Men who have sex with men (MSM) remain the group most at risk of acquiring HIV in the United Kingdom.1 Increasing diagnoses of HIV and other sexually transmitted infections (STIs) and international evidence of increased HIV incidence are of concern.2-5 In Scotland, MSM account for more than a third of all diagnosed HIV cases, and there were more newly diagnosed reported cases in 2005 than in any previous year.6,7 In the 2004 unlinked anonymous genitourinary medicine (GUM) clinic survey, prevalence was 4.2%.8 Community-based surveys of MSM in England, which have included anonymous oral fluid testing for HIV, found an HIV prevalence of up to 14%,9 but prevalence among community samples of MSM in Scotland remains unknown. It is important for strategies for the prevention and control of infection as well as the management of disease to understand better the extent of the HIV epidemic in Scotland, particularly given the low levels of HIV testing and GUM clinic use among Scottish gay men.10-12
Since 1996, we have been surveying the HIV-related sexual behavior of gay men in Scotland.11,13-15 In 2005, we also collected oral mucosal samples to be tested anonymously for HIV antibodies so as to improve the estimates of HIV prevalence and undiagnosed infection in this population. In this article, we examine HIV prevalence and undiagnosed infection among community-recruited gay men in Glasgow and Edinburgh.
Time and location sampling was used to recruit representative samples of men visiting the commercial gay scene in each city, consistent with the approach used in previous years.11,13-15 All but 1 of the exclusively gay bars in each city (6 in Glasgow and 5 in Edinburgh) were included (1 Edinburgh bar declined participation). Bars were surveyed over a 2-week period in the early (7:00-9:00 pm) and late (9:00-11:00 pm) evening, and no bar was visited twice in the same evening. At the end of the 2-week period, each bar had been visited at both time points on each day of the week. A pilot study of saunas was conducted, with 1 in each city surveyed over 2 early evening periods (5:00-7:00 pm). All men present or entering the venues were asked to participate. Trained field workers distributed anonymous self-complete questionnaires and Orasure oral fluid collection kits (OraSure Technologies, Inc., Bethlehem, PA) for men to provide samples to test for HIV antibodies. Barcodes matched the samples to the questionnaires. The Orasure kits have been used successfully in comparable surveys since 2000.9,16 Collection kits were not distributed in the Glasgow sauna by request of the management. Ethical approval was granted by the University of Glasgow Faculty of Medicine Ethics Committee for Nonclinical Research Involving Human Subjects.
The questionnaire included demographics, HIV testing, perceived HIV status, sexual behavior, knowledge of partners' antibody status, experience of STIs, and use of GUM clinics in the previous year. Agreement with 2 standard HIV treatment optimism scales was measured.17
The oral fluid samples were analyzed at the West of Scotland Specialist Virology Center in Glasgow, which is a UK Clinical Pathology Accreditation (CPA) specialist virology laboratory for the main bloodborne viruses. One specimen had insufficient sample volume and was excluded. Specimens were screened for anti-HIV using an enzyme immunoassay (EIA; Vironostika HIV-UniForm II plus 0; BioMerieux UK, Basingstoke, United Kingdom); all that were positive were rescreened, and repeat reactive results were confirmed using a Western blot test (New LAV blot 1; Biorad Laboratories, Hemmel-Hempstead, United Kingdom).
The Pearson χ2 test was used for bivariate comparisons, and multivariate logistic regression was used to produce adjusted odds ratios (ORs) and to assess their significance. Differences between the survey locations and venues are controlled for in the analyses.
A total of 1744 men (66.1% response rate) participated: 1015 in Glasgow (65.6% response rate) and 729 in Edinburgh (66.5% response rate). There were 59 men (3.4%) in the sauna sample and 1685 (96.6%) in the bar sample. Men surveyed in saunas were significantly older than men in bars (mean age of 37 years compared with 32 years, respectively; P < 0.001) and were less likely to live in Glasgow (29.4% compared with 43.3%, respectively; P = 0.041). They reported significantly more sex partners in the previous year than men in the bars (>10 partners: 57.9% compared with 19.2%, respectively; P < 0.001), and more anal intercourse (AI) partners (>10 partners: 20.7% compared with 7.4%, respectively; P = 0.001). There were no differences in unprotected anal intercourse (UAI) with casual or multiple partners or partners of unknown or discordant HIV status. Excluding the sauna sample from the analyses did not change the results.
Oral fluid samples were provided by 1350 men (51.6% response rate): 749 in Glasgow (48.4% response rate) and 601 in Edinburgh (54.8% response rate). A further 394 men (22.6%) completed questionnaires but did not provide oral fluid specimens. When compared with this group, men who provided specimens were more likely to have been surveyed in Edinburgh, live in Edinburgh, and be unemployed (Table 1). In multivariate analysis, only being surveyed in Edinburgh remained significant (OR = 1.66, 95% confidence interval [CI]: 1.16 to 2.38). There were no other significant differences in demographics, sexual health service use, perceived HIV status, or sexual risk behavior (see Table 1).
Of the 1349 men tested, 60 were HIV-positive (4.4%, 95% CI: 3.5% to 5.7%). There were also 5 (0.4%) unconfirmed cases, which had 2 positive EIA screens but were negative on the Western blot confirmation test. HIV prevalence was 5.5% in Edinburgh (95% CI: 3.9% to 7.6%) and 3.6% in Glasgow (95% CI: 2.5% to 5.2%) (Table 2). A positive result was significantly associated with being aged older than 25 years (mean age of 36 years for positive men and 32 years for negative men), and living outside of Scotland. Prevalence was significantly higher among men who had used a GUM clinic in the previous year than among men who had not. Prevalence was highest among men who reported last having an HIV test more than 5 years ago (12.3%) and lowest among men who had never been tested (0.8%).
Differences in sexual risk behaviors are shown in Table 3. Prevalence was significantly higher among men who reported 10 or more sex partners and among men who reported 10 or more AI partners in the previous year than among men who reported fewer partners. Men who reported 2 or more UAI partners in the past year were significantly more likely to be positive than men who did not. There were no differences in reported UAI with casual partners or partners of unknown or discordant HIV status. HIV prevalence was significantly higher among men who reported having had an STI in the previous year than among men who had not.
Multivariate logistic regression was used to assess these associations, and each remained significant after controlling for the demographic differences in age and area of residence as well as for differences in survey location and venue (see Table 3). When the sexual risk behaviors were combined in 1 multivariate regression model (Table 4), older age, having 10 or more AI partners in the previous year, and having an STI in the previous year remained significant. There were no significant interactions between these variables, demonstrating the independent effects of each.
Undiagnosed HIV Infection
Of the 60 men who tested HIV-positive, 58.3% (35 of 60) reported that their last HIV test result was positive, 30.0% (18 of 60) reported that their last test was negative, 5.0% (3 of 60) reported having an HIV test but not the result (1 perceived himself to be positive and 2 did not know their current status), and 6.7% (4 of 60) had never been tested. Therefore, 41.7% of the HIV-positive men were undiagnosed. Of those with undiagnosed HIV, 64.0% (16 of 25) perceived themselves to be HIV-negative, 20.0% (5 of 25) reported that they did not know their current status, and 12.0% (3 of 25) perceived themselves to be HIV-positive (but had last tested negative or not been tested). Overall, more than half of the men with undiagnosed infection (14 men [56.0%]) reported having had an HIV test, that the result of their most recent HIV test was negative, and that they currently perceived themselves to be negative. Of these men, 8 (57.1%) reported that their most recent HIV test was in 2004 or 2005.
The level of undiagnosed infection in the Edinburgh sample was 36.4% compared with 48.1% in the Glasgow sample and 75.0% among men aged <26 years compared with 34.0% among men aged 26 years or older. Only 21.4% of the men who reported that their most recent HIV test was more than 5 years ago were undiagnosed, compared with 44.0% of men tested in the previous year and 45.5% of men tested 1 to 5 years ago. These differences were not statistically significant. There were significant differences in undiagnosed infection by clinic use and STI experience in the previous year; 65.0% of the HIV-positive men who had not used a clinic were undiagnosed compared with 26.3% of men who had used a clinic (OR = 5.20, 95% CI: 1.62 to 16.73); 52.5% of those who had not had an STI were undiagnosed compared with 20.0% of those who had (OR = 4.42, 95% CI: 1.26 to 15.57). There were no significant differences in partner numbers but men who reported UAI with casual partners were significantly less likely to be undiagnosed than men who did not (17.6% compared with 51.2%, respectively, OR = 0.21, 95% CI: 0.05 to 0.82).
Some HIV-positive men who were aware of their status only had UAI with other HIV-positive men. Of the 16 diagnosed HIV-positive men who reported UAI in the previous year, 6 (37.5%) said they always knew their partners' status and that these partners were positive. Risk of other STIs was evident, with 41.2% (7 of 17) of HIV-positive men who reported being diagnosed more than a year ago reporting that they had had an STI in the year since then. Knowledge of partners' HIV status was low among men with undiagnosed infection; 75.0% (6 of 8) of those who reported UAI said that they did not always know their partners' HIV status.
This is the first study to assess HIV prevalence among community samples of gay men in Scotland. It should be noted that this was a bar-based sample, however, so caution should be taken when generalizing to the wider population of gay men. More than three quarters of the men who took part provided oral fluid specimens, demonstrating the acceptability of such a study. There were few differences between the men who did and did not provide samples and, most importantly, there were no differences in sexual risk behavior, HIV testing history, perceived HIV status, or sexual health service use, suggesting that the sample providing specimens was representative of the larger bar-based sample. We have shown that 1 in 25 men tested HIV-positive. Just less than half of these men were undiagnosed. It is also possible that the 5 unconfirmed cases (with 2 positive EIA screens but negative on the Western blot confirmation test) were seroconverting, suggesting that prevalence could be higher. Three main questions arise from these results. What do they add to our understanding of HIV in gay men in Scotland? How comparable are they to other available data? What are the implications for HIV prevention?
These data add to our understanding of HIV among gay men in Scotland by providing previously unknown information on the social and behavioral risk factors for HIV. Age, sexual risk behavior, and sexual health service use were significantly associated with HIV prevalence. Higher prevalence among older men but high levels of undiagnosed infection among younger men (aged <26 years) suggests that both groups require targeted prevention efforts. Increases in sexual risk behavior have been widely reported,11,12,18-20 and although there are now some reports that these have stabilized, levels of sexual risk remain high.21,22 Here, partner numbers were the most important sexual risk factor, and in the final multivariate model, having 10 or more AI partners was particularly significant. Although men who knew their HIV status could have been more likely to use condoms for AI, they and undiagnosed men were just as likely to report high numbers of AI partners. Furthermore, there were low levels of knowledge of partners' HIV status, and for positive men who were aware of their status, there was limited evidence of serosorting (only having UAI with other HIV-positive partners) and evidence of other STI risk. We have previously demonstrated that Scottish gay men exhibited undue confidence that their partners shared the same (assumed) negative HIV status as themselves,11 and there are continuing issues for men around status disclosure and HIV stigma and discrimination.23,24
HIV prevalence was higher among men who had accessed clinics and had STIs in the previous year, but undiagnosed infection was low, as has been noted in similar surveys.9,18 HIV was not confined to these groups, however, and undiagnosed infection was significantly higher among men who had not accessed sexual health services. Until this study, we had no measure of HIV prevalence among men who had never had an HIV test, which is an issue of particular importance given the low uptake of HIV testing among Scottish gay men.11 It was encouraging to discover that prevalence was low, at less than 1%; however, with increased sexual risk behavior among tested and never tested men,11 the potential for further HIV transmission is apparent.
The level of undiagnosed HIV among men who have tested negative in the past and still perceive themselves to be negative is of considerable concern. Perhaps most worryingly, more than half of this prevalent infection was found in men who had tested recently. Similar patterns of undiagnosed infection have been reported in the United States,25-28 where the Centers for Disease Control and Prevention now recommend that sexually active MSM should be tested at least annually.29 It is possible that these men are regular testers (half did report having had 3 or more HIV tests) and could receive a positive diagnosis at some point in the future; however, until that time, they are making incorrect assumptions about their status.
How do our results compare with other prevalence data? Similar community surveys of HIV prevalence among gay men in London, Brighton, and Manchester have been conducted since 2000,9 and although HIV prevalence in Glasgow and Edinburgh was lower, the levels of undiagnosed infection were similar (48.1% in Glasgow and 36.4% in Edinburgh compared with 44.1% in London, 33.3% in Brighton, and 36.7% in Manchester). We compared our results with national HIV prevalence data from the Scottish 2004 unlinked anonymous GUM clinic survey.8 Although prevalence was similar in both surveys (4.4% in our survey compared with 4.2% in the unlinked survey), suggesting consistency between them, undiagnosed infection was substantially higher in the bar survey at 41.7% compared with 27.7% in the unlinked survey (although this difference was not statistically significant). Nevertheless, it is important to note that the Scottish unlinked data come from MSM who attend a GUM clinic, present with a suspected STI, have blood taken for syphilis serology, and do not object to their blood being tested anonymously for HIV. If we look at a more comparable sample from our survey, men who had used a GUM clinic (7.9% prevalence, 26.3% of which was undiagnosed) or had an STI (12.3% prevalence, 20.0% of which was undiagnosed) in the previous year, there were significant differences in prevalence but comparable levels of undiagnosed infection to those of the unlinked survey. Unlinked anonymous HIV testing surveys are an essential part of HIV surveillance and have important public health implications.30 This comparison suggests that the unlinked data represent an accurate estimate of prevalence among gay men but could underestimate the level of undiagnosed infection, and could possibly underestimate prevalence among the clinic population from which the unlinked data come.
Finally, what are the implications of these findings for HIV prevention and sexual health promotion? The sexual behaviors identified here represented a substantial risk for onward transmission of HIV and also placed men at risk of other STIs. Furthermore, the complex issues that surround HIV status disclosure, stigma, and discrimination remain particularly important to HIV prevention in Scotland and must be addressed. Health promotion campaigns should continue to make men aware of these risks. Sexual health services have access to men at high risk for HIV, and they are appropriate locations for continued targeted attempts to increase uptake of testing. Given the known treatment benefits of early diagnosis, there is a need to encourage testing among men not accessing sexual health services and to explore the potential use of alternative community-based HIV testing initiatives. The high level of undiagnosed infection and incorrect assumptions of status among these men are of particular concern, and there is an urgent need for HIV prevention efforts to address these issues, including the promotion of more regular 6-month sexual health checks. There is also a need to understand better what HIV testing means to men and what sexual health and safer sex messages they take away after a negative HIV test result. The design and evaluation of appropriate interventions in community and clinical settings have the opportunity to address these issues and, ultimately, to improve sexual health outcomes for gay men in Scotland.
The authors thank Catherine Ferrell and the survey staff at the MRC Social and Public Health Sciences Unit, Kirsty Roy from Health Protection Scotland, Linda MacDonald from the West of Scotland Specialist Virology Centre, and the field workers employed in each city. Their greatest thanks are to the bar and sauna managers, their staff, and the men who agreed to participate in the survey.
L. Williamson contributed to the design of the study, collected and analyzed the data, and wrote the first and subsequent drafts of the paper. G. Hart initiated and designed the study and commented on each draft of the paper.
1. Health Protection Agency. HIV and AIDS in the United Kingdom quarterly update: data to the end of March 2006. Commun Dis Rep CDR Wkly
2. The UK Collaborative Group for HIV and STI Surveillance. Mapping the Issues. HIV and Other Sexually Transmitted Infections in the United Kingdom: 2005
. London, UK: Health Protection Agency Centre for Infections; 2005:66-75.
3. Lampinen TM, Ogilvie G, Chan K, et al. Sustained increase in HIV-1 incidence since 2000 among men who have sex with men in British Columbia, Canada. J Acquir Immune Defic Syndr
4. Calzavara L, Burchell AN, Major C, et al. Increases in HIV incidence among men who have sex with men undergoing repeat diagnostic HIV testing in Ontario, Canada. AIDS
5. Dukers NH, Spaargaren J, Geskus RB, et al. HIV incidence on the increase among homosexual men attending an Amsterdam sexually transmitted disease clinic: using a novel approach for detecting recent infections [erratum appears in AIDS 2002;16:1707]. AIDS
6. Health Protection Scotland. ANSWER: HIV infection and AIDS: quarterly report to 31st December 2005. HPS Weekly Report
7. Health Protection Scotland. HIV in Scotland review 2005. HPS Weekly Report
8. Health Protection Scotland. Unlinked anonymous HIV testing of genito-urinary medicine clinic attenders in Scotland: data to 2004. HPS Weekly Report
9. Dodds J, Mercey D, Parry J, et al. HIV prevalence and sexual behaviour in community samples of men who have sex with men (MSM) in London, Brighton, and Manchester (UK) [MP-100]. Presented at: 16th Biennial Meeting of the International Society for Sexually Transmitted Diseases Research (ISSTDR); 2005; Amsterdam.
10. Hart GJ, Williamson LM, Flowers P, et al. Gay men's HIV testing behaviour in Scotland. AIDS Care
11. Hart GJ, Williamson LM. Increase in HIV sexual risk behaviour in gay men in Scotland, 1996-2002: prevention failure? Sex Transm Infect
12. Williamson LM, Dodds JP, Mercey DE, et al. Increases in HIV-related sexual risk behavior among community samples of gay men in London and Glasgow: how do they compare? J Acquir Immune Defic Syndr
13. Hart G, Flowers P, Der G, et al. Homosexual men's HIV-related sexual risk behaviour in Scotland. Sex Transm Infect
14. Williamson LM, Hart GJ, Flowers P, et al. The Gay Men's Task Force: the impact of peer education on the sexual health behaviour of gay men in Glasgow. Sex Transm Infect
15. Flowers P, Hart GJ, Williamson LM, et al. Does bar-based, peer-led sexual health promotion have a community-level effect amongst gay men in Scotland? Int J STD AIDS
16. Dodds J, Mercey D, Parry J, et al. Increasing risk behaviour and high levels of undiagnosed HIV infection in a community sample of homosexual men. Sex Transm Infect
17. International Collaboration on HIV Optimism. HIV treatments optimism among gay men: an international perspective. J Acquir Immune Defic Syndr
18. Dodds J, Nardone A, Mercey D, et al. Increase in high risk sexual behaviour among homosexual men, London 1996-8: cross sectional, questionnaire study. BMJ
19. Elford J, Hart G. If HIV prevention works, why are rates of high-risk sexual behavior increasing among MSM? AIDS Educ Prev
20. 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
21. Elford J, Bolding G, Sherr L, et al. High-risk sexual behaviour among London gay men: no longer increasing. AIDS
22. Elford J. Changing patterns of sexual behaviour in the era of highly active antiretroviral therapy. Curr Opin Infect Dis
23. Flowers P, Duncan B, Frankis JS. Community, responsibility and culpability: HIV risk management amongst Scottish gay men. J Community Appl Soc Psychol
24. Flowers P, Knussen C, Church S. Psychosocial factors associated with HIV testing amongst Scottish gay men. Psychol Health
25. Centers for Disease Control and Prevention. HIV prevalence, unrecognized infection, and HIV testing among men who have sex with men-five US cities, June 2004-April 2005. MMWR Morb Mortal Wkly Rep
26. MacKellar D, Valleroy L, Secura G, et al. Unrecognized HIV infection, risk behaviors, and perceptions of risk among young men who have sex with men. J Acquir Immune Defic Syndr
27. Roark RA, Webster RD, Darrow WW, et al. HIV testing among men who have sex with men: how often should one test? J Public Health Manag Pract
28. Valleroy L, MacKellar D, Karon J, et al. HIV prevalence and associated risks in young men who have sex with men. JAMA
29. Centers for Disease Control and Prevention. Sexually transmitted diseases treatment guidelines 2002. MMWR Morb Mortal Wkly Rep
30. Nicoll A, Gill ON, Peckham CS, et al. The public health applications of unlinked anonymous seroprevalence monitoring for HIV in the United Kingdom. Int J Epidemiol