20th International AIDS Conference - Melbourne, Australia


MOAC0103 - Oral Abstract

A videogame increases HIV risk-related knowledge in adolescents

Presented by Lynn Fiellin (United States).

L. Fiellin1,2, K. Hieftje1,2, T. Fakhouri1, L. Duncan1,3, T. Kyriakides2

1Yale play2PREVENT Lab, New Haven, United States, 2Yale University, New Haven, United States, 3McGill University, Montreal, Canada

Background: Nearly 40% and 45% of new HIV infections in the U.S. and globally, respectively, are in adolescents/young adults. Videogames are ubiquitous, foster skill development translating into improved health outcomes, and have the potential to dramatically reduce risk behaviors in youth. Behavior change videogames collect in-game data about the players'' knowledge and skills, how they change in response to gameplay, and how they correlate with out-of-game data collected. PlayForward: Elm City Stories is an iPad-based videogame designed to increase HIV risk knowledge and decrease risk behaviors. The purpose of this study was to assess:
(a) if PlayForward impacted on HIV risk knowledge and
(b) how in-game knowledge data correlated with data collected through standardized assessments.
Methods: As part of a large RCT to evaluate PlayForward, HIV risk-related knowledge data was collected through standardized assessments at baseline, 6 weeks (following gameplay), and 3 months. Software-generated logs of player activity were collected through the iPad, measuring exposure to specific intervention components. Coupled with standardized assessments, the data were analyzed for markers of gains in HIV risk-related knowledge. iPad software data were analyzed using the R statistical computing software package.
Results: One hundred and ninety-eight adolescents have been enrolled in the RCT; 55% boys, mean age is 13 years; 161 have completed 6 weeks of gameplay; 125 have completed 3-month follow-up assessments. There were no significant baseline between-groups differences on a 22-item assessment of HIV risk-related knowledge. After six weeks of gameplay, the intervention group had higher knowledge scores (mean=15 (S.D.=4.8)) than the control group (mean=12.5 (S.D.=4.5); p< 0.001) at six weeks and at three months ( mean=14.4, S.D.=5.5 vs. mean=12.5, S.D.=4.7; p=0.04). Analysis of 1,289,903 events in log files revealed that the number of game levels completed was positively correlated with knowledge gains measured at 6 weeks (r=0.32; p< 0.005) and at three months (r=0.42; p< 0.001).
Conclusions: These findings demonstrate that
(a) PlayForward increases HIV risk-related knowledge among adolescents and
(b) exposure to videogame content is highly correlated with knowledge as assessed by standardized tools used outside of gameplay.
These methods are a new and exciting way to capture evidence for real-world knowledge acquisition.

Back to the Programme-at-a-Glance