Aspects of Gay Male Couples’ Sexual Agreements Vary by Their Relationship Length

Abstract

Few HIV preventing interventions exist for gay male couples in the U.S., yet estimates indicate that between one- and two-thirds of U.S. men who have sex with men acquire HIV while in a primary relationship (e.g., gay male couples). In response to these statistics, research has been conducted to better understand gay male couples’ relationships, including their sexual agreements. Many gay male couples in the U.S. establish a sexual agreement, which is an explicit mutual understanding between two main partners about what sexual and other behaviors they agree to engage in and with whom while in the relationship. Although some research about sexual agreements has been conducted, little is known on whether aspects of gay male couples’ sexual agreements (e.g., establishment, type, and adherence) vary as a function of their relationship length. The present study aimed to fulfill this gap of knowledge, which may lead to a better understanding of how agreements can be used for developing HIV/STI prevention interventions. A national, cross-sectional, Internet-based study was used to collect dyadic data from 361 US gay male couples. Men in each couple completed the questionnaire independently. All analyses were employed at the couple-level. Our findings showed that the longer the couples had been in their relationship, the more likely they would concur about having a sexual agreement. As relationship length increased, the proportion of couples who disagreed about their current agreement type increased. No direct trend was found for recent adherence to an agreement; however, the likelihood of ever breaking an agreement increased as relationship length increased. Findings from this study indicate there is a need to help gay male couples’ establish and maintain their agreements, particularly for those who are in their early formative stages, as well as, for those who are experiencing challenges and/or changes in their relationships.

Estimates from two recent studies indicate that between one-third and two-thirds of gay men and other men who have sex with men (MSM) in the U.S. acquire HIV while in a same-sex relationship (e.g., gay male couples) (Goodreau et al., 2012; Sullivan, Salazar, Buchbinder, & Sanchez, 2009). Rates of new HIV infections among gay male couples may be in part due to their low levels of condom use, greater likelihood of having anal sex, and greater frequency of unprotected anal intercourse (UAI), thereby increasing their cumulative risk for HIV infection over time (Sullivan et al., 2009). In addition, low rates of HIV testing (Chakravarty, Hoff, Neilands, & Darbes, 2012; Mitchell & Petroll, 2012a; Mitchell & Petroll, 2012b), discomfort in communicating about sex (Prestage, Mao, McGuigan, Crawford, Kippax, & Grulich, 2006), and failure to disclose or discuss HIV status may also increase couples’ risk for HIV (Davidovich, de Wit, & Stroebe, 2000; Davidovich, de Wit, & Stoebe, 2004; Elford, Bolding, Maguire, & Sherr, 1999; Sullivan et al., 2009). Unprotected anal intercourse with outside partners may also increase couples’ HIV risk, particularly when UAI is practiced in the relationship and their HIV status is concordantly negative (Davidovich et al., 2000, 2004; Kippax, Crawford, Davis, Rodden, & Dowsett, 1993; Kippax et al., 1997). Additionally, couples’ racial/ethnic background may also influence their HIV risk and prevention efforts. One study with HIV-negative Latino gay male couples found that maintaining knowledgeable about HIV, being exposed to social support groups for Latino gay men, and finding support in their relationship from their main partner had helped lower the couples’ risk for HIV (Beougher, Gomez, & Hoff, 2011).

In response to these findings, researchers have reported that gay male couples’ relationship characteristics must be assessed for development of HIV prevention interventions (Burton, Darbes, & Operario, 2010; El-Bassel et al., 2010; Herbst et al., 2007; Grossman et al., 2011). Thus, a growing body of research has emerged to examine how characteristics of gay male couples’ relationships, including sexual agreements affect their sexual health and risk for HIV. A sexual agreement is an explicit mutual understanding between two main partners about what sexual and other behaviors they agree to engage in and with whom while in the relationship (Hoff & Beougher, 2010; Mitchell, 2013). One specific type of agreement is called ‘Negotiated Safety’. Negotiated safety aims to reduce the possibility of HIV transmission between main partners within a concordant HIV-negative relationship by requiring both partners to know their HIV status, be able to make a reliable agreement about anal intercourse and condom use both inside the relationship and if relevant, with any sexual contacts they might have with other men (Kippax et al., 1997; Prestage et al., 2006). However, studies have found that gay male couples form different types of sexual agreements, and establishing an agreement within the relationship appears to be a fairly common practice (Gass, Hoff, Stephenson, & Sullivan, 2012; Gomez et al., 2012; Hoff & Beougher, 2008; Hoff, Beougher, Chakravarty, Darbes, & Neilands, 2010; Hoff, Chakravarty, Beougher, Neilands, & Darbes, 2012; Hoff et al., 2009; LaSala, 2004a; LaSala, 2004b; Mitchell, 2013; Mitchell, Harvey, Champeau, Moskowitz, & Seal, 2012; Mitchell, Harvey, Champeau, & Seal, 2012; Mitchell & Petroll, 2013; Parsons, Starks, DuBois, Grov, & Golub, 2013; Parsons, Starks, Gamarel, & Grov, 2012; Prestage et al., 2008; Wheldon & Pathak, 2010). For instance, some couples form closed agreements (i.e., behavioral monogamy) while others form open agreements that allows one or both partners to have sex with others, either with or without guidelines (Hoff & Beougher, 2008).

By nature, sexual agreements have direct implications toward couples’ sexual health and risk for HIV/STIs. To better understand agreements, recent studies have assessed what motivates one or both men of the couple to establish an agreement (e.g., promote trust, reduce risk for HIV/STIs) (Gass et al., 2012; Hoff et al., 2010; Mitchell, 2013) and the behaviors that couples allow per their type of agreement, such as to allow UAI within the relationship and/or group play (Mitchell, 2013). Additionally, one recent study noted that some HIV-discordant gay male couples form agreements that allow an acceptable level of HIV risk (as determined by the couple), which may or may not involve using condoms within the relationship (Beougher, Chakravarty, Garcia, Darbes, Neilands, & Hoff, 2012).

Why men break their agreement (e.g., sexually frustrated) (Gass et al., 2012; Hoff et al., 2010; Mitchell, 2013) and whether they disclose having broken the agreement (Hoff et al., 2009; Hoff et al., 2010; Mitchell, 2013) (e.g., fear of relationship dissolution) has also been assessed. Studies have also found that certain characteristics are associated with gay male couples who have broken their sexual agreement, including partners who have had UAI outside of the relationship, and/or reporting lower levels of trust, relationship commitment, and investment in the agreement (Gomez et al., 2012; Mitchell et al., 2012).

Although our understanding of agreements has advanced through these studies, other questions about agreements remain unanswered and warrant further investigation. For instance, research has yet to assess how aspects of couples’ agreements (e.g., establishment, type, adherence) may differ according to their relationship length. Some gay male couples may form an agreement at different stages or time points of their relationship, which is important to consider for development and promotion of prevention programs. The present exploratory study sought to provide a richer understanding of how aspects of agreements (e.g., establishment, current type, recent adherence, ever broken agreement, and change in agreement type since establishment) may differ according to the couples’ relationship length.

Method

Participants and Recruitment

Recruitment for the present study was conducted through Facebook ® banner advertising during a ten-week period in 2011. A total of 7,994 Facebook users clicked on at least one of the advertisements and were then directed to the study webpage. Among those who visited our study webpage, 4,056 (51%) potential participants answered our eligibility questions. Of these 4,056 potential participants, 731 MSM (18%) were deemed ineligible; 1,529 MSM (38%) were eligible, enrolled, but failed to complete at least 80% of the survey items; 1,796 MSM (44%) were eligible, enrolled and had completed the survey. Of the 1,796 MSM, only dyadic data from 361 male couples were included in this subsample of participants. The remaining partners of the 1,074 MSM either did not enroll or complete the survey.

The study webpage briefly described the purpose of the study, what a participant could expect if he chose to participate (e.g., be asked to invite his main, male relationship partner to also participate in the study), and asked eligibility questions. Eligible participants were asked to complete an electronic consent form. Consenting participants advanced to take the 30–40 minute confidential survey, which was hosted on a secure access portal. We embedded a partner referral system in our survey to facilitate dyadic data collection. Every fifth couple that completed the survey received two modest incentives via email (e.g., $20 e-gift card per partner). The BLINDED Institutional Review Board approved the study protocol. The methodology used in this study has previously been reported in detail (blinded refs).

Measures

A variety of measures were used to assess couples’ demographic and relationship characteristics, and engagement of UAI by partner type. Participants were asked to self-report their and their primary partner’s HIV serostatus.

Relationship characteristics assessed included relationship length, cohabitation length, and aspects about the sexual agreement. Aspects of a sexual agreement were the primary variables of interest, and included the following categorical items: establishment of, original and current type of, recent adherence to (i.e., within prior three months), and ever broken, the agreement. Specifically, participants were asked whether they had established a sexual agreement (yes or no) with their main partner, the type of sexual agreement that was initially established and their current type of sexual agreement. Both initial and current type of sexual agreement items were assessed categorically with the following response options, “We only have sex with each other and no one else”, “We have sex with each other, and we are allowed to have sex with others under certain guidelines/rules” and “We have sex with each other, and are allowed to have sex with others without any guidelines/rules”. Additional items about adherence and non-adherence to the sexual agreement assessed whether a participant and/or his main partner had kept or broken their sexual agreement ever and within the 3 months prior to assessment. For the aspects of agreement outcome variables (e.g., formation, original and current type, recent adherence, ever adherence), we compared responses between both partners of the couple and constructed corresponding couple-level variables. For example, comparison of both partners’ responses to “establishment of a sexual agreement” was constructed into a couple-level variable to indicate whether “both partners reported yes”, “partners disagreed about having a sexual agreement”, and “both partners reported no”. A similar approach was used for creating couple-level outcome variables of current agreement type, adherence – last 3 months, broken agreement – ever, and changed agreement type since established (original type vs. current type).

Other details about the sample’s sexual agreements, testing behaviors, attitudes toward use of couples-based HIV testing, and use of risk-reduction strategies have been previously reported (blinded).

Data Analysis

Dyadic data from 361 male couples (722 individuals) were analyzed using Stata v12 (StataCorp, College Station, TX). Several dummy couple-level variables were constructed to characterize the couples by comparing how similar or different each partner’s self-report was for a given variable (e.g., race, UAI, education). Descriptive statistics including means, standard deviations, rates, and percentages were calculated, as appropriate, for the measures. Differences in aspects of gay male couples’ sexual agreements by relationship length were calculated by using tests of associations, including Fisher’s exact and Pearson’s Chi-square; only couple-level variables were examined for these analyses.

Results

Characteristics of the Sample

The majority of male couples (N=361) self-identified as gay (97%), concordantly HIV-negative (76%), White (66%), and living in an urban or suburban environment (89%). Many of them also had one or both partners who were employed (94%), had health insurance (88%), and/or obtained at least a Bachelor’s degree (68%). The mean age for individuals was 33.0 years (SD 10.8).

Most couples practiced UAI within their relationship (84%), and approximately one-third of couples (N = 113, 31%) had one or both men who reported having had sex with a casual MSM partner within the three months prior to assessment. Among these 113 couples, 66% (N=75) had one or both men who engaged in UAI with a casual MSM partner during this same timeframe.

On average, couples’ relationship length was 4.9 years (SD 5.5). Ten percent of couples (N = 35) had been in their relationship for less than 6 months, 33% (N = 118) between 6 months and 2 years, 27% (N = 96) between 2 and 5 years, 15% (N = 57) between 5 and 10 years, and 15% of couples (N = 55) had been in their current relationship for more than 10 years. Among those who lived together (75%), their mean cohabitation length was 5.2 years (SD 5.8).

Over half of couples (57%, N=207) had both partners who concurred about establishing a sexual agreement, and of these couples, 84% (N=174) had both partners who concurred about their current type of agreement and 80% (N=140) who concurred about adhering to their agreement within the last three months. Characteristics of the sample are presented in Table 1.

Table 1.

Sociodemographic and Behavioral Characteristics of 361 Male Couples

Couple-level characteristic	N	%
Sexual orientation
Both men in couple identified as gay	349	97
One or both partners in couple identified as bisexual	12	3
Race
Mixed	124	34
White	237	66
Education: Had a Bachelor’s degree or higher
Both partners	134	37
Only one partner	110	31
Neither partner	117	32
Employment status
Both partners employed	235	65
Only one partner employed	104	29
Neither partner employed	22	6
Had health insurance at time of assessment
Both partners reported yes	227	63
Only one partner reported yes	91	25
Both partners reported no	43	12
HIV serostatus
Concordant negative	275	76
Concordant positive	28	8
Discordant	58	16
Practiced unprotected anal intercourse (UAI) within relationship	304	84
One or both men in couple had sex outside their relationship	113	31
One or both men in couple had UAI with a casual MSM partner a	75	66
Establishment of a sexual agreement
Couple concurred about having an agreement	207	57
Couple disagreed about having an agreement	92	25
Couple did not have an agreement	62	17
Current type of sexual agreement b
Closed agreement	92	44
Open agreement	82	40
Kept sexual agreement within prior three months to assessment
Both partners in couple kept agreement	166	80
Only one partner in couple kept agreement	31	15
Both partners broke their agreement	10	5
Ever broken sexual agreement
Both partners reported yes
Only one partner reported yes
Both partners reported no

	Mean	SD
Individual age [range: 18–68 years]	33.0	10.8
Relationship length [range: 0.25–35 years]	4.9	5.5
Cohabitation duration [range: .08–31.2 years]	5.2	5.8

Note:

^aRepresents the proportion of couples who had one or both men having had sex outside of their relationship (e.g., 75 of the 113 couples).

^bRepresents couples with both partners who concurred about having a sexual agreement (N = 207 dyads) as well as the same type of agreement.

Aspects of Couples’ Sexual Agreements Differ by Relationship Length

Differences in aspects of sexual agreements were noted according to relationship length. For formation of a sexual agreement, the proportion of couples who concurred about forming a sexual agreement appeared to be a function of their relationship length. Specifically, the longer the couples had been in their relationship, the more likely they would be on the “same page” about having a sexual agreement and less likely to disagree or report not having one (χ²(8) = 17.6; p < .05). In addition, compared to couples of other relationship lengths, a higher proportion of couples who had been together 6 months or less had disagreed about whether they had an agreement. In contrast, the opposite trend was noted between couples’ relationship length and current agreement type. As relationship length increased, the proportion of couples who disagreed about their current agreement type also increased (χ²(4) = 27.0; p < .001). For recent adherence to the agreement (< 3 months), fewer couples who had been together less than 6 months and between 2 and 5 years had kept their current agreement (χ²(4) = 16.7; p < .01). However, whether couples had ever broken their agreement appeared to have a negative trend as their relationship length increased. For instance, the longer the couples had been in their relationship, the more likely that one or both partners had reported they had ever broken their agreement (χ²(4) = 37.2; p < .001). Whether couples concurred about changing their agreement type also seemed to differ according to their relationship length. As relationship length increased, the proportion of couples who had either both or one partner who reported having changed the agreement type also increased (χ²(8) = 44.2; p < .001). Table 2 provides additional information about these results.

Table 2.

Differences in Aspects of Gay Male Couples’ Sexual Agreements, by Relationship length

Aspects of sexual agreement	Relationship length
	< 6 m	6 m–2 yr	2–5 yr	5–10 yr	> 10 yr
	% (N)	% (N)	% (N)	% (N)	% (N)	p
Formation/establishment, N=361 dyads	N=35	N=118	N=96	N=57	N=55	< .05
Both partners reported yes, n=207 dyads	54 (19)	55 (65)	53 (51)	58 (33)	71 (39)
Partners disagreed, n=92 dyads	34 (12)	24 (28)	30 (29)	23 (13)	18 (10)
Both partners reported no, n=62 dyads	11 (4)	22 (25)	17 (16)	19 (11)	11 (6)
Current agreement type, N=207 dyads a	N=19	N=65	N=51	N=33	N=39	< .001
Both partners reported same type, n=174 dyads	100 (19)	91 (59)	86 (44)	73 (24)	72 (28)
Partners disagreed about type, n=33 dyads	0 (0)	9 (6)	14 (7)	27 (9)	28 (11)
Kept agreement last 3 months, N=174 dyads b	N=19	N=59	N=45	N=24	N=28	< .01
Both partners reported yes (kept), n=140 dyads	68 (13)	83 (49)	87 (39)	63 (15)	86 (24)
One or both partners reported no, n=34 dyads	32 (6)	17 (10)	13 (5)	37 (9)	14 (4)
Ever broken agreement, N=207 dyads a	N=19	N=65	N=51	N=33	N=39	< .001
Both partners reported no (kept), n=112 dyads	74 (14)	69 (45)	53 (27)	39 (13)	33 (13)
One or both partners reported yes, n=95 dyads	26 (5)	31 (20)	47 (24)	61 (20)	67 (26)
Changed agreement type since formed, N=207 a, c	N=19	N=65	N=51	N=33	N=39	< .001
Both partners reported a change in type, n=19	11 (2)	2 (1)	14 (7)	9 (3)	15 (6)
One partner reported a change in type, n=44	0 (0)	18 (12)	16 (8)	30 (10)	36 (14)
Both partners reported no change in type, n=144	89 (17)	80 (52)	70 (36)	61 (20)	49 (19)

Note:

^aType of current sexual agreement, broke agreement – ever, and changed agreement type since formed original agreement only includes data reported by couples who concurred about forming a sexual agreement within their relationship (N=207 dyads).

^bRecent adherence to the agreement (e.g., kept agreement last 3 months) only includes data from couples who concurred about having the same type of agreement (N=174 dyads).

^cData described in ‘changed agreement type since originally formed the sexual agreement’ does not indicate that partners within the couple concurred about the type of agreement that they changed from originally to their current agreement.

Discussion

Findings from our study describe how aspects of sexual agreements differ by relationship length. Important trends were noted, and have implications toward development and targeting of prevention programs for gay male couples. As relationship length increased, the proportion of couples who concurred about forming an agreement increased, yet the proportion of couples who concurred about their current agreement type decreased. Dynamics within the couples’ relationship may have changed over time, which could have affected their understanding of the agreement type. Partners’ communication and perceptions about what is allowed to occur sexually by the agreement may have shifted as the couples’ relationship length increased. For example, some couples may have had sexual experiences that were uncharacteristic of their current agreement (e.g., closed agreement yet they had a threesome) and did not discuss how this experience could affect their understanding and perception of the agreement. Prior research with partnered gay men in Australia noted that almost half of the men reported some discomfort discussing with their main partner about sex that occurred outside of the relationship (Prestage et al., 2006). Future prevention strategies must consider how couples’ agreements are impacted over the life course of their relationship (e.g., changes in dynamics between partners).

Couples recent adherence to their sexual agreements was also associated with relationship length. Specifically, more breaks were reported by couples who had been together less than 6 months and among those who had been together between 2 and 5 years. The couples who just started their relationship may have not had the opportunity to fully discuss the details of their agreements, including what behaviors may or may not be allowed to occur. For the couples who have been together for more than 2 years but less than 5 years, some couples may broken their agreement due to myriad of reasons including not having their sexual needs met within the relationship to spur-of-the-moment, unplanned situational contexts. Previous studies have reported that often the reasons why men break their sexual agreements are situational in nature (Hoff et al., 2009; Mitchell, 2013); this may be the case for this group of couples. However, other research has noted that partnered gay men who reported having higher levels of certain relationship characteristics, such as trust, communication, commitment, and social support were less likely to report breaking their agreement with their main partners (Gomez et al., 2012). Additional research, such as individual- and couple-level qualitative interviews, would provide a better understanding about how and which motivational and situational factors (e.g., social support, prior relationship/agreement experiences) may influence gay male couples to adhere to their sexual agreements at different time periods of their relationship (e.g., after being together for two years).

Limitations and Suggestions for Future Research

The use of a cross-sectional study design with a convenience sample precludes us from making causal inferences and generalizing our findings to all U.S. gay male couples. Although we did not collect identifying information, participation, social desirability, and recall biases may have influenced participants to inaccurately report information about their relationships. Additionally, participants may have completed the survey with their main partners, despite our request for them to complete it independently and separately, and therefore potentially causing some bias. These biases, if present, may have inflated the number of couples who established and adhered to an agreement. Some participants may have also calculated when their relationship began differently (e.g., when first met vs. used the words “boyfriends/partner” vs. got engaged/married). Moreover, age difference between partners and the possible presence of intimate partner violence may exist and influence couples’ agreements. To address these potential limitations, future Internet-based studies with gay male couples could require each partner to schedule an online video-based appointment to monitor the participant while he completes the questionnaire. The main strengths of our study include the large geographically diverse sample size of Internet-using gay male couples, the use of paid targeted social media advertisements to capture a large sample size in a short period of time, the use of dyadic data, and ascertaining how aspects of agreements vary by relationship length.

Services that help gay male couples establish and maintain their sexual agreements may be particularly important for minimizing their risk for HIV/STIs. Our study provides support that such services are needed for gay male couples during their early formative stages of the relationship, and throughout the relationship because couples – of all ages – may experience challenges and/or changes within their relationship. Future studies that aim to develop interventions for gay male couples should also consider the importance of communication about sexual health and whether couples use other prevention methods in addition to a sexual agreement. Our suggestions for future studies about gay male couples’ sexual agreements may be best accomplished by using a mixed-method, longitudinal study design with dyadic data collection. These advances in research will help develop future HIV prevention programs for gay male couples.