Making condoms available in schools

There are three main controversial approaches to reducing rates of sexually transmitted diseases and unintended pregnancy among North American teenagers: abstinence-only programs, safer sex education, and making condoms available in schools. Which of these is effective? The American Medical Association (AMA) Council on Scientific Affairs recently concluded that “there are no published studies that measure behavioral effects of the `abstinence-only' curricula,” that “evaluations of safer-sex sexuality education show inconsistent but promising results,” and that programs that make condoms available in schools “usually demonstrate increased condom use.”¹

United States Representative Tom Coburn attacked that conclusion, claiming that the distribution of school condoms conflicts with “common sense” and increases condom use but does not necessarily lead to lower rates of either pregnancy or sexually transmitted diseases. The implementation of abstinence education in schools, he maintained, was followed by lower rates of teen pregnancies out of wedlock.²

The AMA report is a carefully conducted and thoughtful review of the evaluations of all these programs. However, some abstinence-only programs may delay sex; there is strong evidence that some safer sex programs reduce unprotected sex; and the evidence that the availability of condoms increases their use is not strong.

In regard to abstinence-only programs, the AMA Council on Scientific Affairs correctly states that many youths are sexually experienced and need the knowledge, motivation, skills, and access to condoms and contraceptives to avoid sexually transmitted diseases and unintended pregnancies. It finds no good evidence that any abstinence-only programs actually delay the onset of sexual intercourse. All of these facts support the position that abstinence-only programs should not be widely implemented in schools in place of programs that cover both abstinence and contraception.

On the other hand, in fairness to abstinence-only programs, only one study has rigorously evaluated an abstinence-only program.³ That study, conducted in California, was designed to detect small changes in delay in the onset of sexual intercourse; it randomly assigned 7,753 youths in middle school to either treatment or control conditions and tracked them for 17 months. It found no significant difference in the initiation of sex or other sexual behaviors.

It is unfair to judge the results of the diverse range of abstinence-only programs on this one study. Some of these programs, especially those that are more comprehensive and those with qualities found in effective safer sex programs, may delay the onset of intercourse, and rigorous research should be completed to find out which are effective. Until then, we should remain cautious about broadly replicating any programs that have not been shown to be effective.

While Coburn's claim that the emergence of abstinence education in recent years “coincides with the first recorded reduction in sexual activity and out-of-wedlock pregnancy among adolescents” is supported by some survey data, the decrease in pregnancy is also due in small part to a stabilization or fall in the percentage of youths who are sexually active and in large part to an increase in the use of contraception among those who are sexually active. These trends support the position that programs covering both abstinence and contraception should be broadly implemented.

Most safer sex programs do cover both abstinence and the use of condoms, and some address other forms of contraception. The AMA review properly notes that some studies of safer sex programs show positive effects on behavior—such as delaying sex, reducing the frequency of sex, and increasing the use of condoms—whereas other studies have not found such effects. However, the evidence for some safer sex programs is a little stronger than that presented by the AMA review. After all, several studies showing positive effects on behavior for a year or more were well designed with random assignment, large sample sizes (up to 3,600 students) and long-term measures of behavior (up to 31 months). One safer sex program has been independently evaluated in several schools in California and Arkansas and was found to be effective in both studies.^4,5 Furthermore, the curricula found to be effective at changing behavior have common characteristics that are thought to contribute to their success.

Both the AMA report and Coburn state that making condoms available in schools leads to greater condom use, but the evidence is not consistent. Only four studies of programs making condoms available in schools have been published.^6,7,8,9 Only one of these studies evaluated the effects of making condoms available in several schools, collected baseline and follow-up data, had a comparison group, and had large sample sizes (7,179 students in 10 intervention schools and 16,296 students in comparison schools in Seattle, Washington).⁶ That study found that students did take a large number of condoms from the schools when condoms were made available without any restrictions in open baskets in school health centers. However, that study also found that condom use among youths who were sexually experienced did not increase; it decreased. Students simply took condoms from the schools' health centers instead of from other sources. To understand the decline in condom use, the authors conducted focus groups with groups of students and examined schoolwide survey data. They found that even before condoms were made available in the schools, condoms were available from other sources in the community. The reasons youths gave for not using condoms did not typically include lack of access.

Of the three other published studies, two found significant increases in condom use,^7,8 and the third found nonsignificant trends in that direction.⁹ Each of these three studies, however, was limited by one or more of the following methodologic problems: lack of baseline data, lack of comparison groups, insufficient sample sizes, or changes in parental consent procedures resulting in serious attrition at follow-up. In addition, two of these studies measured the effects of broader, more comprehensive programs directed at preventing human immunodeficiency virus infection or health promotion programs,^7,8 not solely the availability of condoms in schools.

Three possible conclusions can be made from these studies of making condoms available in schools. First, the differences in results could be due to differences in the research methods. These studies would provide only weak evidence that making condoms available in schools increases their use, because the strongest study failed to find such an effect.

Second, the differences could be caused by differences in the communities and in student needs. If communities do not provide condoms in convenient and confidential locations, then their availability at school may increase their use, whereas if communities already make them available, then adding school availability may not increase their use. Before making condoms available, schools should assess whether doing so would meet a real need.

Third, in two of the studies, the differences in study results could be due to other factors (for example, educational components). This suggests that schools should determine why youths have sex without condoms. If students have little motivation to avoid having sex or to use condoms, or if they lack the skills to refuse sexual advances, insist on condom use, or use condoms properly, then effective programs to promote safer sex should be implemented. If unsafe sex is part of a larger pattern of substance misuse, poor school performance, family dysfunction, and community disorganization—as it often is—then these causes also need to be addressed.

Schools should also consider the costs of making condoms available. The financial costs are small, but the social or political costs may be large. For religious or moral reasons, some people may strongly oppose making condoms available in schools, and both their beliefs and the community conflicts that might ensue should be properly considered.