Background Illicit drug use among men who have sex with men (MSM) has been associated with sexual risk and HIV. Less is documented about associations with other sexually transmissible infections (STIs). The aim of the present study was to determine whether the use of drugs commonly associated with chemsex is associated with increased risk of gonorrhoea among MSM. Methods: Using data from 16065 UK-based respondents to the European MSM Internet Survey (2010), we examined associations between a recent diagnosis of gonorrhoea and three chemsex drugs (crystal methamphetamine, γ-hydroxybutyric acid (GHB)/γ-butyrolactone (GBL) and mephedrone). Univariate logistic regression identified determinants of gonorrhoea diagnosis and multivariate logistic regression models calculated adjusted odds ratios (aORs) for independent associations between chemsex drugs and gonorrhoea. Results: MSM who reported using crystal methamphetamine and GHB/GBL in the previous year had 1.92- and 2.23-fold higher odds of gonorrhoea respectively over the same period (P=0.0001 and P<0.0001; n=15137) after adjusting for age, recruitment website, HIV status, residence and use of other chemsex drugs. MSM reporting the use of all three chemsex drugs had the highest increased odds (aOR 3.58; P<0.0001; n=15174). Mephedrone alone was not associated with gonorrhoea in multivariate models. Conclusions: Use of chemsex drugs is associated with a higher risk of gonorrhoea. The results of this study complement existing research about crystal methamphetamine and indicate a role for GHB/GBL in adverse sexual health outcomes. The use of mephedrone alongside other chemsex drugs may account for its lack of association with gonorrhoea in multivariate models. Future research should use encounter-level data, examine other STIs and attribute pathways through which chemsex leads to infection.

No abstract

IntroductionIn the last decade diagnoses of most STIs have risen among men who have sex with men (MSM). Although a significant proportion of this is likely due to increased STI screening, understanding the role of behavioural drivers remains critical. We measure the associations between stimulant use to enhance and prolong sexual experiences (chemsex) and bacterial STI diagnoses in UK MSM, individually considering HIV-diagnosed MSM, pre-exposure prophylaxis (PrEP) users and other MSM.MethodsWe used the UK 2017–2018 European MSM Internet Survey data (n=9375). We constructed causal inference models using multivariable logistic regression, calculating adjusted OR (aOR) and 95% CI of the associations between participation in recent (≤12 months) exclusively dyadic or multipartner chemsex versus no chemsex and recent self-reported diagnoses of syphilis, gonorrhoea and chlamydia.ResultsAmong MSM with an HIV diagnosis, 25% of users indicated recent multipartner chemsex, vs 28% of PrEP users and 5% of other MSM. Adjusting for age, ethnicity, UK birth, cis-trans status, sexual identity, education, settlement size and relationship status, participation in recent multipartner chemsex versus no chemsex was associated with greater odds of recent syphilis, gonorrhoea and chlamydia diagnosis. aORs for recent syphilis, gonorrhoea and chlamydia diagnoses were 2.6 (95% CI 1.7 to 4.1), 3.9 (95% CI 2.6 to 5.8) and 2.9 (95% CI 1.9 to 4.3), respectively, in HIV-diagnosed MSM; 1.9 (95% CI 1.1 to 3.3), 2.9 (95% CI 2.0 to 4.2) and 1.9 (95% CI 1.3 to 2.8), respectively, in PrEP users; and 4.0 (95% CI 2.3 to 6.9), 2.7 (95% CI 1.9 to 3.8) and 2.3 (95% CI 1.6 to 3.4), respectively, in other MSM. Conversely, exclusively dyadic chemsex had no significant associations with bacterial STI diagnoses among HIV-diagnosed MSM, only gonorrhoea (aOR 2.4, 95% CI 1.2 to 4.7) among PrEP users and syphilis (aOR 2.8, 95% CI 1.4 to 5.6) among other MSM.DiscussionMultipartner chemsex may drive the association between chemsex and bacterial STI diagnoses and thus should be the focus of future tailored chemsex interventions. Additionally, PrEP acceptability among MSM and particularly chemsex participants has generated an emergent group suitable for such interventions.

Contributors: MK carried out the analysis, and wrote the first draft. All other authors provided comments and edits. The original study design, data collection and data management were carried out by ARH, DR, CVP, GH, CHM and JS.