About four in every ten gay men in the UK who have HIV are undiagnosed, the British HIV Association’s Autumn Conference heard on 10 October, for the most part because they do not test often enough rather than because they refuse to test.
The conference also heard details of mathematical modelling which suggests that almost half of the undiagnosed gay men in the UK have acquired HIV in the past year.
One presenter suggested that clinics should recall gay men at least annually for an HIV test.
The conference also heard that an even higher proportion of African people in the UK who have HIV - over six in ten - are undiagnosed. The main reason for this is because they have not tested in their country of origin, often arrive with low CD4 counts and tend not to test until they have symptoms.
The last session of the conference was a joint symposium shared with the British Association for Sexual Health and HIV (BASHH) and sought to unravel the mystery of why so many people remain undiagnosed.
Professor Andrew Phillips of the Royal Free & University College Medical School in London outlined the problem. A recent survey (Williamson 2008) of men in gay venues in five UK cities (London, Brighton, Manchester, Glasgow and Edinburgh) had found that of the 9% who tested positive when their saliva was tested anonymously for HIV antibodies, 40% were unaware of their infection; of these, nearly two-thirds thought they were HIV-negative.
The last annual report by the Health Protection Agency on HIV diagnoses in the UK found that, paradoxically, although the proportion of gay men who take an HIV test when they go for a sexual health check has increased from 45% to 85% in the last ten years, the proportion who walk away with their HIV infection undiagnosed has only declined from 65% to 45% in the same period. This means that it is the gay men who are most likely to have HIV who are least likely to test.
Why is this? There are three possible explanations:
- Some gay men who say they are HIV-negative and refuse a test actually know they have HIV and lie about it, possibly because they fear being stigmatised or even prosecuted if they reveal unsafe sex;
- A sub-population of gay men who are at high risk of HIV are so anxious about it that they actively avoid testing;
- HIV incidence (the new infection rate) among some gay men is so high that taking a test every few years is failing to detect recent infections.
Data that would prove which of these explanations is most pertinent are unavailable, but Prof. Phillips said that a mathematical model which calculates the rates of new infection and new diagnoses in gay men suggests that the third factor is the predominant one. His model found that nearly half (46%) of gay men with HIV who remain undiagnosed would have been infected this year, and only 15% before 2004.
“People are not necessarily resistant to testing, they just need it frequently,” he said.
In contrast, 55% of undiagnosed African people would have been infected before 2004, with diagnosis rates influenced more by the date of arrival in the UK than by the date of infection.
Phillips said the undiagnosed population was a dynamic group. A quarter of gay men infected with HIV in 2008 will have been diagnosed by the end of this year, as will 30% of infected heterosexual men and 40% of women. While only one in six gay men who are diagnosed will do so because they have symptoms caused by immunodeficiency, over 60% of heterosexual men will, because they are more likely to be diagnosed late in the course of their HIV infection.
Professor Graham Hart, Director for HIV and Sexual Health Research at University College London, told the conference that there was some evidence from a previous Scottish study, also by Williamson and colleagues (2007) to back up Phillips’s model. In this survey, 46% of gay men in Glasgow and Edinburgh had never tested for HIV and of those who had, half had not done so for more than a year. When asked why they had never tested, or not tested recently, the two most common reasons (respondents could choose more than one reason) were “I have not been at risk” (45-46% of both categories) and “I know my status” (a third of those who had never tested, and 40% who had not tested for over a year).
Another common reason for not being tested was because men were in the “window period”, i.e. were turning up at a clinic within a month of a risk incident and thought that there was no point in being tested. In the post-seminar discussion, an audience member commented that patients were unaware that the fourth-generation HIV tests recommended by the new BHIVA/BASHH testing guidelines, which include an antigen test, had shrunk the window period to as little as two weeks, and might be using it as an excuse not to get tested. Suggestions were made that clinics should revise their protocols so that patients presenting soon after exposure should be offered opt-out testing regardless and recalled a month later for a second test.
The fact that in the Scottish survey 56% of men who were in fact HIV-positive despite having had a previous negative test described themselves as “HIV negative” – more than the proportion who said “I don’t know”, underlined the idea that a lot of gay men feel that a negative test implies negativity for a long time to come, regardless of further risks.
However a large minority of men also expressed significant fears around testing, with about one in six of respondents saying they were “too frightened” to go for a test (or another test) and about one in nine saying they “didn’t want to know”.
So how do we reduce the proportion of people who are undiagnosed and shorten the interval between infection and diagnosis? Undiagnosed HIV may be due to high incidence but it is also due to low rates of testing, with Hart pointing out that no more than two-thirds of UK gay men have had an HIV test compared with 92% in the USA and 96% in Australia.
The new BHIVA/BASHH guidelines recommend an expansion of testing into primary care, with tests for anyone who presents with a list of indicator conditions, and routine screening in A&E departments and at GP registration for patients in Primary Care Trusts (PCTs) where HIV prevalence in the general population is over 0.2% (currently 21 PCTs fit this criterion).
Hart presented new results from a pilot study called Rapid HIV Assessment in Primary Care (RHIVA) in which all new patients aged 18-55 registering at one London GP surgery between October 2007 and March 2008 were routinely offered an OraQuick saliva HIV test. Out of 85 eligible patients, less than half (38) tested. Of the 55% who did not, just over a third said they had had a previous test, but the other two-thirds said they were “not at risk”.
Most participants were positive about the programme. One Nigerian man commented that “If you go to a GUM clinic you will be stigmatised, but with a GP no-one will know you’ve had a test. I have never, ever been offered an HIV test before.” However a minority of patients were worried that “it’s so quick – it could be a real shock” or that the test was “intrusive”.
Professor George Kinghorn of the Royal Hallamshire Hospital in Sheffield also emphasised the section of the new BHIVA/BASHH testing guidelines that say “any competent healthcare professional” should be able to do an HIV test.
He was critical of medical personnel who still do not think of testing despite symptoms suggestive of AIDS. “These cases have been regarded as medically unfortunate,” he commented, “but in the future they may be regarded as medically negligent.”
While backing the guidelines’ recommendation for routine screening of patients in A&E and at GPs, he asked, “Why is it more acceptable to screen the asymptomatic than to do the obvious and test where symptoms indicate HIV?”
Dr Martin Fisher of Brighton and Sussex University Hospital highlighted a case he had presented the previous day, in which a 26-year-old Spanish man presented with meningitis and fever but was not asked about sexual risks or tested for HIV, and was subsequently found to be the source partner of a man who tested HIV-positive a year later. The man concerned had tried to sue the hospital for not testing him, Dr Fisher noted.
Main references
Phillips A Epidemiology of undiagnosed infection and its implications for onward transmission of HIV in the UK. Joint BHIVA/BASHH Ordinary General Meeting, BHIVA autumn conference, London, 2008.
Hart G Why do some patients refuse HIV testing? Joint BHIVA/BASHH Ordinary General Meeting, BHIVA autumn conference, London, 2008.
Kinghorn G How to reduce the prevalence of undiagnosed HIV infection outside the clinic. Joint BHIVA/BASHH Ordinary General Meeting, BHIVA autumn conference, London, 2008.
Other references
The UK Collaborative Group for HIV and STI Surveillance Testing Times. HIV and other Sexually Transmitted Infections in the United Kingdom: 2007. London: Health Protection Agency, Centre for Infections, 2007.
Williamson LM et al. Sexual risk behaviour and knowledge of HIV status among community samples of gay men in the UK. AIDS 22(9): 1063-1070, 2008.
Williamson LM et al. HIV prevalence and undiagnosed infection among a community sample of gay men in Scotland. JAIDS 45(2):224-30, 2007.