The purpose of the study was to provide information about markets, distribution, and use of methamphetamine in New York City, both inside and outside of the MSM (men who have sex with men)/gay community. The study had four specific aims. (1) Pilot a research design using mixed qualitative/quantitative data collection methods in New York City methamphetamine markets with participants recruited using Respondent Driven Sampling. (2) Document, describe, and analyze the demographics and methamphetamine abuse behaviors of a sample of methamphetamine market participants. (3) Document, describe, and analyze the social networks of New York City methamphetamine market participants as users, buyers, and sellers of methamphetamine. (4) Document, describe, and analyze the market behaviors of New York City methamphetamine market participants as users, buyers, and sellers of methamphetamine.
Researchers piloted a version of Respondent Driven Sampling (RDS) to recruit different types of methamphetamine users. RDS is a chain-referral recruiting method where participants recruit additional participants in recruitment chains. To better understand how social networks among methamphetamine users and distributors are structured, function, and evolve over time, the researchers collected network data from respondents to understand the complexity of linkages in user and distribution networks. The study used a mixed data-collection interview technique that captured both quantitative and qualitative data in a single interview; the interviewer conducted an open-ended, but structured interview, entering data on a computer database form, while simultaneously employing a stereo recorder to record the entire interview for subsequent selective transcription and analysis. After a period of formative research, the data collection phase of the study used RDS to recruit 132 methamphetamine market participants. Each study participant did a roughly 1-2 hour structured interview that recorded participant demographics (Basic Roster Data and Roster Data; Datasets 1 and 2), enumerated and described the members of their drug use and distribution networks including drug buying, selling, trading, and in-kind transaction partners (Ego Network Data and Respondent Driven Sampling Network Data; Datasets 3 and 4), and described drug use patterns and market participation (Methamphetamine Obtaining Data, Providing for Others Data, Criminal Justice Data, and Consumer Data; Datasets 5-8). Each participant was invited to recruit three additional eligible participants according to the RDS protocol; they were given three numbered coupons that they were instructed to pass along to other people they know who currently participate in retail methamphetamine markets, as buyers, sellers, and/or users. Participants were instructed to return one month after their initial interview date to collect recruitment incentives and complete a follow-up interview that focuses on changes in patterns of use, distribution or network membership since the first interview. To determine eligibility, the researchers developed a screener that asked potential participants about drug use and preparation techniques, folkloric beliefs related to methamphetamine use, and market conditions and prices.
The study recruited 132 current methamphetamine users and distributors (persons who had used or sold methamphetamine in the previous 30 days) over a 33-month data collection period, using Respondent Driven Sampling. Respondent Driven Sampling (RDS) is a methodology that is used to recruit samples of hard-to-reach groups by taking advantage of intragroup social connections to build a sample pool. In order to initiate the RDS sample chains, the research team placed 33 ads on craigslist.org from February 2008 (to recruit Formative Research interviewees) through September 2009 (to recruit RDS seeds to initiate new recruitment chains). This drew many phone calls to the toll-free number the research team set up and 105 emails from potential participants. The researchers scheduled more than 300 appointments over the data collection period, with many no-shows and repeated re-schedulings around potential participants' often busy schedules. The research team also regularly placed flyers on bulletin boards at local MSM (men who have sex with men)-oriented organizations, although no participants were recruited as a result of these flyers. Seeds were also recruited through a local harm-reduction program's Crystal Meth Support Group.
Following the initial interview, initial "seeds" and subsequent waves of study participants were given three numbered coupons that they were instructed to pass along to other people they know who currently participate in retail methamphetamine markets, as buyers, sellers, and/or users. The numbers on these coupons allowed the research team to prevent duplication, to identify who recruited each participant and to keep track of subsequent recruitment patterns using the RDS Coupon Manager software. In order to reduce the likelihood that too many respondents would come from a small number of well-connected people, the researchers limited the recruitment potential of each participant with the rationing of coupons. Researchers also included a 30 day expiration date on the coupons in an attempt to ensure a reasonable pace of sample recruitment. Brief recruiter training sessions were conducted with the seeds and each subsequent eligible respondent, explaining: recruitment criteria; incentives for interviews and recruiting ($50 and $20 respectively); how recruits should contact the project; the time limit and sample size limit on recruiting; and discussion of appropriate recruiting techniques to avoid coercion and ensure participation was voluntary. When necessary, researchers selected new seeds because recruitment chains were slow to develop; many participants did not recruit. Of the 132 subjects interviewed, 28 percent (37 individuals) were seeds in the RDS process. Of these 37 seeds, 38 percent (14 individuals) produced referrals. More than seventy persons attempted to participate in the study who were not in fact eligible, as far as could be determined after screening. For a detailed discussion of RDS, please refer to the final report for this study (Wendel et al., 2011; NCJ 236122).
The Basic Demographic Data (Dataset 1) includes information on 157 initial participants, some of whom were determined to be ineligible for the study, the Ego Network Data (Dataset 3) includes information on 466 ego network relationships, the Providing for Others Data (Dataset 6) includes information on 111 participants who self-identified as having provided methamphetamine to others, and the Criminal Justice Data (Dataset 7) includes information on 132 subjects but has 133 cases due to a case that spans two data files due to a computer crash in mid-interview. The Roster Data (Dataset 2), Respondent Driven Sampling Network Data (Dataset 4), Methamphetamine Obtaining Data (Dataset 5), and Consumer Data (Dataset 8) include information on the 132 methamphetamine users and distributers.
All persons who lived or worked in New York City and used, bought, or sold methamphetamine in the 30 days prior to the interview.
In-person interviews with methamphetamine market participants in New York City
Each dataset has a mixture of quantitative and qualitative (open-ended response) data, as well as derived/re-coded variables computed by the principal investigators. All datasets include information on the respondent's sexual orientation, and most include additional demographic variables.
- The Basic Demographic Data (Dataset 1) has 21 variables providing identifying information for the respondents, including gender, sexual orientation, race/ethnicity, and physical descriptions.
- The Roster Data (Dataset 2) has 15 variables with additional demographic information for the respondents including age, sexual orientation, ethnicity, birthplace, and HIV + status (although respondents were not asked about HIV + status, some respondents volunteered this information).
- The Ego Network Data (Dataset 3) has 32 variables about the respondent's ego network, and on the respondent's relationship to the methamphetamine market. Respondents were asked to provide data on up to five members of their ego network, including physical descriptions, the network member's relationship to the market, and relationship to the respondent. Other variables include who the respondent would call to find a new source of methamphetamines, and demographic data.
- The Respondent Driven Sampling Network Data (Dataset 4) includes 45 variables gathered for the Respondent Driven Sampling. These describe the respondent's network, based on people whose cell phone number the respondent know. Variables include number of people who use and/or provide methamphetamine, gender, race and sexual orientation of people in the network, how many provide and how many only use, the respondent's frequency of methamphetamine use, and the relationship between methamphetamine use and sex for the respondent.
- The Methamphetamine Obtaining Data (Dataset 5) has 63 variables regarding the participant's participation in the methamphetamine market in obtaining methamphetamine over the last year. There are a series of variables for each of the three most recent episodes of obtaining methamphetamine. Variables regarding the recent episodes include source, cost, payment type, quality and packaging, purpose of purchase (use or distribution), relationship to the source, and descriptions of each transaction.
- The Providing for Others Data (Dataset 6) has 58 variables describing the participant's role in the methamphetamine market as a provider over the last year. There are a series of variables for each of the three most recent episodes of providing methamphetamine. Variables include the quantity provided, payment, relationship with the receiver, and details of the transaction, including location, delivery method, quality and packaging of the product, whether there were any problems with the transaction.
- The Criminal Justice Data (Dataset 7) has 69 variables about the respondent's experience with the criminal justice system and experience with violence in the course of their participation in methamphetamine markets. Criminal justice variables include arrests for possession and distribution of methamphetamine, non-methamphetamine drug related arrests, whether police have ever confiscated methamphetamine, and whether police and/or other community community factors prevented the respondent from buying methamphetamine. Other variables include whether the respondent owns a weapon and if they carry it when participating in the methamphetamine market, and whether they have experienced violence while participating in the market. There are also demographic variables, including gender, sexual orientation, and ethnicity.
- The Consumer Data (Dataset 8) has 74 variables regarding consumption of methamphetamines. These include mode of ingestion, how the respondent uses methamphetamine (whether they have sex, how long they use it for, how they manage their use, who they use it with), why the respondent uses methamphetamine and whether they use other drugs, dependency status, and access to dealers; demographic variables are also included.