JMIR PUBLIC HEALTH AND SURVEILLANCE Pérez-Romero et al Original Paper Excessive Drinking Among Men Who Have Sex With Men Recruited From Web-Based Resources: Cross-sectional Questionnaire Study César Pérez-Romero1*, MPH; Juan-Miguel Guerras2,3*, PhD; Juan Hoyos4*, PhD; Marta Donat1,3*, MPH; Gregorio Barrio1,3*, PhD; Luis de la Fuente2,3*, PhD; David Palma3,5*, MPH; Patricia García de Olalla3,5*, MPH; María-José Belza1,3*, PhD; Methysos Project Group6* 1 Escuela Nacional de Sanidad, Instituto de Salud Carlos III, Madrid, Spain 2 Centro Nacional de Epidemiología, Instituto de Salud Carlos III, Madrid, Spain 3 Centro de Investigación Biomédica en Red de Epidemiología y Salud Pública, Madrid, Spain 4 Departamento de Salud Pública y Materno-Infantil, Universidad Complutense de Madrid, Madrid, Spain 5 Servicio de Epidemiología, Agència de Salut Pública de Barcelona, Barcelona, Spain 6 see Acknowledgments * all authors contributed equally Corresponding Author: Juan-Miguel Guerras, PhD Centro Nacional de Epidemiología Instituto de Salud Carlos III Avenida Monforte de Lemos 5 Madrid, 28029 Spain Phone: 34 918222699 Email: guerrasmoreira@hotmail.com Abstract Background: US and Northern European studies have found a higher prevalence of alcohol-related problems among men who have sex with men (MSM) than among the general population of men (GPM). However, most of them relied on traditional sampling methods, not profiting from MSM dating apps and websites for recruitment. Besides, analogous comparisons in Southern Europe are lacking. Objective: This study aimed to compare several indicators of excessive drinking between MSM and GPM in Spain. Methods: Overall, 5862 MSM were recruited through dating apps or websites for the Méthysos Project, and 10,349 GPM were recruited using probability sampling via the Household Survey on Alcohol and Drugs in Spain from 2018 to 2020. The outcomes were the prevalence of hazardous or harmful drinking (Alcohol Use Disorders Identification Test [AUDIT] ≥8), hazardous drinking (AUDIT-Consumption ≥4), harmful drinking (AUDIT-Problem ≥4), regular hazardous drinking (>14 standard drinks per week), and monthly binge drinking. The prevalence of excessive drinking indicators was calculated for MSM and GPM and compared using the adjusted prevalence ratio (aPR). Two different aPRs and their 95% CIs were estimated using Poisson regression models with robust variance. The first was adjusted for sociodemographic characteristics, and the second was adjusted for the aforementioned covariates plus other drug use. Results: The prevalence of hazardous or harmful drinking was 15.6% (913/5862) among MSM versus 7.7% (902/10,349) among GPM. After adjusting for sociodemographic covariates, the risk was higher in MSM than in GPM for harmful or hazardous drinking (aPR 1.8, 95% CI 1.6-2.0), harmful drinking (aPR 2.3, 95% CI 2.0-2.7), and binge drinking (aPR 1.7, 95% CI 1.5-1.9); the same in both populations for hazardous drinking (aPR 0.9, 95% CI 0.9-1.0); and higher in GPM than in MSM for regular hazardous drinking (aPR 0.7, 95% CI 0.6-0.9). The relative excess risk of harmful drinking and binge drinking among MSM tended to increase with increasing education level and size of the place of residence, and the opposite was true for the deficit risk in regular hazardous drinking. Additional adjustment for other drug use greatly buffered the relative excess risk in harmful drinking and binge drinking in MSM, while it deepened its deficit risk in regular hazardous drinking. https://publichealth.jmir.org/2022/10/e32888 JMIR Public Health Surveill 2022 | vol. 8 | iss. 10 | e32888 | p. 1 (page number not for citation purposes) XSL• FO RenderX JMIR PUBLIC HEALTH AND SURVEILLANCE Pérez-Romero et al Conclusions: The use of web-based resources allowed recruiting a large sample of MSM. The risk of hazardous or harmful drinking was 80% greater in MSM than in GPM, which was mainly because of the higher risk of harmful drinking and binge drinking among MSM. Nearly 1 in 6 MSM would benefit from early brief alcohol intervention procedures. The subgroup with harmful or binge drinking combined with other drug use is an important contributor to excess MSM risk in hazardous or harmful drinking and must be a priority target for harm reduction interventions. (JMIR Public Health Surveill 2022;8(10):e32888) doi: 10.2196/32888 KEYWORDS alcohol use; men who have sex with men; MSM; dating apps or websites; new recruitment methods; Alcohol Use Disorders Identification Test; AUDIT; hazardous drinking; harmful drinking; binge drinking; alcohol-related problems may overcome this problem and increase the size of the MSM Introduction sample. In recent years, concerns have been raised about health A common problem when comparing drinking indicators disparities due to sexual orientation [1-5]. Surveys in the United between MSM and GPM is the existence of important States have found a higher prevalence of alcohol-related differences between both groups in sociodemographic problems among men who have sex with men (MSM) compared characteristics (such as age, place of residence, immigration, with general population of men (GPM) [6-9]. However, studies or socioeconomic status) that can distort the comparison [24-28]. on the prevalence of excessive drinking and frequency of binge To avoid this, it is necessary to adjust the comparative measures drinking have not found significant increases among MSM for these covariates carefully. In addition, the evidence shows [1,2,7,10-13]. Here may be a disparity in alcohol indicators that the use of other psychoactive drugs is higher in MSM than between MSM and GPM depending on what is measured: in GPM [29], so the difference between comparative measures consumption levels or consequences of alcohol use. It should with and without adjustment for such covariates will allow us be noted that there are some exceptions: some studies describe to estimate which part of the MSM-GPM differences in drinking higher consumption levels among certain subgroups of MSM indicators can be explained by differences in the use of other [14-16]. drugs. To our knowledge, the only studies in Europe that compared The Alcohol Use Disorders Identification Test (AUDIT) is a alcohol use between MSM and GPM have been conducted in 10-item scale that screens for alcohol use disorders to identify Northern or Central European countries [2,17-21]. One study and assist individuals at high risk [30]. It includes several in Sweden showed more alcohol-related problems, but not subscales, the most widespread of which is the increased consumption, among MSM than among heterosexuals AUDIT-Consumption (AUDIT-C) (consumption items, Q1-Q3) [17]. Another study in Sweden, found a higher prevalence of [31-34]. However, although AUDIT-C performs well as a proxy high-risk alcohol use among homosexuals (men and women for hazardous drinking, it is less accurate in identifying jointly) but not among bisexual men [18]. Other studies in the alcohol-related problems [35-38], especially within the United Kingdom [2,20] or The Netherlands [21] did not observe framework of general population surveys [39]. The significant differences in consumption levels, while a recent complementary subscale AUDIT-Problem (AUDIT-P) study in Ireland [22] showed a 3-fold increase in the prevalence (problems items, Q4-Q10) has been shown to screen for these of alcohol use disorder in MSM compared with GPM, although issues as well as full-scale [35]. Some authors have suggested 2 different instruments were used. that this two-factor conceptualization (consumption and Knowledge of alcohol use among MSM living in southern problems, AUDIT-C and AUDIT-P) is a better fit for scale European countries is important given that historical cultural content than the three-factor structure originally proposed [40]. norms regarding alcohol use differ between these countries and To the best of our knowledge, only 2 published studies have those where research has been conducted [23]. The compared AUDIT scores between MSM and GPM [19,41]. The Mediterranean consumption pattern, characterized by almost results were not separated according to consumption or related daily drinking (primarily wine accompanying meals), may be problems. One of the two studies focused on the full-scale score, less followed by MSM living in Southern European countries whereas the other focused only on AUDIT-C. In contrast, other compared with GPM, given their higher openness to the cultural studies that analyzed AUDIT among MSM did not include a habits of other countries. This could affect findings among MSM comparative sample of GPM, and none analyzed AUDIT-P in terms of both alcohol consumption and related problems. separately [42-48]. In contrast, most of the former studies comparing MSM and In this study, we took advantage of dating apps and websites GPM are based on representative population samples commonly used by MSM to recruit a large sample of MSM and [2,7,9,12,16]. This method, although ensuring probabilistic compare various indicators of excessive drinking, including recruitment, entails as a substantial limitation the small sample hazardous and harmful drinking, between MSM and GPM in of MSM frequently included. We consider that recruitment Spain. We used the AUDIT to assess MSM who would methods based on new technologies, such as apps and websites, potentially benefit from early alcohol interventions. In addition, we aimed to discern the differences between the 2 dimensions https://publichealth.jmir.org/2022/10/e32888 JMIR Public Health Surveill 2022 | vol. 8 | iss. 10 | e32888 | p. 2 (page number not for citation purposes) XSL• FO RenderX JMIR PUBLIC HEALTH AND SURVEILLANCE Pérez-Romero et al of hazardous drinking (distinguishing regular hazardous drinking The questionnaire was self-administered, computer- or from binge drinking) and harmful drinking. app-based, and treated for sexual behavior and drug and alcohol use (including AUDIT). The full questionnaire is available in Methods the Multimedia Appendix 1. Study Design In studies using this type of methodology, it is impossible to calculate or estimate the response rate. The apps identify the We conducted a cross-sectional comparison of several measures number of displayed banners and sometimes the number of of excessive drinking between MSM and GPM in Spain. Data clicks on the banner, but not the number of different people were obtained from the results of the AUDIT questionnaire on who see or click on them to obtain some information. Something 2 different samples for MSM and GPM. similar happens with influencers. They may know the number Setting of views on a promotion, but not the number of people, or whether those people meet the criteria to participate. Of the The MSM sample included participants from the Méthysos 6602 MSM who began the questionnaire, 740 abandoned it Project, which aimed to investigate their health status. before completing the AUDIT questions; thus, the final sample Participants were recruited through different web-based comprised 5862 MSM. resources (see Participants, MSM Sample) between May and July 2020. The GPM sample was recruited between February GPM Sample and April 2018 within the framework of the Household Survey The GPM sample was obtained from the anonymous database on Alcohol and Drugs in Spain (Encuesta Sobre Alcohol y Otras of EDADES, 2017 edition provided by the National Plan on Drogas en España [EDADES], 2017 edition) [49]. Drugs. The EDADES is a biennial national survey of a Participants representative sample of the population aged 15-64 years living in Spain. It uses a 3-stage random sampling design (census tract, MSM Sample household, and individual sampling). Census tracts and subjects MSM were invited to participate through 3 types of web-based within households were selected using random probabilistic resources: methods. The sample was stratified by age (15-34/35-64) and living region (19 categories). In this edition, people aged 15 to 1. A total of 7 MSM dating apps using promotional banners 34 years and living in small regions were oversampled, so (Scruff, Grindr), personal messages (GROWLr), or both analyses were weighted to account for strata imbalance (Wapo, Bakala, MachoBB, and Xtudr) contributed 70.5% compared with the universe. The global response rate was (4655/6602) of the participants. 50.6%, and the main causes of nonresponse were not opening 2. Furthermore, 3 influencers largely followed by the MSM the door, preventing interviews, and the absence of all household community (Gabriel José Martin–with a video on YouTube members or the selected person. Before classifying a household available throughout the recruitment period–and or person as absent and selecting another household, the @frewaskachannel and @tigrilloig–with 24-hours available fieldworker should visit the household that was initially selected stories launched twice (@frewaskachannel) or once at least three times on different days and times. (@tigrilloig) during the recruitment period), which contributed 26.3% (1741/6602) of the participants. We selected all men aged 16-64 years who participated in the 3. Finally, a message encouraging the diffusion of the study survey (women and men aged 15 years were excluded). among friends and acquaintances, placed at the end of the EDADES does not have any variables to identify sexual identity questionnaire, and distribution lists from 3 organizations or behavior; thus, we could not estimate the proportion of MSM (Agència de Salut Pública de Barcelona, Pink Peace, and among the group of men participating in the survey, nor Chem-Safe), which contributed with 3.1% (206/6602) of excluding them in our analysis. Therefore, we refer to this group the participants. as GPM. However, we can infer that this sample is principally composed of heterosexual men, relying on the most recent Individuals who decided to participate were addressed to an estimation of the Spanish population on sexual orientation, in initial screen where they were informed about the aim and which 93.9% (2791/2972) of people declare themselves content of the survey. Before starting, they were obliged to heterosexual [50]. The EDADES used a self-administered check on the “I am of legal age for sexual intercourse and want paper-and-pencil questionnaire. The interviewer was present in to participate in the study.” All participants were required to the household to support the participant if needed. The questions meet the inclusion criteria of being male, at least 16 years old, included the AUDIT. We provided the full questionnaire used and having ever had anal intercourse with a male. We provided in the 2017 edition (only in Spanish) in the Multimedia unique links to each recruitment site to identify the recruitment Appendix 2. More information about the survey methodology site for each participant. To avoid multiple participation, the is available on web [49]. Of the 10,576 GPM who began the initial screen also included a request to complete the questionnaire, 227 (2.1%) abandoned it before completing the questionnaire only once if the invitation was received in various AUDIT questions; thus, the final sample comprised 10,349 ways. In addition, to further limit the possibility of multiple GPM. responses from one individual, we used the option provided by the software that only allowed the completion of one questionnaire per electronic device. No incentive was offered for participation, limiting the chances of multiple participation. https://publichealth.jmir.org/2022/10/e32888 JMIR Public Health Surveill 2022 | vol. 8 | iss. 10 | e32888 | p. 3 (page number not for citation purposes) XSL• FO RenderX JMIR PUBLIC HEALTH AND SURVEILLANCE Pérez-Romero et al Variables to include all harm-related questions (dependence and harm) in Only the participants who had consumed alcohol in the past 12 a single variable for harmful drinking. As explained above, this months answered the AUDIT questionnaire. For the remaining was done to accomplish the two-domain division of AUDIT participants, the AUDIT score was imputed as 0. Questionnaire (consumption and consequences). The results of the full-scale skips were automatically included in the resources used by the AUDIT (including questions for hazardous and harmful MSM sample but had to be manually included in the GPM drinking) were categorized as hazardous or harmful drinking. sample. Textbox 1 presents the AUDIT questionnaire. Regarding consumption levels, we differentiated a variable of This study uses labels and concepts from the recently published regular hazardous drinking (cumulative weekly exposure to 11th revision of the International Classification of Diseases [51] alcohol over a pre-established threshold) from binge drinking to refer to the different AUDIT measures analyzed. Hazardous to identify differences in drinking patterns. Although weekly alcohol use is defined as “a pattern of alcohol use that alcohol consumption above a certain threshold entails appreciably increases the risk of harmful physical or mental health-related consequences [52], these are expected to be more health consequences to the user or to others,” while a harmful severe when engaging in binge drinking [53]. Regular hazardous pattern of use of alcohol is that which “has caused damage to drinking has been named “excessive” or “heavy” drinking a person’s physical or mental health or has resulted in behavior elsewhere [15,54], while binge drinking (consumption of more leading to harm to the health of others.” Even if the AUDIT than 5 drinks on one occasion, sometimes equated to 2 hours) tool considers the concept of alcohol dependence [30], we opted is also referred to as “heavy episodic drinking” [12] or “risky single occasion drinking” [25]. https://publichealth.jmir.org/2022/10/e32888 JMIR Public Health Surveill 2022 | vol. 8 | iss. 10 | e32888 | p. 4 (page number not for citation purposes) XSL• FO RenderX JMIR PUBLIC HEALTH AND SURVEILLANCE Pérez-Romero et al Textbox 1. Alcohol Use Disorders Identification Test (AUDIT) questionnaire (English version) [30]. Full-scale AUDIT comprises all questions (score between 0 and 40; utilized threshold: 8), AUDIT-Consumption comprises questions Q1 to Q3 (score 0-12; utilized threshold: 4), and AUDIT-Problem questions Q4 to Q10 (score 0-28; utilized threshold: 4). Q1. How often do you have a drink containing alcohol? 0. Never (skip to Qs 9-10) 1. Monthly or less 2. 2-4 times a month 3. 2-3 times a week 4. 4 or more times a week Q2. How many drinks containing alcohol do you have on a typical day when you are drinking? 0. 1 or 2 1. 3 or 4 2. 5 or 6 3. 7, 8, or 9 4. 10 or more Q3. How often do you have six or more drinks on one occasion? 0. Never 1. Less than monthly 2. Monthly 3. Weekly 4. Daily or almost daily Skip to questions 9 and 10 if total score for questions 2 and 3=0 Q4. How often during the last year have you found that you were not able to stop drinking once you had started? 0. Never 1. Less than monthly 2. Monthly 3. Weekly 4. Daily or almost daily Q5. How often during the last year have you failed to do what was normally expected from you because of drinking? 0. Never 1. Less than monthly 2. Monthly 3. Weekly 4. Daily or almost daily Q6. How often during the last year have you needed a first drink in the morning to get yourself going after a heavy drinking session? 0. Never 1. Less than monthly 2. Monthly 3. Weekly 4. Daily or almost daily Q7. How often during the last year have you had a feeling of guilt or remorse after drinking? 0. Never 1. Less than monthly 2. Monthly https://publichealth.jmir.org/2022/10/e32888 JMIR Public Health Surveill 2022 | vol. 8 | iss. 10 | e32888 | p. 5 (page number not for citation purposes) XSL• FO RenderX JMIR PUBLIC HEALTH AND SURVEILLANCE Pérez-Romero et al 3. Weekly 4. Daily or almost daily Q8. How often during the last year have you been unable to remember what happened the night before because you had been drinking? 0. Never 1. Less than monthly 2. Monthly 3. Weekly 4. Daily or almost daily Q9. Have you or someone else been injured because of your drinking? 0. No 2. Yes, but not in the last year 4. Yes, during the last year. Q10. Has a relative or friend or a doctor or another health worker been concerned about your drinking or suggested you cut down? 0. No 2. Yes, but not in the last year 4. Yes, during the last year. at least monthly in a year). The AUDIT cutoff was originally Quantitative Variables established by the World Health Organization [24], and the Scores for the full-scale AUDIT (hazardous or harmful AUDIT-C has been widely used [27,42]. As there are no drinking), AUDIT-C (hazardous drinking), and AUDIT-P well-established cutoffs for AUDIT-P, we adapted the cutoff (harmful drinking or alcohol-related problems) instruments used by Hansen et al [52] considering the recommendations of were calculated by adding the scores in Q1-10, Q1-Q3, and the lower thresholds [24,29]. We selected the threshold for Q4-Q10, respectively. The weekly average number of standard regular hazardous drinking based on updated recommendations drinks was used as an indicator of the intensity of regular alcohol from national and international guidelines regarding limits on consumption. It was estimated by multiplying the answers for low-risk alcohol consumption [53-55]. questions on drinking frequency (Q1) and quantity (Q2). To this end, the median values of each response category of Q1 The prevalence of the different drinking indicators was and Q2 were used, except for the extreme categories “monthly calculated in both MSM and GPM and compared between the or less” in Q1, where 0.25 drinking days per week was assigned, groups using the adjusted prevalence ratio (aPR). aPR is a and “ten or more were assigned to Q2, where 10 standard drinks measure of relative inequality without units, which indicates per day were assigned. A similar method was used previously the number of times the risk is higher or lower in MSM [11]. The occurrence of binge drinking at least monthly was compared with GPM. Two different aPRs and their obtained from the AUDIT Q3 on binge drinking frequency. corresponding 95% CI were estimated using Poisson regression This method has been used elsewhere [55,56]. models with robust variance [57]. The first was adjusted for age, education level, size of place of residence, country of birth, Statistical Methods and perceived economic status (Model I), and the second for For the GPM sample, to account for strata imbalance compared the aforementioned covariates plus the use of other drugs in the with the universe, data were weighted by region (19 categories), past 12 months (Model II). Analyses were carried out globally size of place of residence (7 categories), age (7 categories), and and by age group (16-24, 25-34, and 35-64 years), education sex (2 categories), resulting in 619 different values for the level (university or no university), and size of their place of ponderation factor (from 0.04 to 3.06). For the MSM sample, residence (<10,000, 10,000-500,000, and >500,000 inhabitants). as the global universe of MSM residing in Spain was unknown, Categories were chosen following usual classifications and to all participants were assigned a ponderation factor of 1. An obtain a differentiated profile in young men (16-24 years) versus initial descriptive analysis of the general characteristics of the other groups, or in rural areas (<10,000 inhabitants), and large MSM and GPM samples was performed, and the statistical cities (>500,000 inhabitants) as opposed to other living significance of the differences was assessed using the Pearson conditions. Differences and comparative measures were χ2 test. The main outcomes were the prevalence of harmful or considered statistically significant if 2-tailed P-values were hazardous drinking (AUDIT score≥8), hazardous drinking <.05. The analysis was not preregistered, and the results should (AUDIT-C score≥4), harmful drinking or alcohol-related be considered exploratory. Statistical analyses were performed problems (AUDIT-P score≥4), regular hazardous drinking (>14 using Stata version 15 (StataCorp). standard drinks per week or >20 g of pure alcohol per day in a year), and monthly binge drinking (occurrence of binge drinking https://publichealth.jmir.org/2022/10/e32888 JMIR Public Health Surveill 2022 | vol. 8 | iss. 10 | e32888 | p. 6 (page number not for citation purposes) XSL• FO RenderX JMIR PUBLIC HEALTH AND SURVEILLANCE Pérez-Romero et al Ethical Considerations possibility of denying participation. For those who wanted to This study was approved by the Research Ethics Committee of participate, the interviewers provided informed consent before the Instituto de Salud Carlos III (Comité de Ética de la starting the study. Investigación [research ethics committee] Proyecto de Investigación [research project; principal investigator] Results 35_2020-v3). Concerning the MSM sample, individuals Baseline Characteristics approved their conformity to participate by clicking on “On my age, I am legally authorized to have sexual relations and I want Participants were more frequently aged 25-34 years (MSM) and to participate in the study.” This message appeared before 35-44 years (GPM). Compared with the GPM, they had a higher starting the questionnaire, as can be observed in the (Multimedia education level, lived more frequently in urban settings, and Appendix 3). Privacy was ensured by not asking for personal declared a better economic status. The prevalence of alcohol data, which led to possible identification. Regarding the GPM use in the past 12 months was similar between MSM and GPM, sample, individuals selected by randomized sampling had the but that of other drugs was much higher among MSM (Table 1). Table 1. Baseline characteristics of the sample: men who have sex with men (MSM) and general population men (GPM), Spain, 2018-2020. Characteristics MSM (N=5862), n (%a) GPM (N=10,349), n (%a,b) P value Age group (years) <.001 16-24 821 (14.0) 1955 (18.9) 25-34 1650 (28.1) 2701 (26.1) 35-44 1625 (27.7) 2665 (25.8) 45-54 1265 (21.6) 1668 (16.1) 55-64 501 (8.5) 1360 (13.1) Education level .001 ≤Lower secondary 586 (10.0) 5109 (49.6) Upper secondary 2104 (35.9) 3424 (33.2) University 3169 (54.1) 1767 (17.2) Place of birth <.001 Spain 4827 (82.3) 9116 (93.0) Latin America 797 (13.6) 110 (1.1) Other 238 (4.1) 580 (5.9) Size of place of residence (inhabitants) <.001 >1 million 1689 (31.6) 638 (6.2) 500,000-1 million 617 (11.5) 770 (7.4) 100,000-500,000 1155 (21.6) 2142 (20.7) 50,000-100,000 528 (9.9) 1613 (15.6) 10,000-50,000 734 (13.7) 2758 (26.6) <10,000 622 (11.6) 2428 (23.5) Perceived economic status <.001 Good 3447 (64.4) 3379 (48.3) Regular 1462 (27.3) 2732 (39.0) Bad 443 (8.3) 886 (12.7) Alcohol use in past 12 months 4729 (80.7) 8129 (78.5) .046 Other drug use in past 12 months 3529 (72.6) 2402 (23.6) <.001 a Percentages were estimated over cases with valid data. b Data were weighted by region, size of place of residence, age, and sex. https://publichealth.jmir.org/2022/10/e32888 JMIR Public Health Surveill 2022 | vol. 8 | iss. 10 | e32888 | p. 7 (page number not for citation purposes) XSL• FO RenderX JMIR PUBLIC HEALTH AND SURVEILLANCE Pérez-Romero et al Hazardous Drinking and Harmful Drinking The prevalence of hazardous drinking was similar for MSM The prevalence of hazardous or harmful drinking among MSM (1950/5862, 33.3%) and GPM (3472/10,349, 34.1%; aPR 0.9, was 15.6% (913/5862) compared with 7.7% (902/10,349) in 95% CI 0.9-1.0). A slight disparity favoring GPM was observed the GPM group, leading to an aPR of 1.8 (95% CI 1.6-2.0). in several subgroups: those aged 35 to 64, with no university Significant MSM-GPM disparities in prevalence in the same education level, and living in places with <10,000 inhabitants. direction were observed in all subgroups, except for those who Moreover, both MSM users and nonusers of other drugs were did not consume other drugs in the past 12 months (Table 2). less frequent hazardous drinkers than their GPM counterparts (Table 2). https://publichealth.jmir.org/2022/10/e32888 JMIR Public Health Surveill 2022 | vol. 8 | iss. 10 | e32888 | p. 8 (page number not for citation purposes) XSL• FO RenderX JMIR PUBLIC HEALTH AND SURVEILLANCE Pérez-Romero et al Table 2. Comparison of the prevalence of 3 Alcohol Use Disorders Identification Test (AUDIT) measures between men who have sex with men (MSM) and general population men (GPM), by age, education level, size of place of residence and other drug use, Spain, 2018-2020a. MSM, n (%) Sample size (n) GPM, n (%a) Sample size (n) aPRb (95% CI)c Hazardous or harmful drinking (AUDIT score ≥8) All participants 913 (15.6) 5862 902 (7.7) 10,349 1.8 (1.6-2.0) Age (years) 16-24 184 (22.4) 821 216 (10.4) 1955 1.5 (1.2-2.0) 25-34 310 (18.8) 1650 275 (9.4) 2701 1.8 (1.5-2.3) 35-64 419 (12.4) 3391 411 (6.7) 5693 1.8 (1.5-2.2) Education level No university 437 (16.2) 2690 773 (8.1) 8533 1.8 (1.5-2.1) University 475 (15.0) 3169 125 (6.0) 1767 1.7 (1.3-2.3) Size of place of residence (inhabitants) <10,000 73 (11.7) 622 213 (7.9) 2428 1.5 (1.1-2.0) 10,000-500,000 338 (14.0) 2417 566 (7.7) 6513 1.8 (1.5-2.1) >500,000 385 (16.7) 2306 123 (7.5) 1408 2.2 (1.7-2.9) Other drug use in past 12 months No 67 (5.0) 1332 443 (5.0) 7797 0.9 (0.7-1.3) Yes 755 (21.4) 3529 450 (16.9) 2402 1.2 (1.0-1.5)d Hazardous drinking (AUDIT-Consumption score ≥4) All participants 1950 (33.3) 5862 3472 (34.1) 10,349 0.9 (0.9-1.0) Age 16-24 316 (38.5) 821 595 (30.2) 1955 1.1 (0.9-1.3) 25-34 633 (38.4) 1650 923 (34.0) 2701 1.1 (1.0-1.3) 35-64 1001 (29.5) 3391 1954 (35.0) 5693 0.9 (0.8-0.9) Education level No university 849 (31.6) 2690 2900 (34.9) 8533 0.9 (0.8-1.0)d University 1100 (34.7) 3169 554 (30.4) 1767 1.0 (0.9-1.1) Size of place of residence (inhabitants) <10,000 188 (30.2) 622 946 (39.2) 2428 0.8 (0.7-0.9) 10,000-500,000 764 (31.6) 2417 2064 (32.9) 6513 1.0 (0.9-1.0) >500,000 815 (35.3) 2306 462 (32.4) 1408 1.0 (0.9-1.2) Other drug use in past 12 months No 219 (16.4) 1332 2303 (30.5) 7797 0.6 (0.5-0.7) Yes 1451 (41.1) 3529 1121 (46.3) 2402 0.9 (0-8-0.9) Harmful drinking (AUDIT-Problem score ≥4) All participants 715 (12.2) 5862 604 (5.0) 10,349 2.3 (2.0-2.7) Age (years) 16-24 173 (21.1) 821 164 (7.6) 1955 2.3 (1.7-3.1) 25-34 231 (14.0) 1650 177 (6.1) 2701 2.1 (1.6-2.7) 35-64 311 (9.2) 3391 263 (4.2) 5693 2.4 (1.9-3.0) Education level No university 366 (13.6) 2690 527 (5.4) 8533 2.2 (1.8-2.6) https://publichealth.jmir.org/2022/10/e32888 JMIR Public Health Surveill 2022 | vol. 8 | iss. 10 | e32888 | p. 9 (page number not for citation purposes) XSL• FO RenderX JMIR PUBLIC HEALTH AND SURVEILLANCE Pérez-Romero et al MSM, n (%) Sample size (n) GPM, n (%a) Sample size (n) aPRb (95% CI)c University 348 (11.0) 3169 74 (3.6) 1767 2.6 (1.7-3.8) Size of place of residence (inhabitants) <10,000 60 (9.6) 622 129 (5.0) 2428 1.8 (1.2-2.6) 10,000-500,000 270 (11.2) 2417 403 (5.2) 6513 2.1 (1.7-2.6) >500,000 296 (12.8) 2306 72 (4.3) 1408 3.4 (2.4-4.7) Other drug use in past 12 months No 61 (4.6) 1332 260 (2.9) 7797 1.4 (1.0-2.0)d Yes 583 (16.5) 3529 336 (12.3) 2402 1.5 (1.2-1.8) a Crude prevalence. For the GPM sample, data were weighted by region, size of place of residence, age, and sex. b aPR: adjusted prevalence ratio. c aPR were obtained from Poisson regression with robust variance in the framework of generalized linear models and adjusted by age, education level, size of place of residence, country of birth and economic status. The reference group (aPR=1) was general population men. d P<.05. disparities in such prevalence in the same direction were Regular Hazardous Drinking and Binge Drinking observed in all subgroups (Table 2). The prevalence of regular hazardous drinking (>14 drinks per week) was lower among MSM (245/5862, 4.2%) than among The aPRs of Model II after further adjustment for the use of GPM (524/10,349, 5.8%), with an aPR of 0.7 (95% CI 0.6-0.9). other drugs in the past 12 months are shown in Tables S1 and A statistically significant MSM-GPM disparity in such S2 in Multimedia Appendix 4. The considerable relative excess prevalence was not observed among individuals aged between prevalence in MSM found in Model I for both harmful and 16 and 34 years, with university education level, or living in binge drinking was greatly dampened in Model II. Thus, the places with >500,000 inhabitants (Table 3). aPR went from 2.3 (95% CI 2.0-2.7) in Model I to 1.5 (95% CI 1.3-1.7) in Model II for harmful drinking, and from 1.7 (95% The prevalence of monthly binge drinking was higher among CI 1.5-1.9) to 1.2 (95% CI 1.0-1.3) for binge drinking. MSM (798/5862, 13.6%) than among GPM (855/10,349, 7.6%), Regarding regular hazardous drinking, the relative prevalence with an aPR of 1.7 (95% CI 1.5-1.9). Significant MSM-GPM deficit in MSM found in Model I increased in Model II. Thus, disparities in prevalence in the same direction were observed aPR went from 0.7 (95% CI 0.6-0.9) in Model I to 0.5 (95% CI in all subgroups, except for other drug users (Table 3). 0.4-0.6) in Model II. In contrast, the aPR for hazardous drinking Finally, the prevalence of harmful drinking was higher among went from 0.9 (95% CI 0.9-1.0) in Model I to 0.8 (95% CI MSM (715/5862, 12.2%) than among GPM (604/10,349, 5.0%), 0.7-0.8) in Model II, and that of hazardous or harmful drinking, with an aPR of 2.3 (95% CI 2.0-2.7) Significant MSM-GPM from 1.8 (95% CI 1.6-2.0) to 1.2 (95% CI 1.0-1.4). https://publichealth.jmir.org/2022/10/e32888 JMIR Public Health Surveill 2022 | vol. 8 | iss. 10 | e32888 | p. 10 (page number not for citation purposes) XSL• FO RenderX JMIR PUBLIC HEALTH AND SURVEILLANCE Pérez-Romero et al Table 3. Comparison of the prevalence of regular hazardous drinking and monthly binge drinking between men who have sex with men (MSM) and general population men (GPM), by age, education level, size of place of residence, and other drug use, Spain, 2018-2020a. MSM, n/N (%) GPM, n/N (%) aPRb (95% CI)c Regular hazardous drinking (>14 drinks per week) All participants 245/5862 (4.2) 524/10,349 (5.8) 0.7 (0.6-0.9) Age (years) 16-24 25/821 (3.1) 43/1955 (2.2) 0.6 (0.3-1.2) 25-34 75/1650 (4.5) 103/2701 (4.2) 0.9 (0.6-1.4) 35-64 145/3391 (4.3) 378/5693 (7.0) 0.6 (0.5-0.8) Education level No university 104/2690 (3.9) 470/8533 (6.4) 0.6 (0.5-0.8) University 141/3169 (4.4) 53/1767 (3.5) 0.8 (0.5-1.2) Size of place of residence (inhabitants) <10,000 13/622 (2.1) 173/2428 (7.2) 0.3 (0.2-0.6) 10,000-500,000 77/2417 (3.2) 286/6513 (5.6) 0.6 (0.5-0.9) >500,000 106/2306 (4.6) 65/1408 (4.8) 1.1 (0.7-1.7) Other drug use in past 12 months No 13/1332 (1.0) 297/7797 (4.4) 0.3 (0.2-0.6) Yes 212/3529 (6.0) 221/2402 (10.6) 0.5 (0.4-0.6) Binge drinking (at least monthly) All participants 798/5862 (13.6) 855/10,349 (7.6) 1.7 (1.5-1.9) Age (years) 16-24 144/821 (17.5) 200/1955 (10.2) 1.7 (1.2-2.3) 25-34 270/1650 (16.4) 266/2701 (9.3) 1.7 (1.3-2.1) 35-64 384/3391 (11.3) 389/5693 (6.7) 1.6 (1.3-2.0) Education level No university 372/2690 (3.8) 740/8533 (8.1) 1.6 (1.4-1.9) University 426/3169 (13.4) 111/1767 (5.5) 1.8 (1.3-2.4) Size of the place of residence (inhabitants) <10,000 70/622 (11.3) 218/2428 (8.1) 1.4 (1.0-1.9)d 10,000-500,000 307/2417 (12.7) 527/6513 (7.5) 1.7 (1.4-2.1) >500,000 314/2306 (13.6) 110/1408 (7.6) 1.8 (1.4-2.4) Other drug use in the past 12 months No 70/1332 (5.3) 453/7797 (5.4) 1.0 (0.7-1.3) Yes 645/3529 (18.3) 393/2402 (15.4) 1.2 (1.0-1.4) a Crude prevalence. For the GPM sample, data were weighted by region, place of residence, age, and sex. b aPR: adjusted prevalence ratio. c aPR was obtained from Poisson regression with robust variance in the framework of generalized linear models and adjusted by age, education level, size of place of residence, country of birth, and economic status. The reference group (aPR=1) was general population men. d P<.05. living in Spain. Nearly 1 in 6 MSM had hazardous or harmful Discussion drinking and would benefit from early brief alcohol intervention Principal Findings procedures (the ultimate goal for which the AUDIT tool was designed) [30]. The MSM’s excess risk was mainly because of Compared with GPM, we found a relative excess risk of harmful drinking and binge drinking. In contrast, both hazardous or harmful drinking (full-scale AUDIT) among MSM populations had similar risks for hazardous drinking https://publichealth.jmir.org/2022/10/e32888 JMIR Public Health Surveill 2022 | vol. 8 | iss. 10 | e32888 | p. 11 (page number not for citation purposes) XSL• FO RenderX JMIR PUBLIC HEALTH AND SURVEILLANCE Pérez-Romero et al (AUDIT-C), resulting from the balance of a higher risk of harmful drinking among MSM compared with GPM. This study regular hazardous drinking among GPM than MSM, and the also found the same clear disparity between the two dimensions. opposite for binge drinking. One study [41] compared only hazardous drinking using The relative excess risk in the GPM of hazardous drinking and AUDIT-C and found a higher but not statistically significant regular hazardous drinking tended to increase among older, less prevalence in MSM compared with heterosexual men, which educated, and rural-dwelling individuals. In contrast, the relative agrees with the results of various studies using other indicators excess risk of harmful and binge drinking in MSM tended to [1,10,11]. However, the MSM sample size used in these studies be larger among individuals aged 16-24 years (only in the case tends to be limited, as is usually observed in studies derived of harmful drinking) or with university education level and from general population surveys. It was not possible to compare living in big cities (in both cases). our results in terms of regular alcohol consumption with those of 2 other studies that also performed this analysis because their Additional adjustment for other drug use greatly buffered the results were presented [2,15]. Our findings suggest that higher considerable relative excess risk in MSM compared to that in binge drinking among MSM contradicts the findings of most GPM for harmful drinking and binge drinking while deepening prior studies, which have reported no such differences between the risk deficit in MSM for regular hazardous drinking. This MSM and GPM [1,2,7,13], or an even lower prevalence among suggests that drinking behaviors and other drug use were certain subgroups of MSM [10,12]. strongly associated in the studied populations. Therefore, the considerable excess of risk in MSM compared with GPM for No other study has analyzed harmful drinking using AUDIT-P; binge drinking and harmful drinking is mainly because of the however, our findings of a high aPR MSM or GPM are similar subgroup having both these drinking patterns and the use of to those previously described in studies using Diagnostic and other drugs. This subgroup represents a much higher proportion Statistical Manual of Mental Disorders criteria [6-8]. of the total in MSM than in GPM, and must clearly be a priority Considering these findings, we can affirm that there is a slight target for harm reduction interventions. difference in hazardous drinking between MSM and GPM, but Comparison With Previous Work a remarkable increase in harmful drinking among MSM also This is the first study to directly compare excessive drinking exists in Mediterranean countries such as Spain. However, the indicators (hazardous and harmful drinking) between MSM and reasons for this discrepancy are not well understood. In the case GPM in a southern European country. Moreover, this is the first of Spain, it could be that MSM adopted Northern European study ever published that differentiates between these 2 drinking patterns more quickly, characterized by a higher dimensions using a validated instrument (AUDIT). Moreover, frequency of binge drinking, which is more likely to produce because of the use of new technologies in the recruitment harmful consequences [53]. As an indicator of more traditional process, namely apps and websites, it includes a larger sample habits, in our sample, the Mediterranean alcohol consumption of MSM than that in most previous studies inside and outside pattern was predominant among older, lower-educated, and Europe [1,2,6,7,9-18,41,58]. rural-dwelling GPM. The changes in consumption patterns among MSM may be related to their greater knowledge of habits Only one study has compared hazardous or harmful drinking in other countries because of their openness to cultural between these 2 populations using full-scale AUDIT [19], and differences and the higher proportion of young, the results showed a lower mean AUDIT score among MSM university-educated, large city dwellers in this group. Attributing than among heterosexual men. In this study, both populations differences between homosexuals and heterosexuals to were recruited using snowball sampling and the study did not differential consumption patterns was mentioned in one of the include a stratified analysis to distinguish between the 2 first studies on this issue, whose author suggested a higher dimensions evaluated here. Other studies [42-48] have estimated frequency of “European and continental drinking patterns” the proportion of MSM who engage in hazardous or harmful among Swedish MSM [17]. Nevertheless, there were no drinking using AUDIT but did not compare it to that of GPM. differences in consumption patterns between MSM and GPM. The proportion of hazardous or harmful drinking (full-scale AUDIT) in nearly all these studies was higher than that in our In contrast, minority stress theory has also been proposed to study, and it is likely because of the fact that these studies explain the increase in harmful and hazardous drinking among included individuals with a higher risk of unhealthy behaviors. MSM [59,60]. According to this theory, alcohol and other drug However, we believe that it does not make much sense to use is a response to psychological discomfort derived from compare these prevalences, nor those of the 2 dimensions stigma or internalized homophobia among lesbian, gay, bisexual, separately, because the levels and patterns of alcohol transgender community [61]. As our study did not explore consumption are primarily determined by the social environment psychological variables or drinking motives, further studies are in which one lives. From our perspective, the most relevant needed to interpret our findings in terms of this theory. measures to use are relative comparisons of MSM and their Other factors may have influenced the results. It is well known heterosexual counterparts, such as aPR. that certain potentially alcohol-linked phenomena, such as sex There are 3 studies [7,12,17] that have used instruments or work [62], chemsex [63], or sex tourism [64] are more frequent indicators other than AUDIT, which estimated both dimensions among MSM. Even though only 13.0% (431/3319) of MSM in separately. Their results showed low or insignificant differences our sample (data not shown, nonrespondents excluded from the in hazardous drinking but higher and significant differences in denominator) declared engaging regularly in binge drinking during chemsex sessions, and only 15.3% (898/5854) affirmed https://publichealth.jmir.org/2022/10/e32888 JMIR Public Health Surveill 2022 | vol. 8 | iss. 10 | e32888 | p. 12 (page number not for citation purposes) XSL• FO RenderX JMIR PUBLIC HEALTH AND SURVEILLANCE Pérez-Romero et al having received economic rewards for sex more than once in had occurred out of the general lockdown in March and April their life course, the influence of these variables warrants further 2020. investigation. Concerning sex tourism, 92.8% (596/642) of Another limitation is related to AUDIT thresholds. Cutoffs are foreign MSM in our sample declared living in Spain for at least needed for screening; however, when different cutoff points are 1 year, so their consideration as tourists can be ruled out. used, epidemiological research is hindered. We offset this Moreover, all analyses were adjusted for country of birth. limitation by estimating new results by applying different cutoffs In summary, it could be that there are differences in consumption used in other studies. In addition, problems related to thresholds patterns between MSM and GPM that have not been properly were not exclusive to AUDIT; they also appeared when analyzed, or, more probably, that MSM are more susceptible establishing the recommended levels of weekly alcohol to the harmful effects of alcohol use or that they have a greater consumption. The final limitation relates to the use of only 2 ability to identify or disclose them. Further research is needed AUDIT questions (quantity and frequency) to estimate alcohol to clarify this last point, which requires the use of large samples consumption. Although this was not the first study to measure for both comparison groups. Studies based on general population this variable using these 2 questions (or similar questions) [11], surveys, which represent the majority to date, do not permit a more complete module concerning consumption in the 2 discerning of these particular issues and often include a small surveys would have provided more detailed results. number of MSM. Conclusions and Implications for Intervention and Finally, although not an objective of this study, our findings Research suggest that AUDIT measures 2 different domains [40]. It also This is the first study to compare indicators of hazardous and illustrates the inadequacy of using the AUDIT-C scale in harmful drinking (problems or consequences) between MSM isolation to assess differences in alcohol-related problems among and GPM living in southern Europe. The use of different MSM [35]. Internet-based resources in recruitment (mainly apps and Limitations influencers) allowed us to obtain information on alcohol use The use of 2 different sampling methods for MSM and GPM from a large and diverse sample of MSM. Using a validated is a limitation of this study. However, using dating apps and brief screening tool (AUDIT), we found an 80% excess risk of websites for the recruitment of MSM was a way to overcome hazardous or harmful drinking among MSM, which was mainly 2 handicaps: the small sample of MSM frequently included in owing to the higher risk of harmful drinking and binge drinking studies based on representative samples of the general among MSM. population and the fact that the national drug survey in Spain From the perspective of the implications for clinical practice, does not include questions about sexual orientation. Although it is important to emphasize that the periodic use of brief this sampling method may have overrepresented hazardous screening tools such as AUDIT would seem to be highly drinkers among MSM, it provided a significantly larger MSM recommended in health consultations of MSM. This would sample than that in most studies with probabilistic sampling result in a significant proportion of MSM benefiting from early methods, in which the inclusion of a large number of MSM was brief alcohol harm reduction interventions and from referral to hindered by the low percentage of men who reported having specialized services when needed. Harm reduction strategies sex with other men. In our study, the MSM questionnaire was should focus on binge drinking. Nursing professionals in primary self-administered, which is not always guaranteed in general care and sexual transmitted infections or HIV clinics (which population surveys. This might have reduced the probability of take care of a large number of MSM) may provide appropriate reporting bias for nonnormative behaviors, such as homosexual services for the implementation of these screening instruments. behavior or alcohol use. From the perspective of future research, it is desirable to enrich Another limitation related to recruitment is the different contexts these findings by conducting studies based on a single in which alcohol was more frequently consumed in these 2 population-based survey. To achieve this goal, surveys in Spain groups. Even though dating apps as well as other Internet-based (and in other countries) should include a variable on sexual resources were used to recruit MSM, a skew toward behavior or gender identity, as it has been common in the United “party-goers” may have taken place. As some questions in States for a decade. Second, we want to highlight that there are AUDIT (notably in AUDIT-P, such as Q6 or Q8) are related to currently no clear explanations for the disparity in the findings nightlife drinking, this point may have had an influence. in the 2 domains of AUDIT. Neither the tendency toward Unfortunately, we did not have a variable to measure the context incorporation of non-Mediterranean consumption patterns of consumption. On the contrary, we do not think that recruiting among MSM nor the minority stress theory satisfactorily MSM, but not GPM, during the COVID-19 pandemic (between explains these results. Therefore, further studies are needed to May and July 2020) may have increased the differences favoring analyze whether differential consumption patterns exist or MSM. It has been observed that hazardous drinking increased whether there are different levels of susceptibility to alcohol during lockdowns [65-68], but AUDIT questions refer to use and its harmful consequences. common habits in the last 12 months, so most times evaluated https://publichealth.jmir.org/2022/10/e32888 JMIR Public Health Surveill 2022 | vol. 8 | iss. 10 | e32888 | p. 13 (page number not for citation purposes) XSL• FO RenderX JMIR PUBLIC HEALTH AND SURVEILLANCE Pérez-Romero et al Acknowledgments This work was funded by the National Drugs Plan (Ministry of Health, Government of Spain, 2019I017). We would like to thank the following websites, companies, and institutions who collaborated in the recruitment: Pink Peace, Wapo, Bakala, MachoBB, Scruff, Xtudr, GROWLr, Grindr, Chem-Safe, and Agencia de Salud Pública de Barcelona. We thank Gabriel José Martin (YouTube), @frewaskachannel, and @tigrilloig (Instagram). The authors would also like to thank the following members of the Methysos Project Group: Lidia Herrero, Instituto de Salud Carlos III, Madrid; José Antonio San Juan Bueno, Asociación Pink Peace, Madrid; Francisca Roman Urrestaruzo, Agència de Salut Pública de Barcelona, Barcelona; Jesus E Ospina, Agència de Salut Pública de Barcelona, Barcelona; Miguel Alarcón Gutiérrez, Agència de Salut Pública de Barcelona, Barcelona; Jorge del Romero, Centro Sanitario Sandoval, Madrid; Oskar Ayerdi, Centro Sanitario Sandoval, Madrid; Carmen Rodríguez, Centro Sanitario Sandoval, Madrid; Sonsoles del Corral Del Campo, Centro Sanitario Sandoval, Madrid; Natividad Jerez Zamora, Centro Sanitario Sandoval, Madrid; Marta Ruiz Fernández, Centro Sanitario Sandoval, Madrid; Montserrat González Polo, Centro Sanitario Sandoval, Madrid; Jorge-Néstor García Pérez, CAP Drassanes, Barcelona; María Jesús Barbera Gracia, CAP Drassanes, Barcelona; Luis López Pérez, CAP Drassanes, Barcelona; Claudia Broto Cortes, CAP Drassanes, Barcelona; Julio Morais Martin, CAP Drassanes, Barcelona. Authors' Contributions All authors made substantial contributions to conception and design, acquisition of data, or analysis and interpretation of data, drafted the article or revised it critically for important intellectual content, and granted final approval of the version to be published. Conflicts of Interest None declared. Multimedia Appendix 1 Men who have sex with men sample questionnaire. [ZIP File (Zip Archive), 4357 KB-Multimedia Appendix 1] Multimedia Appendix 2 General population of men sample questionnaire Encuesta Sobre Alcohol y Otras Drogas en España (Spanish National Survey on Alcohol and Drugs). [PDF File (Adobe PDF File), 768 KB-Multimedia Appendix 2] Multimedia Appendix 3 Message appeared before starting the questionnaire Méthysos. [PDF File (Adobe PDF File), 157 KB-Multimedia Appendix 3] Multimedia Appendix 4 Supplementary tables. [PDF File (Adobe PDF File), 86 KB-Multimedia Appendix 4] References 1. Blosnich JR, Farmer GW, Lee JG, Silenzio VM, Bowen DJ. Health inequalities among sexual minority adults: evidence from ten U.S. states, 2010. Am J Prev Med 2014 Apr;46(4):337-349 [FREE Full text] [doi: 10.1016/j.amepre.2013.11.010] [Medline: 24650836] 2. Mercer CH, Prah P, Field N, Tanton C, Macdowall W, Clifton S, et al. The health and well-being of men who have sex with men (MSM) in Britain: evidence from the third National Survey of Sexual Attitudes and Lifestyles (Natsal-3). BMC Public Health 2016 Jul 07;16:525 [FREE Full text] [doi: 10.1186/s12889-016-3149-z] [Medline: 27386950] 3. Daniel H, Butkus R, Health and Public Policy Committee of American College of Physicians. Lesbian, gay, bisexual, and transgender health disparities: executive summary of a policy position paper from the American College of Physicians. Ann Intern Med 2015 Jul 21;163(2):135-137 [FREE Full text] [doi: 10.7326/M14-2482] [Medline: 25961598] 4. McDermott E, Nelson R, Weeks H. The politics of LGBT+ health inequality: conclusions from a UK scoping review. Int J Environ Res Public Health 2021 Jan 19;18(2):826 [FREE Full text] [doi: 10.3390/ijerph18020826] [Medline: 33478019] 5. Rubin R. Minimizing health disparities among LGBT patients. JAMA 2015 Jan 06;313(1):15-17. [doi: 10.1001/jama.2014.17243] [Medline: 25562250] https://publichealth.jmir.org/2022/10/e32888 JMIR Public Health Surveill 2022 | vol. 8 | iss. 10 | e32888 | p. 14 (page number not for citation purposes) XSL• FO RenderX JMIR PUBLIC HEALTH AND SURVEILLANCE Pérez-Romero et al 6. Chaudhry AB, Reisner SL. Disparities by sexual orientation persist for major depressive episode and substance abuse or dependence: findings from a national probability study of adults in the United States. LGBT Health 2019 Jul;6(5):261-266 [FREE Full text] [doi: 10.1089/lgbt.2018.0207] [Medline: 31184987] 7. McCabe SE, Hughes TL, Bostwick WB, West BT, Boyd CJ. Sexual orientation, substance use behaviors and substance dependence in the United States. Addiction 2009 Aug;104(8):1333-1345 [FREE Full text] [doi: 10.1111/j.1360-0443.2009.02596.x] [Medline: 19438839] 8. Peralta RL, Victory E, Thompson CL. Alcohol use disorder in sexual minority adults: age- and sex- specific prevalence estimates from a national survey, 2015-2017. Drug Alcohol Depend 2019 Dec 01;205:107673. [doi: 10.1016/j.drugalcdep.2019.107673] [Medline: 31707274] 9. Rice CE, Vasilenko SA, Fish JN, Lanza ST. Sexual minority health disparities: an examination of age-related trends across adulthood in a national cross-sectional sample. Ann Epidemiol 2019 Mar;31:20-25 [FREE Full text] [doi: 10.1016/j.annepidem.2019.01.001] [Medline: 30792064] 10. Fish JN. Sexual orientation-related disparities in high-intensity binge drinking: findings from a nationally representative sample. LGBT Health 2019 Jul;6(5):242-249 [FREE Full text] [doi: 10.1089/lgbt.2018.0244] [Medline: 31184966] 11. Gruskin EP, Gordon N. Gay/Lesbian sexual orientation increases risk for cigarette smoking and heavy drinking among members of a large Northern California health plan. BMC Public Health 2006 Oct 03;6:241 [FREE Full text] [doi: 10.1186/1471-2458-6-241] [Medline: 17018152] 12. Schuler MS, Rice CE, Evans-Polce RJ, Collins RL. Disparities in substance use behaviors and disorders among adult sexual minorities by age, gender, and sexual identity. Drug Alcohol Depend 2018 Aug 01;189:139-146 [FREE Full text] [doi: 10.1016/j.drugalcdep.2018.05.008] [Medline: 29944989] 13. Boehmer U, Miao X, Linkletter C, Clark MA. Adult health behaviors over the life course by sexual orientation. Am J Public Health 2012 Feb;102(2):292-300. [doi: 10.2105/AJPH.2011.300334] [Medline: 22390443] 14. Fish JN, Hughes TL, Russell ST. Sexual identity differences in high-intensity binge drinking: findings from a US national sample. Addiction 2018 Apr;113(4):749-758 [FREE Full text] [doi: 10.1111/add.14041] [Medline: 28940778] 15. Gonzales G, Przedworski J, Henning-Smith C. Comparison of health and health risk factors between lesbian, gay, and bisexual adults and heterosexual adults in the united states: results from the national health interview survey. JAMA Intern Med 2016 Sep 01;176(9):1344-1351. [doi: 10.1001/jamainternmed.2016.3432] [Medline: 27367843] 16. Lunn MR, Cui W, Zack MM, Thompson WW, Blank MB, Yehia BR. Sociodemographic characteristics and health outcomes among lesbian, gay, and bisexual U.S. adults using healthy people 2020 leading health indicators. LGBT Health 2017 Aug;4(4):283-294 [FREE Full text] [doi: 10.1089/lgbt.2016.0087] [Medline: 28727950] 17. Bergmark KH. Drinking in the Swedish gay and lesbian community. Drug Alcohol Depend 1999 Sep;56(2):133-143 [FREE Full text] [doi: 10.1016/s0376-8716(99)00026-5] 18. Bränström R, Pachankis JE. Sexual orientation disparities in the co-occurrence of substance use and psychological distress: a national population-based study (2008-2015). Soc Psychiatry Psychiatr Epidemiol 2018 Apr;53(4):403-412 [FREE Full text] [doi: 10.1007/s00127-018-1491-4] [Medline: 29450600] 19. King M, McKeown E, Warner J, Ramsay A, Johnson K, Cort C, et al. Mental health and quality of life of gay men and lesbians in England and Wales: controlled, cross-sectional study. Br J Psychiatry 2003 Dec;183:552-558. [doi: 10.1192/bjp.183.6.552] [Medline: 14645028] 20. King M, Nazareth I. The health of people classified as lesbian, gay and bisexual attending family practitioners in London: a controlled study. BMC Public Health 2006 May 08;6:127 [FREE Full text] [doi: 10.1186/1471-2458-6-127] [Medline: 16681849] 21. Sandfort TG, de Graaf R, Bijl RV, Schnabel P. Same-sex sexual behavior and psychiatric disorders: findings from the Netherlands Mental Health Survey and Incidence Study (NEMESIS). Arch Gen Psychiatry 2001 Jan;58(1):85-91. [doi: 10.1001/archpsyc.58.1.85] [Medline: 11146762] 22. Daly FP, O'Donnell K, Davoren MP, Noone C, Weatherburn P, Quinlan M, et al. Potential alcohol use disorder among MSM in Ireland - findings from the European MSM internet survey (EMIS 2017). Drug Alcohol Depend 2021 Jun 01;223:108698 [FREE Full text] [doi: 10.1016/j.drugalcdep.2021.108698] [Medline: 33878574] 23. Popova S, Rehm J, Patra J, Zatonski W. Comparing alcohol consumption in central and eastern Europe to other European countries. Alcohol Alcohol 2007;42(5):465-473. [doi: 10.1093/alcalc/agl124] [Medline: 17287207] 24. Devaux M, Sassi F. Alcohol consumption and harmful drinking: trends and social disparities across OECD countries. Organisation for Economic Co-operation and Development. 2015. URL: https://www.oecd-ilibrary.org/docserver/ 5js1qwkz2p9s-en.pdf?expires=1664394697&id=id&accname=guest&checksum=7EF1A53E3F082FA36226484CD0E160C6 [accessed 2021-08-20] 25. Grittner U, Kuntsche S, Gmel G, Bloomfield K. Alcohol consumption and social inequality at the individual and country levels--results from an international study. Eur J Public Health 2013 Apr;23(2):332-339 [FREE Full text] [doi: 10.1093/eurpub/cks044] [Medline: 22562712] 26. Bloomfield K, Grittner U, Kramer S, Gmel G. Social inequalities in alcohol consumption and alcohol-related problems in the study countries of the EU concerted action 'Gender, Culture and Alcohol Problems: a Multi-national Study'. Alcohol Alcohol Suppl 2006;41(1):i26-i36. [doi: 10.1093/alcalc/agl073] [Medline: 17030500] https://publichealth.jmir.org/2022/10/e32888 JMIR Public Health Surveill 2022 | vol. 8 | iss. 10 | e32888 | p. 15 (page number not for citation purposes) XSL• FO RenderX JMIR PUBLIC HEALTH AND SURVEILLANCE Pérez-Romero et al 27. Sandoval JL, Leão T, Theler JM, Favrod-Coune T, Broers B, Gaspoz JM, et al. Alcohol control policies and socioeconomic inequalities in hazardous alcohol consumption: a 22-year cross-sectional study in a Swiss urban population. BMJ Open 2019 May 24;9(5):e028971 [FREE Full text] [doi: 10.1136/bmjopen-2019-028971] [Medline: 31129604] 28. Dixon MA, Chartier KG. Alcohol use patterns among urban and rural residents: demographic and social influences. Alcohol Res 2016;38(1):69-77 [FREE Full text] [Medline: 27159813] 29. Guerras JM, Hoyos J, García de Olalla P, de la Fuente L, Herrero L, Palma D, The Methysos Project Group. Comparison of polydrug use prevalences and typologies between men who have sex with men and general population men, in Madrid and Barcelona. Int J Environ Res Public Health 2021 Nov 04;18(21):11609 [FREE Full text] [doi: 10.3390/ijerph182111609] [Medline: 34770122] 30. World Health Organization, Babor T, Higgins-Biddle JC, Saunders JB, Monteiro MG. The Alcohol Use Disorders Identification Test: Guidelines for Use in Primary Care. 2nd edition. World Health Organization. Geneva, Switzerland: World Health Organization. Department of Mental Health and Substance Dependence; 2001. URL: https://apps.who.int/ iris/bitstream/handle/10665/67205/WHO_MSD_MSB_01.6a-eng.pdf?sequence=1&isAllowed=y [accessed 2021-08-20] 31. Charlebois ED, Plenty AH, Lin J, Ayala A, Hecht J. Impact of a structural intervention to address alcohol use among gay bar patrons in San Francisco: the PACE study. AIDS Behav 2017 Nov;21(Suppl 2):193-202 [FREE Full text] [doi: 10.1007/s10461-017-1891-6] [Medline: 28840396] 32. Motos-Sellés P, Cortés-Tomás MT, Giménez-Costa JA. Evaluation of AUDIT consumption items new adaptation to improve the screening of college students binge drinking. Adicciones 2020 Nov 17;32(4):255-264 [FREE Full text] [doi: 10.20882/adicciones.1145] [Medline: 31017997] 33. Liu Y, Ruan Y, Strauss SM, Yin L, Liu H, Amico KR, et al. Alcohol misuse, risky sexual behaviors, and HIV or syphilis infections among Chinese men who have sex with men. Drug Alcohol Depend 2016 Nov 01;168:239-246 [FREE Full text] [doi: 10.1016/j.drugalcdep.2016.09.020] [Medline: 27723554] 34. García Carretero MÁ, Novalbos Ruiz JP, Martínez Delgado JM, O'Ferrall González C. Validation of the alcohol use disorders identification test in university students: AUDIT and AUDIT-C. Adicciones 2016 Mar 02;28(4):194-204 [FREE Full text] [doi: 10.20882/adicciones.775] [Medline: 26990260] 35. Selin KH. Alcohol Use Disorder Identification Test (AUDIT): what does it screen? Performance of the AUDIT against four different criteria in a Swedish population sample. Subst Use Misuse 2006;41(14):1881-1899. [doi: 10.1080/10826080601025532] [Medline: 17162595] 36. Lundin A, Hallgren M, Balliu N, Forsell Y. The use of alcohol use disorders identification test (AUDIT) in detecting alcohol use disorder and risk drinking in the general population: validation of AUDIT using schedules for clinical assessment in neuropsychiatry. Alcohol Clin Exp Res 2015 Jan;39(1):158-165. [doi: 10.1111/acer.12593] [Medline: 25623414] 37. Dawson DA, Grant BF, Stinson FS, Zhou Y. Effectiveness of the derived Alcohol Use Disorders Identification Test (AUDIT-C) in screening for alcohol use disorders and risk drinking in the US general population. Alcohol Clin Exp Res 2005 May;29(5):844-854. [doi: 10.1097/01.alc.0000164374.32229.a2] [Medline: 15897730] 38. Moehring A, Rumpf HJ, Hapke U, Bischof G, John U, Meyer C. Diagnostic performance of the Alcohol Use Disorders Identification Test (AUDIT) in detecting DSM-5 alcohol use disorders in the general population. Drug Alcohol Depend 2019 Nov 01;204:107530. [doi: 10.1016/j.drugalcdep.2019.06.032] [Medline: 31505375] 39. Kriston L, Hölzel L, Weiser AK, Berner MM, Härter M. Meta-analysis: are 3 questions enough to detect unhealthy alcohol use? Ann Intern Med 2008 Dec 16;149(12):879-888. [doi: 10.7326/0003-4819-149-12-200812160-00007] [Medline: 19075207] 40. Moehring A, Krause K, Guertler D, Bischof G, Hapke U, Freyer-Adam J, et al. Measurement invariance of the alcohol use disorders identification test: establishing its factor structure in different settings and across gender. Drug Alcohol Depend 2018 Aug 01;189:55-61. [doi: 10.1016/j.drugalcdep.2018.05.002] [Medline: 29879682] 41. Roxburgh A, Lea T, de Wit J, Degenhardt L. Sexual identity and prevalence of alcohol and other drug use among Australians in the general population. Int J Drug Policy 2016 Mar;28:76-82. [doi: 10.1016/j.drugpo.2015.11.005] [Medline: 26691433] 42. Herrera MC, Konda KA, Leon SR, Deiss R, Brown B, Calvo GM, et al. Impact of alcohol use on sexual behavior among men who have sex with men and transgender women in Lima, Peru. Drug Alcohol Depend 2016 Apr 01;161:147-154 [FREE Full text] [doi: 10.1016/j.drugalcdep.2016.01.030] [Medline: 26896169] 43. Korhonen C, Kimani M, Wahome E, Otieno F, Okall D, Bailey RC, et al. Depressive symptoms and problematic alcohol and other substance use in 1476 gay, bisexual, and other MSM at three research sites in Kenya. AIDS 2018 Jul 17;32(11):1507-1515 [FREE Full text] [doi: 10.1097/QAD.0000000000001847] [Medline: 29734218] 44. Roth EA, Cui Z, Rich A, Lachowsky N, Sereda P, Card K, et al. Repeated measures analysis of alcohol patterns among gay and bisexual men in the momentum health study. Subst Use Misuse 2018 Apr 16;53(5):816-827 [FREE Full text] [doi: 10.1080/10826084.2017.1388259] [Medline: 29172871] 45. Tobin K, Davey-Rothwell M, Yang C, Siconolfi D, Latkin C. An examination of associations between social norms and risky alcohol use among African American men who have sex with men. Drug Alcohol Depend 2014 Jan 01;134:218-221 [FREE Full text] [doi: 10.1016/j.drugalcdep.2013.10.002] [Medline: 24210770] https://publichealth.jmir.org/2022/10/e32888 JMIR Public Health Surveill 2022 | vol. 8 | iss. 10 | e32888 | p. 16 (page number not for citation purposes) XSL• FO RenderX JMIR PUBLIC HEALTH AND SURVEILLANCE Pérez-Romero et al 46. Wirtz AL, Zelaya CE, Latkin C, Stall R, Peryshkina A, Galai N, et al. Alcohol use and associated sexual and substance use behaviors among men who have sex with men in Moscow, Russia. AIDS Behav 2016 Mar;20(3):523-536 [FREE Full text] [doi: 10.1007/s10461-015-1066-2] [Medline: 25893659] 47. Ludford KT, Vagenas P, Lama JR, Peinado J, Gonzales P, Leiva R, Peruvian HIV Sentinel Surveillance Working Group. Screening for drug and alcohol use disorders and their association with HIV-related sexual risk behaviors among men who have sex with men in Peru. PLoS One 2013 Aug 6;8(8):e69966 [FREE Full text] [doi: 10.1371/journal.pone.0069966] [Medline: 23936364] 48. Lu H, Han Y, He X, Sun Y, Li G, Li X, et al. Alcohol use and HIV risk taking among Chinese MSM in Beijing. Drug Alcohol Depend 2013 Dec 01;133(2):317-323. [doi: 10.1016/j.drugalcdep.2013.06.013] [Medline: 23876859] 49. Ministerio de Sanidad. La Encuesta sobre alcohol y otras drogas en España, EDADES. Gobierno de España. URL: https:/ /pnsd.sanidad.gob.es/profesionales/sistemasInformacion/sistemaInformacion/encuestas_EDADES.htm [accessed 2021-05-12] 50. Encuesta sobre relaciones sociales y afectivas en tiempos de pandemia de la COVID-19: Avance de resultados. Mayo-Junio 2021. Centro de Investigaciones Sociológicas (CIS). 2021. URL: http://datos.cis.es/pdf/Es3325marMT_A.pdf [accessed 2021-08-20] 51. ICD-11 for Mortality and Morbidity Statistics (ICD-11 MMS). World Health Organization. URL: https://icd.who.int/ browse11/l-m/en [accessed 2021-05-31] 52. Wood AM, Kaptoge S, Butterworth AS, Willeit P, Warnakula S, Bolton T, Emerging Risk Factors Collaboration/EPIC-CVD/UK Biobank Alcohol Study Group. Risk thresholds for alcohol consumption: combined analysis of individual-participant data for 599 912 current drinkers in 83 prospective studies. Lancet 2018 Apr 14;391(10129):1513-1523 [FREE Full text] [doi: 10.1016/S0140-6736(18)30134-X] [Medline: 29676281] 53. Bagnardi V, Zatonski W, Scotti L, La Vecchia C, Corrao G. Does drinking pattern modify the effect of alcohol on the risk of coronary heart disease? Evidence from a meta-analysis. J Epidemiol Community Health 2008 Jul;62(7):615-619. [doi: 10.1136/jech.2007.065607] [Medline: 18559444] 54. Rehm J. How should prevalence of alcohol use disorders be assessed globally? Int J Methods Psychiatr Res 2016 Jun;25(2):79-85 [FREE Full text] [doi: 10.1002/mpr.1508] [Medline: 27133364] 55. Åhlin J, Hallgren M, Öjehagen A, Källmén H, Forsell Y. Adults with mild to moderate depression exhibit more alcohol related problems compared to the general adult population: a cross sectional study. BMC Public Health 2015 Jun 09;15:542 [FREE Full text] [doi: 10.1186/s12889-015-1837-8] [Medline: 26051511] 56. Aalto M, Alho H, Halme JT, Seppä K. AUDIT and its abbreviated versions in detecting heavy and binge drinking in a general population survey. Drug Alcohol Depend 2009 Jul 01;103(1-2):25-29. [doi: 10.1016/j.drugalcdep.2009.02.013] [Medline: 19395203] 57. Espelt A, Marí-Dell'Olmo M, Penelo E, Bosque-Prous M. Applied Prevalence Ratio estimation with different Regression models: an example from a cross-national study on substance use research. Adicciones 2016 Jun 14;29(2):105-112 [FREE Full text] [doi: 10.20882/adicciones.823] [Medline: 27391846] 58. Kahle EM, Veliz P, McCabe SE, Boyd CJ. Functional and structural social support, substance use and sexual orientation from a nationally representative sample of US adults. Addiction 2020 Mar;115(3):546-558 [FREE Full text] [doi: 10.1111/add.14819] [Medline: 31599027] 59. Pakula B, Carpiano RM, Ratner PA, Shoveller JA. Life stress as a mediator and community belonging as a moderator of mood and anxiety disorders and co-occurring disorders with heavy drinking of gay, lesbian, bisexual, and heterosexual Canadians. Soc Psychiatry Psychiatr Epidemiol 2016 Aug;51(8):1181-1192. [doi: 10.1007/s00127-016-1236-1] [Medline: 27178431] 60. Brewster KL, Tillman KH. Sexual orientation and substance use among adolescents and young adults. Am J Public Health 2012 Jun;102(6):1168-1176. [doi: 10.2105/AJPH.2011.300261] [Medline: 22021322] 61. Meyer IH. Minority stress and mental health in gay men. J Health Soc Behav 1995 Mar;36(1):38-56 [FREE Full text] [doi: 10.2307/2137286] [Medline: 7738327] 62. Tan D, Holloway IW, Gildner J, Jauregui JC, Garcia Alvarez R, Guilamo-Ramos V. Alcohol use and HIV risk within social networks of MSM sex workers in the Dominican Republic. AIDS Behav 2017 Nov;21(Suppl 2):216-227 [FREE Full text] [doi: 10.1007/s10461-017-1896-1] [Medline: 28849279] 63. Torres TS, Bastos LS, Kamel L, Bezerra DR, Fernandes NM, Moreira RI, et al. Do men who have sex with men who report alcohol and illicit drug use before/during sex (chemsex) present moderate/high risk for substance use disorders? Drug Alcohol Depend 2020 Apr 01;209:107908. [doi: 10.1016/j.drugalcdep.2020.107908] [Medline: 32078972] 64. Brooks B, Park SH, Guilamo-Ramos V, Schneider JA, Harry-Hernandez S, Mgbako O, et al. Sex tourism and pre-exposure prophylaxis modality preferences among men who have sex with men. J Sex Res 2019;56(4-5):632-640. [doi: 10.1080/00224499.2018.1515343] [Medline: 30257113] 65. Jacob L, Smith L, Armstrong N, Yakkundi A, Barnett Y, Butler L, et al. Alcohol use and mental health during COVID-19 lockdown: a cross-sectional study in a sample of UK adults. Drug Alcohol Depend 2021 Feb 01;219:108488 [FREE Full text] [doi: 10.1016/j.drugalcdep.2020.108488] [Medline: 33383352] 66. Killgore WD, Cloonan SA, Taylor EC, Lucas DA, Dailey NS. Alcohol dependence during COVID-19 lockdowns. Psychiatry Res 2021 Feb;296:113676 [FREE Full text] [doi: 10.1016/j.psychres.2020.113676] [Medline: 33385782] https://publichealth.jmir.org/2022/10/e32888 JMIR Public Health Surveill 2022 | vol. 8 | iss. 10 | e32888 | p. 17 (page number not for citation purposes) XSL• FO RenderX JMIR PUBLIC HEALTH AND SURVEILLANCE Pérez-Romero et al 67. Barbosa C, Cowell AJ, Dowd WN. Alcohol consumption in response to the COVID-19 pandemic in the United States. J Addict Med 2021;15(4):341-344 [FREE Full text] [doi: 10.1097/ADM.0000000000000767] [Medline: 33105169] 68. Irizar P, Gage SH, Field M, Fallon V, Goodwin L. The prevalence of hazardous and harmful drinking in the UK Police Service, and their co-occurrence with job strain and mental health problems. Epidemiol Psychiatr Sci 2021 Jun 21;30:e51 [FREE Full text] [doi: 10.1017/S2045796021000366] [Medline: 34402422] Abbreviations aPR: adjusted prevalence ratio AUDIT: Alcohol Use Disorders Identification Test AUDIT-C: Alcohol Use Disorders Identification Test-Consumption AUDIT-P: Alcohol Use Disorders Identification Test-Problem EDADES: Encuesta Sobre Alcohol y Otras Drogas en España GPM: general population of men MSM: men who have sex with men Edited by T Sanchez, A Mavragani; submitted 26.08.21; peer-reviewed by A Sarasa-Renedo, R De Boni; comments to author 23.02.22; revised version received 11.05.22; accepted 06.09.22; published 31.10.22 Please cite as: Pérez-Romero C, Guerras JM, Hoyos J, Donat M, Barrio G, de la Fuente L, Palma D, García de Olalla P, Belza MJ, Methysos Project Group Excessive Drinking Among Men Who Have Sex With Men Recruited From Web-Based Resources: Cross-sectional Questionnaire Study JMIR Public Health Surveill 2022;8(10):e32888 URL: https://publichealth.jmir.org/2022/10/e32888 doi: 10.2196/32888 PMID: 36315235 ©César Pérez-Romero, Juan-Miguel Guerras, Juan Hoyos, Marta Donat, Gregorio Barrio, Luis de la Fuente, David Palma, Patricia García de Olalla, María-José Belza, Methysos Project Group. Originally published in JMIR Public Health and Surveillance (https://publichealth.jmir.org), 31.10.2022. This is an open-access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work, first published in JMIR Public Health and Surveillance, is properly cited. The complete bibliographic information, a link to the original publication on https://publichealth.jmir.org, as well as this copyright and license information must be included. https://publichealth.jmir.org/2022/10/e32888 JMIR Public Health Surveill 2022 | vol. 8 | iss. 10 | e32888 | p. 18 (page number not for citation purposes) XSL• FO RenderX