The prevalence of HIV-1 infection among homosexual and bisexual men attending genitourinary medicine clinics, measured through anonymised testing of samples taken to test for syphilis, has declined in recent years.1 This decline may be spurious, however, because the proportion of such men who have serum samples taken to test for syphilis has declined and because specialist HIV services have drawn patients infected with HIV away from the genitourinary medicine clinics that participate in seroepidemiological surveillance.2 We therefore studied the prevalence of HIV-1 in homosexual and bisexual men without a diagnosis of HIV-1 infection who presented with a new episode of an acute sexually transmitted infection: this group is likely to attend genitourinary medicine clinics and undergo testing for syphilis.
Methods, and results
A continuing survey in England and Wales in 15 genitourinary medicine clinics uses specimens left over after testing for syphilis.3 The data collected (time and place of attendance, the patient's characteristics and risks for HIV-1 infection, and the presence of HIV-1 infection or an acute sexually transmitted infection) and the remains of the specimen are unlinked and anonymised before testing for HIV-1. Infections probably acquired through unprotected sex, such as gonorrhoea, chlamydia, and first episodes of viral sexually transmitted infections, were categorised as acute sexually transmitted infections. Trends in the prevalence of HIV-1 infection among homosexual and bisexual men were adjusted for clinic, age group, injecting drug use, and presence of acute sexually transmitted infection and analysed by multivariable logistic regression in GLIM-4 (Numerical Algorithms Group, Oxford). The unit of analysis was each new attendance per quarter. Repeat attendances are not identified but are unlikely to exceed 3% per six month period (Communicable Disease Surveillance Centre, unpublished data).
Altogether, 32 006 attendances were analysed. In London, the observed prevalence of HIV-1 declined among all patients presenting with an acute sexually transmitted infection (figure), from 16.5% (110/667) in 1993 to 9.0% (123/1373) in 1998, an adjusted average decrease of 9% a year (95% confidence interval decrease 1% to 16%). When specimens from men known to be HIV-1 positive were excluded there was no adjusted trend in prevalence, which was 4.9% (64/1314) in 1998. The adjusted decline in prevalence of HIV-1 among homosexual and bisexual men presenting with non-acute sexually transmitted infections was significant: 17% a year (12% to 21%) overall and 11% a year (7% to 16%) when men known to be HIV-1 positive were excluded.
Outside London, the observed prevalence of HIV-1 among homosexual and bisexual men was lower (figure), and the adjusted decrease in prevalence overall was significant only in men presenting with non-acute sexually transmitted infections (average decrease of 11% a year (4% to 18%)).
The proportion of all attenders with an acute sexually transmitted infection increased over time in London, but not outside London. This trend was less apparent when attenders known to be HIV-1 positive were excluded.
Comment
In 1993-8 the prevalence of undiagnosed HIV-1 infection in homosexual and bisexual men presenting with acute sexually transmitted infection did not fall, which indicates a high level of continuing transmission. The smaller increase in this group of men in London is consistent with a progressive drawing away of men infected with HIV from participating clinics or a reduction in syphilis testing of HIV infected attenders without an acute infection.
The difference in trends between men with and without acute sexually transmitted infections may be partly because of repeat sampling of a core group of HIV-1 infected men with repeated new infections. If so, it will represent a marker of significant risk for HIV-1 transmission in the population. These men probably have more partners and engage in riskier sex than those without an acute sexually transmitted infection.4 The facilitatory effect of many acute infections on HIV-1 transmission5 may also have contributed to the higher prevalence. Health promotion directed at this group of men should be intensified.
Figure.
Unadjusted prevalence of HIV-1 for homosexual and bisexual men by year of attendance
Footnotes
Funding: Department of Health.
Competing interests: None declared.
References
- 1.Unlinked Anonymous HIV Surveys Steering Group. Prevalence of HIV in the United Kingdom, data to end 1998. London: Department of Health; 1999. [Google Scholar]
- 2.Nardone A, Rogers P, Joyce C, Nicoll A, Mercey D. The interpretation of trends in HIV seroprevalence amongst homo/bisexual men attending a GUM clinic in central London. MRC AIDS Research Workshop Abstract Book. 1995;4-5:29. [Google Scholar]
- 3.Simms I, Rogers P, Catchpole M, McGarrigle CA, Nicoll A.on behalf of the collaborative group. Trends in undiagnosed HIV-1 infection among attenders at genitourinary medicine clinics: 1990-6 Sex Transm Inf 199975332–336. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Catchpole M, Connor N, Brady A, Kinghorn G, Mercey D, Band B, et al. Behavioural and demographic characteristics of attenders at two genitourinary medicine clinics in England. Genitourin Med. 1997;73:457–461. doi: 10.1136/sti.73.6.457. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Grosskurth H, Mosha F, Todd J, Mwijarubi E, Klokke A, Senkoro K, et al. Impact of improved treatment of sexually transmitted diseases on HIV infection in rural Tanzania: randomised controlled trial. Lancet. 1995;346:530–536. doi: 10.1016/s0140-6736(95)91380-7. [DOI] [PubMed] [Google Scholar]