Identifying subgroups of Chinese men who have sex with men based on sexual behavior and drug use patterns using a clustering analysis approach

Identifying subgroups of Chinese men who have sex with men based on sexual behavior and drug use patterns using a clustering analysis approach | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Identifying subgroups of Chinese men who have sex with men based on sexual behavior and drug use patterns using a clustering analysis approach Bingyang She, Jiajun Sun, Fang Lu, Siqi Lin, Yi Liu, Gaixia Li, and 6 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-5286116/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 10 Apr, 2025 Read the published version in BMC Public Health → Version 1 posted 4 You are reading this latest preprint version Abstract Introduction: Sexualized drug use (SDU) refers to using drugs before and during sex to enhance experiences, increasing high-risk behaviors among men who have sex with men (MSM). This study explores how SDU affects sexual behavior in Chinese MSM. Methods: We collected information on demographics, sexual acts, drug use, and condom attitudes among 890 MSM in six Chinese cities via WeChat ads through community-based organizations from March 23 to April 22, 2022. Cluster analysis using Gower’s distance and hierarchical clustering explored differences in sexual acts among MSM who reported SDU in their last encounter and otherwise. Results: Cluster analysis categorized participants into three clusters. Cluster 3 (n=155) reported 100% SDU in their last sexual encounter (83.87% poppers use), whereas Clusters 1 (n=581) and 2 (n=154) reported none. Compared to other clusters, Cluster 3 had significantly higher PrEP use (34.90% vs. 17.02% vs. 8.00%, p<0.0001), more sexual acts over the past 12 months (35.80-61.30 vs. 31.30-56.10 and 4.37-21.22, p<0.001), more regular (3.16±4.37 vs. 2.27±3.52 vs. 2.51±2.53, p=0.028) and casual partners (4.55±6.55 vs. 2.48±3.21 vs. 2.74±3.66, p<0.0001), more partners with STIs (8.39% vs. 3.79% vs. 3.90%, p=0.029), and lower consistent condom use (48.53% vs. 59.41% vs. 72.28%, p<0.0001). Cluster 1 had moderate frequency in all sexual acts except self-masturbation, which was most common in Cluster 2. Conclusion: SDU is a stratum for identifying MSM subgroups, and MSM who reported SDU demonstrated higher sexual risk behaviors and PrEP usage. Among those not practicing SDU, self-masturbation is a key behavioral indicator for subgrouping. Sexualized drug use MSM Cluster analysis Sexual Behavioral Patterns sexual acts Figures Figure 1 Figure 2 Introduction Sexualized drug use (SDU), which includes chemsex, is defined as the intentional engagement in substance use before or during sexual contact to enhance the experience and has emerged as a worldwide concern. The substances involved in SDU often include mephedrone, crystal methamphetamine, Gamma hydroxybutyrate/Gamma butyrolactone (GHB/GBL), cocaine, and ketamine[ 1 – 5 ]. Increasingly, the interplay between substance use and high-risk sexual behavior has led to public health challenges, particularly among men who have sex with men (MSM). SDU, alongside inherently other high-risk sexual practices, elevates the risk of sexually transmitted infections (STIs) in MSM [ 6 – 8 ]. The prevalence of SDU in MSM varies significantly across different countries and regions, ranging from 34–79% in developed nations such as the United States, United Kingdom, and Germany, whereas in China, prevalence ranges between 18–28% [ 9 , 10 ]. SDU can profoundly affect cognitive functions and increase risk-taking sexual activity. Under the influence of drugs, MSM may engage in prolonged sexual sessions, potentially resulting in physical injuries or traumas [ 11 ]. Moreover, high-risk sexual behavior under the influence of SDU, like condomless sexual behavior, increases the spread of HIV and STIs [ 12 ]. Further, drug-enhanced sense of intimacy can lead individuals to engage in various sexual acts during a single encounter and facilitate group sex, increasing the risk of HIV/STI exposure [ 10 , 12 , 13 ]. To date, only limited research has been conducted to understand the pattern and frequency of sexual acts in MSM who practice SDU. Only a few studies have mentioned a potential association between the frequency of rimming and the use of poppers, methamphetamine, GHB, and ecstasy/MDMA [ 8 , 14 , 15 ]. Noticeable knowledge gap remains in the understanding of the relationship between sexual behavioral patterns and SDU in MSM [ 1 , 16 – 24 ]. Conventional analytical approaches face challenges in identifying the subtle differences across various subpopulations in MSM due to the preference and complex nature of their sexual behaviors. To overcome this, we used cluster analysis, a machine learning technique, to simplify the data and uncover underlying patterns. Cluster analysis is widely used in behavioral studies to identify subpopulations [ 25 – 28 ], making it an appropriate approach for analyzing SDU and related sexual behaviors in MSM. This study aims to use cluster analysis to identify MSM subpopulations according to their SDU behaviors and compare the differences in their sexual behavioral pattern across identified clusters. This approach may provide evidence for targeted intervention of high-risk sexual behavior and drug use during sexual encounters in MSM. Methods Study design and participants recruited Our study was an anonymous cross-sectional survey across China from March 23 to April 22, 2022, through community organizations and social groups. We recruited MSM nationally through advertisements on the WeChat platforms of six major gay service groups including Zhuhai Xutong Volunteer Service Center, Qingdao Qingtong Anti-AIDS Volunteer Service Center, Nanjing Xingyou Volunteer Service Center, Jinan Rainbow, Beijing LGBT + and Positive Peers Group, which are dedicated to HIV and STI education and intervention. The recruitment process provided detailed study information which included a general description of the project and addresses of participating health service stations and CDCs and offered cash incentives for participation. Eligible participants who consented to participate were directed to complete a detailed questionnaire on sexual behavior, substance use and recent sexual activities with men at Weng Juan Xing ( www.wjx.cn ). To be eligible for participation in the questionnaire, individuals had to meet three criteria: 1) biologically assigned as male at birth; 2) have a history of male-to-male anal sex; and 3) be between the ages of 18 and 70. Data collection At the beginning of our study, participants were asked to provide information about their sociodemographic characteristics, including gender, age, income, marital status, educational qualifications, and self-perceived gender identity. Regarding their sexual activity, the questionnaire surveyed the number of days since their last sexual acts and the sequence of their most recent sexual encounter. A significant part of the questionnaire focused on understanding drug use during their most recent sexual encounter. Participants were asked the question, 'Did you use drugs during the last time you had sex with a man?' and were given a list of substances to choose from, including Rush, Crystal Meth, and Ketamine, etc. We collected information on the frequency of sexual encounters with both regular and casual male partners and also the frequency of condom use over the past six months. We also collected the number of days since the most recent sexual act, which ranged from kissing to more intimate acts like insertive and receptive oral sex, anal sex, rimming, and masturbation (self, for a partner, by a partner). Definition of SDU We defined SDU as the behavior where participants use substances during their sexual encounter. These substances include, but are not limited to, Poppers, Ketamine, Ecstasy (MDMA), and GHB/GBL. Definition of Regular and Casual Sexual Partnerships Regular sexual partners were defined as individuals with whom a committed and lasting romantic or sexual relationship was maintained, including boyfriends and other male partners. Casual sexual partners were defined as individuals with whom sexual relationships had been established for a duration of three months or less. This category included male sex workers. Clustering analysis We selected 11 sexual behavioral indicators for clustering analysis, including the number of days since the participant’s last sexual encounter involving sexual acts such as kissing, insertive or receptive oral sex, insertive or receptive anal sex, rimming, masturbation (self, with a partner, or by a partner), and drug use during the last sexual encounter (SDU). Given the combination of continuous variables (e.g., the number of days since the participant’s last sexual acts) and binary categorical variables (e.g., SDU), we employed the Partitioning Around Medoids (PAM) algorithm using Gower’s distance to handle these mixed data types and the silhouette coefficient to determine the optimal number of clusters. Multidimensional scaling (MDS) was used to standardize both categorical and continuous variables [ 29 ]. PAM clustering analysis separated the population into two groups: those with SDU and those without. Heterogeneity within the non-SDU group prompted us to investigate further potential subgroups in this group using a hierarchical clustering approach based on sexual behavior frequency. The combined PAM and hierarchical clustering enabled us to account for both participants’ drug use and sexual behavioral patterns at two different levels. We used t-SNE for dimensionality reduction and visualization of the clustering structure in two dimensions. The visualization helped confirm the existence of clusters and sub-clusters. Estimating the number of sexual acts over the past 12 months We employed a log-normal distribution fitting method to estimate the frequency of various sexual behaviors, as the log-normal distribution effectively simulates the distribution of complex human behaviors such as sexual acts. We have used such a fitting in a previous study [ 29 ]. Based on the log-normal distribution fitting, we estimated the number of different sexual acts based on the number of days since the last sexual acts that were reported [ 30 , 31 ]. We then calibrated the simulated frequency to the observed accumulated frequency of sexual acts in the participants. This calibration enabled us to estimate the distribution of the number of sexual acts by their sexual act types even in the absence of empirical data. The calibration was considered successful when the difference (defined as Mean Square Error (MSE)) between the simulated and observed data was minimized: $$\:MSE=\frac{1}{n}\sum\:_{i=1}^{n}{\left({\stackrel{\prime }{y}}_{i}-{y}_{i}\right)}^{2}$$ Where, \(\:{\stackrel{\prime }{y}}_{i}∼LN\) ( \(\:{\mu\:}_{i,}{\delta\:}_{i}\) ) represents the log-normal distribution curve of the number of days of the last sex behaviors on type \(\:i\) , mean \(\:{\mu\:}_{i}\) and variance \(\:{\delta\:}_{i}\) , and \(\:{\stackrel{\prime }{y}}_{i}\) represents a randomly generated bootstrap datasets corresponding to the original datasets of sexual acts. The Mean Square Error (MSE) was used to compute the mean value of the sum of the squares of errors between the estimated data point and the bootstrapped data point. Bootstrapped data was generated by resampling the original data to create a dataset of similar dimensions as the fitted data. We repeated the estimated process 400 times and selected the curve with the smallest MSE value as the optimized curve. The best 50 fittings were chosen to establish a 95% confidence interval. After estimating the curves, we obtained their log-probability density function values. These probabilities were then converted into weights for days 1 to 365. The weights were used to estimate the number of sexual acts over the last 12 months. This allowed us to calculate the weighted mean number of sexual acts over last 12 months. The entire optimization and fitting data process was conducted using MATLAB R2020a. Statistical analysis Categorical variables in this study were described using numbers and percentages, and the chi-square test was used to compare differences between groups. For normally distributed continuous variables, mean ± standard deviation was used to present the data, and ANOVA (analysis of variance) was performed for comparison. Non-normally distributed continuous variables were reported as median and interquartile range, and differences between groups were assessed using the Wilcoxon rank-sum test. Result Clustering participants by sexual and drug-use behaviors Our study initially surveyed 1,034 individuals, excluding 144 due to lack of consent (n = 100), age discrepancies (n = 42), or duplicate responses (n = 2). Ultimately, 890 participants were included in the analysis (Figure S1 ). The first step of the clustering analysis categorized participants into two clusters: those whose most recent sexual encounter reported SDU (n = 155, cluster 3) and those who did not report SDU (n = 735) (Figure S2 ). Subsequently, the second step of the clustering analysis further divided the non-SDU in their last sexual encounter cluster into two groups, resulting in a total of three distinct clusters (Cluster 1, n = 581; Cluster 2, n = 154; Cluster 3, n = 155) (Fig. 1 , S3). Differences in demographic characteristics across clusters The majority of the study participants were single (88.99%), predominantly held a college or bachelor’s degree (73.37%) and had an income range of RMB 5001–8000 (26.07%). The average age of participants was 28.14 ± 7.37 years. About 17.42% of the participants reported engagement in SDU in their last sexual encounter. (Table 1 ) Table 1 Demographic and behavioral characteristics of study participants, stratified by Clusters Category Overall (N = 890) Cluster1 (N = 581) Cluster2 (N = 154) Cluster3 (N = 155) Chi 2 /F P-value Age (years) 28.14 ± 7.37 28.15 ± 7.22 27.85 ± 8.06 28.40 ± 7.27 0.215 0.806 Marital Status 6.591 0.159 Single 792 (88.99%) 525 (90.36%) 132 (85.71%) 135 (87.1%) Engaged or Married 54 (6.07%) 30 (5.16%) 10 (6.49%) 14 (9.03%) Separated or Divorced or Widowed 44 (4.94%) 26 (4.48%) 12 (7.79%) 6 (3.87%) Education level 8.19 0.085 High school or below 120 (13.48%) 89 (15.32%) 19 (12.34%) 12 (7.74%) College/Bachelors 653 (73.37%) 421 (72.46%) 116 (75.32%) 116 (74.84%) Masters or above 117 (13.15%) 71 (12.22%) 19 (12.34%) 27 (17.42%) Personal monthly income (RMB) 4.708 0.788 < 1500 106 (11.91%) 62 (10.67%) 23 (14.94%) 21 (13.55%) 1500–3000 112 (12.58%) 73 (12.56%) 21 (13.64%) 18 (11.61%) 3001–5000 225 (25.28%) 152 (26.16%) 39 (25.32%) 34 (21.94%) 5001–8000 232 (26.07%) 149 (25.65%) 38 (24.68%) 45 (29.03%) > 8001 215 (24.16%) 145 (24.96%) 33 (21.43%) 37 (23.87%) Use drugs during last sexual encounter Yes 155 (17.42%) 0 (0%) 0 (0%) 155 (100%) 890 <0.0001* No 735 (82.58%) 581 (100%) 154 (100%) 0 (0%) Drug Types Poppers 130 (83.87%) 0 (0%) 0 (0%) 130 (83.87%) 890 <0.0001* DMT 1 (0.65%) 0 (0%) 0 (0%) 1 (0.65%) Triazolam 1 (0.65%) 0 (0%) 0 (0%) 1 (0.65%) Others 23 (14.84%) 0 (0%) 0 (0%) 23 (14.84%) Crystal Meth 0 (0%) 0 (0%) 0 (0%) 0 (0%) Ketamine 0 (0%) 0 (0%) 0 (0%) 0 (0%) Ecstasy 0 (0%) 0 (0%) 0 (0%) 0 (0%) Ya ba pills 0 (0%) 0 (0%) 0 (0%) 0 (0%) Mixing drugs 0 (0%) 0 (0%) 0 (0%) 0 (0%) GHB 0 (0%) 0 (0%) 0 (0%) 0 (0%) Monkey dust 0 (0%) 0 (0%) 0 (0%) 0 (0%) Note: DMT: N, N-Dimethyltryptamine; GHB: Gamma-Hydroxybutyrate. While comparing the demographic characteristics of the participants, we observed that all participants who reported SDU in their last sexual encounter were exclusively found in Cluster 3 (100% versus 0% in both Cluster 1 and Cluster 2, chi-2 test, p < 0.001). In Cluster 3, the most common drug used was poppers (130 participants, 83.87%), followed by other drugs (23 participants, 14.84%). DMT and Triazolam were each used by 1 participant (0.65%). Apart from the engagement in SDU in their last sexual encounter, there were no statistically significant differences among the clusters in terms of age, marital status, education level, and personal monthly income (all p > 0.05). (Table 1 ) Difference in frequency of sexual acts across clusters The median number of days since last sexual activity in the study population ranged from 3.0 to 28.50 days. self-Masturbation had the shortest duration at 3.00 days, while rimming had the longest duration at 28.50 days. The estimated number of sexual acts over the past 12 months was 88.80 (95% CI 59.50-153.18) for self-masturbation, 43.24 (95% CI 41.00-50.35) for insertive anal sex, 38.16 (95% CI 31.19–48.23) for receptive anal sex, 49.38 (95% CI 52.07–73.72) for insertive oral sex, 46.97 (95% CI 51.44–59.77) for receptive oral sex, 34.90 (95% CI 25.03–45.18) for rimming, and 34.99 (95% CI 25.57–46.39) for being rimmed. (Table S1 -2) When comparing the clustering results of the three Clusters, Cluster 3 had a significantly lower median number of days since last sexual activity for sexual acts with a partner, compared to Cluster 1 and Cluster 2 (median days range: 5–10 versus 7–19 and 90–200, Wilcoxon, p < 0.001). Consistently, Cluster 3 had a significantly higher average number of sexual acts except for self-masturbation (average number of acts: 88.80 (95% CI 65.33, 147.79) versus 82.94 (95% CI 62.81, 119.90) and 118.52 (95% CI 70.32, 175.75), Wilcoxon, p < 0.001) over the past 12 months, compared to Cluster 1 and Cluster 2 (average acts range: 35.80–61.30 versus 31.30–56.10 and 4.37–21.22, Wilcoxon, p < 0.001). The difference in sexual behavior frequency between Cluster 1 and Cluster 2 was particularly notable, and neither group reported SDU (sexualized drug use) in their last sexual encounter. Cluster 1 engaged in a variety of sexual acts, including insertive and receptive oral and anal sex, with a moderate level of sexual activity, averaging 35.80 to 61.30 sexual acts over the past year. In contrast, Cluster 2 had much lower levels of sexual activity, with self-masturbation being the predominant sexual act. Sexual acts with partners were rare in Cluster 2, with an average of only 4.37 to 21.22 acts, indicating that sexual activity was primarily limited to self-masturbation, with minimal engagement with partners. Importantly, neither Cluster 1 nor Cluster 2 participants reported SDU in their last sexual encounter, in stark contrast to Cluster 3. The radar chart clearly displayed the differences in clustering results among the three groups. (Fig. 2 ) Difference in sexual history, PrEP Use, and STI testing across clusters When comparing the three Clusters, several key differences were observed. Notably, Cluster 1 and Cluster 2 members were not entirely free from SDU. SDU during sexual encounters over the past life was significantly more common in Cluster 3 (94.84% vs. 41.83% vs. 43.51%, chi-2 test, p < 0.0001). Cluster 3 had the significantly highest proportion of participants who had ever taken PrEP (34.90% vs. 17.02% vs. 8.00%, chi-2 test, p < 0.0001) and had the significantly highest willingness to take it in the future (64.95% vs. 55.98% vs. 47.10%, chi-2 test, p < 0.0001). Drug use during sexual encounters over the past life was significantly more common in this group (94.84% vs. 41.83% vs. 43.51%, chi-2 test, p < 0.0001). In the last six months, Cluster 3 also had significantly more casual male sex partners (4.55 ± 6.55 vs. 2.48 ± 3.21 vs. 2.74 ± 3.66, F = 12.49, p < 0.0001) and regular male sex partners (3.16 ± 4.37 vs. 2.27 ± 3.52 vs. 2.51 ± 2.53, F = 3.58, p = 0.028) and reported the significantly highest percentage of having sexual partners infected with an STI (8.39% vs. 3.79% vs. 3.90%, chi-2 test, p = 0.029). Additionally, Cluster 3 had the significantly highest frequency of participation in group sex (29.03% vs. 7.84% vs. 8.44%, chi-2 test, p < 0.0001) and the significantly lowest rate of always using condoms during anal sex with casual male partners (48.53% vs. 59.41% vs. 72.28%, chi-2 test, p < 0.0001). Despite the lack of statistical significance, Cluster 3 had a higher proportion of participants reporting a positive HIV test result (3.92% vs. 3.10% vs. 2.80%, chi-2 test, p = 0.2) and recent STI diagnoses other than HIV (9.03% vs. 5.85% vs. 5.84%, chi-2 test, p = 0.334) compared to the other clusters. (Table 2 ) Table 2 Sexual history, PrEP Use, and STI testing among study participants, stratified by Clusters Category Cluster 1 (N = 581) Cluster2 (N = 154) Cluster3 (N = 155) Chi 2 /F P-value Have you ever taken PrEP? 38.61 < 0.0001 Yes 96 (17.02%) 12 (8.00%) 52 (34.90%) No 468 (82.98%) 138 (92.00%) 97 (65.10%) If you haven't taken PrEP yet, would you want to? 43.75 < 0.0001 Yes 262 (55.98%) 65 (47.10%) 63 (64.95%) No 206 (44.02%) 73 (52.90%) 34 (35.05%) What was the result of your last HIV test? 8.6 0.2 Negative 514 (93.62%) 136 (95.10%) 145 (94.77%) Positive 17 (3.10%) 4 (2.80%) 6 (3.92%) Prefer not to say 18 (3.28%) 3 (2.10%) 2 (1.31%) Have you been diagnosed with an STD other than HIV in the past 6 months? 2.16 0.334 Yes 34 (5.85%) 9 (5.84%) 14 (9.03%) No 547 (94.15%) 145 (94.16%) 141 (90.97%) Have you had any sexual partners infected with an STD in the last 6 months? 10.79 0.029 Yes 22 (3.79) 6 (3.90%) 13 (8.39%) No 387 (66.61) 103 (66.88%) 85 (54.84%) Not sure 172 (29.6) 45 (29.22%) 57 (36.77%) Have you ever used drugs in your sex life? 142.24 < 0.0001 Yes 243 (41.83) 67 (43.51%) 147 (94.84%) No 338 (58.18%) 87 (56.49%) 8 (5.16%) The role assumed during anal sex with another man in the last six months 12.08 0.017 Insertive 260 (44.75%) 60 (38.96%) 53 (34.19%) Receptive 222 (38.21%) 74 (48.05%) 65 (41.94%) Both 99 (17.04%) 20 (12.99%) 37 (23.87%) The frequency of participation in group sex (with two or more people) in the last six months 37.98 < 0.0001 Never 502 (86.40%) 141 (91.56%) 109 (70.32%) Occasionally 75 (12.91%) 13 (8.44%) 40 (25.81%) Often 4 (0.69%) 0 (0) 5 (3.23%) Always 0 (0) 0 (0) 1 (0.65%) Frequency of condom use during anal sex with casual male partners in the last 6 months 28.75 < 0.0001 Never 39 (7.91%) 9 (7.83%) 12 (8.70%) Occasionally 98 (19.88%) 13 (11.30%) 33 (23.91%) Often 112 (22.72%) 19 (16.52%) 38 (27.54%) Always 244 (49.49%) 74 (64.35%) 55 (39.86%) Frequency of condom use during anal sex with casual male partners in the last 6 months 40.08 < 0.0001 Never 15 (3.71%) 3 (2.97%) 4 (2.94%) Occasionally 55 (13.61%) 7 (6.93%) 25 (18.38%) Often 94 (23.27%) 18 (17.82%) 41 (30.15%) Always 240 (59.41%) 73 (72.28%) 66 (48.53%) Number of casual male sex partners in the last 6 months 2.74 ± 3.66 2.48 ± 3.21 4.55 ± 6.55 12.49 < 0.0001 Number of regular male sex partners in the last 6 months 2.51 ± 2.53 2.27 ± 3.52 3.16 ± 4.37 3.58 0.028 Cluster 1 participants were more likely to have insertive anal sex (44.75% vs. 38.96% vs. 34.19%, chi-2 test, p = 0.017). Cluster 2 participants were more likely to have receptive anal sex (48.05% vs. 38.21% vs. 41.94%, chi-2 test, p = 0.017). Cluster 3 participants were more likely to assume both roles during anal sex in the last six months (23.87% vs. 17.04% vs. 12.99%, chi-2 test, p = 0.017) compared to other Clusters. (Table 2 ) Discussion Among 890 Chinese MSM, we identified three distinct clusters, finding that SDU in last sexual encounter was a significant criterion for differentiating sexual behavioral patterns. Cluster 3 all reported in SDU in last sexual encounter, while cluster 1 and cluster 2 did not. A key finding is that although cluster 1 and cluster 3 exhibited almost identical frequencies across various sexual acts—including oral sex, anal sex, and rimming—their risk profiles differed significantly. Cluster 3 exhibited significantly higher sexual risk behaviors, including having more casual and regular partners, lower consistent condom use, more frequent participation in group sex, and a higher proportion of partners with an STI in the past six months. The primary distinction between these two clusters was SDU in last sexual encounter in cluster 3, which seems to be the driving factor behind the elevated risk behaviors. In contrast, while cluster 1 reported similar sexual act frequencies, it demonstrated fewer risk behaviors, such as more consistent condom use and fewer sexual partner with STIs. On the other hand, cluster 2 primarily engaged in self-masturbation, had the least sexual act frequencies and exhibited the safest sexual practices. All participants in Cluster 3 reported drug use, with the majority using poppers (83.87%). This finding contrasts with studies from other countries, which indicate a higher prevalence of cannabis, GHB/GBL, ketamine, and mephedrone use[ 32 – 36 ]. The lower rate of these drugs observed in this cluster may be influenced by Chinese policies that have effectively curtailed the use of these drugs among MSM populations. Cluster 3 also showed the highest frequency of all types of sexual acts over 12 months. This contradicts a longitudinal study that showed a decline in all forms of anal sex among SDU participants over time[ 37 ]. This discrepancy may be because our study participants primarily used poppers, whereas the longitudinal study focused on MSM using more addictive drugs. Long-term use of such drugs may impair physical capacity, reducing the frequency of sexual acts. Our findings are consistent with a qualitative study where users reported that drug use enhanced their sexual experience and performance, which may psychologically increase the frequency of sexual encounters[ 38 ]. Additionally, 34.9% of participants in Cluster 3 reported had ever taken PrEP, which is slightly higher than the prevalence rates observed in other regions. Previous studies have indicated that MSM who had ever taken PrEP in China typically ranges from 19.5–24.7% [ 39 , 40 ]. However, as noted in earlier research, PrEP use may contribute to lower condom usage, potentially increasing the risk of other STIs [ 10 , 16 – 18 , 41 – 44 ]. Our findings in China mirror global trends where SDU is associated with higher-risk sexual behaviors in MSM populations. For example, studies in the U.S. and Europe have shown that MSM who use poppers, GHB/GBL, and methamphetamine engage in more casual partners, group sex, and lower condom use [ 1 , 7 , 45 – 48 ]. the strong association between SDU and high-risk sexual behavior is evident across different cultural and regulatory contexts, suggesting that public health strategies need to adapt to global trends while considering local policies and cultural norms. Therefore, interventions for this cluster should focus on harm reduction strategies, particularly in relation to SDU. Clear educational messaging is needed to emphasize reducing sexual risks by consistently using condoms, even while on PrEP, and promoting waiting until sober before engaging in sex to reduce impulsive risk-taking. Furthermore, efforts should be made to ensure easy access to condoms and lubricants in high-risk settings, such as MSM-friendly venues, clubs, and group sex parties. Event organizers in these environments should be encouraged to proactively provide safety supplies and promote safer sex practices. Although PrEP usage in this cluster is relatively high, it is crucial to reinforce the importance of combining PrEP with condom use to prevent STIs beyond HIV. Cluster 1 includes approximately 40% of members who had previously used SDU but not recently. This group demonstrates high sexual activity with moderate numbers of sexual partners (3–5), primarily engaging in one-on-one sexual encounters, with little group sex participation. Condom use is inconsistent, with some members reporting occasional use. For this cluster, public health interventions should focus on promoting consistent condom use by offering regular health education programs that emphasize the importance of using condoms with both regular and casual partners. Additionally, ensuring access to condoms, oral condoms, and lubricants through both online and offline distribution channels is crucial. Targeted education on risk communication in one-on-one relationships should be provided, helping this cluster understand that even in stable partnerships, maintaining proper safety measures is critical. Cluster 2 consists of members with more conservative sexual behaviors, fewer sexual partners, and a higher focus on self-masturbation. Although 40% had used SDU in the past, none reported recent use. Their sexual acts are infrequent, and they exhibit high safety awareness, consistently using condoms. This cluster is likely more discreet or closeted and may experience significant social isolation and psychological stress. Therefore, public health interventions for this clsuter should include mental health support, providing anonymous psychological counseling services to help them cope with issues related to sexual identity and societal pressures. Additionally, a mobile app could be developed to help them anonymously locate nearby sexual health clinics and services, enhancing access to essential health resources. Our method for estimating the frequency of sexual acts in MSM, including handling missing data, has been validated in previous study [ 29 ]. This study has used similar approaches to quantify the frequency of sexual acts in MSM populations. Our study revealed that self-masturbation was the most common sexual act among participants, occurring 88.80 times over 12 months, whereas being rimmed and rimming were the least frequent, with 34.90 times respectively. This finding is consistent with previous research, which suggested self-masturbation as the most prevalent and rimming as the least practiced sexual act [ 49 ]. Following self-masturbation, kissing also emerged as a commonly practice sexual act with 50.20 times over 12 months, and this finding is aligned with the widely perceived significance of kissing in intimate relationships [ 50 ]. Our study has several limitations. First, the questionnaire design did not distinguish between participants who refused to answer and those who genuinely did not engage in the behavior. As a result, we could not determine how many participants selected ‘no occurrence’ versus those who refused to respond. To address this, we used available data and log-normal distribution fittings to impute the missing values, providing a more robust estimate and reducing bias from missing data. Second, our study may misclassify habitual SDU users who did not report in SDU during their most recent sexual encounter; this was due to most participants not reporting frequency data despite having a history of SDU. Third, self-reporting could lead to recall or social desirability biases, challenging data authenticity. Fourth, the recruitment method used in this study primarily reached MSM who are actively engaged with sexual health services or community organizations, which may introduce potential selection bias. MSM who are more involved in health-focused activities or community initiatives may be overrepresented. This bias may limit the generalizability of the study’s findings to the broader MSM population. Further, the reliance on online recruitment through WeChat and community groups may exclude MSM who are less active on these platforms. Future research should focus on including MSM who are less connected to these networks to ensure a more comprehensive understanding of the population’s behaviors and risks. Fifth, our analysis used an unsupervised machine learning during clustering. To date, unsupervised models can be difficult to interpret due to the lack of predefined labels, making results highly dependent on the chosen algorithm and parameters, which introduces subjectivity. Additionally, clustering results are sensitive to the choice of algorithm (e.g., PAM vs. hierarchical clustering) and distance metrics (e.g., Gower’s distance), potentially leading to different cluster classifications. Noise or outliers in the dataset may skew results and affect the interpretation of behavioral patterns. Sixth, we acknowledge that t-SNE results are not directly interpretable as clustering outcomes. Therefore, we included an interpretation of Fig. 1 in the Results section to clarify how the t-SNE visualization corresponds to the identified clusters Conclusions Using cluster analysis, we identified three MSM subgroups with varying levels of high-risk sexual behavior. Cluster 3, involving MSM who engaged in SDU, showed the highest sexual risk and requires harm reduction strategies, including PrEP, condom promotion, and safer sex education. Cluster 1, while not reporting SDU, still exhibited higher-risk behaviors than Cluster 2 and would benefit from education on consistent condom use in relationships. Cluster 2, with the lowest risk, primarily engaged in self-masturbation and may need mental health support for social isolation. Future research should explore SDU’s long-term effects and targeted STI prevention strategies. Declarations Conflict of interest None declared. Ethical approval Human Research Ethics Committee of Zhuhai Centre for Disease Control and Prevention (CDC) (Ethics documents ID No. [2022] 11). Author Contributions Statement Bingyang She was responsible for data collection, data analysis, manuscript drafting, and revisions. Jiajun Sun, Fang Lu, Siqi Lin, Yi Liu, Gaixia Li, and Yawu Hu contributed to data collection. Weiming Tang, Rayner Tan, and Jason Ong reviewed the manuscript. Shu Su and Lei Zhang supervised the project and reviewed the manuscript. Funding LZ is supported by National Key R&D Program of China (2022YFC2505100, 2022YFC2505103); Outstanding Young Scholars Support Program (Grant number: 3111500001); Epidemiology modeling and risk assessment (Grant number: 20200344) and Xi’an Jiaotong University Young Scholar Support Grant (Grant number: YX6J004). SS is supported by the National Natural Science Foundation of China (82304246), Natural Science Foundation of Chongqing (CSTB2023NSCQ-MSX0198), Joint Medical Research Project of Chongqing Municipal Science and the Technology Bureau and Health Commission (2024QNXM057) and Emergency special project for COVID-19 (2023IITXG26). Acknowledgments LZ is supported by the National Key R&D Program of China (2022YFC2505100, 2022YFC2505103); Outstanding Young Scholars Support Program (Grant number: 3111500001); Epidemiology modeling and risk assessment (Grant number: 20200344) and Xi’an Jiaotong University Young Scholar Support Grant (Grant number: YX6J004). SS is supported by the National Natural Science Foundation of China (Grant No. 82304246), the Natural Science Foundation of Chongqing (Grant No. CSTB2023NSCQ-MSX0198), and the Chongqing Joint Health Sciences and Technology-Health Medical Research (Grant No. 2024QNXM056), Emergency special project for COVID-19 (2023IITXG26). Consent for Publication All authors have reviewed the manuscript and provided their consent for publication. Availability of Data and Materials The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request. Glossary MSM (Men Who Have Sex with Men) Refers to men who engage in sexual activity with other men, regardless of their sexual orientation. This term includes all men who have sex with men, whether they identify as gay, bisexual, or otherwise. SDU (Sexualized Drug Use) The use of drugs before or during sexual activity to enhance the experience. SDU is often associated with riskier sexual behaviors, such as unprotected sex or group sex, increasing the risk of HIV and sexually transmitted infections (STIs). PrEP (Pre-exposure Prophylaxis) A preventive treatment for HIV where people at high risk of infection take antiretroviral medication daily to reduce the risk of contracting HIV. STI (Sexually Transmitted Infections) Infections transmitted through sexual contact, such as gonorrhea, syphilis, chlamydia, and HIV. Mean Square Error (MSE ) A measure of the average squared difference between estimated values and actual values. It is used to evaluate the accuracy of the model used to estimate sexual behavior frequencies in the study. Gamma Hydroxybutyrate/Gamma Butyrolactone (GHB/GBL): GHB and GBL are substances often used recreationally, sometimes referred to as “club drugs.” GHB is a central nervous system depressant, commonly used in small doses for its euphoric and sedative effects. GBL is a prodrug, meaning it is converted into GHB once ingested. Both substances are frequently associated with increased sexual arousal and disinhibition, and are sometimes used during sexual encounters, which may heighten the risk of unsafe sexual practices. Overuse can lead to loss of consciousness, respiratory issues, or overdose. Partitioning Around Medoids (PAM): PAM is a clustering algorithm that identifies representative objects, or medoids, from a dataset. These medoids are used to group similar objects together, based on a chosen distance metric. Unlike k-means, PAM works well with non-Euclidean distances and is robust to outliers. In this study, Gower’s distance was used, which is suitable for mixed data types (continuous and categorical variables). References Maxwell, S., M. Shahmanesh, and M. Gafos, Chemsex behaviours among men who have sex with men: a systematic review of the literature. International Journal of Drug Policy, 2019. 63 : p. 74-89. Bourne, A., et al., The Chemsex study: drug use in sexual settings among gay and bisexual men in Lambeth, Southwark and Lewisham. 2014. Edmundson, C., et al., Sexualised drug use in the United Kingdom (UK): A review of the literature. International Journal of Drug Policy, 2018. 55 : p. 131-148. Íncera-Fernández, D., M. Gámez-Guadix, and S. Moreno-Guillén, Mental health symptoms associated with sexualized drug use (Chemsex) among men who have sex with men: a systematic review. International journal of environmental research and public health, 2021. 18 (24): p. 13299. Hawkinson, D.E., T.C. Witzel, and M. Gafos, Exploring practices to enhance benefits and reduce risks of chemsex among gay, bisexual, and other men who have sex with men: A meta-ethnography. International Journal of Drug Policy, 2024. 127 : p. 104398. Moreno-Gamez, L., D. Hernandez-Huerta, and G. Lahera, Chemsex and Psychosis: A Systematic Review. Behav Sci (Basel), 2022. 12 (12). Maxwell, S., M. Shahmanesh, and M. Gafos, Chemsex behaviours among men who have sex with men: A systematic review of the literature. Int J Drug Policy, 2019. 63 : p. 74-89. Aguilera-Mijares, S., et al., Variations in Sexual Behaviors by Use of Specific Substances Among Vancouver Gay, Bisexual, and Other Men Who Have Sex with Men: An Event-Level Analysis. Arch Sex Behav, 2021. 50 (7): p. 2875-2886. Wang, H., K.J. Jonas, and T.E. Guadamuz, Chemsex and chemsex associated substance use among men who have sex with men in Asia: A systematic review and meta-analysis. Drug Alcohol Depend, 2023. 243 : p. 109741. Tomkins, A., R. George, and M. Kliner, Sexualised drug taking among men who have sex with men: a systematic review. Perspect Public Health, 2019. 139 (1): p. 23-33. Prevention, C.f.D.C.a. Men Who Have Sex with Men (MSM) . 2021; Available from: https://www.cdc.gov/std/treatment-guidelines/msm.htm. Nevendorff, L., et al., Prevalence of sexualized drug use and risk of HIV among sexually active MSM in East and South Asian countries: systematic review and meta-analysis. J Int AIDS Soc, 2023. 26 (1): p. e26054. Rajasingham, R., et al., A systematic review of behavioral and treatment outcome studies among HIV-infected men who have sex with men who abuse crystal methamphetamine. AIDS patient care and STDs, 2012. 26 (1): p. 36-52. Plankey, M.W., et al., The relationship between methamphetamine and popper use and risk of HIV seroconversion in the multicenter AIDS cohort study. JAIDS Journal of Acquired Immune Deficiency Syndromes, 2007. 45 (1): p. 85-92. Semple, S.J., T.L. Patterson, and I. Grant, Motivations associated with methamphetamine use among HIV men who have sex with men. Journal of substance abuse Treatment, 2002. 22 (3): p. 149-156. Xu, J.-J., et al., Recreational drug use among Chinese men who have sex with men: a risky combination with unprotected sex for acquiring HIV infection. BioMed research international, 2014. 2014 . Chen, X., et al., Club drugs and HIV/STD infection: an exploratory analysis among men who have sex with men in Changsha, China. PloS one, 2015. 10 (5): p. e0126320. Mao, X., et al., Use of multiple recreational drugs is associated with new HIV infections among men who have sex with men in China: a multicenter cross-sectional survey. BMC Public Health, 2021. 21 (1): p. 354. Bourne, A., et al., Illicit drug use in sexual settings (‘chemsex’) and HIV/STI transmission risk behaviour among gay men in South London: findings from a qualitative study. Sexually transmitted infections, 2015. 91 (8): p. 564-568. Curtis, T.J., et al., Patterns of sexualised recreational drug use and its association with risk behaviours and sexual health outcomes in men who have sex with men in London, UK: a comparison of cross-sectional studies conducted in 2013 and 2016. Sexually transmitted infections, 2020. 96 (3): p. 197-203. Guerras, J.-M., et al., Association of sexualized drug use patterns with HIV/STI transmission risk in an internet sample of men who have sex with men from seven European countries. Archives of Sexual Behavior, 2021. 50 : p. 461-477. Heath, J., A. Lanoye, and S.A. Maisto, The role of alcohol and substance use in risky sexual behavior among older men who have sex with men: a review and critique of the current literature. AIDS and Behavior, 2012. 16 : p. 578-589. Folch, C., et al., High prevalence of drug consumption and sexual risk behaviors in men who have sex with men. Medicina Clínica (English Edition), 2015. 145 (3): p. 102-107. Carey, J.W., et al., Drug use, high-risk sex behaviors, and increased risk for recent HIV infection among men who have sex with men in Chicago and Los Angeles. AIDS and Behavior, 2009. 13 : p. 1084-1096. Salazar-Vizcaya, L., et al., ua Clusters of sexual behaviour in HIV-positive men who have sex with men reveal highly dissimilar time trends. Clin Infect Dis Off Publ Infect Dis Soc Am, 2019. 15 . Salazar-Vizcaya, L., et al., Clusters of Sexual Behavior in Human Immunodeficiency Virus-positive Men Who Have Sex With Men Reveal Highly Dissimilar Time Trends. Clin Infect Dis, 2020. 70 (3): p. 416-424. Dishion, T.J., T. Ha, and M.-H. Véronneau, An ecological analysis of the effects of deviant peer clustering on sexual promiscuity, problem behavior, and childbearing from early adolescence to adulthood: an enhancement of the life history framework. Developmental psychology, 2012. 48 (3): p. 703. Blondeel, K., et al., Sexual behaviour patterns and STI risk: results of a cluster analysis among men who have sex with men in Portugal. BMJ open, 2021. 11 (1): p. e033290. Hummel, M., D. Edelmann, and A. Kopp-Schneider, Clustering of samples and variables with mixed-type data. PLoS One, 2017. 12 (11): p. e0188274. Kault, D., The Shape of the Distribution of the Number of Sexual Partners. Statistics in Medicine, 1996. 15 (2): p. 221-230. Zhang, L., E.P. Fung Chow, and D.P. Wilson, Men who have sex with men in China have relatively low numbers of sexual partners. Infect Dis Rep, 2011. 3 (1): p. e10. Brogan, N., et al., Sexually transmitted infections in MSM: Canadian results from the European men-who-have-sex-with-men internet survey (EMIS-2017). Canada Communicable Disease Report, 2019. 45 (11): p. 271. Dolengevich-Segal, H., et al., Drug-related and psychopathological symptoms in HIV-positive men who have sex with men who inject drugs during sex (slamsex): Data from the U-SEX GESIDA 9416 Study. PLoS One, 2019. 14 (12): p. e0220272. Schecke, H., et al., Crystal methamphetamine use in sexual settings among German men who have sex with men. Frontiers in Psychiatry, 2019. 10 : p. 886. Nöstlinger, C., et al., Drug use, depression and sexual risk behaviour: a syndemic among early pre-exposure prophylaxis (PrEP) adopters in Belgium? AIDS care, 2020. 32 (sup2): p. 57-64. Vaccher, S.J., et al., Prevalence, frequency, and motivations for alkyl nitrite use among gay, bisexual and other men who have sex with men in Australia. International Journal of Drug Policy, 2020. 76 : p. 102659. Sewell, J., et al., Changes in chemsex and sexual behaviour over time, among a cohort of MSM in London and Brighton: findings from the AURAH2 study. International Journal of Drug Policy, 2019. 68 : p. 54-61. Marques Oliveira, P., C. Sousa Reis, and M.A. Vieira-Coelho, Getting Inside the Mind of Gay and Bisexual Men Who Have Sex with Men with Sexualized Drug Use–A Systematic Review. International Journal of Sexual Health, 2023. 35 (4): p. 573-595. Huang, W., et al., Prepared for PrEP: preferences for HIV pre-exposure prophylaxis among Chinese men who have sex with men in an online national survey. BMC Infectious Diseases, 2019. 19 (1): p. 1057. Zhang, G., et al., Pre-exposure prophylaxis uptake for high-risk men who have sex with men in China: a multi-city cross-sectional survey. AIDS Research and Therapy, 2023. 20 (1): p. 32. Wei, C., et al., Patterns and levels of illicit drug use among men who have sex with men in Asia. Drug Alcohol Depend, 2012. 120 (1-3): p. 246-9. Tomkins, A., R. George, and M. Kliner, Sexualised drug taking among men who have sex with men: a systematic review. Perspectives in public health, 2019. 139 (1): p. 23-33. He, L., et al., New types of drug use and risks of drug use among men who have sex with men: a cross-sectional study in Hangzhou, China. BMC Infect Dis, 2018. 18 (1): p. 182. Shusen Liu, M., PhD and Roger Detels, MD, MS*, Recreational drug use: an emerging concern among venuebased male sex workers in China , C.C.f.D.C.a.P. National Centre for AIDS/STD Prevention and Control, Beijing China, Editor. 2012: Sex Transm Dis. Blomquist, P., et al., P531 Chemsex and STI clinic use among MSM: results from a large online survey in england . 2019, BMJ Publishing Group Ltd. Glynn, R.W., et al., Chemsex, risk behaviours and sexually transmitted infections among men who have sex with men in Dublin, Ireland. International Journal of Drug Policy, 2018. 52 : p. 9-15. Aguilera-Mijares, S., et al., Variations in Sexual Behaviors by Use of Specific Substances Among Vancouver Gay, Bisexual, and Other Men Who Have Sex with Men: An Event-Level Analysis. Archives of Sexual Behavior, 2021. 50 (7): p. 2875-2886. Brown, R.E., et al., Partner-level substance use associated with increased sexual risk behaviors among men who have sex with men in San Francisco, CA. Drug and Alcohol Dependence, 2017. 176 : p. 176-180. Richters, J., et al., Masturbation, paying for sex, and other sexual activities: the Second Australian Study of Health and Relationships. Sexual health, 2014. 11 (5): p. 461-471. Cornelisse, V.J., et al., The frequency of kissing as part of sexual activity differs depending on how men meet their male casual sexual partners. International journal of STD & AIDS, 2018. 29 (6): p. 598-602. Additional Declarations No competing interests reported. Supplementary Files Appendixcodebook.27May2022.docx appendix1.docx Cite Share Download PDF Status: Published Journal Publication published 10 Apr, 2025 Read the published version in BMC Public Health → Version 1 posted Editorial decision: Revision requested 24 Oct, 2024 Editor assigned by journal 24 Oct, 2024 Submission checks completed at journal 22 Oct, 2024 First submitted to journal 17 Oct, 2024 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-5286116","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":370000354,"identity":"455ec91e-403f-4904-8a13-96859dfae3d3","order_by":0,"name":"Bingyang She","email":"","orcid":"","institution":"Xi'an Jiaotong University","correspondingAuthor":false,"prefix":"","firstName":"Bingyang","middleName":"","lastName":"She","suffix":""},{"id":370000355,"identity":"c3935131-5e2a-4d2f-9669-4ee79784dcb4","order_by":1,"name":"Jiajun Sun","email":"","orcid":"","institution":"Monash University","correspondingAuthor":false,"prefix":"","firstName":"Jiajun","middleName":"","lastName":"Sun","suffix":""},{"id":370000356,"identity":"9d4273f9-3fab-4729-9610-bbfefd07825a","order_by":2,"name":"Fang Lu","email":"","orcid":"","institution":"Xi'an Jiaotong University","correspondingAuthor":false,"prefix":"","firstName":"Fang","middleName":"","lastName":"Lu","suffix":""},{"id":370000357,"identity":"d114bd50-6ef3-4f6b-94af-cc769e5cea2d","order_by":3,"name":"Siqi Lin","email":"","orcid":"","institution":"Xi'an Jiaotong University","correspondingAuthor":false,"prefix":"","firstName":"Siqi","middleName":"","lastName":"Lin","suffix":""},{"id":370000358,"identity":"ec95dcd1-f5a2-4182-8767-aa334337714e","order_by":4,"name":"Yi Liu","email":"","orcid":"","institution":"Xi'an Jiaotong University","correspondingAuthor":false,"prefix":"","firstName":"Yi","middleName":"","lastName":"Liu","suffix":""},{"id":370000359,"identity":"ce951624-3be6-4421-8f24-6150b42420f6","order_by":5,"name":"Gaixia Li","email":"","orcid":"","institution":"Xi'an Jiaotong University","correspondingAuthor":false,"prefix":"","firstName":"Gaixia","middleName":"","lastName":"Li","suffix":""},{"id":370000360,"identity":"7307e44a-c90d-45b7-9655-b7b4fdec2425","order_by":6,"name":"Yawu Hu","email":"","orcid":"","institution":"Xi'an Jiaotong University","correspondingAuthor":false,"prefix":"","firstName":"Yawu","middleName":"","lastName":"Hu","suffix":""},{"id":370000361,"identity":"5bec0af7-b770-4bb3-ba98-cd422d9939d8","order_by":7,"name":"Weiming Tang","email":"","orcid":"","institution":"University of North Carolina at Chapel Hill","correspondingAuthor":false,"prefix":"","firstName":"Weiming","middleName":"","lastName":"Tang","suffix":""},{"id":370000362,"identity":"be6e998a-034b-4dd4-a974-c50652121d30","order_by":8,"name":"Rayner Tan","email":"","orcid":"","institution":"National University of Singapore","correspondingAuthor":false,"prefix":"","firstName":"Rayner","middleName":"","lastName":"Tan","suffix":""},{"id":370000363,"identity":"88d712a3-937b-4b4c-9f41-31060df746c3","order_by":9,"name":"Jason Ong","email":"","orcid":"","institution":"Monash University","correspondingAuthor":false,"prefix":"","firstName":"Jason","middleName":"","lastName":"Ong","suffix":""},{"id":370000364,"identity":"2c77c7a5-5cd4-4760-bbce-376a6f722141","order_by":10,"name":"Shu Su","email":"","orcid":"","institution":"Second Affiliated Hospital of Chongqing Medical University","correspondingAuthor":false,"prefix":"","firstName":"Shu","middleName":"","lastName":"Su","suffix":""},{"id":370000365,"identity":"22bb8bca-971c-4796-ab1c-be1967937836","order_by":11,"name":"Lei Zhang","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA/0lEQVRIie3RsWoCMRjA8S8EvCVy6yeF3itEnEpFB1/kglAXEaFjh6YIcVF8BF+hRxfHOzK43AN0PCk4OTiJtzV3FsfEUWj+Qwgf+RFIAHy+O4wDkbSUZkdpCljPYjeBVkWCRgwY30RMNQkZ/zvtIsFMQnvTm0QDdsKns4YwGHMoNxbCMgkiH762dfOLY6yhtThwsswtBIWETFGRzJpJURH+PeaUKAuJdpJ8qHdDWJFWpO8kaB6ZKC3WlJHLLegiTEhN1FZ80kaH48uIYb6fZksL6c+3u59SvYn1Su8fsPv8GM6HSVFaSFVaX2dWan6fXSfOIglAjjcd9fl8vv/WL9b/Tv40uNkAAAAAAElFTkSuQmCC","orcid":"","institution":"Alfred Health","correspondingAuthor":true,"prefix":"","firstName":"Lei","middleName":"","lastName":"Zhang","suffix":""}],"badges":[],"createdAt":"2024-10-18 03:38:12","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-5286116/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-5286116/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1186/s12889-025-22388-x","type":"published","date":"2025-04-10T16:04:54+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":68885313,"identity":"f1fc4b6a-5fa9-4202-b090-b13d8435eb6d","added_by":"auto","created_at":"2024-11-13 06:36:53","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":192257,"visible":true,"origin":"","legend":"\u003cp\u003eReduced dimensional scatter plot of three clustering results\u003c/p\u003e","description":"","filename":"floatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-5286116/v1/92bc9f58784d6e91c8bbebb3.png"},{"id":68884794,"identity":"a12d4ba4-9dc7-42db-bf0c-75c8e1c6e4ba","added_by":"auto","created_at":"2024-11-13 06:28:53","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":421166,"visible":true,"origin":"","legend":"\u003cp\u003eRadar plot demonstrating the number of days since the most recent sexual encounter and the estimated frequency of sexual acts over the past twelve months\u003c/p\u003e","description":"","filename":"floatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-5286116/v1/a5326a710e52bba48d26cf97.png"},{"id":80558676,"identity":"32b288a6-ac5b-4747-b1ea-3d81d81decb6","added_by":"auto","created_at":"2025-04-14 16:16:01","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":2085144,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-5286116/v1/d63da2a8-656e-4348-83cf-f567ccf64694.pdf"},{"id":68884796,"identity":"734b7345-1637-45c8-8f74-cb36076e6b52","added_by":"auto","created_at":"2024-11-13 06:28:53","extension":"docx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":58867,"visible":true,"origin":"","legend":"","description":"","filename":"Appendixcodebook.27May2022.docx","url":"https://assets-eu.researchsquare.com/files/rs-5286116/v1/b5d092135e730a076f615951.docx"},{"id":68884797,"identity":"95b76e43-522c-48a6-8f79-1029bd85de55","added_by":"auto","created_at":"2024-11-13 06:28:53","extension":"docx","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":325204,"visible":true,"origin":"","legend":"","description":"","filename":"appendix1.docx","url":"https://assets-eu.researchsquare.com/files/rs-5286116/v1/e5398a71cd73ea9fda1e23e5.docx"}],"financialInterests":"No competing interests reported.","formattedTitle":"Identifying subgroups of Chinese men who have sex with men based on sexual behavior and drug use patterns using a clustering analysis approach","fulltext":[{"header":"Introduction","content":"\u003cp\u003eSexualized drug use (SDU), which includes chemsex, is defined as the intentional engagement in substance use before or during sexual contact to enhance the experience and has emerged as a worldwide concern. The substances involved in SDU often include mephedrone, crystal methamphetamine, Gamma hydroxybutyrate/Gamma butyrolactone (GHB/GBL), cocaine, and ketamine[\u003cspan additionalcitationids=\"CR2 CR3 CR4\" citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e]. Increasingly, the interplay between substance use and high-risk sexual behavior has led to public health challenges, particularly among men who have sex with men (MSM). SDU, alongside inherently other high-risk sexual practices, elevates the risk of sexually transmitted infections (STIs) in MSM [\u003cspan additionalcitationids=\"CR7\" citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e]. The prevalence of SDU in MSM varies significantly across different countries and regions, ranging from 34\u0026ndash;79% in developed nations such as the United States, United Kingdom, and Germany, whereas in China, prevalence ranges between 18\u0026ndash;28% [\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e, \u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eSDU can profoundly affect cognitive functions and increase risk-taking sexual activity. Under the influence of drugs, MSM may engage in prolonged sexual sessions, potentially resulting in physical injuries or traumas [\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e]. Moreover, high-risk sexual behavior under the influence of SDU, like condomless sexual behavior, increases the spread of HIV and STIs [\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e]. Further, drug-enhanced sense of intimacy can lead individuals to engage in various sexual acts during a single encounter and facilitate group sex, increasing the risk of HIV/STI exposure [\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e, \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e, \u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eTo date, only limited research has been conducted to understand the pattern and frequency of sexual acts in MSM who practice SDU. Only a few studies have mentioned a potential association between the frequency of rimming and the use of poppers, methamphetamine, GHB, and ecstasy/MDMA [\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e, \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e, \u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e]. Noticeable knowledge gap remains in the understanding of the relationship between sexual behavioral patterns and SDU in MSM [\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e, \u003cspan additionalcitationids=\"CR17 CR18 CR19 CR20 CR21 CR22 CR23\" citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eConventional analytical approaches face challenges in identifying the subtle differences across various subpopulations in MSM due to the preference and complex nature of their sexual behaviors. To overcome this, we used cluster analysis, a machine learning technique, to simplify the data and uncover underlying patterns. Cluster analysis is widely used in behavioral studies to identify subpopulations [\u003cspan additionalcitationids=\"CR26 CR27\" citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e], making it an appropriate approach for analyzing SDU and related sexual behaviors in MSM. This study aims to use cluster analysis to identify MSM subpopulations according to their SDU behaviors and compare the differences in their sexual behavioral pattern across identified clusters. This approach may provide evidence for targeted intervention of high-risk sexual behavior and drug use during sexual encounters in MSM.\u003c/p\u003e"},{"header":"Methods","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e\n \u003ch2\u003eStudy design and participants recruited\u003c/h2\u003e\n \u003cp\u003eOur study was an anonymous cross-sectional survey across China from March 23 to April 22, 2022, through community organizations and social groups. We recruited MSM nationally through advertisements on the WeChat platforms of six major gay service groups including Zhuhai Xutong Volunteer Service Center, Qingdao Qingtong Anti-AIDS Volunteer Service Center, Nanjing Xingyou Volunteer Service Center, Jinan Rainbow, Beijing LGBT\u0026thinsp;+\u0026thinsp;and Positive Peers Group, which are dedicated to HIV and STI education and intervention. The recruitment process provided detailed study information which included a general description of the project and addresses of participating health service stations and CDCs and offered cash incentives for participation. Eligible participants who consented to participate were directed to complete a detailed questionnaire on sexual behavior, substance use and recent sexual activities with men at Weng Juan Xing (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ewww.wjx.cn\u003c/span\u003e\u003c/span\u003e). To be eligible for participation in the questionnaire, individuals had to meet three criteria: 1) biologically assigned as male at birth; 2) have a history of male-to-male anal sex; and 3) be between the ages of 18 and 70.\u003c/p\u003e\n\u003c/div\u003e\n\u003ch3\u003eData collection\u003c/h3\u003e\n\u003cp\u003eAt the beginning of our study, participants were asked to provide information about their sociodemographic characteristics, including gender, age, income, marital status, educational qualifications, and self-perceived gender identity. Regarding their sexual activity, the questionnaire surveyed the number of days since their last sexual acts and the sequence of their most recent sexual encounter. A significant part of the questionnaire focused on understanding drug use during their most recent sexual encounter. Participants were asked the question, \u0026apos;Did you use drugs during the last time you had sex with a man?\u0026apos; and were given a list of substances to choose from, including Rush, Crystal Meth, and Ketamine, etc.\u003c/p\u003e\n\u003cp\u003eWe collected information on the frequency of sexual encounters with both regular and casual male partners and also the frequency of condom use over the past six months. We also collected the number of days since the most recent sexual act, which ranged from kissing to more intimate acts like insertive and receptive oral sex, anal sex, rimming, and masturbation (self, for a partner, by a partner).\u003c/p\u003e\n\u003ch3\u003eDefinition of SDU\u003c/h3\u003e\n\u003cp\u003eWe defined SDU as the behavior where participants use substances during their sexual encounter. These substances include, but are not limited to, Poppers, Ketamine, Ecstasy (MDMA), and GHB/GBL.\u003c/p\u003e\n\u003ch3\u003eDefinition of Regular and Casual Sexual Partnerships\u003c/h3\u003e\n\u003cp\u003eRegular sexual partners were defined as individuals with whom a committed and lasting romantic or sexual relationship was maintained, including boyfriends and other male partners. Casual sexual partners were defined as individuals with whom sexual relationships had been established for a duration of three months or less. This category included male sex workers.\u003c/p\u003e\n\u003ch3\u003eClustering analysis\u003c/h3\u003e\n\u003cp\u003eWe selected 11 sexual behavioral indicators for clustering analysis, including the number of days since the participant\u0026rsquo;s last sexual encounter involving sexual acts such as kissing, insertive or receptive oral sex, insertive or receptive anal sex, rimming, masturbation (self, with a partner, or by a partner), and drug use during the last sexual encounter (SDU). Given the combination of continuous variables (e.g., the number of days since the participant\u0026rsquo;s last sexual acts) and binary categorical variables (e.g., SDU), we employed the Partitioning Around Medoids (PAM) algorithm using Gower\u0026rsquo;s distance to handle these mixed data types and the silhouette coefficient to determine the optimal number of clusters. Multidimensional scaling (MDS) was used to standardize both categorical and continuous variables [\u003cspan class=\"CitationRef\"\u003e29\u003c/span\u003e]. PAM clustering analysis separated the population into two groups: those with SDU and those without. Heterogeneity within the non-SDU group prompted us to investigate further potential subgroups in this group using a hierarchical clustering approach based on sexual behavior frequency. The combined PAM and hierarchical clustering enabled us to account for both participants\u0026rsquo; drug use and sexual behavioral patterns at two different levels. We used t-SNE for dimensionality reduction and visualization of the clustering structure in two dimensions. The visualization helped confirm the existence of clusters and sub-clusters.\u003c/p\u003e\n\u003cdiv id=\"Sec8\" class=\"Section2\"\u003e\n \u003ch2\u003eEstimating the number of sexual acts over the past 12 months\u003c/h2\u003e\n \u003cp\u003eWe employed a log-normal distribution fitting method to estimate the frequency of various sexual behaviors, as the log-normal distribution effectively simulates the distribution of complex human behaviors such as sexual acts. We have used such a fitting in a previous study [\u003cspan class=\"CitationRef\"\u003e29\u003c/span\u003e]. Based on the log-normal distribution fitting, we estimated the number of different sexual acts based on the number of days since the last sexual acts that were reported [\u003cspan class=\"CitationRef\"\u003e30\u003c/span\u003e, \u003cspan class=\"CitationRef\"\u003e31\u003c/span\u003e]. We then calibrated the simulated frequency to the observed accumulated frequency of sexual acts in the participants. This calibration enabled us to estimate the distribution of the number of sexual acts by their sexual act types even in the absence of empirical data. The calibration was considered successful when the difference (defined as Mean Square Error (MSE)) between the simulated and observed data was minimized:\u003c/p\u003e\n \u003cdiv id=\"Equa\" class=\"Equation\"\u003e\n \u003cdiv class=\"mathdisplay\" id=\"FileID_Equa\" name=\"EquationSource\"\u003e$$\\:MSE=\\frac{1}{n}\\sum\\:_{i=1}^{n}{\\left({\\stackrel{\\prime }{y}}_{i}-{y}_{i}\\right)}^{2}$$\u003c/div\u003e\n \u003c/div\u003e\n \u003cp\u003e\u003cbr\u003e\u003c/p\u003e\n \u003cp\u003eWhere, \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{\\stackrel{\\prime }{y}}_{i}\u0026sim;LN\\)\u003c/span\u003e\u003c/span\u003e(\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{\\mu\\:}_{i,}{\\delta\\:}_{i}\\)\u003c/span\u003e\u003c/span\u003e) represents the log-normal distribution curve of the number of days of the last sex behaviors on type \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:i\\)\u003c/span\u003e\u003c/span\u003e, mean \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{\\mu\\:}_{i}\\)\u003c/span\u003e\u003c/span\u003e and variance \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{\\delta\\:}_{i}\\)\u003c/span\u003e\u003c/span\u003e, and \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{\\stackrel{\\prime }{y}}_{i}\\)\u003c/span\u003e\u003c/span\u003e represents a randomly generated bootstrap datasets corresponding to the original datasets of sexual acts. The Mean Square Error (MSE) was used to compute the mean value of the sum of the squares of errors between the estimated data point and the bootstrapped data point. Bootstrapped data was generated by resampling the original data to create a dataset of similar dimensions as the fitted data. We repeated the estimated process 400 times and selected the curve with the smallest MSE value as the optimized curve. The best 50 fittings were chosen to establish a 95% confidence interval. After estimating the curves, we obtained their log-probability density function values. These probabilities were then converted into weights for days 1 to 365. The weights were used to estimate the number of sexual acts over the last 12 months. This allowed us to calculate the weighted mean number of sexual acts over last 12 months. The entire optimization and fitting data process was conducted using MATLAB R2020a.\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec9\" class=\"Section2\"\u003e\n \u003ch2\u003eStatistical analysis\u003c/h2\u003e\n \u003cp\u003eCategorical variables in this study were described using numbers and percentages, and the chi-square test was used to compare differences between groups. For normally distributed continuous variables, mean\u0026thinsp;\u0026plusmn;\u0026thinsp;standard deviation was used to present the data, and ANOVA (analysis of variance) was performed for comparison. Non-normally distributed continuous variables were reported as median and interquartile range, and differences between groups were assessed using the Wilcoxon rank-sum test.\u003c/p\u003e\n\u003c/div\u003e"},{"header":"Result","content":"\u003cdiv id=\"Sec11\" class=\"Section2\"\u003e \u003ch2\u003eClustering participants by sexual and drug-use behaviors\u003c/h2\u003e \u003cp\u003eOur study initially surveyed 1,034 individuals, excluding 144 due to lack of consent (n\u0026thinsp;=\u0026thinsp;100), age discrepancies (n\u0026thinsp;=\u0026thinsp;42), or duplicate responses (n\u0026thinsp;=\u0026thinsp;2). Ultimately, 890 participants were included in the analysis (Figure \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e). The first step of the clustering analysis categorized participants into two clusters: those whose most recent sexual encounter reported SDU (n\u0026thinsp;=\u0026thinsp;155, cluster 3) and those who did not report SDU (n\u0026thinsp;=\u0026thinsp;735) (Figure \u003cspan refid=\"MOESM2\" class=\"InternalRef\"\u003eS2\u003c/span\u003e). Subsequently, the second step of the clustering analysis further divided the non-SDU in their last sexual encounter cluster into two groups, resulting in a total of three distinct clusters (Cluster 1, n\u0026thinsp;=\u0026thinsp;581; Cluster 2, n\u0026thinsp;=\u0026thinsp;154; Cluster 3, n\u0026thinsp;=\u0026thinsp;155) (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e, S3).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec12\" class=\"Section2\"\u003e \u003ch2\u003eDifferences in demographic characteristics across clusters\u003c/h2\u003e \u003cp\u003eThe majority of the study participants were single (88.99%), predominantly held a college or bachelor\u0026rsquo;s degree (73.37%) and had an income range of RMB 5001\u0026ndash;8000 (26.07%). The average age of participants was 28.14\u0026thinsp;\u0026plusmn;\u0026thinsp;7.37 years. About 17.42% of the participants reported engagement in SDU in their last sexual encounter. (Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e)\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eDemographic and behavioral characteristics of study participants, stratified by Clusters\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"7\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCategory\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eOverall (N\u0026thinsp;=\u0026thinsp;890)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eCluster1 (N\u0026thinsp;=\u0026thinsp;581)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eCluster2 (N\u0026thinsp;=\u0026thinsp;154)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eCluster3 (N\u0026thinsp;=\u0026thinsp;155)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003eChi\u003csup\u003e2\u003c/sup\u003e/F\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c7\"\u003e \u003cp\u003eP-value\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAge (years)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e28.14\u0026thinsp;\u0026plusmn;\u0026thinsp;7.37\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e28.15\u0026thinsp;\u0026plusmn;\u0026thinsp;7.22\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e27.85\u0026thinsp;\u0026plusmn;\u0026thinsp;8.06\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e28.40\u0026thinsp;\u0026plusmn;\u0026thinsp;7.27\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.215\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e0.806\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"5\" nameend=\"c5\" namest=\"c1\"\u003e \u003cp\u003e\u003cb\u003eMarital Status\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e6.591\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e0.159\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSingle\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e792 (88.99%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e525 (90.36%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e132 (85.71%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e135 (87.1%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eEngaged or Married\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e54 (6.07%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e30 (5.16%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e10 (6.49%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e14 (9.03%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSeparated or Divorced or Widowed\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e44 (4.94%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e26 (4.48%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e12 (7.79%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e6 (3.87%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"5\" nameend=\"c5\" namest=\"c1\"\u003e \u003cp\u003e\u003cb\u003eEducation level\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e8.19\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e0.085\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHigh school or below\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e120 (13.48%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e89 (15.32%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e19 (12.34%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e12 (7.74%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCollege/Bachelors\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e653 (73.37%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e421 (72.46%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e116 (75.32%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e116 (74.84%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMasters or above\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e117 (13.15%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e71 (12.22%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e19 (12.34%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e27 (17.42%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"5\" nameend=\"c5\" namest=\"c1\"\u003e \u003cp\u003e\u003cb\u003ePersonal monthly income (RMB)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e4.708\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e0.788\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;1500\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e106 (11.91%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e62 (10.67%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e23 (14.94%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e21 (13.55%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e1500\u0026ndash;3000\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e112 (12.58%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e73 (12.56%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e21 (13.64%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e18 (11.61%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e3001\u0026ndash;5000\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e225 (25.28%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e152 (26.16%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e39 (25.32%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e34 (21.94%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e5001\u0026ndash;8000\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e232 (26.07%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e149 (25.65%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e38 (24.68%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e45 (29.03%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u0026gt;\u0026thinsp;8001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e215 (24.16%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e145 (24.96%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e33 (21.43%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e37 (23.87%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"4\" nameend=\"c4\" namest=\"c1\"\u003e \u003cp\u003e\u003cb\u003eUse drugs during last sexual encounter\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eYes\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e155 (17.42%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0 (0%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0 (0%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e155 (100%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e890\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e\u0026lt;0.0001*\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNo\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e735 (82.58%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e581 (100%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e154 (100%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0 (0%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eDrug Types\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePoppers\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e130 (83.87%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0 (0%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0 (0%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e130 (83.87%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e890\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e\u0026lt;0.0001*\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eDMT\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1 (0.65%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0 (0%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0 (0%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e1 (0.65%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTriazolam\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1 (0.65%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0 (0%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0 (0%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e1 (0.65%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eOthers\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e23 (14.84%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0 (0%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0 (0%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e23 (14.84%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCrystal Meth\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0 (0%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0 (0%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0 (0%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0 (0%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eKetamine\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0 (0%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0 (0%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0 (0%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0 (0%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eEcstasy\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0 (0%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0 (0%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0 (0%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0 (0%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eYa ba pills\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0 (0%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0 (0%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0 (0%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0 (0%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMixing drugs\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0 (0%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0 (0%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0 (0%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0 (0%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eGHB\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0 (0%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0 (0%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0 (0%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0 (0%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMonkey dust\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0 (0%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0 (0%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0 (0%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0 (0%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003ctfoot\u003e \u003ctr\u003e\u003ctd colspan=\"7\"\u003eNote: DMT: N, N-Dimethyltryptamine; GHB: Gamma-Hydroxybutyrate.\u003c/td\u003e\u003c/tr\u003e \u003c/tfoot\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eWhile comparing the demographic characteristics of the participants, we observed that all participants who reported SDU in their last sexual encounter were exclusively found in Cluster 3 (100% versus 0% in both Cluster 1 and Cluster 2, chi-2 test, p\u0026thinsp;\u0026lt;\u0026thinsp;0.001). In Cluster 3, the most common drug used was poppers (130 participants, 83.87%), followed by other drugs (23 participants, 14.84%). DMT and Triazolam were each used by 1 participant (0.65%). Apart from the engagement in SDU in their last sexual encounter, there were no statistically significant differences among the clusters in terms of age, marital status, education level, and personal monthly income (all p\u0026thinsp;\u0026gt;\u0026thinsp;0.05). (Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e)\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec13\" class=\"Section2\"\u003e \u003ch2\u003eDifference in frequency of sexual acts across clusters\u003c/h2\u003e \u003cp\u003eThe median number of days since last sexual activity in the study population ranged from 3.0 to 28.50 days. self-Masturbation had the shortest duration at 3.00 days, while rimming had the longest duration at 28.50 days. The estimated number of sexual acts over the past 12 months was 88.80 (95% CI 59.50-153.18) for self-masturbation, 43.24 (95% CI 41.00-50.35) for insertive anal sex, 38.16 (95% CI 31.19\u0026ndash;48.23) for receptive anal sex, 49.38 (95% CI 52.07\u0026ndash;73.72) for insertive oral sex, 46.97 (95% CI 51.44\u0026ndash;59.77) for receptive oral sex, 34.90 (95% CI 25.03\u0026ndash;45.18) for rimming, and 34.99 (95% CI 25.57\u0026ndash;46.39) for being rimmed. (Table \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e-2)\u003c/p\u003e \u003cp\u003eWhen comparing the clustering results of the three Clusters, Cluster 3 had a significantly lower median number of days since last sexual activity for sexual acts with a partner, compared to Cluster 1 and Cluster 2 (median days range: 5\u0026ndash;10 versus 7\u0026ndash;19 and 90\u0026ndash;200, Wilcoxon, p\u0026thinsp;\u0026lt;\u0026thinsp;0.001). Consistently, Cluster 3 had a significantly higher average number of sexual acts except for self-masturbation (average number of acts: 88.80 (95% CI 65.33, 147.79) versus 82.94 (95% CI 62.81, 119.90) and 118.52 (95% CI 70.32, 175.75), Wilcoxon, p\u0026thinsp;\u0026lt;\u0026thinsp;0.001) over the past 12 months, compared to Cluster 1 and Cluster 2 (average acts range: 35.80\u0026ndash;61.30 versus 31.30\u0026ndash;56.10 and 4.37\u0026ndash;21.22, Wilcoxon, p\u0026thinsp;\u0026lt;\u0026thinsp;0.001).\u003c/p\u003e \u003cp\u003eThe difference in sexual behavior frequency between Cluster 1 and Cluster 2 was particularly notable, and neither group reported SDU (sexualized drug use) in their last sexual encounter. Cluster 1 engaged in a variety of sexual acts, including insertive and receptive oral and anal sex, with a moderate level of sexual activity, averaging 35.80 to 61.30 sexual acts over the past year. In contrast, Cluster 2 had much lower levels of sexual activity, with self-masturbation being the predominant sexual act. Sexual acts with partners were rare in Cluster 2, with an average of only 4.37 to 21.22 acts, indicating that sexual activity was primarily limited to self-masturbation, with minimal engagement with partners. Importantly, neither Cluster 1 nor Cluster 2 participants reported SDU in their last sexual encounter, in stark contrast to Cluster 3. The radar chart clearly displayed the differences in clustering results among the three groups. (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e)\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec14\" class=\"Section2\"\u003e \u003ch2\u003eDifference in sexual history, PrEP Use, and STI testing across clusters\u003c/h2\u003e \u003cp\u003eWhen comparing the three Clusters, several key differences were observed. Notably, Cluster 1 and Cluster 2 members were not entirely free from SDU. SDU during sexual encounters over the past life was significantly more common in Cluster 3 (94.84% vs. 41.83% vs. 43.51%, chi-2 test, p\u0026thinsp;\u0026lt;\u0026thinsp;0.0001). Cluster 3 had the significantly highest proportion of participants who had ever taken PrEP (34.90% vs. 17.02% vs. 8.00%, chi-2 test, p\u0026thinsp;\u0026lt;\u0026thinsp;0.0001) and had the significantly highest willingness to take it in the future (64.95% vs. 55.98% vs. 47.10%, chi-2 test, p\u0026thinsp;\u0026lt;\u0026thinsp;0.0001). Drug use during sexual encounters over the past life was significantly more common in this group (94.84% vs. 41.83% vs. 43.51%, chi-2 test, p\u0026thinsp;\u0026lt;\u0026thinsp;0.0001). In the last six months, Cluster 3 also had significantly more casual male sex partners (4.55\u0026thinsp;\u0026plusmn;\u0026thinsp;6.55 vs. 2.48\u0026thinsp;\u0026plusmn;\u0026thinsp;3.21 vs. 2.74\u0026thinsp;\u0026plusmn;\u0026thinsp;3.66, F\u0026thinsp;=\u0026thinsp;12.49, p\u0026thinsp;\u0026lt;\u0026thinsp;0.0001) and regular male sex partners (3.16\u0026thinsp;\u0026plusmn;\u0026thinsp;4.37 vs. 2.27\u0026thinsp;\u0026plusmn;\u0026thinsp;3.52 vs. 2.51\u0026thinsp;\u0026plusmn;\u0026thinsp;2.53, F\u0026thinsp;=\u0026thinsp;3.58, p\u0026thinsp;=\u0026thinsp;0.028) and reported the significantly highest percentage of having sexual partners infected with an STI (8.39% vs. 3.79% vs. 3.90%, chi-2 test, p\u0026thinsp;=\u0026thinsp;0.029). Additionally, Cluster 3 had the significantly highest frequency of participation in group sex (29.03% vs. 7.84% vs. 8.44%, chi-2 test, p\u0026thinsp;\u0026lt;\u0026thinsp;0.0001) and the significantly lowest rate of always using condoms during anal sex with casual male partners (48.53% vs. 59.41% vs. 72.28%, chi-2 test, p\u0026thinsp;\u0026lt;\u0026thinsp;0.0001). Despite the lack of statistical significance, Cluster 3 had a higher proportion of participants reporting a positive HIV test result (3.92% vs. 3.10% vs. 2.80%, chi-2 test, p\u0026thinsp;=\u0026thinsp;0.2) and recent STI diagnoses other than HIV (9.03% vs. 5.85% vs. 5.84%, chi-2 test, p\u0026thinsp;=\u0026thinsp;0.334) compared to the other clusters. (Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e)\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab2\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eSexual history, PrEP Use, and STI testing among study participants, stratified by Clusters\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"6\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCategory\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eCluster 1 (N\u0026thinsp;=\u0026thinsp;581)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eCluster2 (N\u0026thinsp;=\u0026thinsp;154)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eCluster3 (N\u0026thinsp;=\u0026thinsp;155)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eChi\u003csup\u003e2\u003c/sup\u003e/F\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003eP-value\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"4\" nameend=\"c4\" namest=\"c1\"\u003e \u003cp\u003e\u003cb\u003eHave you ever taken PrEP?\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e38.61\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.0001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eYes\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e96 (17.02%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e12 (8.00%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e52 (34.90%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNo\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e468 (82.98%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e138 (92.00%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e97 (65.10%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"4\" nameend=\"c4\" namest=\"c1\"\u003e \u003cp\u003e\u003cb\u003eIf you haven't taken PrEP yet, would you want to?\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e43.75\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.0001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eYes\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e262 (55.98%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e65 (47.10%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e63 (64.95%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNo\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e206 (44.02%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e73 (52.90%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e34 (35.05%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eWhat was the result of your last HIV test?\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e8.6\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.2\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNegative\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e514 (93.62%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e136 (95.10%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e145 (94.77%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePositive\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e17 (3.10%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e4 (2.80%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e6 (3.92%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePrefer not to say\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e18 (3.28%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e3 (2.10%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e2 (1.31%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"4\" nameend=\"c4\" namest=\"c1\"\u003e \u003cp\u003e\u003cb\u003eHave you been diagnosed with an STD other than HIV in the past 6 months?\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e2.16\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.334\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eYes\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e34 (5.85%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e9 (5.84%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e14 (9.03%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNo\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e547 (94.15%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e145 (94.16%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e141 (90.97%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"4\" nameend=\"c4\" namest=\"c1\"\u003e \u003cp\u003e\u003cb\u003eHave you had any sexual partners infected with an STD in the last 6 months?\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e10.79\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.029\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eYes\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e22 (3.79)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e6 (3.90%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e13 (8.39%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNo\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e387 (66.61)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e103 (66.88%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e85 (54.84%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNot sure\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e172 (29.6)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e45 (29.22%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e57 (36.77%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"4\" nameend=\"c4\" namest=\"c1\"\u003e \u003cp\u003e\u003cb\u003eHave you ever used drugs in your sex life?\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e142.24\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.0001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eYes\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e243 (41.83)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e67 (43.51%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e147 (94.84%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNo\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e338 (58.18%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e87 (56.49%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e8 (5.16%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"4\" nameend=\"c4\" namest=\"c1\"\u003e \u003cp\u003e\u003cb\u003eThe role assumed during anal sex with another man in the last six months\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e12.08\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.017\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eInsertive\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e260 (44.75%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e60 (38.96%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e53 (34.19%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eReceptive\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e222 (38.21%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e74 (48.05%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e65 (41.94%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eBoth\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e99 (17.04%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e20 (12.99%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e37 (23.87%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"4\" nameend=\"c4\" namest=\"c1\"\u003e \u003cp\u003e\u003cb\u003eThe frequency of participation in group sex (with two or more people) in the last six months\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e37.98\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.0001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNever\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e502 (86.40%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e141 (91.56%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e109 (70.32%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eOccasionally\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e75 (12.91%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e13 (8.44%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e40 (25.81%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eOften\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e4 (0.69%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0 (0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e5 (3.23%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAlways\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0 (0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0 (0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e1 (0.65%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"4\" nameend=\"c4\" namest=\"c1\"\u003e \u003cp\u003e\u003cb\u003eFrequency of condom use during anal sex with casual male partners in the last 6 months\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e28.75\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.0001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNever\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e39 (7.91%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e9 (7.83%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e12 (8.70%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eOccasionally\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e98 (19.88%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e13 (11.30%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e33 (23.91%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eOften\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e112 (22.72%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e19 (16.52%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e38 (27.54%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAlways\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e244 (49.49%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e74 (64.35%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e55 (39.86%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"4\" nameend=\"c4\" namest=\"c1\"\u003e \u003cp\u003e\u003cb\u003eFrequency of condom use during anal sex with casual male partners in the last 6 months\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e40.08\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.0001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNever\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e15 (3.71%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e3 (2.97%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e4 (2.94%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eOccasionally\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e55 (13.61%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e7 (6.93%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e25 (18.38%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eOften\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e94 (23.27%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e18 (17.82%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e41 (30.15%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAlways\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e240 (59.41%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e73 (72.28%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e66 (48.53%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eNumber of casual male sex partners in the last 6 months\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e2.74\u0026thinsp;\u0026plusmn;\u0026thinsp;3.66\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e2.48\u0026thinsp;\u0026plusmn;\u0026thinsp;3.21\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e4.55\u0026thinsp;\u0026plusmn;\u0026thinsp;6.55\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e12.49\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.0001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eNumber of regular male sex partners in the last 6 months\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e2.51\u0026thinsp;\u0026plusmn;\u0026thinsp;2.53\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e2.27\u0026thinsp;\u0026plusmn;\u0026thinsp;3.52\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e3.16\u0026thinsp;\u0026plusmn;\u0026thinsp;4.37\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e3.58\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.028\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eCluster 1 participants were more likely to have insertive anal sex (44.75% vs. 38.96% vs. 34.19%, chi-2 test, p\u0026thinsp;=\u0026thinsp;0.017). Cluster 2 participants were more likely to have receptive anal sex (48.05% vs. 38.21% vs. 41.94%, chi-2 test, p\u0026thinsp;=\u0026thinsp;0.017). Cluster 3 participants were more likely to assume both roles during anal sex in the last six months (23.87% vs. 17.04% vs. 12.99%, chi-2 test, p\u0026thinsp;=\u0026thinsp;0.017) compared to other Clusters. (Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e)\u003c/p\u003e \u003c/div\u003e"},{"header":"Discussion","content":"\u003cp\u003eAmong 890 Chinese MSM, we identified three distinct clusters, finding that SDU in last sexual encounter was a significant criterion for differentiating sexual behavioral patterns. Cluster 3 all reported in SDU in last sexual encounter, while cluster 1 and cluster 2 did not. A key finding is that although cluster 1 and cluster 3 exhibited almost identical frequencies across various sexual acts\u0026mdash;including oral sex, anal sex, and rimming\u0026mdash;their risk profiles differed significantly. Cluster 3 exhibited significantly higher sexual risk behaviors, including having more casual and regular partners, lower consistent condom use, more frequent participation in group sex, and a higher proportion of partners with an STI in the past six months. The primary distinction between these two clusters was SDU in last sexual encounter in cluster 3, which seems to be the driving factor behind the elevated risk behaviors. In contrast, while cluster 1 reported similar sexual act frequencies, it demonstrated fewer risk behaviors, such as more consistent condom use and fewer sexual partner with STIs. On the other hand, cluster 2 primarily engaged in self-masturbation, had the least sexual act frequencies and exhibited the safest sexual practices.\u003c/p\u003e \u003cp\u003e All participants in Cluster 3 reported drug use, with the majority using poppers (83.87%). This finding contrasts with studies from other countries, which indicate a higher prevalence of cannabis, GHB/GBL, ketamine, and mephedrone use[\u003cspan additionalcitationids=\"CR33 CR34 CR35\" citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e]. The lower rate of these drugs observed in this cluster may be influenced by Chinese policies that have effectively curtailed the use of these drugs among MSM populations. Cluster 3 also showed the highest frequency of all types of sexual acts over 12 months. This contradicts a longitudinal study that showed a decline in all forms of anal sex among SDU participants over time[\u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e]. This discrepancy may be because our study participants primarily used poppers, whereas the longitudinal study focused on MSM using more addictive drugs. Long-term use of such drugs may impair physical capacity, reducing the frequency of sexual acts. Our findings are consistent with a qualitative study where users reported that drug use enhanced their sexual experience and performance, which may psychologically increase the frequency of sexual encounters[\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e]. Additionally, 34.9% of participants in Cluster 3 reported had ever taken PrEP, which is slightly higher than the prevalence rates observed in other regions. Previous studies have indicated that MSM who had ever taken PrEP in China typically ranges from 19.5\u0026ndash;24.7% [\u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e, \u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e40\u003c/span\u003e]. However, as noted in earlier research, PrEP use may contribute to lower condom usage, potentially increasing the risk of other STIs [\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e, \u003cspan additionalcitationids=\"CR17\" citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e, \u003cspan additionalcitationids=\"CR42 CR43\" citationid=\"CR41\" class=\"CitationRef\"\u003e41\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e44\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eOur findings in China mirror global trends where SDU is associated with higher-risk sexual behaviors in MSM populations. For example, studies in the U.S. and Europe have shown that MSM who use poppers, GHB/GBL, and methamphetamine engage in more casual partners, group sex, and lower condom use [\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e, \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e, \u003cspan additionalcitationids=\"CR46 CR47\" citationid=\"CR45\" class=\"CitationRef\"\u003e45\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR48\" class=\"CitationRef\"\u003e48\u003c/span\u003e]. the strong association between SDU and high-risk sexual behavior is evident across different cultural and regulatory contexts, suggesting that public health strategies need to adapt to global trends while considering local policies and cultural norms. Therefore, interventions for this cluster should focus on harm reduction strategies, particularly in relation to SDU. Clear educational messaging is needed to emphasize reducing sexual risks by consistently using condoms, even while on PrEP, and promoting waiting until sober before engaging in sex to reduce impulsive risk-taking. Furthermore, efforts should be made to ensure easy access to condoms and lubricants in high-risk settings, such as MSM-friendly venues, clubs, and group sex parties. Event organizers in these environments should be encouraged to proactively provide safety supplies and promote safer sex practices. Although PrEP usage in this cluster is relatively high, it is crucial to reinforce the importance of combining PrEP with condom use to prevent STIs beyond HIV.\u003c/p\u003e \u003cp\u003eCluster 1 includes approximately 40% of members who had previously used SDU but not recently. This group demonstrates high sexual activity with moderate numbers of sexual partners (3\u0026ndash;5), primarily engaging in one-on-one sexual encounters, with little group sex participation. Condom use is inconsistent, with some members reporting occasional use. For this cluster, public health interventions should focus on promoting consistent condom use by offering regular health education programs that emphasize the importance of using condoms with both regular and casual partners. Additionally, ensuring access to condoms, oral condoms, and lubricants through both online and offline distribution channels is crucial. Targeted education on risk communication in one-on-one relationships should be provided, helping this cluster understand that even in stable partnerships, maintaining proper safety measures is critical.\u003c/p\u003e \u003cp\u003eCluster 2 consists of members with more conservative sexual behaviors, fewer sexual partners, and a higher focus on self-masturbation. Although 40% had used SDU in the past, none reported recent use. Their sexual acts are infrequent, and they exhibit high safety awareness, consistently using condoms. This cluster is likely more discreet or closeted and may experience significant social isolation and psychological stress. Therefore, public health interventions for this clsuter should include mental health support, providing anonymous psychological counseling services to help them cope with issues related to sexual identity and societal pressures. Additionally, a mobile app could be developed to help them anonymously locate nearby sexual health clinics and services, enhancing access to essential health resources.\u003c/p\u003e \u003cp\u003eOur method for estimating the frequency of sexual acts in MSM, including handling missing data, has been validated in previous study [\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e]. This study has used similar approaches to quantify the frequency of sexual acts in MSM populations. Our study revealed that self-masturbation was the most common sexual act among participants, occurring 88.80 times over 12 months, whereas being rimmed and rimming were the least frequent, with 34.90 times respectively. This finding is consistent with previous research, which suggested self-masturbation as the most prevalent and rimming as the least practiced sexual act [\u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e49\u003c/span\u003e]. Following self-masturbation, kissing also emerged as a commonly practice sexual act with 50.20 times over 12 months, and this finding is aligned with the widely perceived significance of kissing in intimate relationships [\u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e50\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eOur study has several limitations. First, the questionnaire design did not distinguish between participants who refused to answer and those who genuinely did not engage in the behavior. As a result, we could not determine how many participants selected \u0026lsquo;no occurrence\u0026rsquo; versus those who refused to respond. To address this, we used available data and log-normal distribution fittings to impute the missing values, providing a more robust estimate and reducing bias from missing data. Second, our study may misclassify habitual SDU users who did not report in SDU during their most recent sexual encounter; this was due to most participants not reporting frequency data despite having a history of SDU. Third, self-reporting could lead to recall or social desirability biases, challenging data authenticity. Fourth, the recruitment method used in this study primarily reached MSM who are actively engaged with sexual health services or community organizations, which may introduce potential selection bias. MSM who are more involved in health-focused activities or community initiatives may be overrepresented. This bias may limit the generalizability of the study\u0026rsquo;s findings to the broader MSM population. Further, the reliance on online recruitment through WeChat and community groups may exclude MSM who are less active on these platforms. Future research should focus on including MSM who are less connected to these networks to ensure a more comprehensive understanding of the population\u0026rsquo;s behaviors and risks. Fifth, our analysis used an unsupervised machine learning during clustering. To date, unsupervised models can be difficult to interpret due to the lack of predefined labels, making results highly dependent on the chosen algorithm and parameters, which introduces subjectivity. Additionally, clustering results are sensitive to the choice of algorithm (e.g., PAM vs. hierarchical clustering) and distance metrics (e.g., Gower\u0026rsquo;s distance), potentially leading to different cluster classifications. Noise or outliers in the dataset may skew results and affect the interpretation of behavioral patterns. Sixth, we acknowledge that t-SNE results are not directly interpretable as clustering outcomes. Therefore, we included an interpretation of Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e in the Results section to clarify how the t-SNE visualization corresponds to the identified clusters\u003c/p\u003e"},{"header":"Conclusions","content":"\u003cp\u003eUsing cluster analysis, we identified three MSM subgroups with varying levels of high-risk sexual behavior. Cluster 3, involving MSM who engaged in SDU, showed the highest sexual risk and requires harm reduction strategies, including PrEP, condom promotion, and safer sex education. Cluster 1, while not reporting SDU, still exhibited higher-risk behaviors than Cluster 2 and would benefit from education on consistent condom use in relationships. Cluster 2, with the lowest risk, primarily engaged in self-masturbation and may need mental health support for social isolation. Future research should explore SDU\u0026rsquo;s long-term effects and targeted STI prevention strategies.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eConflict of interest\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNone declared.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eEthical approval\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eHuman Research Ethics Committee of Zhuhai Centre for Disease Control and Prevention (CDC) (Ethics documents ID No. [2022] 11).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthor Contributions Statement\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eBingyang She was responsible for data collection, data analysis, manuscript drafting, and revisions. Jiajun Sun, Fang Lu, Siqi Lin, Yi Liu, Gaixia Li, and Yawu Hu contributed to data collection. Weiming Tang, Rayner Tan, and Jason Ong reviewed the manuscript. Shu Su and Lei Zhang supervised the project and reviewed the manuscript.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eLZ is supported by National Key R\u0026amp;D Program of China (2022YFC2505100, 2022YFC2505103); Outstanding Young Scholars Support Program (Grant number: 3111500001); Epidemiology modeling and risk assessment (Grant number: 20200344) and Xi\u0026rsquo;an Jiaotong University Young Scholar Support Grant (Grant number: YX6J004). SS is supported by the National Natural Science Foundation of China (82304246), Natural Science Foundation of Chongqing (CSTB2023NSCQ-MSX0198), Joint Medical Research Project of Chongqing Municipal Science and the Technology Bureau and Health Commission (2024QNXM057) and Emergency special project for COVID-19 (2023IITXG26).\u003cbr\u003e\u003cbr\u003e\u003cstrong\u003eAcknowledgments\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eLZ is supported by the National Key R\u0026amp;D Program of China (2022YFC2505100, 2022YFC2505103); Outstanding Young Scholars Support Program (Grant number: 3111500001); Epidemiology modeling and risk assessment (Grant number: 20200344) and Xi\u0026rsquo;an Jiaotong University Young Scholar Support Grant (Grant number: YX6J004). SS is supported by the National Natural Science Foundation of China (Grant No. 82304246), the Natural Science Foundation of Chongqing (Grant No. CSTB2023NSCQ-MSX0198), and the Chongqing Joint Health Sciences and Technology-Health Medical Research (Grant No. 2024QNXM056), Emergency special project for COVID-19 (2023IITXG26).\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConsent for Publication\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAll authors have reviewed the manuscript and provided their consent for publication.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAvailability of Data and Materials\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eGlossary\u003c/strong\u003e\u003c/p\u003e\n\u003col\u003e\n \u003cli\u003e\u003cstrong\u003eMSM (Men Who Have Sex with Men)\u003c/strong\u003e\u003c/li\u003e\n\u003c/ol\u003e\n\u003cp\u003eRefers to men who engage in sexual activity with other men, regardless of their sexual orientation. This term includes all men who have sex with men, whether they identify as gay, bisexual, or otherwise.\u003c/p\u003e\n\u003col start=\"2\"\u003e\n \u003cli\u003e\u003cstrong\u003eSDU (Sexualized Drug Use)\u003c/strong\u003e\u003c/li\u003e\n\u003c/ol\u003e\n\u003cp\u003eThe use of drugs before or during sexual activity to enhance the experience. SDU is often associated with riskier sexual behaviors, such as unprotected sex or group sex, increasing the risk of HIV and sexually transmitted infections (STIs).\u003c/p\u003e\n\u003col start=\"3\"\u003e\n \u003cli\u003e\u003cstrong\u003ePrEP (Pre-exposure Prophylaxis)\u003c/strong\u003e\u003c/li\u003e\n\u003c/ol\u003e\n\u003cp\u003eA preventive treatment for HIV where people at high risk of infection take antiretroviral medication daily to reduce the risk of contracting HIV.\u003c/p\u003e\n\u003col start=\"4\"\u003e\n \u003cli\u003e\u003cstrong\u003eSTI (Sexually Transmitted Infections)\u003c/strong\u003e\u003c/li\u003e\n\u003c/ol\u003e\n\u003cp\u003eInfections transmitted through sexual contact, such as gonorrhea, syphilis, chlamydia, and HIV.\u003c/p\u003e\n\u003col start=\"5\"\u003e\n \u003cli\u003e\u003cstrong\u003eMean Square Error (MSE\u003c/strong\u003e)\u003c/li\u003e\n\u003c/ol\u003e\n\u003cp\u003eA measure of the average squared difference between estimated values and actual values. It is used to evaluate the accuracy of the model used to estimate sexual behavior frequencies in the study.\u003c/p\u003e\n\u003col start=\"6\"\u003e\n \u003cli\u003e\u003cstrong\u003eGamma Hydroxybutyrate/Gamma Butyrolactone (GHB/GBL):\u003c/strong\u003e\u003c/li\u003e\n\u003c/ol\u003e\n\u003cp\u003eGHB and GBL are substances often used recreationally, sometimes referred to as \u0026ldquo;club drugs.\u0026rdquo; GHB is a central nervous system depressant, commonly used in small doses for its euphoric and sedative effects. GBL is a prodrug, meaning it is converted into GHB once ingested. Both substances are frequently associated with increased sexual arousal and disinhibition, and are sometimes used during sexual encounters, which may heighten the risk of unsafe sexual practices. Overuse can lead to loss of consciousness, respiratory issues, or overdose.\u003c/p\u003e\n\u003col start=\"7\"\u003e\n \u003cli\u003e\u003cstrong\u003ePartitioning Around Medoids (PAM):\u003c/strong\u003e\u003c/li\u003e\n\u003c/ol\u003e\n\u003cp\u003ePAM is a clustering algorithm that identifies representative objects, or medoids, from a dataset. These medoids are used to group similar objects together, based on a chosen distance metric. Unlike k-means, PAM works well with non-Euclidean distances and is robust to outliers. In this study, Gower\u0026rsquo;s distance was used, which is suitable for mixed data types (continuous and categorical variables).\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n \u003cli\u003eMaxwell, S., M. Shahmanesh, and M. Gafos, \u003cem\u003eChemsex behaviours among men who have sex with men: a systematic review of the literature.\u003c/em\u003e International Journal of Drug Policy, 2019. \u003cstrong\u003e63\u003c/strong\u003e: p. 74-89.\u003c/li\u003e\n \u003cli\u003eBourne, A., et al., \u003cem\u003eThe Chemsex study: drug use in sexual settings among gay and bisexual men in Lambeth, Southwark and Lewisham.\u003c/em\u003e 2014.\u003c/li\u003e\n \u003cli\u003eEdmundson, C., et al., \u003cem\u003eSexualised drug use in the United Kingdom (UK): A review of the literature.\u003c/em\u003e International Journal of Drug Policy, 2018. \u003cstrong\u003e55\u003c/strong\u003e: p. 131-148.\u003c/li\u003e\n \u003cli\u003e\u0026Iacute;ncera-Fern\u0026aacute;ndez, D., M. G\u0026aacute;mez-Guadix, and S. Moreno-Guill\u0026eacute;n, \u003cem\u003eMental health symptoms associated with sexualized drug use (Chemsex) among men who have sex with men: a systematic review.\u003c/em\u003e International journal of environmental research and public health, 2021. \u003cstrong\u003e18\u003c/strong\u003e(24): p. 13299.\u003c/li\u003e\n \u003cli\u003eHawkinson, D.E., T.C. Witzel, and M. Gafos, \u003cem\u003eExploring practices to enhance benefits and reduce risks of chemsex among gay, bisexual, and other men who have sex with men: A meta-ethnography.\u003c/em\u003e International Journal of Drug Policy, 2024. \u003cstrong\u003e127\u003c/strong\u003e: p. 104398.\u003c/li\u003e\n \u003cli\u003eMoreno-Gamez, L., D. Hernandez-Huerta, and G. Lahera, \u003cem\u003eChemsex and Psychosis: A Systematic Review.\u003c/em\u003e Behav Sci (Basel), 2022. \u003cstrong\u003e12\u003c/strong\u003e(12).\u003c/li\u003e\n \u003cli\u003eMaxwell, S., M. Shahmanesh, and M. Gafos, \u003cem\u003eChemsex behaviours among men who have sex with men: A systematic review of the literature.\u003c/em\u003e Int J Drug Policy, 2019. \u003cstrong\u003e63\u003c/strong\u003e: p. 74-89.\u003c/li\u003e\n \u003cli\u003eAguilera-Mijares, S., et al., \u003cem\u003eVariations in Sexual Behaviors by Use of Specific Substances Among Vancouver Gay, Bisexual, and Other Men Who Have Sex with Men: An Event-Level Analysis.\u003c/em\u003e Arch Sex Behav, 2021. \u003cstrong\u003e50\u003c/strong\u003e(7): p. 2875-2886.\u003c/li\u003e\n \u003cli\u003eWang, H., K.J. Jonas, and T.E. Guadamuz, \u003cem\u003eChemsex and chemsex associated substance use among men who have sex with men in Asia: A systematic review and meta-analysis.\u003c/em\u003e Drug Alcohol Depend, 2023. \u003cstrong\u003e243\u003c/strong\u003e: p. 109741.\u003c/li\u003e\n \u003cli\u003eTomkins, A., R. George, and M. Kliner, \u003cem\u003eSexualised drug taking among men who have sex with men: a systematic review.\u003c/em\u003e Perspect Public Health, 2019. \u003cstrong\u003e139\u003c/strong\u003e(1): p. 23-33.\u003c/li\u003e\n \u003cli\u003ePrevention, C.f.D.C.a. \u003cem\u003eMen Who Have Sex with Men (MSM)\u003c/em\u003e. 2021; Available from: https://www.cdc.gov/std/treatment-guidelines/msm.htm.\u003c/li\u003e\n \u003cli\u003eNevendorff, L., et al., \u003cem\u003ePrevalence of sexualized drug use and risk of HIV among sexually active MSM in East and South Asian countries: systematic review and meta-analysis.\u003c/em\u003e J Int AIDS Soc, 2023. \u003cstrong\u003e26\u003c/strong\u003e(1): p. e26054.\u003c/li\u003e\n \u003cli\u003eRajasingham, R., et al., \u003cem\u003eA systematic review of behavioral and treatment outcome studies among HIV-infected men who have sex with men who abuse crystal methamphetamine.\u003c/em\u003e AIDS patient care and STDs, 2012. \u003cstrong\u003e26\u003c/strong\u003e(1): p. 36-52.\u003c/li\u003e\n \u003cli\u003ePlankey, M.W., et al., \u003cem\u003eThe relationship between methamphetamine and popper use and risk of HIV seroconversion in the multicenter AIDS cohort study.\u003c/em\u003e JAIDS Journal of Acquired Immune Deficiency Syndromes, 2007. \u003cstrong\u003e45\u003c/strong\u003e(1): p. 85-92.\u003c/li\u003e\n \u003cli\u003eSemple, S.J., T.L. Patterson, and I. Grant, \u003cem\u003eMotivations associated with methamphetamine use among HIV men who have sex with men.\u003c/em\u003e Journal of substance abuse Treatment, 2002. \u003cstrong\u003e22\u003c/strong\u003e(3): p. 149-156.\u003c/li\u003e\n \u003cli\u003eXu, J.-J., et al., \u003cem\u003eRecreational drug use among Chinese men who have sex with men: a risky combination with unprotected sex for acquiring HIV infection.\u003c/em\u003e BioMed research international, 2014. \u003cstrong\u003e2014\u003c/strong\u003e.\u003c/li\u003e\n \u003cli\u003eChen, X., et al., \u003cem\u003eClub drugs and HIV/STD infection: an exploratory analysis among men who have sex with men in Changsha, China.\u003c/em\u003e PloS one, 2015. \u003cstrong\u003e10\u003c/strong\u003e(5): p. e0126320.\u003c/li\u003e\n \u003cli\u003eMao, X., et al., \u003cem\u003eUse of multiple recreational drugs is associated with new HIV infections among men who have sex with men in China: a multicenter cross-sectional survey.\u003c/em\u003e BMC Public Health, 2021. \u003cstrong\u003e21\u003c/strong\u003e(1): p. 354.\u003c/li\u003e\n \u003cli\u003eBourne, A., et al., \u003cem\u003eIllicit drug use in sexual settings (\u0026lsquo;chemsex\u0026rsquo;) and HIV/STI transmission risk behaviour among gay men in South London: findings from a qualitative study.\u003c/em\u003e Sexually transmitted infections, 2015. \u003cstrong\u003e91\u003c/strong\u003e(8): p. 564-568.\u003c/li\u003e\n \u003cli\u003eCurtis, T.J., et al., \u003cem\u003ePatterns of sexualised recreational drug use and its association with risk behaviours and sexual health outcomes in men who have sex with men in London, UK: a comparison of cross-sectional studies conducted in 2013 and 2016.\u003c/em\u003e Sexually transmitted infections, 2020. \u003cstrong\u003e96\u003c/strong\u003e(3): p. 197-203.\u003c/li\u003e\n \u003cli\u003eGuerras, J.-M., et al., \u003cem\u003eAssociation of sexualized drug use patterns with HIV/STI transmission risk in an internet sample of men who have sex with men from seven European countries.\u003c/em\u003e Archives of Sexual Behavior, 2021. \u003cstrong\u003e50\u003c/strong\u003e: p. 461-477.\u003c/li\u003e\n \u003cli\u003eHeath, J., A. Lanoye, and S.A. Maisto, \u003cem\u003eThe role of alcohol and substance use in risky sexual behavior among older men who have sex with men: a review and critique of the current literature.\u003c/em\u003e AIDS and Behavior, 2012. \u003cstrong\u003e16\u003c/strong\u003e: p. 578-589.\u003c/li\u003e\n \u003cli\u003eFolch, C., et al., \u003cem\u003eHigh prevalence of drug consumption and sexual risk behaviors in men who have sex with men.\u003c/em\u003e Medicina Cl\u0026iacute;nica (English Edition), 2015. \u003cstrong\u003e145\u003c/strong\u003e(3): p. 102-107.\u003c/li\u003e\n \u003cli\u003eCarey, J.W., et al., \u003cem\u003eDrug use, high-risk sex behaviors, and increased risk for recent HIV infection among men who have sex with men in Chicago and Los Angeles.\u003c/em\u003e AIDS and Behavior, 2009. \u003cstrong\u003e13\u003c/strong\u003e: p. 1084-1096.\u003c/li\u003e\n \u003cli\u003eSalazar-Vizcaya, L., et al., \u003cem\u003eua Clusters of sexual behaviour in HIV-positive men who have sex with men reveal highly dissimilar time trends.\u003c/em\u003e Clin Infect Dis Off Publ Infect Dis Soc Am, 2019. \u003cstrong\u003e15\u003c/strong\u003e.\u003c/li\u003e\n \u003cli\u003eSalazar-Vizcaya, L., et al., \u003cem\u003eClusters of Sexual Behavior in Human Immunodeficiency Virus-positive Men Who Have Sex With Men Reveal Highly Dissimilar Time Trends.\u003c/em\u003e Clin Infect Dis, 2020. \u003cstrong\u003e70\u003c/strong\u003e(3): p. 416-424.\u003c/li\u003e\n \u003cli\u003eDishion, T.J., T. Ha, and M.-H. V\u0026eacute;ronneau, \u003cem\u003eAn ecological analysis of the effects of deviant peer clustering on sexual promiscuity, problem behavior, and childbearing from early adolescence to adulthood: an enhancement of the life history framework.\u003c/em\u003e Developmental psychology, 2012. \u003cstrong\u003e48\u003c/strong\u003e(3): p. 703.\u003c/li\u003e\n \u003cli\u003eBlondeel, K., et al., \u003cem\u003eSexual behaviour patterns and STI risk: results of a cluster analysis among men who have sex with men in Portugal.\u003c/em\u003e BMJ open, 2021. \u003cstrong\u003e11\u003c/strong\u003e(1): p. e033290.\u003c/li\u003e\n \u003cli\u003eHummel, M., D. Edelmann, and A. Kopp-Schneider, \u003cem\u003eClustering of samples and variables with mixed-type data.\u003c/em\u003e PLoS One, 2017. \u003cstrong\u003e12\u003c/strong\u003e(11): p. e0188274.\u003c/li\u003e\n \u003cli\u003eKault, D., \u003cem\u003eThe Shape of the Distribution of the Number of Sexual Partners.\u003c/em\u003e Statistics in Medicine, 1996. \u003cstrong\u003e15\u003c/strong\u003e(2): p. 221-230.\u003c/li\u003e\n \u003cli\u003eZhang, L., E.P. Fung Chow, and D.P. Wilson, \u003cem\u003eMen who have sex with men in China have relatively low numbers of sexual partners.\u003c/em\u003e Infect Dis Rep, 2011. \u003cstrong\u003e3\u003c/strong\u003e(1): p. e10.\u003c/li\u003e\n \u003cli\u003eBrogan, N., et al., \u003cem\u003eSexually transmitted infections in MSM: Canadian results from the European men-who-have-sex-with-men internet survey (EMIS-2017).\u003c/em\u003e Canada Communicable Disease Report, 2019. \u003cstrong\u003e45\u003c/strong\u003e(11): p. 271.\u003c/li\u003e\n \u003cli\u003eDolengevich-Segal, H., et al., \u003cem\u003eDrug-related and psychopathological symptoms in HIV-positive men who have sex with men who inject drugs during sex (slamsex): Data from the U-SEX GESIDA 9416 Study.\u003c/em\u003e PLoS One, 2019. \u003cstrong\u003e14\u003c/strong\u003e(12): p. e0220272.\u003c/li\u003e\n \u003cli\u003eSchecke, H., et al., \u003cem\u003eCrystal methamphetamine use in sexual settings among German men who have sex with men.\u003c/em\u003e Frontiers in Psychiatry, 2019. \u003cstrong\u003e10\u003c/strong\u003e: p. 886.\u003c/li\u003e\n \u003cli\u003eN\u0026ouml;stlinger, C., et al., \u003cem\u003eDrug use, depression and sexual risk behaviour: a syndemic among early pre-exposure prophylaxis (PrEP) adopters in Belgium?\u003c/em\u003e AIDS care, 2020. \u003cstrong\u003e32\u003c/strong\u003e(sup2): p. 57-64.\u003c/li\u003e\n \u003cli\u003eVaccher, S.J., et al., \u003cem\u003ePrevalence, frequency, and motivations for alkyl nitrite use among gay, bisexual and other men who have sex with men in Australia.\u003c/em\u003e International Journal of Drug Policy, 2020. \u003cstrong\u003e76\u003c/strong\u003e: p. 102659.\u003c/li\u003e\n \u003cli\u003eSewell, J., et al., \u003cem\u003eChanges in chemsex and sexual behaviour over time, among a cohort of MSM in London and Brighton: findings from the AURAH2 study.\u003c/em\u003e International Journal of Drug Policy, 2019. \u003cstrong\u003e68\u003c/strong\u003e: p. 54-61.\u003c/li\u003e\n \u003cli\u003eMarques Oliveira, P., C. Sousa Reis, and M.A. Vieira-Coelho, \u003cem\u003eGetting Inside the Mind of Gay and Bisexual Men Who Have Sex with Men with Sexualized Drug Use\u0026ndash;A Systematic Review.\u003c/em\u003e International Journal of Sexual Health, 2023. \u003cstrong\u003e35\u003c/strong\u003e(4): p. 573-595.\u003c/li\u003e\n \u003cli\u003eHuang, W., et al., \u003cem\u003ePrepared for PrEP: preferences for HIV pre-exposure prophylaxis among Chinese men who have sex with men in an online national survey.\u003c/em\u003e BMC Infectious Diseases, 2019. \u003cstrong\u003e19\u003c/strong\u003e(1): p. 1057.\u003c/li\u003e\n \u003cli\u003eZhang, G., et al., \u003cem\u003ePre-exposure prophylaxis uptake for high-risk men who have sex with men in China: a multi-city cross-sectional survey.\u003c/em\u003e AIDS Research and Therapy, 2023. \u003cstrong\u003e20\u003c/strong\u003e(1): p. 32.\u003c/li\u003e\n \u003cli\u003eWei, C., et al., \u003cem\u003ePatterns and levels of illicit drug use among men who have sex with men in Asia.\u003c/em\u003e Drug Alcohol Depend, 2012. \u003cstrong\u003e120\u003c/strong\u003e(1-3): p. 246-9.\u003c/li\u003e\n \u003cli\u003eTomkins, A., R. George, and M. Kliner, \u003cem\u003eSexualised drug taking among men who have sex with men: a systematic review.\u003c/em\u003e Perspectives in public health, 2019. \u003cstrong\u003e139\u003c/strong\u003e(1): p. 23-33.\u003c/li\u003e\n \u003cli\u003eHe, L., et al., \u003cem\u003eNew types of drug use and risks of drug use among men who have sex with men: a cross-sectional study in Hangzhou, China.\u003c/em\u003e BMC Infect Dis, 2018. \u003cstrong\u003e18\u003c/strong\u003e(1): p. 182.\u003c/li\u003e\n \u003cli\u003eShusen Liu, M., PhD and Roger Detels, MD, MS*, \u003cem\u003eRecreational drug use: an emerging concern among venuebased male sex workers in China\u003c/em\u003e, C.C.f.D.C.a.P. National Centre for AIDS/STD Prevention and Control, Beijing China, Editor. 2012: Sex Transm Dis.\u003c/li\u003e\n \u003cli\u003eBlomquist, P., et al., \u003cem\u003eP531 Chemsex and STI clinic use among MSM: results from a large online survey in england\u003c/em\u003e. 2019, BMJ Publishing Group Ltd.\u003c/li\u003e\n \u003cli\u003eGlynn, R.W., et al., \u003cem\u003eChemsex, risk behaviours and sexually transmitted infections among men who have sex with men in Dublin, Ireland.\u003c/em\u003e International Journal of Drug Policy, 2018. \u003cstrong\u003e52\u003c/strong\u003e: p. 9-15.\u003c/li\u003e\n \u003cli\u003eAguilera-Mijares, S., et al., \u003cem\u003eVariations in Sexual Behaviors by Use of Specific Substances Among Vancouver Gay, Bisexual, and Other Men Who Have Sex with Men: An Event-Level Analysis.\u003c/em\u003e Archives of Sexual Behavior, 2021. \u003cstrong\u003e50\u003c/strong\u003e(7): p. 2875-2886.\u003c/li\u003e\n \u003cli\u003eBrown, R.E., et al., \u003cem\u003ePartner-level substance use associated with increased sexual risk behaviors among men who have sex with men in San Francisco, CA.\u003c/em\u003e Drug and Alcohol Dependence, 2017. \u003cstrong\u003e176\u003c/strong\u003e: p. 176-180.\u003c/li\u003e\n \u003cli\u003eRichters, J., et al., \u003cem\u003eMasturbation, paying for sex, and other sexual activities: the Second Australian Study of Health and Relationships.\u003c/em\u003e Sexual health, 2014. \u003cstrong\u003e11\u003c/strong\u003e(5): p. 461-471.\u003c/li\u003e\n \u003cli\u003eCornelisse, V.J., et al., \u003cem\u003eThe frequency of kissing as part of sexual activity differs depending on how men meet their male casual sexual partners.\u003c/em\u003e International journal of STD \u0026amp; AIDS, 2018. \u003cstrong\u003e29\u003c/strong\u003e(6): p. 598-602.\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"bmc-public-health","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"pubh","sideBox":"Learn more about [BMC Public Health](http://bmcpublichealth.biomedcentral.com/)","snPcode":"","submissionUrl":"https://www.editorialmanager.com/pubh/default.aspx","title":"BMC Public Health","twitterHandle":"@BMC_series","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"em","reportingPortfolio":"BMC Series","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"Sexualized drug use, MSM, Cluster analysis, Sexual Behavioral Patterns, sexual acts","lastPublishedDoi":"10.21203/rs.3.rs-5286116/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-5286116/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003e\u003cstrong\u003eIntroduction:\u003c/strong\u003e Sexualized drug use (SDU) refers to using drugs before and during sex to enhance experiences, increasing high-risk behaviors among men who have sex with men (MSM). This study explores how SDU affects sexual behavior in Chinese MSM.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eMethods:\u003c/strong\u003e We collected information on demographics, sexual acts, drug use, and condom attitudes among 890 MSM in six Chinese cities via WeChat ads through community-based organizations from March 23 to April 22, 2022. Cluster analysis using Gower’s distance and hierarchical clustering explored differences in sexual acts among MSM who reported SDU in their last encounter and otherwise.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eResults:\u003c/strong\u003e Cluster analysis categorized participants into three clusters. Cluster 3 (n=155) reported 100% SDU in their last sexual encounter (83.87% poppers use), whereas Clusters 1 (n=581) and 2 (n=154) reported none. Compared to other clusters, Cluster 3 had significantly higher PrEP use (34.90% vs. 17.02% vs. 8.00%, p\u0026lt;0.0001), more sexual acts over the past 12 months (35.80-61.30 vs. 31.30-56.10 and 4.37-21.22, p\u0026lt;0.001), more regular (3.16±4.37 vs. 2.27±3.52 vs. 2.51±2.53, p=0.028) and casual partners (4.55±6.55 vs. 2.48±3.21 vs. 2.74±3.66, p\u0026lt;0.0001), more partners with STIs (8.39% vs. 3.79% vs. 3.90%, p=0.029), and lower consistent condom use (48.53% vs. 59.41% vs. 72.28%, p\u0026lt;0.0001). Cluster 1 had moderate frequency in all sexual acts except self-masturbation, which was most common in Cluster 2.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConclusion:\u003c/strong\u003e SDU is a stratum for identifying MSM subgroups, and MSM who reported SDU demonstrated higher sexual risk behaviors and PrEP usage. Among those not practicing SDU, self-masturbation is a key behavioral indicator for subgrouping.\u003c/p\u003e","manuscriptTitle":"Identifying subgroups of Chinese men who have sex with men based on sexual behavior and drug use patterns using a clustering analysis approach","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2024-11-13 06:28:48","doi":"10.21203/rs.3.rs-5286116/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2024-10-24T13:14:25+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2024-10-24T12:23:54+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2024-10-23T00:11:26+00:00","index":"","fulltext":""},{"type":"submitted","content":"BMC Public Health","date":"2024-10-18T03:31:47+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"bmc-public-health","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"pubh","sideBox":"Learn more about [BMC Public Health](http://bmcpublichealth.biomedcentral.com/)","snPcode":"","submissionUrl":"https://www.editorialmanager.com/pubh/default.aspx","title":"BMC Public Health","twitterHandle":"@BMC_series","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"em","reportingPortfolio":"BMC Series","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"00d5e497-81b8-4f1f-ab69-86296b4a1a3c","owner":[],"postedDate":"November 13th, 2024","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[],"tags":[],"updatedAt":"2025-04-14T16:11:48+00:00","versionOfRecord":{"articleIdentity":"rs-5286116","link":"https://doi.org/10.1186/s12889-025-22388-x","journal":{"identity":"bmc-public-health","isVorOnly":false,"title":"BMC Public Health"},"publishedOn":"2025-04-10 16:04:54","publishedOnDateReadable":"April 10th, 2025"},"versionCreatedAt":"2024-11-13 06:28:48","video":"","vorDoi":"10.1186/s12889-025-22388-x","vorDoiUrl":"https://doi.org/10.1186/s12889-025-22388-x","workflowStages":[]},"version":"v1","identity":"rs-5286116","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-5286116","identity":"rs-5286116","version":["v1"]},"buildId":"qtupq5eGEP_6zYnWcrvyt","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}