Complete information on HIV status, sexual risk behavior, and potential confounders was provided by 6064 men over the study period (Table 1): HIV-positive 1001 (16.5%), HIV-negative 3866 (63.8%), never-tested 1197 (19.7%) (range, 482–834 per year; estimated response rate, 50%–60%).20 The percentage of men who reported ever having had an HIV test increased from 72.1% in 1998 to 90.1% in 2008 (P < 0.001), with 50.8% (329/648) of gay men in 2008 reporting a HIV test within the previous 12 months. Between 1998 and 2008, median age increased from 35 to 41 years for HIV-positive men (P < 0.001) and from 33 to 37 years for HIV negative men (P < 0.001); there was no such increase for never-tested men (P = 0.31). The percentage of men who said they used the Internet to look for sex increased from 27.2% in 2000 to 59.1% in 2008 (P = 0.01). There was no significant trend over time in recreational drug use, steroid use, HIV treatment optimism, or being in a relationship (Table 1).
After excluding 2780 men who had completed a previous questionnaire, 3287 respondents remained for the independent samples analysis. Detailed analysis of the sexual behavior of those respondents included and excluded from the independent samples analysis showed no systematic differences between the 2 groups (data available from the authors on request).
Unprotected Anal Intercourse
The overall percentage of gay men who reported engaging in any UAI in the previous 3 months (i.e., nonconcordant or concordant) increased from 24.3% in 1998 to 36.6% in 2008 (P = 0.07; Table 2, Fig. 1). However, this overall increase conceals some important differences in trends for nonconcordant and concordant UAI.
Overall, the percentage of men who engaged in ncUAI increased significantly between 1998 and 2001 (P < 0.001), followed by a significant decrease between 2001 and 2005 (P < 0.001) and a leveling off between 2005 and 2008 (P = 0.23; Table 2, Fig. 1). In contrast, the percentage of men engaging in concordant UAI (i.e., serosorting) showed a steady increase between 1998 and 2008 (P = 0.01; Table 2, Fig. 1). In 2005, the percentage of men reporting cUAI exceeded the percentage reporting ncUAI for the first time since data collection began in 1998, and has remained higher ever since.
Nonconcordant Unprotected Anal Intercourse
Although the percentage of men who engaged in ncUAI increased significantly between 1998 and 2001, decreased between 2001 and 2005, and then leveled off (Table 2, Fig. 1), this overall trend conceals some important differences according to the HIV status of the respondent and the type of partner (main or casual; Table 3).
The overall trend in ncUAI has been largely driven by changing patterns of ncUAI with casual partners. The percentage of men reporting ncUAI with a casual partner increased from 6.7% to 15.2% between 1998 and 2001 (P < 0.001), then decreased to 11.6% in 2005 (P < 0.05), falling further to 8.6% in 2008 (P = 0.05; Table 3, Fig. 2A).
The increase in ncUAI with a casual partner between 1998 and 2001 was reported among all men irrespective of HIV status (HIV-positive, negative, and never tested; Table 3, Fig. 2B). The decline in the percentage of men reporting ncUAI with a casual partner between 2001 and 2005 was especially notable among HIV-positive men (P < 0.05) and to a lesser extent among HIV negative men (P = 0.09). Between 2005 and 2008, however, ncUAI with a casual partner among HIV-positive (P = 0.9) and never-tested men (P = 0.6) remained stable while the percentage of HIV negative men reporting ncUAI with a casual partner continued to fall (from 10.3% to 5.5%; P < 0.05), reaching the lowest level reported over the 10-year survey period.
The percentage of men reporting ncUAI with a main partner alone decreased significantly from 7.8% in 1998 to 3.9% in 2005 (P < 0.05) followed by a significant increase to 7.1% in 2008 (P < 0.05; Table 3, Fig. 2A). Between 2005 and 2008, the percentage of men reporting ncUAI with a main partner increased from 2.5% to 8.1% (P < 0.05) for HIV-positive gay men and from 2.1% to 5.5% (P = 0.06) for HIV negative men (Table 3, Fig. 2C). This is the first time we have recorded an increase in ncUAI with a main partner since data collection began in 1998.
As a result of the decrease in ncUAI with a casual partner and an increase in ncUAI with a main partner, in 2008 there was little difference in the overall percentage of men reporting ncUAI with a main partner alone (7.1%) or with a casual partner (8.6%) (Table 3, Fig. 2A).
Discordant Versus Status Unknown Partner.
The majority of men who reported ncUAI in 2008 said that this had occurred with a man (or men) of unknown rather than discordant serostatus. In most cases where men did not know the HIV status of their partners, they had assumed it was the same as theirs. Among the 24 HIV-negative men who reported ncUAI with a main partner, only one knew his partner was HIV-positive, whereas the remaining 23 did not know their partners status. On the other hand, among the 12 HIV-positive men reporting ncUAI with a main partner, 6 reported that they knew their partners were HIV-negative.
Concerning casual partners, all 28 HIV-positive men and 23 out of 24 HIV-negative men reporting ncUAI with a casual partner said their partners were of unknown serostatus. Only 1 HIV negative men said he knew his casual ncUAI partner was discordant (i.e., HIV-positive).
Concordant Unprotected Anal Intercourse—“Serosorting”
Overall, the percentage of men engaging in cUAI increased significantly from 9.8% in 1998 to 20.8% in 2008 (P = 0.01) (Table 2, Fig. 1). However, this increasing trend conceals key differences depending upon the HIV status of the respondent and the type of partner (main or casual; Table 3).
The percentage of HIV-positive men reporting cUAI with a casual partner increased significantly between 1998 and 2005 (6.8%, 17.7%; P < 0.01), and then decreased between 2005 and 2008 (17.7%, 14.2%; P = 0.05). On the other hand, only a small number of HIV negative men reported cUAI (serosorting) with a casual partner throughout the study period with no significant trend between 1998 and 2008 (1.7%, 1.6%; P = 0.9) (Table 3, Fig. 3A).
The percentage of HIV negative men reporting cUAI with a main partner alone increased significantly from 12.4% in 1998 to 21.1% in 2008 (P < 0.05). HIV-positive men were less likely to report cUAI with a main partner than HIV negative men, but nonetheless, there was also an increasing trend among positive men over time (5.1%, 10.1%; P < 0.05) (Table 3, Fig. 3B).
In 2008, of the 135 men who reported cUAI (i.e., serosorting), 127 reported how they knew the HIV status of their partner(s). Among the HIV negative men who reported cUAI, only 7 reported cUAI with a casual partner, whereas 85 reported cUAI with a main partner. Of these 85 men, 53 (62.1%) said they knew their main partner's HIV status through verbal disclosure, while 35 (41.2%) reported HIV testing together. Nearly all the 35 HIV-positive men who serosorted said their partners had told them about their status, but over a quarter (n = 10) also relied on a person's online profile or website.
In this article, we have described trends in UAI between 1998 and 2008 among gay men in London. Overall, the percentage of gay men in our study who engaged in UAI steadily increased between 1998 and 2008, a finding reported in other behavioral surveys in London and elsewhere.25–29 In 2008, over a third of all men reported UAI in the previous 3 months, compared to a quarter in 1998.
The overall trend in UAI masked a more complex picture, which became evident when UAI was classified as concordant (cUAI) or nonconcordant (ncUAI). ncUAI increased rapidly from 1998 to 2001, after which it decreased. By 2008, ncUAI had almost fallen back to the level reported in 1998. By way of comparison, cUAI, or serosorting, has shown an increasing trend among London's gay men since 1998. As a consequence, by 2008, more London gay men in our survey reported cUAI (serosorting) than ncUAI.
We have also found that patterns of ncUAI varied considerably by partner type, with distinctly different trends for casual and main partners. With casual partners, there was an increase in ncUAI between 1998 and 2003, after which it steadily decreased. However, ncUAI with a main partner has shown the opposite trend, initially decreasing up until 2005, and then increasing. This is the first time that we have seen an increase in the percentage of men reporting ncUAI with their main partner since data collection began in 1998. As a result, in 2008, there was little difference between the levels of risk reported with main or casual partners.
Main partners have been implicated in HIV transmission among gay men in the Netherlands, Germany, and the United States.14,15,30,31 Our data suggest that main partners may have become an important source of HIV risk among London's gay men, highlighting a key, and as yet, underserved area of health promotion. In addition, we have shown that the majority of men who engaged in ncUAI with their main partner reported not knowing the status of their partner, rather than knowing that they were discordant. This suggests that work may also need to be done around HIV testing within partnerships.
Recent trends in ncUAI varied according to the HIV status of the respondent. The increase in ncUAI with a main partner since 2005 was seen mostly among HIV-positive men, and to a slightly lesser extent among HIV negative men. However, ncUAI with a casual partner remained stable among HIV-positive and never-tested gay men between 2005 and 2008, but decreased among HIV negative men. By 2008, ncUAI with a casual partner reached the lowest level reported among HIV negative men over the 10-year survey period. This is an encouraging trend and is likely to reflect consistent and sustained health promotion campaigns targeting gay men. Similar decreases in ncUAI have been reported among MSM in San Francisco10 and Sydney.32
Over the 10 years of the survey, the overall percentage of men reporting UAI with men of the same HIV status (serosorting) doubled, although this varied according to partner type and HIV status of respondent.
Among HIV negative men serosorting with a casual partner has remained consistently low over the survey period in contrast to trends reported among HIV-negative gay men in Sydney.33,34 This is a reassuring finding and reflects how difficult it is for HIV negative men to reliably establish seroconcordance with casual partners.6 In marked contrast, the percentage of HIV-positive men who reported serosorting with a casual partner increased between 1998 and 2005, but then decreased between 2005 and 2008. Nonetheless, the overall percentage of HIV-positive men who serosort with casual partners remains high (1 in 7 in 2008). An increase in serosorting has also been reported among HIV-positive MSM in Seattle.8
Enhanced surveillance for STIs in the United Kingdom has revealed an increasing proportion of STIs among those already infected with HIV.1 This mirrors the trends in serosorting among HIV-positive men reported here. In 2008, 1 in 7 HIV-positive men in our study reported serosorting with casual partners. It remains to be seen whether the reduction since 2005 in the percentage of HIV-positive men serosorting with casual partners in this study translates into a reduction in new STI diagnoses among this group, or whether STIs will continue to circulate among networks of HIV-positive MSM as has been seen in other European countries.35 It is therefore important that targeted and innovative campaigns continue to highlight the implications of additional STIs in those already infected with HIV.
Serosorting relies on men disclosing their HIV status, and reliably establishing the HIV status of their sexual partners. Men described a variety of mechanisms through which they established the HIV status of their cUAI partners. Most frequently, this involved verbal disclosure, cotesting for HIV and to a lesser extent, the use of online profiles and websites.
Among HIV-negative men, we continued to see a steady increase in the percentage who reported UAI only with a main partner of the same HIV status. However, in 2008, almost two-thirds of HIV negative men who reported concordant UAI with their main partner relied on verbal disclosure to establish seroconcordance and less than half reported testing with their partner for HIV. Verbal disclosure among HIV-negative men is not entirely satisfactory, since it relies on an accurate knowledge of their own HIV status, and will depend on the time since their last test and their subsequent risk behavior. Although an increasing percentage of HIV-negative men reported concordant UAI with a main partner, the mechanism through which many of them establish concordance is unreliable. Rather than serosorting or even seroguessing, many of these men appear to be “sero-hoping.”8 It is therefore crucial that future health promotion campaigns promote HIV testing among couples, both at the outset of the relationship to establish seroconcordance and at intervals during the relationship depending upon the context (monogamous or otherwise).
The majority of HIV-positive men who serosorted relied on verbal disclosure with almost a third also using online profiles and websites, highlighting the importance of the internet as a medium through which HIV-positive men can disclose their status.36 In marked contrast to HIV negative men, however, HIV-positive men can reliably establish concordance through mutual disclosure.
It is encouraging that there has been a steady increase in the percentage of men ever tested for HIV in our study, reaching 90% in 2008. This is likely to reflect the success of universally offering the HIV test in sexual health clinics in Britain. HPA data indicate that uptake of HIV testing among MSM in this setting increased substantially between 2003 and 2008 (Leong, G. Testing offer and uptake among MSM. Health Protection Agency, Centre for Infections, Personal communication, 24 September 2009).
There are a number of limitations to the study. No information is available on trends in receptive and insertive UAI over time, nor did we ask about withdrawal before ejaculation. Furthermore, we did not collect information on the viral load of HIV-positive men which men may use to inform their decisions around risk.
In conclusion, the patterns of sexual behavior among London's gay men between 1998 and 2008 appear to be dynamic and complex. We have seen clear differences in trends in sexual behavior according to the type of partner (casual or main) and the HIV status of the partner (concordant or nonconcordant). These findings throw into sharp focus the importance of differentiating between seroconcordant and nonconcordant partners when tracking trends in UAI, as well as understanding the context in which risk occurs.37 Our study highlights the importance of conducting behavioral surveillance, based on repeat cross-sectional studies in sentinel populations, to monitor trends in risk behavior and partnership patterns over time.16,17
For the first time since we began to survey gay men attending central London gyms, we have seen an increase in HIV risk behavior with a main partner. In addition, although we have seen an encouraging increase over time in the percentage of HIV-negative men who serosort with a main partner, the mechanism by which they establish seroconcordance is often unreliable. These recent trends in sexual behavior among gay men suggest that HIV risk with a main partner and HIV testing among couples should now be given greater priority by health promotion programmes in London and elsewhere in the United Kingdom.
1.The UK collaborative Group on HIV and STI Surveillance. Sexually transmitted infections and man who have sex with men in the UK: 2008 Report. London, United Kingdom: Health Protection Agency, Centre for Infections, 2008.
2.Elford J. Changing patterns of sexual behaviour in the era of highly active antiretroviral therapy. Curr Opin Infect Dis 2006; 19:26–32.
3.Dodds JP, Johnson AM, Parry JV, et al. A tale of three cities: Persisting high HIV prevalence, risk behaviour and undiagnosed infection in community samples of men who have sex with men. Sex Transm Infect 2007; 83:392–396.
4.Cohen MS. HIV and sexually transmitted diseases: Lethal synergy. Top HIV Med 2004; 12:104–107.
5.Rietmeijer CA, Lloyd LV, McLean C. Discussing HIV serostatus with prospective sex partners: A potential HIV prevention strategy among high-risk men who have sex with men. Sex Transm Dis 2007; 34:215–219.
6.Wilson DP, Regan DG, Heymer KJ, et al. Serosorting may increase the risk of HIV acquisition among men who have sex with men. Sex Transm Dis 2010; 37:13–17.
7.Cassels S, Menza TW, Goodreau SM, et al. HIV serosorting as a harm reduction strategy: evidence from Seattle, Washington. AIDS 2009; 23:2497–2506.
8.Golden MR, Stekler J, Hughes JP, et al. HIV serosorting in men who have sex with men: is it safe?[erratum in: J Acquir Immun Defic Syndr 2008; 49:464] J Acquir Immun Defic Syndr 2008; 49:212–218.
9.Marks G, Millett GA, Bingham T, et al. Prevalence and protective value of serosorting and strategic positioning among black and Latino men who have sex with men. Sex Transm Dis 2010; 37:325–327.
10.Truong HM, Kellogg T, Klausner JD, et al. Increases in sexually transmitted infections and sexual risk behaviour without a concurrent increase in HIV incidence among men who have sex with men in San Francisco: A suggestion of HIV serosorting? [erratum in: Sex Transm Infect 2007; 83:76. Truong, H-H M corrected to Truong, H M] Sex Transm Infect 2006; 82:461–466.
11.van der Bij AK, Stolte IG, Coutinho RA, et al. Increase of sexually transmitted infections, but not HIV, among young homosexual men in Amsterdam: Are STIs still reliable markers for HIV transmission? Sex Transm Infect 2005; 81:34–37.
12.Nusbaum MR, Wallace RR, Slatt LM, et al. Sexually transmitted infections and increased risk of co-infection with human immunodeficiency virus. J Am Osteopath Assoc 2004; 104:527–535.
13.Rottingen JA, Cameron DW, Garnett GP. A systematic review of the epidemiologic interactions between classic sexually transmitted diseases and HIV: how much really is known? Sex Transm Dis 2001; 28:579–597.
14.Sullivan PS, Salazar L, Buchbinder S, et al. Estimating the percentage of HIV transmissions from main sex partners among men who have sex with men in five US cities. AIDS 2009; 23:1153–1162.
15.Xiridou M, Geskus R, De Wit J, et al. The contribution of steady and casual partnerships to the incidence of HIV infection among homosexual men in Amsterdam. AIDS 2003; 17:1029–1038.
16.Elford J, Jeannin A, Spencer B, et al; HIV and STI Behavioural Surveillance Mapping Group. HIV and STI behavioural surveillance among men who have sex with men in Europe. Euro Surveill 2009; 14:pii 19414.
17.McGarrigle CA, Fenton KA, Gill ON, et al. Behavioural surveillance: The value of national coordination. Sex Transm Infect 2002; 78:398–405.
18.Elford J, Bolding G, Sherr L, et al. High-risk sexual behaviour among London gay men: No longer increasing. AIDS 2005; 19:2171–2174.
19.Elford J, Bolding G, Davis M, et al. Trends in sexual behaviour among London homosexual men 1998–2003: Implications for HIV prevention and sexual health promotion. Sex Transm Infect 2004; 80:451–454.
21.Elford J, Bolding G, Sherr L. High risk sexual behaviour increases among London gay men between 1998–2001: What is the role of HIV optimism? AIDS 2002; 16:1537–1544.
22.Bolding G, Davis M, Hart G, et al. Gay men who look for sex on the Internet: Is there more HIV/STI risk with online partners? AIDS 2005; 19:961–968.
23.Bolding G, Hart G, Sherr L, et al. Use of crystal methamphetamine among London gay men. Addiction 2006; 101:1622–1630.
24.Honaker J, King G, Blackwell M. Amelia: A Program for Missing Data. Cambridge, Harvard University. Available at: http://GKing.Harvard.edu
. Accessed December, 2008.
25.Dodds JP, Mercey DE, Parry JV, et al. Increasing risk behaviour and high levels of undiagnosed HIV infection in a community sample of homosexual men. Sex Transm Infect 2004; 80:236–240.
26.Dodds JP, Nardone A, Mercey DE, et al. Increase in high risk sexual behaviour among homosexual men, London 1996–8: Cross sectional, questionnaire study [erratum in: BMJ 2000; 321:675]. BMJ 2000; 320:1510–1511.
27.Van de Ven P, Prestage G, Crawford J, et al. Sexual risk behaviour increases and is associated with HIV optimism among HIV-negative and HIV-positive gay men in Sydney over the 4 year period to February 2000. AIDS 2000; 14:2951–2953.
28.Grulich A. HIV risk behaviour in gay men: On the rise? BMJ 2000; 320:1487–1488.
29.Osmond DH, Pollack LM, Paul JP, et al. Changes in prevalence of HIV infection and sexual risk behavior in men who have sex with men in San Francisco: 1997–2002. Am J Public Health 2007; 97:1677–1683.
30.Davidovich U, de Wit J, Albrecht N, et al. Increase in the share of steady partners as a source of HIV infection: A 17-year study of seroconversion among gay men. AIDS 2001; 15:1303–1308.
31.Bochow M. Schwule Manner, AIDS und Safer Sex: neue Entwicklungen. Berlin, Germany: Deutche AIDS-Hilfe e.V, 2001.
32.Zablotska IB, Prestage G, Grulich AE, et al. Differing trends in sexual risk behaviours in three Australian states: New South Wales, Victoria and Queensland, 1998–2006. Sex Health 2008; 5:125–130.
33.Elford J, Bolding G, Sherr L, et al. No evidence of an increase in serosorting with casual partners among HIV negative gay men in London 1998–2005. AIDS 2007; 21:243–245.
34.Mao L, Crawford J, Hospers H, et al. “Serosorting” in casual anal sex of HIV negative gay men is noteworthy and is increasing in Sydney, Australia. AIDS 2006; 20:1204–1206.
35.Dougan S, Evans BG, Elford J. Sexually transmitted infections in western Europe among HIV positive men who have sex with men. Sex Transm Dis 2007; 34:783–790.
36.Davis M, Hart G, Bolding G, et al. Sex and the internet: Gay men, risk reduction and serostatus. Cult Health Sex 2006; 8:161–174.
© Copyright 2011 American Sexually Transmitted Diseases Association
37.Xia Q, Molitor F, Osmond DH, et al. Knowledge of sexual partner's HIV serostatus and serosorting practices in a California population-based sample of men who have sex with men. AIDS 2006; 20:2081–2089.