Roughly one out of every two arrestees in Anchorage tests positive for recent drug use, and marijuana seems to be the illicit "drug of choice" among arrestees in Anchorage—particularly arrestees under the age of 30— according to several years of data assembled under the Arrestee Drug Abuse Monitoring (ADAM) program.
Since 1999, the Justice Center at the University of Alaska-Anchorage has served as the site contractor for the ADAM program. ADAM is a national, multi-site drug monitoring program funded by the National Institute of Justice (NIJ) that measures the extent and nature of alcohol and drug use among those who have been recently arrested. ADAM collects information by conducting in-depth interviews with recent arrestees and collecting urine samples.
ADAM possesses several unique characteristics that distinguish it from other alcohol and drug monitoring data collection systems. First, while most alcohol and drug research relies almost solely on respondents' self-reported behavior, ADAM incorporates both arrestee self-reported behavior and an objective measure of alcohol and drug use: the presence of alcohol and drugs in arrestees' urine. The ADAM program uses the Enzyme Multiplied Immunoassay Testing (EMIT) system, one of the most accurate drug testing methods available, to screen for the presence of drugs and alcohol in urine. In combination, arrestee self-reports and urine samples provide an unparalleled breadth and depth of knowledge about alcohol and drug use.
Second, unlike other drug use monitoring efforts, information from ADAM is based on local rather than national estimates. National surveys such as the National Household Survey on Drug Abuse provide extensive information on state and national drug use trends, but are not designed to assess the nature of alcohol and drug use for smaller social collectivities such as counties, cities or neighborhoods. Criminologists, policy makers, and lay persons alike have long known that alcohol and drug use behaviors are not randomly distributed in society, but instead display wide variation across different localities. By providing contextualized knowledge of drug and alcohol use, ADAM helps local public policy makers to develop programs and policy especially relevant to local communities.
Third, ADAM focuses on a group of people known to be at high-risk for substance abuse and addiction: recent arrestees. Individuals brought within the scope of the criminal justice system demonstrate rates of substance abuse and dependence greater than the general population; yet, prior to the implementation of the ADAM program, only limited knowledge existed about the nature and extent of alcohol and drug use among those incarcerated in correctional institutions in the United States—especially jails. As the number of persons incarcerated in jails and prisons—now over two million—continues to rise, it becomes increasingly important to collect more extensive information about alcohol and drug use behaviors for those entering the criminal justice system.
Finally, ADAM uses probability-based sampling techniques that allow researchers to provide accurate and representative estimates of alcohol and drug use among arrestees. The significance of using probability sampling methods for ADAM is that local policy makers and their constituents can be assured that the data gathered and reported by the ADAM program truly represent the underlying arrestee population.
Early in its history ADAM relied on convenience samples of arrestees. (Prior to 1997, ADAM was known as the Drug Use Forecasting (DUF) program.) With no probability-based sampling protocol, ADAM sites could not determine whether deviations in drug use trends were significant or simply random variations to be expected with any phenomenon. In other words, if a site found a 10 percent increase in the proportion of arrestees who tested positive for cocaine from one time period to the next, it could not be reliably determined if that change was truly an increase or simply a random perturbation in the data. This began to change in 1997 when the ADAM program underwent an extensive re-design.
By 2000 each ADAM research site had an explicit sampling plan in place detailing the selection of male arrestees into the ADAM sample. (At present, female arrestees are not selected using a probability-based sampling plan. Convenience sampling is used to select female arrestees.) To begin, arrestees can only be interviewed within 48 hours of their arrest. (In order to maximize the reliability of urinalysis results, those who have been incarcerated for more than 48 hours are not included in ADAM sample.) The second requirement for inclusion is that arrestees must be jailed for committing a new offense. Individuals held for a change of venue or awaiting transport to another facility, transfers from another facility, those awaiting extradition, and court-ordered remands following trial are not eligible for the study. In addition, federal detainees are not included in the ADAM sample. Third, juveniles under the age of 18, even if they are charged as adults for their current offense, are not eligible for the study.
Once eligibility is determined, arrestees are grouped according to the time they were formally booked into jail. Individuals are selected by sampling at designated intervals, using a random starting point. Nationally certified interviewers then ask the selected arrestees to participate in the ADAM study. Participation in ADAM is completely voluntary; arrestees can refuse outright, withdraw from the interview once it has begun, or refuse to answer any particular questions along the way. Upon completion of the interview, the arrestee is asked to provide a urine sample for analysis by an independent laboratory. As with the interview, the arrestee has the right to refuse to submit a specimen.
All the information provided by the arrestee, including the urine sample, is confidential and subject to strict guidelines specified by federal and state laws, as well as the local Institutional Review Board. All identifying information is removed from the questionnaire and urine specimen to protect individual identities; by law, individual results cannot, be made available to any criminal justice official or agency.
Due to the sampling strategy employed by ADAM-Anchorage, coupled with respondents' right to refuse to answer questions or provide a urine sample, the urine assay data reported comes from only a relatively small sample of all eligible participants. To illustrate, for the three years reported here (1999-2001) there were a total of 3,259 male arrestees and 1,005 female arrestees eligible for the study. Out of these respective respondent pools, less than 50 percent of the men and less than 40 percent of females provided a urine sample to ADAM-Anchorage researchers. Table 1 summarizes the number of participants at each stage of the ADAM research process.
Despite the relatively small proportion of respondents that actually provide urine for analysis, both the male and female samples are highly representative of the population of Anchorage arrestees in general. In a separate analysis conducted by Myrstol and Langworthy (in press), it was found that the male sample evidenced no significant sample bias across age, race, offense seriousness, offense type, or area of arrestee residence. Similarly, the study discovered that even for the female sample, which was not selected using a formal sampling plan, there was no discernable sample bias present. Thus, the data reported are generalizable to the entire Anchorage male and female arrestee population.
Patterns and Trends of Drug Use Among Anchorage Arrestees
Table 2 and Table 3 display the percent of male and female respondents providing a urine sample who tested positive for marijuana, cocaine, opiates and methamphetamine by age and race/ethnicity for the years 1999, 2000, and 2001. (Marijuana, cocaine, opiates and methamphetamine comprise 4 out of the 5 drugs designated by the National Institute of Drug Abuse's primary substances of abuse, commonly referred to as the "NIDA-5 drugs." The fifth drug is PCP, which is not discussed here because of its extremely low prevalence in Anchorage.) In addition, each table also presents the percentage of respondents who tested positive for any of the ten drugs screened by ADAM for the same years. (ADAM screens for: amphetamines, with gas chromatography/mass spectrometry confirmation for methamphetamine; barbiturates; benzodiazepines; cocaine; detla-9-detrahydrocan-nabinol, i.e., marijuana; opiates; methadone; methaqualone; phencyclidine, i.e., PCP; and propoxyphene. At present, EMIT screening is not performed for alcohol.) Data for 1999 are included in the table but highlighted in grey to distinguish data collected prior to the ADAM program redesign implemented in 2000.
Comparison across the three years shows remarkable stability in the percentage of arrestees who test positive for marijuana, cocaine, opiates and methamphetamine, as well as the percent testing positive for any of the ten drugs screened by ADAM. This pattern holds, even when the data are broken down into their respective collection quarters. The trend lines for male arrestees in Figure 1 illustrate this graphically.
For male arrestees, marijuana was the most prevalent drug detected in urinalysis, followed by cocaine, opiates and methamphetamine. Similarly, for females the rate of positive test results for marijuana was greater, on average, than for cocaine, opiates and methamphetamine, respectively. Thus, the data suggest, at least tentatively, that marijuana is the illicit "drug of choice" among adult arrestees in Anchorage. Such a general statement comes with many qualifications, one of which is evident in Table 2 and Table 3. When results are broken down according to age, we see that the percentage of arrestees testing positive for marijuana in the urinalysis is much greater for those younger than 30 years of age. The rate of positive test results drops off drastically for arrestees aged 31 and over. Considering this pattern then, it might be more accurate to describe marijuana as the "drug of choice" for young adult arrestees in Anchorage. Finally, there were no definitive patterns of use across categories of race/ethnicity detected in the ADAM data; that is, no one cultural group displayed a consistent or significantly higher rate of marijuana use, as detected through EMIT screening.
Cocaine, while not detected as often as marijuana among male and female arrestees, was found in the urine of about one-quarter of the male sample and about one-fifth of the female sample. Notably, comparison of the marijuana results and cocaine results shows that the prevalence of cocaine use among these samples of arrestees increases at roughly the same age that the prevalence of marijuana use declines. These preliminary results hint that both marijuana and cocaine use patterns are at least somewhat age-dependent. In addition to the divergence in patterns of use across age categories, another difference between the cocaine and marijuana results was found in comparisons across racial/ethnic groups. For both males and females, arrestees of African descent were significantly more likely than any other ethnic group to test positive for cocaine. (The small number of Hispanic females—n=1; n=4; n=4—and Asian females—n=4; n=4; n=2—renders their prevalence estimates highly erratic and susceptible to drastic variation with the addition/subtraction of only one or two positive test results. Therefore, for comparative purposes, the percentage of Hispanic and Asian females who tested positive for drug use were excluded.) Conversely, Alaska Native and American Indian arrestees were found to be the least likely of the ethnic groups studied to test positive for cocaine. It should also be noted that while the cocaine screen detects the recent use of both powder cocaine and rock cocaine, it does not differentiate between the two forms of the drug; therefore the method of ingestion is not discernable from the assay data.
Urinalysis results for opiates, methamphetamine and PCP (not presented here) converge on one point: among the Anchorage arrestee population, there is a remarkably low prevalence of opiate, methamphetamine and PCP use. While it does appear that female arrestees might have a higher rate of opiate use, the number of positive test results for both men and women booked into Anchorage jails is so low that making such distinctions is somewhat premature and dubious.
The final column in each table provides the urinalysis results for male and female arrestees who tested positive for any of the 10 drugs screened in the ADAM program. On the whole, female arrestees appear to have a greater probability than males of testing positive for any of 10 drugs. Moreover, there is evidence that women brought to Anchorage jails in 2001 were more likely to test positive for any of the 10 drugs tested than in 2000, especially white and Native women under the age of 30. In the aggregate, the pattern of drug use among males across the 10 drugs tested did not change noticeably across age or race/ethnicity groups. Despite the subtle differences in drug use patterns between women and men arrested in Anchorage, female and male arrestees share one important drug use characteristic: roughly one out of every two arrestee intakes will have used at least one of the ten drugs screened by ADAM within days of their incarceration.
Brad Myrstol is a research associate with the Justice Center.