Heterosexual anal sex represents a relatively new area of public health inquiry that has emerged in response to increased rates of HIV infection among women.1 Anal sex is a health concern primarily because it is associated with a greater risk of STD transmission than oral and vaginal sex, though estimates of that risk vary widely.2–6 Prevalence estimates suggest that 30–40% of adults of reproductive age in the United States have engaged in heterosexual anal sex.7,8 Moreover, multiple investigators have found that the prevalence of heterosexual anal sex increases substantially between midadolescence (less than 5% among 15–17-year-olds) and the mid-20s (about 30% among 22–24-year-olds).8–11

Whereas a substantial body of epidemiologic, behavioral and intervention research focuses on anal sex among men who have sex with men, research concerning heterosexual anal sex is in its infancy. To our knowledge, no estimates based on rigorous calculations are available of the extent to which heterosexual anal sex contributes to the HIV epidemic in the United States. However, existing prevalence research, coupled with the relatively high HIV transmission probabilities associated with anal sex, suggests that its contribution may not be trivial. These concerns are compounded by studies that have shown consistently low rates of condom use during anal sex among heterosexual couples.7,11–13

Most recent research on heterosexual anal sex has been quantitative, based on clinical and community samples, and focused on three sets of correlates: co-occurring risk behaviors, demographic characteristics, and relationship and gender dynamics.5 Nearly all of it provides evidence that engagement in anal sex is associated with other risk behaviors, particularly drug or alcohol use and having multiple sexual partners, either concurrently or over the course of a lifetime.9,13–15 Comparisons of the prevalence of anal sex among various demographic groups have produced mixed results. For example, some studies have found whites to be more likely than minority groups to report anal sex,11,13 while others have found no association between ethnicity and anal sex.7,9 Some studies have found that the prevalence of anal sex increases with age, is equally likely to be reported by men and women, and is more common with “main” partners than with casual partners,7,9,11,13 but others have reported contrary results for each of these associations.7,14,16

The findings related to gender relations are also mixed. For example, Lescano and colleagues found an association among young women between recent heterosexual anal sex and experience with forced sex;13 others have found that engaging in anal sex may signify risk for engaging in transactional sex.9 Some studies have demonstrated that anal sex within more serious relationships may reflect commitment or love. For example, Kaestle and Halpern found that in a U.S. sample of young couples, male respondents’ love for a partner (reciprocated or not) was positively associated with having anal sex with that partner.10 Maynard and colleagues’ qualitative study involving a cohort of women with histories of unprotected anal sex showed that anal sex is sometimes prompted by a desire to please a partner and is viewed as a reflection of intimacy.17 And in a study of couples of reproductive age, Billy, Grady and Sill identified positive associations between recent anal sex and two measures of men’s power advantage in the relationship.16

The limited and largely inconclusive research findings on heterosexual anal sex inspire exploration of additional behavioral issues. For example, what distinguishes those who engage in anal sex from those who do not? Is anal sex perceived as a “deviant” behavior? What does anal sex mean to the people who engage in it? Examining these kinds of questions among young adults is particularly important, given their relatively high reported prevalence of both heterosexual anal sex and STDs.18,19 To further explore these fundamental aspects of heterosexual anal sex, we present results from a mixed methods research project involving community samples of urban Puerto Rican and black young adults. Results include survey findings about the prevalence and predictors of anal sex, as well as qualitative findings on the study population’s experience with anal sex.

METHODS

Project PHRESH.comm

We analyzed data from the Philadelphia and Hartford Research and Education on Sexual Health and Communication (PHRESH.comm) project, which was funded by the Centers for Disease Control and Prevention (CDC) and carried out from 2004 through 2008 by the University of Connecticut and the Family Planning Council of Southeastern Pennsylvania. Because sexually active Puerto Rican and black men and women aged 18–25 have increased risk of unintended pregnancy and STDs, these groups were the target population. The focus areas were decision making, communication and behaviors related to contraceptive use, condom use, pregnancy and STD risk, as well as the broader socioeconomic and cultural contexts of those behaviors. The project used complementary data collection strategies to obtain a rich set of qualitative and quantitative information on these topics. For this study, we relied on two data sets from the larger project that provided distinct perspectives on heterosexual anal sex: qualitative information from 70 participants who kept coital diaries and completed debriefing interviews in 2006–2007, and quantitative data from a 2007–2008 survey of 483 men and women. The study protocol was approved by the institutional review boards of the CDC and the partner institutions.

Recruitment and Sampling

Study teams recruited participants from selected Hartford and Philadelphia neighborhoods that had relatively high STD and teenage pregnancy rates and large black or Puerto Rican populations. Eligibility criteria included self-identification as Puerto Rican or black, being 18–25, not being pregnant, having been born on the U.S. mainland or in Puerto Rico, being fluent in English and ever having had heterosexual sex. For the survey cohort, all participants also had to have had sex with a heterosexual partner in the preceding six months. For the qualitative cohort, participants also had to have had, in the previous month, either more than one heterosexual partner or one such partner with whom they had had sex at least three times. By design, similar numbers of male, female, Puerto Rican and black participants were recruited for each cohort.

For the qualitative cohort, participants were recruited through street outreach efforts targeting venues frequented by the general population (e.g., libraries, supermarkets), word of mouth or referrals, and fliers or newspaper advertisements. All qualitative interviews were conducted in study-related community centers or offices.

Survey participants were recruited through venue-based sampling. Study teams in the two cities mapped neighborhood sites that members of the target population frequented or gathered at on a regular basis (e.g., laundromats, bus stops, corner stores, parks). They used that information to create sampling frames from which they randomly selected venues to visit during the interviewing slots on their fieldwork calendars. Most venues were eligible for selection for all sampling periods, which were mainly during the day and included both weekdays and weekends. During sampling periods, research staff approached as many individuals as possible who appeared to be from the target population. Most people who were eligible and agreed to participate completed the survey in a private area at or adjacent to the recruitment venue. In Hartford, the sampling frame included 41 venues in nine neighborhoods. In total, 478 people were screened, and 241 (50%) completed surveys; 108 were ineligible, 76 eligible individuals declined to participate and 53 did not show up for interview appointments (which were scheduled owing to inclement weather on some recruitment days). The Philadelphia study team created three sampling frames crossing seven zip codes, with a total of 207 venues. The interview teams screened 325 people, of whom 242 (74%) completed surveys; 74 were ineligible, and nine eligible individuals declined to participate. Study teams did not track whether survey participants had also been in the qualitative cohort, but given the time lag between the two study components and the different recruitment strategies, they believe that few, if any, individuals participated in both.

Data and Measures

Most interviewers were women. Qualitative interviewers had a bachelor’s or master’s degree in anthropology, public health or a related discipline; the educational backgrounds of survey interviewers were more mixed. All interviewers were extensively trained and supervised, especially to ensure rapport-building and minimize possible social desirability bias. The qualitative interviews were recorded with the consent of participants and were transcribed verbatim. Participants received financial compensation for their time ($25 per qualitative interview and $35 for the survey).

The survey used a standardized closed-ended format. Participants were asked whether they had ever had anal sex (not specifying whether with a partner of the same or opposite sex) and, if so, their age at first anal sex. They were also asked about anal sex experiences with their most recent serious or casual heterosexual partner (both as defined by the participant, but excluding one-night stands) from the preceding six months. Participants were specifically asked whether they had ever had anal sex with that partner and, if so, the condom use pattern during anal sex for the previous six months. Finally, they were asked about condom use during vaginal sex with those same partners in that time period.

The assessment of descriptive characteristics was exploratory and based on previous research. Six measures of demographic characteristics were included: race or ethnicity, gender, age, educational attainment, parity and whether participants lived with a parent. To assess sexual history, the survey asked participants their lifetime number of partners, whether they had ever had a same-sex partner and whether they had ever had an STD.

Three measures examined characteristics of participants’ most recent serious relationship: the relationship’s duration, an index of intimacy and whether they had ever lived with that partner. Relationship duration was derived from respondents’ reports of the month and year that the relationship started (and, in some cases, ended), and coded as 0–3 months, 4–6 months, 7–9 months, 10–24 months, 24–48 months or more than 48 months; it was assessed both continuously and categorically. The intimacy index was created using the following six questions: “How important to you is your relationship with [partner]?”“How much of the time do you believe that [partner] is telling the truth about where she/he has been or the people she/he has been with?”“How much of the time does [partner] believe that you are telling the truth about where you have been or the people you have been with?”“When you are upset about something, how often do you feel that [partner] is there for you?”“How much does [partner] care about you?”“How much do you care about [partner]?” Likert-scale responses were dichotomized to distinguish the most positive response category for each question (coded as 1) from all others (coded as 0), and these individual scores were then summed to yield an overall index score of 0–6; this measure was also assessed continuously and categorically.

Participants in the qualitative cohort were asked to use a diary provided by the interviewer to chronicle their sexual communications and behaviors on a daily basis for five weeks. They returned for weekly debriefing sessions, which included an open-ended review of the previous week’s diary and a discussion of the events that had taken place. Discussion topics were wide-ranging, including minor and major events in participants’ sex lives and relationships. Interviewers probed for information about use of condoms and other contraceptives, and often asked about the type of sex the participants had had during the sexual episodes noted in their diaries (e.g., “Did you have vaginal, oral or anal sex?”). This question often elicited comments about anal sex, and some participants then discussed their thoughts and experiences in more detail. Thus, while interviewers did not systematically ask all participants about their experiences with or attitudes about anal sex, they used follow-up questions to elicit more detailed information about it, when appropriate.

Analysis

For the survey data, we calculated descriptive statistics, compared bivariate frequencies using Pearson’s chi-square tests and conducted multivariate logistic regression analysis to assess predictors of anal sex with the most recent serious heterosexual partner; analysis was limited to this partner because few respondents reported anal sex with their last casual partner. The initial model included all variables significant at the 10% level in the bivariate tests. We identified high correlation among the three sexual history variables and found that age, education and cohabitation were not significant or added little to the model fit. Hence the final model included only gender, race or ethnicity, and lifetime number of partners. We performed logistic regression diagnostics (e.g., examination of outliers and residuals), applied various measures of model fit and assessed the robustness of the final model.

For the qualitative data, we identified interview passages related to anal sex through word searches and by reading transcripts. We then summarized each identified passage, developed and applied codes to describe themes, and grouped the passages accordingly. Most themes captured attitudes about anal sex in general or rationales for having or avoiding anal sex. Codes included, for example, “condom use,”“pain” and “to please the partner.” Codes and passages were compared to identify patterns in the data regarding gender, ethnic subgroup and partner type. Partners were defined in broad categories as “serious” or “casual,” based largely on participants’ own descriptions. Serious relationships tended to involve some degree of commitment, emotional connection and expectation of monogamy, while casual relationships largely lacked these qualities.

RESULTS

Quantitative

•Full sample. Survey participants’ average age was 21 (range, 18–25), and one-third were living with a parent. About half had completed a high school education, and one in eight had some postsecondary schooling. Nearly all were unmarried, and half had at least one child.

Thirty-four percent of all respondents had ever had anal sex; a higher proportion of Puerto Ricans than of blacks reported such experience (42% vs. 26%, p<.01). The mean age at first anal sex was 17.6 (standard deviation, 2.8); it was higher among women than among men (18.2 vs. 16.9, p<.01), but did not differ by race or ethnicity. All gender and ethnic subgroups had had anal sex for the first time approximately 3.5–4.0 years after first having vaginal sex. No respondent reported having had anal sex before vaginal sex.

•Sample with recent serious relationship. Among the 400 participants who reported on their most recent serious heterosexual partner in the preceding six months, the median number of lifetime sex partners was six; 20% had had an STD. About half of these participants had lived, or were living with, a serious partner; the median duration of serious relationships was about two years.

Twenty-two percent of participants had ever had anal sex with their most recent serious partner; by comparison, only 8% of the 130 respondents with a recent casual partner reported having had anal sex with that partner. Of those who had had anal sex with a serious partner in the last six months, 67% said they had never used a condom during anal sex in that time period, and 23% said they had used one every time; the remaining 10% reported inconsistent use. The level of inconsistent condom use during vaginal sex with a serious partner in this period was even higher: Thirty-four percent had used condoms inconsistently, 54% had never used a condom and 12% had used one every time. Concordance in condom use patterns was fairly high: Two-thirds of these respondents reported the same pattern of condom use for both anal and vaginal sex. Among the 10 respondents who had had anal sex with a casual partner in the previous six months, five had used a condom every time with that partner, and one never had.

In bivariate analyses, Puerto Rican respondents were more likely than blacks to report having had anal sex with their most recent serious partner (31% vs. 13%), and women were more likely than men to report this behavior (25% vs. 17%—Table 1). Puerto Ricans were also more likely than blacks to have had anal sex with a casual partner (14% vs. 3%, p<.05—not shown), but women did not differ significantly from men in their reports regarding casual partners (6% vs. 8%). Respondents aged 22–25 were more likely to report anal sex with a serious partner than were those aged 18–21 (27% vs. 18%); anal sex experience also varied significantly by education level, but not in a consistent pattern.

Table 1. Percentage of adults aged 18–25 reporting anal sex with their most recent serious heterosexual partner, by selected characteristics, and adjusted odds ratios (and 95% confidence intervals) from logistic regression analyses assessing predictors of anal sex, Hartford and Philadelphia, 2007–2008
Characteristic N % reporting anal sex Odds ratio
  1. *p≤.05.

  2. **p≤.01.

  3. Score indicates the number of Likert-scaled items (out of six) to which participants gave the most positive response; see page 269 for details. Notes: na=not applicable, because characteristic was not included in the model. ref=reference category.

DEMOGRAPHIC      
Race/ethnicity      
Puerto Rican (ref) 200 31** 1.00
Black 200 13 0.30 (0.17–0.51)**
Gender      
Male (ref) 180 17* 1.00
Female 220 25 2.92 (1.58–5.39)**
Age      
18–21 231 18* na
22–25 169 27 na
Education      
<high school 143 28* na
High school 202 17 na
>high school 53 23 na
SEXUAL HISTORY      
Lifetime no. of partners      
1–3 (ref) 106 15 1.00
4–6 100 23 2.03 (0.98–4.24)
7–12 101 19 2.38 (1.09–5.17)*
≥13 93 30 5.47 (2.39–12.51)**
Ever had same-sex partner      
No 359 20* na
Yes 41 34 na
Ever had an STD      
No 315 20 na
Yes 83 29 na
RELATIONSHIP WITH INDEX PARTNER      
Intimacy score      
0 21 19 na
1 33 27 na
2 66 26 na
3 80 25 na
4 85 16 na
5 70 24 na
6 39 15 na
Ever lived together      
No 174 17* na
Yes 223 26 na
Relationship duration (mos.)      
0–3 45 13 na
4–6 28 21 na
7–9 28 29 na
10–24 99 20 na
25–48 101 21 na
>48 97 27 na
Pseudo-R2=.08      

Of the three sexual history measures, only one was significant: Respondents who had ever had a same-sex partner were more likely than others to report anal sex with their recent serious partner (34% vs. 20%). Among the 41 individuals who reported having had a same-sex partner, 93% were women; thus, this association cannot be attributed to men who have sex with both men and women.

With regard to relationship factors, the only significant finding was that those who had ever cohabited with the index partner were more likely than those who had not to report having had anal sex with that person (26% vs. 17%). We found no associations for the intimacy index or relationship duration, despite assessing these measures both continuously and categorically.

In the regression analysis, blacks were less likely than Puerto Ricans to report having had anal sex with their most recent serious partner (odds ratio, 0.3), and women were more likely than men to report such experience (2.9). Compared with respondents who had had 1–3 sexual partners, those who had had 7–12 or 13 or more had elevated odds of having had anal sex with a serious partner (2.4 and 5.5, respectively). However, our final model poorly predicted this outcome, explaining approximately 8% of the variation in anal sex.

Qualitative

•Participant characteristics. The demographic profile of participants in the diary cohort was similar to that of survey respondents. One-third were living with one or both parents, and 13% were living with a partner. About a third had less than a high school education, a third had at least one child and none were married. Most participants reported having had multiple sexual partners in the prior 12 months (83%).

None of the participants mentioned anal sex without interviewers’ first raising the issue. Some respondents laughed when asked about anal sex; others seemed unwilling or uncomfortable using the term. Of the 70 participants, 46 provided varying amounts of information about anal sex. Twenty provided more extensive information on anal sex, and of those, five discussed their anal sex experiences in great detail (three black females, one Puerto Rican female and one Puerto Rican male). While comments about anal sex came from participants describing both serious and casual sexual relationships (including one-night stands), in-depth discussions tended to be limited to serious relationships. Overall, black females and Puerto Rican males provided most of the information collected regarding anal sex.

•Reasons not to have anal sex. Negative views toward anal sex dominated. Some participants viewed it as unappealing, while others said they found or expected it to be too painful (for the female partner). Comments included benign statements like this one from a 19-year-old black male: “I never did anal sex. I’ve never been interested.” A 24-year-old Puerto Rican male revealed both the discomfort some participants showed in talking about anal sex and the pain that some partners experience. Explaining why he did not have anal sex with one of his regular partners, he stated, “She ain’t—it’s not something—she was—it didn’t work. It didn’t happen.… It hurts her, and she’s like stop, stop.” And a 20-year-old black female, responding to a question about whether she had had anal sex with one of her steady casual partners, laughed, saying:

“No … I can’t. No. I think I’d go crazy or something.… I just said [to my partner], ‘No, I’m not into that.’… He didn’t get mad. It didn’t really matter. It just was a question, a curiosity would I do it or not. And not!”

Negative comments did not come only from participants who were sexually conservative or inexperienced (e.g., a woman who did not want to try oral sex either). A number of participants who expressed dislike for anal sex reported engaging in a variety of other sexual behaviors (e.g., use of pornography, “rough sex”) or having concurrent partners. For example, one 19-year-old black woman worked as a model, dancer and personal escort, and exchanged sex for money. When the interviewer asked her whether she had had vaginal and anal sex during a recent encounter, she replied, “I never do anal sex.… I’m traditional.”

A few participants cited an association between anal sex and homosexuality, even when a woman is the receptive partner, and indicated that this association prevents them from engaging in the behavior. This was exemplified by a 22-year-old black female’s response to a question about whether anal sex is “something that most men expect women to do”:

“No, no. They want … no, I can’t even say they want it. I know most of them might want it…. It’s weird because … one time, I had a partner that wanted it so much, and I’m like, ‘Are you gay?… Why you want that so much? Is the vagina not enough for you?… You makin’ me think stuff…. Makin’ me think you gay.”

Such statements were complemented by a small number of male participants who made clear that contact with their anus was off-limits during sex.

No participants described negative repercussions for refusing to engage in anal sex. Both men and women gave examples of refusing anal sex, either at all or during a particular sexual encounter. Such refusals were said to be acceptable and respected. This attitude is illustrated by a 21-year-old Puerto Rican man who had two regular partners; he often had anal sex with one, and he described the other as simply “not into the anal game.” The man described a one-night stand with a third woman, but noted that they did not have anal sex because “she said that was her forbidden zone.”

•Reasons to have anal sex. Most participants who said they had had anal sex did not elaborate on why. Rather, statements regarding anal sex were often made in the context of describing enjoyable sexual encounters. Thus, many participants who had had anal sex presumably had done so for sexual gratification. Some respondents confirmed this directly. For example, one 25-year-old black female, who regularly engaged in anal sex with both serious and casual partners, said plainly, “I like havin’ anal sex. I like havin’ it.” A 21-year-old black female said that she had occasional anal sex with her serious partner, but that they “just laugh during anal sex, which is hilarious.” A 24-year-old Puerto Rican female who had used a sex toy the first time she and her boyfriend had anal sex explained its appeal, saying, “To me, it felt nice because it was different…. It increases my sex drive.” A 24-year-old Puerto Rican male said that he preferred women who had vaginal and anal sex, adding, “I just like [anal sex], so I try to get them into it. But if they don’t like it, there’s nothing I can do about it.” Finally, a 19-year-old Puerto Rican male described meeting a woman at a college party and having sex with her that night. He was “happy to oblige” when she requested anal sex from him, saying, “She said she liked [anal sex].”

Participants also suggested that anal sex could carry particular meaning within more serious relationships. For example, a few women described having anal sex with their serious partners because they thought that sexual variety would prevent their partners from looking to other women for sexual satisfaction. Reflecting this sentiment, one 22-year-old black female described why she did not think her partner was having sex with other people:

“I do my part as far as sex-wise. I know he’s satisfied…. Instead of him going to somebody else, like, ‘Well, I wanted to do this, and you wouldn’t do it, and she was in my face with it.’ So, I try to do everything…. I did anal sex with him, and I never did that with nobody else.”

Some respondents also felt that anal sex communicated intimacy. For example, a 21-year-old Puerto Rican female described having anal sex for the first time with a long-term casual partner, with whom she wished she could be in a serious relationship. She described having had mixed feelings about having anal sex with him, since she had had a painful experience in the past:

"A large minority of this young adult population had had anal sex, and had engaged in it with their most recent serious partners."

“He was like, ‘If you don’t want to do this, I’m not pressuring you to do it.’ And I’ll say, ‘No, I want you to do it, like I want to experience it, but then on top of that I want to do it ’cause I really like you,’ so I let him do it.… I did not like it, but I just did it to make him feel happy.”

When the interviewer asked if she had really wanted to have anal sex, she affirmed her decision, further describing what the act had meant to her: “When he was done, I was like, ‘See, this should mean how much I care about you.’ And he was like, ‘I care about you baby. I know, baby.’” Nevertheless, she decided against having anal sex again, at least for now, saying, “[He’s] got to be my husband for me to do that [again].” Another Puerto Rican female said she would have to be married to her long-term partner before she would agree to have anal sex with him.

In some instances, anal sex was seen as a marker of relationship progression, as exemplified by a 24-year-old Puerto Rican male who reported on his initial anal sex encounter with a serious partner: “I guess it makes the relationship … deeper, or I guess we get closer…. Well, we wanted to experiment, it seems. We experimented.” This couple continued to have anal sex during the remaining diary period, while also drawing closer to one another.

The link between anal sex and intimacy in serious relationships was also clear to a 21-year-old Puerto Rican male, who explained, “We want anal sex from our significant others because it’s better. It’s a new hole. It hasn’t been touched. If it has been touched, it still hasn’t been touched thoroughly. Why else would you do it?” These participants expressed some of the ways in which anal sex may signify emotional intimacy in serious relationships: It can be a gift from one partner to the other, a reflection of mutual willingness to experiment and something exclusive to that relationship.

Across relationship types, participants usually had anal sex in addition to vaginal (and often oral) sex during a sexual encounter. Anal sex rarely substituted for vaginal sex, and only one participant indicated that she opted for anal sex instead of vaginal sex because it carries no pregnancy risk. Pregnancy prevention did not appear to be part of the motivation to have anal sex among these participants.

•Condom use. Only 11 respondents provided information about their use of condoms during anal sex. Among those who expressed reasons for having used condoms, one participant said he had done so unintentionally, as he had expected to have only vaginal sex that time. Two others used condoms because of STD concerns, and two used condoms for hygiene reasons unrelated to STDs. A 23-year-old black female said, “It just seems weird to me for some reason to not have a condom having anal sex, knowing the types of things that come out of there.”

The primary reason for not using condoms during anal sex was similar to that cited for not using condoms during vaginal sex: little worry about STDs with a particular partner. One 21-year-old Puerto Rican female was using the contraceptive patch and did not use condoms during vaginal or anal sex with her serious partner. The interviewer asked if not using condoms during anal sex was an issue for her:

“No, not at that point, ’cause I’ve done it before with him with no condom. I mean it should be a big issue, ’cause I don’t know what he’s doing [with other women], but at that point I wasn’t even thinking about using a condom.”

•Subgroup comparisons. While rigorous comparisons were limited because of unequal amounts of data about anal sex across the four subgroups defined by gender and ethnicity, each subgroup provided support for all of the dominant themes identified here, thereby suggesting that the groups did not differ much regarding those aspects of anal sex. However, exceptions arose for two minor themes: The majority of comments about a link between anal sex and homosexuality came from black participants; the feeling that anal sex provides sexual variety that could help prevent a partner from cheating was mentioned only by women.

DISCUSSION

Consistent with previous findings,5,8,13 our survey data showed that a large minority of this young adult population had had anal sex, and had engaged in it with their most recent serious partners. The negative perceptions of and painful experiences with anal sex (whether actual or anticipated) expressed by some participants in the qualitative cohort may help explain why anal sex was not more common. Also, as suggested by the apparent need for direct interviewer questioning to initiate discussion of anal sex, many participants seemed somewhat uncomfortable with the topic.

Only a small proportion of survey respondents reported having had anal sex with a recent casual partner, possibly because this population knew that anal sex is a relatively high-risk behavior, or because they saw it as a behavior more commonly reserved for more serious relationships. The qualitative data illustrated that anal sex can communicate emotional intimacy and serve to build and maintain a serious relationship, largely because it is less common than other types of sex. However, anal sex did not seem to be a requirement for emotional intimacy, and it was largely considered an optional sexual behavior. The relatively low prevalence of anal sex within serious relationships reported in the survey, as well as the lack of an association between the index of intimacy and anal sex, may reflect this.

Qualitative findings also suggest why we should not be surprised that anal sex was not rare in this population. These data highlighted that anal sex can be pleasurable for both men and women, and it seemed within the scope of sexual possibilities, as Leichliter asserts,1 particularly in serious relationships. These findings debunk the characterization of anal sex as primarily a behavior of “high-risk” populations and individuals. Although as in previous research,5 we found that people with anal sex experience had had, on average, more sexual partners than others, some of our results caution against using participation in anal sex as a marker of broader sexual risk-taking. First, not all participants who reported engaging in anal sex also reported other risky sexual behaviors, and vice versa. Second, the qualitative results suggest that both expectations of sexual pleasure or pain and experience with anal sex may be more important predictors of participation in anal sex than underlying risk-taking propensity. Finally, it is difficult to know what the association with the lifetime number of partners indicates; more partners could reflect just more sexual experience, rather than an underlying risk propensity. Regardless, the number of partners accounted for little of the variation in reports of anal sex. Future research should attempt to assess the contribution of anal sex experience and perceptions to anal sex behaviors.

The link between anal sex and serious relationships helps explain condom use patterns. Like other studies,5,7,11,13,20 the survey found low condom use during anal sex, and the qualitative data suggest that this may reflect that serious partners are often not considered sources of STD risk. Extensive research on condom use during vaginal sex affirms this logic.21 These data also suggest that hygiene concerns may help explain some of the condom use that was reported.

The qualitative data did not provide much insight into the ethnic and gender differences in anal sex experience identified in the survey data. Nevertheless, it is notable that the few respondents who described having anal sex to keep a partner sexually satisfied were all women. This rationale was also reported by Maynard and colleagues,17 and together these findings underscore the need for additional research on the gender and relationship dynamics of anal sex.

Limitations

Despite the strengths of the two study components, the survey and qualitative results are not necessarily generalizable to other population groups—for example, adolescents, or even urban black and Puerto Rican young adult populations elsewhere in the United States. The survey data are representative of sexually active black and Puerto Rican young adults who can be found at mainstream venues in those targeted neighborhoods of Philadelphia and Hartford. Given the strategies used to recruit the qualitative cohort, those data are not representative in a statistical sense. However, they are illustrative of a wide range of thoughts and experiences reported by this population. Furthermore, the survey data included a limited number of potential explanatory variables, and the survey did not include items on the frequency of anal sex. The qualitative data were limited in that heterosexual anal sex was not a major focus of inquiry; as a result, collection of data about anal sex was not uniform across participants. This fact constrained comparisons by subgroups and our ability to identify more fully the important themes regarding this behavior. Finally, participants from both study cohorts may have underreported anal sex because of discomfort with the topic.

"These findings debunk the characteriza-tion of anal sex as primarily a behavior of “high-risk” populations and individuals."

Conclusions

For researchers, these findings highlight particular areas for future work and help reframe heterosexual anal sex as a matter not just of “risk” and “risk behavior,” but of sexual health. An overemphasis on the epidemiologic risks of heterosexual anal sex would not serve this study population well. For health care providers, the findings indicate that many young adults in heterosexual relationships may be thinking about, and having, anal sex. Yet our experience indicates that to address anal sex, providers must initiate discussions in this area, as clients are likely to be uncomfortable doing so. Moreover, discussing the pros and cons of condom use during anal sex and related health concerns with clients is likely worthwhile, although, as for vaginal sex, expectations of consistent condom use during anal sex with all partners are likely unrealistic. Overall, we hope these data help broaden the collective understanding of why some young adults may or may not engage in anal sex, and in turn lead to more productive exchanges on this topic, both among researchers and between clients and health care providers.