Chemsex is top of the list of risk factors for HIV infection in gay and bisexual men in England


A study tracking new cases of HIV among gay and bisexual men in London and Brighton has identified the key risk factors predicting subsequent HIV infection in recent years. By far the most important was injecting drugs, largely in a chemsex context. Although only 3% of the men in the study reported injecting drugs, 16% of those who did subsequently tested HIV positive.

Men reporting the use of chemsex drugs, condomless sex with multiple partners or HIV–positive partners, group sex, and STIs were at increased risk of acquiring HIV. Men with less education were also significantly more likely to test positive. Nadia Hanum of University College London and colleagues have published the results from this prospective observational cohort study in PLoS Medicine.

The study was unusual in that participants completed detailed behavioural questionnaires in 2013-2016 and gave consent for the researchers to check Public Health England’s data up to 2019 to see if they were later diagnosed with HIV.



The use of recreational drugs such as mephedrone, GHB/GBL and crystal meth before or during sex.


Having sex without condoms, which used to be called ‘unprotected’ or ‘unsafe’ sex. However, it is now recognised that PrEP and U=U are effective HIV prevention tools, without condoms being required. Nonethless, PrEP and U=U do not protect against other STIs. 


Expresses the risk that, during one very short moment in time, a person will experience an event, given that they have not already done so.

hazard ratio

Comparing one group with another, expresses differences in the risk of something happening. A hazard ratio above 1 means the risk is higher in the group of interest; a hazard ratio below 1 means the risk is lower. Similar to ‘relative risk’.

post-exposure prophylaxis (PEP)

A month-long course of antiretroviral medicines taken after exposure or possible exposure to HIV, to reduce the risk of acquiring HIV.

This avoids the problem of participants tending to drop out of cohort studies as time goes on. While Public Health England’s records of new HIV diagnoses are anonymised, the researchers had permission to match study participants with new diagnosis data on the basis of a code based on the person’s surname, date of birth, ethnicity, country of birth, and so on.

The researchers invited gay and bisexual men having routine STI and HIV tests at London's 56 Dean Street and Mortimer Market Centre clinics, and Brighton's  Claude Nicol Clinic, to take part. The 1,162 participants were predominantly White (82%), university educated (74%) and employed (83%).

Over the period the study was being run, HIV infections among gay and bisexual men in England fell by two thirds. Reflecting this, there were only 33 HIV infections in the cohort – which is fewer infections than the researchers had anticipated when planning the study.

This otherwise welcome development reduced the study’s statistical power but the researchers were still able to identify risk factors for HIV infection, based on the men’s responses to the questionnaire when they joined the cohort.

Two of the most important were linked to drug use – 13 of the 321 men using mephedrone, GHB/GBL or crystal meth (drugs often used in chemsex) tested positive. This corresponds to a hazard ratio of 6.45, meaning that this factor increased the risk of later testing HIV positive six-fold. The lower than expected number of infections meant that the 95% confidence interval for this hazard ratio was wide – 1.84 to 22.64. This means that the 'true' level of increased risk, taking into account chance factors, most likely fell somewhere between these two figures.

Results were even more striking for injecting drugs – six of the 38 men who reported this tested positive (hazard ratio 27.96, 95% confidence interval 6.99 to 111.85). While the questionnaire did not ask which drugs were injected, all 38 injectors also reported using the chemsex drugs. The findings show that that nearly half of the 13 men who tested positive and reported using chemsex drugs also reported injecting.

Recent condomless anal sex was reported by over half the study participants, including 26 of those who subsequently tested positive (HR 3.75, 95% CI 1.31 to 10.74). There was a step-wise relationship between the number of sexual partners and the risk of acquiring HIV. For example, 5 of the 60 men who had between five and ten condomless partners in the past three months tested positive (HR 9.60, 95% CI 2.58 to 35.76).

Other sexual risk factors were:

  • Recent condomless sex with a partner known to be HIV positive – 14 of 147 men reporting this were diagnosed (HR 6.45, 95% CI 3.15 to 13.22).
  • Sometimes being top and sometimes bottom in anal sex - 21 of 362 men reporting this tested positive (HR 6.35, 95% CI 2.18 to 18.51).
  • Group sex – 64 of 500 men tested positive (HR 8.81, 95% CI 3.07 to 25.24).
  • Having sex for drugs or money – 4 of 55 men tested positive (HR 3.27, 95% CI 1.14 to 9.38).
  • Using post-exposure prophylaxis (PEP) – 11 of 240 men tested positive (HR 2.29, 95% CI 1.09 to 4.81).
  • Having a bacterial STI – 21 of the 440 men reporting this tested positive (3.95, 95% CI 1.81 to 8.63).

The only demographic factor that was significantly associated with an HIV diagnosis was less education.

HIV infection was not associated with age, housing status, financial status, relationship status, recent HIV testing, fisting, sex toys, PrEP, smoking, alcohol, depressive symptoms or anxiety symptoms.

In a context of increasing access to prompt HIV treatment and PrEP, the results highlight the contexts and behaviours which remain risky. The elevated risk associated with drug use could reflect the sexual behaviour which happens afterwards, but transmission through sharing needles is also a possibility.