Disagreements about agreements – many gay couples lack clarity on sex outside their relationship

This article is more than 9 years old. Click here for more recent articles on this topic

The majority of American gay men in relationships say they establish a ‘sexual agreement’ with their partner, both to minimise HIV risks and to maintain the quality of their relationship, according to research published online ahead of print by the Journal of Sex Research last month.

However, partners do not always agree on whether they have an agreement, on whether it was explicitly discussed, or on what sex is allowed with other people. And the agreement had been broken by one or both partners in just under half the couples studied.

The study has also found that around a quarter of the HIV-negative men who have casual sex attempt to 'serosort' or use 'strategic positioning' when doing so. However, regular HIV testing was far from universal in this group, making such practices potentially unreliable. 



A serodifferent couple is one in which one partner has HIV and the other has not. 

strategic positioning

Another term for seropositioning.

unprotected anal intercourse (UAI)

In relation to sex, a term previously used to describe sex without condoms. However, we now know that protection from HIV can be achieved by taking PrEP or the HIV-positive partner having an undetectable viral load, without condoms being required. The term has fallen out of favour due to its ambiguity.


Studies aim to give information that will be applicable to a large group of people (e.g. adults with diagnosed HIV in the UK). Because it is impractical to conduct a study with such a large group, only a sub-group (a sample) takes part in a study. This isn’t a problem as long as the characteristics of the sample are similar to those of the wider group (e.g. in terms of age, gender, CD4 count and years since diagnosis).


Choosing sexual partners of the same HIV status, or restricting condomless sex to partners of the same HIV status. As a risk reduction strategy, the drawback for HIV-negative people is that they can only be certain of their HIV status when they last took a test, whereas HIV-positive people can be confident they know their status

“From a public health perspective, we need to encourage gay couples to have more honest and explicit discussions when establishing and honouring sexual agreements,” commented Dr Jason Mitchell of the University of Michigan, who conducted the research. “Sexual agreements are not only advantageous from a prevention standpoint for couples, but the agreement can also help strengthen their relationship.”

A sexual agreement is made between two individuals, and concerns what sexual behaviour may occur within and outside their relationship. Some agreements may simply clarify that sex is not permitted with other people. Other agreements may concern the relationship being ‘open’ to a lesser or greater extent. Many couples use agreements, in part, to limit their risk of HIV infection. 

A couples study

Jason Mitchell’s study recruited both partners in couples and compared their responses, rather than relying on just one partner’s perspective. Men were recruited to this cross-sectional online survey via adverts on Facebook in 2011 – the advertising was targeted to reach individuals whose profile information indicated that they might be a man in a relationship with a man. The first participants provided the email address of their partners, who were then contacted and asked to complete the survey.

The study recruited 722 men (361 couples). This was a predominantly white (77%) and relatively well-educated sample, with a mean age of 33, all living in the United States. Couples had been together for an average of an average of five years; three-quarters were living together.

One in eight men had been diagnosed with HIV. Half of them were in a relationship with another man with HIV, and half with an HIV-negative man (i.e. in a 'serodifferent' or ‘serodiscordant’ relationship).

Most men reported that they were HIV negative. (Just 3% of the sample reported that they did not know their HIV status.) However, one in five of the ‘HIV-negative’ men had not tested since the beginning of their relationship, and on average, HIV-negative men had last taken an HIV test two years ago. There was considerable diversity in the men’s HIV testing practices, with a minority testing much more frequently than others.

On average, men said that they had discussed their HIV status 12 days into the relationship, and this occurred before the couple started having unprotected sex (a mean of 81 days into the relationship). However, when the couple made a sexual agreement, this typically occurred much later – 174 days into the relationship.  


Although seven in ten men reported having a sexual agreement, this perspective was not always shared by the man’s partner. In 57% of couples both men agreed that they had a sexual agreement, but in 25% of couples, one man thought there was an agreement, while his partner said that there wasn’t.

Amongst those couples who agreed that they had agreed, 58% said that there had been an explicit discussion, while 11% reported that their understanding was more implied or assumed. In a further 31% of cases, one partner had thought they had had an explicit discussion, while his partner thought that it was implicit.

For 56% of men who thought they had an agreement, it was that the relationship was monogamous.

For a further 41%, the agreement was to permit sex with casual partners, but with some rules or guidelines. For the last 3%, there was an open relationship, without any conditions.

Responses detailing what was permitted for those with open relationship guidelines showed that around a quarter actually permitted unprotected anal sex with casual partners. While the data shows that some couples had different rules for receptive and insertive sex, and for withdrawal before ejaculation, respondents were not asked whether there were conditions based on the partner’s HIV status and seroadaptive behaviours.

Far more couples allowed oral sex and masturbation.

Open relationship guidelines were not just about the risk of infections, but also about intimacy and context. Having sex with a casual partner on more than one occasion, physically sleeping together and dating were all permitted by half or less of couples, whereas threesomes or group sex were allowed by 81%.

Men’s motivations for making agreements were not just about minimising the risk of HIV or sexually transmitted infections, although this was the most common primary reason, cited by 23%. Other important reasons included wanting monogamy or exclusivity; fulfilling sexual desires; establishing guidelines so as to manage expectations; and building and maintaining trust.

Jason Mitchell did not analyse motivations by type of agreement, but in a separate study, another group of researchers previously found that the themes of trust, honesty and strengthening the relationship were predominant in the motivations both of men with open agreements and men with monogamy agreements.

Breaking agreements

Looking at couples, including those with an agreement to monogamy, in 46% of cases, either one or both partners had broken the rules at some point during the relationship.

But in the previous three months, 80% of couples had stuck to their rules. In 15% of the couples, one partner had broken the agreement, and in 5% both partners had.

The main reasons for breaking agreements were sexual frustration and the ‘heat of the moment’. Only a minority of men (30%) told their partner that they had broken the agreement. Reasons given for not disclosing included not giving the partner a reason not to trust the respondent and fearing that this could lead to the relationship ending.

Risk reduction strategies

In a separate article, published in AIDS & Behavior in December, Jason Mitchell has also reported on the risk-reduction strategies employed by this group of men, both inside and outside of their primary relationships.

He was interested in strategies such as serosorting, strategic positioning, 100% condom use and taking undetectable viral load into account.

This analysis is especially interesting because Mitchell actually asked respondents whether they had used strategies to reduce their risk of HIV or sexually transmitted infections. In contrast, most other studies on this topic have examined the pattern of men’s sexual practices with partners of different HIV statuses, and attempted to infer whether there was a strategy in place or not.

It therefore hasn’t been clear from previous research whether the conscious and deliberate use of strategies such as strategic positioning is something widely practiced by gay men, or just a minority pursuit.

For this couple-based analysis, those couples in which both partners had diagnosed HIV were excluded. Participants were asked about strategies used in the last previous three months – they could name more than one strategy, either because strategies were combined, or because different strategies were used in different situations. 

Always using condoms for anal sex, or always doing so with an HIV-positive partner, was reported by a minority of men. Within the main relationship, 15% of HIV-negative couples (i.e. in which both partners thought they were negative) always used condoms, rising to 38% of serodifferent couples (i.e. in which one partner had HIV and the other did not).

When having sex outside the relationship, 38% of men always used a condom for anal sex.

Only having oral sex and never anal sex was reported, for the main relationship, by 23% of HIV-negative couples and 31% of serodifferent couples. This strategy was more commonly reported for sex with casual partners – by 51% of men.

'Serosorting' (having unprotected anal intercourse [UAI] with a partner because he was thought to have the same HIV status) was reported by 66% of HIV-negative couples. Moreover, this was also reported for sex with casual partners, by 27% of men.

‘Strategic positioning’ (only having UAI with the HIV-positive partner in the receptive role) was reported, for the main relationship, by 32% of serodifferent couples. During casual sex, it was also reported by 23% of men.

Having unprotected sex because the HIV-positive partner was either taking HIV treatment or had an undetectable viral load was reported by 24% of serodifferent couples.

It was much less commonly employed as a strategy with casual partners – by 1% of men in an HIV-negative couple and 14% of men in a serodifferent couple.

Finally, it’s important to note that a significant proportion of men did not have a risk-reduction strategy at all, especially with their main partner. One of the answers men could choose was ‘‘regardless of HIV-status, we never use condoms and ejaculate inside’’.

For their primary relationship, 24% of HIV-negative couples and 22% of serodifferent couples chose this answer. It was also chosen by 9% of men having sex outside the relationship.

Overall, having unprotected sex within the main relationship (in the last three months) was reported by 87% of HIV-negative couples and 69% of serodifferent couples. Moreover, for 16% of couples, there had been unprotected sex both with the main partner and at least one casual partner during the same time frame.

Improving the quality of agreements

Clearly, with a number of couples permitting unprotected anal sex outside of the relationship and with a significant minority of men breaking agreements in one way or another, there are risks that HIV can be brought into relationships.

The research highlights a number of other limitations of some men’s sexual agreements, in terms of the frequency of HIV testing, the discrepancy in partners’ perceptions of whether there is an agreement and what it entails, and the quality of communication following breaks in an agreement.

This suggests that HIV-prevention interventions which support men in strengthening their relationships and in making better agreements would be warranted.

One such approach that is being piloted is couples voluntary counselling and testing, in which couples take an HIV test and receive the results together. The focus of the counselling is not on past sexual history, but on how the couple wish to manage the risk of HIV in the future. Quantitative and qualitative research suggests that this is an intervention that many men in relationships would be interested in.


Mitchell JW Characteristics and Allowed Behaviors of Gay Male Couples' Sexual Agreements. Journal of Sex Research, online ahead of print, 2013. (Abstract here)

Mitchell JW, Petroll AE Patterns of HIV and sexually transmitted infection testing among men who have sex with men couples in the United States. Sexually Transmitted Diseases 39: 871-876, 2012. (Abstract here)

Mitchell JW HIV-Negative and HIV-Discordant Gay Male Couples' Use of HIV Risk-Reduction Strategies: Differences by Partner Type and Couples' HIV-Status. AIDS & Behavior, online ahead of print, 2012. (Abstract here)