A US study on heterosexual couples at high risk of HIV finds that individuals’ interpretation of their relationship, their reporting of behaviour within the relationship and, especially, their recollection of whether they’ve disclosed sexual risks differ substantially between men and women.
Dr Kathy Hageman of the US Centers for Disease Control told the Tenth AIDS Impact conference that, even in situations where the same proportion of men and women in the study agreed that a particular behaviour (such as condom use, anal sex or domestic violence) had happened, only about 50% of these behaviours were reported by both partners in the relationship.
This is one of the first studies attempting to quantify the degree of over- or under-reporting of sexual risks and other behaviours in couples,
The study was the Heterosexual Partner Study (HPS), a substudy of the high-risk heterosexual section of the National HIV Behavioral Surveillance System (NHBSS), a series of annual surveys that are conducted in the US annually in various high-prevalence areas. Surveys are carried out in different locations amongst men who have sex with men, injecting drug users and heterosexuals at high risk of HIV; another study from the NHBSS presented at AIDS Impact documented awareness of pre-exposure prophylaxis in gay men.
The HPS had an innovative design. It recruited 855 women from the African-American and Hispanic communities. They had to be involved in at least one relationship with a man in the last year and to have had sex with that partner within the last three months. After completing the NHBSS and HPS questionnaires and being tested for HIV, they were asked to bring in at least one male partner to the research centre. The men then answered the same HPS questionnaire as the women.
The women brought in 926 male partners, with 71 women (7.6%) bringing in two partners [NB these figures are different from the ones reported in the conference abstract]. Eighteen per cent of the women were Hispanic and the rest black; 13% of male partners were Hispanic, 81% black and 11% of other ethnicity. The average length of couples’ relationships was three years. The average age of the women and the men was 33 and 35 respectively, but 37% of couples featured an age difference of five years or more (‘cross-generational’ sex is a known risk factor for HIV).
The male partners had many characteristics both of impoverished urban populations in the USA and of high HIV risk. No less than three-quarters of the men had been in jail at some time. Sixty-two per cent reported sexual partners outside the relationship, 31% crack use, 15% injecting drugs and 30% had been diagnosed with a sexually transmitted infection (STI). Two-thirds had ever taken an HIV test.
Women’s awareness of these factors varied a lot. If the man had been in jail, 77% of their female partners knew it; but only 43% knew whether their partner had taken an HIV test and only 27% if he had had an STI. In general, couples talked more freely about drugs than sex.
The study compared women’s and men’s responses on a number of behaviours. It then looked at whether the male and female partners agreed whether a specific behaviour did or did not happen, or whether there was disagreement about whether it had happened.
Consistent condom use, for instance, was uncommon. Ten per cent of women and 9% of men reported using condoms consistently in vaginal sex. However even fewer, only 4%, of male/female pairs jointly agreed that they used condoms consistently. While 85% of couples agreed that they did not use them consistently, this left 12% of coupleswho did not agree on whether they used condoms consistently or not.
Similarly, 20% per cent of women and 25% of men said that they had had anal sex, but only 12% of couples jointly agreed they had. Two-thirds agreed that had not had anal sex, leaving 22% of couples in disagreement over whether they had or not (consistent condom use, by the way, was no more common in anal than vaginal sex).
The biggest disparity was over domestic violence – which was common. Twenty-eight per cent of both women and men agreed that it had occurred within the relationship. But only 13% of pairs jointly agreed that it had happened, while 57% agreed it had not. This meant that 29% of couples disagreed on whether (to use the questionnaire’s wording) the man had ever physically hurt the woman.
There was even less agreement on whether the couples had discussed specific STI and HIV risks. For instance, 22% of couples said yes, they had discussed whether the male partner had other lovers and 34% said no, they hadn’t discussed it, but 44% disagreed on whether the subject had been raised. Half the women and 45% of the men recalled discussing HIV status within the relationship, but only 27% of pairs agreed that they had discussed it, 36% agreed they hadn’t discussed in, and 43% couldn’t agree on whether they’d discussed it or not.
Men were consistently less likely to report that they had discussed a specific risk factor than women, but the gender difference varied according to the stigma attached to the subject. For instance approximately similar numbers of men and women recalled talking about HIV test results, which may be regarded as a responsible thing to do. But while 40% of women said they’d discussed whether their man had ever had sex with another man, only 16% of men said they had - and only 9% of couples agreed jointly that they had.
Dr Hageman commented on the wide disparity seen between individual and couple responses in the study, and said that efforts to address HIV prevention at the couple level had not been utilised as effectively as individual-level prevention efforts.
“Investigating couple agreement is an important step to understanding how partner-level dynamics impact HIV-related risk behavior and prevention efforts,” she said.
Hageman K et al. Couple agreement of HIV-related behaviors, communication, and knowledge: Heterosexual Partner Study, 16 U.S. cities, 2006-2007. Tenth AIDS Impact conference, Santa Fe, New Mexico. Abstract 87. 2011.
