In Britain, sex between men is a major mode of HIV transmission and the incidence of sexually transmitted infections (STI) among men who have sex with men (MSM) is increasing [1–4]. We have previously reported briefly on the overall increase in the prevalence of reported sex between men in Britain in the decade since 1990 . These factors, combined with increases in the prevalent pool of diagnosed HIV-positive MSM, in part, due to the increased life expectancy from highly active antiretroviral therapy , may be increasing the risk of HIV transmission in the general population.
Convenience sample studies of sexual behaviour among self-identified MSM suggest that the prevalence of HIV-risk behaviours is increasing [7–9]. Increases in rates of partner acquisition and reported unprotected anal intercourse (UAI), especially with partners of unknown or discordant HIV status, are of concern for onward transmission of HIV infection [7–9]. However, these studies, which sample from gay venues or from volunteer samples, may not be representative of MSM in the general population. Some studies are only undertaken in London [7–9], where HIV-risk behaviours are generally more prevalent . In addition, some studies have recruited sexually-transmitted disease (STD) clinic attenders [7,8,10] who tend to engage in more HIV/STI risk behaviours than the general population . Moreover, studies that recruit MSM with a close affiliation to, or identification with, the gay community, are unable to assess whether the overall proportion of men in the population who have same-sex experiences is changing, and therefore whether the number of people in the population at risk from homosexually-acquired HIV is changing.
In this study we used data from two large, national probability sample surveys of sexual attitudes and lifestyles carried out in 1990 and 2000. We estimate the prevalence and timing of homosexual experience among British men; describe the sociodemographics of MSM; examine patterns of sexual practices, partnerships and HIV-risk behaviours among MSM; and examine behavioural and attitudinal changes among MSM between 1990 and 2000.
The National Surveys of Sexual Attitudes and Lifestyles (`Natsal') are stratified probability sample surveys of the general population, resident in Britain. Details of the methodology and question wording are published elsewhere [5,11,12]. Briefly, Natsal 1990 interviewed 13 765 people aged 16 to 44 years of whom 6000 were men, and Natsal 2000 interviewed 11 161 people in this age range of whom 4762 were men. Natsal 1990 and Natsal 2000 achieved similar response rates, 63.3 and 65.4%, respectively.
Respondents were interviewed in their homes with a questionnaire consisting of a face-to-face interview carried out by trained interviewers, and a self-completion module containing more sensitive questions, using pen-and-paper interviewing (`PAPI') in 1990 and computer-assisted self-interviewing (`CASI') in 2000. To facilitate comparisons between surveys, questions in Natsal 2000 were identical to those in Natsal 1990 [11,12]. In addition, in Natsal 2000 new questions were added including those on partnership formation and STI diagnoses. Additional questions on condom use were also included in Natsal 2000. In both surveys, respondents with no sexual experience of any kind, and those aged 16 and 17 years with some heterosexual experience but no heterosexual intercourse or homosexual experience reported in screening questions, were not given the self-completion module.
As in previous publications [4,5,11–14] all analyses were performed using STATA version 7.0 to account for stratification, clustering, and weighting of the samples . The data in each survey were weighted to correct for unequal selection probabilities and to match the corresponding age/sex population profile [5,11]. Natsal 1990 data were weighted for differential selection probabilities and then post-stratified to the 1991 census estimates, thereby differing slightly from the method reported in previous publications .
We used logistic regression to obtain age-adjusted odds ratios (OR) to compare estimates between Natsal 1990 and Natsal 2000, controlling for variation in the age distribution between the two surveys. Statistical significance was considered as P < 0.05 for all analyses.
The study was approved by the University College Hospital and North Thames Multi-Centre Research Ethics Committee and all the Local Research Ethics Committees in Britain.
Homosexual experience ever (defined as any experiences with men such as kissing and touching that did not necessarily lead to genital contact) was reported by 8.4% [95% confidence interval (CI), 7.6–9.3%] of men in 2000 compared with 6.0% (95% CI, 5.4–6.7%) of men in 1990 (age-adjusted OR, 1.38; 95% CI, 1.17–1.63; P < 0.0001). As previously reported, 5.4% (95% CI, 4.8–6.1%) of all men surveyed in 2000 reported homosexual intercourse ever (defined as oral or anal sex or any other genital contact with a man) compared with 3.6% (95% CI, 3.1–4.2%) in 1990; this was also a significant increase in comparison with 1990 (age-adjusted OR, 1.47; 95% CI, 1.21–1.80; P < 0.0001) .
Homosexual experience before age 16 years was reported by 4.5% (95% CI, 3.9–5.2%) of men in 2000 in comparison with 3.7% (95% CI, 3.2–4.3%) of men in 1990 (age-adjusted OR, 1.18; 95% CI, 0.95-1.45). In 2000, 2.7% (95% CI, 2.3–3.2%) of men reported homosexual intercourse before age 16 (age at first homosexual intercourse was not asked in Natsal 1990).
A total of 44.7% (95% CI, 38.4–51.1%) of men in 2000 who reported homosexual intercourse ever reported only one male partner ever. Among those men with one partner, 23.4% (95% CI, 15.4–34.1%) reported homosexual anal sex (insertive and/or receptive); 69.1% (95% CI, 58.1–78.3%) reported homosexual oral sex (insertive and/or receptive); and 65.8% (95% CI, 55.5–74.8%) reported other genital contact with a man. Among men who reported homosexual intercourse ever 53.9% had not had a male partner in the past 5 years, and almost all (95.8%; 95% CI, 91.4–98.0%) of these men reported at least one female partner during this time. We assume that some men who reported homosexual intercourse ever may have had a transient experience so we limited further analyses to MSM, defined as men reporting at least one male partner with whom they had genital contact in the 5 years prior to interview. In 2000, 2.8% (95% CI, 2.3–3.3%) of all men were defined as MSM in comparison with 1.5% (95% CI, 1.2–1.9%) of all men in 1990 (age-adjusted OR, 1.83; 95% CI, 1.36–2.47). This prevalence estimate differs slightly from the estimate previously reported , as here, the numerator includes men who reported any partner in the 5 years prior to interview but did not give a precise number, rather than just men reporting a specific number of partners.
The mean age of MSM in Natsal 2000, 31.5 years, was similar to the mean age of non-MSM men, 31.0 years. A total of 28.4% (95% CI, 21.0–37.1%) of MSM reported currently cohabiting with a male partner and a further 54.8% (95% CI, 46.3–63.0%) described themselves as ‘single, never married'. MSM were similar to the general population in terms of ethnicity, with 94.0% (95% CI, 89.4–96.6%) white, compared with 91.3% (95% CI, 90.3–92.0%) white of non-MSM. The number of MSM who had a university degree (38.5%; 95% CI, 30.2–47.5%) was significantly greater than for non-MSM (20.9%; 95% CI, 19.5–22.3%; P < 0.0001). A total of 30.1% (95% CI, 23.5–37.5%) of MSM lived in Greater London compared with 14.4% (95% CI, 13.3–14.7%) of non-MSM (P < 0.0001).
The distribution of partnerships reported by MSM in Natsal 2000 was highly skewed (Table 1). The median number of male partners in the past 5 years reported by MSM was four, but the mean was 24.1 partners. One-third of MSM reported one male partner in the past 5 years whereas nearly one in ten reported at least 50 male partners in this time. The median reported number of male partners for the past 5 years in 2000 was double the 1990 estimate (four versus two), although the corresponding age-adjusted OR for the change in partner numbers was not statistically significant (1.19; 95% CI, 0.67–2.12). A total of 43.2% of MSM reported at least one female partner in the past 5 years.
In the past year, 77.8% of MSM reported having had at least one male partner. The median reported number of male partners in the past year was one in both 1990 and 2000. Of note, although not significant, was the proportion of MSM reporting at least 10 partners in the past year, which has increased from 4.9% (95% CI, 2.0–11.3%) in 1990 to 15.1% in 2000 (age-adjusted OR, 2.70; 95% CI, 0.93–7.88). One-third of MSM reported at least one female partner in this time.
In 2000, half the MSM reported at least one new male partner in the past year. In the 4 weeks prior to interview, 24.0% (95% CI, 17.5–32.5%) of MSM reported at least one new male partner. A total of 53.2% (95% CI, 44.6–61.7%) of MSM reported intercourse within 1 week of meeting their most recent partner; 35.0% (95% CI, 27.6–43.3%) reported intercourse within 24 hours.
More than half the MSM in 2000 reported experiencing each of the sexual practices in Table 2 in the past year. The proportion of MSM reporting each practice in the past year was greater in 2000 compared with 1990. Over half of MSM in 2000 reported receptive anal sex in the past year, which was significantly greater than in 1990. In 2000, 59.8% of MSM who reported anal sex in the past year reported UAI in this time, and 48.4% of MSM who reported anal sex in the past 4 weeks reported UAI in this time. We found no significant association between condom use at last intercourse and partnership type. In all, 38.1% (95% CI, 30.2–46.6%) of MSM reported their most recent partner as not regular and 38.4% (95% CI, 27.1–51.2%) of these men reported condom use at last intercourse in comparison with 37.8% (95% CI, 27.4–49.5%) of MSM who described their most recent partner as regular. We also found no significant association between condom use and the time between meeting and first intercourse with 34.3% (95% CI, 27.1–42.3%) of MSM reporting intercourse within 24 hours of meeting their most recent partner. A total of 48.5% (95% CI, 35.1–62.1%) of these men reported using condom(s) on this first occasion in comparison with 48.0% (95% CI, 36.9–59.3%) of MSM who had intercourse at least 24 hours after meeting.
Despite some increases in risk behaviour, there has been no change in the proportion of MSM who perceive themselves as ‘greatly’ or ‘quite a lot’ at risk of HIV (12.6%; 95% CI, 8.0–19.2%, of MSM in 2000 compared with 12.6%; 95% CI, 7.4–20.7%, of MSM in 1990).
There has been no significant increase since 1990 in the proportion of MSM reporting, in the past 5 years, having tested for HIV (36.9%; 95% CI, 28.8–45.9% in 2000 compared with 32.9%; 95% CI, 23.0–44.7% in 1990), or having attended a STD clinic (35.6%; 95% CI, 27.2–44.9% in 2000 compared with 29.6%; 95% CI, 20.1–41.2% in 1990). However, the proportion of all men who are MSM and who attended a STD clinic in the past 5 years has significantly increased between 1990 and 2000 (age-adjusted OR, 2.18; 95% CI, 1.26–3.79). In 2000, 10.4% (95% CI, 6.5–16.1%) of MSM reported STI diagnosis/es in the past 5 years in comparison with 2.7% (95% CI, 2.2–3.3%) of non-MSM (P < 0.0001). Questions about STI diagnosis were not asked in Natsal 1990.
There has been a significant increase in the proportion of MSM who consider that homosexual sex ‘is not wrong at all’ from 74.1% (95% CI, 63.1–82.6%) in 1990 to 88.7% (95% CI, 82.6–92.8%) in 2000. We also observed a highly significant increase in the proportion reporting this opinion among non-MSM from 21.2% (95% CI, 20.1–22.5%) in 1990 to 34.8% (95% CI, 33.2–36.5%) in 2000.
Data from large-scale, probability sample surveys of the general population provide evidence that the proportion of British men having sex with men has increased between 1990 and 2000. Among MSM, the prevalence of some risk behaviours has increased since 1990, which is of concern for continuing HIV transmission given that STI incidence is increasing, HIV prevalence is increasing due to improved survival, and antiretroviral resistance is increasing.
The Natsal surveys are unique in having a quantifiable sampling frame from which we can derive estimates of the population at risk and estimate change over time. This contrasts with surveys focussing exclusively on gay men who are typically drawn from gay venues, STD clinics, and internet chat-rooms [7–10,16–19]. Although these community surveys include larger samples of MSM than Natsal, they do not provide information on the size of the population at risk. In addition, comparability over time is compromised if the population attending services or using gay venues alters over time. It is interesting to note then that increases in the reporting of risk behaviours among MSM in the 1990s have been observed in both Natsal and gay community surveys.
It is important to acknowledge that the increases in the reporting of risk behaviours presented here, and in other studies, may not be entirely due to changes in behaviour but may, in part, result from changes in respondents’ willingness to report such socially-censured behaviours. Changes in two factors in particular might be expected to have increased this willingness to report. First, attitudinal changes may be important, and we have shown here a substantial increase in the reported acceptance of homosexuality among MSM, as well as the general population. Second, an increased willingness to report may have also been facilitated by the improvements in the Natsal survey methodology from using PAPI in 1990 to using CASI in 2000 [5,11,12]. However, from a feasibility study we found no overall significant difference in the rates of reporting sensitive behaviours using CASI in comparison to PAPI . Clearly, definitive estimation of the change in willingness to report between repeated surveys is never possible, and so it is not possible to reliably correct for this to arrive at an estimate of the population change. Elsewhere, we have attempted to explore this issue for Natsal [13,14], and concluded that based on the Natsal data alone we cannot be absolutely certain of population change in the prevalence of MSM. However, in light of the additional evidence of increasing STI incidence, behavioural surveillance data [1–4,7–9] and the increased prevalence of MSM estimated from the US’ General Social Surveys from 1988–2000 for which the study methodology was constant , some increase in the population numbers of MSM and/or the prevalence of risk behaviours among MSM is overwhelmingly likely.
These results are of considerable concern given evidence of continuing HIV transmission among MSM  and the growing HIV disease burden in the population. Because these results are drawn from general population surveys, they emphasize that not only are risk practices increasing among MSM, but that our estimate of the number of MSM in the population needs revising upwards, indicating that a higher proportion of men are at risk of homosexually-acquired HIV. These data emphasize the need to maintain investment in, and to continue to investigate the effectiveness of, risk-reduction programmes among MSM.
Catherine Mercer was the lead writer of this paper and undertook all the statistical analyses. Anne Johnson, Kevin Fenton, Bob Erens and Kaye Wellings were principal investigators and participated in the design and management of the main study and preparation of this manuscript. Andrew Copas, Sally McManus, Kiran Nanchahal and Wendy Macdowall contributed to the drafting of the manuscript.
We would like to thank the study participants, the team of interviewers and operations, and computing staff from the National Centre for Social Research who carried out the interviews.
Sponsorship: The study was supported by a grant from the Medical Research Council with funds from the Department of Health, the Scottish Executive and the National Assembly for Wales.
1. Health Protection Agency (HIV/STI Division, Communicable Disease Surveillance Centre) and the Scottish Centre for Infection and Environmental Health. AIDS/HIV Quarterly Surveillance Tables
No. 58:03/1. London: Health Protection Agency; 2003.
2. Public Health Laboratory Service, DHSS and PS, Scottish ISD(D)S Collaborative Group, Trends in Sexually Transmitted Infections in the United Kingdom, 1990 to 1999
. London: Public Health Laboratory Service; 2000.
3. Gill N. Syphilis transmission in homo/bisexual men: new outbreak in London, continuing outbreak in Dublin.Eurosurveillance Weekly
June 2001; 26.
4. Fenton KA, Mercer CH, Johnson AM, Korovessis CJ, McManus S, Erens B, et al
. Reported STD clinic attendance and sexually transmitted infections in Britain: prevalence, risk factors, and proportionate population burden.J Infect Dis
5. Johnson AM, Mercer CH, Erens B, Copas AJ, McManus S, Wellings K, et al
. Sexual behaviour in Britain: partnerships, practices, and HIV-risk behaviours.Lancet
6. Palella FJ, Delaney KM, Moorman AC, Loveless MO, Fuhrer J, Satten GA, et al.Declining morbidity and mortality among patients with advanced human immunodeficiency virus infection.N Engl J Med
7. Dodds JP, Nardone A, Mercey DE, Johnson AM. Increase in high risk sexual behaviour among homosexual men, London 1996-8: cross sectional, questionnaire study.BMJ
8. Dodds JP, Mercey DE. London Gay Men's Survey: 2001 results
. London: Department of Sexually Transmitted Diseases, Royal Free and University College Medical School; 2002.
9. Elford J, Bolding G, Sherr L. Gay Men's Survey in London Gyms
. City University Report. London: City University; 2002.
10. Dukers NHTM, Spaargaren J, Geskus RB, Beijnen J, Coutinho RA, Fennema HAS. HIV incidence on the increase among homosexual men attending an Amsterdam sexually transmitted disease clinic: using a novel approach for detecting recent infections.AIDS
11. Erens B, McManus S, Field J, Korovessis C, Johnson AM, Fenton KA, et al
. National Survey of Sexual Attitudes and Lifestyles II: Technical Report.
London: National Centre for Social Research; 2001.
12. Johnson AM, Wadsworth J, Wellings K, Field J. Sexual Attitudes and Lifestyles
. Oxford: Blackwell Scientific Press; 1994.
13. Copas AJ, Wellings K, Erens B, Mercer CH, McManus S, Fenton KA, et al
. The accuracy of reported sensitive sexual behaviour in Britain: exploring the extent of change 1990–2000.Sex Transm Inf
14. Copas AJ, Farewell VT, Mercer CH, Yao G. The sensitivity of estimates of the change in population behaviour to realistic changes in bias in repeated surveys.Royal Statistical Society Series A
15. StataCorp Stata Statistical Software: Release 7.0
. Texas: Stata Corporation; 2001.
16. Hickson F, Reid D, Weatherburn P, Stephens M, Brown D. Time for More: Findings from the National Gay Men's Sex Survey 2000
. London: Sigma Research; 2001.
17. Reid D, Weatherburn P, Hickson F, Stephens M. Know the Score: Findings from the National Gay Men's Sex Survey 2001
. London: Sigma Research; 2002.
18. Hart G, Flowers P, Der GJ, Frankis JS. Homosexual men's HIV related sexual risk behaviour in Scotland.Sex Transm Infect
19. Weatherburn P, Stephens M, Reid D, Hickson F, Henderson L Brown D. Vital Statistics Scotland 2001: Findings from the Gay Men's Sex Survey.
London: Sigma Research; 2002.
20. Johnson AM, Copas AJ, Erens B, Mandalia S, Fenton KA, Korovessis C. et al
. Effect of computer-assisted self interviews on reporting of sexual HIV-risk behaviours in a general population sample: a methodological experiment.AIDS
21. Anderson JE, Stall R. Increased reporting of male-to-male sexual activity in a national survey.Sex Transm Dis
22. Department of Health. Prevalence of HIV & Hepatitis infection in the United Kingdom 2001 – annual report of the Unlinked Anonymous Prevalence Monitoring Programme. London: DoH; 2002.