Delayed initiation of intercourse is associated with reductions in the likelihood of some risky behaviors, but not necessarily with the kinds of physical and emotional health benefits that proponents of abstinence-only education suggest it is, according to analysts who examined data from a 1996 survey of U.S. adults.1 Respondents who had first had sex later than the norm for individuals of their gender and background had a reduced likelihood of reporting risky partners and of denying that they were at risk for STDs; associations between delayed sexual onset and reductions in other risky behaviors were found for men or women but not both. However, respondents who had begun having sex later than the norm were no more satisfied with their sexual relationships than were those who had started at a normative age and did not rate their general health any better; males reporting a late start had elevated odds of saying that they had erectile, sexual arousal or orgasm problems.

The National Sexual Health Survey was conducted among English- and Spanish-speaking individuals aged 18 or older who lived in the 48 contiguous states. In telephone interviews, participants answered questions about their background characteristics; sexual behavior and functioning; current relationships; and general physical health. Analysts categorized timing of first sex by examining distributions of age at first intercourse within subgroups defined by gender and racial or ethnic and educational background; ages in the top and bottom quartiles of each distribution were considered late and early for that subgroup, respectively, and those in the middle two quartiles were considered normative. To assess long-term correlates of the timing of sexual debut, the analysts used logistic regression to compare outcomes among respondents who had first had sex at an early or a late age for their subgroup with those who had done so at the normative age. Data were weighted so that results can be generalized to all U.S. adults.

Men and women were about equally represented among the 8,466 respondents; 74% of participants were white, and 55% had no more than a high school education. On average, male respondents were 43 years old, and females were 46. Some 65% and 59%, respectively, were involved in sexual relationships, and most of these were married. The overall and normative mean age at first intercourse was 18; the mean was 14 for participants who had begun sexual activity early and 22 for those who had begun late.

For men, early intercourse was positively associated with reports of almost every behavioral risk examined and of erectile, sexual arousal and orgasm problems; it was negatively associated with participants’ assessments of their general health and was not related to their views of their sexual relationships. Among women, those who had started having sex early had an increased likelihood of reporting most risk-related behaviors, but not of other adverse outcomes.

By contrast, both men and women who had started having sex late had reduced odds of reporting any risky partners in the previous year or in the previous five years (odds ratios, 0.6–0.7), and were less likely than their peers with a normative start to deny their vulnerability to STDs (coefficient, –0.17 for each gender). Additionally, men who had first had sex late reported fewer partners in the past five years than those who had done so at a normative age (–0.16); women with a late start had reduced odds of saying that they had frequently had sex while under the influence of alcohol or drugs (odds ratio, 0.6) and that they had had an STD (0.5).

Participants’ satisfaction with their sexual relationship was not associated with timing of first intercourse, but men who had started having sex late considered their relationships more solid than did their counterparts who had started at a normative age (coefficient, 0.07). Delayed intercourse was associated with reduced odds of problems with sexual arousal among women (odds ratio, 0.8), but with elevated odds of arousal, erectile and orgasm problems among men (1.5–1.7). Respondents’ overall assessments of their health were unrelated to late initiation of intercourse.

Comparisons between respondents who began intercourse before marriage and those who did not have sex until they were married yielded similar results to those for normative versus late initiation. One notable exception is that women whose first intercourse occurred within marriage had an increased likelihood of reporting arousal problems (odds ratio, 1.4).

The analysts note that their study was limited by the use of an existing data set that did not include potentially important outcome variables, and by other characteristics of the data and the sample. They also point out that because the study was cross-sectional, they cannot draw causal inferences from their findings. Nonetheless, they conclude that the results have important implications for sex education in the United States. In particular, the finding of problems with sexual functioning associated with late sexual initiation “lends credence to research showing that abstinence-only education may actually increase health risks and that strategies designed to promote relevant sexual health information, motivation, and skills are likely to be more effective than abstinence-only messages in helping young people avoid short- as well as long-term health consequences.” Greater understanding of the relationship between timing of first intercourse and long-term outcomes, they write, “is urgently needed to inform adolescent health policies and programs.”


1. Sandfort TGM et al., Long-term health correlates of timing of sexual debut: results from a national US study, American Journal of Public Health, 2008, 98(1):155–161.