This is an open-access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work, first published in JMIR Research Protocols, is properly cited. The complete bibliographic information, a link to the original publication on http://www.researchprotocols.org, as well as this copyright and license information must be included.
The human immunodeficiency virus (HIV) and acquired immune deficiency syndrome (AIDS) continue to be a major public health problem in Sub-Saharan Africa (SSA), particularly in Swaziland, which has the highest HIV prevalence in this region. A wide range of strategies and interventions have been used to promote behavior change, though almost all such interventions have involved mass media. Therefore, innovative behavior change strategies beyond mass media communication are urgently needed. Serious games have demonstrated effectiveness in advancing health in the developed world; however, no rigorous serious games interventions have been implemented in HIV prevention in SSA.
We plan to test whether a serious game intervention delivered on mobile phones to increase HIV risk perception, increase intention to reduce sexual partnerships, and increase intention to know own and partners HIV status will be more effective compared with current prevention efforts.
This is a two-arm randomized intervention trial. We will recruit 380 participants who meet the following eligibility criteria: 18-29 years of age, own a smartphone running an Android-based operating system, have the WhatsApp messaging app, live in Swaziland, and can adequately grant informed consent. Participants will be allocated into a smartphone interactive, educational story game, and a wait-list control group in a 1:1 allocation ratio. Subsequently, a self-administered Web-based questionnaire will be issued at baseline and after 4 weeks of exposure to the game. We hypothesize that the change in HIV risk perception between pre- and post-intervention assessment is greater in the intervention group compared with the change in the control group. Our primary hypothesis is based on the assumption that increased perceived risk of HIV provides cues to engage in protective behavior. Our primary outcome measure is HIV risk perceived mean change between pre- and post-intervention compared with the mean change in the wait-list control group at 4-weeks post-intervention. We will use standardized regression coefficients to calculate the effect of the intervention on our primary outcome with
This study is funded by Hayao Nakayama Foundation for Science & Technology and Culture; Grant number H26-A2-41. The research and development approval has been obtained from Kyoto University Graduate School and Faculty of Medicine Ethics Committee, Japan, and Swaziland’s Ministry of Health Ethics and Scientific committee. Results are expected in February 2017.
This study will provide evidence on the efficiency of a mobile phone interactive game in increasing HIV risk perception in Swaziland. Our findings may also be generalizable to similar settings in SSA.
University Hospital Medical Information Network Clinical Trial Registry ID number (UMIN-CTR):UMIN000021781; URL:https://upload.umin.ac.jp/cgi-open-bin/ctr_e/ctr_view.cgi?recptno=R000025103 (Archived by WebCite at http://www.webcitation.org/6hOphB11a).
It is estimated that 35.3 million people are living with human immunodeficiency virus (HIV) globally [
Other key drivers have been highlighted in the Extended National Multi-sectoral HIV/AIDS Framework for 2014–2018 (eNSF) as: low rates of HIV testing (only 40% of people aged 15-49 had tested for HIV 12 months preceding a household survey) [
HIV is the leading public health concern in Swaziland [
Furthermore, anecdotal information suggests that there is information fatigue from the target audience in receiving HIV prevention messages from the mass media because most prevention campaigns have been dominantly delivered through mass media. One strategy that can break this perceived fatigue is the use of target audiences’ mobile phones. In developing countries, decreasing costs and increasing mobile network coverage provide a wide range of opportunities for apps using mobile phones [
In this study, we adopt the definition of Serious Games as proposed by Alvarez and Djaouti [
Utilitarian scenario + gaming scenario => Serious Games.
Current literature suggests that serious games are effective in changing behavior. For example, a randomized trial (in the United States) designed to improve treatment adherence among 13- to 29-year-old patients with malignancies including acute leukemia, lymphoma, and soft-tissue sarcoma found that among 200 participants who were prescribed oral trimethoprim-sulfamethoxazole and 6-mercaptopurine, 16% indicted an increase in adherence for the serious games intervention group compared with the control group. Mixed-effect linear model analyses of chemotherapy metabolite concentrations showed that patients in the intervention group maintained significantly higher chemotherapy metabolite levels over time than did patients in the control group (significant group × time interaction;
The Swaziland Serious Games–Based HIV Prevention Trial (SGprev trial) will be a 4-week, two-arm randomized intervention trial. Participants will be randomized into 2 groups (the intervention group and a wait-list control group) in a 1:1 allocation ratio (
We plan to test whether a serious game intervention delivered on mobile phones to increase HIV risk perception, increases the intention to reduce multiple sexual partnerships, intention to know own HIV status, and intention to know all sexual partners’ HIV status will be more effective compared with current prevention efforts. Therefore, our hypotheses are that
1. The change in HIV risk perception between pre- and post-intervention assessment is greater in the intervention group compared with the change in the control group.
2. The change in HIV risk perception between pre- and post-intervention assessment will be greater among those reporting high HIV risk behavior in the intervention group compared with the control group.
3. The change in intention to have an HIV test between pre- and post-intervention assessment will be greater in the intervention group compared with the change in the control group.
4. The change in intention to reduce multiple concurrent partnerships between pre- and post-intervention assessment will be greater in the intervention group compared with the change in the control group.
5. The intervention group will report higher rate of condom use in the last sexual encounter compared with the control group.
CONSORT diagram for Swaziland serious games-based trial.
Our target population is Swazi males and females between 18- to 29-years old currently in Swaziland. Our intervention study targets people between 18- and 29-years old for the following reasons: (1) according to Bicego and colleagues [
For this study, we will include males and females if they meet the following criteria: (1) are between 18 and 29 years of age, (2) own a smartphone running an Android-based operating system, (3) currently have the WhatsApp messaging app, (4) currently live in Swaziland, and (5) are able to adequately grant informed consent.
To recruit participants, we will post a targeted (based on our inclusion criteria), clickable banner advertisement on Facebook. After clicking on the advertisement, potential participants will be redirected to our website. Those who meet our eligibility criteria and have granted informed consent will be sent a unique trial verification code via text message and email. This unique trial verification code will be used to take our survey. Moreover, participants eligible for this trial will be entered into a lottery draw with a 1:100 chance of receiving US $20.
The Kingdom of Swaziland, situated in Southern Africa, is a small land-locked country, the area of Swaziland is estimated to be 17,364 km2 with an estimated population of 1,146,050 (2006) [
SwaziYolo (a smartphone game) is an interactive, educational story game that puts the player in the role of a young adult looking for love in Mbabane (the capital city of Swaziland), making important choices about relationships and sexual health (see
The goal of the game is to maintain relationships with the characters, while staying healthy and happy. Once all the interactions with the characters have been completed, the game will give feedback on choices made and the risks those choices might carry. The game is expected to have an immense appeal to the youth, as an exciting new way to use their smartphones (see
Player’s curiosity to “know what happens when they make a choice” is key to user engagement. The game’s narrative is primarily concerned with matters of sexual health, especially as it relates to HIV. Players will usually find themselves in situations where they have to make important decisions about their health, for example, resisting the pressure to have unprotected sex. The game keeps track of how well a player’s relationship is going with other characters using “TRUST” ratings (intimacy ratings), and a “YOLO” rating: how safe (safe relates to making choices that do not expose player to HIV risk) they have been during the course of the game. While players enjoy game play, they are exposed to valuable learning situations and are encouraged to care more about the various characters. Some will give good advice, while others will find themselves in difficult situations where they ask other players for help and guidance.
SwaziYolo screenshots.
The trial will be powered on the primary outcome measure and based on comparison of the change in HIV risk perception score from pre- to post-intervention assessment in the intervention group at 4 weeks. A study conducted in Uganda estimated mean HIV risk perception of M=3.27, SD
Upon confirmation of participants’ trial registration, participants will be assigned another unique code called “game unlock code,” which will be used for randomization using secure, remote, Web-based computer software within 24 hours of recruitment. As stated in the research design, participants will be randomized into 2 groups; the intervention group and the wait-list control group in a 1:1 ratio. The data analysis team will be blinded in this study, however, participants will not be blinded.
A self-administered structured Web-based questionnaire was created based on review of both Swazi and international literature. For example, questions relating to sociodemographic characteristics were adopted from the 2007 Swaziland Demographic Health Survey, and those related to risky sexual behavior and intention to change behavior were developed from our formative studies. Additionally, the Perceived Risk of HIV Infection Scale (PRHS; found to have excellent internal consistency Cronbach alpha=0.88) [
Adding one or more comparison groups to a pre- and post-intervention assessment will result in a stronger intervention design than having a single intervention group to a pre- and post-intervention assessment [
The secondary outcomes for this intervention are self-reported intention to have an HIV test; intention to reduce multiple concurrent sexual partnerships; and an increase in reported condom use in the last sexual encounter (
Secondary outcome measures.
Measures | Baseline | Follow-up at 4 weeks | |
Condom use in the last sex | Xa | X | |
Number of sexual partners | X | X | |
Intention to test for HIV and know partners’ HIV status | X | X | |
Intention to reduce multiple concurrent partnerships | X | X | |
Intention to use a condom | X | X | |
Has current sexual partner ever tested for HIV? | X | X | |
Knowledge of current partner’s HIV status | X | X | |
Age | X | ||
Level of education | X | ||
Employment status | X | ||
Income level | X | ||
Marital status | X | ||
Ever tested for HIV | X | X | |
Cellphone number and email | X | ||
Would you recommend the game to your friends? | X | ||
How did you hear about this game? | X | ||
Number of times player reached the end of the game | X | ||
Level of satisfaction about the game | X |
aTiming of assessment.
Participants will be recruited from Facebook users in Swaziland via a targeted Facebook advertisement. Potential participants will be directed to the study Web page where information about the intervention trial will be given and if willing, screening for eligibility will be done. After screening for eligibility, eligible individuals interested in participating in the trial will have an opportunity to ask detailed questions via free text message service offered by the WhatsApp app, Facebook messenger, or calling us. Sufficient time will be allowed for making an informed decision about participation in the study. Recruitment into this study will continue until our sample size is achieved.
After informed consent, a trial verification code will be sent to the participants via their mobile phones to prevent multiple identities in line with the CONSORT-EHEALTH guidelines 4a(ii) [
Trial participants will be followed-up for 4 weeks, the game will collect log data and send this data to our servers when the participant goes on the Internet, this will allow us to assess the exposure to the intervention without over burdening our participants to manually send us their usage data. Data captured will be limited to login data. In addition to this, a Web-based questionnaire will be sent to participants at the end of the follow-up period. Participants who will miss their 4-week follow-up assessment will be actively traced though phone calls and text messages.
First, Web-based questionnaires must be usable even for less experienced and knowledgeable Internet users [
Initially, descriptive statistics for the sample characteristics will be done for the intervention group and the wait-list control group to assess the distribution of important predictors of the outcome between both groups at baseline.
First, we will use bivariate analysis to calculate the mean between baselines and follow-up. Next, to estimate the difference between the 2 groups, we will calculate the difference between the mean change of the intervention group and the mean change of the wait-list control group using two-sample paired
Secondly, in the case that, even after randomization, we observe some baseline differences, we will use multiple linear regression to adjust for those differences; where the outcome will be the follow-up score and the independent variables will be the intervention group, baseline scores, age, gender, marital status, level of education employment status, current monthly income, and number of times players played SwaziYolo. We will present our results in standardized regression coefficients for the intervention effect on the outcome variable as previously done for this type of hypothesis [
Although great effort will be put to minimize attrition, it is common for eHealth trials to typically have substantial attrition [
Two-sample generalization McNemar’s test will be done to assess whether a significant change occurred between the pre- and the post-intervention assessments for dichotomous variables such as: intention to know self and partners HIV status, intention to reduce multiple sexual partners, and intention to use a condom the next time a participant has sexual intercourse. Each of these, outcomes will be assessed separately (individually). In order to judge the change, we will calculate the proportions of the dichotomous variable pre- and post-intervention in both groups. After that, we will obtain the pre- and post-intervention difference percentage at a
All participants will be required to give Web-based informed consent (
The following measures will be taken to protect participant’s personal information:
1. Permission will be sought from study participants to collect game usage data (login data) automatically.
2. All participants’ data will be stored under encrypted servers to protect participants’ information.
3. Participants’ cellphone numbers will be stored in a password protected file and will not be used for purposes other than those outlined in this protocol. After the trial, all cellphone numbers will be deleted.
During or after the study, participants may develop psychological distress or embarrassment. All efforts to prevent this psychological distress or embarrassment will be put in place. If despite our efforts any psychological issue arises during the intervention and data collection, the research team will refer the participants’ to the nearest counselor (who is well vest on psychological issues) for appropriate psychological care and support via text messaging or calling. Participants will be encouraged to self-report any feelings of distress or discomfort to the research team using Web-based tools such as the WhatsApp app, Facebook private message, or via our contact details provided in the study information sheet including a toll-free number for HIV counseling.
All data will be stored in the password-encrypted servers. Upon completion of the survey, all data will be exported to a password protected desktop computer at Kyoto University Department of Global Health and Socio-Epidemiology. Persons not part of the research team will not have any access to the collected data.
A lottery draw at baseline with a 1:100 chance of receiving US $20 will be given to all participants at the end of data collection as an incentive for their time in taking part in this trial. This amount was chosen carefully not to cause undue influence to the target population in that it is not excessive and is fair considering the country’s socioeconomic status.
The status of the study is in preinitiation stage. Results are expected in February 2017. We will present results as percentages, observed means with 95% confidence intervals, mean difference and 95% confidence intervals, standardized regression coefficients, and
The study will be conducted according to the principles outlined in the Declaration of Helsinki International Guidelines for Ethical Review of Epidemiological Studies (CIOMS, 1991 Geneva). Furthermore, the research and development approval has been obtained from Kyoto University Graduate School and Faculty of Medicine Ethics Committee, Japan, and Swaziland’s Ministry of Health Ethics and Scientific committee. Caution will be taken to protect participant’s privacy during the data collection, data handling, and data reporting.
Development of “SwaziYolo” serious game was funded by Hayao Nakayama Foundation for Science & Technology and Culture; Grant number H26-A2-41. The research implementation will be sponsored by the Department of Global Health and Socio-Epidemiology, Kyoto University, Japan.
The risk of HIV infection is high among young people who practice risky sexual behavior, often they do not perceive their risk to be high, a phenomenon termed optimistic bias [
In line with the guidance form UNAIDS, this trial will provide a robust and rigorous evaluation of the efficacy of mobile serious games in increasing HIV risk perception in a resource limited setting such as Swaziland. Findings from this study will be made available to Swaziland authorities and stakeholders working to improve HIV prevention in Swaziland. We envision that the results of this study will be highly relevant to HIV prevention interventions in Swaziland and will inform future innovative strategies for HIV prevention. We are hopeful that our results will be generalizable to other settings in SSA. To our knowledge this is the first randomized control trial of a mobile serious games–based study to increase HIV protective behaviors in Swaziland and SSA; therefore, our findings will be a timely contribution to literature.
Intervention development.
Pre- and post-intervention questionnaire for the SGpriv trial.
Informed consent form.
Trial information sheet.
acquired immune deficiency syndrome
Extended National Multi-sectoral HIV/AIDS Framework for 2014–2018
human immunodeficiency virus
mean
perceived risk of HIV infection scale
Sub-Saharan Africa
United Nations Programme on HIV/AIDS
We would like to acknowledge Formula D Interactive for development of the game and Net Storm Technologies for the website development. Further, we thank Mr Norihisa Wada for his advice and support.
MK, MOK and BWL led study conception, intervention design, and methodology, statistical design, programing platform, and intervention user interface design, and will lead study implementation. CE participated in the study design and intervention development. PPM participated in the study conception, statistical design, and intervention development. TT and SPS participated in the study conception, design and statistic design, and intervention development. BWL will establish recruitment and randomization of participants. BWL and MK drafted the protocol, and all authors read and edited drafts of the protocol and approved the final protocol manuscript.
The authors declare that they have no competing interests. Though as stated earlier, BWL received funding for intervention development from Hayao Nakayama Foundation for Science & Technology and Culture, our funding agreement gives us full control over primary data, statistical analysis, and the freedom to publish findings whether negative or positive, as is standard precaution to ensure potential competing interests are kept in check [