From chemsex to a broader drug-use for sex landscape: Comparing two cross-sectional surveys on men who have sex with men in Italy Authors: Maria Gabriella Donà, Mauro Zaccarelli, Laura Gianserra, Christof Stingone, Eugenia Giuliani, Francesca Di Tullio, Massimo Giuliani, Alessandra Latini Published: 2026 ABSTRACT BackgroundDrug-use in sexual settings among men who have sex with men (MSM) is increasing across Europe and is a key driver of sexually transmitted infections (STIs). Chemsex is evolving with increasing use of new psychoactive substances and diversification of context, shifting from private homes to diverse settings. Our aim was to assess changes in drug-use for sex among MSM who attended our STI/HIV reference center in two different periods over the last 8 years.MethodsWe compared two cross-sectional surveys conducted at the STI/HIV center of the San Gallicano Dermatological Institute (Rome, Italy) in 2018 (using paper-based, self-administered questionnaires) and 2024–2025 (using online self-administered questionnaires). The main outcome was the self-reported sexualized drug-use and chemsex (i.e., use of any of the three classic “chems”: crystal methamphetamine, mephedrone, and GHB/GBL) in the previous year. Proportions were compared using Fisher’s exact test or chi-square test, as appropriate.ResultsThe surveys included 354 MSM in 2018 and 331 MSM in 2024–2025. Overall, chemsex significantly increased from 5.1% in 2018 to 19.0% in 2024–2025 (p < 0.0001), both among MSM living with HIV (p < 0.0001) and MSM not living with HIV (p = 0.0005). A significant increase was observed in the use of crystal meth (2.3% vs 8.2%, p = 0.0005), mephedrone (4.0% vs 9.4%, p = 0.005), and GHB/GBL (3.4% vs 16.9%, p < 0.0001). Synthetic cathinones (from 0 to 2.7%, p = 0.001) and cannabinoids (from 0 to 0.9%, p = 0.112) emerged as new substances reported only in the most recent survey.ConclusionsIn our setting, drug-use for sex increased over time among MSM, with expansion of chemsex and diversification of substances. This behavioral change may increase STI/HIV transmission. Public health & STI/HIV services should update screening, counselling and harm reduction strategies to address these evolving patterns. Introduction Sexualized drug use (SDU) among men who have sex with men (MSM) is a growing challenge for STI/HIV services in Europe. At the same time, this behavior is not fully described by the classic and relatively narrow concept of chemsex alone. 1 Definitions still vary across studies, and the traditional drug triad (crystal methamphetamine, mephedrone, and GHB/GBL) is now often part of a wider set of drugs used in sexual contexts. This wider set can include ketamine, cocaine, poppers, cannabinoids, synthetic cathinones, and other new psychoactive substances. 2 This broader range of drugs matters because it can change prevalence estimates, risk interpretation, and how STI/HIV services should screen and counsel their patients. The clinical and behavioral picture of SDU and chemsex is also changing. With respect to early descriptions of prolonged sessions in private homes, more recent studies describe a wider range of situations, from well-organized group sessions to shorter and more fragmented encounters, also involving newly met partners, often contacted through dating apps.2,3 These encounters may happen in nightlife venues, saunas, clubs, private rooms, and other settings. Literature data, including Italian ones, also suggest that SDU differs by HIV status, psychosocial vulnerability, sexual networks, and local culture, supporting the idea that SDU/chemsex scenario is changing over time rather than staying as a fixed pattern.1,2,3,4,5 For instance, recreational use of Erectile Dysfunction Agents (EDA) among MSM is increasingly reported. They are Phosphodiesterase-5 inhibitors (PDE5Is), the four main PDE5Is being Avanafil (Stendra), Sildenafil (Viagra), Tadalafil (Cialis), and Vardenafil, which are used by MSM to prolong sexual activity and reduce erectile dysfunction linked to other drugs and alcohol. 6 In 2018, we investigated recreational drug use and STI diagnoses through a cross-sectional survey among MSM attending an STI/HIV reference center in Rome, Italy. We found that drug use for sex was common, while classic chemsex drugs were less frequent than cocaine, poppers, and other recreational substances. 7 In order to investigate whether SDU and chemsex have changed over the last years, we conducted a new cross-sectional survey among MSM attending the same center, and compared the findings with those from the former survey. Methods This cross-sectional study has been carried out at the STI/HIV center of the San Gallicano Dermatological Institute IRCCS (Rome, Italy). This center is one of the 12 sentinel sites in the Italian STI Surveillance System and is the largest center for STI diagnosis and care in central and southern Italy. It also offers HIV counselling and testing and antiretroviral treatment for people living with HIV (PLWH). Participants in the current survey were recruited among MSM consecutively attending the center for STI/HIV counselling/testing, follow-up visits, or clinical management between June 2024 and October 2025. Eligible participants were cisgender men aged ≥18 years who reported at least one lifetime sexual intercourse with another man. The anonymous online questionnaire was available on the STI/HIV center web portal https://pensarapido.ifo.it/. To describe changes over time in SDU/chemsex, we compared data from the 2018 survey with those from the current 2024–2025 survey. The first survey was conducted between March and September 2018 and used an anonymous self-administered paper questionnaire. Both surveys were designed to collect socio-demographic data, sexual behavior (condom use, group sex, use of dating apps), STI history, SDU and chemsex. A predefined list of substances (including variants of their names, e.g., M-Cat, meph, meow meow for mephedrone) from which respondents could select those they used was provided, in addition to an open ‘other’ field that allowed them to report any additional drugs not included in the list. The 2024–2025 was developed to collect additional data that have gained relevance in recent years, such as HIV Pre-Exposure Prophylaxis (PrEP), doxycycline use for pre- and post-exposure STI prophylaxis (doxy-PrEP/PEP), vaccination for mpox. Data on doxy-PrEP/PEP use have been previously published. 8 Participants in neither the first nor the most recent survey received any form of incentive for their participation. The main outcomes were self-reported SDU (any drug use for sex) and chemsex (at least one of the three drugs of the classic triad) in the previous year. We compared proportions using chi-square test and Fisher’s exact test as appropriate, and median values using the Mann-Whitney test. Statistical significance was set at p < 0.05. Analyses were performed using SPSS version 27.0 (IBM Corp., Armonk, NY, USA). Results Overall, 354 MSM were included in the 2018 survey and 331 in the 2024–2025 survey. The median age was 37 years (IQR 29–44) in the first survey and 41 years (IQR 32–50) in the second one (p < 0.001). The proportion of PLWH was higher in 2018 than in 2024–2025 (65.5% vs 32.9%, p < 0.0001). Educational level was comparable across surveys: 45.2% of respondents in 2018 had a high-school education or less, compared with 43.8% in the most recent survey (p = 0.72). Condomless sex increased significantly, being reported by 44.1% of the 2018 respondents versus 90.3% of the 2024–2025 respondents (p < 0.0001). Both the use of dating apps to seek for sexual partners (71.2% in 2018 vs 77.9% in 2024–2025, p = 0.048) and the practice of group sex (23.7% in 2018 vs 74.3% in 2024–2025, p < 0.0001) were more frequently reported in the most recent survey. Conversely, the proportion of participants who reported any STI diagnosis in the previous year was nearly twice as high in 2018 compared to 2024–2025 (36.4% vs 17.2%, p < 0.0001). PrEP use, not investigated in the 2018 survey, was reported by 41/222 MSM not LWH participating in the most recent survey (18.5%). Results of the two surveys in terms of SDU, chemsex and specific substances are shown in Table 1. Considering the overall population, SDU increased from 20.9% to 27.2%, but this difference did not reach statistical significance (p = 0.054). Among PLWH, changes were even larger. SDU increased from 18.5% to 29.4% (p = 0.024), whereas no significant difference was observed in MSM not LWH (25.4% in 2018 vs 26.1% in 2024-2025, p = 0.89). Use of chemsex-related drugs increased clearly. Crystal methamphetamine use rose from 2.3% to 8.2% (p = 0.0005), mephedrone from 4.0% to 9.4% (p = 0.005), and GHB/GBL from 3.4% to 16.9% (p < 0.0001), resulting the third most frequently used substance in the 2024–2025 survey. As a result, chemsex increased from 5.1% to 19.0% (p < 0.0001). The increase was significant both among MSM LWH (from 5.6% to 23.9%, p < 0.0001) and MSM not LWH (from 4.1% to 16.7%, p = 0.0005).Table 1.Comparison between the 2018 and 2024–2025 surveys in terms of sexualized drug use (SDU), chemsex, and use of specific substances among MSM attending an STI/HIV reference center in Rome, Italy.Variable2018 survey n/N (%)2024–2025 survey n/N (%)p valueSDU All MSM74/354 (20.9)90/331 (27.2)0.054 MSM living with HIV43/232 (18.5)32/109 (29.4) 0.024 MSM not living with HIV31/122 (25.4)58/222 (26.1)0.89Chemsex All MSM18/354 (5.1)63/331 (19.0) <0.0001 MSM living with HIV13/232 (5.6)26/109 (23.9) <0.0001 MSM not living with HIV5/122 (4.1)37/222 (16.7) 0.0005 Specific SDU substances Crystal methamphetamine8/354 (2.3)27/331 (8.2) 0.0005 Mephedrone14/354 (4.0)31/331 (9.4) 0.005 GHB/GBL12/354 (3.4)56/331 (16.9) <0.0001 Ketamine16/354 (4.5)20/331 (6.0)0.38 MDMA/Amphetamines18/354 (5.1)29/331 (8.8)0.056 PHP/Synthetic Cathinones0/354 (0.0)9/331 (2.7) 0.001 Cocaine/Crack47/354 (13.3)59/331 (17.8)0.104 Synthetic cannabinoides0/354 (0.0)3/331 (0.9)0.112Other substances Poppers46/354 (13.0)69/331 (20.8) 0.006 Cannabinoides84/354 (23.7)56/331 (16.9) 0.028 EDA35/354 (9.9)73/331 (22.0) <0.0001 SDU: sexualized drug use; MSM: men who have sex with men; LWH: living with HIV; GHB/GBL: gamma-hydroxybutyrate/gamma-butyrolactone; MDMA: methylenedioxymethamphetamine; PHP: pyrrolidinohexanophenone; EDA: erectile dysfunction agents. Significant differences are highlighted in bold (p<0.05). Table 1.Comparison between the 2018 and 2024–2025 surveys in terms of sexualized drug use (SDU), chemsex, and use of specific substances among MSM attending an STI/HIV reference center in Rome, Italy.Variable2018 survey n/N (%)2024–2025 survey n/N (%)p valueSDU All MSM74/354 (20.9)90/331 (27.2)0.054 MSM living with HIV43/232 (18.5)32/109 (29.4) 0.024 MSM not living with HIV31/122 (25.4)58/222 (26.1)0.89Chemsex All MSM18/354 (5.1)63/331 (19.0) <0.0001 MSM living with HIV13/232 (5.6)26/109 (23.9) <0.0001 MSM not living with HIV5/122 (4.1)37/222 (16.7) 0.0005 Specific SDU substances Crystal methamphetamine8/354 (2.3)27/331 (8.2) 0.0005 Mephedrone14/354 (4.0)31/331 (9.4) 0.005 GHB/GBL12/354 (3.4)56/331 (16.9) <0.0001 Ketamine16/354 (4.5)20/331 (6.0)0.38 MDMA/Amphetamines18/354 (5.1)29/331 (8.8)0.056 PHP/Synthetic Cathinones0/354 (0.0)9/331 (2.7) 0.001 Cocaine/Crack47/354 (13.3)59/331 (17.8)0.104 Synthetic cannabinoides0/354 (0.0)3/331 (0.9)0.112Other substances Poppers46/354 (13.0)69/331 (20.8) 0.006 Cannabinoides84/354 (23.7)56/331 (16.9) 0.028 EDA35/354 (9.9)73/331 (22.0) <0.0001 SDU: sexualized drug use; MSM: men who have sex with men; LWH: living with HIV; GHB/GBL: gamma-hydroxybutyrate/gamma-butyrolactone; MDMA: methylenedioxymethamphetamine; PHP: pyrrolidinohexanophenone; EDA: erectile dysfunction agents. Significant differences are highlighted in bold (p<0.05). Comparison between the 2018 and 2024–2025 surveys in terms of sexualized drug use (SDU), chemsex, and use of specific substances among MSM attending an STI/HIV reference center in Rome, Italy. SDU: sexualized drug use; MSM: men who have sex with men; LWH: living with HIV; GHB/GBL: gamma-hydroxybutyrate/gamma-butyrolactone; MDMA: methylenedioxymethamphetamine; PHP: pyrrolidinohexanophenone; EDA: erectile dysfunction agents. Significant differences are highlighted in bold (p<0.05). In the 2024–2025 survey, some substances were reported that were not reported by any respondents in the 2018 survey, including synthetic cathinones (0.0% vs 2.7%, p = 0.001) and synthetic cannabinoids (0.0% vs 0.9%, p = 0.112). Poppers increased from 13.0% to 20.8% (p = 0.006), while cannabinoids decreased from 23.7% to 16.9% (p = 0.028). Ketamine and MDMA/amphetamines also increased, but not significantly. Finally, slamsex (not investigated in 2018) was reported by 6/331 MSM (1.8%) in the most recent survey, but not within the previous year, solely as a lifetime experience. Discussion Our findings suggest that drug use for sex has increased over time among MSM in our clinical setting, particularly for substances more closely linked to chemsex, such as crystal methamphetamine, mephedrone, and GHB/GBL. This trend is consistent with recent literature showing that SDU and chemsex remain highly prevalent among MSM in Europe, with substantial variability across settings but an overall picture of persistent and, in some contexts, expanding use.1,9 A recent meta-analysis, which used the same chemsex definition of our investigation, estimated a pooled prevalence of 22%, with a wide range of prevalence (3%–79%) according to MSM HIV status, setting, and geographic location, among others. 9 Another meta-analysis, which included eight cross-sectional studies performed across 16 different European countries, reported a chemsex prevalence between 3% and 26%. 1 When comparing our two survey rounds, chemsex prevalence significantly increased among both MSM LWH and not LWH. In addition, MSM LWH showed a higher prevalence of chemsex than MSM not LWH, especially in the most recent survey, in line with literature data that report a consistently higher engagement in chemsex for PLWH. 10 Recent evidence also suggests that drug use in sexual contexts extends beyond a restricted group of “classic” chemsex users, supporting the need to interpret these behaviors within a broader and evolving framework. 11 In this perspective, our data suggest that the traditional concept of chemsex may now be too narrow to fully capture current patterns of SDU. Alongside the rise in the classic chemsex substances, we observed the emergence of newer compounds, such as synthetic cathinones and synthetic cannabinoids, as well as a broader increase in substances used to facilitate sex-related disinhibition and stimulation. Our findings reflect the rapidly evolving dynamics of the European clandestine drug market. Over recent years, a large number of Novel Psychoactive Substances (NPS) has entered the European market, with new compounds being detected each year and creating a significant public health challenge. 12 In 2025, 50 NPS were notified for the first time, in addition to around 400 already detected on the market in 2024. Both synthetic cannabinoids and synthetic cathinones belong to NPS group. According to the European Drug Report 2026, potent synthetic cannabinoids are used to adulterate cannabis products, and the health risks from consuming these novel compounds are poorly understood. 13 The same report affirms that synthetic cathinones have become established in parts of Europe as affordable alternatives for illicit stimulants. In fact, depending on the particular substance, the effects of synthetic cathinones are claimed to be similar to those of cocaine, amphetamine or MDMA (ectstasy). One of the most common unregulated cathinone derivatives is mephedrone, the use of which has significantly increased when comparing our two survey rounds. Prevalence of mephedrone use among MSM is quite heterogeneous in Western Europe, emerging as the most or least commonly used “chems” of the classic triad, depending on the setting (e.g., MSM recruited from sexual health clinics or not). 10 MDPV (3, 4-methylenedioxypyrovalerone) seems to be an even more potent molecule, since the available data suggest that it is used in much lower doses than mephedrone. Stimulant effects similar to those of MDPV have been reported for pyrrolidinohexanophenone (PHP), which appeared on the illicit market of NPS in 2014. A recent French survey on adults engaging in chemsex reported synthetic cathinones as the most frequently used substances during chemsex (almost 80% of the respondents), in line with a UK study (almost 65%) and a Belgian study (almost 70%).14,15,16 The use of synthetic cannabinoids in this latter study was much more limited (2%). 16 In addition to chemsex-associated substances, other substances emerged as frequently used. Among them, alkyl nitrates (poppers), which ranked at the third place in the 2018 survey, while being at first place in the most recent survey. Noteworthy, a significantly higher proportion of MSM reported EDA use in the most recent survey compared to the previous one. Because EDA use often accompanies behaviors like group sex, condomless sex, and multiple partners, it is viewed as a proxy for elevated sexual exposure and higher STI vulnerability.17,18 Chemsex seems to be increasingly embedded within a wider drug-sex landscape that includes both “core” chemsex and more occasional or situational substance use, linked to casual encounters, digital partner-seeking, and varied sexual settings. 19 This broader framework is relevant for several reasons. Different substances may still converge toward similar effects on sexual behavior, including increased arousal, prolonged intercourse, reduced inhibition, greater partner turnover, unprotected anal intercourse, sex with multiple partners, sex with serodiscordant (differing HIV status) partners and lower risk perception. Focusing only on a restrictive chemsex definition may underestimate sexualized drug use, which remains highly relevant for the strong association with risky sexual behaviors and increased vulnerability to HIV and other STIs. This needs to be taken into account when counselling, behavioral or pharmacological interventions are offered by STI/HIV services, although the effects of these interventions still need to be clearly established. 20 This study has several limitations. First, the two study populations are not fully comparable, particularly because the proportion of PLWH differs significantly between 2018 and 2024–2025, as well as the median age, and this may have contributed to some of the differences observed in SDU and chemsex. Second, the change from paper‐based questionnaires in the first survey to online questionnaires in the second one may have introduced selection and reporting biases, potentially affecting the respective findings and, thus, their comparison. Third, because we analyzed two independent cross‐sectional surveys, our findings may reflect differences between participants enrolled at the two time points rather than longitudinal changes in individual behavior. Fourth, data on SDU and chemsex were based on anonymous self‐report and are therefore subject to recall bias and possible under‐reporting due to stigma, which could lead to underestimates of prevalence. Fifth, because the surveys were fully anonymous, we were unable to verify whether the same individual(s) submitted more than one questionnaire. Nonetheless, this scenario appears improbable, as respondents are regular attendees of our center and typically show strong engagement with our research initiatives. Finally, newer or broader patterns of sexualized drug use may not have been fully captured, particularly given the continous emergence of additional NPS. In conclusion, our findings support a shift from a binary “chemsex versus no chemsex” model towards a more pragmatic continuum of sexualized drug use, which may better reflect current behaviors. STI/HIV services should broaden history-taking beyond classic chemsex drugs and explicitly explore sexualized use of stimulants, poppers, cocaine, ketamine, and emerging substances across different sexual contexts in order to improve prevention, screening, and harm reduction strategies for MSM, also given the increasing availability of new substances with largely unknown health harms.