Guerras et al. Harm Reduct J (2021) 18:86 https://doi.org/10.1186/s12954-021-00531-2 RESEARCH Open Access Sexualized drug injection among men who have sex with men in Madrid and Barcelona as the first episode of drug injecting Juan‑Miguel Guerras1,2, Patricia García de Olalla2,3, María José Belza2,4* , Luis de la Fuente1,2, David Palma2,3, Jorge del Romero5, Jorge‑Néstor García‑Pérez6 and Juan Hoyos7the Methysos Project Group Abstract Background: We estimate the prevalence of drug injection, the variables associated with having ever injected and the proportion of ever injectors whose first drug injection was for having sex; we describe the first drug injection episode, analyze the drugs most frequently injected and estimate the prevalence of risky injecting behaviors. Methods: The participants were 3387 MSM without a previous HIV diagnosis attending four HIV/STI diagnosis services in Madrid and Barcelona. Lifetime prevalence and prevalence ratios (PRs) by different factors were calculated using Poisson regression models with robust variance. We compared the characteristics of first drug injection episode, lifetime injection and risky injecting behaviors of those whose first injection was for sex (FIS) with those whose was not (non-FIS). Results: Lifetime prevalence of injection was 2.1% (CI 1.7–2.7). In the multivariate analysis, it was strongly associated with having been penetrated by more than five men in the last 12 months (aPR = 10.4; CI 2.5–43.4) and having met most of their partners at private parties (aPR = 7.5; CI 4.5–12.3), and less strongly with other factors. Of those who had ever injected drugs, 81.9% injected for sex the first time they injected drugs (FIS). At first injection, FIS participants had a mean age of 31 years, 62.7% used mephedrone and 32.2% methamphetamine on that occasion. Of this FIS group 39.0% had ever shared drugs or equipment and 82.6% had always shared for sex. Some 30.8% of non-FIS reported having also injected drugs for sex later on. Conclusions: Only two out of a hundred had ever injected, most to have sex and with frequent drug or injecting equipment sharing. Injecting for sex is the most common first episode of drug injection and is the most efficient risky behavior for the transmission of HIV, hepatitis B or C and other blood-borne infections. MSM participating in private parties should be considered a priority group for prevention policies. Keywords: Drug injection, MSM, Chemsex, Slamming Introduction epidemic (mostly heroin) [1]; immediately followed by During the last two decades of the 20st century, Spain an HIV epidemic [2, 3]. As a consequence, Spain was for was arguably the developed country most heavily affected years one of the countries with the highest rates of AIDS by two intertwined epidemics: first, a drug injection cases linked to intravenous drug use [4]. The incidences of these two epidemics peaked before most drug users *Correspondence: mbelza@isciii.es were even aware of the HIV-infection risk. Moreover, 4 Escuela Nacional de Sanidad, Instituto de Salud Carlos III, Monforte de also before users became generally aware, a transition Lemos 5, 28029 Madrid, Spain in heroin administration route had begun in Southwest Full list of author information is available at the end of the article © The Author(s) 2021. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://​creat​iveco​mmons.​org/​licen​ses/​by/4.​0/. The Creative Commons Public Domain Dedication waiver (http://​creat​iveco​ mmons.​org/​publi​cdoma​in/​zero/1.​0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data. Guerras et al. Harm Reduct J (2021) 18:86 Page 2 of 11 Spain which gradually spread to the northeast; injection it is one of the strongest correlates for overdose [21, 22] was replaced by “chasing the dragon” (inhaling the vapors and the development of dependence [23]. which result when the drug is heated) as both the initial This original is focused on MSM without a previous and usual administration route [5]. Madrid and Barce- positive HIV diagnosis, while it is still possible to pre- lona experienced very different patterns in the develop- vent this infection; HCV, whose prevalence is extremely ment of injection practices. Madrid began the transition low in HIV negative MSM [24] may also be prevented. It from injection to chasing the dragon before Barcelona has been carried out in Madrid and Barcelona: the two and the latter maintained a higher level of administration most populated cities in Spain which also have high prev- by injection [6]. HIV-prevention programs in the sec- alences of MSM, because they attract MSM both from ond part of the eighties and especially the development other parts of Spain and abroad. of harm reduction policies during the nineties [3] were We estimate the prevalence of drug injection in differ- decisive for a radical decrease in the prevalence and espe- ent periods and the variables associated with having ever cially the incidence of injection. These policies also led injected; we also estimate the proportion of ever injectors to a reduction in new HIV diagnoses linked to this risky whose first drug injection in their lives was for having behavior: only 2.6% in 2019 [7]. sex; we describe this first injection episode, analyze the In the first two decades of the current century, there drugs that have ever and most frequently been injected, has not been any consistent evidence of an increase in and estimate the prevalence of risky injecting behaviors. drug-injection incidence in Spain. However, in the last decade several studies—either conducted only in Spain, Methods or as part of European studies—have shown that Span- Design, recruitment, data collection instruments ish men who have sex with other men (MSM) are adopt- and variables ing new patterns of sexualized drug-use characterized by This survey was the first design included in the Methysos the use of specific substances (mainly methamphetamine, Project, which is devoted to analyzing the prevalence and mephedrone, GHB/GBL or ketamine) generally labeled characteristics of drug use (including sexualized drug “chemsex drugs” (ChSD) [8–11]. Although the charac- use) in MSM in Spain. A cross-sectional survey was car- teristics of the participants in the studies are very hetero- ried out in four facilities: the two most important sexu- geneous, practically all of those that addressed the topic ally transmitted infection clinics in Spain—Sandoval, in showed some evidence of associations between chemsex Madrid and Drassanes, in Barcelona—and two commu- behavior and higher prevalence of HIV, HCV and other nity programs for rapid HIV-testing: Pink Peace Program, sexually transmitted infection (STI) [8, 11–14]. Admin- in Madrid, and Agencia de Salut Pública, in Barcelona. istration routes for these substances are generally snort- The STI clinics basically provide on-demand services and ing or ingesting, but they are also sometimes injected, a perform classical testing for all STIs, whereas the com- practice known as “slamming.” This administration route munity programs also carry out various different kinds of is motivated by similar reasons to chemsex, but also for active recruitment, including via ads and profiles on dat- some specific reasons [14]. Available epidemiological ing apps for MSM, and only offer rapid testing for HIV, analysis of slamming is limited, with many studies being syphilis and, sometimes, for HCV. qualitative or having a fairly limited sample size [14–18]. The recruitment began in May 2018 and finished in Moreover, nearly all the research overall is focused on December 2020. The study was restricted to MSM with- the association with sexual risk practices, and, as far as out a previous HIV diagnosis, because they make up we know, no analysis of initiation circumstances has yet the vast majority of attendees and the priority group for been undertaken. Thus, the proportion of MSM who prevention of both HIV and HCV. Therefore, only those have ever injected drugs who report that their first epi- MSM without a previous HIV test in their life or those sode of drug injecting was for having sex (slamming) whose last HIV test had been negative accessing these remains unknown. Likewise, the percentage who report facilities were offered to participate. If they accepted, having injected for the first time for purposes other they answered a self-administered online questionnaire than having sex is also unknown. Nor has it been stud- without personal identifiers on a tablet. The question- ied whether and to what extent the process of preparing naire had different sections: sociodemographics, history and injecting the substances implies sharing of needles, of HIV testing, sexual risk behaviors and both recrea- syringes or other devices, although it has been suggested tional and sexualized drug use. It included a short set that at least certain subpopulations often share syringes of questions related with drug injection for recreational [19]. It is well known that this risky behavior is much use or in a sexual context: “Have you ever injected any more efficient than sexual practices at transmitting HIV, drug?”, characteristics of first injection, some character- HBV and HCV [20], and, for pharmacokinetic reasons, istics of having ever injected, and of “sharing” material to Guerras et al. Harm Reduct J (2021) 18:86 Page 3 of 11 prepare drugs or sharing injection equipment. The term were under 40 years of age, 37.9% were born abroad, “sharing” includes several behaviors, because in the ques- 75.6% lived in the municipalities of Madrid or Barcelona, tionnaire it was defined as “having injected with a syringe 59.8% had university-level studies, 59.8% had a comfort- previously used by another person; or, having taken the able economic situation and 40.4% had lived alone during dissolved drug from a syringe previously used by another the last 12 months. person; or, from the recipient in which another person With respect to sexual and risk behavior: 62.6% had has previously inserted their syringe.” The study was only ever had sex with men, 17.6% had had their first sex- approved by the Research Ethics Committee of the Insti- ual relationship with a man before 16 years of age, 61.5% tuto de Salud Carlos III (CEI PI 44_2018_subproyecto1- lived their sexual life with men openly, 64.2% had met v2 and CEI PI 44_2018_subproyecto2). most of their partners through websites or dating apps; 5.3% had never been penetrated and 25.5% had been pen- Statistical analysis etrated by more than 50 men in their lives; 21.5% had Nearly all of the variables were collected in a more dis- been paid for sex; 17.2% had paid for sex; and 6.2% had aggregated form than presented here, since some of the injected steroids. Concerning HIV and STI testing: 49.1% original categories have been grouped together based had been tested for HIV in the last 6 months and only on their frequencies and the rationale for the analy- 6.4% had never been tested before; 2.5% were diagnosed sis. See Additional file 1 for more information on how with an HIV infection in that consultation, and 71.5% had original variables and categories from the questionnaire been diagnosed with an STI at some point. were managed to obtain the final variables and catego- Lifetime injection prevalence for any drug in the global ries used in the analysis. Most of the final variables were sample was 2.1% (CI 1.7–2.7), which was higher, but not based only on one original variable. Characteristics (n significantly, in Madrid (2.4%; CI 1.9–3.1) than Barcelona and percentages) of the participants were described: (1.6; CI 0.9–2.6) and in the community programs (2.5%, sociodemographics, sexual risk behaviors, history of CI 1.9–3.4) than in the STI centers (1.7%; CI 1.2–2.5). In HIV testing and STI diagnosis. Then, we calculated the the crude analysis (Table 1), the prevalence of lifetime prevalence of drug injection for different periods, and the injection was significantly associated (when the 95% CI of prevalence of having ever injected stratified by different the cPR did not include the null value, 1) with being born variables. Comparisons of independent variables were in Latin America (cPR = 1.7; CI 1.0–2.7), being unem- assessed using Pearson’s χ2 and Fisher’s exact tests. For ployed (cPR = 2.7; CI 1.4–5.2), ever having had sex with the analysis of correlates of ever having injected, Pois- women (cPR = 2.1; CI 1.3–3.3), having had their first sex- son regression models with robust variance were used ual intercourse with a man before age 20 (cPR = 1.8; CI [25, 26]. Both crude and adjusted Prevalence Ratios (cPRs 1.0–3.1), living their sexual life openly (cPR = 1.9; CI 1.1– and aPRs) and 95% Confidence Intervals (95%CI) were 3.2), having found most of their sexual partners in private calculated. Variables with a significance level of < 0.25 parties (cPR = 10.8; CI 6.5–18.1), having been penetrated in the bivariate analysis were introduced into the multi- by more than fifty men in their lives, (cPR = 8.2; CI 1.1– variate model—after collapsing the number of categories 59-9), having been penetrated by more than five men for many of these variables, due to the limited number of in the last 12 months (cPR = 14.5; CI 3.5–59.5), having participants who had ever injected. We used the Akaike been paid for sex in the last 12 months (cPR = 3.5; CI Information Criteria to perform model comparisons in 2.1–5.9), ever having injected steroids (cPR = 3.7; CI 2.0 order to select the final model. Finally, we stratified the 6.5), and having been diagnosed with an STI (cPR = 3.2; participants into two groups: those whose first injection CI 1.5–6.6). was for sex (FIS) and those whose was not (non-FIS), and In the multivariate analysis (Table 2) higher prevalence we made a comparison of the different characteristics of injection remained independently and strongly associ- of the first injection episode, drugs injected for all pur- ated with two variables: Having been penetrated by more poses and specifically for sexual purposes, frequency of than five men in the last 12 months (aPR = 10.4; CI 2.5– injection and “sharing” of drugs or injection equipment 43.4), and having found most of their sexual partners in between the two groups. private parties (aPR = 7.5; CI 4.5–12.3). It also remained associated with another four variables, but with much Results lower aPRs, around two: ever having injected steroids, As shown in Table 1, the study included 3387 partici- ever having been diagnosed with an STI, ever having pants: 2356 in Madrid (1224 in the STI center and 1132 been paid for sex, and having also had sex with women. in the community center) and 1031 in Barcelona (418 in When we focused on the first injection episode of the STI center and 613 in the community center). Con- those who had ever injected, we observed that 81.9% cerning sociodemographics: 74.0% of the participants injected for sex at first injection (FIS) versus 18.1% who Guerras et al. Harm Reduct J (2021) 18:86 Page 4 of 11 Table 1 Sample characteristics and bivariate analysis of factors associated with having ever injected drugs among MSM* in Madrid and Barcelona Sample characteristics (N = 3387) Prevalence of having ever Crude prevalence (95% CI**) injected drugs (2.1%) ratio N % % Recruitment City of testing Barcelona 1031 30.4 1.6 1.0 Madrid 2356 69.6 2.4 1.5 0.9–2.7 Kind of testing program STI diagnostic center 1642 48.5 1.7 1.0 Comunity program 1745 51.5 2.5 1.5 0.9–2.4 Sociodemographics Age (years) ≤ 24 559 16.5 2.0 1.3 0.6–3.0 25–39 1949 57.5 2.5 1.6 0.9–3.1 ≥ 40 879 26.0 1.5 1.0 Country of birth Spain 2105 62.1 1.8 1.0 Latin-America 927 27.4 3.0 1.7 1.0–2.7 Others 355 10.5 1.7 0.9 0.4–2.1 Years residing in Spain (for those born outside of Spain) ≤1 332 29.2 2.4 2–4 281 24.7 2.9 ≥5 526 46.2 2.9 Size of city of residence (last 12 months) ≤ 100.000 389 11.5 1.5 1.0 100.000–1 million 434 12.9 1.8 1.2 0.4–3.4 > 1 million 2547 75.6 2.3 1.5 0.6–3.4 Level ofeducation Up to upper secondary 229 6.8 3.1 1.8 0.8–4.0 Post secondary 1129 33.5 2.7 1.5 0.9–2.5 University 2016 59.8 1.7 1.0 Employment status (last 12 months)*** Employed 1737 74.1 2.3 1.0 Unemployed 183 7.8 6.0 2.7 1.4–5.2 Others 424 18.1 1.4 0.6 0.3–1.5 Economic situation (last 12 months) Comfortable/it is OK 2022 59.8 2.2 1.0 Tight 1070 31.7 2.1 0.9 0.6–1.6 Dificult/very difficult 287 8.5 2.1 1.0 0.4–2.3 Cohabitation (last 12 months)*** Alone 947 40.4 2.9 1.0 With some people 1400 59.6 2.1 0.7 0.4–1.2 Sexual and risk behavior Gender of sex partners (ever) Only men 2119 62.6 1.5 1.0 Men and women 1268 37.4 3.2 2.1 1.3–3.3 Age at first sexual intercourse with another men (years) < 16 596 17.6 3.5 7.3 1.7–31.0 16–20 1826 54.0 2.2 4.5 1.1–18.7 21–24 549 16.2 1.6 3.4 0.7–15.6 Guerras et al. Harm Reduct J (2021) 18:86 Page 5 of 11 Table 1 (continued) Sample characteristics (N = 3387) Prevalence of having ever Crude prevalence (95% CI**) injected drugs (2.1%) ratio N % % ≥ 25 412 12.2 0.5 1.0 Lives sex-life with men… Not Openly 1297 38.5 1.4 1.0 Openly 2075 61.5 2.6 1.9 1.1–3.2 Place where the largest number of partners were met Dicos/clubs/bars 509 15.6 1.4 Saunas 320 9.8 1.9 Apps/websites 2093 64.2 1.8 Cruising places 103 3.2 1.9 Private parties 112 3.4 17.9 Others/no search 122 3.7 0.0 Place where the largest number of partners were met Others 3147 96.6 1.7 1.0 Private parties 112 3.4 17.9 10.8 6.5–18.1 Number of men penetrated by (ever) None 178 5.3 0.6 1.0 1–49 2345 69.2 1.3 2.4 0.3–17.2 ≥ 50 864 25.5 4.6 8.2 1.1–59.9 Number of men penetrated by (last 12 months) None 691 20.4 0.3 1.0 ≤5 1569 46.4 1.5 5.1 1.2–21.5 >5 1123 33.2 4.2 14.5 3.5–59.5 Been paid for sex Never 2660 78.6 1.5 1.0 > 12 months ago 321 9.5 2.8 1.8 0.9–3.7 Last 12 months 405 12.0 5.4 3.5 2.1–5.9 Paid for sex Never 2805 82.9 2.0 1.0 > 12 months ago 290 8.6 2.1 1.0 0.4–2.4 Last 12 months 290 8.6 3.4 1.7 0.9–3.4 Ever injected steroids No 3177 93.8 1.8 1.0 Yes 210 6.2 6.7 3.7 2.0–6.5 History of HIV and other STI testing Time since last HIV test Never tested before 218 6.4 0.5 1.0 < 6 months 1661 49.1 2.2 4.9 0.7–35.4 > 6 months 1502 44.4 2.3 4.9 0.7–36.0 HIV new diagnosis in this consultation No 3253 97.5 2.0 1.0 Yes 82 2.5 3.7 1.9 0.6–5.9 STI diagnosis (ever) No 955 28.5 0.8 1.0 Yes 2399 71.5 2.7 3.2 1.5–6.6 *MSM: men who have sex with men **95% CI: 95% confidence interval ***These question were not included in Barcelona Guerras et al. Harm Reduct J (2021) 18:86 Page 6 of 11 Table 2 Multivariate regression analysis of factors associated had moderate levels of injection for both groups of drugs with having ever injected drugs among MSM* in Madrid and (ChSD and non ChSD) except opioids, MDMA and Barcelona (N = 3387) mephedrone. aPR** (95% CI***) 39.0% of FIS had shared drugs or injecting equipment during their lifetimes, 78% in the last 6 months; while Number of men who had penetrated him (last 12 months) only 15% of the non-FIS had done so and none of them None 1.0 in the last 6 months. The 47.8% non-FIS who had shared ≤5 5.1 1.3–21.7 had done so with more than five people, and 82.6% had >5 10.4 2.5–43.4 always shared for sex. Place where the largest number of partners were met Others 1.0 Private parties 7.5 4.5–12.3 Discussion Ever injected steroids Main results and comparisons with other studies No 1.0 As far as we know, this is the first study to provide clear Yes 2.3 1.3–4.2 empirical evidence that injecting drugs to have sex is STI diagnosis (ever) being the most common first episode of injection for No 1.0 MSM. Previous studies [17] have documented the use of Yes 2.2 1.1–4.6 injection as an administration route for various different drugs consumed for having sex. However, none of these Been paid for sex studies asked whether the first injection was for sex or Never 1.0 for any other purpose. This study shows that more than > 12 months ago 1.3 0.7–2.7 four out of five participants, who had injected drugs dur- Last 12 months 2.1 1.3–3.5 ing their lifetimes, had done so for sex when injecting Gender of sex partners (ever) for the first time. In addition, of those whose first time Only men 1.0 injecting was not for sex, one third had injected for sex Men and women 2.0 1.3–3.1 later. It has been well known for several years that self- *MSM: men who have sex with men intravenous injection is a difficult to adopt behavior, as **aPR: adjusted Prevalence Ratio injection generates fear and instinctive rejection, and also ***95% CI: 95% confidence interval requires training [27, 28]. For both reasons, the first few times novices are injected by or at least under the super- vision of a more experienced injector. We could say that injected with other purposes. Some 30.8% of non-FIS a prerequisite for the spread of injecting is that there are also injected drugs for sex later on. FIS participants skilled injectors. became injection initiates when they were more than The study shows that the proportion of MSM who 7 years older than non-FIS (31.1 vs 23.9); 83.1% of FIS had ever injected is small: Only two per cent. However, had injected for the first time in the last 3 years versus when assessing its significance, it is necessary to bear in only 46.2% in non-FIS, around 70% in both groups did mind that the spread of injection among MSM is rela- not self-inject but were injected by someone else; meth- tively recent, and therefore its growth potential may be amphetamine was the first drug injected by a third of the substantial. In fact, more than four out of five FIS in the participants from both groups, mephedrone was the ini- study began injecting in the last 3 years. The prevalence tiate drug for 62.7% of FIS, while non-FIS used other dif- found in our study is higher than the 1.5% found in Spain ferent drugs, like cocaine, heroin and amphetamine. The as a whole by the EMIS study [29], despite the fact that drug used for the first injection was significantly different the latter included more than 10% of HIV-positives in its by city: in Madrid mephedrone was the drug of choice sample, who can be expected to have a higher injection for 77.8%, while in Barcelona it was methamphetamine prevalence. This prevalence found by EMIS (an Internet- (78.6%) (Table 3). based survey) in Spain was lower than in some countries, When lifetime injection was analyzed, it appears that such as the UK, the Netherlands, Belgium and France. most of the participants in both groups were sporadic Likewise, the prevalence found in the present study is injectors since 32.2% of FIS and 53.8% of non-FIS had lower than that found in a different Internet-based sur- only injected once and 28.8% FIS and 15.4% non-FIS had vey in the UK (2.9%) [15] or in a study in five large cit- done it twenty or more times. Both groups had similar ies in France using time-location (3.1%) [14], but higher patterns of drug use for the first injection: FIS had used than another Internet-based survey in Ireland (1.6%) the ChSD, especially mephedrone (67.8%) and practically [30]. The comparison with the other studies included in a never had injected any of the other drugs while non-FIS recent review [18] makes little sense as those studies are Guerras et al. Harm Reduct J (2021) 18:86 Page 7 of 11 Table 3 Characteristics of drug injection by purpose of first injection among MSM* in Madrid and Barcelona First injection for sex First injection NON for sex Total p value N = 59 (81.9%) N = 13 (18.1%) N = 72 (2.1%) N % N % N % First injection Age: Median (1–3 IQR) 31.1 26.0–35.0 23.9 20.0–25.0 29.8 24.0–34.0 0.002** Years since first injection 0.005 ≤3 49 83.1 6 46.2 55 76.4 >3 10 17.0 7 53.9 17 23.6 Who performed theinjection 0.888 Self 17 28.8 4 30.8 21 29.2 Other person 42 71.2 9 69.2 51 70.8 Who was the injector 0.003 Stable partner 0 0.0 1 11.1 1 2.0 Casual partner 27 64.3 1 11.1 28 54.9 Friend/acquaintance 15 35.7 7 77.8 22 43.1 Drug injected < 0.001 Cocaine 0 0.0 2 15.4 2 2.8 Heroine or other opiods 0 0.0 1 7.7 1 1.4 Anphetamine (speed) 0 0.0 1 7.7 1 1.4 MDMA 1 1.7 0 0.0 1 1.4 Metanphetamine 19 32.2 4 30.8 23 31.9 Mephedrone 37 62.7 0 0.0 37 51.4 Ketamine (K, keta, kei) 1 1.7 1 7.7 2 2.8 Others 1 1.7 4 30.8 5 6.9 Lifetime injection Number of days with an injection 0.405 1 19 32.2 7 53.8 26 36.1 2–4 11 18.6 1 7.7 12 16.7 5–19 12 20.3 3 23.1 15 20.8 ≥ 20 17 28.8 2 15.4 19 26.4 Last injection 0.188 Last month 28 48.3 3 23.1 31 43.7 Last 6 months 17 29.3 4 30.8 21 29.6 Last 12 months 6 10.3 4 30.8 10 14.1 > 12 months 7 12.1 2 15.4 9 12.7 Drugs ever injected Cocaine 2 3.4 4 30.8 6 8.3 < 0.001 Heroine or other opiods 1 1.7 1 7.7 2 2.8 0.234 Anphetamine 0 0.0 2 15.4 2 2.8 0.002 MDMA 2 3.4 1 7.7 3 4.2 0.482 Metanphetamine 27 45.8 4 30.8 31 43.1 0.323 Mephedrone 40 67.8 1 7.7 41 56.9 < 0.001 Ketamine 12 20.3 2 15.4 14 19.4 0.683 Others 3 5.1 4 30.8 7 9.7 0.005 Sharing drug or injection equipment Ever shared 0.106 No 36 61.0 11 84.6 47 65.3 Yes 23 39.0 2 15.4 25 34.7 Last time shared 0.061 Last month 13 56.5 0 0.0 13 52.0 Guerras et al. Harm Reduct J (2021) 18:86 Page 8 of 11 Table 3 (continued) First injection for sex First injection NON for sex Total p value N = 59 (81.9%) N = 13 (18.1%) N = 72 (2.1%) N % N % N % Last 6 months 5 21.7 0 0.0 5 20.0 More than 6 months 5 21.7 2 100.0 7 28.0 With how many people 0.908 1 2 8.7 0 0.0 2 8.0 2–4 10 43.5 1 50.0 11 44.0 ≥5 11 47.8 1 50.0 12 48.0 Proportion with they shared for sex 0.009 All 19 82.6 0 0.0 19 76.0 Not all 4 17.4 2 100.0 6 24.0 *MSM: men who have sex with men **Student’s T Test qualitative, either with a fairly limited sample size or with had started at least 35 years old, while that of non-FIS very different inclusion criteria. Considering that most was roughly 24, more similar to trends in the general injectors in Spain are recent initiates, it could be that population when drugs are injected for any purpose [6]. this difference in prevalence is mainly due to the fact that As is common with drug use, there are differences injection among MSM has spread later in Spain than in in local patterns. The prevalence in Madrid was higher other countries, especially the UK. In any case, we should than in Barcelona, and in Madrid mephedrone was remember the impact that the adoption of this adminis- the most injected, both for lifetime and the first injec- tration route can have on MSM’s health. It is well known tion, while in Barcelona it was methamphetamine. The that this risky behavior is much more efficient than sexual higher prevalence of injecting in Madrid than in Bar- practices at transmitting HIV, HBV and HCV [20], and, celona is another indicator that injecting among MSM in addition, for pharmacokinetic reasons, injection is one is a phenomenon quite independent of developments of the strongest correlates for overdose [21, 22] and for in the general population. In fact, in Barcelona inject- the development of dependence [23]. ing heroin or cocaine as a primary route of use is still The two strongest correlates with having ever injected almost twice as high as in Madrid in people requesting drugs were the number of partners participants had treatment for the use of these drugs [33, 34]. recently been penetrated by and having met those part- The high prevalence of sharing drugs or injection ners mainly at private parties. Both variables clearly indi- equipment should be noted: Almost two out of every cate that injecting occurs in MSM who frequent what are five FIS had done so, more than double that of non- known as chemsex sessions. In addition, injecting was FIS. It is not possible to compare this finding with that also associated with having injected steroids, since those offered by EMIS [29], since that study presents the who do so have probably lost their "fear of the needle" prevalence of sharing without differentiating whether [27, 31, 32]. The association found with having been paid participants have injected drugs or steroids and does for sex indicates that this behavior possibly occurs in the not offer this information by country. The Spanish context of sex work. A key point is that the association EMIS report [35] found 14.8% sharing for all injec- with having also had relationships with women is a wake- tors (of drugs or steroids). The study in France [14] up call since some MSM could act as a bridge population mentioned above found 21.5%. Comparisons with the for the dissemination of this behavior among women, as other studies in the review we have already discussed happened in the injection epidemic of the 80 s and 90 s [16] do not seem useful for the reasons already out- in the general population, when most the women who lined. Finally, it is worth noting that among FIS, more injected were introduced to this administration route by than four out of five have only ever injected for sex. It their injecting partner [27]. is unknown to what extent the current low percent- A surprising finding is that the mean age of the first age of those who have injected for other purposes may injection for FIS was over 30 and more than a quarter increase in the future, once they have crossed the bar- rier into needle use. Guerras et al. Harm Reduct J (2021) 18:86 Page 9 of 11 Limitations receive money in exchange for sex. The most approxi- First of all, it should be borne in mind that like almost all mate ones are the harm reduction guidelines for slam- studies in this population, this study employed conveni- ming during sex sessions [38]. Taking into account the ence sampling, in this case MSM accessing HIV testing. interventions that have been suggested by the partici- In order to increase the sample’s heterogeneity and rep- pants themselves can also be a good starting point [39, resentativeness, a sample of significant size was recruited 40]. However, these initiatives should take into account and two programs with very different client recruitment for the challenge posed by the likely existence of cul- characteristics were chosen in each city. Still, caution ture of “counterpublic health.” This culture has been should be exercised in generalizing results even among found to be underpinned by forms of “sex-based social- HIV-negative MSM, because convenience samples tend ity,” which gives primacy to the priorities and practices to sample higher-risk MSM than general population sur- of gay and bisexual men, as a recent study in Australia veys [36, 37]. However, by not including HIV-positive has shown [41]. MSM, it is almost certain that some findings are under- estimates of the true prevalences should we generalize Abbreviations the results to all MSM, since HIV-positive MSM have a ChSD: Chemsex drugs; CI: Confidence intervals; FIS: First injection was for sex; higher prevalence of injecting in all studies [14, 15, 18]. MSM: Men who have sex with other men; non-FIS: Non first injection was for Likewise, due to the low prevalence of injection overall, sex; PRs: Prevalence ratios; STI: Sexually transmitted infection. the comparison between FIS and non-FIS has certain limitations, with the FIS results being more accurate, Supplementary Information due to its higher prevalence. It would be very desirable The online version contains supplementary material available at https://​doi.​ org/​10.​1186/​s12954-​021-​00531-2. that the findings of this study could be confirmed by other studies carried out in MSM populations recruited Additional file 1. Equivalence of original questions and variables used in by different methods, not limited to large cities and also the analysis. including HIV-positive individuals. Acknowledgements The authors are grateful to Pink Peace whose rapid HIV testing programme Conclusions was responsible for the recruitment of participants in Madrid. To the best of our knowledge, this is the first study show- Methysos Project Group: Marta Donat, María del Carmen Burgos, César Pérez Romero (Instituto de Salud Carlos III, Madrid); José Antonio San Juan ing that among MSM, injecting for sex is being the most Bueno (Asociación Pink Peace, Madrid); Francisca Román Urrestarazu, Jesus common first episode of drug injection, which is also E Ospina, Miguel Alarcón Gutiérrez (Agència de Salut Pública de Barcelona, the most efficient behavior for the transmission of HIV, Barcelona); Oskar Ayerdi, Carmen Rodríguez, Sonsoles del Corral Del Campo, Natividad Jerez Zamora, Marta Ruiz Fernández, Montserrat González Polo Hepatitis B or C and other blood-borne infections. How- (Centro Sanitario Sandoval, Madrid); María Jesús Barbera Gracia, Luis López ever, only two out of a hundred had injected, most very Pérez, Claudia Broto Cortes, Julio Morais Martin (UITS Drassanes, Barcelona). occasionally but for the purpose of having sex, and shar- Authors’ contributions ing drugs or injecting equipment was frequent in these LdF, PGO and MJB conceived, designed and supervised the study. JMG and events. However, in each of the cities, a different drug PGO organized the recruitment of participants. DP, JdR and JNGP contributed predominated. It was not explicitly asked about, but the to the recruitment of subject and data collection. JMG and JH performed the main analyses. LdF, JMG and JH wrote the manuscript. LdF and MJB were correlates found point to it occurring in the context of responsible for drafting and critical revisions of the manuscript. JMG, JH, PGO, chemsex sessions with multiple partners and often when LdF, DP, JdR, JNGP, MJB and the additional members of the Methysos Project the new injector is paid for sex. Finally, MSM who also Group made substantive contributions to the current article. All authors read and approved the final manuscript. have sex with women are more involved, so there is a risk that it will become the bridge population for the spread Funding of injecting to women, and indirectly to non-MSM, as This study was supported by Delegación del Gobierno para el Plan Nacional sobre Drogas (2019I017). Delegación del Gobierno para el Plan Nacional happened in the injecting epidemic at the end of the last sobre Drogas had no further role in study design, in the collection, analysis century [27]. and interpretation of data, in the writing of the report, and in the decision to MSM participating in private parties, especially those submit the paper for publication. who are paid to have sex, should be considered a prior- Availability of data and materials ity group for prevention and harm reduction policies. The datasets used and/or analyzed during the current study are available from To prevent this risky behavior and minimize its harms the corresponding author on reasonable request. it will be necessary to design imaginative programs, since there are no harm reduction strategies that have been evaluated or even designed specifically to target MSM who inject drugs in private sex parties and who Guerras et al. Harm Reduct J (2021) 18:86 Page 10 of 11 Declarations high-risk sexual behaviors and sexually transmitted infections in hiv- positive men who have sex with men: data from the U-SEX GESIDA 9416 Ethics approval and consent to participate study. AIDS Patient Care STDS. 2018;32(3):112–8. https://​doi.​org/​10.​1089/​ Informed consent was obtained from all subjects involved in the study. apc.​2017.​0263. The study was approved by the Research Ethics Committee of the Insti‑ 11. Guerras JM, Hoyos J, Agusti C, Chanos S, Pichon F, Kuske M, et al. Associa‑ tuto de Salud Carlos III (CEI PI 44_2018_subproyecto1-v2 and CEI PI tion of sexualized drug use patterns with HIV/STI transmission risk in an 44_2018_subproyecto2). internet sample of men who have sex with men from Seven European Countries. Arch Sex Behav. 2021;50(2):461–77. https://​doi.​org/​10.​1007/​ Consent for publication s10508-​020-​01801-z. Not applicable. 12. Guerras JM, Hoyos J, Agusti C, Casabona J, Sordo L, Pulido J, et al. Sub‑ stance use in sexual context among spanish resident men who have sex Competing interests with men. Adicciones. 2020. https://​doi.​org/​10.​20882/​adicc​iones.​1371. The authors declare that they have no competing interests. 13. Hampel B, Kusejko K, Kouyos RD, Boni J, Flepp M, Stockle M, et al. Chem‑ sex drugs on the rise: a longitudinal analysis of the Swiss HIV Cohort Author details Study from 2007 to 2017. HIV Med. 2019. https://​doi.​org/​10.​1111/​hiv.​ 1 Centro Nacional de Epidemiología, Instituto de Salud Carlos III, Madrid, Spain. 12821. 2 CIBER Epidemiologia Y Salud Pública (CIBERESP), Madrid, Spain. 3 Servicio 14. Trouiller P, Velter A, Saboni L, Sommen C, Sauvage C, Vaux S, et al. Inject‑ de Epidemiología, Agència de Salut Pública de Barcelona, Barcelona, Spain. ing drug use during sex (known as “slamming”) among men who have 4 Escuela Nacional de Sanidad, Instituto de Salud Carlos III, Monforte de Lemos sex with men: results from a time-location sampling survey conducted 5, 28029 Madrid, Spain. 5 Centro Sanitario Sandoval, Instituto de Investigación in five cities, France. Int J Drug Policy. 2020;79: 102703. https://​doi.​org/​10.​ Sanitaria San Carlos, Hospital Clínico San Carlos, Madrid, Spain. 6 Unidad de 1016/j.​drugpo.​2020.​102703. ITS de Vall D’Hebron‑Drassanes, Hospital Vall D’Hebron, Barcelona, Spain. 15. Melendez-Torres GJ, Bourne A, Hickson F, Reid D, Weatherburn P. Cor‑ 7 Departamento de Salud Pública Y Materno‑Infantil, Universidad Complutense relates and subgroups of injecting drug use in UK gay and bisexual men: de Madrid, Madrid, Spain. findings from the 2014 Gay Men’s Sex Survey. Drug Alcohol Depend. 2018;187:292–5. https://​doi.​org/​10.​1016/j.​druga​lcdep.​2018.​03.​014. Received: 29 March 2021 Accepted: 27 July 2021 16. Heinsbroek E, Glass R, Edmundson C, Hope V, Desai M. Patterns of inject‑ ing and non-injecting drug use by sexual behaviour in people who inject drugs attending services in England, Wales and Northern Ireland, 2013–2016. Int J Drug Policy. 2018;55:215–21. https://​doi.​org/​10.​1016/j.​ drugpo.​2018.​02.​017. References 17. Scheibein F, Wells J, Henriques S, Van Hout MC. “Slam Sex”—Sexualized 1. Sanchez-Niubo A, Fortiana J, Barrio G, Suelves JM, Correa JF, Domingo- Injecting Drug Use (“SIDU”) Amongst Men Who Have Sex with Men Salvany A. Problematic heroin use incidence trends in Spain. Addiction. (MSM)-A Scoping Review. J Homosex. 2020. https://​doi.​org/​10.​1080/​ 2009;104(2):248–55. https://​doi.​org/​10.​1111/j.​1360-​0443.​2008.​02451.x. 00918​369.​2020.​18042​58. 2. Castilla J, de la Fuente L. Trends in the number of human immunodefi‑ 18. Schreck B, Victorri-Vigneau C, Guerlais M, Laforgue E, Grall-Bronnec M. ciency virus infected persons and AIDS cases in Spain: 1980–1998. Med Slam Practice: A Review of the Literature. Eur Addict Res. 2020. https://​ Clin (Barc). 2000;115(3):85–9. https://​doi.​org/​10.​1016/​s0025-​7753(00)​ doi.​org/​10.​1159/​00051​1897 71473-4. 19. Glass R, Hope VD, Tanner C, Desai M, “Slamming” among men who have 3. de la Fuente L, Bravo MJ, Barrio G, Parras F, Suarez M, Rodes A, et al. Les‑ sex with men accessing general drug services, in response to Schmidt, sons from the history of the human immunodeficiency virus/acquired AJ et al., Illicit drug use among gay and bisexual men in 44 cities: Find‑ immunodeficiency syndrome epidemic among Spanish drug injectors. ings from the European MSM Internet Survey (EMIS). Int J Drug Policy. Clin Infect Dis. 2003;37:S410–5. https://​doi.​org/​10.​1086/​377562. 2016;2017(49):24–5. https://​doi.​org/​10.​1016/j.​drugpo.​2017.​07.​028. 4. European Centre for the Epidemiological Monitoring of AIDS. HIV/AIDS 20. Patel P, Borkowf CB, Brooks JT, Lasry A, Lansky A, Mermin J. Estimating per- Surveillance in Europe. End-year report 1999. 2000 [cited 2021 Mar 28]. act HIV transmission risk: a systematic review. AIDS. 2014;28(10):1509–19. https://​www.​who.​int/​hiv/​strat​egic/​en/​euroh​iv62.​pdf?​ua=1 https://​doi.​org/​10.​1097/​QAD.​00000​00000​000298. 5. de la Fuente L, Barrio G, Royuela L, Bravo MJ. The transition from injecting 21. Brugal MT, Barrio G, de la Fuente L, Regidor E, Royuela L, Suelves JM. to smoking heroin in three Spanish cities. Addiction. 1997;92(12):1749– Factors associated with non-fatal heroin overdose: assessing the effect of 63. https://​doi.​org/​10.​1111/j.​1360-​0443.​1997.​tb028​95.x. frequency and route of heroin administration. Addiction. 2002;97(3):319– 6. de la Fuente L, Bravo MJ, Toro C, Brugal MT, Barrio G, Soriano V, et al. 27. https://​doi.​org/​10.​1046/j.​1360-​0443.​2002.​00058.x. Injecting and HIV prevalence among young heroin users in three Spanish 22. Degenhardt L, Bucello C, Mathers B, Briegleb C, Ali H, Hickman M, et al. cities and their association with the delayed implementation of harm Mortality among regular or dependent users of heroin and other opioids: reduction programmes. J Epidemiol Community Health. 2006;60(6):537– a systematic review and meta-analysis of cohort studies. Addiction. 42. https://​doi.​org/​10.​1136/​jech.​2005.​037333. 2011;106(1):32–51. https://​doi.​org/​10.​1111/j.​1360-​0443.​2010.​03140.x. 7. Unidad de Vigilancia de VIH y Comportamientos de Riesgo. Vigilancia 23. Barrio G, de la Fuente L, Lew C, Royuela L, Bravo MJ, Torrens M. Differences Epidemiológica del VIH y sida en España 2019: Sistema de Información in severity of heroin dependence by route of administration: the impor‑ sobre Nuevos Diagnósticos de VIH y Registro Nacional de Casos de Sida. tance of length of heroin use. Drug Alcohol Depend. 2001;63(2):169–77. Plan Nacional sobre el Sida - D.G. de Salud Pública, Calidad e Innovación https://​doi.​org/​10.​1016/​s0376-​8716(00)​00204-0. / Centro Nacional de Epidemiología - ISCIII. Madrid. [cited 2021 Mar 28]. 24. Jin F, Dore GJ, Matthews G, Luhmann N, Macdonald V, Bajis S, et al. https://​www.​mscbs.​gob.​es/​ciuda​danos/​enfLe​siones/​enfTr​ansmi​sibles/​ Prevalence and incidence of hepatitis C virus infection in men who have sida/​vigil​ancia/​Infor​me_​VIH_​SIDA_​20201​130.​pdf sex with men: a systematic review and meta-analysis. Lancet Gastroen‑ 8. Rosinska M, Gios L, Nostlinger C, Vanden Berghe W, Marcus U, Schink S, terol Hepatol. 2021;6(1):39–56. https://​doi.​org/​10.​1016/​S2468-​1253(20)​ et al. Prevalence of drug use during sex amongst MSM in Europe: Results 30303-4. from a multi-site bio-behavioural survey. Int J Drug Policy. 2018;55:231– 25. Barros AJ, Hirakata VN. Alternatives for logistic regression in cross-sec‑ 41. https://​doi.​org/​10.​1016/j.​drugpo.​2018.​01.​002. tional studies: an empirical comparison of models that directly estimate 9. Schmidt AJ, Bourne A, Weatherburn P, Reid D, Marcus U, Hickson F. Illicit the prevalence ratio. BMC Med Res Methodol. 2003;3:21. https://​doi.​org/​ drug use among gay and bisexual men in 44 cities: Findings from the 10.​1186/​1471-​2288-3-​21. European MSM Internet Survey (EMIS). Int J Drug Policy. 2016;38:4–12. 26. Espelt A, Mari-Dell’Olmo M, Penelo E, Bosque-Prous M. Applied Preva‑ https://​doi.​org/​10.​1016/j.​drugpo.​2016.​09.​007. lence Ratio estimation with different Regression models: An example 10. Gonzalez-Baeza A, Dolengevich-Segal H, Perez-Valero I, Cabello A, from a cross-national study on substance use research. Adicciones. Tellez MJ, Sanz J, et al. Sexualized drug use (Chemsex) Is associated with 2016;29(2):105–12. https://​doi.​org/​10.​20882/​adicc​iones.​823 Guerras et al. Harm Reduct J (2021) 18:86 Page 11 of 11 27. Bravo MJ, Barrio G, de la Fuente L, Royuela L, Domingo L, Silva TC. Reasons Available from: https://​sigma​resea​rch.​org.​uk/​files/​local/​EMIS-​2017_​Natio​ for selecting an initial route of heroin administration and for subsequent nal-​Report_​ES.​pdf transitions during a severe HIV epidemic. Addiction. 2003;98(6):749–60. 36. Schwarcz S, Spindler H, Scheer S, Valleroy L, Lansky A. Assessing repre‑ https://​doi.​org/​10.​1046/j.​1360-​0443.​2003.​00393.x. sentativeness of sampling methods for reaching men who have sex with 28. Gicquelais RE, Werb D, Marks C, Ziegler C, Mehta SH, Genberg BL, et al. men: a direct comparison of results obtained from convenience and Prevalence and correlates of providing and receiving assistance with the probability samples. AIDS Behav. 2007;11(4):596–602. https://​doi.​org/​10.​ transition to injection drug use. Epidemiol Rev. 2020;42(1):4–18. https://​ 1007/​s10461-​007-​9232-9. doi.​org/​10.​1093/​epirev/​mxaa0​08. 37. Prah P, Hickson F, Bonell C, McDaid LM, Johnson AM, Wayal S, et al. 29. EMIS Network. EMIS-2017: The European men-who-have-sex-with-men Men who have sex with men in Great Britain: comparing methods and internet survey. Key findings from 50 countries. 2019 [cited 2021 Mar 28]. estimates from probability and convenience sample surveys. Sex Transm Available from: https://​www.​ecdc.​europa.​eu/​sites/​defau​lt/​files/​docum​ Infect. 2016;92(6):455–63. https://​doi.​org/​10.​1136/​sextr​ans-​2015-​052389. ents/​Europ​ean-​MSM-​inter​net-​survey-​2017-​findi​ngs.​pdf 38. Grupo de Trabajo sobre Tratamientos del VIH (gTt-VIH). Guía para la 30. Barrett P, O’Donnell K, Fitzgerald M, Schmidt AJ, Hickson F, Quinlan M, reducción de daños asociados al uso de drogas inyectables en las et al. Drug use among men who have sex with men in Ireland: Prevalence sesiones de sexo. 2018 [cited 2021 Jul 19]. Available from: http://​gtt-​vih.​ and associated factors from a national online survey. Int J Drug Policy. org/​files/​active/​1/​Guia_​slamm​ing_​oct20​19_​web.​pdf 2018;64:5–12. https://​doi.​org/​10.​1016/j.​drugpo.​2018.​11.​011. 39. Birch R, Jollimore J, Howard T, Lal A, Cui Z, Rich AJ, et al. Hosts as Gate‑ 31. McBride AJ, Pates RM, Arnold K, Ball N. Needle fixation, the drug user’s keepers for North American Gay and Bisexual Men’s Private Group Sex perspective: a qualitative study. Addiction. 2001;96(7):1049–58. https://​ Parties. J Homosex. 2021:1–16. https://​doi.​org/​10.​1080/​00918​369.​2021.​ doi.​org/​10.​1046/j.​1360-​0443.​2001.​96710​4914.x. 19139​19 32. Strang J, Griffiths P, Gossop M. Heroin smoking by ’chasing the dragon’: 40. Nimbi FM, Rosati F, Esposito RM, Stuart D, Simonelli C, Tambelli R. origins and history. Addiction. 1997;92(6):673–83; discussion 85–95. Chemsex in Italy: experiences of men who have sex with men consum‑ https://​doi.​org/​10.​1046/j.​1360-​0443.​1997.​92667​34.x ing illicit drugs to enhance and prolong their sexual activity. J Sex Med. 33. Agència de Salut Pública de Catalunya. Sistema d’Informació sobre 2020;17(10):1875–84. https://​doi.​org/​10.​1016/j.​jsxm.​2020.​07.​001. Drogodependències de Catalunya. Informe anual 2019. 2020 [cited 2021 41. Drysdale K, Bryant J, Dowsett GW, Lea T, Treloar C, Aggleton P, et al. Mar 28]. Available from: https://​drogu​es.​gencat.​cat/​web/.​conte​nt/​minis​ Priorities and practices of risk reduction among gay and bisexual men ite/​drogu​es/​profe​ssion​als/​epide​miolo​gia/​docs/​INFOR​ME-​SIDC-​2019_​ok.​ in Australia who use crystal methamphetamine for sex. Int J Drug Policy. pdf 2021;93: 103163. https://​doi.​org/​10.​1016/j.​drugpo.​2021.​103163. 34. Comunidad de Madrid. Indicadores del consumo de drogas en la comu‑ nidad de Madrid año 2018. 2019 [cited 2021 Mar 28]. Available from: http://​www.​madrid.​org/​bvirt​ual/​BVCM0​20290.​pdf Publisher’s Note 35. EMIS Spain. Encuesta europea on-line para hombres que tienen sexo con Springer Nature remains neutral with regard to jurisdictional claims in pub‑ hombres (EMIS-2017): resultados en España. 2020 [cited 2021 Mar 28]. lished maps and institutional affiliations. Ready to submit your research ? Choose BMC and benefit from: • fast, convenient online submission • thorough peer review by experienced researchers in your field • rapid publication on acceptance • support for research data, including large and complex data types • gold Open Access which fosters wider collaboration and increased citations • maximum visibility for your research: over 100M website views per year At BMC, research is always in progress. Learn more biomedcentral.com/submissions