Inaccuracies in Self-Reports and Urinalysis Tests: Impacts on Monitoring Marijuana Use Trends among Arrestees
Golub, Andrew, Liberty, Hilary James, Johnson, Bruce D., Journal of Drug Issues
Measured trends in drug use can potentially reflect changes in drug use, changes in the accuracy of the measurement instrument, or both. This paper compares marijuana use trends from 1987 to 2001 using self-report and urinalysis data from arrestees interviewed at 23 sites served by the Arrestee Drug Abuse Monitoring (ADAM) Program. Overall, 60% of the variation in reported use reflected changes in detected use. Most notably, reported and detected use suggested different dynamics to the increase in marijuana use during the 1990s. The growth in detected use started later, increased more, and lasted longer. Several factors appear to have clearly contributed to this divergence between the measures: the percentage of marijuana users that disclosed their activity changed over time, the accuracy of ADAM's urinalysis test increased with time, and the percentage of infrequent users changed over time (urinalysis tests are less likely to detect infrequent users). The paper concludes with recommendations for the careful analysis of marijuana use trends using self-report data, biological data, or both. Trends in cocaine, crack, heroin, and methamphetamine are also considered.
Our ability to identify major drug use trends was greatly advanced in the 1970s with the establishment of ongoing drug use surveys. The National Household Survey on Drug Abuse (NHSDA) was established in 1971 as "the primary source of information on the prevalence and incidence of illicit drugs, alcohol, and tobacco use in the civilian non-institutionalized population aged 12 or older in the United States" (SAMHSA, 2003, p. 1). This program collected drug use information from a nationally representative sample every two or three years at first and annually since 1990. In 2003, the name of the program was changed to the National Survey on Drug Use and Health (NSDUH). The Monitoring the Future (MTF) program was established in 1975 and has provided annual estimates of drug use from a nationally representative sample of high school seniors and since 1991 for eighth and tenth graders too (Johnston, O'Malley, & Bachman, 2003). In 1987, the National Institute of Justice (NIJ) established the Arrestee Drug Abuse Monitoring (ADAM, formerly Drug Use Forecasting or DUF) Program to monitor drug use among arrestees (ADAM, 2003; Golub & Johnson, 2002). Given the drugs-crime nexus, ADAM provides information about a particularly high-risk population of great interest to criminal justice agencies as well as a wide variety of other constituencies concerned with drug-related problems. Since January 29, 2004, the ADAM program has been on hiatus as a Federal cost-saving measure (National Institute of Justice [NIJ], 2004).
The U.S. Office of National Drug Control Policy (ONDCP, 2004) uses data from the NSDUH, MTF, and ADAM surveys as well as several other data collection programs to closely monitor prevailing drug use trends over time. Moreover, ONDCP (2004) uses the MTF and NSDUH surveys to measure its progress towards meeting its goals of reduced drug use. However, the usefulness of selfreports of drug use (hereafter reported use) is limited to the extent that respondents provide accurate information. Prior research indicates that sometimes as few as 50% of recent users accurately disclose their activity and that disclosure can vary substantially across drugs, interview settings, race/ethnicity, gender, age, and other demographic characteristics (for reviews see Golub, Johnson, Taylor, & Liberty, 2002; Harrison, 1997; Magura & Kang, 1996). This variation in disclosure calls into question the validity of comparisons across groups. For example, a finding that youths reported less drug use than young adults could reflect lower rates of use among youths. However, the same finding could reflect a lower willingness to report use by youths.
ADAM data have been an outstanding resource for studying the validity of reported use because ADAM collects urine samples to objectively test for drugs (hereafter detected use). …