Plankey, Michael W PhD*; Ostrow, David G MD, PhD†; Stall, Ron PhD‡; Cox, Christopher PhD§; Li, Xiuhong MS§; Peck, James A PsyD∥; Jacobson, Lisa P PhD§
The use of methamphetamine, a powerful central nervous system stimulant associated with sexual enhancement, has been popular among men who have sex with men (MSM) for many years.1-3 Behavioral research has demonstrated that gay male methamphetamine users are more likely to engage in high-risk sexual practices for the transmission of HIV and other sexually transmitted infections and to be HIV-seropositive.4-18
Even though a substantial literature exists to demonstrate the cross-sectional relationship between methamphetamine use and risk for HIV transmission among MSM, only a few studies have examined the relation between methamphetamine use and HIV seroincidence. Chesney et al19 demonstrated 1.02 and 2.89 relative risks of HIV seroconversion for current methamphetamine users versus nonusers and chronic methamphetamine users versus nonusers, respectively, after adjusting for unprotected anal intercourse in 337 seronegative gay men followed for 3 years from the San Francisco Men's Health Study. Most recently, Buchacz et al20 examined the association of methamphetamine use and HIV seroincidence in 2991 MSM who were tested anonymously for HIV in San Francisco. Thirty-four of 290 methamphetamine users (within the past year) had recently seroconverted, yielding a relative risk of HIV seroconversion associated with methamphetamine use of 2.5 (95% confidence interval [CI]: 0.9 to 6.9), adjusted for age, race, ethnicity, use of other noninjected drugs in the past year (including poppers), marijuana use, and alcohol use. Although important in documenting a possible link between methamphetamine use and increased risk of HIV sexual transmission, these studies have been limited by the small number of seroconversions and self-reported data collected over a short follow-up period, inadequate adjustment for confounding factors, or limited geographic diversity, and thus only provide a limited measure of the methamphetamine-HIV seroconversion relation.
A better understanding of the drug use-HIV seroincidence relationships in general and the methamphetamine-HIV seroincidence relation in particular among vulnerable populations is needed for effective preventive interventions. An examination of this relationship across multiple sites and over longer periods of time, taking into account important risk factors, might provide an important replication of the finding that methamphetamine use is associated with new HIV infections among MSM as well as potential clues to the mechanism(s) of this relationship. Because methamphetamine use can highly disengage sexual pleasure from cognition21 (the well-known “behavioral disinhibition” mechanism), its recent spread among MSM across geographically diverse areas has led to its association with the intentional practice of unprotected anal sex (“barebacking”).22-25 This recent trend only intensifies the potential impact of methamphetamine use among MSM and others in maintaining the current unacceptably high rate of new HIV infections.7
Our examination of popper use, in addition to methamphetamine use, in this study was the result of previous findings demonstrating popper use as one of the most consistently and strongly associated drug use behaviors in HIV seroconversion among Multicenter AIDS Cohort Study (MACS) participants26 and in the study by Chesney et al19 as well. Additionally, popper use has been shown to be frequently used, along with methamphetamine, to “enhance sexual pleasure, get a better high, or take the edge off of methamphetamine.”27 More significantly, however, may be the fact that popper use can enable an individual to experience multiple orgasms during a methamphetamine-fueled extended sexual session.
In this study, we examined the association of methamphetamine and other drug use, along with risky sexual behavior, on HIV seroconversion using data from MSM who were initially HIV-seronegative and were followed over time in the MACS.
Population and Study Design
The MACS is an ongoing prospective study of the natural history of HIV infection among MSM in the United States. A total of 6973 men were recruited (4954 in 1984 to 1985, 668 in 1987 to 1991, and 1351 in 2001 to 2003) at 4 centers located in Baltimore-Washington, DC; Chicago; Los Angeles; and Pittsburgh. The study design has been described previously,28,29 and only methods relevant to the present study are presented here. The study questionnaires are available at http://www.statepi.jhsph.edu/macs/forms.html. The MACS study protocols were approved by institutional review boards of each of the participating centers, their community partners, and community advisory boards, and informed consent was obtained from all participants on entry into the MACS and with each add-on substudy for which they volunteered.
Participants returned every 6 months for a detailed interview, a physical examination, and collection of blood for laboratory testing and storage in a central repository. The interview included questions about medical conditions, medical treatments, sexual behavior, and recreational drugs (eg, marijuana, poppers, cocaine, crack, heroin, methamphetamine, ecstasy) and alcohol consumption. All questions concerning sexual and drug use behaviors were assessed using audio computer-assisted self-interviewing (ACASI) during phase 2 of the study (October 1996-September 2004), a methodology shown to yield more accurate assessments of “sensitive behaviors” than interviewer-administered questionnaires.30 Enzyme-linked immunosorbent assays (ELISAs) with confirmatory Western blot tests were performed on all participants initially and at every semiannual visit thereafter for initially seronegative participants.
A prospective cohort design was used to examine the effects of recreational drugs, particularly methamphetamine and poppers, on the risk of HIV seroconversion among initially HIV-negative participants. The cohort included all participants who were HIV-negative at enrollment and had data on methamphetamine use at baseline or follow-up visits (n = 4003). Because methamphetamine use was not directly assessed at visits 16 to 25 (April 1992-April 1996), data from these visits were excluded. Specifically, the analysis used MACS data at visits 1 to 15 (April 1984-September 1991), hereafter designated as “phase 1,” and at visits 26 to 41 (October 1996-September 2004), hereafter designated as “phase 2.”
Time to HIV seroconversion was the outcome of interest. The date of seroconversion was defined as the midpoint between the dates of the last HIV-seronegative visit and the first HIV-seropositive visit. Participants who did not seroconvert during 1 or both observation periods also contributed time to the analysis. In visits 1 to 15 (phase 1), the person-time contributed to the analysis was the time from study entry to the date of seroconversion for those who seroconverted before September 1991 (end of visit 15) or to the earlier date of the last visit seen or visit 15 (if no seroconversion). Participants who were HIV-seronegative as of October 1, 1996 (visit 26) contributed time to the analysis of visits 26 to 41. Specifically, in phase 2, such participants were treated as late entries; entry time was the date of visit 26 for those enrolled before 1996 and the date of the baseline visit for the 2001 to 2003 cohort, and exit time was the date of seroconversion if subsequent to 1996 or the earlier of the last visit seen or visit 41 (if no seroconversion).
The primary exposure of interest was the use of methamphetamine (yes/no), which was defined as “yes” if a participant reported using it at the current visit or the previous visit. The questions eliciting information about use of methamphetamine differed in the 2 phases. In the early phase, we used an affirmative answer to the question about use of “amphetamines, speed, crystal, or other uppers,” whereas in the second phase, use of methamphetamine (along with speed) was a unique drug category in the questionnaire.
Demographics and other behaviors were included to adjust for possible confounding. Age at the time of enrollment was calculated using self-reported date of birth and was treated as a continuous covariate centered at the approximate median of 33.4 years. Race was self-reported at baseline and categorized as white non-Hispanic, white Hispanic, black non-Hispanic, black Hispanic, and other. Self-reported highest level of education completed at baseline was categorized as grade 12 or less, college, and post-college graduate.
Using the interview data collected at each visit, the number of sexual partners since the prior visit was categorized as none, 1, 2 to 4, and 5 or more. The number of anal sexual partners with whom the participants reported always using a condom was subtracted from the total number of partners for each respective activity (receptive or insertive) to obtain the number of partners with whom the participant engaged in unprotected anal sex and was also categorized as none, 1, 2 to 4, and 5 or more. Any insertive oral/anal (rimming) sex reported at the visit was treated as a dichotomous variable. Current or previous visit use of poppers, cocaine, or ecstasy and any needle use also were dichotomized. Type of alcohol use was classified using frequency of drinking and average number of drinks the participant drank per day since the last visit. Binge drinking was defined as 5 or more drinks per occasion occurring at least monthly. Moderate to heavy drinking was at least weekly drinking of 3 to 4 drinks or drinking 5 or more drinks less than monthly. The remaining participants who had low to moderate or no drinking comprised the third group of alcohol use in this analysis. Likelihood of clinical depression was indicated by a score of 16 or higher using the Center for Epidemiologic Study of Depression symptom checklist.31 Except for age, race, education, cohort membership, and study center, other factors were all time varying (ie, they were updated at each visit).
Associations between the risk factors and time to seroconversion were estimated using the Cox proportional hazards regression model with time-dependent covariates.32 Because the model for the proportionality factor is log linear, effects on the relative hazard are multiplicative. An important feature of this model is that it can accommodate late entries (technically, left-truncated observations or subjects for whom data are available only if they survive to a known observation point).33 In our analysis, subjects having data during phase 2 were treated as late entries at their first available visit during this period. This prevented bias caused by events, particularly use of methamphetamine, occurring during the period between the 2 phases. Time-dependent covariates are treated in a similar fashion; available subjects are included in the risk set at each time point of observation according to their covariate status (eg, use of a given drug or not), which can change during different intervals of time. Predicted survival curves were calculated by using a nonparametric estimate of the baseline survival function and applying the parametric proportionality factors for the hazard corresponding to particular choices for the values of the covariates.
Data collected at the time of study enrollment were used to characterize the men who remained persistently seronegative and those who seroconverted during the follow-up period. The remaining variables were computed using questions asked at each visit. Each exposure variable was first evaluated in a univariate proportional hazards model. The multivariate approach used methamphetamine and popper use with all other covariates to test the most recent antecedent exposure to the outcome event. Interactions terms such as recent methamphetamine use × popper use, recent methamphetamine use × number of unprotected receptive anal sex partners, and popper use × number of unprotected receptive anal sex partners were also tested.
The baseline characteristics of the cohort are presented in Table 1. Men who seroconverted were similar to seronegative men in terms of age at baseline, race/ethnicity, and educational level. The proportions of seroconverters (22%-28%) and seronegative men (23%-28%) were similar across study centers. Seroconverters were more likely to have enrolled in the first phase of the cohort; to ever have used methamphetamine, poppers, cocaine, or ecstasy during any of the cohort visits; and to have practiced unprotected insertive and receptive anal sex before enrollment. However, the prevalences of having ever used poppers, cocaine, or ecstasy or needle use during any of the cohort visits among the seroconverters who recently used methamphetamine were 93%, 51%, 38%, and 16%, respectively, and were consistent throughout the study periods. Additionally, 98% and 94% of these seroconverters reported having had unprotected insertive or receptive anal sex at least once.
Table 2 presents univariate and multivariate predictors of HIV seroconversion. As expected, in this cohort of MSMs, the number of unprotected anal receptive sex partners was the primary risk factor, as demonstrated by an increasing “dose-response” relation to HIV seroconversion. Protective univariate predictors of HIV seroconversion were older age at baseline and late versus early cohort, whereas the number of sexual partners, number of unprotected insertive anal sex partners, insertive rimming, methamphetamine use, use of cocaine, ecstasy, or poppers, any needle use, and alcohol use were positive risk factors for HIV seroconversion.
After adjusting for all the covariates, there was an independent (1.46 [95% CI: 1.12 to 1.92]) increased relative hazard rate of HIV seroconversion associated with methamphetamine use and a 2-fold increased risk associated with popper use. The number of sexual partners, cocaine or ecstasy use, and race/ethnicity (black vs. non-Hispanic white) were also independent predictors of HIV seroconversion in the multivariate model. There was a significant increased risk with the increasing number of unprotected receptive anal sexual partners; however, this pattern was not evident with the number of unprotected insertive anal sexual partners. In the multivariate model, the covariation of polydrug use attenuated the increased risk of seroconversion of each drug assessed separately. The combined multiplicative effects of the relative hazards of methamphetamine, popper, and cocaine use exceeded by far the effect of any sociodemographic effect such as race/ethnicity.
Additionally, given the high prevalence of popper use and the number of unprotected receptive anal sex partners among the seroconverters who used methamphetamine, we focused our results on these phenomena in assessing the risk of HIV seroconversion. In doing so, we found that the joint relative hazard rate for methamphetamine and popper use versus no use was 3.05 (95% CI: 2.12 to 4.37; Fig. 1). Most notably, there was a significant dose-response relation between methamphetamine use and 1, 2 to 4, or 5+ unprotected receptive anal sex partners, with the joint relative hazard rates being 2.71 (95% CI: 1.81 to 4.04), 7.79 (95% CI: 5.17 to 11.74), and 13.57 (95% CI: 8.43 to 21.84), respectively (Fig. 2). These associations were not attenuated during later periods of study in the MACS (ie, the relative hazards were nearly identical when we studied phase 1 data alone). Also, sensitivity testing demonstrated that the relative hazard rate for methamphetamine use unadjusted for the sexual practice covariates was similar (2.10 [95% CI: 1.60 to 2.71]) to that of the full model. Likewise, the relative hazard rates for the number of unprotected insertive and receptive anal sex partners unadjusted for methamphetamine, popper, cocaine, ecstasy, and needle use did not change compared with those from the full model. None of the 3 interaction terms tested in the full multivariate model was statistically significant.
We found a significant association between methamphetamine use and HIV seroconversion after adjusting for other important risk factors in the MACS. In addition, we found independent effects of popper use and the number of unprotected receptive anal sex partners on HIV seroconversion. Individually, these increased relative hazards for HIV seroconversion were 1.46 and 2.10, respectively; however, among men who used both drugs, the multiplicative effect reached 3.05. There was a significant joint dose-response relation between methamphetamine use and the number of unprotected anal sex partners, ranging from 2.71 to 13.57. Although the joint relative hazard associated with an increasing number of unprotected anal sex partners is not surprising, the 2.71 increased risk of HIV seroconversion for recent methamphetamine users with only 1 unprotected receptive anal sex partner is noteworthy, particularly in this cohort, in which 26% of seronegative men reported only 1 unprotected receptive anal sex encounter since their last visit. It should be noted, however, that there were relatively few seroconverters who used only methamphetamine; more than 90% also used poppers, more than half used cocaine, and a significant proportion (16%) were also intravenous drug users in a cohort in which the overall rates of injection drug use fluctuates between 1% and 4%. Thus, it is difficult to attribute all or even most of the increased seroconversion risk seen among methamphetamine users to that drug alone.
We examined other potential covariates of HIV seroconversion in this cohort based on earlier findings.34,35 Alcohol use per se or heavy/binge drinking did not contribute significantly to increased rates of HIV seroconversion in the multivariate model; other drug groups tested were also nonsignificant, as was needle use. These findings are consistent with other investigators' earlier findings19,20 and extend those findings to a well-characterized, large, longitudinal, multisite cohort of MSM. We also found that late entry into the cohort (2001-2003) was an independent protective predictor against HIV seroconversion. This effect may possibly be attributable to lower viral loads in the population because of the availability of highly active antiretroviral therapy (HAART), the overall decreased prevalence of stimulant use during phase 2 as compared to phase 1, decreased rates of unprotected receptive anal sex, and the adoption of alternative forms of sexual risk reduction among the later cohort, or perhaps the existence of different viral or host genetic characteristics in the HAART era. Regardless of any differences in these factors between early and late entry into the cohort, the association of methamphetamine use and HIV seroconversion remained the same between the entry phases. Ultimately, this protective cohort effect may represent a spurious finding because of the relatively low number of seroconversions among late-entry cohort members.
Although the MACS participants were diverse in terms of age, socioeconomic status, race/ethnicity, geographic region, and baseline levels of HIV risk, they may not be nationally representative of MSM. The drug and sexual behavior questions attempted to identify aggregate levels of behavior within the previous 6 months, and hence were not linked directly to each other. Although this is a documented drawback of general behavioral data collection,36 because the seroconversion outcome is not misclassified, any underreporting of methamphetamine (including that attributable to different wording of questions over the cohort visits) and popper use or sexual practices or their concurrence by the highly vulnerable men would underestimate the risks presented here. The strongest advantage of these data is that all the exposures were ascertained consistently across the cohort and were obtained before the seroconversion outcome. Hence, the inclusion of a large number of previously uninfected but sexually active MSM has permitted us to examine a cumulative “natural history” of predictors of HIV seroconversion. Additional studies are needed to expand this natural history approach to the acute and chronic health effects of methamphetamine and other drug use among HIV-seropositive and HIV-seronegative men in the MACS.
Although we cannot conclude from this study the exact mechanism(s) for the increased likelihood of HIV seroconversion with methamphetamine or popper use associated with unprotected anal sex, it is likely to be a multifactorial process involving behavioral disinhibition, anal trauma, and selection bias for higher risk sexual partners. Drumright et al37 have postulated a conceptual model of methamphetamine and popper use among MSM, with the alteration of mental state,38,39 reduction of pain,40,41 enhancement of sexual function,40,42 and increased vasodilation43,44 leading to tissue damage or increased bleeding, reduction of condom use, or increased number of sexual partners or duration of sexual encounter as potential causative risk factors for HIV infection. Such a multifactorial model would help to explain the difficulty of intervening in sexual situations, because once methamphetamine and poppers are used, most inhibitions are discarded just as the other covariates of HIV transmission are brought into play.
The results from this large study have replicated those found in a small cluster of previous studies. Although 90% of the MACS participants have never used methamphetamine, the approximately 3-fold increased risk of HIV seroconversion associated with methamphetamine and popper use provides a clear rationale for improving access to information for methamphetamine- and popper-using MSM to reduce or eliminate sex with partners from drug-using networks characterized by high HIV seroprevalence rates resulting in a higher probability of HIV exposure.
To date, the methamphetamine-HIV concerns have been primarily focused within the white gay/bisexual community; however, swift community-based public health prevention strategies with follow-up assessment of their efficacy need to be developed and implemented before methamphetamine use captures other vulnerable populations.45,46 Ultimately, our task is to move beyond descriptive research of the methamphetamine-seroconversion relationship to understand better the dynamics of behavior and vulnerabilities to drug abuse and addiction and their intertwining with the HIV epidemic.2 It is time to address these issues in designing primary and secondary interventions to disrupt these complex and multifactorial links.
The MACS includes the following: (Baltimore) The Johns Hopkins University Bloomberg School of Public Health: Joseph B. Margolick (Principal Investigator), Haroutune Armenian, Barbara Crain, Adrian Dobs, Homayoon Farzadegan, Joel Gallant, John Hylton, Lisette Johnson, Shenghan Lai, Justin McArthur, Ned Sacktor, Ola Selnes, James Shepard, and Chloe Thio; (Washington, DC) Michael W. Plankey; (Chicago) Howard Brown Health Center, Feinberg School of Medicine, Northwestern University and Cook County Bureau of Health Services: John P. Phair (Principal Investigator), Joan S. Chmiel (Coprincipal Investigator), Sheila Badri, Bruce Cohen, Craig Conover, Maurice O'Gorman, David G. Ostrow, Frank Palella, Daina Variakojis, and Steven M. Wolinsky; (Los Angeles) University of California, Los Angeles Schools of Public Health and Medicine: Roger Detels (Principal Investigator), Barbara R. Visscher (Coprincipal Investigator), Aaron Aronow, Robert Bolan, Elizabeth Breen, Anthony Butch, Thomas Coates, Rita Effros, John Fahey, Beth Jamieson, Otoniel Martínez-Maza, Eric N. Miller, John Oishi, James A. Peck, Paul Satz, Harry Vinters, Dorothy Wiley, Mallory Witt, Otto Yang, Stephen Young, and Zuo Feng Zhang; (Pittsburgh) University of Pittsburgh, Graduate School of Public Health: Charles R. Rinaldo (Principal Investigator), Lawrence Kingsley (Coprincipal Investigator), James T. Becker, Robert L. Cook, Robert W. Evans, John Mellors, Sharon Riddler, Anthony Silvestre, and Ron Stall; Data Coordinating Center, The Johns Hopkins University Bloomberg School of Public Health: Lisa P. Jacobson (Principal Investigator), Alvaro Muñoz (Coprincipal Investigator), Haitao Chu, Stephen R. Cole, Christopher Cox, Stephen J. Gange, Janet Schollenberger, Eric C. Seaberg, and Sol Su; National Institutes of Health, National Institute of Allergy and Infectious Diseases: Robin E. Huebner; National Cancer Institute: Geraldina Dominguez; and National Heart, Lung, and Blood Institute: Cheryl McDonald. The Web site is located at http://www.statepi.jhsph.edu/macs/macs.html.
2. 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
3. Gorman M, Halkitis P. Methamphetamine and club drug use and HIV. Focus
5. Halkitis PN, Parsons JT, Wilton L. Barebacking among gay and bisexual men in New York City: explanations for the emergence of intentional unsafe behavior. Arch Sex Behav
6. Morin SF, Steward WT, Charlebois ED, et al. Predicting HIV transmission risk among HIV-infected men who have sex with men: findings from the Healthy Living Project. J Acquir Immune Defic Syndr
7. Bimbi DS. Factors related to barebacking, or intentional unprotected anal sex, among HIV-positive and HIV-negative gay and bisexual men. Presented at: International AIDS Conference, Bangkok, Thailand, 2004.
8. Colfax G, Coates TJ, Husnik MJ, et al. Longitudinal patterns of methamphetamine, popper (amyl nitrite), and cocaine use and high-risk sexual behavior among a cohort of San Francisco men who have sex with men. J Urban Health
. 2005;82(Suppl 1):i62-i70.
9. Crosby GM, Stall RD, Paul JP, et al. Condom use among gay/bisexual male substance abusers using the timeline follow-back method. Addict Behav
10. Dolezal C, Meyer-Bahlburg HFL, Remien RH, et al. Substance use during sex and sensation seeking as predictors of sexual risk behavior among HIV+ and HIV− gay men. AIDS Behav
11. Halkitis PN, Parsons JT, Stirratt MJ. A double epidemic: crystal methamphetamine drug use in relation to HIV transmission among gay men. J Homosex
12. Halkitis PN, Green KA, Mourgues P. Longitudinal investigation of methamphetamine use among gay and bisexual men in New York City: findings from Project BUMPS. J Urban Health
. 2005;82(Suppl 1):i18-i25.
13. Molitor F, Truax SR, Ruiz JD, et al. Association of methamphetamine use during sex with risky sexual behaviors and HIV infection among non-injection drug users. West J Med
14. Pendo M, DeCarlo P. CAPS Fact Sheet: How Do Club Drugs Impact HIV Prevention?
San Francisco, CA: University of California, San Francisco; 2004.
15. Rekart ML, Marchand R. The “Sex Now Survey”: HIV risk and prevention for gay men in Vancouver. Presented at: International AIDS Conference, Bangkok, Thailand, 2004.
16. Stopka TJ, Webb DS. Gay men/MSM, methamphetamine use and HIV: the California perspective. Presented at: National HIV Prevention Conference, Atlanta, Georgia, 2003.
17. Swanson J, Cooper A. Dangerous liaison: club drug use and HIV/AIDS. IAPAC Mon
18. Ven PVD, Campbell D, Kippax S, et al. Factors associated with unprotected anal intercourse in gay men's casual partnerships in Sydney, Australia. AIDS Care
19. Chesney MA, Barrett DC, Stall R. Histories of substance use and risk behavior: precursors to HIV seroconversion in homosexual men. Am J Public Health
20. Buchacz K, McFarland W, Kellogg TA, et al. Amphetamine use is associated with increased HIV incidence among men who have sex with men in San Francisco. AIDS
21. McKirnan DJ, Vanable PA, Ostrow DG, et al. Expectancies of sexual “escape” and sexual risk among drug and alcohol-involved gay and bisexual men. J Subst Abuse
22. Mansergh G, Colfax GN, Marks G, et al. The Circuit Party Men's Health Survey: findings and implications for gay and bisexual men. Am J Public Health
23. Parsons JT, Bimbi DS. Intentional unprotected anal intercourse among sex who have sex with men: barebacking-from behavior to identity. AIDS Behav
. June 15, 2006 (e-pub ahead of print).
24. Mansergh G, Marks G, Colfax GN, et al. “Barebacking” in a diverse sample of men who have sex with men. AIDS
25. Mansergh G, Shouse RL, Marks G, et al. Methamphetamine and sildenafil (Viagra) use are linked to unprotected receptive and insertive anal sex, respectively, in a sample of men who have sex with men. Sex Transm Infect
26. Ostrow DG, Beltran ED, Joseph JG, et al. Recreational drugs and sexual behavior in the Chicago MACS/CCS cohort of homosexually active men. Chicago Multicenter AIDS Cohort Study (MACS)/Coping and Change Study. J Subst Abuse
27. Patterson TL, Semple SJ, Zians JK, et al. Methamphetamine-using HIV-positive men who have sex with men: correlates of polydrug use. J Urban Health
. 2005;82(Suppl 1):i120-i126.
28. Kaslow RA, Ostrow DG, Detels R, et al. The Multicenter AIDS Cohort Study: rationale, organization, and selected characteristics of the participants. Am J Epidemiol
29. Dudley J, Jin S, Hoover D, et al. The Multicenter AIDS Cohort Study: retention after 9½ years. Am J Epidemiol
30. Gribble JN, Miller HG, Cooley PC, et al. The impact of T-ACASI interviewing on reported drug use among men who have sex with men. Subst Use Misuse
31. Radloff LS, Rae DS. Susceptibility and precipitating factors in depression: sex differences and similarities. J Abnorm Psychol
32. Cox DR, Oakes R. Analysis of Survival Data
. London: Chapman and Hall; 1984.
33. Klein JP, Moeschberger ML. Survival Analysis: Techniques for Censored and Ttruncated Data
. 2nd ed. New York: Springer-Verlag; 2003.
34. Ostrow DG, DiFranceisco WJ, Chmiel JS, et al. A case-control study of human immunodeficiency virus type 1 seroconversion and risk-related behaviors in the Chicago MACS/CCS Cohort, 1984-1992. Multicenter AIDS Cohort Study. Coping and Change Study. Am J Epidemiol
35. Detels R, English P, Visscher BR, et al. Seroconversion, sexual activity, and condom use among 2915 HIV seronegative men followed for up to 2 years. J Acquir Immune Defic Syndr
36. Leigh BC, Stall R. Substance use and risky sexual behavior for exposure to HIV. Issues in methodology, interpretation, and prevention. Am Psychol
37. Drumright LN, Patterson TL, Strathdee SA. Club drugs as causal risk factors for HIV acquisition among men who have sex with men: a review. Subst Use Misuse
38. Murray JB. Psychophysiological aspects of amphetamine-methamphetamine abuse. J Psychol
39. Nordahl TE, Salo R, Leamon M. Neuropsychological effects of chronic methamphetamine use on neurotransmitters and cognition: a review. J Neuropsychiatry Clin Neurosci
40. Green AI. “Chem friendly”: the institutional basis of “club drug” use in a sample of urban gay men. Deviant Behav
41. Romanelli F, Smith KM, Thornton AC, et al. Poppers: epidemiology and clinical management of inhaled nitrite abuse. Pharmacotherapy
42. Kurtz SP. Post-circuit blues: motivations and consequences of crystal meth use among gay men in Miami. AIDS Behav
43. French RS, Power R. Self-reported effects of alkyl nitrate use: a qualitative study amongst targeted groups. Addict Res
44. Mayer KH. Medical consequences of the inhalation of volatile nitrites. In: Ostrow DG, Sandholzer TA, Felman YM, eds. Sexually Transmitted Diseases in Homosexual Men: Diagnosis, Treatment and Research
. New York: Plenum Press; 1983:237-242.
45. Jacobs A. Battling HIV where sex meets crystal meth. New York Times
. February 21, 2006.
46. Krawczyk C. Methamphetamine use and HIV risk behaviors among heterosexual men-preliminary results from five Northern California counties, December 2001-November 2003. MMWR Morb Mort Wkly Rep
© 2007 Lippincott Williams & Wilkins, Inc.