This project, funded by the National Institute of Mental Health, will test the efficacy of an electronic game to prevent HIV among African adolescents (aged 12-17), delivered via inexpensive Android smartphones. This study involves a sample of 912 young people and 500 of their parents in Kenya's former Nyanza province, where 11.4% of young women and 3% of men ages 15-24 are HIV-infected. This study will be carried out by Emory University and the Kenya Medical Research Institute (KEMRI).
The 912 study participants (youth ages 12-14) will be randomized to the intervention or control arm upon completion of the baseline survey. The assignments will be generated with the use of a pseudo-random-number generator with permuted blocks, used to ensure a sex-balanced and age-balanced assignment to each arm. All participants will be given a low-cost (\<$40) Android phone on which the study game will be loaded. Intervention arm phones will have app for Tumaini programmed on them, while control arm phones will have a commercially available game to be determined prior to the start of the randomized clinical trial. Preference will be given to a game or other app with educational content, for example, a general knowledge quiz or strategy game. Each app will be standalone and not require data access or connection to the Internet to function. All participants will be invited to play their assigned game for at least 10 hours during the November-December school holiday. While we anticipate that access to smartphones will be widespread at time of roll-out of an efficacious game, this is not yet the case: for the purposes of this study, phones need to be provided in order to ensure consistency of technology and avoid SES bias. All other phone functions aside from the alarm, will be disabled for safety reasons. The game will automatically record the time spent playing and the choices made in the context of gameplay. When returning the phones, all participants will be asked to fill out a form identifying others with whom the game or app was shared and creators of each profile on their study phones. In line with the preferences expressed by parents during the feasibility test, the phones will be provided to participants during the long November-December school holidays and collected at the end of the holiday. Providing there is no evidence of contamination, participants will (a) keep the phone throughout the long November-December school holiday; and (b) receive the phone, loaded with the game, again during the November-December holidays of the following years.
Study Type
INTERVENTIONAL
Allocation
RANDOMIZED
Purpose
PREVENTION
Masking
NONE
Enrollment
1,000
Tumaini is a scenario-based role-playing game application. Participants will be instructed to engage in a minimum of 10 hours of gameplay over the holiday period each year. The game will automatically collect for analysis data related to participants' in-game behavior, e.g. time spent playing, scores on knowledge-based mini-games, choices made in the narrative game, components to which the player was exposed. The game is designed to: educate players about sexual health and HIV/AIDS; build risk-reduction skills and related self-efficacy for prevention of HIV/STIs and unintended pregnancy; challenge HIV stigma; and promote parent-child dialogue.
Independent gameplay of control game. Total respondent burden: participants will be invited to engage in at least 10 hours of gameplay during each long school holiday (November-December).
Kenya Medical Research Institute
Kisumu, Kenya
Age at first sex
Age at first sex encounter.
Time frame: Up to month 46
Percentage of participants reporting condom use at first sex
Percentage of participants reporting condom use at first sex
Time frame: Up to month 46
Binary Risk outcome of "risk" vs "low risk"
Binary outcome of "risk" group (experienced sexual debut during the study period without condom use at first sex) vs "low risk" group (not yet experienced sexual debut by the end of the study, or used a condom at sexual debut during the study period).
Time frame: Up to month 46
Change in percentage of participants reporting recent sexual activity
For participants reporting previous sexual intercourse, self-report of sexual intercourse since previous behavioral survey.
Time frame: Months 1, 3, 6, 9, 12, 17, 22, 25, 30, 34, 37, 42, 46
Change in percentage of participants reporting recent condom use
For participants reporting recent sexual activity, self-report of consistent condom use since previous behavioral survey.
Time frame: Months 1, 3, 6, 9, 12, 17, 22, 25, 30, 34, 37, 42, 46
Change in number of sexual partners
For participants reporting any past sexual activity, number of different sexual partners
Time frame: Months 1, 3, 6, 9, 12, 17, 22, 25, 30, 34, 37, 42, 46
Change in percentage of participants with recent alcohol and drug use
Percentage of participants that report recent alcohol and drug use.
Time frame: Months 1, 3, 6, 9, 12, 17, 22, 25, 30, 34, 37, 42, 46
This platform is for informational purposes only and does not constitute medical advice. Always consult a qualified healthcare professional.
Change in knowledge
Knowledge measures will be assessed via Yes/No survey questions and will address HIV, sexually transmitted infections (STIs), pregnancy, and condoms. Answers to knowledge questions will be coded as 0 or 1, with higher scores indicating more accurate knowledge. A positive value for the change from baseline score indicate an increase in knowledge from the baseline assessment.
Time frame: Months 1, 3, 6, 9, 12, 17, 22, 25, 30, 34, 37, 42, 46
Change in self-efficacy
Self-efficacy will be assessed via a Likert scale, and focus on self-efficacy to use a condom correctly; to communicate with a partner about protected sex; to reject peer, partner, and adult pressure to engage in risk behaviors; and, for female participants, to manage menstruation. Self-efficacy will be scored on a 0-1 scale with higher scores indicating stronger self-efficacy. A positive value for the change from baseline score will indicate an increase in self-efficacy since the baseline assessment.
Time frame: Months 1, 3, 6, 9, 12, 17, 22, 25, 30, 34, 37, 42, 46
Change in risk assessment
Risk assessment will be assessed via a Likert scale, and address perceived risk of certain risk situations/behaviors. Risk items will be scored on a 0-1 scale, with higher scores indicating higher risk assessment. A positive value for the change from baseline score will indicate an increase in assessing risky situations as risky.
Time frame: Months 1, 3, 6, 9, 12, 17, 22, 25, 30, 34, 37, 42, 46
Change in attitudes
Attitudes will be assessed via Likert survey questions relating to puberty, HIV stigma, condoms, sex, and gender. Attitudes items will be scored on a 0-1 scale, with higher scores indicating more desirable attitudes. A positive value for the change from baseline score will indicate an increase in desirable attitudes.
Time frame: Months 1, 3, 6, 9, 12, 17, 22, 25, 30, 34, 37, 42, 46
Change in behavioral intentions
Behavioral intention will be assessed via Likert survey questions. The measure will address intention to avoid risk situations and to engage in health protective behaviors. Behavioral intention items will be scored on a 0-1 scale with higher scores indicating more intention to partake in health protective behaviors and a positive value for the change from baseline score indicating an increase in intention to engage in health protective behaviors.
Time frame: Months 1, 3, 6, 9, 12, 17, 22, 25, 30, 34, 37, 42, 46
Total length of gameplay
Total time playing from game log files
Time frame: Months 3,18, 28
Number of participants with new HSV-2 diagnosis
Incidence rates over the course of study of HSV-2 among uninfected participants will be compared by exposure to intervention using Poisson and/or Cox regression methods.
Time frame: Months 1 and 46
Number of participants with new HIV diagnosis
Incidence rates over the course of study of HIV among uninfected participants will be compared by exposure to intervention using Poisson and/or Cox regression methods.
Time frame: Months 1 and 46