Timing, Amount of Teenage Alcohol or Marijuana Use May Make Future Risky Sex More Likely

C. Coren

First published online:

Specific patterns of substance use during adolescence can predict risky sexual behavior in young adulthood. Researchers in Seattle base this conclusion on their analysis of data from a longitudinal study of urban youths.1 For example, compared with study participants reporting no teenage use of marijuana, persons reporting late-onset use of marijuana had significantly more sexual partners at age 21 and were less likely to use condoms consistently. However, unlike some previous studies, this study found no association between use of illicit hard drugs, such as cocaine or heroine, and risky sexual behavior.

The data come from a project that recruited fifth graders at 18 Seattle elementary schools located in high-crime areas in 1985; the youngsters were followed through 1996, when they were 21 years old. Of the 808 participants in this study, 52% lived in low-income households. The sample had a diverse racial and ethnic makeup (46% white, 24% black, 21% Asian American and 9% other) and roughly equal numbers of male and female participants.

The participants were surveyed about their substance use at multiple points during their adolescence: about their use of alcohol, cigarettes, marijuana and other illicit drugs at ages 13, 14, 15 and 16, and about their use of all these substances except hard drugs at age 18. The follow-up survey at age 21 asked participants about their sexual behavior in the past year (i.e., the number of partners they had had and whether they had used condoms in all sexual encounters), whether they were involved in a stable relationship (i.e., lived with a partner) and whether they had used condoms at first intercourse. In all of the surveys except those at ages 13 and 17, participants were asked their age at first sexual intercourse.

For each type of teenage substance use examined, the researchers used a statistical model to group participants according to patterns of use. The resultant categories distinguished participants' behaviors according to the onset and frequency of use and changes in use over time. The researchers used negative binomial regressions and logistic regressions, respectively, to assess whether patterns of substance use in adolescence were independently related to number of partners and consistency of condom use in young adulthood.

Statistical modeling resulted in the following categories of adolescent drinking: chronic (steady) bingeing (3% of participants), binge drinking begun during the mid-teens and then steadily increasing over time (4%), binge drinking begun in the middle to late teen years (23%) and no binge drinking (70%). For use of cigarettes, the patterns that emerged were chronic heavy smoking (1%), smoking that escalated over time (8%), smoking begun after age 14 (11%), experimental smoking (7%) and no smoking (73%). Categories of marijuana use were use that began in the early teen years (3%), use with sharply increasing frequency after age 15 (5%), use begun in the late teen years (19%) and nonuse (74%). Hard-drug use was categorized as having begun in the early teens (7%) or later in adolescence (4%), and nonuse (89%).

On average, at age 21, study participants reported having had 1.9 sexual partners in the previous year. Young adults whose teenage patterns of substance use reflected no alcohol bingeing reported the lowest number (1.7), and those with chronic binge drinking reported the highest (3.0). The mean had a smahller range within the other types of substance use: for cigarette smoking, 1.9 (for nonsmokers and experimental smokers) to 2.2 (for all other types of smokers); for marijuana use, 1.8 (nonusers) to 2.5 (those who began using in their late teens); and for hard-drug use, 1.8 (users beginning in late adolescence) to 2.3 (users beginning in the early teens). Participants whose first sexual intercourse occurred at or after age 13 (88% of the total sample) reported an average of 1.9 recent partners at age 21; participants younger than 13 at first intercourse reported an average of 2.4.

Eighty-two percent of the total sample reported not always using condoms for sexual intercourse at age 21. Ninety percent of participants who had been chronic adolescent binge drinkers reported inconsistent use, compared with 80% of those who had never binged. The level of inconsistent condom use was markedly lower among participants who had experimented with cigarette smoking (65%) than among nonsmokers (83%) or regular smokers (85-86%). Inconsistent condom use was reported by 88% of participants who started using marijuana in the late teen years, 81% of nonusers of marijuana and 77% of those who began using marijuana in their early teens. Eighty-eight percent of participants who started using illicit hard drugs in the late teen years, compared with 82% of nonusers, reported inconsistent condom use.

Among participants who had not used a condom at first intercourse, 92% reported not always using a condom as a young adult; for those who had used a condom for their first sexual experience, 78% reported inconsistent use. Ninety-two percent of participants in stable relationships, and 78% of others, had not consistently used condoms for recent intercourse.

In the multivariate analyses examining the associations between number of sexual partners and each type of substance use, the researchers controlled for the effects of the other types of substance use and early initiation of sexual intercourse (before age 13). They found that the number of sexual partners for participants who had been binge drinkers throughout adolescence or in their late teens was significantly larger than that for participants who had not binged (coefficients, 0.5 and 0.3, respectively). Participants who had begun using marijuana in their late teens had significantly more sexual partners than nonusers of marijuana (0.3).

The logistic regression analyses of substance use patterns and condom use were performed by controlling for the effects of other teenage substance use, condom use at first intercourse and involvement in a stable relationship at age 21. Compared with not smoking, experimental smoking in adolescence was associated with a significantly decreased risk of inconsistent condom use in young adulthood (odds ratio, 0.3). The authors suggest that the same sort of self-control that allowed some young people to smoke cigarettes occasionally without escalating to chronic smoking may explain their relatively high rate of consistent condom use. For participants who began using marijuana in their late teens, the risk was nearly double (1.9) that for nonusers. Use of illicit hard drugs was not correlated with either measure of risky sexual behavior.

According to the authors, a limitation of this study is that it did not explore possible factors affecting both substance use and risky sexual behavior. In addition, the study sample included a higher proportion of low-income persons than would be found in the general population. Nonetheless, the authors believe that their findings can provide useful information for the design of programs to curb the risky sexual practices of young adults: "Specifically, interventions that prevent the onset of binge-drinking and marijuana use during high school may be of particular utility in preventing later risky sexual behavior."--C. Coren


1. Guo J et al., Developmental relationships between adolescent substance use and risky sexual behavior in young adulthood, Journal of Adolescent Health, 2002, 31(4): 354-362.