Gay and bisexual men represent approximately 58% of men living with AIDS in the United States, and an additional 8% of men who have sex with men (MSM) also use injection drugs . Men of color have been particularly affected by the epidemic, and they represent more than half of new AIDS cases among MSM in the USA every year since 1997 [1,2]. Although gay and bisexual men continue to be concerned about HIV prevention in their communities, recent data in North America, Australia, and Europe suggest a resurgence of unprotected anal sex, sexually transmitted diseases (STD), and HIV infections among these men [3–7]. Although there are many potential explanations for these trends, substance use is one factor frequently found to be associated with sexual risk behavior and the acquisition or transmission of STD or HIV among gay and bisexual men [8–11], including HIV-positive men .
Drinking alcohol and using illicit substances are both important to the health of HIV-positive MSM because they can impair immune functioning, react adversely with prescribed medications , are related to worse psychosocial and cognitive functioning , and potentially decrease adherence to HIV medications . Data from almost 10 000 HIV-positive gay and bisexual men about their alcohol and other substance use during the past 5 years found that approximately 29% of the men had ‘possible’ alcohol abuse, 58% had used a substance other than alcohol, and 9% had injected drugs . Other recent data of substance use in the past 3 months show robust rates of alcohol use (64%), marijuana (36%), nitrate inhalants or ‘poppers’ (27%), cocaine (13%) and amphetamines (12%) among urban, HIV-positive, gay and bisexual men . Qualitative data from HIV-positive gay and bisexual men show that the relationship between substances and sexual risk is complex, and is related to at least three personal factors (behavioral, cognitive, and emotional), and three contextual factors (setting, partner type, and ethnicity) .
The potential link between substance use and risky sexual behavior among HIV-positive gay and bisexual men is important to understand so that interventions can be developed to minimize harms to both the infected individual and his partners (such as STD, HIV, and potential re-infection with HIV). Cross-sectional and longitudinal studies have documented an association between drinking or drug use and sexual risk behavior among gay and bisexual men [12,18–25], and longitudinal research shows that men who use certain substances are more likely to become HIV positive [8,26,27]. Studies that look at the link between overall substance use and sexual behavior (without requiring that the two occur at the same time) and studies examining substance use in the context of sexual behavior commonly find an association between substance use and sexual risk, whereas the findings from event-level data have been more mixed .
Recent studies have found that the use of specific ‘party drugs’ such as methamphetamine [21,29,30], poppers [12,17,23,31–33], ecstasy , and cocaine  are likely to be associated with sexual risk, whereas substances such as alcohol or marijuana are often not. Many of the substances associated with unprotected anal intercourse enhance sensations or ease anal penetration (by relaxing sphincter muscles, enhancing stimulation, or lowering inhibitions) [17,30,36,37]. Some of these drugs, particularly methamphetamines, inhibit sexual functioning, and have been associated with viagra use among HIV-positive gay and bisexual men . Substance use also appears to have a greater effect on sexual risk with casual partners compared with primary or main partners [12,38,39].
In this study, we describe the alcohol and non-injection drug use among a diverse group of 1168 HIV-positive gay and bisexual men recruited for an HIV prevention intervention trial. First, we describe the prevalence of drinking alcohol and the use of illicit, non-injection drugs, as well as the prevalence of drinking and the use of illicit substances before or during sex. We include the use of viagra here because men often appear to use it with other party drugs, and we were interested in testing for independent effects on sexual risk. Next, we examined the univariate and multivariate associations between substance use variables and sexual risk. For all analyses, we focused on casual partners and on the sexual behaviors most likely to transmit HIV or STD; unprotected insertive anal intercourse (UIAI) and unprotected receptive anal intercourse (URAI). This study extends the current literature by describing risk practices by partner serostatus separately (HIV-negative partners, unknown-serostatus partners, and HIV-positive partners). This allowed us to examine whether substance use, or substance use in the context of sex, is differentially associated with risk based on the HIV-positive individual's knowledge or perception of the HIV serostatus of his casual partners.
Data reported here are from the baseline quantitative survey gathered using audio-computer assisted self-interviewing technology from a diverse sample of 1168 HIV-positive gay and bisexual men in New York City and San Francisco. Data were collected as part of the Seropositive Urban Men's Intervention Trial, a multi-site randomized controlled trial of an HIV prevention intervention funded by the Centers for Disease Control and Prevention. This study was approved by the Centers for Disease Control and Prevention Institutional Review Board and by the Institutional Review Boards at both sites. The recruitment methods, eligibility, and demographics of the overall sample are described elsewhere in this issue .
Men were first asked if they drank any alcohol in the past 90 days. Men who drank alcohol in the past 90 days were then asked about how many times they drank before or during sex during those 90 days.
Non-injection drug use
Men were first asked if they had used any non injection drugs in the past 90 days. Those who reported non-injection drug use were asked to indicate whether they had used any of the following nine types or classes of drugs: (i) speed/amphetamines; (ii) crystal (methamphetamine); (iii) barbiturates/tranquilizers; (iv) cocaine; (v) ecstasy; (vi) special K (ketamine); (vii) marijuana (pot, hash); (viii) poppers (amyl nitrate); and (ix) GHB (gamma hydroxybutyrate). Correlations between the use of speed/amphetamines and crystal were robust (r = 0.45) but were low enough that we decided to look at the relationships with each category separately. In addition to recording whether the men used each drug or not, we also calculated the number of these illicit, non-injection drugs used in the past 90 days (range from 0 to 9).
Regarding the link between substance use and sexual behavior, we asked the men who used illicit, non-injection drugs about how many times they had used any drugs before or during sex in the past 90 days. This question was a general measure of the use of substances before or during sex and was not specifically linked to specific sexual behaviors or partner serostatus. Finally, because of recent interest in the use of viagra and sexual risk among gay and bisexual men, we asked the men if they were ‘currently taking’ viagra. The use of viagra was not included in the total number of non-injection drugs described above.
Sociodemographic and health characteristics
Demographic and health variables were used as control variables and were usually collapsed into categories for analyses; city (New York and San Francisco), race/ethnicity (black, Hispanic, white, other), sexual orientation (gay, non-gay), educational background (high school or less, some college or more), personal income (less than US$20 000, US$20 000 or more), and ever been diagnosed with AIDS (yes, no). Age was analysed as a continuous variable.
Sexual risk behaviors
Sex behaviors of interest (insertive anal and receptive anal sex) were assessed by asking participants to indicate the frequency in the 3 months before baseline of these behaviors, with and without condoms. Separate frequencies were obtained for main partners (defined as ‘a partner you would call your boyfriend, spouse, significant other, or life partner’) and casual partners. For these analyses, frequencies were dichotomized into binary outcomes of ‘never’ or ‘ever in past 3 months’. Sexual risk behavior was analysed in terms of six binary outcomes with casual partners: two risk behaviors; (1) any UIAI, and (2) any URAI, by three serostatuses of casual partners; (1) HIV negative, (2) unknown serostatus, or (3) HIV positive.
In preliminary analyses, we determined the prevalence of substance use and substance use in the context of sex. Then we used logistic regression to determine associations between the six outcome variables (sex behavior × partner serostatus) and participant characteristics such as sociodemographics and health status to identify potential control variables needed in analyses of substance use and risk behavior. Sociodemographic and health variables that were significantly associated with a particular outcome (α = 0.05) at the univariate level were used to develop multiple regression models of risk behaviors.
The relationship between substance use and risk behavior was also assessed using logistic regression. We first determined the association between several substance use variables (individual substances, number of substances used, and substance use before or during sex) and each of the six outcomes (sex behavior × partner serostatus) in univariate models. Substance use variables that were significantly associated with a particular outcome (α = 0.05) were then used to develop multiple regression models of the risk behaviors.
Variables for the final six multiple regression models were determined using the automated step-wise selection process in the SAS Logistic Procedure (version 8.2). The main effects for substance use, demographics, and health status were added to the model if they were significant at the α = 0.05 level. After each additional step, effects already in the model were dropped if they did not remain significant (α = 0.05). This step-wise process was used to select an appropriate and parsimonious model given the large number of variables that were significantly associated with the risk behaviors in univariate models and the likely presence of intercorrelations among the substance use variables. Parameters were estimated using maximum likelihood with Fisher scoring.
Table 1 shows the prevalence of substance use for the overall sample and by city. Three-quarters of the sample drank alcohol in the past 90 days. Almost 60% of the sample had used an illicit drug in the past 90 days, and the most commonly used drugs were marijuana (42%), poppers (26%), cocaine (18%), and methamphetamine (10%). The current use of viagra was 12%, making it one of the more popular substances used by men. HIV-positive men in San Francisco were significantly more likely to have used marijuana, methamphetamine, speed/amphetamines, and GHB, whereas men in New York City were approximately two and a half times more likely to have used cocaine.
Approximately 35% of the sample used two or more illicit drugs and 15% used more than three drugs in the past 3 months. Drinking before or during sex was reported by 57% of men, and illicit non-injection substance use before or during sex was reported by over 47% of men. Men in San Francisco were significantly more likely to report using illicit, non-injection drugs before or during sex than were men in New York City.
Univariate associations between participant characteristics and sexual risk behavior
We conducted univariate logistic regression models for each of the six outcomes, with participant characteristics including city, race, income, education, sexual orientation, age, and whether the participant had ever been diagnosed with AIDS (data not shown). Significant predictors for each particular outcome were entered into the multiple regression models described below.
Univariate associations between substance use and sexual risk behavior
Tables 2–4 present crude odds ratios (OR) and 95% confidence intervals (CI) from univariate models of sexual risk by partner serostatus and substance use. We examined the relationships of 11 substance use variables (alcohol, the nine illicit, non-injection drugs, and viagra) with sexual risk. The results indicate that a majority of the substance use variables were significantly associated with sexual risk behaviors with casual partners, regardless of serostatus (Tables 2–4).
Multiple regression models predicting sexual risk behavior
Tables 5–7 present adjusted OR and 95% CI from multiple regression models of the six outcomes of interest. The final models include only the control (sociodemographic and health) factors and substance use variables that were significantly associated with each risk behavior.
With respect to control variables, health status did not appear to be significantly associated with any risk behavior with casual partners regardless of partner serostatus. However, Tables 5–7 show several general patterns of association between other sociodemographic variables and risk behavior, and where significant, the direction of associations with risk was consistent across specific types of behavior and partner serostatus. Overall, younger participants had greater odds of engaging in risk than older participants. Gay-identified participants had greater odds of engaging in risk than non-gay-identified participants. Higher income participants had greater odds of engaging in risk than lower income participants. More-educated (more than high school) participants had greater odds of engaging in risk than less-educated participants. With respect to race, Hispanic individuals had greater odds of engaging in risk compared with whites, whereas African Americans and participants of other races did not significantly differ from whites.
After controlling for significant sociodemographic variables, the final multiple regression models show several significant associations between substance use and risk behaviors. For HIV-negative casual partners (Table 5), several substance use variables were significantly associated with risk behavior with these partners. In particular, the use of special K (ketamine) was significantly associated with UIAI (OR = 3.1), gamma hydroxybutyrate use was associated with UIAI (OR = 2.2) and URAI (OR = 3.2), and methamphetamine use was associated with URAI (OR = 2.0). Neither drinking nor illicit substance use before or during sex were associated with unprotected sexual behaviors with HIV-negative casual partners.
With casual partners of unknown HIV serostatus, Table 6 shows that several substance use variables were significantly associated with risk behavior with these partners. The use of gamma hydroxybutyrate was significantly associated with UIAI (OR = 5.2), the use of poppers was associated with UIAI (OR = 1.8) and URAI (OR = 1.5), and methamphetamine use was associated with URAI (OR = 2.3). Drinking alcohol before or during sex was associated with UIAI (OR = 1.7) and illicit drug use before or during sex was associated with URAI (OR = 1.8).
With HIV-positive casual partners (Table 7), a number of substance use variables were significantly associated with risk behavior with these partners. The use of gamma hydroxybutyrate was significantly associated with UIAI (OR = 2.4), the use of poppers was associated with URAI (OR = 1.8), and methamphetamine use was associated with UIAI (OR = 2.8) and URAI (OR = 4.3). Drinking alcohol before or during sex was also associated with both unprotected sexual behaviors, and illicit drug use before or during sex was associated with UIAI (OR = 1.6).
HIV-positive gay and bisexual men report significant drinking and substance use as well as drinking and substance use in the context of sex. Drinking and drug use, by themselves, can threaten physical and emotional functioning, as well as the use of medical care and adherence to HIV medications, among HIV-positive individuals [14,41,42]. In addition, the use of certain party drugs, such as methamphetamine, and gamma hydroxybutyrate was associated with sexual risk behaviors with casual partners, regardless of serostatus, even after controlling for the use of other substances. The use of poppers was associated with sexual risk only with HIV-positive or unknown-serostatus casual partners, whereas ketamine was associated with sexual risk only with HIV-negative casual partners. The contextual variables (drinking and drug use before or during sex) were associated with risk behaviors only with HIV-positive and unknown-serostatus casual partners. These findings pose a public health challenge for ensuring the health of HIV-positive gay and bisexual men and their sexual partners.
Recent media and research reports have particularly focused on methamphetamine as a drug that is fueling high-risk sexual behavior and HIV seroconversion among gay and bisexual men [27,30,43,44]. Our data suggest a more complex scenario, with men who report any use of various party drugs reporting more sexual risk behavior, mostly regardless of serostatus. In terms of the method of action, these drugs have some similarities in that they allow individuals to dissociate from reality and to escape the more mundane concerns of life, at least to some extent, and for varying lengths of time . We can thus say that the use of these drugs is a marker of potential risk behavior, but we cannot determine whether these drugs cause risk or whether a third variable is associated with substance use and risk (e.g. individuals who have certain personality characteristics may be drawn to party drugs and risk).
These party drugs are often used in the context of sex and in sexually charged settings, such as dance clubs, circuit parties, sex clubs, and sex parties . Interestingly, in univariate analyses, men who used drugs and drank alcohol before or during sex in the past 3 months were more likely to report engaging in risk behavior in the past 3 months with partners of all serostatuses. However, in multivariate analyses, men who used certain club drugs were more likely to report engaging in risk behavior, but men who reported drinking or using non-injection drugs in the context of sexual behavior did not report more risk with HIV-negative casual partners, only with HIV-positive and serostatus-unknown casual partners. This pattern may suggest the importance of the knowledge of a partner's serostatus among HIV-positive men. Qualitative data looking at sexual risk and substance use among HIV-positive men support this finding at the level of the sexual episode . It may be that, with knowledge of a partner's HIV-negative serostatus, HIV-positive men are more able to enact their plans for safer sex, even if they use substances. The fact that a casual partner says he is ‘negative’, as opposed to not saying anything, and therefore being ‘unknown’ to the participant, may make one or both partners take more responsibility for keeping the HIV-negative partner uninfected. Interventions that attempt to encourage serostatus disclosure by HIV-positive individuals to their sexual partners have been found to reduce risk .
For unknown-serostatus partners, drinking before sex was associated with the riskiest sexual behavior (UIAI), and drug use before or during sex was associated with URAI. The pattern of results for unknown-serostatus partners is similar to that reported previously for analyses that combined HIV-negative and unknown-serostatus casual partners into one group . With partners whose serostatus is unknown, substance use in the context of sex is related differently to particular sexual behaviors, suggesting that men may use substances in situations with partners of unknown serostatus to engage in unprotected behaviors, which may be more desirable for some men. For example, many party drugs facilitate receptive anal sex. In a sexual situation in which serostatus is unknown, drug use may allow the positive partner to make assumptions that his partner is also HIV positive, and thus make riskier behavior acceptable. Research shows that in the absence of knowledge, gay and bisexual men make assumptions about their partner's serostatus on the basis of a variety of contextual cues and characteristics [48,49]. Substance use in such a situation may allow men to ignore the fact that they do not know their partner's serostatus, and are actually making a potentially erroneous assumption. Finally, for HIV-positive partners, substance use and contextual variables were also associated with risk. It may be that substances are the social lubricant making risk more likely or facilitating certain sexual behaviors, but the men may have already chosen to accept the risk of unprotected sex with HIV-positive partners, not being as concerned about STI or re-infection as they would be about infecting an HIV-negative man. Substance use may also help HIV-positive men to find and have sex with other HIV-positive partners, which then leads to intentional or accidental unprotected sex.
The overall limitations of the SUMIT study are discussed elsewhere , but there are a few additional limitations of these particular analyses. First, the data are cross-sectional, so causation cannot be implied. Therefore, we cannot know whether substance use leads to unprotected sex, whether such sex is planned in advance and substances help it to happen, or whether some third variable such as having a risk-taking personality explain both substance use and transmission risk. In addition, we do not have episode-level data, so we do not know whether substance use during specific sexual episodes was associated with sexual risk behavior among the men or the serostatus of men in any particular sexual episode. Despite these limitations, this study adds to the literature by demonstrating that among HIV-positive MSM: (i) certain party drugs appear to be associated with unprotected sexual behavior with casual partners, regardless of the serostatus of the partner; (ii) focusing on one drug as the cause of problems among any group, such as the current focus on methamphetamine use among gay and bisexual men, is short-sighted and may ignore certain broader dynamics that transcend any particular drug; and (iii) the associations between ‘any substance use in the context of sex’ and ‘any sexual risk behavior in the past 3 months’ vary by the serostatus of the partner.
Substance use among HIV-positive gay and bisexual men is a potential threat to both personal and public health. To understand more about the complex dynamics for gay and bisexual men between HIV status, sexuality, and sexual risk behavior, it would be helpful to conduct additional quantitative and qualitative studies regarding these issues. Focusing on the role of serostatus disclosure by both HIV-positive and HIV-negative men is sorely needed. In addition, qualitative studies about substance use and sexual behavior will allow for a more nuanced understanding of the complex dynamics between sex, substance use, and risk. It is also important for doctors to talk about substance use with their patients, and the challenges that substance use may cause with utilizing medical care, adhering to HIV treatments, and engaging in safer sexual behavior. From a broader perspective, it is also important to understand substance use and sexual behavior in the context of HIV-positive mens’ lives, which may include an abundance of free time (as a result of disability) and potential boredom. Structural solutions that address some of the challenges HIV-positive gay and bisexual men face and tap into some of their altruism may help decrease the link between substance use and risk with HIV-unknown-serostatus partners. By starting to understand these individual and structural dynamics, we can address how some of these intertwined issues contribute to the HIV/AIDS epidemic among gay and bisexual men.
The authors would also like to acknowledge the following people who contributed to this research: Cynthia Gómez, Colleen Hoff, Perry Halkitis, Ann O'Leary, David Bimbi, Rich Wolitski, Tim Matheson, Byron Mason, Carmen Mandic, Bonnie Faigeles, Nick Alvarado, Andrew Nelson Peterson, Eric Rodriguez, Paul Galatowitsch, Michael Marino, Aongus Burke, Michael Stirratt, Eric Martin, Gloria Abitol, Caroline Bailey, and Cindy Lyles.
Sponsorship: Research on SUMIT was funded by the Centers for Disease Control and Prevention through cooperative agreements with New Jersey City University (UR3/CCU216471) and the University of California, San Francisco (UR3/CCU916470).
2. Centers for Disease Control and Prevention. HIV/AIDS among racial/ethnic minority men who have sex with men – United States, 1989–1998
3. Centers for Disease Control and Prevention. Primary and secondary syphilis among men who have sex with men – New York City, 2001
4. Centers for Disease Control and Prevention. Increases in HIV diagnoses – 29 states, 1999-2002
5. Chen SY, Gibson S, Weide D, McFarland W. Unprotected anal intercourse between potentially HIV-serodiscordant men who have sex with men in San Francisco. J Acquir Immune Defic Syndr 2003; 33:166–170.
6. Van De Ven P, Rawstorne P, Crawford J, Kippax S. Increasing proportions of Australian gay and homosexually active men engage in unprotected anal intercourse with regular and with casual partners. AIDS Care 2002; 14:335–341.
7. Wolitski RJ, Valdiserri RO, Denning PH, Levine WC. Are we headed for a resurgence in the HIV epidemic among men who have sex with men? Am J Public Health 2001; 91:883–888.
8. Chesney MA, Barrett DC, Stall R. Histories of substance use and risk behavior precursors to seroconversion in homosexual men. Am J Public Health 1998; 88:113–116.
9. Stall R, Purcell DW. Intertwining epidemics: a review of research on substance use among men who have sex with men and its connection to the AIDS epidemic. AIDS Behav 2000; 4:181–192.
10. Klitzman RL, Pope HG, Hudson JI. MDMA (“ecstasy”) abuse and high-risk sexual behaviors among 169 gay and bisexual men. Am J Psychiatry 2000; 157:1162–1164.
11. McNall M, Remafedi G. Relationship of amphetamine and other substance use to unprotected intercourse among young men who have sex with men. Arch Pediatr Adolesc Med 1999; 153:1130–1135.
12. Purcell DW, Parsons JT, Halkitis PN, Mizuno Y, Woods WJ. Substance use and sexual transmission risk behavior of HIV-positive men who have sex with men. J Subst Abuse 2001; 13:185–200.
13. Harrington RD, Woodward JA, Hooton TM, Horn JR. Life threatening interactions between HIV-1 protease inhibitors and illicit MDMA and gamma-hydroxybutyrate. Arch Intern Med 1999; 159:2221–2224.
14. Green JE, Saveanu RV, Bornstein RA. The effect of previous alcohol abuse on cognitive function in HIV infection. Am J Psychiatry 2004; 161:249–254.
15. Flaks RC, Burman WJ, Gourley PJ, Rietmeijer CA, Cohn DL. HIV transmission risk behavior and its relation to antiretroviral treatment adherence. Sex Trans Dis 2003; 30:399–404.
16. Sullivan PS, Nakashima AK, Purcell DW, Ward JW, and the Supplement to the HIV/AIDS Surveillance Study Group. Geographic differences in noninjection and injection substance use among HIV-seropositive men who have sex with men: western United States versus other regions. J Acquir Immune Defic Syndr Hum Retrovirol 1998; 19:266–273.
17. Purcell DW, Ibañez GE, Schwartz D. Under the influence: alcohol and drug use and sexual behavior among HIV-positive gay and bisexual men
. In: HIV+ Sex: The Psychological and Interpersonal Dynamics of HIV-Seropositive Gay and Bisexual Men's Relationships
. Edited by Halikitis PN, Wolitski RJ, Gómez CA. Washington, DC: American Psychological Association; 2005. pp. 163–182.
18. Barrett DC, Bolan G, Joy D, Counts K, Doll L, Harrison J. Coping strategies, substance use, sexual activity, and HIV sexual risks in a sample of gay male STD patients. J Appl Soc Psychol 1995; 25:1058–1072.
19. Crosby GM, Stall R, Paul JP, Barrett DC, Midanik LT. Condom use among gay/bisexual male substance abusers using the timeline follow-back method. Addict Behav 1996; 21:249–257.
20. Dolezal C, Meyer-Bahlburg HFL, Remien RH, Petkova E. Substance use during sex and sensation seeking as predictors of sexual risk behaviors among HIV positive and HIV negative gay men. AIDS Behav 1995; 1:19–28.
21. Molitor F, Truax S, Ruiz JD, Sun RK. Association of methamphetamine use during sex with sexual behaviors and HIV infection among non-injection drug users in California. West J Med 1998; 268:93–97.
22. Mulry G, Kalichman SC, Kelly JA. Substance use and unsafe sex among gay men: global versus situational use of substances. J Sex Educ Ther 1994; 20:175–184.
23. Ostrow DG, Beltran ED, Joseph JG, DiFrancisco W, MACS/CCD Study Group. Recreational drugs and sexual behavior in the Chicago MACS/CCS cohort of homosexually active men. J Subst Abuse 1993; 5:311–325.
24. Paul J, Stall R, Davis F. Sexual risk for HIV transmission among gay/bisexual men in substance abuse treatment. AIDS Educ Prevent 1993; 5:11–24.
25. Stall R, McKusick L, Wiley J, Coates TJ, Ostrow DG. Alcohol and drug use during sexual activity and compliance with safe sex guidelines for AIDS: the AIDS behavioral research project. Health Educ Q 1986; 13:359–371.
26. Penkower L, Dew M, Kingsley L, Becker J, Satz P, Schaerf F, et al
. Behavioral, health and psychosocial factors and risk for HIV infection among sexually active homosexual men: The Multicenter AIDS Cohort Study. Am J Public Health 1991; 81:194–196.
27. Silvestre A, Lyter D, Valdiserri R, Huggins J, Rinaldo C. Factors related to seroconversion among homo- and bisexual men after attending a risk-reduction educational session. AIDS 1989; 3:647–650.
28. Leigh BC, Stall R. Substance use and risky sexual behavior for exposure to HIV. Am Psychol 1993; 48:1035–1045.
29. Halkitis PN, Parsons JT, Wilton L. An exploratory study of contextual and situational factors related to methamphetamine use among gay and bisexual men in New York City. J Drug Issues 2003; 33:413–432.
30. Reback CJ, Larkins S, Shoptaw S. Changes in the meaning of sexual risk behaviors among gay and bisexual male methamphetamine abusers before and after drug treatment. AIDS Behav 2004; 8:87–98.
31. Ekstrand ML, Stall RD, Paul JP, Osmond DH, Coates TJ. Gay men report high rates of unprotected anal sex with partners of unknown or serodiscordant status. AIDS 1999; 13:1525–1533.
32. Halkitis PN, Parson JT. Recreational drug use and HIV-risk sexual behavior among men frequenting gay social venues. J Gay Lesbian Soc Serv 2003; 14:19–38.
33. Woody GE, Donnell D, Seage GR, Metzger D, Marmor M, Koblin B, et al
. Non-injection substance use correlates with risky sex among men having sex with men. Drug Alcohol Depend 1999; 53:197–205.
34. Klitzman RL, Pope HG, Hudson JI. MDMA (“ecstasy”) abuse and high-risk sexual behaviors among 169 gay and bisexual men. Am J Psychiatry 2000; 157:1162–1164.
35. McNall M, Remafedi G. Relationship of amphetamine and other substance use to unprotected intercourse among young men who have sex with men. Arch Pediatr Adolesc Med 1999; 153:1130–1135.
36. Lowry TP. Psychosexual aspects of the volatile nitrites. J Psychoactive Drugs 1982; 14:77–79.
37. Romanelli F, Smith KM, Thornton AC, Pomeroy C. Poppers: epidemiology and clinical management of inhaled nitrite abuse. Pharmacotherapy 2004; 24:69–78.
38. Seage GR, Mayer KH, Wold G, Lenderking WR, Goldstein R, Cai B, et al
. The social context of drinking, drug use, and unsafe sex in the Boston Young Men Study. J Acquir Immune Defic Syndr Hum Retrovirol 1998; 17:368–375.
39. Paul JP, Stall RD, Crosby GM, Barrett DC, Midanik LT. Correlates of sexual risk-taking among gay male substance abusers. Addiction 1994; 89:971–983.
40. Wolitski RJ, Parsons JT, Gómez CA, Purcell DW, Hoff CC, Halkitis PN. Prevention with gay and bisexual men living with HIV: rationale and methods of the Seropositive Urban Men's Intervention Trial. AIDS 2005; 19(suppl. 1):S1–S11.
41. McNaghten AD, Hanson DL, Dworkin MS, Jones JL. Adult/Adolescent Spectrum of Disease Project. Differences in prescription of antiretroviral therapy in a large cohort of HIV-infected patients. J Acquir Immune Defic Syndr 2003; 32:499–505.
42. Ickovics JR, Meade CS. Adherence to HAART among patients with HIV: breakthroughs and barriers. AIDS Care 2002; 14:309–318.
43. Heredia C. Dance of death: crystal meth fuels HIV
. San Francisco Chronicle
4 May 2003, A-1.
44. Altman LK. Gays’ use of Viagra and methamphetamine is linked to diseases
. New York Times
11 March 2004, A-23.
45. Freese TE, Miotto K, Reback CJ. The effects and consequences of selected club drugs. J Subst Abuse Treatm 2003; 23:151–156.
46. Mansergh G, Colfax G, Marks G, Rader M, Guzman R, Buchbinder S. The circuit party men's health survey: findings and implications for gay and bisexual men. Am J Public Health 2001; 91:953–958.
47. Kalichman SC, Rompa D, Cage M, DiFonzo K, Simpson D, Austin J, et al
. Effectiveness of an intervention to reduce HIV transmission risks in HIV-positive people. Am J Prevent Med 2001; 21:84–92.
48. Gold RS, Skinner MJ, Hinchy J. Gay men's stereotypes about who is HIV infected: a further study. Int J STD AIDS 1999; 10:600–605.
49. Gold RS, Skinner MJ, Ross MW. Situational factors and thought processes associated with unprotected intercourse in young gay men. AIDS 1992; 6:1021–1030.