Dyadic qualitative analysis of condom use scripts among emerging adult gay male couples

Abstract

Sexual risk reduction among gay male couples has received increasing attention in light of evidence that primary partners account for many – possibly most – new HIV infections. This study examined the content of condom use scripts in interviews conducted with both members of17 HIV-negative gay male couples. In each couple, at least one partner was an emerging adult (aged 18 to 29). Three scripts were identified: romantic love, unanticipated condomless anal intercourse (CAI), and negotiated safety. Scripts varied in their emphasis on emotional factors versus HIV risk reduction, the salience of sexual agreements, and the presence of an explicit communication goal. Results indicated that condom use may vary for couples as a result of script content and from the fluid adoption of scripts across contexts. Results highlighted potential tensions between emotional closeness and HIV prevention. Condom use cessation and sexual agreements – a potential mechanism for HIV risk reduction – may also serve as expressions of intimacy. This implies interventions which facilitate direct communication about sexual and relational goals – as well as those which expand couples’ repertoire for expressing emotional closeness – may enhance sexual health for gay couples, particularly during the period of emerging adulthood.

INTRODUCTION

Recent years have seen increased attention directed towards main partnerships as a context for HIV transmission among gay and bisexual men (GBM) (Goodreau et al., 2012; Sullivan, Salazar, Buchbinder, & Sanchez, 2009). This attention is due in part to persistent and disproportionate rates of HIV infection among GBM and other men who have sex with men (MSM) (CDC, 2015a). While they account for an estimated 4% of the U.S. population, MSM accounted for 63% of new HIV infections in the US and are the only group to show an increased incidence in HIV infection (CDC, 2015a).

The focus on young GBM in relationships specifically (rather than MSM more generally) reflects the confluence of several factors. First, epidemiological surveillance highlights that younger GBM face disproportionately high rates of infection relative to older GBM as well as heterosexual men of similar age(CDC, 2015b). Second, studies have indicated that main partners account for 35-68% of new HIV infections among MSM (Goodreau et al., 2012; Sullivan et al., 2009). Younger partners are at greater risk of contracting HIV from a main partner and this risk increases with the age disparity between partners (Goodreau et al., 2012; Mustanski, Newcomb, & Clerkin, 2011; Sullivan et al., 2009). Sullivan et al. (2009) estimated that as many as 79% of HIV infections among MSM between the ages of 18 and 24 occur between main partners. This age-range largely corresponds to the period identified as emerging adulthood (approximately age 18 to 29)(Arnett, 2000). Finally, research suggests that GB couples may perceive and navigate HIV-related risk in ways that are distinct from the larger group of MSM in same-sex relationships. Goldenberg, Finneran, Sullivan, Andes and Stephenson (2017) found that perceptions of partners’ sexual orientation and outness(i.e., others’ awareness of the individual’s sexual minority identity) influence perceptions of HIV-related risk as well as sexual behavior. HIV risk perceptions were diminished, and the likelihood of condomless sex increased, when partner’s identified as heterosexual, were less out, and were engaged in sex with female (as well as male) partners.

Together, these findings point to the need to understand how gay couples, especially those involving emerging adult GBM, navigate HIV prevention and condom use. Starks, Tuck, Millar, and Parsons (2016) used data from 100 gay couples to examine communication processes that are central to the enactment of negotiated safety. Negotiated safety is a well-studied method of couples’ HIV prevention in which a couple discusses their HIV status and establishes rules for allowing condomless anal intercourse (CAI) within the relationship while placing restrictions on sex with outside partners (e.g., monogamy) or requiring condom use with outside partners (Davidovich, de Wit, & Stroebe, 2000; Guzman et al., 2005; Kippax, Crawford, Davis, Rodden, & Dowsett, 1993; Kippax et al., 1997). Starks et al. (2016) found that 63% of gay couples reported using condoms the first time they had anal sex and partners concurred that they disclosed HIV status prior to first CAI in 82% of couples. Couples who reported more syndemic stress were less likely to use condoms at first intercourse and less likely to discuss HIV status prior to first CAI.

Starks et al. (2016) hypothesized that syndemic burden may diminish a couple’s capacity to engage in effective prevention communication at critical points in their relationship development. While useful, their quantitative study provided little insight into the sexual scripts which may govern HIV prevention generally and condom use specifically for gay couples at these formative points in development. A sexual script refers to an “organized cognitive schema” which “defines the situation, names the actors, and plots the behavior” (Gagnon, 1990, p. 6) of individuals in a sexual context. Simon and Gagnon (1973) suggested that sexual scripts may exist at the cultural, interpersonal, and intrapersonal levels. These levels correspond respectively to broad social norms, specific social interactions, and intrapsychic beliefs.

Mutchler (2000) identified four scripts which emerged from the stories of the sexual experiences of young GBM. These included themes of sex as an expression of romantic love, erotic adventure, sexual safety, and sexual coercion. Of particular interest to the current paper, Mutchler (2000) observed that the “sexual safety” script, which proscribes condom use and discourages “risky” behaviors such as CAI, was a relatively new schema which may exist largely on the cultural level. Enactment of sexual safety scripts was related to their degree of integration into interpersonal and intrapsychic scripts.

More recent work by Campbell et al. (2014) highlighted racial differences in the enactment of sexual scripts reported by concordant negative and serodiscordant same-sex male couples. Couples in which both members identified as Black commonly reported implicitly defaulting to condom use. In contrast, no White couples and a minority of inter-racial couples reported this implicit understanding that condoms would be used. Most couples in which both members identified as Black used condoms regularly. A small number of Black couples reported CAI which occurred, “in the heat of the moment” without discussion or planning. In these instances, unplanned CAI was generally followed by an explicit discussion of HIV prevention and a return to condom use. In contrast, many White and inter-racial couples reported that the initial occurrence of CAI was viewed as implicit permission to forego condom use. When unexpected CAI triggered an explicit discussion of condom use in White couples; discussions generally resulted in an understanding to forego condom use after these experiences in contrast to their Black counterparts.

Together, these results suggest that the enactment of sexual safety scripts is contextualized by other sexual scripts and directly involves interpersonal communication processes. These scripts of primary interest are interpersonal in nature – they involve exchanges between partners in the dyad and emerge from narratives that are shared. Lewis et al. (2006) presented evidence suggesting that the sexual health of partners within a relationship can be thought of as linked, and HIV prevention therefore becomes a joint or shared goal of the couple consistent with the tenets of Couples Interdependence Theory (CIT) (Rusbult & Van Lange, 2003). The existing literature conducted within this framework suggests that sexual agreements and emotional factors are likely to emerge as relevant within couples’ condom use scripts.

A sexual agreement refer to the couples’ understanding about the boundaries and limitations on sexual behavior with people outside of the relationship (Hoff & Beougher, 2010). Research suggests that HIV risk reduction is one potential motivation for negotiating a sexual agreement (Hoff & Beougher, 2010; Hoff, Beougher, Chakravarty, Darbes, & Neilands, 2010; Lewis et al., 2006). At the same time, a substantial amount of research has indicated that decisions around condom use are associated with emotional factors (Bauernmeister, 2012; Goldenberg, Finneran, Andes, & Stephenson, 2015; Golub, Starks, Payton, & Parsons, 2012; Starks, Golub, Payton, Weinberger, & Parsons, 2014; Worth, Reid, & McMillan, 2002). Feelings of love, intimacy, and trust have been linked to diminished perceptions of HIV infection risk and increased likelihood of sexual risk taking (Goldenberg et al., 2015). The perception that condoms interfere with intimacy has been shown to account for CAI above and beyond the perception that condoms reduce physical pleasure and HIV risk (Golub et al., 2012).

While the broad issue of condom use has been well-studied in gay men, the existing literature on condom use scripts is limited in two key ways. First, few studies have utilized dyadic data, which means most of what is known about condom use scripts in gay couples is inferred from individual interviews. Second, few studies have examined how condom use emerges and fluctuates across time within a relationship. We know relatively little about how relational development shapes the content of condom use scripts. The current study therefore set out to conduct a dyadic examination of condom use scripts in gay male couples, with specific attention to how condom use has fluctuated across time within the relationship.

Based on the existing literature, it was expected that couples’ condom use would evolve over time in the relationship and vary across situations and emotional connection, HIV risk perceptions, and sexual agreements evolve. Finally, given the unique risk faced by emerging adult GBM in relationships, it was deemed essential to examine these issues with a special focus on this age-group.

METHODS

Participants and procedures

The 21 couples who participated in this study were drawn from participants in a large online study of same-sex relationships (Starks, Millar, & Parsons, 2015). Participants for the larger study were recruited through a variety of mechanisms involving in-person and online venues focused primarily on reaching GBM in the New York City area (Starks et al., 2015). In-person recruitment activities included attendance by study staff at community and social events frequented by GBM in the New York City area. Online recruitment activities included the distribution of study information via listservs and websites targeting GBM. Online recruitment materials were also sent to partnered men who had completed or were ineligible for participation in other studies conducted at our research center and indicated an interest in being contacted for future studies. Online recruitment materials contained a direct link to a pre-eligibility survey, as well as our contact information. The online survey was designed to accommodate the joint participation of both members of the couple using an “index case” approach. Index participants were those who accessed the study link through any of the recruitment methods described above. After providing personal contact information, index participants were given the option to provide their partners’ contact information and send the study link directly to their partner. If they chose to do so, the survey generated an automatic email, which the participant was allowed to modify prior to sending.

To be eligible for this study, both members of the couple had to complete the online survey, report currently being in a relationship whose duration was at least 3 months, and engage in anal sex together (at least 1 time in the previous 3 months). At least one member of the couple had to be 18-29 years old and both members of the couple needed to reside in the New York City metropolitan area and be able to communicate in oral and written English. Those eligible couples were contacted at a later date, rescreened for eligibility, and then scheduled for an in-person assessment. Of the 21 couples who participated, 4 included one member who was HIV positive. The current analyses focus on data from 17 couples in which both partners identified as HIV negative.

The data drawn upon for this paper are derived from a series of 4 interviews conducted with each couple during a single assessment appointment. This series included an initial joint interview with both members of the couple. Subsequently, each member of the couple was interviewed separately. Finally, the couple was brought back together to complete an exit interview. This structure yielded data from a total of 68 different interviews (17 couples each contributing 4 interviews). These qualitative interviews were conducted during the baseline assessments for this project which occurred between February and September, 2012. Assessments were conducted at a community-based research center in private rooms to ensure confidentiality. All interviews were digitally recorded and transcribedverbatim.

Measures

The semi-structured interview utilized critical incident measures to draw out specific narratives about participants’ sexual experiences with their partner. In general, critical incident approaches ask respondents to describe a specific experience in their life (e.g., the first time or most recent time something occurred) and then subsequently probe for details related to this incident. Critical incident measures reduce recall bias and provide context for behaviors rooted within specific events (Leonard & Ross, 1997). Participants provided descriptive narratives of the first time they had sex together as well the first time they discussed condom use as a couple. Additionally, the couples’ current as well as past history of using and not using condoms within their relationship was explored and discussed. Interviewers probed around patterns and trajectories of condom usage as well as what specific factors and context motivated particular condom use practices.

Analysis

A thematic analysis was utilized in order to generate in depth descriptive understandings of condom usage among gay couples. Thematic analysis is an inductive research method utilized to identify and analyze patterns in the description of particular phenomenon and involves a deep immersion within the data (Boyatzis 1998). Preliminary review of the transcripts produced key themes that were used to develop a coding scheme that focused on discussions of condom usage. Upon establishing a systematized codebook in NVIVO, a team of 3 researchers were trained on coding protocols. This team of coders underwent three rounds of coding an initial interview until > 85% agreement on relevant themes was reached. Throughout this process, inconsistent codes were discussed and appropriate revisions were made. The coding team then applied these codes to the remainder of the interviews and met regularly as a team to discuss coding protocols and aid in the development of analytic frameworks.

The iterative process of engaging in thematic analyses enabled the emergence of patterns from the narratives provided by individuals, and as such the reported results reflect the themes endorsed by multiple respondents. Codified identifiers for the 17 couples are indicated with quotations used. The letter “P1” refers to partner one while “P2” refers to the second partner, and finally the letter “I” refers to the interviewer.

Results

Demographic data are provided in Table 1. At the individual-level, approximately half (52.9%) of the sample indicated a majority-White racial and ethnic identity. Most couples (n_couples = 10; 58.8%) were comprised of one individual who identified as majority-White and another who indicated a racial or ethnic minority identity. Both partners identified as White in 4 couples (23.5%); and 3 couples (17.6%) included members who both indicated a racial or ethnic minority identity. The average age of the sample was 27.3 years (SD = 6.5 years); and most couples (58.2%) had been together for 2 years or more. Most participants were either employed full time (44.1%) or self-employed (35.3%); and most participants (61.8%) reported an annual individual income of less than $30,000.

Table 1.

Demographic Characteristics

	n	%
Race
White	18	52.9
Black/African American	4	11.8
Latino	7	20.6
Other	5	14.7
Employment
Full-time	15	44.1
Part-time	6	17.6
Self-employed	12	35.3
Unemployed	1	2.9
Annual income
Less than $30,000	21	61.8
$30,000 - $49,000	7	20.6
$50,000 or more	6	17.6
Relationship Length
2 years or less	7	41.2
More than 2 years	10	58.2
	M	SD
Age	27.3	6.5

The data that emerged from our respondent’s narratives were organized into three distinct and definable condom use scripts that relate to notions of romance, unanticipated CAI, and negotiated safety. Eleven of our couples drew upon sexual scripts rooted in romantic love when discussing decisions to use condoms during anal intercourse. Ten couples referenced scripts grounded in notions of unanticipated CAI and nine of the couples talked about putting into practice specific sexual agreements strategically aimed at reducing sexual risk.

Romantic Scripts

More than half of the couples discussed CAI as a sign of trust and intimacy. Relationship goals or progression were used to help explain their history of condom usage. More specifically, 12 (57%) couples articulated how they were explicitly looking for a partner to establish an emotional connection with and not just a sexual connection. As such, these partners detailed situations in which the couple waited to have sex in order to develop a more intimate and emotional connection with one another.

I didn’t want to have sex with him until we were more serious. Because I knew it was special and I wasn’t looking for just a sex partner. (2010, P1)

We waited like three weeks. The thing is that for me I wanted to be more serious, so you know, when I was single I used to have sex with people the same day we met, but since I wanted him to be like in a serious relationship with me I wanted to give the impression that it’s not just sex. So I waited. (2024, P2)

These statements are consistent with interpersonal scripts emphasizing romantic love observed by Mutchler (2000). Mutchler observed that the enactment of sexual safety scripts was dependent upon their integration into these types of interpersonal scripts. In the current sample, these instances of integration typically resulted in a script where emotional connection promotes foregoing sexual safety measures. Romantic scripts shaped condom use in two ways. Initially, they tended to delay sex in order to facilitate emotional development of the relationship.

P1: When we started having sex we weren’t using protection because you know, it was like I think that we had developed something that we were saying that we wanted to be intimate with each other. P2: Yeah it wasn’t something we jumped into lightly. We just realized that it was appropriate for where we were and our situation. P1: I would never put myself in the situation of saying I’m gonna have bareback sex, but you know the fact that we started, you know, developing trust between each other. You know where it was like I trust you enough [to not use a condom]. (2016)

Thereafter, foregoing condom use (in some cases at first intercourse) became a behavioral expression of this established emotional connection. The articulation that condoms inhibit couples from fully experiencing these symbolic moments reveals how sexual scripts can shape decisions to use condoms. In fact, one couple’s conception of love and its expression through the act of having sex is something that, at times, prevents them from using condoms. Meanwhile, other couples frame condom use as being a turn off within their relationship due to its perceived ability in preventing the intimacy brought about through fluid exchange.

It makes that connection a little bit deeper when there is no physical barrier. There’s a sharing factor that you don’t get with a condom, especially exchange of bodily fluids in a relationship, like I want that. I want to experience that person as they are. (2031, P2)

Whenever I started really defining this as “we’re a couple” is when we stopped using condoms cause we both fell into what we believed to be a monogamous [relationship]. I think we really trusted each other that we felt comfortable not using condoms. (2015, P1)

Unanticipated CAI

A number of couples indicated condom use that was informed by very sudden and unexpected events, such as getting “caught in the heat of the moment.” This theme was characterized by notions of passion and not wanting to “ruin the mood.” Decisions to use condoms in these circumstances were shaped by a script that defined spontaneous sex as devoid of rational thought and consumed by passion and pleasure. The following interviews demonstrate facets of condom use scripts that reflect notions of spontaneity and passion leading to unanticipated CAI:

P2: Almost out of the blue we can’t know what kind of feeling we’re gonnaget that evening. If I grab it [a condom] or sometimes I’ll be slick if I’m like “okay I just want to do a quickie”.

P1: Yes, but it’s not even a thought process, it’s kind of like just okay this is his mood, I already know what it is he wants so okay I’m gonna go ahead and not use one.

P2: It’s just like we don’t really think about it or have a conversation about it, it just kind of either happens or it doesn’t. (2018)

I feel like when we use a condom now, it’s because we say, “Hey, Should we use a condom?’ ‘Sure.’ And then if it’s an undecided thing, if it’s more just like a feeling thing, then I feel like that’s when we typically don’t use a condom; when we don’t actually specifically talk about it, when it’s a heat-of-the-moment kind of thing. (2040 P2)

Sporadic condom use is an inherent aspect of this script. This variable pattern of condom use highlights the importance of situational factors in deciding whether or not the use of a condom would be appropriate or not for their current sexual situation. Similar to romantic scripts, communication emerges as an important aspect of this script. However, for romantic scripts, the non-use of condoms purposively arose as a demonstration of intimacy and trust, or was directly intended as a mechanism for establishing intimacy and trust. Here, explicit and implicit communication about the desired length of sex and the intensity of physical sensations lead to condom decisions which participants described as largely unplanned and highly variable. Note, Campbell et al., (2014) identified racial and ethnic differences in the implicit vs. explicit nature of couples’ communication and understandings about condom use. The racial and ethnic composition of this sample did not support that kind of between-group analyses; however, the salience of implicit communication with this script indicates that further examination of such differences in samples stratified to support such analyses may yield informative results.

Negotiated Safety Scripts

The final script to have emerged from our couples is typified by navigating a strategic plan whose aim is to reduce the couple’s risk of contracting HIV while engaging in CAI. This strategy was characterized by the cessation of condom use only after HIV tests that confirm the couple’s status as concordantly negative. The integration of HIV testing as a proactive, calculated, and strategic tool into couples’ routine sexual practices is a crucial aspect of this sexual script. In addition to the importance of getting HIV tested, couples also described how establishing a sexual agreement that outlines the boundaries of permissible sexual practices with outside partners is a necessary negotiation that needs to be put into place before engaging in CAI. Variations of these agreements included but were not limited to practicing monogamy or using condoms with outside partners. The following excerpts highlight the details of this strategic script:

Well after we had sex initially we decided to go get tested together. So then when it both came back negative we decided well if we’re gonna be faithful then we’ll still remain negative. I think we really trust each other, and so we know that we’d never put each other at risk. So as long as we have the agreement that we are monogamous and not doing anything then we feel comfortable not using condoms, but if not then we’ll deal with that when we come to it, but that’s pretty much where it was. We decided we were gonna be together after three days, we used condoms, we went and got tested, we’re both negative so going from then on it was fine. (2018)

This quote emphasizes the use of HIV testing and clearly articulated sexual agreements as essential steps that need to be navigated and negotiated prior to engaging in CAI. As such, this sexual script essentially aims to transform potentially high risk sexual acts like CAI into relatively low risk acts. Couples communicated the importance of traversing these specific strategic steps as a responsible way in which to practice safer sex with their partner:

P2: We used condoms until we were tested and knew that we were exclusive.

P1: It wasn’t even an option of not using condoms. I haven’t even ever had a partner who suggested to not have condoms before we were both tested. So we did it [got HIV tested] again and again just to be absolutely sure, but after that three month [window] period then we started to have sex without condoms. (2004)

These couples frame their condom use decisions as being informed by harm reduction scripts that seek to significantly reduce the risks of engaging in sex without a condom within the context of a relationship.

Similarly to romantic condom use scripts, these negotiated safety scripts also frame condomless sex as an act that signifies the seriousness of that relationship. What distinguishes the negotiated safety script is a distinct set of practices that aid in the disclosure of HIV status and the co-construction of a sexual agreement. Just like the act of having sex without a condom in the romantic script represents an act of trust, navigating these risk reduction practices can also be seen as ways in which to develop and communicate trust and commitment within the relationship. These concepts of trust and commitment as navigated within the negotiated safety script were found to be especially important in that it allowed particular partners to feel comfortable engaging in a sexual behavior that is typically defined as deviant and highly risky within both the gay and public health communities.

Fluidity and script salience

Throughout couples’ descriptions of condom use decisions, it was apparent that scripts intersect. Couples condom use decisions were influenced by multiple scripts across time and contexts. The dynamic potential of script salience, and the corresponding variability in condom use, is well-illustrated in the following quote:

P1: I remember that when we weren’t using condoms when we were first together I finally said, “I think we need to be using condoms”. I mean, it just was like, “We’re having unprotected sex, and I’m not supposed to do that and so I think we should”. And I just remember [my partner] being like, “If that’s what you want to do, let’s do that then”, you know, “That’s fine”.

P2: I felt bad for not talking about sex or using a condom those first two times, and then we started using condoms.

P1: And so I remember, we started using condoms more until we got tested together and made sure that we were being safe, and then, after that, we decided that we would stop using condoms again.

P2: So we don’t use condoms right now, but we’re monogamous right now. When we are not as monogamous we use condoms anytime we are messing around with anybody else and then we also used condoms ourselves throughout those times and then when we go back to monogamy we wait three months and get tested and then go back to not using condoms.

P1: It’s [our sexual agreement and condom usage] definitely an evolving aspect of our relationship

P2: Yeah it evolves with us. (2015)

This story depicts how condom usage within a relationship may be prone to fluctuations as each partner draws from different scripts to help inform their sexual expectations and behavior. For this couple, condom use practices are a result of an ongoing evolutionary process. Their reliance on a particular script is not only rooted in certain contextual cues but is also co-constructed and established over time through a process of communication and mutual understanding.

Conclusions

These results provide insight into the content of sexual scripts for GBM couples in the age range of emerging adulthood. Themes of romance, unanticipated CAI, and negotiated safety emerged as salient. Notably, these results highlight an emphasis on emotional factors versus HIV prevention concerns; the relevance of sexual agreements; and the role of communication in condom use decisions. They also illustrate the potential for variability in condom use inherent in some scripts as well as that resulting from fluctuations in script salience across contexts.

Scripts varied in the extent to which they emphasized emotional factors versus HIV prevention. The romantic script observed in the current data prioritized a subset of emotional factors related to closeness. Consistent with previous research (Blechner, 2002; Mutchler, 2000; Shernoff, 2005), this script emphasized that condom cessation may serve to establish a sense of emotional closeness and/or serve as a concrete expression of that felt emotion. Similarly, Goldenberg et al. (2015) observed that GB men’s perceptions of HIV risk and sexual risk taking behavior covaried with love, intimacy and trust. In the unanticipated CAI script, emotional factors centered on “passion” or the intensity of physical desire “in the moment.” Goldenberg et al. (2015) conceptualized these feelings under the term “Lust.” They noted that, while lust did not diminish perceptions of HIV risk (as did love, intimacy and trust), Lust was related to sexual risk taking behavior through an increased willingness to overlook perceived risk and engage in CAI despite it. In contrast, the negotiated safety script de-emphasized emotional decision making. Changes in condom use were portrayed as driven by HIV-related communication, including: HIV status disclosure, HIV testing, and the establishment of sexual agreements in a manner that minimized risk of HIV infection.

Sexual agreements also emerged as salient across identified scripts. Consistent with CIT formulations of sexual health as a joint goal (Lewis et al., 2006), the role of sexual agreements in the observed negotiated safety script was consistent with generally recognized negotiated safety procedures (Davidovich et al., 2000; Guzman et al., 2005; Kippax et al., 1993). Couples in this sample elected to either proceed with a monogamous agreement or to use condoms during anal sex with any partners outside the relationship. While not salient in content from these interviews, others have found evidence suggesting that sexual agreements may impact the ongoing success of negotiated safety. Individuals in open agreements (who achieve negotiated safety through a commitment to condom use with outside partners) are more likely to engage in regular HIV testing than those in monogamous agreements (Stephenson, White, Darbes, Hoff, & Sullivan, 2015), which serves to insure the ongoing success of negotiated safety by confirming partners’ HIV-negative status.

Mention of sexual agreements cut across observed scripts, suggesting they have a potentially important emotional significance that goes beyond their use as an HIV prevention strategy. Couples often discussed monogamy along with condom cessation as an indicator of emotional connection. Even in the context of the negotiated safety script – where the theme of HIV prevention was more salient than emotional factors – some couples framed their agreement about sex with outside partners as a mechanism for expressing and maintaining the unique emotional commitment they shared as a couple. This finding is highly consistent with observations by Hoff and colleagues who found that relational motivations, (e.g., expressing/enhancing love and trust or providing structure and meaning) were commonly reported and often ranked as more important than HIV prevention motivations (Hoff & Beougher, 2010; Hoff et al., 2010).

Finally, scripts varied with respect to the deliberate versus spontaneous nature of communication. In both romantic and negotiated safety scripts, couples described a clear communication goal. In the former, the goal was to communicate love, trust, and emotional connection. In the latter, the goal was to communicate information necessary for HIV risk reduction. Notably, the presence of a clear intention did not always lead to the use of explicit communication strategies, which involve a direct conversation between partners (Campbell, 2014). In fact, romantic scripts often framed CAI as an implicit, non-verbal method of communicating love and trust consistent with qualitative findings linking emotional closeness to HIV-related risk perceptions and sexual behavior (Goldenberg et al., 2015) and hypotheses generated from quantitative studies of condom use beliefs (Golub et al., 2012; Starks, Payton, Golub, Weinberger, & Parsons, 2014). In contrast, communication goals in the unanticipated CAI script were less focused. Couples often relied on indirect or implicit (Campbell, 2014) means of communicating a desire to have sex. Deliberate and direct communication was often trumped in this script by an emphasis on “passion,” “pleasure,” or the “heat of the moment.” The result was a variable pattern of condom use in which CAI “just happened” without a strong sense of intention. As mentioned above, the available data did not support a detailed examination of racial and ethnic difference in communication; however, these findings, in conjunction with Campbell et al’s (2014) study indicate that such work may be useful.

Results indicated that condom use may vary for couples as a consequence of both the content of these scripts and from the fluid adoption of scripts across contexts. Each sexual script contained some elements which would lead to the occurrence of CAI over time. For romantic and negotiated safety scripts, this evolution involves the progression to CAI as emotional closeness is established or HIV risk reduction procedures are followed. In the unanticipated CAI script, condom use fluctuates constantly across time as priorities such as time and intensity vary across each sexual encounter. In addition to the potential for fluidity within each of these scripts, the salience of particular scripts varied across contexts. All concordant HIV-negative couples indicated more than one script. The script most salient at the moment was determined by a confluence of factors including the couples’ development, individual priorities, and messages communicated between partners.

While the themes observed in their condom use scripts are not necessarily unique to emerging adult GB in relationships, they have implications for interventions targeting this highly vulnerable population. Perhaps counter-intuitively, content within the romantic script indicated that “waiting to have sex” was not necessarily a protective behavior. Ironically, the enhanced sense of emotional closeness cultivated during this waiting process actually inhibited condom use for these couples. This has important implications for interventions which emphasize abstinence until particular relationship milestones (such as marriage) have been achieved, which have been examined in primarily heterosexual samples of youth (e.g., Sherr & Dyer, 2010). This act of “waiting until the relationship is serious” may actually reduce the use of condoms as a risk reduction strategy for GBM couples when/if initial sex occurs.

These results suggest that interventions focused on enhancing direct communication may be more appropriate for GBM emerging adults. Identifying alternative mechanisms for communicating and establishing love, trust and emotional closeness may diminish the salience of romantic scripts which lead to CAI. Developing clear and explicit strategies for communicating about sex together may serve to slow down “in the heat of the moment” decision making which is part of the unanticipated CAI script in a manner that enhances sexual safety practices. Couples HIV Testing and Counseling(Sullivan, White, et al., 2014) provides a straightforward example of an intervention which utilizes direct communication to reduce HIV risk. Inherent in the CHTC protocol is a guided explicit discussion about the couples’ shared HIV prevention goals as well as their sexual agreement(Sullivan, Stephenson, et al., 2014).

These findings suggest that researchers studying the sexual health of gay couples as well as clinicians working with gay couples should be open to the possibility that couples may use sexual agreements to achieve a number of goals, including but not limited to HIV risk reduction. Such a finding does not preclude a regulatory role for sexual agreements in couples risk reduction plans; however, it does mean that clinicians should be aware that many couples may think of their sexual agreement as primarily a mechanism for communication closeness as opposed to serving the primary purpose of reducing HIV risk.

These findings must be understood in light of several limitations. First, just over half of this sample identified as majority-White and 10 of the 16 racial or ethnic minority identified individuals were in relationships with a majority-White partner. This composition limited our ability to examine racial and ethnic differences in thematic content. Campbell et al. (2014) illustrated meaningful associations between the racial and ethnic composition of couples and their communication about condom use. While the current study provides an initial look at the narratives gay couples construct about their use of condoms across time in their relationship, future studies should utilize stratified samples that would permit an examination of racial and ethnic differences in these narratives. Second, the sample was limited to couples in which at least one member was between the ages of 18-29 and reported an HIV negative sero-status. While these limitations were reasonable given the study’s focus on condom use as an HIV prevention strategy for young GBM in relationships, these findings may not apply to older couples or couples both members are HIV positive. Seeing as Romantic, unanticipated CAI, and negotiated safety scripts, in which some CAI occurred, were only analyzed among our concordant HIV-negative couples, important distinctions may exist among sero-discordant couples and merits additional investigation. Third, future studies should include a more detailed focus on substance use. Qualitative research focused on the role of alcohol has identified specific alcohol-related scripts associated with unprotected sex (Parsons et al., 2004). Future studies should examine differences in script content and enactment between couples who do and do not use substances. Finally, the current study did not examine the use of or attitudes towards pre-exposure prophylaxis (PrEP). PrEP has been shown to reduce the HIV transmission risk of CAI(CDC, 2014; Grant et al., 2010; Thigpen et al., 2012). Despite this limitation, condom use remains a behavior with significant public health implications. Condomless sex has the potential to transmit a wide range of sexually transmitted infections (STIs) other than HIV, even when taking PrEP (Baeten et al., 2012; Grant et al., 2010; Thigpen et al., 2012; Volk et al., 2015). Furthermore, understanding scripts related to condom use may meaningfully inform PrEP targeting. Intimacy motivations for CAI have been linked to PrEP receptivity among partnered MSM (Gamarel & Golub, 2015). Future studies should examine the role of PrEP in the content and enactment of condom use scripts.

Despite these limitations, the present study provides information about the content of condom use scripts for young HIV-negative GB couples. Findings suggest that couples’ narratives around condom use may be organized along a continuum from emotion-focused to risk-reduction focused. Furthermore, results highlight ways in which script content and fluid adoption of different scripts may lead to situational variability in condom use across time within couples. Finally, these results highlight the prominent role of sexual agreements in condom use scripts across the spectrum of focus from emotion to risk-reduction.