Understanding experiences of potential harm among MSM (cis and trans) using HIV self-testing in the SELPHI randomised controlled trial in England and Wales: a mixed-methods study.

BACKGROUND: The potential of HIV self-testing (HIVST) to cause harm is a concern hindering widespread implementation. The aim of this paper is to understand the relationship between HIVST and harm in SELPHI (An HIV Self-testing Public Health Intervention), the largest randomised trial of HIVST in a high-income country to date. METHODS: 10 111 cis and trans men who have sex with men (MSM) recruited online (geolocation social/sexual networking apps, social media), aged 16+, reporting previous anal intercourse and resident in England or Wales were first randomised 60/40 to baseline HIVST (baseline testing, BT) or not (no baseline testing, nBT) (randomisation A). BT participants reporting negative baseline test, sexual risk at 3 months and interest in further HIVST were randomised to three-monthly HIVST (repeat testing, RT) or not (no repeat testing, nRT) (randomisation B). All received an exit survey collecting data on harms (to relationships, well-being, false results or being pressured/persuaded to test). Nine participants reporting harm were interviewed in-depth about their experiences in an exploratory substudy; qualitative data were analysed narratively. RESULTS: Baseline: predominantly cis MSM, 90% white, 88% gay, 47% university educated and 7% current/former pre-exposure prophylaxis (PrEP) users. Final survey response rate was: nBT=26% (1056/4062), BT=45% (1674/3741), nRT=41% (471/1147), RT=50% (581/1161).Harms were rare and reported by 4% (n=138/3691) in exit surveys, with an additional two false positive results captured in other study surveys. 1% reported harm to relationships and to well-being in BT, nRT and RT combined. In all arms combined, being pressured or persuaded to test was reported by 1% (n=54/3678) and false positive results in 0.7% (n=34/4665).Qualitative analysis revealed harms arose from the kit itself (technological harms), the intervention (intervention harms) or from the social context of the participant (socially emergent harms). Intervention and socially emergent harms did not reduce HIVST acceptability, whereas technological harms did. DISCUSSION: HIVST harms were rare but strategies to link individuals experiencing harms with psychosocial support should be considered for HIVST scale-up. TRIAL REGISTRATION NUMBER: ISRCTN20312003.

Predictors of starting and stopping chemsex in men who have sex with men in England: findings from the AURAH2 prospective study.

BACKGROUND: Chemsex (the use of psychoactive drugs in sexual contexts) has been associated with HIV acquisition and other STIs, so there is benefit in identifying those most likely to start chemsex to offer risk reduction interventions such as pre-exposure prophylaxis (PrEP). To date, there have been no data from a longitudinal study analysing factors most associated with starting and stopping chemsex. METHODS: The prospective cohort study, Attitudes to and Understanding Risk of Acquisition of HIV over Time (AURAH2), collected 4 monthly and annual online questionnaire data from men who have sex with men (MSM) from 2015 to 2018. We investigate the association of sociodemographic factors, sexual behaviours and drug use with starting and stopping chemsex among 622 men who completed at least one follow-up questionnaire. Poisson models with generalised estimating equations were used to produce risk ratios (RRs) accounting for multiple starting or stopping episodes from the same individual. Multivariable analysis was adjusted for age group, ethnicity, sexual identity and university education. FINDINGS: In the multivariable analysis, the under 40 age group was significantly more likely to start chemsex by the next assessment (RR 1.79, 95% CI 1.12 to 2.86). Other factors which showed significant association with starting chemsex were unemployment (RR 2.10, 95% CI 1.02 to 4.35), smoking (RR 2.49, 95% CI 1.63 to 3.79), recent condomless sex (CLS), recent STI and postexposure prophylaxis (PEP) use in the past year (RR 2.10, 95% CI 1.33 to 3.30). Age over 40 (RR 0.71, 95% CI 0.51 to 0.99), CLS, and use of PEP (RR 0.64, 95% CI 0.47 to 0.86) and PrEP (RR 0.47, 95% CI 0.29 to 0.78) were associated with lower likelihood of stopping chemsex by the next assessment. INTERPRETATION: Knowledge of these results allows us to identify men most likely to start chemsex, thus providing an opportunity for sexual health services to intervene with a package of risk mitigation measures, especially PrEP use.

Causes of hospitalisation among a cohort of people with HIV from a London centre followed from 2011 to 2018.

BACKGROUND: We describe the spectrum of ICD-10 classified causes for hospitalisations occurring between 2011 and 2018 in a cohort of people living with HIV (PLHIV). METHODS: This sub-study includes 798 PLHIV participating in the Antiretroviral, Sexual Transmission Risk and Attitudes (ASTRA) questionnaire study who were recruited from a large London centre. A medical record review identified the occurrence and causes of hospitalisation from the date of questionnaire completion (February-December 2011) until 1 June 2018. Up to five causes were classified by an HIV clinician using the ICD-10 system. RESULTS: There were 274 hospitalisations in 153 people (rate = 5.8/100 person-years; 95% CI: 5.1, 6.5). Causes were wide-ranging; the most common were circulatory (16.8%), digestive (13.1%), respiratory (11.7%), infectious diseases (11.0%), injury/poisoning (10.6%), genitourinary diseases (9.9%) and neoplasms (9.1%). A tenth (27/274) of hospitalisations were related to at least one AIDS-defining illness. Median duration of hospitalisation was 5 days (IQR 2-9). At the time of hospitalisation, median CD4 count was high (510 cells/µl; IQR: 315-739), while median CD4 nadir was relatively low (113 cells/µl; IQR: 40-239). At admission, half of individuals (51%) had a previous AIDS-defining illness and 21% had viral load > 50 copies/ml. Individuals admitted for infectious diseases were particularly likely to have unfavourable HIV-related clinical characteristics (low CD4, viral non-suppression, not on antiretroviral therapy (ART), previous AIDS). CONCLUSIONS: In the modern combination antiretroviral therapy era, the spectrum of causes of hospitalisation in PLHIV in the UK is wide-ranging, highlighting the importance of holistic care for PLHIV, including prevention, early detection and treatment of comorbidities.

Prospective association of social circumstance, socioeconomic, lifestyle and mental health factors with subsequent hospitalisation over 6-7 year follow up in people living with HIV.

BACKGROUND: Predictors of hospitalisation in people with HIV (PLHIV) in the contemporary treatment era are not well understood. METHODS: This ASTRA sub-study used clinic data linkage and record review to determine occurrence of hospitalisations among 798 PLHIV from baseline questionnaire (February to December 2011) until 1 June 2018. Associations of baseline social circumstance, socioeconomic, lifestyle, mental health, demographic and clinical factors with repeated all-cause hospitalisation from longitudinal data were investigated using Prentice-Williams-Peterson models. Associations were also assessed in 461 individuals on antiretroviral therapy (ART) with viral load ≤50 copies/ml and CD4 count ≥500 cells/ µl. FINDINGS: Rate of hospitalisation was 5.8/100 person-years (95% CI: 5.1-6.5). Adjusted for age, demographic group and time with diagnosed HIV, the following social circumstance, socioeconomic, lifestyle and mental health factors predicted hospitalisation: no stable partner (adjusted hazard ratio (aHR)=1.59; 95% CI=1.16-2.20 vs living with partner); having children (aHR=1.50; 1.08-2.10); non-employment (aHR=1.56; 1.07-2.27 for unemployment; aHR=2.39; 1.70-3.37 for sick/disabled vs employed); rented housing (aHR=1.72; 1.26-2.37 vs homeowner); not enough money for basic needs (aHR=1.82; 1.19-2.78 vs enough); current smoking (aHR=1.39; 1.02-1.91 vs never); recent injection-drug use (aHR=2.11; 1.30-3.43); anxiety symptoms (aHRs=1.39; 1.01-1.91, 2.06; 1.43-2.95 for mild and moderate vs none/minimal); depressive symptoms (aHRs=1.67; 1.17-2.38, 1.91; 1.30-2.78 for moderate and severe vs none/minimal); treated/untreated depression (aHRs=1.65; 1.03-2.64 for treated depression only, 1.87; 1.39-2.52 for depressive symptoms only; 1.53; 1.05-2.24; for treated depression and depressive symptoms, versus neither). Associations were broadly similar in those with controlled HIV and high CD4. INTERPRETATION: Social circumstance, socioeconomic disadvantage, adverse lifestyle factors and poorer mental health are strong predictors of hospitalisation in PLHIV, highlighting the need for targeted interventions and care. FUNDING: British HIV Association (BHIVA) Research Award (2017); SMR funded by a PhD fellowship from the Royal Free Charity.

Recreational drug use and use of drugs associated with chemsex among HIV-negative and HIV-positive heterosexual men and women attending sexual health and HIV clinics in England.

BACKGROUND: There is little information on the prevalence of recreational drug use among UK heterosexual men and women, in particular on use of drugs associated with 'chemsex' within gay communities. The aim of this study was to examine among HIV-negative and HIV-positive heterosexual men and women in England: (i) the prevalence of recreational drug use (including use of drugs associated with chemsex), (ii) socio-economic/lifestyle correlates of drug use, and (iii) the association of drug use with sexual behavior measures and mental health symptoms. METHODS: Data are from the AURAH study of HIV-negative individuals attending sexual health clinics across England (2013-2014) and the ASTRA study of HIV-positive individuals attending HIV outpatient clinics in England (2011-2012). Prevalence of recreational drug use (past three months) and associations are presented separately among the four sample groups: HIV-negative (N = 470) and HIV-positive (N = 373) heterosexual men and HIV-negative (N = 676) and HIV-positive (N = 637) women. RESULTS: The age standardized prevalence of any drug use was 22.9%, 17.1%, 15.3%, and 7.1% in the four sample groups respectively. In all groups, cannabis was the drug most commonly used (range from 4.7% to 17.9%) followed by cocaine (1.6% to 8.5%). The prevalence of use of drugs associated with chemsex was very low among HIV-negative participants (1.0% heterosexual men, 0.2% women) and zero among HIV-positive men and women. In age-adjusted analysis, factors linked to drug use overall and/or to cannabis and cocaine use specifically in the four sample groups included Black/mixed Caribbean and white (vs. Black/mixed African) ethnicity, lower level of education , cigarette smoking, and higher risk alcohol consumption. Associations of recreational drug use with measures of condomless sex, depression, and anxiety were observed in the four groups, but were particularly strong/apparent among women. CONCLUSION: Providers need to be aware of cannabis and cocaine use and its potential link with sexual risk behavior and symptoms of depression and anxiety among heterosexual men and women attending sexual health and HIV clinics.

Exploring Mechanisms of Action: Using a Testing Typology to Understand Intervention Performance in an HIV Self-Testing RCT in England and Wales.

SELPHI involves two interventions: (A) It provides one HIV self-testing (HIVST) kit; (B) It offers 3-monthly repeat HIVST kits if participants report ongoing risk. A logic model underpinned by the Behaviour Change Wheel informed the design of the intervention. SELPHI recruited 10,135 cis-men and trans people in England and Wales, all reporting anal sex with a man. This paper explores how the interventions were experienced and the pathways to impact for different groups of trial participants. In-depth interviews with 37 cis-men who have sex with men (MSM) were used to inductively categorise participants based on sexual and HIV testing histories. Themes relating to intervention experiences and impacts were mapped onto SELPHI-hypothesised intermediate outcomes to consider intervention impacts. Three groups were identified: 'Inexperienced testers' engaged with SELPHI to overcome motivational and social and physical opportunity testing barriers. For 'pro self-testers', testing frequency was constrained by psychological and social barriers and lack of opportunity. 'Opportunistic adopters' engaged in HIVST for novelty and convenience. Perceived impacts for inexperienced testers were most closely aligned with the logic model, but for opportunistic adopters there was little evidence of impact. Distinctive groups were discernible with divergent intervention experiences. Using COM-B as a model for understanding behaviour change in relation to HIVST, our results indicate how HIVST interventions could be adapted to respond to different needs based on the target population's demographic and behavioural features.

Protocol, rationale and design of SELPHI: a randomised controlled trial assessing whether offering free HIV self-testing kits via the internet increases the rate of HIV diagnosis.

BACKGROUND: Among men who have sex with men (MSM) in the UK, an estimated 28% have never tested for HIV and only 27% of those at higher risk test at least every 6 months. HIV self-testing (HIVST), where the person takes their own blood/saliva sample and processes it themselves, offers the opportunity to remove many structural and social barriers to testing. Although several randomised controlled trials are assessing the impact of providing HIVST on rates of HIV testing, none are addressing whether this results in increased rates of HIV diagnoses that link to clinical care. Linking to care is the critical outcome because it is the only way to access antiretroviral treatment (ART). We describe here the design of a large, internet-based randomised controlled trial of HIVST, called SELPHI, which aims to inform this key question. METHODS/DESIGN: The SELPHI study, which is ongoing is promoted via social networking website and app advertising, and aims to enroll HIV negative men, trans men and trans women, aged over 16 years, who are living in England and Wales. Apart from the physical delivery of the test kits, all trial processes, including recruitment, take place online. In a two-stage randomisation, participants are first randomised (3:2) to receive a free baseline HIVST or no free baseline HIVST. At 3 months, participants allocated to receive a baseline HIVST (and meeting further eligibility criteria) are subsequently randomised (1:1) to receive the offer of regular (every 3 months) free HIVST, with testing reminders, versus no such offer. The primary outcome from both randomisations is a laboratory-confirmed HIV diagnosis, ascertained via linkage to a national HIV surveillance database. DISCUSSION: SELPHI will provide the first reliable evidence on whether offering free HIVST via the internet increases rates of confirmed HIV diagnoses and linkage to clinical care. The two randomisations reflect the dual objectives of detecting prevalent infections (possibly long-standing) and the more rapid diagnosis of incident HIV infections. It is anticipated that the results of SELPHI will inform future access to HIV self-testing provision in the UK. TRIAL REGISTRATION: DOI 10.1186/ISRCTN20312003 registered 24/10/2016.

Changes in recreational drug use, drug use associated with chemsex, and HIV-related behaviours, among HIV-negative men who have sex with men in London and Brighton, 2013-2016.

OBJECTIVES: The objective of this study was to compare the prevalence of polydrug use, use of drugs associated with chemsex, specific drug use, and HIV-related behaviours, between two time periods, using two groups of HIV-negative men who have sex with men (MSM) attending the same sexual health clinics in London and Brighton, in two consecutive periods of time from 2013 to 2016. METHODS: Data from MSM in the cross-sectional Attitudes to and Understanding Risk of Acquisition of HIV (AURAH) study (June 2013 to September 2014) were compared with baseline data from different MSM in the prospective cohort study Attitudes to and Understanding Risk of Acquisition of HIV over Time (AURAH2) (November 2014 to April 2016). Prevalence of polydrug use, drug use associated with chemsex and specific drug use, and 10 measures of HIV-related behaviours including condomless sex, post-exposure prophylaxis (PEP) use, pre-exposure prophylaxis (PrEP) use, and HIV testing, were compared. Prevalence ratios (PRs) for the association of the study (time period) with drug use and HIV-related behaviour measures were estimated using modified Poisson regression analysis, unadjusted and adjusted for sociodemographic factors. RESULTS: In total, 991 MSM were included from AURAH and 1031 MSM from AURAH2. After adjustment for sociodemographic factors, use of drugs associated with chemsex had increased (adjusted PR (aPR) 1.30, 95% CI 1.11 to 1.53) and there were prominent increases in specific drug use; in particular, mephedrone (aPR 1.32, 95% CI 1.10 to 1.57), γ-hydroxybutyric/γ-butryolactone (aPR 1.47, 95% CI 1.15 to 1.87) and methamphetamine (aPR 1.42, 95% CI 1.01 to 2.01). Use of ketamine had decreased (aPR 0.54, 95% CI 0.38 to 0.78). Certain measures of HIV-related behaviours had also increased, most notably PEP use (aPR 1.50, 95% CI 1.21 to 1.88) and number of self-reported bacterial STI diagnoses (aPR 1.24, 95% CI 1.08 to 1.43). CONCLUSIONS: There have been significant increases in drug use associated with chemsex and some measures of HIV-related behaviours among HIV-negative MSM in the last few years. Changing patterns of drug use and associated behaviours should be monitored to enable sexual health services to plan for the increasingly complex needs of some clients.

Multicenter European Prevalence Study of Neurocognitive Impairment and Associated Factors in HIV Positive Patients.

We conducted a cross-sectional study in 448 HIV positive patients attending five European outpatient clinics to determine prevalence of and factors associated with neurocognitive impairment (NCI) using computerized and pen-and-paper neuropsychological tests. NCI was defined as a normalized Z score ≤-1 in at least 2 out of 5 cognitive domains. Participants' mean age was 45.8 years; 84% male; 87% white; 56% university educated; median CD4 count 550 cells/mm3; 89% on antiretroviral therapy. 156 (35%) participants had NCI, among whom 26 (17%; 5.8% overall) reported a decline in activities of daily living. Prevalence of NCI was lower in those always able to afford basic needs (adjusted prevalence ratio [aPR] 0.71, 95% confidence interval [CI] 0.54-0.94) or with a university education (aPR 0.72, 95% CI 0.54-0.97) and higher in those with severe depressive symptoms (aPR 1.53, 95% CI 1.09-2.14) or a significant comorbid condition (aPR 1.40, 95% CI 1.03-1.90).

RESUMEN: Reportamos un estudio de tipo corte transversal que incluye 448 pacientes VIH seropositivos vistos en cinco clínicas especializadas en Europa con el objetivo de medir la prevalencia del trastorno neurocognitivo asociado al VIH (NCI por sus siglas en inglés) y los factores de riesgo asociados a éste. Se usaron pruebas neuropsicológicas computarizadas y en papel para determinar la presencia de NCI, definido como puntuación Z ≤ 1 en al menos 2 de los 5 dominios cognitivos evaluados. La media de edad de los pacientes fue 45,8 años, 84% eran hombres, 87% blancos y 56% tenían educación universitaria. La media de CD4 fue de 550 cel/mm3 y 89% de los pacientes recibían terapia antiretroviral. Un total de 156 (35%) participantes tenían NCI, de los cuales 26 (17%, 5,8% de la población de estudio) reportaron deterioro en actividades de la vida diaria. La prevalencia de NCI fue menor en participantes capaces de cubrir sus necesidades básicas (Razón de prevalencia ajustada [aPR] 0,71; Intervalo de confianza del 95% [95% CI] 0,54-0,94) o con educación universitaria (aPR 0,72; 95%CI 0,54-0,97) pero fue mayor en aquellos con síntomas de depresión severa (aPR 1,53; 95%CI 1,09-2,14) o alguna comorbilidad importante (aPR 1,40; 95%CI 1,03-1,90).

Minimal cognitive impairment in UK HIV-positive men who have sex with men: effect of case definitions and comparison with the general population and HIV-negative men.

BACKGROUND: To determine the prevalence of neurocognitive impairment (NCI) in UK HIV-positive and HIV-negative men who have sex with men (MSM). METHODS: HIV-positive and HIV-negative participants were recruited to a cross-sectional study from 2 London clinics and completed computer-assisted neuropsychological tests and questionnaires of depression, anxiety, and activities of daily living. Published definitions of HIV-associated neurocognitive disorders (HAND) and global deficit scores were used. Age- and education-adjusted neuropsychological test scores were directly compared with reference population data. RESULTS: A total of 248 HIV-positive and 45 HIV-negative MSM participated. In the HIV-positive group, median time since diagnosis was 9.4 years, median CD4 count was 550 cells per cubic millimeter, and 88% were on antiretroviral therapy. Prevalence of HAND was 21.0% in HIV-positive MSM (13.7% asymptomatic neurocognitive impairment, 6.5% mild neurocognitive disorder, and 0.8% HIV-associated dementia). Using a global deficit score threshold of 0.5, the prevalence of NCI was 31.5% (when averaged over 5 neuropsychological domains) and 40.3% (over 10 neuropsychological test scores). These results were not significantly different from the HIV-negative study sample. No consistent pattern of impairment was seen in HIV-positive patients relative to general male population data (n = 380). CONCLUSIONS: We found a prevalence of HAND and degree of impairment on neuropsychological testing of HIV-positive MSM that could represent a normal population distribution. These findings suggest that NCI may be overestimated in HIV-positive MSM, and that the attribution of NCI to HIV infection implied by the term HAND requires revision.

Health-related quality-of-life of people with HIV in the era of combination antiretroviral treatment: a cross-sectional comparison with the general population.

BACKGROUND: Combination antiretroviral therapy has substantially increased life-expectancy in people living with HIV, but the effects of chronic infection on health-related quality of life (HRQoL) are unclear. We aimed to compare HRQoL in people with HIV and the general population. METHODS: We merged two UK cross-sectional surveys: the ASTRA study, which recruited participants aged 18 years or older with HIV from eight outpatient clinics in the UK between Feb 1, 2011, and Dec 31, 2012; and the Health Survey for England (HSE) 2011, which measures health and health-related behaviours in individuals living in a random sample of private households in England. The ASTRA study has data for 3258 people (response rate 64%) and HSE for 8503 people aged 18 years or older (response rate 66%). HRQoL was assessed with the Euroqol 5D questionnaire 3 level (EQ-5D-3L) instrument that measures health on five domains, each with three levels. The responses are scored on a scale where a value of 1 represents perfect health and a value of 0 represents death, known as the utility score. We used multivariable models to compare utility scores between the HIV and general population samples with adjustment for several sociodemographic factors. FINDINGS: 3151 (97%) of 3258 of participants in ASTRA and 7424 (87%) of 8503 participants in HSE had complete EQ-5D-3L data. The EQ-5D-3L utility score was lower for people with HIV compared with that in the general population (marginal effect in utility score adjusted for age, and sex/sexuality -0·11; 95% CI -0·13 to -0·10; p < 0·0001). HRQoL was lower for people with HIV for all EQ-5D-3L domains, particularly for anxiety/depression. The difference in utility score was significant after adjustment for several additional sociodemographic variables (ethnic origin, education, having children, and smoking status) and was apparent across all CD4 cell count, antiretroviral therapy, and viral load strata, but was greatest for those people diagnosed with HIV in earlier calendar periods. Reduction in HRQoL with age was not greater in people with HIV than in the general population (pinteraction > 0·05). INTERPRETATION: People living with HIV have significantly lower HRQoL than do the general population, despite most HIV positive individuals in this study being virologically and immunologically stable. Although this difference could in part be due to factors other than HIV, this study provides additional evidence of the loss of health that can be avoided through prevention of further HIV infections. FUNDING: UK National Institute for Health Research.

Recreational drug use, polydrug use, and sexual behaviour in HIV-diagnosed men who have sex with men in the UK: results from the cross-sectional ASTRA study.

BACKGROUND: Recreational drug use in men who have sex with men (MSM) is of concern because it might be linked to the transmission of HIV and other sexually transmitted infections. Evidence about drug use in HIV-diagnosed MSM in the UK is limited by representativeness of the study populations. We describe patterns of drug use and associations with sexual behaviours in HIV-diagnosed MSM in the UK. METHODS: We used data from the cross-sectional ASTRA study, which recruited participants aged 18 years or older with HIV from eight HIV outpatient clinics in the UK between Feb 1, 2011, and Dec 31, 2012. We examined data for MSM, assessing the prevalence of recreational drug use and polydrug use in the previous 3 months and associations with sociodemographic and HIV-related factors. We examined the association of polydrug use with measures of condomless sex in the previous 3 months and with other sexual behaviours. FINDINGS: Our analysis included data for 2248 MSM: 2136 (95%) were gay, 1973 (89%) were white, 1904 (85%) were on antiretroviral treatment (ART), and 1682 (76%) had a viral load of 50 copies per mL or lower. 1138 (51%) used recreational drugs in the previous 3 months; 608 (27%) used nitrites, 477 (21%) used cannabis, 460 (21%) used erectile dysfunction drugs, 453 (20%) used cocaine, 280 (13%) used ketamine, 258 (12%) used 3,4-methylenedioxy-N-methylamphetamine (MDMA), 221 (10%) used gamma-hydroxybutyrate or gamma-butyrolactone, 175 (8%) used methamphetamine, and 162 (7%) used mephedrone. In the 1138 individuals who used drugs, 529 (47%) used three or more drugs and 241 (21%) used five or more. Prevalence of injection drug use was 3% (n = 68). Drug use was independently associated with younger age (p < 0·0001), not being religious (p = 0·001), having an HIV-positive stable partner (p = 0·0008), HIV-serostatus disclosure (p = 0·009), smoking (p < 0·0001), evidence of harmful alcohol drinking (p = 0·0001), and ART non-adherence (p < 0·0001). Increasing polydrug use was associated with increasing prevalence of condomless sex (prevalence range from no drug use to use of five or more drugs was 24% to 78%), condomless sex with HIV-seroconcordant partners (17% to 69%), condomless sex with HIV-serodiscordant partners (10% to 25%), and higher-HIV-risk condomless sex after taking viral load into account (4% to 16%; p ≤ 0·005 for all). Associations were similar after adjustment for sociodemographic and HIV-related factors. Methamphetamine was more strongly associated with higher-HIV-risk condomless sex than were other commonly used drugs. INTERPRETATION: Polydrug use is prevalent in HIV-diagnosed MSM and is strongly associated with condomless sex. Specialist support services for MSM with HIV who use recreational drugs might be beneficial in the reduction of harm and prevention of ongoing transmission of HIV and other sexually transmitted infections. FUNDING: National Institute for Health Research.