During the year following a visit to one of three public STD clinics, clients who returned for follow-up visits frequently reported engaging in heterosexual anal sex, mostly without using condoms.1 Multivariate analyses identified several characteristics that were positively associated with the odds of having anal sex during a three-month period, including number of partners, total number of sex acts and sexual activity with a main partner. By far the strongest predictor of consistent condom use during anal sex was consistent use during vaginal sex.
In a secondary analysis of data from an STD prevention intervention trial, investigators examined the sexual behavior of clients attending clinics in Denver; Long Beach, California; and Newark, New Jersey. Trial participants were recruited in 1999–2000, during a clinic visit for an STD examination; to be eligible for the study, clients had to be 15–39 years old and HIV-negative. Those who enrolled received prevention counseling and agreed to return for follow-up visits every three months for a year. During each visit, they used audio computer-assisted self-interviewing to provide information on their sexual behavior with up to three partners during the preceding three months. The analytic sample was restricted to the 2,357 men and women who reported at least one partner of the opposite sex at baseline and were sexually active during the follow-up period. Together, they contributed data on more than 6,000 three-month intervals and more than 9,000 partnerships.
Participants reported heterosexual anal sex in 18% of three-month intervals following their baseline clinic visit. Virtually all of these intervals also included vaginal sex; on average, participants reported 4.7 occurrences of anal sex and 38.0 episodes of vaginal sex per interval. In multivariate analyses, the likelihood that anal sex occurred during a given interval was elevated if the participant reported more than one partner (odds ratio, 1.5), more than 13 sex acts (2.1–3.8) or any unprotected vaginal sex (1.3) during that interval; it also was elevated if the participant reported having bought or sold sex (1.5). Age, gender and race—which have been identified as predictors of heterosexual anal sex in the general population2—were not significant for this sample.
Fourteen percent of all reported partnerships included at least one episode of heterosexual anal sex. Main partnerships were more likely than casual ones to involve anal sex (odds ratio, 1.4), and the odds of this activity were elevated if either partner had been high on alcohol or other drugs during sex (1.4). Compared with partnerships that had included 1–13 episodes of sex during a three-month period, those that had included 14 or more had higher odds of involving anal sex (2.2–3.5). New and established partnerships were equally likely to involve anal sex, and reports of anal sex were not related to whether a partner had been treated for an STD.
Condom use during heterosexual anal sex was rare. Only 27% of intervals in which anal sex occurred contained reports of consistent use; 63% contained no reports of use. The strongest predictor of consistent use was protective behavior during vaginal sex. Intervals in which condoms were always used for vaginal sex were considerably more likely to involve consistent use for anal sex than were intervals marked by inconsistent condom use for vaginal sex (odds ratio, 17.4). Only one other characteristic included in a multivariate analysis was associated with consistent condom use: Participants were more likely to report consistent use during intervals in which they had had multiple partners than in periods in which they had had only one (1.6).
Of all partnerships that had included heterosexual anal sex, 30% had involved consistent condom use. This protective behavior was more likely in main partnerships than in casual ones (odds ratio, 2.1), and was more prevalent in new partnerships than in established ones (1.4). No other partnership characteristics included in the analysis predicted consistent condom use during anal sex.
The researchers acknowledge that because of possible underreporting and the select nature of their sample, the findings should be interpreted with caution. Nevertheless, they note, the results "suggest that clinicians should ask their patients about anal sex, tell them unprotected anal sex is an efficient path for HIV/STD transmission, and recommend consistent condom use for both anal and vaginal sex." Commenting on the broader implications of the study, the author of an accompanying editorial writes, "At a minimum, public health researchers should consider continual monitoring of the prevalence of anal sex, the frequency with which it occurs, and associated condom use in high-risk and the general population."2
1. Tian LH et al., Heterosexual anal sex activity in the year after an STD clinic visit, Sexually Transmitted Diseases, 2008, 35(11):905–909.
2. Leichliter JS, Heterosexual anal sex: part of an expanding sexual repertoire? editorial, Sexually Transmitted Diseases, 2008, 35(11):910–911.