Pleasure, Prophylaxis and Procreation: A Qualitative Analysis of Intermittent Contraceptive Use And Unintended Pregnancy

Abstract

CONTEXT

Although pregnancy ambivalence is consistently associated with poorer contraceptive use, little is known about the sexual, social and emotional dynamics at work in pregnancy ambivalence.

METHODS

During in-depth sexual and reproductive history interviews conducted in 2003, 36 women and men were asked about the relational and emotional circumstances surrounding each pregnancy, as well as their thoughts about conceiving a baby with both current and previous partners. An ethnographic, inductive approach was used to analyze the data.

RESULTS

Half of respondents had experienced at least one unintended pregnancy. Respondents described three categories of pleasure related to pregnancy ambivalence: active eroticization of risk, in which pregnancy fantasies heightened the charge of the sexual encounter; passive romanticization of pregnancy, in which people neither actively sought nor prevented conception; and an escapist pleasure in imagining that a pregnancy would sweep one away from hardship. All three categories were associated with misuse or nonuse of coitus-dependent methods.

CONCLUSIONS

For some individuals, the perceived emotional and sexual benefits of conception may outweigh the goal of averting conception, even when a child is not wholly intended. Future behavioral studies should collect more nuanced data on pregnancy-related pleasures. Clinicians and patients would benefit from clearer guidelines for assessing ambivalence and for linking ambivalent clients with longer-acting methods that are not coitus-dependent.

The prevalence, health consequences and political salience of unintended pregnancy have made it a source of significant policy concern for several decades. Nearly half (48%) of all U.S. pregnancies are unintended, giving the United States one of the highest unintended pregnancy rates in the industrialized world.^1,2 Despite recent declines in unintended pregnancy rates among middle-and upper-class women, rates are rising among the most socially disadvantaged women.¹

Significant evolution in the research in recent years has sparked disagreement on the definition of, the precursors to and even the usefulness of the concept of pregnancy intendedness. For example, women often report being happy about an unintended pregnancy, and a third of pregnancies resulting from contraceptive failures are subsequently classified as intended.^3,4 Furthermore, for a large subset of women, pregnancy intention does not appear to be linked to contraceptive behavior.⁵

Though the social and demographic patterns of unintended pregnancy are well documented,^1,2,6 the psychological predictors are far less clear. A growing number of studies associate intermittent contraceptive use with pregnancy ambivalence—that is, unresolved feelings about whether one wants to have a child at a particular time.^5,7 Using a nationally representative sample of adult women, Frost and colleagues found a strong association between pregnancy ambivalence and contraceptive nonuse or gaps in use.^6,8 Several studies of adolescents similarly demonstrate that ambivalence is associated with reduced odds of contraceptive use.^9–13 And Zabin and colleagues discovered that women’s childbearing intentions and behaviors varied across partners, events and time periods.¹⁴ However, that study was not designed to explore the psychological dynamics through which partner and context influence pregnancy ambivalence. Nor do similar studies help explain the sexual, social and emotional processes at work in shaping pregnancy ambivalence, or how those processes are shaped by gender, social class or other types of social inequality.

Indeed, reflecting the field’s general neglect of the role of sexuality in reproductive behaviors,¹⁵ few researchers have examined whether unprotected sex or ambivalence about pregnancy may heighten the sexual experience, or whether the romantic notion of creating a child with someone may deter the use of contraceptives. We know little about the emotional, physical and cognitive states that contribute to situations in which lack of contraceptive use is pleasurable or purposeful to women and men.

In contrast, the AIDS literature has examined ways in which sexual pleasure-seeking motivates HIV risk-taking among men who have sex with men.^16–18 Deliberately unprotected anal sex, or barebacking, is eroticized among certain men who have sex with men, even those who may be well aware of the attendant risk of HIV.^19–22 In some circumstances, riskier sex may be “hotter” or closer sex; some men may also desire to share a disease with a loved one in order to facilitate closeness or connection.^*

This HIV scholarship provides a useful explanatory model for the eroticization of unsafe sex, but the same concepts cannot be mapped perfectly onto pregnancy risk. Heterosexual couples are more likely to encounter gendered power differentials than same-sex couples; further, risking an unintended pregnancy carries far different consequences than risking an HIV infection. We want to explore the degree to which women and men find pleasure in the possibility of a pregnancy with a particular partner or at a particular moment; how this helps people meet certain sexual, social and emotional needs; and whether this could help explain the link between ambivalence, contraceptive use and unintended pregnancy.

METHODS

Sample Strategy and Construction

We explored our research question as part of a larger study on the effects of sexual pleasure-seeking on contraceptive use. We collected the data for this project in 2003, using theoretical sampling²³ to recruit respondents from metropolitan Atlanta. Similar to purposive or quota-driven sampling, theoretical sampling is used to select participants on the basis of variables most likely to affect the outcomes of interest, as established in the literature and any experience with the population. The variables are used to create cells within a sampling frame that are filled as recruitment ensues.

In this case, gender and social class were the primary selection variables. Women, the focus of our study, made up the majority of the sample (24 participants); men were included for comparative purposes (12 participants). The women’s mean age was 36, and the men’s was 32. Given well-documented social class differences in rates of unintended pregnancy, we sought respondents from different backgrounds to investigate whether there were class-based differences in their experiences of sexual pleasure that might help explain how inequality translates into health disparities.

While the majority of public health research assesses education level²⁴ or poverty status² as a proxy for social class, our approach incorporates both financial and cultural resources.²⁵ Cultural dimensions of social class—for example, tastes and habits stemming from childhood socialization processes—may be at least as important as income and education in shaping sexual behavior.^26,27 Cultural resources may be identified in part through participants’ social class of origin and upbringing, current neighborhood and habitat, or occupation (e.g., an employee of a nonprofit agency may have the same salary as a grocery store clerk but occupy a different class position).

During the screening call, we asked participants about their education level (categorized as any college or no college), occupational status (white-collar, blue-collar, unemployed or homemaker) and neighborhood (middle-class, working-class or poor). We created two class labels—socially advantaged (middle-class) and less socially advantaged (working-class and poor)—and assigned participants the one that corresponded to their responses on at least two of these measures. During the interviews, we collected information on additional social class criteria: social class of origin, as determined by early home environment, and current financial situation and cultural resources, including needs for housing, food, clothing or other basic necessities. If the first interview revealed information that contradicted our original assignment, particularly on social class of origin, we reassigned the class label accordingly. Reassignment occurred in only two cases.^†

In keeping with theoretical sampling, we also selected participants to represent a range of other variables that are related to contraceptive use and unintended pregnancy: race and ethnicity, age, marital status and parity.²⁸ Thus, within each main cell of the sample (socially advantaged women, socially advantaged men, less socially advantaged women and less socially advantaged men), we tried to capture whites and blacks, a range of ages, ever-married and never-married participants, and parous and non-parous participants (Table 1).

TABLE 1.

Number of participants in a study of pregnancy risk-taking, by selected demographic characteristics, according to sex and social class, Atlanta, 2003

Characteristic	All (N=36)	Less socially advantaged women (N=12)	Socially advantaged women (N=12)	Men* (N=12)
Marital status
Ever-married	12	4	3	5
Never-married	24	8	9	7
Race
White	17	3	8	6
Other†	19	9	4	6
Parity
0	23	6	8	9
≥1	13	6	4	3

^*Six men were socially advantaged; the other six were less advantaged.

^†The sample contained one Latina. Blacks accounted for the remainder of this sub-group.

Participants were recruited through several mechanisms, including flyers distributed in numerous Atlanta neighborhoods and word-of-mouth referrals from neighborhood contacts and other study participants. We also sent notices to community-based electronic mailing lists, including those of parent-teacher associations, community advisory boards, youth organizations and church groups. The notices were then forwarded, as intended, to many people beyond the original recipients. Potential participants called the telephone number provided and were informed of the inclusion criteria: Participants had to be 18 or older and had to have used some type of pregnancy prophylaxis in the past 12 months. Individuals also provided information on the sampling variables of interest.

Interview Protocol

Before beginning the interview, respondents read and signed a consent form. The study protocol and interview instruments were reviewed and approved by the institutional review board at Emory University.

Interviews were conducted in participants’ homes or in public places near their homes or workplaces. Each interview took approximately three hours to complete, so a total of 104 hours of interviews were digitally recorded and transcribed. At the completion of the interview, which usually took place over two or three sessions, participants were paid $40.

The semistructured interview guide elicited information on relationship, sexual and reproductive histories, including questions on all contraceptive methods used and on every pregnancy experienced, whether intended or unintended. Respondents reported on the circumstances surrounding each pregnancy, including the relationship, financial and emotional contexts. Further, for each sexual or romantic partner described, regardless of whether a pregnancy was involved, respondents were asked to report any thoughts they had had about having a baby with that person. The guide was designed so that topics moved from less to more sensitive, as a way to enhance rapport and data validity. We also administered closed-ended questionnaires that collected information on income level, receipt of public assistance and highest level of education completed.

Analysis

An ethnographic, inductive approach was used in analyzing the data, meaning that the analyses were informed by both preexisting themes from the literature and the research questions and themes that arose from the data themselves. We first read, reread and summarized the transcripts and field notes from each interview. We then extracted excerpts of transcribed interviews relating to each pregnancy in the sample and coded these detailed observations using the preexisting classifications of intendedness, unintendness, happiness and unhappiness. In working with the data, we identified three types of pleasure relating to pregnancy ambivalence and unintended pregnancy: active eroticization of pregnancy risk, passive romanticization of the notion of having a baby with a particular partner and desire to be swept away from one’s current life circumstances. We then used these categories to code other relevant areas of the transcripts that might contain these themes (e.g., descriptions of romantic partners, or reasons offered for intermittent use of contraceptives). Once coding was complete, we compared both individuals and subgroups on the basis of gender and class using descriptive and analytic cross-case analysis.²⁹

RESULTS

Past Unintended Pregnancies

At least 17 respondents had been involved in at least one unintended pregnancy (Table 2). Further, since several respondents had been involved in multiple unintended pregnancies, more than half of reported pregnancies (28 of 49) were unintended. In this sample, class and gender discrepancies were apparent: Whereas nine of the 12 less socially advantaged women had experienced at least one unintended pregnancy, only five of the 12 socially advantaged women had done so. Among the men, three out of 12 confirmed their involvement in an unintended pregnancy, but the reports of another three suggested that they likely had been involved in one, as well. For example, two men said that a partner had indicated to them that she was pregnant, but that she had had an abortion or miscarriage before paternity had been confirmed. Only one socially advantaged man reported involvement in a suspected or confirmed unintended pregnancy (not shown).

TABLE 2.

Selected reproductive history measures, by respondents’ sex and social class

Measure	Less socially advantaged women (N=12)	socially advantaged women (N=12)	Men (N=12)
≥1 lifetime unintended pregnancy	9	5	3 (3)
Total pregnancies	25	13	11 (1)
Unintended pregnancies, by outcome
All	16	6	6 (5)
Abortion	6	4	2
Birth	9	0	3
Miscarriage/stillbirth	1	2	1
Births	18	7	7

Notes: Numbers in the last column are low estimates, as some men did not know whether their partners had experienced these events. Numbers in parentheses represent unconfirmed but likely pregnancies.

Pregnancy and Pleasure

Relationships between pleasure and pregnancy shaped contraceptive practices. Some of the respondents’ nonuse or intermittent use of contraceptives was associated with ambivalence about pregnancy or the pleasures associated with pregnancy, even when they did not actually desire a child or the responsibilities of parenthood.

Active eroticization of pregnancy risk

The least common but most direct articulation of the pleasures of pregnancy ambivalence took form in the eroticization of pregnancy risk. In these cases, participants described increased sexual arousal at the prospect of conception. A detailed illustration of this eroticization comes from Alex, a 27-year-old, socially advantaged woman who experienced an unintended pregnancy at the age of 22. At the time, she was in a serious relationship with a man whom she imagined marrying. Because Alex disliked the side effects and “unnaturalness” of hormonal methods, she and her partner used a condom or diaphragm. At the beginning of the relationship, they were “religious” about using contraceptives consistently, but as the relationship progressed, they occasionally used withdrawal or nothing at all. When asked to explain the decrease in consistent use and what characterized those occasions when they did not use anything, Alex responded:

“Sometimes when I was having sex with him, I would just kind of lose my mind a little bit and want to have a baby with him so badly. It was like I can’t get close enough to him or connected enough with him, and conceiving a child would be the closest we could get.”

In fantasizing about what it could be like to create a life with her partner, Alex eroticized the risk of pregnancy, thus leading her to occasionally abandon contraception. Alex said that when she discovered she was pregnant, she was “horrified, of course,” but “part of me was kind of thrilled about [the pregnancy], almost in awe of it.” She continued:

“It seemed amazing to me that we could create this, this life together. And I suppose a part of us did want to keep it, especially since at that point we imagined we’d be together forever. It was … romantic to imagine building a permanent life with him.”

Nevertheless, Alex explained that since she was in her first year of law school, “I wasn’t ready to have a child. I wasn’t ready to give up my plans.” Thus, although it had felt sexy to flirt with the idea of pregnancy with her partner, Alex came up sharply against her more immediate goals and decided to terminate the pregnancy.

Jo Jo, a 36-year-old, less socially advantaged man, also had eroticized pregnancy risk. He recounted a passionate relationship with a woman who had moved away after they had been together only briefly. They used condoms “sometimes, for sure. Definitely at the beginning. But less over time. Maybe, like, half the time.” When asked if he had ever thought about having a baby with this woman, Jo Jo unhesitatingly replied, “Yes, absolutely. A gorgeous, smart woman like that? Of course.” Later, Jo Jo spoke of the kinds of communication, both verbal and nonverbal, that occurred between them when they did not use condoms:

“Sometimes she said it was OK not to, because she wasn’t at a time in her cycle where she was going to get pregnant. And sometimes she wouldn’t say anything. So I wouldn’t put one on, and we’d have sex anyway. You know, all unspoken. [Pause.] Sometimes I wondered if she wanted me to get her pregnant those times. Man, it would turn me on so much.”

Although the idea that a pregnancy could occur heightened Jo Jo’s pleasure during sex, as far as he knew, no unintended pregnancy resulted from these encounters.

Occasionally, respondents described the sexual charge they associated with actively trying to become pregnant. A 25-year-old, less socially advantaged woman said, “I enjoyed sex so much when I was trying to conceive my son.” Another 25-year-old woman, this one socially advantaged, had not yet had a child but eagerly anticipated a time when she would discontinue contraception and fully embrace procreative sex. “I think it would be so intense,” she said.

Passive romanticization of procreation

In many cases, ambivalence manifested not as heat-of-the-moment arousal, but as a less dramatic romanticization of the general idea of a pregnancy with a particular partner. Certain respondents flirted with pregnancy in the form of a subtle romantic fantasy that also contributed to intermittent use or nonuse of contraceptives. While these respondents did not actively intend to conceive, they did not stringently avoid pregnancy, either.

Joseph, a 30-year-old, less socially advantaged man, began an intense relationship with a woman while in his mid-20s. His partner had been taking oral contraceptives at the start of their relationship, but she discontinued them several months later for reasons he could not recall or never knew. They used no other method. She soon became pregnant, and she decided immediately to have an abortion—a decision he supported and felt “secretly relieved” about. After the termination, the couple used condoms “about half of the time.” When asked about the intermittent use, Joseph replied:

“If I don’t know a girl at all, I’ll definitely use a condom. But if I’ve been with a girl for a while, and if we’re really into each other.… [Pause.] A pregnancy with her wouldn’t have been terrible, even though we weren’t trying to have kids.”

Even though he said he had neither desired nor felt prepared for fatherhood, Joseph had imagined, at least occasionally, having a baby with his partner. His on-again, off-again contraceptive use pattern reflected his fluctuating pregnancy desire.

Another illustration of the romantic notion of procreation comes from Lydia, a 32-year-old socially advantaged woman and a scrupulous contraceptive user, who described a pregnancy scare that she had had when her period was late. At the time, she was taking oral contraceptives and dating the man who later became her husband. She was convinced that he was “the one”; she wanted to be with him indefinitely and have children with him. But they had not been planning a pregnancy.

Lydia described crying when she tested negative with a home pregnancy test. “I was relieved, but also sad,” she said. “I had almost gotten excited about the idea of making a baby with him. It was something I wanted to do with him, just not right then.” While alluding here to the romantic pleasures involved in creating life with a loved one, even when the event is unplanned and feared, Lydia spoke at other points about how much she valued the careful planning of the births of her two children.

Most cases of romanticizing the notion of pregnancy occurred within long-term relationships. However, shorter-term relationships occasionally became fertile ground for such romanticization. Few women spoke of yearning for pregnancies with short-term partners, but men occasionally mentioned the appeal of impregnating a desirable woman. One example comes from Martin, a 38-year-old, less socially advantaged man who claimed to have had sex with approximately 500 women. He spoke in great detail and very emotionally about one of his first sexual partners, a 15-year-old, whom he had been involved with at age 17. Martin did not use a condom when the two first had vaginal intercourse, but not because he lacked pregnancy-related knowledge: “I certainly realized she could get pregnant. In a way, I was hoping she would.”

Although this encounter did not lead to an unintended pregnancy (as far as he knows or reported), it could have. Martin had not wanted to embrace the emotional and financial responsibilities of fatherhood and had not eroticized risk. Rather, he had wanted to impregnate an attractive, socially desirable young woman: “She was real pretty—looked like an Indian. Her attitude was just lovely. I felt like it would be nice to have a baby with a person like that,” he explained.

Escapist pleasures

Even when they were not planning or hoping for a baby, several respondents came to embrace an unintended pregnancy as a way to foster a relationship, cultivate a new family and potentially escape the hardships of their lives. Not surprisingly, only less socially advantaged women—especially those who had become pregnant at a young age—described this phenomenon. Pregnancies represented temporary hopes that things would get better and that their unborn children would enjoy brighter futures. This perspective has surfaced in other research, as well.^30,31

At the age of 14, Destiny, a 25-year-old, less socially advantaged respondent, met a man who lived in her public housing complex. “Mm, I thought he was somethin’ else,” she said with a laugh. They did not talk about wanting a pregnancy, nor did they discuss or use contraceptives. She soon contracted gonorrhea; she also became pregnant. “When I found out about [the gonorrhea],” she said, “I was so in love with him that I didn’t want to believe it.” When she learned about the pregnancy, Destiny experienced a mix of anxiety and excitement. “At first, I was sad and worried,” she said. “I thought, ‘What my mama gonna do to me?’ I wanted to keep going to school, but I also loved him and wanted to have his child.” She and her partner never considered abortion, as her mother was against it and “we just don’t believe in that.”^* She continued, “It was scary, you know, being pregnant at 14. [My partner] was excited, though, and this meant a lot. I got ready to have his baby.” Throughout most of the pregnancy, Destiny thought she would move in with him and his grandmother, but “it just didn’t work out,” and she stayed in her mother’s apartment. The birth of her baby girl precipitated Destiny’s experience with intermittent homelessness and public assistance.

Although Destiny ended up raising her daughter single-handedly, the pregnancy had briefly represented great possibilities: a connected home life with a partner she loved, an escape route from her undesirable living situation, and a way to build a new life and new family.

Rose, a 50-year-old, less socially advantaged respondent, told a similar story about becoming pregnant for the first time at age 14. She met her boyfriend and future husband at the Magnolia Club, a popular dance spot for young people. She was taken with the charming, attractive young man and his attentions:

“He was real handsome. He was popular. Every girl likin’ him and everything. … He asked me to dance. I said no, and he called me a wallflower. He danced like he was having sex—all rolling and everything. He kept on coming over, aggravating me and everything. Then it was last time—the last dance—and he asked me again. I said yes. We danced, and afterwards he asked if I wanted him to walk me home.”

Although she felt slightly pressured and even badgered by the young man, and was disarmed by his overt sexuality, she also enjoyed the attention and found his overtures intriguing. “He didn’t like the other girls,” she said. “He liked me.”

In addition to his flattery, the young man offered Rose an escape from her harsh life at home, which was marred by poverty, a strict and exhausted mother, nine siblings and an absent father. She remembered feeling impoverished by a lack of material and status items, and while taken with the young man himself, she was perhaps even more seduced by the new life and new things he promised: “There were some basic things I didn’t have then. The Magnolia Ballroom—it was a place where I could just … get away. Get nice things and whatnot.” When asked how she got nice things, Rose replied, “He bought them for me. New clothes. Jewelry sometimes. Flowers. He gave me something to ease me from what I had at home. Something to get me away. Have better things in life. I only had five dresses.”

As their sexual relationship developed, Rose played the role of gatekeeper and “wouldn’t let him cross certain lines.” She was interested in presenting herself as a good girl, who did not say yes to sex easily. But the young man continued to court her, and eventually they had intercourse. Soon, she missed her period and acknowledged she was pregnant. She hid at her boyfriend’s house. “I wanted to avoid my mother,” she said. “Also, I had fallen in love with him.”

Despite her mother’s disappointment in her, Rose expressed happiness about the pregnancy:

“I felt good when I found out I was pregnant. I felt like … I wanted two children. I wanted to get out of my mother’s house. I wanted to make a family. [Pause.] Sometime it don’t work out like that.”

Rose’s story exemplifies some of the ways in which poverty affects sexual and procreative goals. The pregnancy with this charming, popular young man provided the illusory promise of being “swept away” into a better life. The couple wed before the baby was born, but the marriage turned physically abusive shortly after the wedding and ended after the birth of their fourth child. At least two of their births were unplanned.

Because of the paucity of social, financial and, often, emotional resources at home, Destiny, Rose and other less socially advantaged women turned to their partners and their potential children to fulfill their needs, even when they did not hope to become pregnant at the time of conception. In contrast, several socially advantaged women described using dual or even triple methods during their early sexual experiences, reporting that a pregnancy at that stage of their lives would have been disastrous.

Cases of being “swept away” by pleasure were also marked by gender inequality. Women depended on their romantic relationships in a way that few men did. Male respondents used sex to uphold their identities as potent, seasoned and skilled lovers, but they did not rely on long-term relationships for social mobility and material resources in the same way as poor women. Jamara, a 19-year-old, less socially advantaged respondent, articulated this issue even though she had not experienced an unintended pregnancy. Her parents had moved from a housing project in New York to a working-class community in Atlanta to secure a better life for their family. Her mother spoke openly with her daughter about sex—its pleasures as well as its disappointments—and Jamara had postponed having vaginal intercourse until age 19. However, a number of her high school classmates had become pregnant unintentionally and carried their pregnancies to term. She offered the following insights into how teenage women relied on these pregnancies to publicly solidify their relationships with and “claim to” their boyfriends:

“Usually the person that they got pregnant with was somebody that they had been with for a while. So, it wasn’t just ‘I got pregnant,’ but ‘This is his child.’ So, you know, it’s … it’s somebody that they were in love with, so you know, that’s … ‘our baby.’”

In Jamara’s narrative, her female peers invoked a narrative of romance and love to justify and seek acceptance for the pregnancies. The young women took pleasure and pride in carrying the product of their connection with a particular young man. Their lack of contraceptive use may have been influenced by their sexual goals and their sense that unintended pregnancy could be a reasonable outcome—as opposed to an ill-planned or unfortunate consequence.

DISCUSSION

We have outlined some of the ways in which women and men gain pleasure from pregnancy ambivalence and how contraceptive use can be undermined as a result. Respondents sometimes placed more value on emotional and physical intimacy than on the goal of averting pregnancy or disease through contraceptive use. Sometimes, they even deliberately avoided facing realities such as the risk of pregnancy or disease. Doing so helped women and men meet certain sexual, emotional and social needs, including sexual arousal and fulfillment, closeness and connection with their partner, and a more emotionally and materially secure future. Given the existence of these needs, intermittent or nonexistent contraceptive use—practices that are consistently portrayed in our field as failure to do something, the consequence of an “unmet need” for family planning or a health risk that rational people would want to avoid—may represent purposeful action. That is, participants … sometimes had more immediately salient goals than averting pregnancy, even in the absence of pregnancy intention. The social, emotional and sexual benefits they cite contradict the notion that their behavior is irrational or self-destructive.^23,32

For some respondents, sex occasionally became a way to flirt with and eroticize pregnancy risk, often as an avenue for seeking ultimate closeness with one’s partner. Aroused during sex by the idea of pregnancy, these respondents dispensed with contraceptives. At least one other preliminary study has suggested that in the heat of the sexual moment, a couple’s or individual’s temporary desire for a pregnancy could lead to unprotected sex, even if a baby is not fully or rationally intended.³³ Likewise, abortion clinic clients often describe a temporary surrender to the fantasy of a pregnancy.³⁴ Luker, in her study of abortion in the United States, found that conservative, religious women often claim that contraception devalues their sexual intimacy and reduces their physical pleasure because their excitement is heightened by the possibility of a conception.³⁵ Our findings also echo some of the literature on bare-backing among gay men,^19–22 for whom the perceived closeness and greater physical pleasure of skin-on-skin sex, or even the prospect of sharing a disease with one’s partner, may heighten the sexual encounter.

In a second type of pleasure, respondents romanticized pregnancy in a more removed, abstracted way. Although pregnancy was neither intended nor not intended, the notion of creating a baby was compelling—a situation in which some couples used contraceptives inconsistently or not at all. This finding echoes the work of Stanford and colleagues, who have used the term “passive preceptors” to describe fecund couples who do not use contraceptives but are not trying to conceive.³⁶ (One of our colleagues described this situation as “not not trying to have a baby.”) Unprotected sex could meet people’s need for connection with and love and support from their partners. This phenomenon dovetails with an analysis by Bartz and colleagues, in which half of coital events were unprotected, even among adolescent women who reported they were committed to not getting pregnant.¹³ The authors found that day-to-day factors, such as respondents’ daily assessments of partners’ support and feelings of being in love, could help predict lack of contraceptive use. Feeling loved and supported are two social benefits that can carry great weight in the heat of the sexual moment.

A third category of pleasure, described only by socially disadvantaged women, pertained to pregnancy’s imagined promise of altering one’s life circumstances, even when conception is not intended. Pregnancies could be seen as a way to solidify commitment from men, build kinship networks, create social affirmation and transform one’s current living situation. In resource-poor settings, where young women have few educational and professional opportunities and depend on men for social affirmation and sometimes financial support, unintended pregnancies could hold promise, however tenuous, for positive change.

Although we have identified three types of pleasure, sample size restrictions mean that we could neither capture the many nuances of these categories nor describe all the differences within and across gender and class groups. For example, with cell sizes larger than six, we could have made more systematic comparisons between socially advantaged and disadvantaged men. However, even when small, samples whose selection is theoretically driven hold great merit for policy-relevant research because of their ability to draw attention to certain social phenomena or relationships. Rather than showing causation between variables or definitive comparisons between groups, we wanted to give name and voice to an association not thoroughly articulated in previous research: how the pleasures of ambivalence shape contraceptive use and unintended pregnancy, and how gender and class influence that relationship. Toward that end, our sample—albeit small—served us well.

Given that themes discussed here appeared to influence the ways in which contraceptives were used or not used, we suggest that future behavioral studies of contraceptive use and unintended pregnancy consider how flirting with the idea of procreation may decrease one’s motivation to use prophylaxis. Women and men benefit psychologically and socially from sexual risking-taking—a notion missing from the family planning literature, which, like public health literature more broadly, tends to ascribe qualities of future orientation, rationality and safety-consciousness to people’s reproductive health behaviors.^23,32,37 These benefits may apply particularly to pregnancy risk-taking, which leads not to infection, but potentially to closeness, affirmation and a new life. More research is needed on developing reasonable programmatic and clinical guidelines addressing ambivalence, including the assessment of clients’ potential for ambivalence and an emphasis, when appropriate, on methods that are not coitus-dependent.

This study also provides further evidence for how and why men should be included in research on unintended pregnancy. Few studies attempt to measure men’s pregnancy ambivalence and its possible influence on contraceptive use.³⁸ Our analysis demonstrates that men, too, experience pregnancy ambivalence that affects the consistency with which couples use condoms, withdrawal and other methods that require men’s involvement.

Pregnancy ambivalence does not account for all unintended pregnancies. A full understanding of the pleasures of pregnancy ambivalence will do little to improve insufficient reproductive health services, let alone provide disadvantaged women and men with motivations, such as educational opportunities and rewarding employment, to prevent unintended childbearing. Nonetheless, we argue that the field would benefit from considering the social, psychological and sexual benefits of pregnancy risk-taking, even in the absence of active desire for a child. A more detailed canvas of these benefits and pleasures will ultimately improve programmatic responses to people’s short-term and long-term reproductive goals. Particularly if unintended pregnancy prevention remains an ongoing policy priority, these concepts deserve attention within sexual and reproductive health models.