on the basis of changes in self-reported risky sexual behaviors such as frequency of unprotected sexual intercourse.
The overwhelming proportion of peer-reviewed evaluations of the effectiveness of sex and AIDS and STD education programs have reported positive changes or no effect of programs on sexual health behaviors such as condom use and onset of sexual intercourse (Kirby et al., 1994). In a few studies based on national survey data, mixed outcomes (positive, negative, or no effect) have also been documented (Dawson, 1986; Marsiglio and Mott, 1986; Ku et al., 1993). Kirby and colleagues recently completed a comprehensive review of 23 studies of school-based sex and AIDS and STD education programs that were published in peer-reviewed journals and evaluated their impact on sexual behavior (Kirby et al., 1994) The authors found that some but not all programs were effective and that programs having the following six characteristics had a clear impact on behavior:
The authors concluded that, contrary to the concerns of some individuals and groups, such educational programs do not increase sexual activity among students. Studies of specific programs found that programs that included instruction on contraception either delayed the onset of sexual intercourse or had no effect on onset. In addition, a previous IOM committee (IOM, 1995:233) evaluated 23 local programs, including school-based programs, related to unintended pregnancy and concluded that:
Sexuality education programs that provide information on both abstinence and contraceptive use neither encourage the onset of sexual intercourse nor increase the frequency of intercourse among adolescents. In fact, programs that provide both messages appear to be effective in delaying the onset of sexual intercourse and encouraging contraceptive use once sexual activity has begun, especially among younger adolescents.
A recent nationwide survey revealed that respondents wanted information on STDs from multiple sources and rated health care professionals and school-based