Increased high-risk sexual behavior among gay men has been widely reported.1-4 This places gay men at risk of HIV and other sexually transmitted infections (STIs); indeed, increases in transmission of HIV, syphilis, and gonorrhea in particular have been reported among UK gay men.5 Although higher levels of sexual risk behavior may be common across different locations and communities, there may be city-specific factors associated with such increases. Sexual health interventions shown to be effective in one setting may not necessarily transfer to another, and understanding city-specific differences could aid the development of future interventions.2
In this paper, we describe trends in HIV-related sexual risk behavior among men recruited from England s largest city (London) and the largest city in Scotland (Glasgow), using comparable surveys of social venues.1,4 We examine changes in unprotected anal intercourse (UAI), number of UAI partners, and UAI with partners of unknown/discordant HIV status. Our aim is to determine whether city-level differences could alert us more specifically to the prevention needs of communities at risk and to inform the health promotion efforts of the agencies that work with them.
In London, data were collected in the annual sexual health survey of gay social venues, during which a representative sample of bars, clubs, and saunas were surveyed (58venues in 1996, 49 in 1999, and 54 in 2002).6 In Glasgow, surveys of gay men were conducted in 1996, 1999, and 2002, using time and location sampling to recruit representative samples of men visiting the commercial gay scene (5 exclusively gay bars in 1996 and 1999; 6 in 2002).7 Anonymous, self-complete questionnaires were distributed and collected by fieldworkers trained to administer them in these settings. Ethical approval was granted, in Glasgow by the University of Glasgow Ethics Committee for Non-clinical Research Involving Human Subjects and in London by the University College London/University College London Hospitals Committees on the Ethics of Human Research.
Comparable data were collected on demographics, health-seeking behavior [HIV testing, STIs, and genitourinary medicine (GUM) clinic use], and recent (last year) sexual behavior (UAI, number of partners, and knowledge of own and partner s HIV antibody status; UAI with partners of unknown/discordant HIV status was based on any UAI in the last year at which own or partner s status was unknown or discordant). The Pearson χ2 test was used for bivariate comparisons, and multiple logistic regression was used to produce adjusted odds ratios and to assess their significance. Differences between and within cities were examined, and significant results are reported throughout.
A total of 8247 men completed the questionnaires (response rates in brackets). In 1996, 1895 (75%) men participated in London and 1245 (77%) men participated in Glasgow; 1368 (74%) in London and 1442 (75%) in Glasgow in 1999; and 1325 (77%) in London and 972 (63%) in Glasgow in 2002.
Survey demographics, health-seeking behavior, and sexual behavior are shown in Table 1. Glasgow men were consistently younger than the London sample (P < 0.01). Whereas the proportion of 15- to 25-year-old men decreased in London over time, it increased in Glasgow. More London men than Glasgow men were currently employed (P < 0.05), but the proportion employed increased 'in Glasgow.
Levels of HIV testing were considerably higher in London than in Glasgow (P < 0.001), and whereas testing increased in London, there was no change in Glasgow. In 2002, more than 10% of London men who had been tested reported being HIV positive compared with 7% of Glasgow men. Sexually transmitted infections in the previous year were consistently higher among London men (P < 0.001) but increased significantly among Glasgow men.
In both cities, the number of AI and UAI partners increased; London men reported higher numbers of partners than Glasgow men (P < 0.01). The proportion of men reporting UAI with partners of unknown/discordant HIV status increased in both cities. Overall, UAI with partners of unknown/discordant status was higher in Glasgow (P < 0.001). Unprotected anal intercourse with more than 1 partner increased significantly in both samples. Comparisons of sexual behavior data from the 2 cities demonstrate increasing partner numbers and trends toward higher sexual risk behavior (significant behavior change was not apparent in Glasgow until 2002).
Multivariate logistic regression was used to assess the trends in HIV testing, UAI, UAI with partners of unknown/discordant status, and UAI with more than 1 partner (controlling for age and employment status, and HIV testing in the sexual behavior models) (Table 2). Using 1996 as the baseline, the odds of ever having had an HIV test increased significantly in London in 2002, but there was no significant increase in Glasgow. The odds of any UAI, of UAI with partners of unknown/discordant status, and of UAI with more than 1 partner increased significantly in 1999 and 2002 in London. In Glasgow, there was no change in each of the risk behaviors between 1996 and 1999, but there were significant increases in 2002. The increases in risk behavior were independent of HIV testing status.
In this paper, we compared data from more than 8000 men surveyed in social venues in London and Glasgow in 1996, 1999, and 2002 to demonstrate that there have been significant changes in HIV-related sexual risk behavior. Although the same pattern of increase was apparent in the cities, there were some differences between them that are worth highlighting. Firstly, the increases began earlier in London than in Glasgow; secondly, HIV testing levels were considerably lower in Glasgow; and finally, although overall risk levels were higher in London, UAI with partners of unknown/discordant HIV status was higher among Glasgow men.
In London, significant increases in sexual risk behavior were apparent in 1999, but in Glasgow, there was no change in risk between 1996 and 1999. During that time, a specific sexual health promotion intervention was targeted to gay men in Glasgow-the Gay Men s Task Force. Our evaluation of the Gay Men s Task Force demonstrated that it had little effect on behavior 6 months after its end, but there was certainly no increase in risk behavior.7,8 Nevertheless, by 2002, sexual health promotion with gay men had been considerably reduced in Glasgow, and we have argued elsewhere that this may have contributed to the increase in levels of risk reported.4
In 2002, less than half of the men in Glasgow had had an HIV test compared with three quarters of London men. In each of the surveys, testing levels were higher in London and increased significantly, whereas testing levels in Glasgow did not change. Free and confidential HIV testing is widely available in both cities, and the number of gay men being tested at GUM clinics has actually increased in Glasgow, partly as a result of clinics recommending rather than only offering testing.9 However, the increase at the clinic level has not translated into an increase at the community level. Barriers to HIV testing among Scottish gay men have been the focus of attention, and it is recognized that there is a need to combat the stigma and discrimination that still surround HIV.10 This remains a specific challenge for sexual health promotion in Glasgow.
Overall, levels of sexual risk behavior were higher in London than in Glasgow. Men in London reported more UAI and greater numbers of partners, but Glasgow men reported more partners of unknown/discordant HIV status. It is possible that this is a direct result of the lower HIV testing levels in Glasgow. Never-tested men have previously been reported to be more likely to assume that their partners are HIV negative.10 Never-tested men were more likely to have discordant partners, but around half of tested men reported nonconcordant UAI and so were also at risk.
Glasgow and London have distinct gay communities and very different commercial gay scenes (6 exclusively gay bars in Glasgow compared with more than 100 in London), yet our results demonstrate similar patterns of change occurring among bar-based samples of gay men. Nevertheless, there are some factors distinguishing the 2 cities that argue for quite different interventions. Glasgow has consistently lower levels of HIV testing than London, resulting in higher rates of UAI with partners of unknown/discordant HIV status. The interpretation of this sexual risk behavior is determined by HIV status, a factor that will influence intervention design. This highlights the need for prevention work to target the continued stigma and discrimination surrounding HIV testing and to encourage HIV testing in Glasgow. The rate of increase in UAI with partners of unknown/discordant status was higher in London, suggesting that this behavior requires highly focused health promotion. It has been shown that the effectiveness of interventions is influenced by setting, place, time, and other demographic and cultural factors.2 The transferability of sexual health interventions remains uncertain, and recognition of such differences could aid the development of more effective interventions. Regular behavioral surveillance combined with data on incident HIV and other STIs should inform the development of appropriate interventions at the city level.
The UK Medical Research Council funded the Glasgow surveys, and the Camden and Islington Primary Care Trusts and the Department of Health funded the London surveys. Paul Flowers and Jamie Frankis contributed to the 1996 and 1999 Glasgow surveys, and Anthony Nardone contributed to the 1996 London survey. Venue managers, staff, and most importantly their customers are due our deepest gratitude for their willingness to complete questionnaires.
1. 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
2. Elford J, Hart G. If HIV prevention works, why are rates of high-risk sexual behavior increasing among MSM? Aids Educ Prev
3. 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
4. Hart GJ, Williamson LM. Increase in HIV sexual risk behaviour in gay men in Scotland, 1996-2002: prevention failure? Sex Transm Infect
5. Health Protection Agency, SCIEH, ISD, National Public Health Service for Wales, CDSC Northern Ireland, and the UASSG. Renewing the focus. HIV and other Sexually Transmitted Infections in the United Kingdom in 2002
. London: Health Protection Agency; 2003.
6. Nardone A, Dodds J, Mercey D, et al. Active surveillance of sexual behaviour among homosexual men in London. Commun Dis Public Health
7. Flowers P, Hart G, Williamson L, et al. Does bar-based, peer-led sexual health promotion have a community-level effect amongst gay men in Scotland? Int J STD AIDS
8. Williamson L, Hart G, 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
9. SCIEH. Focus: HIV prevalence among non-IDU heterosexuals, IDUs and men who have sex with men who undergo named HIV testing in Scotland: 2003 update. SCIEH Weekly Report
10. Flowers P, Knussen C, Church S. Psychosocial factors associated with HIV testing amongst Scottish gay men. Psychol Health