Diagnoses of sexually transmissible infections in HIV-serodiscordant male couples in Australia, Brazil and Thailand
James Gray
A * , Fengyi Jin A , Nittaya Phanuphak B , Ruth K. Friedman C , Christopher K. Fairley D E , David J. Templeton A F G , Iryna Zablotska-Manos H I , Jennifer Hoy J , David Baker K , Mayara Secco Torres Silva C , Andrew Grulich A , Benjamin Bavinton A and A
B
C
D
E
F
G
H
I
J
K
Abstract
In an era of biomedical prevention of HIV transmission, prevention of sexually transmissible infections (STIs) remains a priority for male HIV-serodiscordant couples. Three common STIs in populations of gay and bisexual men are chlamydia, gonorrhoea and syphilis.
We examined the incidence of these three infections in 343 male HIV-serodiscordant couples in Australia, Brazil and Thailand. The couples participated in a prospective, observational cohort study that included regular STI testing and behavioural surveys for HIV-negative partners (HNP).
HIV-positive partners (HPP; IR = 22.9/100 person-years, 95% CI: 19.3–27.0) had a higher incidence rate of any STI (22.9/100 person-years, 95% CI: 19.3–27.0) compared with HNP (14.9/100 person-years, 95% CI: 12.1–18.3). It was more common for only the HPP (n = 107/343, 7.1%) or only the HNP (n = 60, 4.0%) to have an STI than for both to have an STI (n = 29/343, 1.9%). Higher STI incidence was associated with vocational education (HNP: aHR = 2.19, P = 0.005), full-time employment (HPP: aHR = 1.61, P = 0.016), living in Brazil (HPP: aHR = 1.85, P = 0.017) or Thailand (HPP: aHR = 1.98, P = 0.008), having anal sex with other partners (HNP: aHR = 2.42, P = 0.003; HPP: aHR = 2.87, P < 0.001) and recent PrEP use (HPP: aHR = 1.51, P = 0.044). Lower STI incidence was associated with age >40 years (HNP: aHR = 0.49, P = 0.021; HPP: aHR = 0.46, P = 0.006), full-time employment (HNP: aHR = 0.56, P = 0.016) and >5 years since first sex with the study partner (HNP: aHR = 0.50, P = 0.036).
There were differences in STI diagnoses within HIV-serodiscordant couples, particularly based on their agreements about sex outside the relationship. Interventions aimed at strengthening these agreements with commitments to STI testing could be effective in reducing the incidence of STIs.
Keywords: agreements, chlamydia, gonorrhoea, HIV, serodiscordant, STD, STI, syphilis.
Introduction
Responding to sexually transmissible infections (STIs) among sexually active gay and bisexual men (GBM) is a global health priority.1 In this population, HIV has received the larger focus due to its greater severity in terms of health and social stigma, in comparison with other STIs. STIs are known to increase HIV transmission risk in untreated people with HIV and their partners;2 however, improved access to treatment as prevention (TasP)3–5 and pre-exposure prophylaxis (PrEP)6 has reduced this risk.
Among the curable STIs, the most common in GBM are chlamydia, gonorrhoea and syphilis. In three major Australian cities in 2021–2022, the self-reported 12-month diagnosis of chlamydia in GBM was between 9.3% and 13.1%, gonorrhoea between 6.1% and 11.3%, and syphilis between 3.8% and 4.3%.7–9 In Brazil in 2017, a study of dating app users reported an STI prevalence of 11% in the previous year, with the most common being syphilis (3.1%), herpes (1.5%) and gonorrhoea (1.4%).10 Among GBM screened by nucleic acid amplification testing at community-led clinics in Bangkok, Pattaya, Chiang Mai and Songkhla in 2015–2016, 21.7% had chlamydia and 15.5% had gonorrhoea at any anatomical site.11
Comprehensive STI testing has been normalised in Australia, where testing for STIs among gay sexually active men with multiple partners has been recommended every 3 months.12 Condoms are effective at reducing the risk of STI transmission when used correctly,13 although issues with breakage and slippage have been reported in this population.14 Consistent condom use is declining within gay communities, as both TasP and PrEP provide people with new ways to reduce HIV transmission risk.15 These recent changes in HIV prevention efforts to focus on biomedical interventions have coincided with increases in STI transmission, although these increases began prior to those approaches being formally adopted.16
Limited data are available on the experience of STIs in GBM in HIV-serodiscordant relationships, defined as couples where one partner is HIV-positive and the other is HIV-negative. The literature exploring these relationships focuses on the dynamics of HIV transmission risk, with STIs only being considered as a potential function of that risk. Many GBM serodiscordant relationships have agreements that allow for sex outside the relationship,17 and even where agreements for monogamy exist, behaviour does not always align. These relationships are an important context for STI transmission due to their potential to increase the risk of HIV transmission,18 although STIs do not appear to decrease the effectiveness of TasP.19 Previous research in this group has demonstrated that agreements for condomless anal intercourse increase within these relationships over time, for sex both within the relationship, as well as with others.17
STIs cause harm for individuals and have implications for the provision of health care to GBM in HIV-serodiscordant couples. As TasP and PrEP reduce the likelihood of HIV transmission, more attention should be focused on reducing STI transmission and its associated harms. The aim of this analysis was to estimate incidence rates of diagnoses of three common, curable bacterial STIs in both partners in gay male HIV-serodiscordant relationships, including the impact of relationship agreements and the congruence of diagnosis between the two partners.
Materials and methods
Participants
Participants were cisgender GBM in the Opposites Attract study, for which the design and methods have previously been published.20 Briefly, data were collected as part of a prospective, observational cohort study of HIV-serodiscordant male couples recruited through high HIV caseload clinics, sexual health clinics, and hospitals in Australia, Brazil and Thailand. The primary goal of the study was to determine the impact of antiretroviral therapy on the prevention of HIV transmission among gay and bisexual men within relationships.4
To be eligible to participate, both men in the couple, defined here as study partners, had to be aged at least 18 years, partner be a man with HIV (HPP) and the other partner HIV-negative (HNP) at baseline, have anal sex with each other at least once a month on average, and agree to attend clinic visits at least twice a year. Enrolment occurred between 2012 and 2014, with follow up through to the end of 2016. Couples were followed up until the end of the study (n = 230) or until they became ineligible (n = 72), withdrew or were lost to follow up (n = 41). Among those that became ineligible, 60 (83%) ceased within-couple anal intercourse entirely or broke up, 10 (14%) reported anal intercourse less than once per month on average and two (3%) died in separate couples.
Procedures
Couples completed online computer-assisted self-interview questionnaires that collected behavioural and attitudinal information, and were aligned to the time of each study visit. The questionnaires for the two partners differed, with less detail about sexual agreements and acts related to condomless anal intercourse asked of the HPP compared with the HNP due the potential impact of public health legislation in place at the time of the study in Australia. The questionnaires were available in English, Brazilian Portuguese and Thai. Clinical data were collected via electronic case report forms. Ethics approvals were obtained in all three countries.
Both study partners were screened for common bacterial STIs at each study visit. Different testing technology was used in different locations, and results of both culture testing and nucleic acid amplification tests for gonorrhoea have been combined in the diagnosis variable. The results of rectal and urethral testing are analysed in this paper. Due to resourcing constraints, participants in Brazil and Thailand were not screened for STIs in the throat; therefore, this site has not been included for analysis. If positive results were detected, participants were contacted for interim visits and prescribed antimicrobial treatment according to local guidelines. Additional STI testing performed outside regular study visits was not considered for this analysis.
Measures
Demographic characteristics including age, education and ethnicity were collected from both partners, as well as information about the length of time since they first had sex with each other and whether they lived together.
Participants were asked whether they had been diagnosed with chlamydia, gonorrhoea and/or syphilis at any time in the past 3 months. For each STI, participants were asked about the infection location (anus, penis, throat, any other body site and/or diagnosed by blood test). STIs were diagnosed via STI screening during their study visits. For the purpose of this paper, the term ‘any STI’ refers to incident infections of chlamydia, gonorrhoea and syphilis. In cases where it was unclear based on test results whether a syphilis test results was due to previous infection or an active infection, these were classified based on clinical assessment by the treating clinician in consultation with the study team.
The HNP was asked about their clear, spoken agreements with their study partner, about sex with other men, including what this involved (may not have sex at all, may not have anal sex, may only have sex with condoms, may have sex without condoms, other). The HNP was asked about sexual behaviour with others in the previous 3 months, whether they had anal intercourse with other men, and how many times they had anal intercourse with other partners (never, once, twice, 3–5 times, 6–10 times, 11–30 times, 31–50 times, >50 times) including both insertive and receptive anal intercourse. They were also asked how many times they had anal intercourse with their study partner or other partners (never, once, twice, 3–5 times, 6–10 times, 11–30 times, 31–50 times, >50 times). The HPP was not asked these questions due to legal risks related to phylogenetic testing conducted as part of the broader study.
Several key variables were constructed from the items above. This included the diagnosis of any STI, the diagnosis of one of either gonorrhoea or chlamydia in any location, and combined diagnosis by location for both chlamydia and gonorrhoea.
Statistical analysis
Data were analysed using Stata 15.1 (Stata Corporation, College Station, Texas, USA). Couples were excluded if they attended the baseline visit, but had no follow-up visits (n = 13), resulting in 343 couples being included in the final analysis. Descriptive analyses of STI diagnoses were conducted by STI type, relationship agreement type, country and whether one or both study partners had an infection. Incidence rates (IR) of STIs were calculated as the total number of STIs diagnosed divided by the total years of follow up, and presented as the number of events per 100 person-years. For this calculation, follow up started from 2012 and ended in 2016 to allow multiple diagnoses in the same individuals across follow up. When STI testing was not conducted at a study visit, these visits were categorised as not having an STI diagnosis.
Bivariable and multivariable generalised linear models using Cox regression were used to examine associations over follow up between a range of variables and STI diagnosis. The variables were selected based on associations found in previous research21–23 and initial exploratory analyses within this data set. All variables significant in the bivariable analysis (P < 0.05) were block entered. The strengths of the associations were presented as incidence rate ratios (IRR) along with their corresponding 95% confidence intervals (95% CI) and P-values. These analyses were stratified by each of the three countries to examine differences.
Results
A total of 343 couples provided data during follow up, for a total of 686 participants. Demographic and baseline behavioural characteristics of the participants have previously been published.4
At baseline, 39 (11.4%) of HNPs were diagnosed with an STI, whereas 46 (13.4%) of HPPs had an STI (Table 1). Twenty (5.3%) HNPs and 26 (7.6%) HPPs were diagnosed with chlamydia, eight (2.3%) HNPs and 10 (2.9%) HPPs were diagnosed with gonorrhoea, and 14 (4.1%) HNPs and 17 (5.0%) HPPs were diagnosed with infectious syphilis. Within the total cohort of 343 couples, there were differences in baseline prevalence of STIs in Australia (153 couples), Brazil (93 couples) and Thailand (97 couples). The differences were found for any STI diagnosis for the HNPs (P = 0.003) and HPPs (P = 0.013), for chlamydia diagnoses for the HNPs (P = 0.003) and HPPs (P < 0.001), and for infectious syphilis for the HNPs (P = 0.001) and HPPs (P = 0.002). No differences were found between countries for gonorrhoea diagnoses for either partner.
| Total (n = 343) | Australia (n = 153) | Brazil (n = 93) | Thailand (n = 97) | Chi-squared, P-value | ||
|---|---|---|---|---|---|---|
| HIV-negative partner | ||||||
| Any STIA | 39 (11.4%) | 9 (5.9%) | 16 (17.2%) | 14 (14.4%) | 8.616, P = 0.003 | |
| Chlamydia | 20 (5.3%) | 3 (2.0%) | 5 (5.4%) | 12 (12.4%) | 11.765, P = 0.003 | |
| Gonorrhoea | 8 (2.3%) | 5 (3.3%) | 2 (2.2%) | 1 (1.0%) | 1.323, P = 0.516 | |
| Infectious syphilis | 14 (4.1%) | 2 (1.3%) | 10 (10.8%) | 2 (2.1%) | 14.590, P = 0.001 | |
| HIV-positive partner | ||||||
| Any STIA | 46 (13.4%) | 12 (7.8%) | 15 (16.1%) | 19 (19.6%) | 7.863, P = 0.013 | |
| Chlamydia | 26 (7.6%) | 6 (3.9%) | 4 (4.3%) | 16 (16.5%) | 15.354, P < 0.001 | |
| Gonorrhoea | 10 (2.9%) | 4 (2.6%) | 2 (2.2%) | 4 (4.1%) | 0.742, P = 0.690 | |
| Infectious syphilis | 17 (5.0%) | 3 (2.0%) | 11 (11.8%) | 3 (3.1%) | 12.952, P = 0.002 | |
Across follow up, 89 (24.8%) HNPs and 115 (33.5%) HPPs were diagnosed with at least one STI. Incident rates of overall STI diagnosis were analysed for all participants, by country (Table 2). Across all countries, HPPs had a higher incidence (22.9/100 person-years, 95% CI: 19.3–27.0) of any STI compared with HNPs (14.9/100 person-years, 95% CI: 12.1–18.3). The HPPs also had a higher incidence of chlamydia (12.5/100 person-years, 95% CI: 9.9–15.7 vs 6.5/100 person-years, 95% CI: 4.8–8.9) and syphilis (8.5/100 person-years, 95% CI: 6.47–11.2 vs 5.2/100 person-years, 95% CI: 3.6–7.4), but not for gonorrhoea (4.2/100 person-years, 95% CI: 2.8–6.2 vs 4.3/100 person-years, 95% CI: 3.0–6.4).
| Total (n = 343) | Australia (n = 153) | Brazil (n = 93) | Thailand (n = 97) | ||||||
|---|---|---|---|---|---|---|---|---|---|
| HIV-negative partner | HIV-positive partner | HIV-negative partner | HIV-positive partner | HIV-negative partner | HIV-positive partner | HIV-negative partner | HIV-positive partner | ||
| Any STIA diagnosis | 14.9 (12.1–18.3) | 22.9 (19.3–27.0) | 15.6 (11.9–20.5) | 17.8 (13.8–22.9) | 13.7 (8.8–21.2) | 24.6 (17.8–34.1) | 14.2 (8.8–22.7) | 34.9 (25.8–47.2) | |
| Chlamydia | |||||||||
| Any site | 6.5 (4.8–8.9) | 12.5 (9.9–15.7) | 7.5 (5.1–11.1) | 10.8 (7.8–15.0) | Nil | 7.5 (4.2–13.6) | 11.6 (6.9–19.6) | 23.3 (16.1–33.7) | |
| Rectal | 4.2 (2.8–6.2) | 10.0 (7.8–12.9) | 5.1 (3.2–8.2) | 7.5 (5.1–11.1) | Nil | 6.2 (3.2–11.8) | 6.6 (3.3–13.3) | 21.6 (14.7–31.7) | |
| Urethral | 2.7 (1.6–4.4) | 3.0 (1.9–4.8) | 2.4 (1.2–4.8) | 4.2 (2.5–7.1) | Nil | 1.4 (0.3–5.5) | 6.6 (3.3–13.3) | 1.7 (0.4–6.6) | |
| Gonorrhoea | |||||||||
| Any site | 4.3 (3.0–6.4) | 4.2 (2.8–6.2) | 6.0 (3.9–9.3) | 3.0 (1.6–5.6) | 2.1 (0.7–6.4) | 4.1 (1.8–9.1) | 2.5 (0.8–7.7) | 7.5 (3.9–14.4) | |
| Rectal | 3.5 (2.3–5.4) | 3.3 (2.2–5.2) | 5.1 (3.2–8.2) | 2.4 (1.2–4.8) | 1.4 (0.3–5.5) | 2.1 (0.8–6.4) | 1.7 (0.4–6.6) | 7.5 (3.9–14.4) | |
| Urethral | 1.5 (0.8–2.9) | 1.0 (0.5–2.2) | 1.5 (0.6–3.6) | 0.6 (0.2–2.4) | 1.4 (0.3–5.5) | 2.1 (0.8–6.4) | 1.7 (0.4–6.6) | 0.8 (0.1–5.9) | |
| Infectious syphilis | 5.2 (3.6–7.4) | 8.5 (6.47–11.2) | 3.3 (1.8–6.0) | 6.0 (3.9–9.3) | 11.6 (7.2–18.7) | 13.0 (8.3–20.4) | 2.5 (0.8–7.7) | 10.0 (5.7–17.6) | |
In Australia, the incidence was similar in HPPs (17.8/100 person-years, 95% CI: 13.8–22.9) and HNPs (15.6/100 person-years, 95% CI: 11.9–20.5) for overall STI diagnoses, and for chlamydia (10.8/100 person-years, 95% CI: 7.8–15.0 vs 7.5/100 person-years, 95% CI: 5.1–11.1), gonorrhoea (3.0/100 person-years, 95% CI: 1.6–5.6 vs 6.0/100 person-years, 95% CI: 3.9–9.3) and syphilis (6.0/100 person-years, 95% CI: 3.9–9.3 vs 3.3/100 person-years, 95% CI: 1.8–6.0).
In Brazil, STI incidence was higher in the HPP (24.6/100 person-years, 95% CI:17.8–34.1) than in the HNP (13.7/100 person-years, 95% CI: 8.8–21.2). Chlamydia incidence in the HPP was 7.5/100 person-years (95% CI: 402–13.6), whereas there were no diagnoses in among the HPPs. Incidence rates were similar between partners for gonorrhoea (4.1/100 person-years, 95% CI: 1.8–9.1 vs 2.1/100 person-years, 95% CI: 0.7–6.4) and syphilis (13.0/100 person-years, 95% CI: 8.3–20.4 vs 11.6/100 person-years, 95% CI: 7.2–18.7).
In Thailand, the incidence rate of STI diagnosis was higher in HPPs (34.9/100 person-years, 95% CI: 25.8–47.2) compared with HNPs (14.1/100 person-years, 95% CI: 8.8–22.7). Chlamydia incidence in HPPs was marginally higher (23.3/100 person-years, 95% CI: 16.1–33.7 vs 11.6/100 person-years, 95% CI: 6.9–19.6). Gonorrhoea incidence in HPPs was higher (7.5/100 person-years, 95% CI: 3.9–14.4 vs 2.5/100 person-years, 95% CI: 0.8–7.7). Syphilis incidence for HNPs was higher (10.0/100 person-years, 95% CI: 5.7–17.6 vs 2.5/100 person-years, 95% CI: 0.8–7.7).
Across follow up, there were 1317 (87.1%) study visits when couples had congruent negative STI tests, and 29 (1.9%) when couples were both diagnosed with any STI (Table 3). Of the remaining study visits, 60 (4.0%) HNPs and 107 (7.1%) HPPs were diagnosed with an STI when their study partner was not. By country, Thailand had a lower level of congruence (79.9% both no STI, 3.2% both with STI) compared with Australia (89.5% both no STI, 1.6% both with STI) and Brazil (85.8% both no STI, 1.7% both with STI). Congruence stratified by infection type is also presented in Table 3.
| Any STIA | Chlamydia | Gonorrhoea | Infectious syphilis | ||||||
|---|---|---|---|---|---|---|---|---|---|
| HPP No Dx | HPP Yes Dx | HPP No Dx | HPP Yes Dx | HPP No Dx | HPP Yes Dx | HPP No Dx | HPP Yes Dx | ||
| All participants | |||||||||
| HNP No Dx | 1317 (87.1%) | 107 (7.1%) | 1410 (93.2%) | 64 (4.2%) | 1468 (97.0%) | 19 (1.3%) | 1439 (95.1%) | 43 (2.8%) | |
| HNP Yes Dx | 60 (4.0%) | 29 (1.9%) | 28 (1.6%) | 11 (0.7%) | 20 (1.3%) | 6 (0.4%) | 24 (1.6%) | 7 (0.5%) | |
| Australia | |||||||||
| HNP No Dx | 817 (89.5%) | 44 (4.8%) | 860 (94.2%) | 28 (3.1%) | 866 (97.0%) | 7 (0.8%) | 883 (96.7%) | 19 (2.1%) | |
| HNP Yes Dx | 37 (4.1%) | 15 (1.6%) | 17 (1.9%) | 8 (0.9%) | 17 (1.9%) | 3 (0.3%) | 10 (1.1%) | 1 (0.1%) | |
| Brazil | |||||||||
| HNP No Dx | 301 (85.8%) | 30 (8.6%) | 340 (96.9%) | 11 (1.4%) | 343 (97.7%) | 5 (1.4%) | 318 (90.6%) | 16 (4.6%) | |
| HNP Yes Dx | 14 (4.0%) | 6 (1.7%) | 0 (0.0%) | 0 (0.0%) | 2 (0.6%) | 1 (0.3%) | 14 (4.0%) | 3 (0.9%) | |
| Thailand | |||||||||
| HNP No Dx | 199 (79.9%) | 33 (13.3%) | 210 (84.3%) | 225 (10.0%) | 239 (96.0%) | 7 (2.8%) | 238 (95.6%) | 8 (3.2%) | |
| HNP Yes Dx | 9 (3.6%) | 8 (3.2%) | 11 (4.4%) | 3 (1.2%) | 1 (0.4%) | 2 (0.8%) | 0 (0.0%) | 3 (1.2%) | |
HNPs: In bivariable analysis, the following factors (Table 4) were associated with increased STI incidence: having vocational level education compared with having tertiary education, having an agreement to have sex outside the relationship with condoms, having anal sex with other partners and the HIV negative partner having used PrEP in the previous 3 months. Reduced risk of STI diagnosis was associated with age >40 years, being in full-time employment and >5 years since having first sex with the study partner. In multivariable analysis, adjusting for all variables that were significant in the bivariate analysis, the increased risk of STI diagnosis was associated with having vocational education and having anal sex with other partners. Reduced risk of STI diagnosis was associated with age >40 years, being in full-time employment and >5 years since having first sex with the study partner.
| Person-years | STI diagnosesA | Rate per 100 PY | Any STI – bivariable (HR, 95% CI) | P-value | Any STI – multivariable (aHR, 95% CI) | P-value | ||
|---|---|---|---|---|---|---|---|---|
| Country | ||||||||
| Australia | 332.3 | 52 | 15.6 | 1 | ||||
| Brazil | 146.3 | 20 | 13.6 | 0.93 (0.54–1.60) | 0.801 | |||
| Thailand | 119.9 | 17 | 14.2 | 0.97 (0.55–1.73) | 0.919 | |||
| Age of HIV-negative partner | ||||||||
| <30 years | 144.7 | 32 | 22.1 | 1 | 1 | |||
| 30–39 years | 188.0 | 35 | 13.4 | 0.77 (0.47–1.26) | 0.295 | 1.06 (0.61–1.83) | 0.681 | |
| ≥40 | 262.5 | 22 | 8.4 | 0.33 (0.19–0.58) | <0.001 | 0.49 (0.27–0.90) | 0.021 | |
| Education | ||||||||
| High school or less | 154.0 | 24 | 15.6 | 1 | 1 | |||
| Vocational | 106.2 | 32 | 30.1 | 1.90 (1.12–3.23) | 0.017 | 2.19 (1.27–3.77) | 0.005 | |
| University | 305.5 | 31 | 10.1 | 0.63 (0.37–1.08) | 0.093 | 0.70 (0.41–1.22) | 0.214 | |
| Employed full-time | ||||||||
| No | 214.9 | 46 | 21.4 | 1 | 1 | |||
| Yes | 351.4 | 41 | 11.7 | 0.53 (0.35–0.82) | 0.004 | 0.56 (0.35–0.90) | 0.016 | |
| Live together full-time | ||||||||
| No | 149.7 | 30 | 20.0 | 1 | ||||
| Yes | 415.0 | 57 | 13.7 | 0.66 (0.42–1.03) | 0.070 | |||
| First sex within the couple | ||||||||
| <12 months | 216.4 | 42 | 19.4 | 1 | 1 | |||
| 1–5 years | 185.6 | 30 | 16.2 | 0.84 (0.52–1.33) | 0.459 | 0.73 (0.45–1.19) | 0.206 | |
| ≥ 5 years | 164.2 | 15 | 9.1 | 0.46 (0.26–0.85) | 0.012 | 0.50 (0.27–0.96) | 0.036 | |
| HNP perceived HPP to have detectable viral load | ||||||||
| No | 400.0 | 65 | 16.2 | 1 | ||||
| Yes | 167.1 | 22 | 13.1 | 0.85 (0.52–1.40) | 0.522 | |||
| Agreement to have CLAI inside relationship | ||||||||
| No | 252.2 | 41 | 16.3 | 1 | ||||
| Yes | 189.6 | 26 | 13.7 | 0.83 (0.72–1.36) | 0.453 | |||
| Having CLAI with study partner | ||||||||
| No | 246.0 | 38 | 15.4 | 1 | ||||
| Yes | 320.0 | 49 | 15.3 | 0.98 (0.64–1.51) | 0.943 | |||
| Agreement to about sex outside relationship | ||||||||
| No sex | 355.3 | 44 | 12.4 | 1 | 1 | |||
| Yes – condoms | 157.5 | 32 | 20.3 | 1.65 (1.04–2.60) | 0.033 | 1.32 (0.81–2.14) | 0.265 | |
| Yes – CLAI | 51.5 | 11 | 21.4 | 1.57 (0.81–3.09) | 0.182 | 1.18 (0.58–2.42) | 0.647 | |
| Anal sex with other partners | ||||||||
| No | 222.7 | 17 | 7.6 | 1 | 1 | |||
| Yes | 375.7 | 72 | 19.1 | 2.54 (1.50–4.31) | 0.001 | 2.42 (1.35–4.34) | 0.003 | |
| HNP PrEP use in prev 3 months | ||||||||
| No | 430.9 | 58 | 13.5 | 1 | 1 | |||
| Yes | 135.3 | 29 | 21.4 | 1.61 (1.03–2.52) | 0.038 | 1.37 (0.86–2.20) | 0.179 | |
HPPs: In bivariable analysis, (Table 5) the following factors were associated with increased risk of STI diagnosis: living in Thailand, being in full-time employment, 1–5 years since having first sex with the study partner, having anal sex with other partners, and the HNP having used PrEP within the previous 3 months. Age >40 years was associated with decreased risk of STI diagnosis. In multivariable analysis, adjusting for all variables that were significant in the bivariate analysis, increased risk of STI diagnosis was associated with living in Brazil or Thailand, being employed full-time, having anal sex with other partners and the HNP having used PrEP within the previous 3 months. Reduced risk of STI diagnosis was associated with being aged >40 years.
| Person-years | STI diagnosesA | Rate per 100 PY | Any STI – bivariable (HR, 95% CI) | P-value | Any STI – multivariable (aHR, 95% CI) | P-value | ||
|---|---|---|---|---|---|---|---|---|
| Country | ||||||||
| Australia | 329.5 | 59 | 17.9 | 1 | ||||
| Brazil | 145.6 | 36 | 24.7 | 1.49 (0.96–2.30) | 0.072 | 1.85 (1.11–3.05) | 0.017 | |
| Thailand | 120.3 | 41 | 34.1 | 2.25 (1.46–3.47) | <0.001 | 1.98 (1.20–3.27) | 0.008 | |
| Age of HIV-negative partner | ||||||||
| <30 years | 138.3 | 46 | 33.3 | 1 | 1 | |||
| 30–39 years | 186.4 | 46 | 24.7 | 0.72 (0.48–1.09) | 0.118 | 0.67 (0.42–1.08) | 0.098 | |
| ≥40 years | 270.7 | 44 | 16.3 | 0.44 (0.28–0.67) | <0.001 | 0.46 (0.26–0.80) | 0.006 | |
| Education | ||||||||
| High school or less | 156.9 | 39 | 24.9 | 1 | ||||
| Vocational | 137.6 | 35 | 25.4 | 1.03 (0.65–1.63) | 0.900 | |||
| University | 264.5 | 54 | 20.4 | 0.82 (0.54–1.23) | 0.333 | |||
| Employed full-time | ||||||||
| No | 266.9 | 47 | 17.6 | 1 | ||||
| Yes | 294.44 | 81 | 27.3 | 1.55 (1.08–2.22) | 0.017 | 1.61 (1.09–2.36) | 0.016 | |
| Live together full-time | ||||||||
| No | 155.3 | 40 | 1 | |||||
| Yes | 419.3 | 94 | 0.83 (0.57–1.21) | 0.342 | ||||
| First sex within the couple | ||||||||
| <12 months | 207.8 | 39 | 18.8 | 1 | ||||
| 1–5 years | 173.2 | 52 | 30.0 | 1.58 (1.04–2.40) | 0.032 | 1.49 (0.97–2.30) | 0.069 | |
| ≥5 years | 161.5 | 34 | 21.1 | 1.11 (0.70–1.76) | 0.669 | 1.36 (0.81–2.28) | 0.242 | |
| HNP perceived HPP to have detectable viral load | ||||||||
| No | 396.5 | 82 | 20.7 | 1 | ||||
| Yes | 167.7 | 46 | 27.4 | 1.44 (0.99–2.09) | 0.057 | |||
| Agreement to have CLAI inside relationship | ||||||||
| No | 251.3 | 59 | 23.5 | 1 | ||||
| Yes | 188.2 | 39 | 20.7 | 0.86 (0.57–1.30) | 0.474 | |||
| Having CLAI with study partner | ||||||||
| No | 244.9 | 60 | 24.5 | 1 | ||||
| Yes | 318.2 | 68 | 21.4 | 0.85 (0.60–1.21) | 0.374 | |||
| Agreement to sex outside relationship | ||||||||
| No sex | 353.6 | 72 | 20.3 | 1 | ||||
| Yes – condoms | 157.3 | 43 | 27.3 | 1.37 (0.94–2.01) | 0.100 | |||
| Yes – CLAI | 50.7 | 13 | 25.6 | 1.25 (0.69–2.25) | 0.469 | |||
| Anal sex with other partners | ||||||||
| No | 330.5 | 48 | 14.5 | 1 | 1 | |||
| Yes | 243.9 | 85 | 34.8 | 2.40 (1.69–3.42) | <0.001 | 2.87 (1.96–4.21) | <0.001 | |
| HNP PrEP use in previous 3 months | ||||||||
| No | 472.7 | 86 | 20.1 | 1 | 1 | |||
| Yes | 135.6 | 42 | 31.0 | 1.56 (1.08–2.26) | 0.019 | 1.51 (1.01–2.25) | 0.044A | |
When stratifying STI diagnosis incidence rates by relationship agreement type, having an agreement to not have sex outside the relationship resulted in the lowest STI diagnosis rates for both the HNP (6.2/100 person-years, 95% CI: 3.4–11.2) and HPP (10.7/100 person-years, 95% CI: 6.6–16.8). When there was no agreement, the STI diagnosis rates for the HNP (17.6/100 person-years, 95% CI: 12.0–25.6) and HPP (26.7/100 person-years, 95% CI: 19.7–36.4) were similar to when the couple had an agreement that sex could occur outside the relationship, but a condom was required, or no anal sex (HNP: 18.4/100 person-years, 95% CI: 12.4–27.2; HPP: 29.5/100 person-years, 95% CI: 21.6–40.2). The highest STI diagnosis rates were when there was an agreement allowing condomless anal intercourse (CLAI) outside the relationship, for both the HNP (27.2/100 person-years, 95% CI: 12.9–57.0) and HPP (39.7/100 person-years, 95% CI: 21.4–73.8).
Discussion
This analysis highlights important dynamics in the diagnosis of STIs in HIV-serodiscordant gay male couples. It extends knowledge of the needs of these couples by exploring STI transmission where existing research has focused predominantly on HIV transmission risk.21,24,25 Participants with HIV had higher incidence of concomitant bacterial STIs during study follow up, highlighting their syndemic interaction and the importance of combined prevention approach.26 We reported low congruence of STI diagnosis among study partners, and different forms of relationship agreements were associated with different rates of STI diagnosis. Different STI rates across country mirror regional data, markedly the high burden of syphilis in Brazil and of chlamydia in Thailand. Overall, younger ages and more frequent reports of anal sex with other partners, which are known vulnerability factors for STI diagnosis, were independently associated with bacterial STI acquisition during study follow up, regardless of HIV status.
The higher rates of STIs among the HPPs, particularly in Brazil and Thailand, could reflect different regional variabilities, as well as interconnection of different sexual networks, encompassing potential differences in sexual practices among enrolled participants. For instance, participants may have been having CLAI with other people with HIV, whereas the HNP may have been using condoms at a higher rate with other partners to minimise the risk of HIV transmission. Within some sexual networks of people with HIV, condoms are less used,9 and it has been shown in Australia that HIV-positive gay men have higher rates of syphilis and anal gonorrhoea.27
Access to regular, culturally appropriate STI testing is essential for men in HIV-serodiscordant couples. A person-centred approach is important to assess the risk of STI acquisition for each participant individually, and this should guide decisions on STI testing and other prevention strategies. For those with monogamous agreements, testing may still be important, given that behaviour does not always match the agreement,17 but will need to be promoted in a constructive way. The higher levels of STI diagnosis at baseline in Brazil and Thailand may reflect lower levels of accessible STI testing and treatment in these countries compared with Australia, rather than substantial differences in behaviour. The higher rates of STI diagnosis among HPPs has implications for regular HIV monitoring and STI testing. In many cases, viral load and treatment monitoring is now occurring on a 6-monthly basis, which may represent fewer routine opportunities for STI testing. Gay men with HIV who are sexually active should continue to be encouraged to undertake quarterly STI testing.12
There was low congruence between partners for STI diagnoses, which may reflect a range of factors. First, to participate in this study, the study partners only had to be having sex with each other once a month, so it may be that there were instances where the study partners were having more sex outside the relationship than within, so the opportunity for STI transmission to occur within the relationship was lower. Second, the couples may have been using condoms within the relationship primarily to prevent HIV transmission, as the period of data collection was before widespread acceptance of undetectable = untransmissable.14 Data collection occurred prior to TasP being confirmed in gay men,4,28 so in some cases, the couples would not have been comfortable relying on an undetectable viral load to prevent transmission.29 Third, even if the couple was having sex in the absence of condoms, STIs are not guaranteed to be transmitted with every act of sex.30,31 Finally, although the timing of STI testing for each partner was close, they were not necessarily at the same time, so a partner could have acquired an STI in the gap between each individual’s visit. This highlights the need to focus on the risk of STI acquisition from outside partners. In the past, concepts, such as negotiated safety32 for HIV, were promoted to enable couples to safely have condomless sex within the relationship by consistently using condoms outside the relationship. If STI rates continue to climb within this population, couples may choose to prioritise condomless sex within the relationship and agree to use condoms outside the relationship. The use of doxycycline as post-exposure prophylaxis is also being explored,33 although it appears to not be effective for gonorrhoea due to antimicrobial resistance.
Several other variables impacted on the likelihood of STI diagnosis. The decrease in STI diagnosis as the age of both partners increased could point to a change in sexual practice, whether number of partners or the form of protection used. This aligned with the finding that HNPs who reported >5 years since first sex with their study partner being less likely to have STI. However, it has been reported that is common for gay men to move from monogamy (or assumed monogamy) to have an agreement for sex outside the relationship as the length of the relationship increases.34 It was expected that PrEP use by the HNP would lead to additional STI diagnoses, and this was the case for both partners in bivariable analysis. In multivariable analysis, there was no longer a statistically significant association, although there was for the HPP. This difference may reflect a lack of power in the study, but may also reflect that the introduction of PrEP may have changed the type of sex happening both inside and outside the relationship.
Relationship agreements had an impact on STI transmission, although this is moderated by whether behaviour matches the stated agreement.17 An agreement for monogamy greatly reduced the incidence of STIs. Not having an agreement and an agreement to use condoms or not have anal intercourse had similar rates. The participants who had an agreement for CLAI had the highest rate of STI diagnosis, which was expected. Given that consistent condom use is decreasing due to greater uptake of other biomedical interventions,15 it is not clear that having agreements for condom use outside the relationship will be as common in the future. Where there may be significant gains is addressing the ‘no agreement’ group, and particularly the assumed monogamy to be formalised as a strategy to reduce STI transmission. Additionally, there is scope for relationship agreements to have a stronger focus on STI testing where CLAI is occurring with other partners. Treatment as prevention and PrEP are having an impact on relationship agreements,35 which will flow on to STI risk. By removing HIV risk in the context of condomless anal intercourse, the impetus to use condoms is reduced. As the risk of HIV transmission within their relationship has diminished, it is possible that this has changed risk perceptions for STIs within the relationship, as well as for both HIV and STIs with other partners.
There were clear differences in STI diagnosis by country. In multivariable analysis, HPPs in Brazil and Thailand were more likely to be diagnosed with an STI than their Australian counterparts, with syphilis being the most common STI in Brazil, and chlamydia the most common in Thailand. However, differences in the number of other partners between countries may also impact on these results. Brazil and Thailand predominantly undertake syndromic management, which differs from Australia’s system of large-scale screening. Although condoms remain the most effective tool available for preventing the spread of each STI, a stronger focus on recognising symptoms is important for health promotion activities, while increasing access to both symptomatic and asymptomatic testing. Different STIs have different health implications for individuals and for the health system response, and this is a reminder that it is important to monitor which STIs are prevalent in particular communities, and provide education and testing as required.
This analysis had a number of limitations. Although this was a relatively large sample in the context of serodiscordant couples, the sample size led to large confidence intervals, and a larger sample may have uncovered additional associations. The differences in the number of couples recruited in Australia compared with Brazil and Thailand meant that factors where there was an association in Australia, but not the other countries, may reflect the difference in statistical power. The participating couples may not be representative of male HIV-serodiscordant couples in the three countries, as they were drawn from high case load clinics in urban areas and were therefore more likely to be connected to care, resulting in selection bias, and they were more likely to be engaged with community health promotion focused on sexual health. Finally, either nucleic acid amplification tests or culture-based tests were used for chlamydia and gonorrhoea screening based on local availability, and the difference in sensitivity could have influenced the diagnosis rate.
This study highlights important differences in STI diagnoses within HIV-serodiscordant couples between three different countries, as well as the impact of different types of agreements about sex outside the relationship. Interventions aimed at strengthening these agreements with commitments to STI testing will be valuable.
Consent to participate
Participants were provided detailed information on the study, including written materials, prior to signing consent forms. Participants were required to sign consent forms while their study partner was out of the room to reduce the risk of coerced participation. Participants could withdraw from the study at any time.
Consent for publication
Consent to publish de-identified, aggregate participant data was obtained when participants consented to participate.
Data availability
A Research Data Management Plan has been developed for this research. All data is held within a university data repository to ensure transparency.
Conflicts of interest
Nittaya Phanuphak and Benjamin Bavinton are Associate Editors for Sexual Health but were not involved in the peer review or decision-making process for this paper. The authors have no further conflicts of interest to declare.
Declaration of funding
Opposites Attract was funded by: NHMRC, the American Foundation for AIDS Research, Gilead Sciences, and ViiV Healthcare.
Author contributions
All authors contributed to the development of the manuscript. All authors, with the exception of Graham Brown and Mayara Secco Torres Silva, were involved in the study development and data collection.
Acknowledgements
We thank all participants in the study, staff in the participating clinics, laboratory staff, and staff in gay and HIV community organisations for supporting this research. Opposites Attract was funded by: NHMRC, the American Foundation for AIDS Research, Gilead Sciences and ViiV Healthcare.
References
2 Galvin SR, Cohen MS. The role of sexually transmitted diseases in HIV transmission. Nat Rev Microbiol 2004; 2(1): 33-42.
| Crossref | Google Scholar | PubMed |
3 Cohen MS, Chen YQ, McCauley M, Gamble T, Hosseinipour MC, Kumarasamy N, et al. Antiretroviral therapy for the prevention of HIV-1 transmission. N Engl J Med 2016; 375(9): 830-39.
| Crossref | Google Scholar | PubMed |
4 Bavinton BR, Pinto AN, Phanuphak N, Grinsztejn B, Prestage GP, Zablotska-Manos IB, et al. Viral suppression and HIV transmission in serodiscordant male couples: an international, prospective, observational, cohort study. Lancet HIV 2018; 5(8): e438-47.
| Crossref | Google Scholar | PubMed |
6 Grant RM, Lama JR, Anderson PL, McMahan V, Liu AY, Vargas L, et al. Preexposure chemoprophylaxis for HIV prevention in men who have sex with men. N Engl J Med 2010; 363(27): 2587-99.
| Crossref | Google Scholar | PubMed |
10 Queiroz AAFLN, Matos MCB, Araújo TMEd, Reis RK, Sousa ÁFL. Sexually transmitted infections and factors associated with condom use in dating app users in Brazil. ACTA Paulista De Enfermagem 2019; 32(5): 546-53.
| Crossref | Google Scholar |
11 Hiransuthikul A, Sungsing T, Jantarapakde J, Trachunthong D, Mills S, Vannakit R, et al. Correlations of chlamydia and gonorrhoea among pharyngeal, rectal and urethral sites among Thai men who have sex with men: multicentre community-led test and treat cohort in Thailand. BMJ Open 2019; 9(6): e028162.
| Google Scholar |
13 Crosby R, Bounse S. Condom effectiveness: where are we now? Sex Health 2012; 9(1): 10-7.
| Crossref | Google Scholar | PubMed |
14 Gray J, Prestage G, Jin F, Phanuphak P, Friedman RK, Fairley CK, et al. Condom failure among HIV-negative men in serodiscordant relationships in Australia, Brazil, and Thailand. AIDS Behav 2024; 28(11): 3587-95.
| Crossref | Google Scholar |
15 Holt M, Broady TR, Mao L, Chan C, Rule J, Ellard J, et al. Increasing preexposure prophylaxis use and ‘net prevention coverage’ in behavioural surveillance of Australian gay and bisexual men. AIDS 2021; 35(5): 835-40.
| Crossref | Google Scholar | PubMed |
17 Gray J, Prestage G, Jin F, Phanuphak N, Friedman RK, Fairley CK, et al. Agreements, behaviour, and change: sex outside the relationship in male HIV-negative partners in HIV serodiscordant relationships in Australia, Brazil, and Thailand. AIDS Behav 2023; 27(9): 3098-108.
| Crossref | Google Scholar |
18 Johnson LF, Lewis DA. The effect of genital tract infections on HIV-1 shedding in the genital tract: a systematic review and meta-analysis. Sex Transm Dis 2008; 35(11): 946-59.
| Crossref | Google Scholar | PubMed |
19 Champredon D, Bellan SE, Delva W, Hunt S, Shi C-F, Smieja M, et al. The effect of sexually transmitted co-infections on HIV viral load amongst individuals on antiretroviral therapy: a systematic review and meta-analysis. BMC Infect Dis 2015; 15: 249.
| Crossref | Google Scholar | PubMed |
20 Bavinton BR, Jin F, Prestage G, Zablotska I, Koelsch KK, Phanuphak N, et al. The Opposites Attract Study of viral load, HIV treatment and HIV transmission in serodiscordant homosexual male couples: design and methods. BMC Public Health 2014; 14: 917.
| Crossref | Google Scholar | PubMed |
21 Philpot SP, Prestage G, Ellard J, Grulich AE, Bavinton BR, Opposites Attract Study Group. How do gay serodiscordant couples in Sydney, Australia negotiate undetectable viral load for HIV prevention? AIDS Behav 2018; 22(12): 3981-90.
| Crossref | Google Scholar | PubMed |
22 Goddard SL, Poynten IM, Petoumenous K, Jin F, Hillman RJ, Law C, et al. Prevalence, incidence and predictors of anal Chlamydia trachomatis, anal Neisseria gonorrhoeae and syphilis among older gay and bisexual men in the longitudinal Study for the Prevention of Anal Cancer (SPANC). Sex Transm Infect 2019; 95(7): 477-83.
| Crossref | Google Scholar | PubMed |
23 Traeger MW, Cornelisse VJ, Asselin J, Price B, Roth NJ, Willcox J, et al. Association of HIV preexposure prophylaxis with incidence of sexually transmitted infections among individuals at high risk of HIV infection. JAMA 2019; 321(14): 1380-90.
| Crossref | Google Scholar | PubMed |
24 Bavinton BR, Jin F, Mao L, Zablotska I, Prestage GP, Grulich AE. Homosexual men in HIV serodiscordant relationships: implications for HIV treatment as prevention research. J Int AIDS Soc 2015; 18(1): 19884.
| Crossref | Google Scholar | PubMed |
25 Starks TJ, Gamarel KE, Johnson MO. Relationship characteristics and HIV transmission risk in same-sex male couples in HIV serodiscordant relationships. Arch Sex Behav 2014; 43(1): 139-47.
| Crossref | Google Scholar | PubMed |
26 Gandhi M, Spinelli MA, Mayer KH. Addressing the sexually transmitted infection and HIV syndemic. JAMA 2019; 321(14): 1356-58.
| Crossref | Google Scholar | PubMed |
27 Jin F, Prestage GP, Zablotska I, Rawstorne P, Kippax SC, Donovan B, et al. High rates of sexually transmitted infections in HIV positive homosexual men: data from two community based cohorts. Sex Transm Infect 2007; 83(5): 397-99.
| Crossref | Google Scholar | PubMed |
28 Cohen MS, Chen YQ, McCauley M, Gamble T, Hosseinipour MC, Kumarasamy N, et al. Prevention of HIV-1 infection with early antiretroviral therapy. N Engl J Med 2011; 365(6): 493-505.
| Crossref | Google Scholar | PubMed |
29 Holt M, MacGibbon J, Bear B, Lea T, Kolstee J, Crawford D, et al. Trends in belief that HIV treatment prevents transmission among gay and bisexual men in Australia: results of national online surveys 2013–2019. AIDS Educ Prev 2021; 33(1): 62-72.
| Crossref | Google Scholar |
30 Stoltey JE, Cohen SE. Syphilis transmission: a review of the current evidence. Sex Health 2015; 12(2): 103-09.
| Crossref | Google Scholar | PubMed |
31 Althaus CL, Heijne JCM, Low N. Towards more robust estimates of the transmissibility of Chlamydia trachomatis. Sex Transm Dis 2012; 39(5): 402-4.
| Crossref | Google Scholar | PubMed |
32 Kippax S, Noble J, Prestage G, Crawford JM, Campbell D, Baxter D, et al. Sexual negotiation in the AIDS era: negotiated safety revisited. AIDS 1997; 11(2): 191-7.
| Crossref | Google Scholar | PubMed |
33 Sokoll PR, Migliavaca CB, Döring S, Traub U, Stark K, Sardeli AV. Efficacy of postexposure prophylaxis with doxycycline (Doxy-PEP) in reducing sexually transmitted infections: a systematic review and meta-analysis. Sex Transm Infect 2025; 101(1): 59-67.
| Crossref | Google Scholar | PubMed |
34 Philpot SP, Duncan D, Ellard J, Bavinton BR, Grierson J, Prestage G. Negotiating gay men’s relationships: how are monogamy and non-monogamy experienced and practised over time? Cult Health Sex 2018; 20(8): 915-28.
| Crossref | Google Scholar | PubMed |
35 Gray J, Prestage G, Jin F, Phanuphak N, Friedman RK, Fairley CK, et al. Characteristics of agreements to have condomless anal intercourse in the presence of an undetectable viral load among HIV serodiscordant male couples in Australia, Brazil and Thailand. AIDS Behav 2021; 25(12): 3944-54.
| Crossref | Google Scholar | PubMed |
Footnotes
§ Opposites Attract Study Group: Andrew E. Grulich (Chief Investigator), Iryna B. Zablotska, Garrett P. Prestage, Fengyi Jin, Benjamin R. Bavinton (project leader), Beatriz Grinsztejn, Nittaya Phanuphak, David A. Cooper, Anthony Kelleher, Sean Emery, Christopher K. Fairley, David Wilson, Kersten K. Koelsch, Kathy Triffitt, Nicolas Doong, David Baker, Mark Bloch, David J. Templeton, Anna McNulty, Catherine Pell, Jennifer Hoy, Ban Kiem Tee, Richard Moore, Norm Roth, David Orth and Angie N. Pinto.
