The Effects of a Computer-Based Pregnancy, STD, and HIV Prevention Intervention: A Nine-School Trial

Abstract

A computer-based intervention was designed to change perceived threat, perceived efficacy, attitudes, and knowledge regarding pregnancy, STD, and HIV prevention in rural adolescents. The intervention, which was guided largely by the extended parallel process model (Witte, 1992), was implemented and evaluated in nine rural high schools using an institutional cycle pretest–posttest control-group design (Campbell & Stanley, 1963; Cook & Campbell, 1979). Eight-hundred eighty-seven ninth-graders completed the survey at both points in time. Process evaluation results indicated that the intervention was implemented as intended, and that over 91% of students in the treatment group completed at least one of the six computer-based activities (M = 3.46, SD = 1.44 for those doing at least one activity). Two-way mixed-model repeated-measures analysis of variance revealed that students in the treatment group outperformed students in the control group on knowledge, condom self-efficacy, attitude toward waiting to have sex, and perceived susceptibility to HIV. These results suggest that computer-based programs may be a cost-effective and easily replicable means of providing teens with basic information and skills necessary to prevent pregnancy, STDs, and HIV.

ACKNOWLEDGMENTS

The research reported here was funded by National Institute of Mental Health Grant R01 MH16876 awarded to the University of Kentucky, Rick S. Zimmerman, principal investigator. The authors thank the students, parents, teachers, and administrators at the schools where data collection took place. We also thank the University of Kentucky Department of Communication's Research Technology staff (Scott Johnson, Carrie Johnson, Shea McCombs, Brian Raney, and Derek Rockhill) for their technical support throughout this project. Finally, we thank Lisa Murray-Johnson, Kim Witte, and both anonymous reviewers for their valuable feedback. An earlier version of this manuscript received a Top-Three Paper Award from the Health Communication Division of the International Communication Association in May 2005.

Notes

¹This study was part of a much larger 28-school five-year evaluation of an adaptation of the Reducing the Risk curriculum (Barth, 1996) to rural populations. Although Reducing the Risk (and by extension this computer-based intervention) focused to some extent on perceived threat, it is primarily a skills-based intervention meant to increase perceived efficacy regarding pregnancy, STD, and HIV prevention. Although considerable time and space were devoted to measuring the efficacy-related variables on all surveys, the surveys contained only one severity item (regarding pregnancy) and three susceptibility items (one each regarding pregnancy, STDs, and HIV). Thus, results for all perceived threat variables should be interpreted with caution.

²Because of the modest nature of the intervention and the relatively short-term nature of this study (i.e., 3 months between the end of the intervention and the posttest), a priori hypotheses regarding behavior change did not seem warranted. However, the survey did contain behavior measures regarding initiation of sexual activity and condom use at last intercourse. The initiation of sexual activity item asked: “Have you ever had sexual intercourse (‘going all the way’ or vaginal sex)?” The condom use at last intercourse item asked “The last time you had sexual intercourse did you (or your partner) use a condom?” The response categories in both cases were “yes” and “no.” Logistic regression was used to assess the impact of the intervention on the dichotomous behavioral measure. In both cases, the analyses were run controlling for time between the pretest and the posttest. Further, for the initiation of sexual activity item, analysis was conducted using only individuals who were not sexually active at Time 1. For condom use at last intercourse, analysis was conducted using only individuals who were sexually active in the last six months. Results indicate there were no significant differences between the treatment and control group on either behavioral measure.

Roberto, A. J., Zimmerman, R. S., Cupp, P. K., Hansen, G. L., Banerjee, B., Mazloomdoost, Y., (2003, July). Developing a small-media campaign to prevent HIV and pregnancy in rural adolescent populations: Formative evaluation insights and implications. Presentation at the National HIV Prevention Conference, Atlanta, GA.