Introduction

The HIV epidemic continues to expand among men who have sex with men (MSM) in China. The prevalence of HIV/AIDS among MSM had increased from 0.9% in 2003 to 7.8% in 2014 [1]. In some less economically developed regions in China, the HIV prevalence among MSM is over 20% [2]. Drug use and HIV/AIDS are proven to be intertwined epidemics, particularly among MSM [3]. A meta-analytic review of studies showed that MSM have a 2.4-fold higher risk for drug abuse, according to the criteria of the DSM-IV (fourth edition of the Diagnostic and Statistical Manual of Mental Disorders, published by the American Psychiatric Association), than heterosexual people [4]. In China, the prevalence of illicit drug use among MSM ranged from 12.7 to 47.3%, depending on the types of drug examined [5,6,7,8]. It is widely believed that drug use among MSM is associated with HIV-related risk behaviors [9,10,11] and HIV transmissions [12, 13].

While a number of studies have focused on the relationship between drug use and HIV infection among MSM in general [14, 15], fewer studies have investigated the prevalence and correlates of drug use among HIV-positive MSM (HIVMSM). The issue of drug use among HIVMSM is of particular concern because of several reasons. First, drug use and polydrug use are more prevalent among HIV-positive MSM than their HIV-negative or unknown serostatus counterparts [16,17,18], even when cannabis, which has potential medicinal use for HIV, is excluded [17, 18]. Moreover, substance use can result in more adverse health outcomes among HIV-infected population, either directly worsening immune and cognitive functioning or indirectly due to decreased adherence to and effectiveness of antiretroviral medications [19, 20]. In addition, drug use may lead HIVMSM to take sexual risk behaviors, such as group sex and condomless anal sex. Condomless anal sex has a high risk of HIV transmission in the absence of biomedical intervention such as effective antiretroviral therapy (ART) among those living with HIV and pre-exposure prophylaxis (PrEP) among those without HIV infection and sexually transmitted infections [21]. Despite the significance, few studies have investigated the factors associated with drug use among this population beyond Western societies. To our knowledge, no research has investigated the prevalence and correlates of drug use among HIVMSM in China. Such studies are urgently warranted for the design of related interventions.

Previous studies have shown that depression and negative affect are risk factors of drug use among MSM [22]. Based on the minority stress model [23], drug use may be a potential maladaptation of HIVMSM to stressful situations caused by internalized homophobia, expectations of stigma, and experiences of prejudice and violence [24, 25]. However, various forms of mental health problems (e.g. depression and anxiety) have not been consistently associated with drug use in prior research among MSM in Western societies. While some studies found that drug use is more common in MSM with depression/anxiety symptoms [9, 21, 26], others found no association between mental health and drug use [27, 28]. Moreover, the relationship between mental health and drug use may depend on the level of drug use, with frequent or dependent users having poorer mental health [27]. The findings are hence mixed. More studies are needed to ascertain the relationship between mental health and drug use among MSM.

In addition to the contention that maladaptation potentially contributes to drug use among HIVMSM, another contention is that HIVMSM may use drugs to achieve pleasurable sexual experiences and disinhibition [29]. Certain drugs, such as methamphetamine and inhaled nitrites (popper), may have an effect on the central nervous system that increases sexual arousal, alleviates anxiety, and enhances sexual pleasure [28]. The prevalence of drug use for sex among MSM in China is rather high. A study showed that 24.1% of 655 MSM in Beijing, China reported using inhaled nitrites (popper) during anal intercourse in the last 3 months [30]. A study in Changsha, China reported that about 21.4% of MSM used drugs at some time before or during sex in the past 6 months [31].

A limitation of previous investigations is that they mainly focus on risk factors (e.g., depression and sexual risk behavior) that are associated with drug use among HIVMSM, whereas few studies have focused on positive factors that might reduce the chance of drug use. Empirical studies have shown that positive affect predicts reduced levels of substance use among adolescents [32]. Previous studies have also found a significant effect of life satisfaction on substance use [33,34,35], relapse among chronic drug abusers [36], and remission from drugs among treatment-seeking drug users [37]. To our knowledge, no study has examined the influence of life satisfaction and positive affect on substance use among HIVMSM.

This study aims to investigate the prevalence and risk and protective factors associated with global and sexual situation-specific drug use in the past 30 days among HIVMSM in China. Based on the literature review, the studied risk factors included mental distress (negative affect, HIV-related stress, and depression) and the number of male sex partners (regular and nonregular male sex partners); factors that might reduce the chance of drug use included life satisfaction and positive affect. We hypothesized that (1) negative affect, HIV-related stress, and depression were positively associated with drug use among HIVMSM; (2) life satisfaction and positive affect were negatively associated with drug use among HIVMSM; and (3) numbers of regular and nonregular male sex partners were positively associated with drug use among HIVMSM.

Methods

Study Population and Procedure

The inclusion criteria used were as follows: (1) men who have sex with at least one man in the last six months, (2) with Chinese nationality, (3) aged 18 years old or above, and (4) diagnosed as HIV positive for at least 3 months.

A cross-sectional survey was conducted in Chengdu, China from July to October 2013. Participants were recruited by the staff of a local non-governmental organization (NGO). The NGO is one of the largest organizations serving MSM in China, including approximately 600 local HIVMSM, which was about 60% of reported HIVMSM in the city (the data was collected by the online reporting system of the local CDC). The NGO publicized the study by displaying posters in the public area of its office and social media to their service users. The fieldworkers of the NGO approached prospective participants, briefed them about the study and logistics, and invited them to participate in the study after confirming their eligibility. Interested participants were invited to visit the office of the NGO and met with a research assistant there. The participants were reassured that their refusal would not affect their right to use any services the NGO offered and that they can quit the interview any time they want. After obtaining the participants’ written informed consent, anonymous face-to-face interviews using a pilot-tested and structured questionnaire were conducted in a room with ensured privacy. Upon completion of the interview, an honorarium of RMB 50 (approximately USD 7.5) was provided to the participants to compensate for their time. A total of 450 HIVMSM were successfully contacted by the peer fieldworkers, among which 415 (92%) completed the interview with written informed consent. Ethics approval was obtained from the Survey and Behavioral Research Ethics Committee of The Chinese University of Hong Kong.

Measures

Dependent Variables

The two dependent variables were the use of any illicit drugs in the past 30 days and the use of such drugs during anal intercourse in the past 30 days. The questions for the former was “Did you use any drugs (e.g., heroin, meth, cocaine, popper/RUSH) in the past 30 days” and for the latter was “Did you use any drugs (e.g., heroin, meth, cocaine, popper/RUSH) during anal intercourse in the past 30 days?”

Independent Variables

Positive Affect and Negative Affect

The 20-item Positive and Negative Affect Schedule scale [38] was administered to assess positive affect and negative affect. The scale consists of 20 items, in which 10 items reflect positive affect (e.g., excited, enthusiastic) and 10 for negative affect (e.g., ashamed, scared). Participants were asked to rate on a 5-point Likert scale to reflect the extent they feel the listed emotions at the present moment (1 = not at all to 5 = extremely). Scores for each subscale range from 10 to 50, with higher scores representing higher levels of positive affect and negative affect. In this study, the Cronbach’s alpha value was 0.93 for the Positive Affect Subscale and 0.87 for the Negative Affect Subscale.

Life Satisfaction

The Satisfaction with Life Scale [39] was used to assess the participants’ level of satisfaction with life as a whole. The Chinese version has been shown to be reliable and valid among Chinese populations [40, 41]. The respondents were asked to rate on a 7-point Likert scale how much they agree with the statements (1 = strongly disagree to 7 = strongly agree). Sample items included “In most ways my life is close to my ideal,” and “So far I have gotten the important things I want in life.” A higher sum score of the items indicated a higher level of life satisfaction (alpha = 0.83).

Depression

The 20-item Center for Epidemiologic Studies Depression Scale (CES-D) [42] was used to assess the presence of depressive symptoms. The scale has shown good internal reliability among PLWH [43, 44] and MSM [45, 46]. The Chinese version of the CES-D has been validated [47, 48]. The participants were asked to rate on a 4-point Likert scale of how often they experienced depressive symptoms in the past 7 days (1 = not at all to 4 = nearly every day). In this study, we used a sum score of all the items to represent depression, with a higher score indicating more depressive symptoms. The scale showed good reliability in this sample (alpha = 0.92).

HIV-Related Stress

Validated Perceived Stress Scale for People living with HIV/AIDS was used to assess HIV-related stress [49]. The scale includes eight subscales, namely, social/psychological problems, sexual relationship, functional problems, social acceptance/rejection issues, work-related issues, family/offspring issues, accessibility to treatment, and treatment outcomes. One item of each of the eight subscales was selected to construct an 8-item scale on perceived stress. The internal reliability of this scale is good (alpha = 0.86).

Regular and Nonregular Male Sex Partners

Regular male sex partners referred to spouse, boyfriends, or lovers, whereas nonregular partners included commercial, casual, or anonymous sex partners [50]. The participants were asked whether they had regular and/or nonregular male partners in the past 30 days. If the participants responded “yes”, they were further asked to give the number of their regular and nonregular male partners.

Background Variables

Other socio-demographic variables, including age, education (high school or below vs. college or above), marital status (single vs. married/cohabited/divorced), residence (local vs. non-local), sexual orientation (homosexual vs. heterosexual/bisexual/not sure), and personal monthly income, were also measured. Moreover, HIV- and health-related information were collected, including HIV diagnosed years (< 6 months vs. ≥ 6 months) and self-perceived health (very poor/poor/fair vs. good/very good).

Statistical Analysis

All the analyses were performed using Stata 14.2. Descriptive statistics were used to characterize the study population. Both dependent variables (drug use in the past 30 days and drug use during anal intercourse in the past 30 days) were binary; thus, we fit multiple logistic regression models to study the associations. We first examined the relationships between multiple socio-demographic variables and the outcome variables using multiple logistic regression models. The significant background variables were adjusted in subsequent analyses. Next, we performed a univariate analysis between each independent variable and the two dependent variables. Afterward, we investigated the effects of each independent variable on the dependent variable when adjusting for socio-demographic variables that were significant in previous regression models. The odds ratios (OR) and their 95% confidence intervals were reported.

Results

Descriptive Statistics

Among the 415 participants, 10.8% had used drugs and 8.2% had used drugs during anal intercourse in the past 30 days, respectively. Among the 45 participants who had used drugs in the past 30 days, 39 used Rush/popper; eight used methamphetamine; five used Viagra; two used marijuana; one used heroin; one used ketamine. Of the past 30-day drug users, 29 used Rush/popper only. Considering some HIVMSM may not be involved in sexual activities in the past month, we restricted our analysis of drug use during anal intercourse to those having anal intercourse with men in the past 30 days only. Among the 269 participants who had anal intercourse with men in the past 30 days, 31 participants (11.5%) used drugs during anal intercourse.

The descriptive statistics were shown in Table 1. Over half of the participants aged between 26 and 35 (50.8%), were single (65.1%), were local residents (55.4%), reported a homosexual sex orientation (85.3%), and possessed a college degree or above (60.7%). Of the participants, 11.1% had a monthly income below 1000 RMB (approximately USD 150), whereas 41.4% and 47.5% of the participants had a monthly income between 1001 and 3000 RMB (approximately USD 150–445) and over 3000 RMB (approximately USD 445), respectively. As for the HIV-related characteristics, more than 80% of HIVMSM had been diagnosed for more than 6 months; 57.1% of the participants perceived their health as very poor/poor/fair. While more than half (55.7%) of the participants had no regular male partners, 37.6% and 6.7% of the participants had one or more than two regular male partners, respectively. As for nonregular male partners, 10.6% of the participants had one and 19.3% had more than two, while others had no nonregular partners.

Table 1 Descriptive statistics (N = 415)
Full size table

Association Between Background Variables and Drug Use and Drug Use During Anal Intercourse

The logistic regression models of the associations between background variables and drug use behaviors were presented in Table 2. None of these variables were significantly associated with either drug use in the past 30 days or drug use during anal intercourse. The relationship between income and the past 30-day drug use approached marginal significance (1001–3000 RMB, ORu = 3.72; reference: ≤ 1000, p = 0.09). The effects of income (1001–3000 RMB, ORu = 6.29; reference: ≤ 1000, p = 0.09) and education (college or above, ORu = 2.54; reference: high school or below, p = 0.05) on drugs during anal intercourse in the past 30 days were marginally significant. We adjusted these potential correlates of drug use among HIVMSM in the following analysis.

Table 2 Association between background variables and drug use in the past 30 days/drug use during anal intercourse in the past 30 days among HIVMSM
Full size table

Factors Associated with Drug Use in the Past 30 Days

The results of factors associated with drug use in the past 30 days were shown in Table 3. In accordance with previous studies among other populations, life satisfaction was negatively associated with the past 30-day drug use among HIVMSM in China (ORa = 0.90, p < 0.001). After adjusting for relevant background variables, participants with more nonregular male sex partners were more likely to use drugs in the past 30 days (1, ORa = 4.76; ≥ 2, ORa = 4.79; reference: 0, p < 0.001), while the number of regular male partners were not statistically associated with past 30-day drug use. Surprisingly, no mental distress variables (i.e., negative affect, HIV-related stress, and depression) were associated with past 30-day drug use among HIVMSM in China.

Table 3 Logistic regression of drug use in the past 30 days/drug use during anal intercourse in the past 30 days among HIVMSM
Full size table

Factors Associated with Drug Use During Anal Intercourse in the Past 30 Days

The results of the correlates of drug use during anal intercourse in the past 30 days were shown in Table 3. Similar to the past 30-day drug use, the effect of mental distress variables on drug use during anal intercourse was not significant. Life satisfaction kept showing a protective effect on not using drugs during anal intercourse (ORa = 0.89, p < 0.001). As expected, participants with more nonregular male partners were more likely to use drugs during anal intercourse (1 nonregular partner, ORa = 7.87; ≥ 2 nonregular partners, ORa = 7.51; reference: 0, p < 0.001). However, participants with one regular partner were less likely to use drugs during anal intercourse compared to those without regular partners (1 regular partner, ORa = 0.27; reference: 0, p < 0.01).

Discussion and Conclusion

This research is among the first to investigate various factors that may be associated with global and sexual situation-specific drug use among HIVMSM in China. Our findings showed that 10.8% of the participants had used drugs in the past 30 days. Among the 269 participants who had anal intercourse with men in the past 30 days, 11.5% had used drugs during anal intercourse. The prevalence of drug use among Chinese HIVMSM was lower than that in Western societies. Previous studies on drug use in general among HIVMSM showed a 3-month prevalence of 51% in the UK [51] and 70% in the U.S. [52]. About 1/3 of sexually active HIVMSM used drugs before or during anal intercourse in the U.K. [26] and about 51% used drugs in conjunction with sex in the U.S. [52]. Other studies conducted in mainland China involving MSM in general reported higher prevalence of drug use: 27.7% (lifetime use) in Sichuan Province [50], 18.2% (3-month) in Hangzhou [13], and 21% (12-month) nationally [8]. Such prevalence was higher than what was found in this study, possibly due to longer window periods and regional differences.

We employed logistic regression models to examine the factors associated with drug use among Chinese HIVMSM. One important finding was that the studied mental distress variables were not significantly associated with drug use. It might be because, for some MSM, drug use (especially during anal intercourse) is mainly a pursuit of (sexual) pleasure, rather than an escape from mental distress [53]. Such pleasure-seeking behaviors may not be considered problematic and pathological among the gay community, even when they involve risk [54, 55]. Thus, drug use is not necessarily associated with self-reported mental health problems among HIVMSM. Nevertheless, previous research suggested that varying levels of drug addiction among MSM were associated with very different mental health outcomes, with chronic or dependent users showing higher levels of stress and mental health problems compared to those who occasionally use drugs [56, 57]. Future research should further examine the relationship between mental health and different frequencies and levels of drug use and among HIVMSM in China.

In addition, we found that the number of nonregular sex partner was strongly associated with drug use in general and drug use during anal intercourse. Such finding was consistent with those found among HIVMSM in the West that recreational drug use is more prevalent among those with multiple sex partners [58]. Drug use among MSM in China was also associated with multiple sexual partnerships [30] and those used gay mobile apps to seek sexual partners [8]. MSM with more male sex partners may be more likely to be sexually active and use drugs as a way to enhance sexual performance and pleasure. MSM who are diagnosed with HIV also commonly rationalize their use of drugs as a strategy to escape from concerns about rejection by sexual partners (following HIV status disclosure) [59]. Moreover, MSM with more sexual partners may be more connected to the gay communities and have more gay friends. Studies of MSM in the West have found that higher level of gay community involvement or greater affiliation to gay male culture is associated with increased use of illicit drugs [60]. Our study further differentiated the effect of regular and nonregular partners on drug use. While participants with more nonregular were involved in more drug use, having a regular sex partner seemed to be a protective factor against drug use, especially during anal intercourse. That may be because those with regular sex partners possess other protective factors not included in the study, such as social support, intimacy, better physical and mental health, and lower level of loneliness [61]. Future studies may further explore the mechanisms by which regular partnership reduces global and sexual situation-specific drug use among HIVMSM.

Our findings also showed the importance to focus on subjective wellbeing that might reduce the chance of drug use and to go beyond the traditional pathology-focused model of addiction that narrowly assesses disease-specific and mental health factors [62, 63]. Our findings showed that participants who were more satisfied with their lives were less likely to have used drugs in the last 30 days or to use drugs during anal intercourse, possibly because satisfying life conditions represent something one valued that can be lost if substance use starts or continues, thereby acting as a deterrent to using drugs by “raising the price” of doing so [37]. In fact, life satisfaction is negatively correlated with a wide range of risk-taking behaviors, including physical fighting, carrying weapons, suicide ideation and self-harming behaviors, sexual risk-taking, and substance abuse [64]. Prevention programs for drug use in HIVMSM should place more emphasis on helping them find greater meaning in life and develop life goals to increase their satisfaction and happiness without the use of drugs. Specific prevention interventions such as positive psychotherapy [65] dedicated to developing resources of positive emotion, character strengths, and a sense of meaning in life may help individuals to set realistic life goals and equip them skills for enhancing their psychological wellbeing and overall life satisfaction.

This study has several limitations. First, the cross-sectional research design precluded us from making causal inferences about the relationship between psychosocial and sexual relationship variables and drug use among HIVMSM. Longitudinal investigations are warranted to elucidate whether various factors predict drug use among this population. Second, our data on illicit drug use may be subject to reporting bias, although we kept strict anonymity in order to reduce the potential inaccuracies. The use of other measures, such as biological markers and urine test may provide a useful approach to examine the criterion-related validity of self-report data. Third, we did not investigate the factors associated with the use of any specific drug due to a limited number of participants who report using each type of drug. Considering various drugs may have different patterns of use and distinct health effects, future studies may further explore whether factors associated with drug use among HIVMSM vary by different drug categories. Fourth, our participants were recruited by non-probabilistic sampling and may not be representative of the overall HIVMSM population in China. However, studies using similar sampling approaches have yielded valuable information on MSM in China [e.g., 30, 31]. Fifth, our data were collected in 2013 while some potential changes may have occurred since then (e.g., type of substance used, pattern of drug use, treatment as prevention policy). Hence, the implication of drug use on HIV transmission may also change. More recent data are needed to determine the prevalence and correlates of drug use among HIVMSM in China. Lastly, given that this study was conducted in Chengdu, cautions should be made when generalizing the findings to other cities in China. Future studies may test these associations using other samples to examine whether these mechanisms are cultural and context specific.

Limitations notwithstanding, the findings of this study are informative for designing effective substance-use intervention programs targeting HIVMSM, and to reduce new HIV infection and other risky behaviors among MSM. This study provided evidence that positive wellbeing factors, particularly life satisfaction, were protective factors against drug use among HIVMSM. Thus, investigations and interventions for drug use among HIVMSM must focus on the role of positive functioning, other than the traditional pathology-focused model assessing disease-specific and mental health factors [62, 63]. Moreover, the number of nonregular partners was positively associated with global and sexual situation-specific measures of drug use among HIVMSM, whereas having a stable partner seemed to act as a protective factor against drug use during anal intercourse. Future studies should further examine the potential differential effects of partner type on drug use among HIVMSM.