Objectives: To conduct a meta-analysis of computer technology-based HIV prevention behavioral interventions aimed at increasing condom use among a variety of at-risk populations.
Design: Systematic review and meta-analysis of existing published and unpublished studies testing computer-based interventions.
Methods: Meta-analytic techniques were used to compute and aggregate effect sizes for 12 randomized controlled trials that met inclusion criteria. Variables that had the potential to moderate intervention efficacy were also tested.
Results: The overall mean weighted effect size for condom use was d = 0.259 (95% confidence interval = 0.201, 0.317; Z = 8.74, P < 0.001; N = 4639), indicating a statistically significant impact of the interventions. This effect size compares favorably to previously tested interventions delivered by human facilitators. Statistically significant effect sizes were also found for frequency of sexual behavior, number of partners, and incident sexually transmitted diseases. In addition, interventions were significantly more efficacious when they were directed at men or women (versus mixed sex groups), utilized individualized tailoring, used a Stages of Change model, and had more intervention sessions.
Conclusion: Computer technology-based HIV prevention interventions have similar efficacy to more traditional human-delivered interventions. Given their low cost to deliver, ability to customize intervention content, and flexible dissemination channels, they hold much promise for the future of HIV prevention.
aDepartment of Communication, USA
bDepartment of Marketing, USA
cDepartment of Educational and Counseling Psychology, University of Kentucky, Kentucky, USA.
Received 11 June, 2008
Revised 15 September, 2008
Accepted 29 September, 2008
Correspondence to Seth M. Noar, PhD, Associate Professor, Department of Communication, University of Kentucky, 248 Grehan Building, Lexington, KY 40506-0042, USA. Tel: +1 859 257 7809; fax: +1 859 257 4103; e-mail: email@example.com