Men who have sex with men (MSM) usually show poorer health indicators than exclusively heterosexual men (HM) or general population men (GPM, that includes both MSM and HM), including mental disorders (e.g., depression or anxiety), chronic diseases or disabilities, and higher prevalence of human immunodeficiency virus (HIV), hepatitis C virus (HCV) and other blood-borne or sexually transmitted infections (STIs) [1,2,3,4,5,6]. Although less family and social support, discrimination, stigma, and victimization, as well as greater barriers to accessing optimal health care across their life spans, could contribute to this health disadvantage, it is probable that the greater frequency of particular unhealthy behaviours such as drug use, including tobacco and alcohol, or high-risk sexual practices are very important contributory factors to these health inequalities [1,4,7].

During the last decade a large number of research papers has been published, which have focused on the use of different drugs immediately before or during sex (sexualised drug use, SDU), especially centred on certain drugs whose use has spread recently among MSM and that have been commonly labelled as “chem drugs”: crystal methamphetamine, gamma-hydroxybutyrate/gamma-hydroxybutirolactone (GHB/GBL), mephedrone and, sometimes, ketamine. However, all these studies show that the prevalences of SDU of more traditional drugs (such as cannabis, cocaine, ecstasy or poppers) tend to be much higher than the so called “chem drugs” [8,9,10,11,12,13,14]. That said, while a very important factor, sex is only one of the reasons for using drugs, there are many others, and—with the exception of alkyl nitrites and the four chem drugs—psychoactive drugs are probably used more frequently by most MSM for purposes other than sex. In addition, some recent articles suggest that drug use, both for sex and recreational purposes, is increasing, in both HIV-positive [15] and HIV-negative MSM [16].

It is always assumed that the prevalence of drug use is much higher in MSM than in their heterosexual counterparts, although there are not many studies outside the USA comparing prevalence-of-use or related problems. First, it is not uncommon for authors to claim in their conclusions that MSM have higher levels of drug use than their heterosexual counterparts while having only MSM populations in their studies and no heterosexual men [17,18,19]. So these conclusions are only based on a rough direct comparison of their findings with the global figures for all males included in general population surveys, without any consideration of age structure differences. Second, many of these studies are restricted to alcohol, tobacco, cannabis or illicit/recreational drugs taken together, without considering the specific drugs involved [4,20,21,22,23,24,25]. All of these latter studies are based exclusively on general population surveys where the percentage of men who report having sex with other men is very low, making it very difficult to compare the prevalence-of-use of specific illicit drugs between MSM and other men. Practically all of these studies come from the USA except one from the UK [4], one from Australia [26] and another from Sweden [22] and these three have been published since 2016, because previous population surveys did not include questions about sexual behaviour. As far as we know, only four studies that actually compare particular levels of disaggregated prevalences of different drugs have been published [27,28,29,30]. One [28] was carried out among men (MSM and HM) attending two sexual health clinics in the UK, although this was in a fairly limited sample and did not include age adjustment. The other three studies come from the USA and present data for groups of drugs, but do not provide separate information for some specific substances of great interest: ecstasy, methamphetamine, ketamine, GHB/GBL and mephedrone [27,29,30].

On the other hand, in recent years there has been a growing interest in studying polydrug use (PDU) to classify users into different profiles or typologies (two or more subgroups with meaningful homogeneous characteristics) which take advantage of new methodologies, especially Latent Class Analysis (LCA). This methodology offers substantial advantages compared with more traditional techniques such as Cluster Analysis. LCA has been described as a person-oriented approach, because it focus on similarities and differences among individuals instead of relations among variables [31]. This approach has been used more frequently to identify subgroups of SDU among MSM, but its use in the analysis of PDU for any purpose is very scarce [32,33,34,35]. None of the aforementioned studies that have compared the prevalence between MSM and HM have performed LCA in the two populations to analyse how similar the subgroups of polydrug users and their prevalences are in the different populations.

The areas of Madrid and Barcelona are the two most populated in Spain and have higher prevalences of MSM [13,36,37]. Previous studies have shown that they are among the European cities with the highest SDU prevalences [8,9]. However, no study in these cities, nor in Spain, has ever compared drug-use prevalence for any purpose between MSM and GPM.

In this context, the objectives of this study were: first, to compare drug-use prevalence between a wide sample of HIV-negative MSM attending HIV/STI services in Madrid and Barcelona, and GPM (as a proxy of HM) from a general population survey; second, to identify the main subgroups of polydrug users in the two populations.