A trial comparing several sex education approaches found a theory-based abstinence-only intervention to be the only one associated with delayed sexual initiation among urban black adolescents.1 The intervention differed from many abstinence programs that until recently received federal funding in that it “did not contain inaccurate information, portray sex in a negative light, or use a moralistic tone.” Participants were less likely than those in a control intervention to initiate sex during the 24-month follow-up period and to have had sex in the three months prior to the final follow-up (risk ratios, 0.7 and 0.9, respectively). While the abstinence-only intervention had no effects on other -outcomes, including condom use, two comprehensive interventions were associated with a reduced risk that participants had had multiple sex partners in the three months prior to the last follow-up (0.95–0.96).

Between September 2001 and March 2002, the researchers recruited sixth- and seventh-grade students from four urban public middle schools in the northeastern United States. Participants were randomly assigned to one of five interventions. The eight-hour abstinence-only intervention focused on increasing knowledge of HIV and other STDs, promoting abstinence as a way to prevent pregnancy and STDs and foster the attainment of future goals, and increasing skills related to abstaining from sex. An eight-hour safer sex–only intervention focused on increasing knowledge of STDs, including HIV, and enhancing beliefs and increasing skills promoting condom use. Two comprehensive interventions—one eight and the other 12 hours long—covered abstinence, safer sex and HIV risk reduction. Lastly, an eight-hour general health promotion intervention served as a control. Each program was based on social cognitive theory, the theory of reasoned action and the theory of planned behavior, and was -implemented in four-hour sessions. In addition to the interventions, some students were randomly selected to participate in intervention--maintenance programs. These adolescents received two additional three-hour sessions, six issues of a newsletter and six short one-on-one counseling sessions with the facilitator who led their original intervention.

In follow-up questionnaires administered three, six, 12, 18 and 24 months after the intervention, students reported whether they had ever had sexual intercourse and whether, in the previous three months, they had had intercourse, had sex with multiple partners, had unprotected sex and used condoms every time they had sex.

Some 662 students participated in the trial, the majority of whom were female (54%). Participants’ mean age was 12.0, and one-third lived with both parents. Some 23% had ever had sex, and in the previous three months, 12% had had sex and 6% had had multiple partners. Of those who had had sex in the three months prior to the final follow-up, 67% reported consistent condom use. Attendance was high at both the intervention sessions (96-–100%) and the maintenance sessions (84–96%).

At the 24-month follow-up, 33% of students who had participated in the abstinence-only intervention reported having initiated sex since baseline, as did 41–42% of those in the comprehensive interventions, 52% of those in the safer-sex intervention and 47% of controls. The proportion of students who reported having had sex in the three months prior to the final follow-up ranged from 30% among those in the abstinence-only group to 40% among those in the safer-sex intervention. Eleven percent of students who had taken part in the eight-hour comprehensive program and 13–16% of others reported having had multiple sex partners in the past three months; 1–6% reported having had unprotected sex. Among participants who had had sex in those three months, the proportion who reported consistent condom use ranged from 29% in the safer-sex group to 93% in the abstinence-only group.

In a generalized linear regression analysis, only students who had participated in the abstinence-only intervention were significantly less likely than controls to have initiated sexual intercourse during the next 24 months and to have had sex in the three months before the final follow-up (risk ratios, 0.7 and 0.9, respectively). Participants’ probability of having had first sex by the 24-month follow-up was 34% among the abstinence-only group and 49% among controls. The probabilities that participants in these groups had had sex in the three months prior to follow-up, averaged over follow-up surveys, were 21% and 29%, respectively. The risk of having had multiple sex partners in the three months prior to the last follow-up was reduced only among students in the two comprehensive interventions (0.95–0.96). The probability that students had had multiple sex partners in that time period was 9% in each of these groups, compared with 14% among controls.

Students from the abstinence-only and 12-hour comprehensive intervention groups who participated in the intervention-maintenance program were less likely to have had multiple sex partners than were participants from those groups who did not participate in the maintenance program (risk ratio, 0.9 for each).

The researchers note that this is the first randomized controlled trial to establish a link between an abstinence-only sex education intervention and sustained levels of reduced sexual initiation among adolescents. They caution that the association between participation in the abstinence-only intervention and delayed sexual initiation does not imply that all abstinence-only sex education programs are effective, or that this intervention would necessarily be effective among a different population of young people. Yet, they posit that an effective, theory-based, appropriately targeted abstinence-only intervention is valuable, because “abstinence is the only approach that is acceptable in some communities and settings.” The researchers add that nonetheless, “tackling the problem of [STDs] among young people requires an array of approaches.”

—H. Ball


1. Jemmott JB III, Jemmott LS and Fong GT, Efficacy of a theory-based abstinence-only intervention over 24 months, Archives of Pediatrics and Adolescent Medicine, 2010, 164(2):152–159.