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 the Journal of Medical Internet Research, is properly cited. The complete bibliographic information, a link to the original publication on http://www.jmir.org/, as well as this copyright and license information must be included.
Human immunodeficiency virus (HIV) disproportionately impacts minority youth. Interventions to decrease HIV sexual risk are needed.
We hypothesized that an engaging theory-based digital health intervention in the form of an interactive video game would improve sexual health outcomes in adolescents.
Participants aged 11 to 14 years from 12 community afterschool, school, and summer programs were randomized 1:1 to play up to 16 hours of an experimental video game or control video games over 6 weeks. Assessments were conducted at 6 weeks and at 3, 6, and 12 months. Primary outcome was delay of initiation of vaginal/anal intercourse. Secondary outcomes included sexual health attitudes, knowledge, and intentions. We examined outcomes by gender and age.
A total of 333 participants were randomized to play the intervention (n=166) or control games (n=167): 295 (88.6%) were racial/ethnic minorities, 177 (53.2%) were boys, and the mean age was 12.9 (1.1) years. At 12 months, for the 258 (84.6%) participants with available data, 94.6% (122/129) in the intervention group versus 95.4% (123/129) in the control group delayed initiation of intercourse (relative risk=0.99, 95% CI 0.94-1.05,
An interactive video game intervention improves sexual health attitudes and knowledge in minority adolescents for at least 12 months.
Clinicaltrials.gov NCT01666496; https://clinicaltrials.gov/ct2/show/NCT01666496 (Archived by WebCite at http://www.webcitation.org/6syumc9C0).
Human immunodeficiency virus (HIV) disease and sexually transmitted infections (STIs) significantly impact young people, with racial/ethnic minority youth disproportionately affected. Nearly 10,000 youth were diagnosed with HIV in 2014 in the United States, accounting for 22% of new infections [
Serious games, defined as video games for a primary purpose beyond pure entertainment [
Our objective was to test the efficacy of a digital health intervention in the form of a theory-driven interactive video game intervention compared to a set of control games on sexual risk behaviors, knowledge, attitudes, and intentions in a population of racial/ethnic minority adolescents. We hypothesized that a highly engaging theory-based video game intervention would have a greater impact on our outcomes of interest than a set of control games.
We conducted a randomized controlled trial in 12 urban community-based settings consisting of seven school-based afterschool programs, four independent afterschool programs, and one summer camp. A description of the development of the PlayForward: Elm City Stories (PlayForward) intervention and the trial design have been published [
Participant eligibility included ages 11 to 14 years, speaks English, able to provide assent and parental/legal guardian consent, and willing to play video games for up to 75 minutes twice weekly for 6 weeks. Although the primary outcome of this study was delay of initiation of sexual intercourse, we enrolled participants who had already initiated intercourse because we did not want potential participants to provide incorrect information regarding their sexual activities to gain or avoid access to the study. These participants (n=6) were not included in these analyses. Participants were provided with an age-appropriate study description and parents/legal guardians were informed that the study focused on promoting healthy behaviors and reducing risk in adolescents. If an adolescent or parent/guardian did not wish to participate, they were not included in the study. All data were collected from 2013 to 2015 and analyzed in 2016.
Enrolled participants were randomized in an unmasked fashion to the PlayForward group or to a set of 12 attention-time control video games with all participants playing the games on iPad tablets. A single randomization scheme was generated and written in TrialDB, a customizable Web-based clinical trials database system [
Participants played PlayForward or a set of control games on the iPad. PlayForward is a two-dimensional, role-playing adventure video game [
The attention/time control games consisted of 12 video games such as Angry Birds, Dragonbox, and Subway Surfer. They were devoid of content relevant to our study goals and mirrored the number of sessions and length of gameplay in the experimental group [
Participants played their assigned game(s) for two sessions per week, approximately 1 hour per session, for 6 weeks on-site at their program [
The PlayForward game software records in-game data assessing intervention exposure and fidelity [
The primary outcome was delay of initiation of sexual intercourse (defined as initiation of vaginal or anal intercourse) at 12 months post-baseline [
Power calculations were based on data from published studies [
The primary comparison evaluated the effect of PlayForward compared with the control video games on the delay of initiation of sexual intercourse measured at 12 months postrandomization and constructed as a binary outcome (delayed beyond 12 months/initiated before 12 months). The primary analysis was carried out as an intent-to-treat analysis and adjusted for gender and age (the randomization stratification variables). Sensitivity analyses were carried out with missing responses being assigned both as delay and not delay of initiation of intercourse.
Differences in scores in the secondary outcomes were compared between the two groups at the time points using longitudinal mixed-effects models. Changes in secondary outcome measures since baseline were assessed in repeated measures models (with unstructured covariance), with the assigned baseline values, study group, gender, age, study time point, and study group*time interactions used as covariates. Least squares (LS) means and standard error were plotted for each secondary outcome at each time point. Statistical analyses were done using SAS version 9.4 (SAS Institute Inc, Cary, NC, USA). All protocols were reviewed at intervals by a DSMB.
A total of 333 participants were recruited and enrolled into the study between February 26, 2013 and May 16, 2014; 166 were assigned to the PlayForward intervention and 167 were assigned to the control condition (
The primary outcome was delay of initiation of sexual intercourse. Six participants (PlayForward: n=4; control: n=2) who had engaged in intercourse before baseline were removed from the primary analysis because they had already reached the primary outcome (
Overall, the rates of initiating sexual intercourse were low in both groups. There were no differences in rates of delaying initiation of intercourse at each time point in the PlayForward versus the control groups. At 12 months, for the 258 of 304 (84.9%) participants from whom data were available, 122 of 129 (94.6%, 95% CI 89.1%-97.8%) in the PlayForward group versus 123 of 129 (95.4%, 95% CI 90.2%-98.3%) in the control group had delayed the initiation of intercourse (relative risk=0.99, 95% CI 0.94-1.05,
Over the 12-month follow-up period, the PlayForward group demonstrated an improvement in attitudes about sexual health compared to the control group (LS means difference 0.37, 95% CI 0.01-0.72,
Over the 12-month follow-up period, the PlayForward group demonstrated an increase in sexual health knowledge compared to the control group (LS means difference 1.13, 95% CI 0.64-1.61,
Over the 12-month follow-up period, there were no differences in intentions to delay the initiation of intercourse between the two groups (LS means difference 0.10, 95% CI –0.23 to 0.43,
Enrollment and follow-up flow diagram for videogame intervention trial for sexual risk reduction. Note: Assessment refers to primary outcome assessment (delay of initiation of sexual intercourse);% is of active participants. Participants who did not initiate game play were still considered active and assessed for study outcomes. A total of six participants (four in PlayForward; two in control) who had initiated sexual intercourse (per primary outcome definition) at baseline were removed from the analysis of primary outcome.
Baseline demographic and clinical characteristics of participants.
Characteristics | Control (n=167) | PlayForward (n=166) | Total (N=333) | |
Male | 89 (53.3) | 88 (53.0) | 177 (53.2) | |
Female | 78 (46.7) | 78 (47.0) | 156 (46.8) | |
Age (years), mean (SD) | 12.9 (1.1) | 12.9 (1.1) | 12.9 (1.1) | |
11 | 41 (24.6) | 42 (25.3) | 83 (24.9) | |
12 | 45 (26.9) | 45 (27.1) | 90 (27.0) | |
13 | 45 (26.9) | 45 (27.1) | 90 (27.0) | |
14 | 36 (21.6) | 34 (20.5) | 70 (21.0) | |
White | 14 (8.6) | 17 (10.4) | 31 (9.5) | |
Black | 65 (40.1) | 70 (42.7) | 135 (41.4) | |
Other | 80 (49.4) | 76 (46.3) | 156 (47.9) | |
Unknown | 3 (1.9) | 1 (.6) | 4 (1.2) | |
Hispanic | 87 (55.1) | 85 (55.9) | 172 (55.5) | |
Non-Hispanic | 71 (44.9) | 67 (44.1) | 138 (44.5) | |
Sexual health attitudes score, mean (SD) | 10.1 (2.4) | 10.1 (2.5) | 10.1 (2.5) | |
Sexual health knowledge score, mean (SD) | 6.5 (2.8) | 6.1 (2.6) | 6.3 (2.7) | |
Intentions to delay initiation of sex score, mean (SD) | 14.9 (2.0) | 14.6 (2.2) | 14.8 (2.1) |
Delay of initiation of sexual intercourse by study condition.a
Behavior | Control | PlayForward | Total, n (%) | ||||
n (%) | 95% CIb | n (%) | 95% CIb | ||||
Delay of initiation of sexual intercourse | 165 (100) | 162 (100) | 327 (100)a | ||||
>.99 | |||||||
Delay of initiation of sexual intercourse | 132 (97.8) | 93.6-99.5 | 132 (98.5) | 94.7-99.8 | 264 (98.1) | ||
Initiation of sexual intercourse | 3 (1.8) | 2 (1.2) | 5 (1.5) | ||||
.72 | |||||||
Delay of initiation of sexual intercourse | 133 (97.8) | 93.7-99.54 | 127 (97.0) | 92.4-99.2 | 260 (97.4) | ||
Initiation of sexual intercourse | 3 (1.8) | 4 (2.5) | 7 (2.1) | ||||
.72 | |||||||
Delay of initiation of sexual intercourse | 126 (97.7) | 93.4-99.5 | 120 (96.8) | 92.0-99.1 | 246 (97.2) | ||
Initiation of sexual intercourse | 3 (1.8) | 4 (2.5) | 7 (2.1) | ||||
>.99 | |||||||
Delay of initiation of sexual intercourse | 123 (95.4) | 90.2-98.3 | 122 (94.6) | 89.1-97.8 | 245 (95.0) | ||
Initiation of sexual intercourse | 6 (3.6) | 7 (4.3) | 13 (4.0) |
aA total of six participants (control: n=2; PlayForward: n=4), who at baseline were identified (per primary outcome definition) as being engaged in sexual intercourse, were removed from the primary analysis because they had already reached the primary outcome.
bExact 95% confidence intervals are provided for the main outcome (delay of initiation of sexual intercourse).
c
Changes in attitudes, knowledge, and intentions by study condition for total group by gender and by age. PF: PlayForward.
In a cohort of community-based racial/ethnic minority adolescents, we found low rates of initiation of sexual intercourse over the 12-month follow-up in the PlayForward and control groups. Those who played PlayForward demonstrated greater improvement in attitudes around sexual health and greater increases in sexual health knowledge over 12 months than the control group.
This study is unique in that it is based in community settings targeting HIV risk in teens using engaging and novel methods that possess the potential for widespread dissemination and impact. To our knowledge, this is the first randomized controlled trial demonstrating the efficacy of a portable sexual health serious video game intervention impacting sexual attitudes and knowledge. The PlayForward game includes many components of programs that have been shown to be effective in reducing sexual risk in adolescents [
Our findings are consistent with and complement findings from studies demonstrating the effectiveness of school-based interventions targeting sexual risk reduction in adolescents of similar ages and racial/ethnic backgrounds [
Although there have been long-standing efforts for comprehensive sex education, many challenges for school-based programs remain. Barriers to implementation include competing priorities, lack of parental and administration support, and lack of training [
Although we adhered to standards regarding evaluating behavioral interventions [
Despite these limitations and the low event rate of sexual initiation in our cohort, we demonstrated a compelling and persistent impact on attitudes and knowledge. There is evidence for a correlation between attitudes and behavior [
Serious video games as digital health interventions offer the unique opportunity to increase the accessibility and reach of theory-driven and tested interventions. The PlayForward intervention demonstrated efficacy in improving sexual attitudes and knowledge in racial/ethnic minority adolescents that persisted up to 12 months. Future research should assess PlayForward’s efficacy in populations with higher rates of sexual initiation and its comparative efficacy, effectiveness, and implementation.
PlayForward homescreen.
A Challenge Stack: 7 Minutes in Heaven.
Know Sense mini-game.
People Sense mini-game.
CONSORT-EHEALTH checklist (v1.6.1).
acquired immune deficiency syndrome
data and safety monitoring board
human immunodeficiency virus
least squares
sexually transmitted infection
We acknowledge all authors who contributed significantly to this work. We also acknowledge our game developer partners at Schell Games; our research staff and students, particularly Erica Miller, Orli Florsheim, and Christina Mezes; Gail Slap, MD, MS and Peter Salovey, PhD for their input and support; and Alex Seropian and Noah Falstein for their guidance on the games aspects of this study. We also thank our Data and Safety Monitoring Board members. We offer special thanks to the afterschool and youth programs and their staff and all the adolescents that participated and made a major contribution to this study. The study was supported by grant R01HD062080 from the Eunice Kennedy Shriver National Institute of Child Health and Human Development.
None declared.