In 2004, the University of Colorado at Denver developed an Internet-based intervention method, the Youthnet Trials, for teens and young adults (ages 18-24) who are at risk for contracting and transmitting HIV and other STDs. Participants were either enrolled online or at clinics in the Denver metropolitan area. From each of the online and clinic groups, participants were placed in either intervention or control groups (RCT). The goal is to use online-based modules to promote protected sex and prevent HIV transmission among teens.

Enrolled participants in the intervention group took part in a single online session composed of five modules, in which they answered questions about HIV risks and watched clips of role model stories that included visuals and audio. Role models were matched to participants based on gender and race/ethnicity. Modules were 60-90 seconds long, and addressed attitudes, norms, HIV and STD risk, and self-efficacy regarding condom negotiation and usage. All answered follow-up questions at the end of the modules. Those in the control group did not watch clips of role model stories, but instead read generic text-based HIV information. Participants were given up to $35 to remain in the study.

The goal of the progarm is to use online-based modules to promote protected sex and prevent HIV transmission among teens.

A randomized controlled trial was used to evaluate this study. From each of the online and clinic groups, participants were placed in intervention or control groups. The Internet group consisted of 514 people in the intervention group and 447 people in the control group. In the clinic group, 285 people were placed in the intervention group and 289 people in the control group. In total, 1565 people participated in the study.
The investigators designed two different statistical and theoretical models that were able to explain 42% and 59% of the variance in the proportion of protected sexual acts between baseline and follow-up in the Internet and clinic samples, respectively. However, the impact of the intervention on the variance was not significant.

The intervention in the Internet sample had a very small effect (with a standardized path coefficient of 0.06) on the condom norms construct. This construct was also the only factor with a direct and significant impact on the proportion of protected sexual acts, with a path coefficient of 0.42. There were no indirect effects on the study outcome, since the path coefficient between self-efficacy for condom use and the study outcome was not significant.

In the clinic sample, the intervention had no significant effects on the study outcome. Those in the intervention were significantly less likely to feel confident they could use condoms with their partner at follow-up, than those in the control (R2= -.10). However, the authors note that self-efficacy did not have an impact on the study outcome. As in the Internet sample, only the norms construct had a direct effect on the study outcome in the clinic sample.