorders, and substance use disorders (SUDs) are common diagnoses among veterans from recent conflicts in Iraq and Afghanistan [
Young adult veterans report multiple barriers to seeking and receiving behavioral health services in traditional settings and formats. These include inconvenience of appointments, concerns about high costs, perceived stigma from peers, beliefs that they can handle symptoms on their own, and living in rural settings that are far from care settings [
Video games have the potential to improve the reach of behavioral health services [
In this exploratory study, we replicated analyses from two cross-sectional, community-based surveys to explore the plausibility of video games as a modality for behavioral health services for young adult US veterans. The lack of data on veteran video game playing precluded us from making clear a priori hypotheses regarding the prevalence of video game playing in the sample. As video game-based interventions appeal more to those who play video games generally [
Data presented in this manuscript are from two surveys conducted as part of a larger randomized controlled trial (RCT) of a Web-based normative feedback intervention for heavy drinking young adult veterans [
We recruited nontreatment seeking young adult (age, 18-34 years) veteran participants in both samples through advertisements on Facebook that did not mention video games or behavioral health. We have previously reported comprehensive details of the recruitment strategy and methods for validating veteran participants for Study 1 [
Study 1 involved a survey on the behavioral health symptoms of a large general sample of young adult veterans recruited outside of VHA settings. We targeted a series of Facebook ads to users between the ages of 18 and 40 years who expressed an interest in (ie, “liked”) specific veteran or military Facebook pages as well as media (movies, TV shows, and video games) related to military (eg, Act of Valor, Generation Kill, Call of Duty). Interested Facebook users who clicked on ads were directed to a Web-based informational statement and consent form. Eligible participants who consented to participate were first verified to be actual veterans using data check procedures we have described in detail elsewhere [
Study 2 involved a screening and baseline survey for an RCT of a brief, Web-based, personalized normative drinking intervention, where participants saw feedback about their drinking (intervention) or video game playing behavior (control) compared with their peers. As with Study 1, participants clicked on targeted Facebook ads, although we did not include ads targeting any media regarding video games (eg, Call of Duty). The eligibility criteria were the same across studies, except that participants in Study 2 needed to score at least a 3 (females) or 4 (males) on the 10-item Alcohol Use Disorders Identification Test (AUDIT) [
We recruited 1023 young adult veterans overall, of whom 552 (53.9%) reported playing video games at least 1 hour per week. To match the eligibility criteria of Study 2, we restricted the subsample who reported playing video games at least 1 hour per week to the 350 veterans who also had scores of at least 3 (females) or 4 (males) on AUDIT.
We recruited 784 young adult veterans overall. Because we were interested in veterans who reported any video game use for analyses in the current study, we restricted our sample to 582 veterans (74.2%, 582/784) who reported playing video games at least 1 hour per week.
For both studies, we collected information on demographics, behavioral health symptoms, behavioral health services use, and video game behaviors. In this manuscript, we report analyses on similar constructs assessed in both Study 1 and Study 2, although we operationalized several constructs using different measures (
Participants in both studies filled out the same measures regarding age, gender, ethnicity or race, education, marital status, number of children, annual household income, and branch of military service.
Participants completed brief screening measures for behavioral health problems.
In Study 1, we assessed PTSD symptoms with the 4-item Primary Care PTSD scale (PC-PTSD). A score of 3 or higher (ie, participants endorsed “yes” for 3 of the 4 PTSD symptoms) on PC-PTSD indicated a probable diagnosis of PTSD [
In Study 1, we assessed depression symptoms with the 2-item Patient Health Questionnaire (PHQ-2). Participants rated two symptoms (ie, “little interest or pleasure in doing things” and “feeling down, depressed, or hopeless”) from 0 “not at all” to 3 “nearly every day” for the past 2 weeks. A score of 2 on the PHQ-2 indicated screening for a depression diagnosis [
In both studies, we assessed alcohol use disorder (AUD) symptoms in the past year using AUDIT [
We asked participants if they used any cannabis or marijuana in the past 6 months (yes or no), and if so, how many days in the past month did they use. The Study 1 survey referred to the drug as cannabis and the Study 2 survey referred to it as marijuana.
In both studies, participants indicated whether they had attended any appointments (in any setting: VHA, Vet Centers, or community providers) for mental health concerns or substance use concerns since discharge from active duty in the past month or year.
In both studies, participants indicated the typical number of hours they played video games per day, hours they played video games per week, and days they played video games per week in the past 30 days using slightly different methods. In Study 1, participants indicated how many hours on each day of the week they typically played video games, while in Study 2, they responded to 3 single items about the hours per day, hours per week, and days per week they typically played video games. In both studies, participants were asked to consider computer-based games, console video games, arcade video games, mobile phone or tablet games, or Web-based JavaScript games.
We first calculated descriptive statistics for, and differences in, demographics between Study 1 and Study 2 samples. Then, we used Welch’s
We included 350 participants from Study 1 and 582 participants from Study 2 (see
Participant characteristics.
Demographics | Study 1 (N=350) | Study 2 (N=582) | ||||||||
Age, mean (SD) | 28.4 (3.4) | 28.7 (3.4) | .11 | |||||||
Male, n (%) | 323 (92.6) | 505 (86.8) | .009 | |||||||
Hispanic ethnicity, n (%) | 72 (20.6) | 60 (10.3) | <.001 | |||||||
White | 273 (78.0) | 499 (85.7) | ||||||||
Other | 77 (22.0) | 83 (14.3) | ||||||||
Some college or less | 286 (81.7) | 437 (80.1) | ||||||||
College graduate | 64 (18.3) | 116 (19.9) | ||||||||
No | 187 (53.4) | 382 (65.6) | ||||||||
Yes | 163 (46.6) | 200 (34.4) | ||||||||
<US $10,000 | 29 (8.3) | 32 (5.5) | ||||||||
US $10,000 to US $14,999 | 29 (8.3) | 46 (7.9) | ||||||||
US $15,000 to US $24,999 | 70 (20.0) | 90 (15.5) | ||||||||
US $25,000 to US $49,999 | 121 (34.6) | 238 (40.9) | ||||||||
US $50,000 to US $99,999 | 84 (24.0) | 142 (24.4) | ||||||||
US $100,000 to US $149,999 | 14 (4.0) | 28 (4.8) | ||||||||
US $150,000 to US $199,999 | 3 (0.9) | 3 (0.9) | ||||||||
US $200,000 + | 0 (0.0) | 1 (0.2) | ||||||||
Married, n (%) | 186 (53.1) | 278 (47.8) | .13 | |||||||
Number of children, mean (SD) | 1.4 (1.5) | 1.3 (1.4) | .14 | |||||||
Number of children living at home, mean (SD) | 1.7 (1.1) | 1.7 (1.3) | .82 | |||||||
Air Force | 22 (6.3) | 57 (9.8) | ||||||||
Army | 215 (61.4) | 348 (59.8) | ||||||||
Marine Corps | 87 (24.9) | 129 (22.2) | ||||||||
Navy | 26 (7.4) | 48 (8.2) | ||||||||
Positive screen for posttraumatic stress disorder | 152 (43.4) | 227 (39.0) | .21 | |||||||
Positive screen for depression | 169 (48.3) | 271 (46.6) | .66 | |||||||
Positive screen for disorder, n (%) | 151 (43.1) | 174 (29.9) | <.001 | |||||||
Total drinking days, mean (SD) | 10.4 (9.1) | 12.4 (8.8) | .001 | |||||||
Drinks per drinking day, mean (SD) | 4.8 (4.3) | 4.7 (3.3) | .77 | |||||||
Heavy drinking occasions, mean (SD) | 4.6 (6.2) | 5.8 (7.2) | .01 | |||||||
Max drinks on a drinking day, mean (SD) | 8.5 (6.0) | 9.4 (6.0) | .03 | |||||||
Alcohol consequences, mean (SD) | 7.8 (7.1) | 7.6 (6.8) | .58 | |||||||
Cannabis use in past 6 months, n (%) | 106 (41.9) | 169 (29.0) | <.001 | |||||||
Total number of cannabis use days, mean (SD) | 9.9 (11.4) | 3.3 (8.6) | <.001 | |||||||
Services since discharge | 183 (52.3) | 338 (58.1) | .10 | |||||||
Services in past year | 119 (34.0) | 238 (40.9) | .04 | |||||||
Services in past month | 42 (12.0) | 101 (17.4) | .04 | |||||||
Services since discharge | 175 (50.0) | 320 (55.0) | .16 | |||||||
Services in past year | 116 (33.1) | 228 (39.2) | .08 | |||||||
Services in past month | 42 (12.0) | 98 (16.8) | .06 | |||||||
Services since discharge | 67 (19.1) | 121 (20.8) | .60 | |||||||
Services in past year | 31 (8.9) | 66 (11.3) | .28 | |||||||
Services in past month | 1 (0.3) | 18 (3.1) | .007 | |||||||
Total hours spent playing per day | 2.3 (1.8) | 3.5 (3.2) | <.001 | |||||||
Total hours spent playing per week | 12.8 (13.5) | 18.4 (21.9) | <.001 | |||||||
Total days spent playing per week | 5.0 (2.3) | 4.7 (2.2) | .046 |
a
Similar proportions of patients screened positive for PTSD (Study 1, 152/350, 43.4%; Study 2, 227/582, 39.0%) and depressive disorder (Study 1, 169/350, 48.3%; Study 2, 271/582, 46.6%). Participants consumed about 5 drinks per drinking day. Screening positive for an AUD was more likely for participants in Study 1 (151/350, 43.1%) than in Study 2 (174/582, 29.9%), χ21=16.3,
Participants in Study 1 reported playing video games fewer hours per day (Study 1 mean 2.3 [SD 1.8]; Study 2 mean 3.5 [SD 3.2];
Within each study, participants with any positive screen (PTSD, depression, AUD, or cannabis use) did not differ significantly from participants without any positive screen on video game behavior, while participants with either positive mental health screen (PTSD, depression) spent more hours per day and per week playing video games than those without a positive mental health screen (
No association was found between video game use and either any services receipt, mental health services receipt, or substance use services receipt within both studies. Pooling data across studies, participants with any positive screen for a behavioral health issue who had received any type of behavioral health services (mental health services, substance use services) since discharge from active duty spent 2.67 more hours per week (95% CI 0.14-5.20) and 0.48 more days per week (95% CI 0.14-0.82) playing video games than participants with any positive screen who had not received any type of behavioral health services since discharge. Participants with any positive SUD screen who had received any type of substance use services since discharge spent 0.75 more days per week (95% CI 0.28-1.21) playing video games than participants with any positive SUD screen who had not received any type of substance use services since discharge.
We examined the video game playing behavior of two separate samples of young adult veterans recruited online. First, we found evidence across the two samples that most young veterans played video games: 54% of a general sample of young veterans and 74% of a sample of young adult veteran drinkers reported playing video games at least 1 hour per week. Next, among the video game players, we found that young adult veterans spent about 13-18 hours per week playing video games and about 2.5-3.5 hours per day playing video games. In a typical week, young adult veterans played video games on most days of the week. These findings suggest that video games might be a feasible intervention modality for young veterans generally and for behavioral health concerns specifically.
While most analyses did not yield differences that were replicated across both studies, we did find several replicated differences in video game behaviors among young adult veterans depending on their screening positive for a behavioral health issue as well as their receiving services for a behavioral health need. Regarding screening positive for a behavioral health issue, we identified the strongest evidence for more hours per day and per week playing video games among participants with a positive screen for both PTSD and depression compared with those without positive screens for these conditions. Regarding the receipt of services for a behavioral health need, we identified the strongest evidence that participants who had a positive SUD screen and received substance use services since discharge spent more days per week playing video games than those with a positive SUD screen who had not received substance use services.
Although most veterans in our sample played video games and those that did played quite often, our findings aimed to identify the potential groups of young adult veterans for the development and evaluation of video games as a modality for behavioral health services. Specifically, with respect to playing video games more per day and per week, young adult veterans screening positive for PTSD and depression may be more familiar with and dedicate more time to behavioral health services delivered via video games because these veterans already play more video games and more frequently than those without these issues and those not receiving services. For this population, relatively more intensive video game-based interventions might be acceptable. For example, previous research has demonstrated the feasibility of incorporating the core components of traditional cognitive behavioral therapies and exercise-based interventions for mental health into engaging, video games in Web-based, computer, console, and application-based formats [
Several strengths and limitations are worth noting. Strengths of this study include replicating analyses from two independent samples to reduce the rate of false positives [
Findings suggest several avenues of future research as well as collaborations between researchers and video game developers. First, analyses from this study would benefit from direct, preregistered replications to strengthen the credibility of our findings. Direct assessment of acceptability and willingness to engage with video games for behavioral health services should be incorporated into this research [
Codebook for Grant, Spears, and Pedersen (Study 1 and Study 2).
Comparison of measures in Study 1 and Study 2.
Exploratory replication analyses of video game use by young adult veterans.
Markdown analytic code and output for exploratory analyses.
De-identified dataset.
alcohol use disorder
Alcohol Use Disorders Identification Test
posttraumatic stress disorder
randomized controlled trial
substance use disorder
Veterans Health Administration
This research was supported by grant R34AA022400 awarded to ERP from the National Institute on Alcohol Abuse and Alcoholism (NIAAA).
SG’s spouse is a salaried-employee of Eli Lilly and Company and owns stock. SG has accompanied his spouse on company-sponsored travel. All other authors declare no conflicts of interest.