Urban, sexually active female adolescents who have recently had a main or casual sexual relationship with a partner they met through a venue other than school have significantly increased odds of being dissimilar in age to their partner (odds ratios, 3.2-21.5). Females who have ever had a sexually transmitted disease (STD) have elevated odds of having had a recent main partner whose ethnicity or history of hard drug use is different from their own (2.2-3.1). For males, no such associations are apparent when all of these factors are considered simultaneously. These findings come from a 1997-1999 study of teenage clinic clients in San Francisco.1
Participants were 14-19-year-olds recruited at two clinics (a municipal STD clinic and a general adolescent medicine clinic). In a baseline interview, research assistants elicited information on participants' demographic characteristics and social and sexual histories. At a second interview six months later, data were collected on characteristics of the participants' main sexual partners ("someone...you are serious about") and casual partners ("not a main partner") since baseline and on where or how they had met.
The analyses included the 473 teenagers who reported at follow-up having had intercourse in the previous six months. Among the 151 male participants, the average age was 17.2. Half were black or Hispanic (29% and 22%, respectively); the rest were Asian (16%), white (12%), or of other or mixed ethnicities (22%). Ninety-eight percent considered themselves heterosexual; the mean lifetime number of sexual partners was 8.2. One-third of males had been with main and casual partners in the previous six months; 7% had ever had an STD. Eighty-four percent had used drugs; 79% of males' recent main partners, and 70% of their casual partners, had never used hard drugs.
The average age among the 322 females was 16.9. Blacks comprised the largest female ethnic subgroup (27%); next came whites (20%), Hispanics (18%) and Asians (17%). Eighteen percent reported other or mixed ethnicities. Ninety-four percent identified themselves as heterosexual; the mean lifetime number of partners was 5.4. In the past six months, 29% had had main and casual partners. Twenty-two percent had had an STD. Most females had used drugs (81%) but thought that their recent main (68%) and casual (68%) partners had never used hard drugs.
Among males, the most popular venues for meeting main partners were social networks (31%) and school (30%), followed by street locations, such as public transportation or parks (17%); work (15%); and clubs or parties (7%). The most frequently cited venue for meeting casual partners was school (31%); then came social networks (21%), clubs or street locations (18% each), and work (13%). Females had met more than one-third of recent partners--main (39%) or casual (35%)--through social networks. For meeting main partners, the next most popular venues were school (23%), street locations (18%), work (13%) and clubs (8%); for casual partners, street locations, work or school (18% each), and clubs (11%).
Males' partners were generally age-concordant (i.e., partners were less than two years older or younger than they were). However, 50% of casual partnerships among males with a history of STDs and 60% of those among males who had met their partner at a street venue involved an age-discordant partner. After adjustment for age, ethnicity, drug use, meeting venue and STD history, logistic regression analysis revealed no significant associations with age discordance (although in bivariate analysis, meeting a casual partner at a street venue was associated with significantly higher odds than meeting at school).
Among females, most recent relationships involved age discordance (a difference of three years or more). The subgroups with the greatest prevalence of discordance were females who had met their main partner at a street location, females who had met their casual partner at a club and females with an STD history who were in casual partnerships (76-83%). Only three female subgroups had age discordance in fewer than half of the relationships: blacks with casual partners and females with main or casual partners they had met at school (20-49%). In multivariate analysis, the odds of age dissimilarity were significantly elevated for females in main or casual partnerships involving a meeting venue other than school (odds ratios, 3.2-21.5), for female hard drug users in main partnerships (2.1) and for females in casual partnerships who had had an STD (3.0).
Most relationships were between partners of the same ethnicity. The highest proportions of ethnicity discordance among males' relationships were seen between casual partners who had met through friends and main partners who had met at work (61-65%). In most subgroups, 20-50% of males reported ethnicity discordance, but the proportions were even lower among males who had met their partners at street locations and blacks with main and casual partners (13-16%). Bivariate analysis revealed decreased odds of discordance for black males in main partnerships and for males meeting casual partners at street locations, and showed increased odds for hard drug users in main partnerships; however, these relationships did not retain significance after adjustment for potential confounders.
Among females, Asians and Hispanics with casual partners had the highest proportions of ethnicity discordance (71-73%), blacks with main or casual partners had the lowest (13-14%), and most other subgroups reported discordance in 24-48% of relationships. For black females with main partners, the odds of ethnicity discordance were one-10th those for whites (odds ratio, 0.1). The odds of dissimilar partnering were elevated for females who had met their main partner at a street location or work (3.8-5.5) and for those with a main partner and an STD history (2.2). Although no associations were found for females' casual partnerships in the multivariate analysis, meeting at work was associated in the bivariate analysis with decreased odds of ethnicity discordance; Hispanics and Asians had increased odds of discordance with casual partners.
The majority of partnerships involved couples with the same drug use status (i.e., both had tried, or not tried, hard drugs). Most male subgroups had a prevalence discordance falling in the range of 13-48%; the exceptions were blacks in casual partnerships and males in main relationships who had never tried hard drugs (6-8%) and whites and drug users in main relationships (56-59%). In bivariate analysis, black males had reduced odds of discordance with main partners; in main and casual partnerships, drug users' odds of discordance were higher than those of nonusers. However, in logistic regression analysis, no variables were associated with drug use discordance in males' partnerships.
Among female subgroups, proportions of drug use discordance ranged from 11% to 43%; the subgroups with the greatest proportions of discordant couples were females with main partners they had met at work and drug users with casual partners. Multivariate analysis showed that compared with white females, black females had decreased odds of drug use discordance with their main partners (odds ratio, 0.2). Females had increased odds of discordance if they had met their main partner at work (5.1) or had had an STD (3.1).
The researchers had sought to identify teenagers likely to have sex with persons within and outside of their own sexual networks--which presumably consist of sexually active persons who share a demographic trait or social behavior; who have mutual friends; or who work, attend school or congregate together. They note that persons with dissimilar partners may spread STDs between demographically and socially disparate populations. However, as the researchers point out, their study did not examine whether discordant partnering was associated with STD transmission.
The researchers believe that adolescents should be considered at increased risk for dissimilar partnering, and possibly for STDs, if they have a history of drug use or STDs. They also conclude that interventions for STD prevention should concentrate on venues associated with elevated likelihoods of discordant partnering.--C. Coren
1. Lee JK, Jennings JM and Ellen JM, Discordant sexual partnering: a study of high-risk adolescents in San Francisco, Sexually Transmitted Diseases, 2003, 30(3): 234-240.