Numbers of gay men and other men that have sex with men (MSM)
Our population of concern is gay men and and other men who have sex men (MSM), currently and in the future. At mid-2009 the Office for National Statistics estimated that there were 20,170,000 men aged 16 and over normally resident in England. The number in the UK was 24,021,600. In the second National Survey of Sexual Attitudes and Lifestyles (whose fieldwork took place in 2000) the proportion of men aged 16-44 who had genital contact with another man in the last five years was 2.8% (95% confidence range 2.3-3.3%, Mercer et al. 2004).This is likely to be an under-estimate of the true figure due to underreporting but because it relates to men aged 16-44, it is likely to be lower for all adult males as sexual activity declines with age.
In the absence of further data to estimate the size of these errors, we take this figure of 2.8% as being the proportion of adult males who are homosexually active. This suggests there are 564,760 (463,910-665,610) MSM in England and 672,605 (552,499-792,713) MSM in the UK.
Prevalence of HIV among MSM
The Health Protection Agency (2011) recently estimated that there were 40,100 (35,300 - 46,700) MSM living with HIV infection in the UK in 2010, including all MSM both diagnosed and undiagnosed. This suggests that the overall prevalence of HIV infection among MSM in the UK in 2010 was 7.1% (5.3% - 10.1%).
The same report (Health Protection Agency 2011) estimated 10,300 (credible interval 5,500 - 16,800) MSM with undiagnosed HIV in the UK in 2010. This equates to just over a quarter (26%) of MSM with HIV in the UK not knowing they are infected (credible interval 16% to 36%). This suggests the overall prevalence of undiagnosed HIV infection was in the range 0.8% - 3.6%. In comparison, residual syphilis blood samples from MSM attending sentinel GUM clinics across the UK during 2009 measured 2.4% of MSM having previously undiagnosed HIV infection (Health Protection Agency 2009).
The prevalence of HIV among MSM varies by age, geography, ethnicity and social class. Although approximately half of MSM with HIV in the UK live in London, over the last few years rises in HIV prevalence have been greater outside London.
HIV deaths and diagnoses
This graph shows the number of MSM with HIV who have died each year in England. The rising tide of deaths was severely curtailed by the introduction of anti-retroviral therapy (ART) in the mid-1990s. However, around 200 MSM still die each year with HIV across the UK (Health Protection Agency Centre for Infections 2010). The majority of these deaths occur because men were diagnosed too late for treatment to be effective (Chadborn et al. 2005). Reducing late diagnoses is therefore a major route for increasing the health of men with HIV.
The graph also shows the number of new HIV diagnoses made in England each year among MSM. Since 2003 there have been substantial and undiminished levels of newly diagnosed HIV (and other STIs) in MSM. The pattern of declining HIV diagnoses in MSM in the late 1990s followed by a consistent increase has been seen across Western Europe, North America and Australia (Sullivan et al. 2009).
Across the UK in 2009 there were 1,497 MSM diagnosed with HIV acquired in the UK, a further 314 MSM diagnosed with HIV acquired abroad (including MSM who moved to the UK with HIV), and 660 MSM diagnosed with HIV whose country of acquisition was not determined, making a total of 2,471 men joining the diagnosed positive population that year (Health Protection Centre for Infections 2010).
Length of time spent undiagnosed
It is important to distinguish these diagnosis events from the infection events that preceded them. All men who acquire HIV spend some time with undiagnosed infection before being diagnosed. The profile of this length of time is poorly understood. However, counts of CD4 cells in the blood can be used as a marker for this length of time. The longer people have untreated HIV, the lower their CD4 cell count falls. So a higher CD4 count suggests a shorter period of time with HIV.
The Health Protection Agency collate CD4 count data at diagnoses of HIV. The average (median) CD4 count among MSM in the UK has been rising for some years (see graph). Correspondingly, the proportion of MSM diagnosed with HIV who have a CD4 count below 200 (the definition of ‘late’ diagnosis) has been falling and the proportion that are of recent infection has been rising (Fisher et al. 2007). This suggests that the average length of time spent undiagnosed has been getting shorter. Among MSM diagnosed with HIV in the UK in 2006 (the most recent year for which an estimate was made) the median length of time spent living with HIV before diagnosis was estimated at 4.7 years (Tim Chadborn, Health Protection Agency, personal communication to Ford Hickson, May 2008).
The Recent Infection Testing Algorithm (RITA), which can distinguish recently acquired infection (within the last six months) from longer standing infections, suggest around 20% of MSM diagnosed with HIV in 2008 were within the first six months of their infection (Health Protection Agency 2009). Ideally, this figure should be 100%, that is, all men who acquire HIV being diagnosed very promptly.
It is problematic to use changes in the rate of diagnoses as an indicator of change in HIV incidence as it is also influenced by changes in the length of time between infection and diagnosis. Recent years have seen a large increase in the number of HIV tests being taken by MSM, as well as the proportion of MSM who test.
Serological testing of men attending GUM clinics with previously undiagnosed HIV infection is able to distinguish recent from longer standing infections. In 2007 a survey (Health Protection Agency 2009) estimated HIV incidence among MSM clinic attenders to be in the range 1.0% to 3.4%.
The large increase in HIV diagnoses over the past ten years is in part the result of changes in HIV testing policy and practices (Dougan et al. 2007). This effect is unsustainable as the undiagnosed population is diminished and the average length of time spent undiagnosed is reduced. The increase in diagnoses may also be a result of increase in HIV incidence, but the contribution of these two factors to the overall increase in diagnoses is not clear. If the former is the predominant explanation, then diagnoses should soon start dropping.
On the other hand, although there is no direct evidence of significant change in HIV incidence among all MSM in the UK since 2003, the increase in diagnoses has also occurred among men under the age of 25 and among men under the age of 20 (see graph below). These are much more likely to be recent infections.
The lack of direct evidence for an increase in incidence may be a reflection of the lack of a sufficiently powerful system for measuring it. This may change in the near future with the introduction of RITA (Health Protection Agency 2009) which was planned to reach national coverage by the end of 2010.
For HIV transmission to occur between individuals, a number of specific behavioural and biological conditions must be met. These are the primary causes of HIV transmission. The combination of these conditions, result in HIV incidence in the population. The relative contributions of various primary causes vary among different populations across the globe.
This section describes the conditions we think are contributing to new infections among MSM in the UK. We acknowledge that there is a limited consensus about several of the issues raised here. The purpose of laying down the arguments is to better understand them and to encourage investigations in order to provide a firmer evidence base for our theorising.
The parameters of the challenge
The figure below illustrates HIV prevalence and HIV incidence in a population, here gay men and other men that have sex with men (MSM) in England. The large triangle represents men in England who have sex with men. Men are joining and leaving this population all the time as they become homosexually active, finish their sex life with men (through choice, circumstance or death) and as they enter and leave the country.
The large inner triangle at the top represents those MSM who have HIV infection (and the remaining rhomboid below are those who are uninfected). The smaller triangle to the right are those MSM with HIV who have not yet had it diagnosed.
Men join the HIV infected MSM population when uninfected MSM acquire HIV (incidence, shown by the arrow from uninfected to undiagnosed infected) and when MSM with HIV move into the country (these men with diagnosed HIV are retested and reported as a case new to the UK and are shown as joining the undiagnosed group in the first instance).
Men move from the undiagnosed infected MSM group to the diagnosed infected MSM group by being diagnosed (shown by the arrow). Men can only leave the diagnosed infected MSM group by stopping having sex with men, dying or leaving the country.
The number of MSM living with HIV in the UK has only ever increased. This is because the rate at which MSM have become infected with HIV has always been greater than the rate at which MSM with HIV have died.
The figure concerns HIV transmission during sex between men. MSM who are also injecting drug users (IDU) have a much higher incidence of HIV infection than MSM who are not IDU. On acquiring HIV MSM-IDU also have a worst prognosis. MSM-IDU are a relatively small group very disproportionately suffering harm from HIV.
Page last updated: 26 March 2012