HIV in the UK 2017: a shift to eliminate AIDS

Simon Collins, HIV i-Base

Two important new reports from Public Health England (PHE) review the latest data on key aspects of HIV diagnoses, testing and care across the UK. [1, 2]

Both are essential reading, as are the surveillance tables for 2016 that are the basis for the reports and that are also published online. [3]

Similar EU surveillance data was also published by the European Centre for Disease Prevention and Control (ECDC), giving comparative demographic breakdowns by country. [4]


The 2017 UK HIV surveillance report is notable for reporting the most significant drop in HIV incidence for over a decade: fewer people were diagnosed HIV positive, even though more people were tested. This was not part of a steady decline but was driven by a rapid drop in new diagnoses in gay men.

In response, this year the main report has been restructured to emphasise the potential to end the HIV epidemic in the UK rather than just presenting a summary of the surveillance data.

This is organised into three main sections (plus appendices tables):

  1. HIV diagnosis and transmission, AIDS and HIV-related mortality
  2. Towards elimination of HIV transmission, AIDS and HIV-related mortality
  3. Living with diagnosed HIV

Throughout, there is an emphasis on the importance of new approaches that are likely to explain the drop in incidence.

  • More frequent HIV testing as part of routine sexual health.
  • Earlier linkage to care reduces the period when people are unaware of their HIV status,
  • Earlier use of HIV treatment (ART) for those who are diagnosed HIV positive.
  • Greater awareness and use of PrEP by HIV negative gay and bisexual men.

Summary of new HIV diagnoses

The big news is that new diagnoses last year fell by 16% – against a background of consistently high incidence from 2000 to 2015 – with approximately 6000 new diagnoses reported each year.

The HIV testing report (only for England) notes that more than one million tests were carried out in 2016 (approximately half in antenatal care), with 100,000 tests in gay and bisexual men.

However, although overall estimated numbers of people living with HIV (both in care and undiagnosed) steadily increased from 2000 to 2013, this total has changed little or fallen in the last few years (from 107,000 in 2013 to 102,000 in 2016). This is only partly accounted for by having fewer people who are undiagnosed. See Table 1.

In 2016, a total of 5164 people were diagnosed HIV positive, 931 fewer than 2015. The largest change was a 22% drop among gay and bisexual men, with 2810 new diagnoses in 2016 compared to 3570 in 2015. Reductions were also reported for heterosexually acquired HIV,

Although this decrease was driven by significant drops at five highest incidence clinics in London, similar percentage reductions reported in all regions of the UK, were likely driven by centres in other cities with high HIV expertise, following a similar shift to earlier testing and ART.

Also, although the report places a heavy emphasis on the reductions in gay men diagnosed in London – rightly because of numbers of diagnoses at very high incidence clinics – these declines occurred in every region in the UK. The impressive 29% reduction in London was relatively modest in percentage terms, compared to drops of 61% in Wales, 43% in Northern Ireland, 34% in the East Midlands and 33% in Scotland. (See Table 2).

The breakdowns by gender, age, route of transmission and geographical region show that diagnoses in heterosexual men and women also fell, partly related to different patterns of migrations. Approximately half of heterosexually acquired HIV occurred in the UK and half outside the UK.

Approximately 2% of HIV diagnoses were related to injecting drug use – roughly constant each year.

Of the 41 children diagnosed HIV positive, only five were born in the UK and only two were born in 2016.

Late diagnosis – still a considerable problem

Late diagnosis, defined as having a CD4 <350 cells/mm3, is still a significant concern, perhaps more so than is reported.

So although the report describes a 45% decrease in late diagnoses compared to 2007, which might sound impressive, this approach to comparing percentages doesn’t adjust for the higher numbers of people who were diagnosed in 2007.

Using the total diagnosed for each year as a denominator, the figures for late diagnosis have changed little and remain worryingly high.

In 2016, 42% (2159/5164) of people diagnosed HIV positive still had CD4 counts <350 – very little change from the 50% (3930/7777) in 2007. So rather than a 45% reduction, this data could therefore be presented as an 8% reduction over seven years – or worse still, given this is a key performance indicator, only reducing late diagnoses by just over 1% a year.

Here the demographic breakdowns are very useful to inform programmes that might have a great impact.

Younger people were less likely to be diagnosed late (31% of those aged 15-24, compared to 57% of those aged 50-64 and 63% aged >65 years old. Regional differences were also significantly different with London at the lowest rate (36%) with other regions varying between 42% to 47%.

While overall mortality for people diagnosed promptly has now normalised to the general population (1.22 vs 1.39 per 1000, respectively) rates are more than 20 times higher (26.1/1000) for those diagnosed late.

Overall, while the results are groundbreaking, they only take us back to 2010 levels, we still have a long way to go. (see Table 1).

Table 1: UK HIV surveillance data 2007-2016

Data year Total +ve* Undiag-nosed* In care* New dx New dx MSM Change in MSM dx.
2016 102,300 10,400 92,000 5164 2810 – 22 %
2015 101,200 13,500 89,200 6286 3570 + 6 %
2014 103,700 18,100 85,200 6200 3360 + 4 %
2013 107,800 26,100 81,500 5973 3230 + 1 %
2012   98,400 21,900 77,300 6204 3180 + 7 %
2011   96,000 22,600 73,600 6146 2010 + 3%
2010   91,500 22,200 69,900 6319 2860 + 3%
2009   86,500 22,200 64,900 6583 2850   0 %
2008   83,000 22,400 60,800 7156 2800   0 %
2007   77,400 21,700 56,000 7277 2850 + 5%

Key: dx – diagnosed; MSM – gay and bisexual men;

* rounded to nearest 100

Source: Public Health England, UK HIV surveillance data (2017)

Table 2: Regional declines in new diagnoses in gay and bisexual men

2015 2016 % change
London overall 1555 1096 – 29%
Five high incidence London clinics 1034 672 – 35%
Other London clinics 520 424 – 16%
East Midlands 110 73 – 34%
West Midlands 192 145 – 24%
South East 259 233 – 10%
South West 145 110 – 24%
Wales 85 33 – 61%
N. Ireland 61 35 – 43%
Scotland 131 90 – 33%

Source: Public Health England, UK HIV surveillance data (2017)

Living with HIV

The third section of the report highlights the generally excellent level of care available to HIV positive people who are diagnosed: approximately 96% are on ART and 97% of those on ART have undetectable viral load.

With median age of 46 the report highlights that more than one-third (38%) are now older than 50.

Demographic breakdown by ethnicity includes that 14% of gay and bisexual men were from black and minority ethnic (BAME) groups and among heterosexuals, 58% were black African and 25% were white.

Summary data are also included on TB coinfection (2015 data) and transmitted drug resistance (7.5% mainly to a single class, with rates stable since 2005),

Next year, the experiences of HIV positive people in these reports will also be included. In 2018, for the first time, the report will include results from a cross-sectional survey of approximately 4,400 demographically representative people receiving care.


The annual UK surveillance reports are amongst the most timely and comprehensive national datasets and they have tracked patterns in HIV incidence and mortality since the 1980s. The more recent testing reports is an essential new development.

While the reports are essential summaries, on a few generally small points the narrative doesn’t necessarily correlate with the data.

One is the way some percentages are used to compare results from different years (including the late diagnoses example above), rather than using the number of people diagnosed each year as the denominator.

Another is to report the drop in diagnoses in gay and bisexual men as having occurred “for the first time in over thirty years”. Annual diagnoses have sometimes gone down (though only modestly) and the extent of the drop this year is really significant.

The challenges for subsequent years are whether this improved signal can continue or whether the combination of frequent testing, early ART and PrEP simply reduced transmission to the people at highest risk – who were easiest to reach but being seen at large urban clinics who had responded quickly to incorporate new services.

An optimistic signal comes from the similar percentage reductions that were reported from all regions of the UK. A pessimistic response comes from the lack of central government investment in health care, especially sexual health, such that some clinics (anecdotally) are being told by health providers to cap the number of HIV tests, even if this limit is reached early in the year.

Another is that the long-awaited PrEP IMPACT study in England, might only have a limited effect on HIV incidence if it primarily enrolls those who were previously buying PrEP online.

Clearly HIV services, together with all other areas of health, need to resist policies that progressively destabilise and dismantle the NHS.


  1. Public Health England. Towards elimination of HIV transmission, AIDS and HIV-related deaths in the UK: 2017 report. (November 2017).
  1. Public Health England. HIV testing in England: 2017 report. (November 2017)
  1. Public Health England. HIV: annual data tables, 2017. (Excel file)
  2. European Centre for Disease Prevention and Control (ECDC). HIV/AIDS surveillance in Europe 2017 – 2016 data. (28 November 2017). (PDF)

Links to other websites are current at date of posting but not maintained.