Our study is the first statewide description of MA use among newly diagnosed HIV-positive young men in the Southeastern US, and demonstrates an increasing trend in MA use among young men diagnosed with HIV in NC from 2000 to 2005. Although these cross-sectional surveillance data cannot demonstrate causation, certain findings have implications for public health policy in NC and other states in the region.

MA users in this sample are similar to those described in samples from major cities in previous years. They were often self-identified as MSM[21], [22] and tended to engage in more HIV-associated risk behaviors than their non-using counterparts [23], [24], [25], [26]. They frequented clubs or bars[21] and had anonymous sex partners[27] – often met in a bar, club, or online. Internet sex-seeking was prominent among both users and non-users.

In particular, two findings were noteworthy: coinfection with early syphilis and the HIV testing behavior of MA users. NC is currently experiencing a major syphilis outbreak, and MSM are disproportionately impacted [28]. Studies from urban settings demonstrate MA use as a risk factor for incident syphilis infection[29], [30], and previous work in NC suggests increased sexual risk-taking behavior among stimulant users [31].

Frequent HIV testers may regard this as a method of primary prevention; while acknowledging their high-risk sexual activities, they may use a recent negative HIV test as evidence of their “safety” as a partner for unprotected intercourse [32]. Currently, counselors at HIV testing centers (including sexually transmitted infection (STI) clinics) in NC record very limited information about sexual and injection drug HIV risk behaviors on surveillance forms. Our findings underscore the need to expand the inquiry and to more broadly discuss and record non-injection drug use, the circumstances and environment in which drugs are used, and whether or not sexual activity occurs while intoxicated. More comprehensive drug-use assessments at STI clinics could greatly improve our understanding of the expansion of new drugs like MA in the absence of a robust regional surveillance system for drug use.

While relatively few men in this cohort acknowledged MA use in the period prior to their HIV diagnosis, the proportion who did increased significantly over the five-year study period – the same period over which MA infiltration into NC began in earnest. US DEA annual “busts” of clandestine MA laboratories in NC grew from 14 in 2000 to 322 in 2005, but these numbers do not account for higher-purity “crystal” MA imported via established cocaine trafficking routes from Mexico [12]. Thus, these users with newly diagnosed HIV infection may be sentinels for what we will see in the future, as MA use expands in our state.

The inverse association of black race with MA use is noteworthy in our study, and echoes qualitative data collected elsewhere in the Southeastern US. Community and cultural norms among black drug users favor the use of crack or powder cocaine and fuel a strong suspicion of MA as being an impure and unpredictable drug [33]. This trend should not provide a false sense of security, however; although our total number of MA users was small, a third of them were black.

Travel that included sexual activity was associated with MA use in our analyses. Since almost 75% of our sample self-identified as either MSM or having sex with both men and women, many of these young men may intentionally explore their sexual identities away from the close-knit, small communities in which they live. Studies of MSM in San Francisco demonstrated increased HIV-associated risk behavior when traveling away from home for weekend-long dance (“circuit”) parties [34]. In the case of MA, drug availability may also be an issue; the small numbers of users identified in our sample suggests that the drug had yet to widely disseminate among young men in NC by 2005.

Our results have several limitations. Only a small percentage of the entire cohort were MA users. Data are self-reported by individuals being interviewed by an officer of the NC-DHHS, and thus social desirability bias has likely resulted in underestimation of the prevalence of both drug use and sexual risk behavior. Data collection bias may have improved the capture of information on MA use toward the end of the period, as awareness of expanded availability and use of the drug diffused through to more DIS officers. Additionally, ascertainment bias may have been introduced through the use of chart abstraction, although those reviewing the charts were trained for the task. Finally, our data pertain only to men aged 18 to 30, so we cannot characterize MA use or associated factors in men over age 30 in NC.

In summary, the use of MA among young men newly diagnosed with HIV underscores the fact that MA use exists in the Southeast US, and may be expanding among groups at high risk for HIV acquisition and transmission. HIV risk-reduction messages for MA users should be directed at young, college-aged men, including MSM – with special attention to any point of intersection with the healthcare system that includes STI or HIV testing.