In a randomized trial examining two middle school sex education programs, a risk reduction intervention that emphasized delaying sexual initiation and promoted condom use reduced the likelihood that students had sex during follow-up, while a risk avoidance program that advised abstinence until marriage was effective only among certain subgroups.1 Compared with a control group that received regular health education, youth assigned to the risk reduction program were less likely to report in ninth grade that they had begun having vaginal, oral or anal sex (odds ratio, 0.7) or had had unprotected vaginal sex in the past three months (0.6). Neither benefit was apparent for the full sample of students in the risk avoidance program, although initiation of some behaviors was reduced among Hispanic youth and females. Youth in both intervention groups were less likely than those in the control group to have had unprotected sex at last intercourse (0.7 for both).
The study, conducted in 2006–2010 in 15 randomly selected middle schools in a large Southern city, focused on two theory-based interventions. The risk reduction program, called It’s Your Game … Keep It Real, had been shown in a previous trial to be effective; it emphasized delaying sexual initiation (though not necessarily until marriage) and provided skills training in condom use (including tailored material for sexually experienced youth). The other program was modeled on It’s Your Game, but took a risk avoidance approach; it emphasized abstaining from sex until marriage and promoted character development. Both programs consisted of 24 lessons and included group classroom activities, interactive computer activities (e.g., skills-training exercises) and homework assignments designed to foster parent-child communication. Half of each program was presented in seventh grade, and half in eighth. Five schools were assigned to the risk reduction program, and five to the risk avoidance intervention; students in the remaining schools, who served as a control group, had their regular health classes. About half of eligible students returned a parental consent form allowing them to participate in the trial.
Using laptop computers, participants completed a baseline survey at the beginning of seventh grade and follow-up surveys in eighth and ninth grades (the last was done 26 months after baseline). For students who were sexually inexperienced at study entry, the primary outcome measure was sexual initiation, defined as having had oral, vaginal or anal sex for the first time; other behavioral outcomes included measures of unprotected sex and number of partners. The researchers also examined a range of psychosocial measures concerning participants’ knowledge, beliefs and intentions. They assessed relationships between program assignment and behavioral outcomes using generalized linear models that controlled for social, demographic and psychosocial measures, as well as sexual experience at baseline (students in the control group were more likely than those in the intervention groups to have had sex at study entry). Sexually experienced youth were excluded from analyses of sexual initiation, but were included in other analyses.
About a fourth of the 1,742 students who completed the baseline survey did not complete the ninth-grade follow-up survey or had missing or inconsistent responses, yielding an analytic sample of 1,258 youth. Most participants were Hispanic (48%) or black (39%), and three-fifths were female. On average, youth in the two intervention groups attended about two-thirds of the 24 lessons.
Compared with their counterparts in the control group, sexually inexperienced youth who participated in the risk reduction program had 35% lower odds of initiating any type of sex by ninth grade (odds ratio, 0.7). In subgroup analyses, the odds of initiation were reduced among females and blacks (0.4 each), but not among males and Hispanics. In contrast, youth assigned to the abstinence-based risk avoidance program were generally no less likely than controls to initiate sex; the only subgroup for whom the odds were reduced was Hispanic youth (0.4).
Outcomes for specific types of sex were also more positive among youth in the risk reduction group than among those in the risk avoidance group. Youth in the risk reduction program were less likely than controls to have had vaginal sex by ninth grade (odds ratio, 0.6); in subgroup analyses, the odds of such initiation were reduced among blacks (0.3) and females (0.5). Moreover, young women in this group were less likely than females in the control group to initiate oral sex (0.4). Overall, youth in the risk avoidance group did not differ from controls in initiation of vaginal or oral sex; the odds of initiating oral sex were reduced among females in this group (0.6), but were elevated among males (1.7). In addition, Hispanic youth in the risk avoidance program had reduced odds of initiating vaginal sex (0.4).
Both interventions were associated with a reduced likelihood that respondents had not used a condom at last vaginal intercourse (odds ratio, 0.7 each). However, the odds of having had unprotected vaginal sex in the past three months were reduced only among participants in the risk reduction group (0.6). Youth in the risk avoidance program were more likely than controls to have had at least two vaginal sex partners in the past three months (1.7).
Finally, both programs had positive effects on psychosocial outcomes at the final follow-up. Compared with their peers in the control group, youth in the two intervention groups reported greater self-efficacy for condom use, had more positive beliefs about abstinence, could cite more reasons for not having sex and had stronger intentions to use condoms in the next three months. In addition, relative to controls, youth in the risk reduction program were more knowledgeable about condoms and had stronger intentions to remain abstinent through high school and to get tested for STDs if at risk. Youth in the risk avoidance group, meanwhile, had greater knowledge of STDs and rated their parents’ and friends’ beliefs about waiting to have sex more positively.
Limitations of the study, according to the authors, included its reliance on self-reported data and its exclusion of youth who did not obtain parental consent. Nonetheless, the findings suggest that a risk reduction intervention that discusses abstinence but also provides information and training on condom use can delay middle school students’ sexual initiation until at least ninth grade. Although the risk avoidance intervention delayed sexual initiation among some subgroups, such benefits were not apparent among the full sample, and the investigators note that the elevated risk of oral sex initiation among males is "worrying" and warrants "further investigation to understand how males and sexually experienced youth process risk avoidance messages to better tailor activities to their needs."—P. Doskoch