Endnotes

1. The concept of patterns or pathways of problem behavior is widely discussed in the literature (Newcomb and Bentler, 1990; Cohen et al., 1990; Elliott, Huizinga, and Menard, 1989; Kandel and Yamaguchi, 1993; Kandel and Logan, 1984; Loeber and Hay, 1997; Loeber et al., 1991, 1993; Moffitt, 1993a, 1993b; Rutter, 1996; Yamaguchi and Kandel, 1984a, 1984b).

2. A large body of research has questioned the validity of self-reported drug use data from adult arrestees (Golub et al., 2002; DeJong and Wish, 2000; Gray and Wish, 1999; Harrison, 1995; Mieczkowski, 1990; Mieczkowski et al., 1991; Mieczkowski and Newel, 1993) and detained juveniles (Dembo et al., 1987, 1993, 1999; Feucht, Stephens, and Walker, 1994; Magura, Kang, and Shapiro, 1995; Mieczkowski, Newel, and Wraight, 1998; Wislar and Fendrich, 2000).

3. Saliva analysis has problems of contamination and different absorption rates for different drugs. Sweat collection is impractical in many settings because the subject must be available for 24 to 48 hours. Collection of blood requires a licensed health care professional. Semen presents obvious collection problems and is, of course, gender specific. Both urine and hair can be collected by trained lay personnel. Urine has additional advantages: many substances are concentrated in the urine by normal kidney functions, and, compared with other fluids, urine is relatively free of cellular components that can interfere with drug testing. The collection of any fluid or tissue, including urine and hair, poses special problems for juvenile justice settings: because subjects are detained, they may feel coerced and are more likely to be noncompliant. In fact, collection for forensic purposes may require a court order.

4. However, the detection period for urinalysis can be as long as 3 weeks for heavy use of cannabis and as long as 30 days for use of phencyclidine (PCP), benzodiazepines, and some long-acting barbiturates.

5. How drugs become part of the hair follicle is not clear. They may pass from the blood to the follicle as it is formed, or they may be transferred to the follicle from sweat and sebum (in which case environmental exposure rather than actual drug use could result in a positive finding).

6. The mean time between intake and sampling was 1.5 days. The median time was 1 day.

7. Fourteen subjects refused to provide a sample, 17 were unable to urinate, 7 samples were too small to analyze, and 8 were not collected because too much time had elapsed since intake. Thirty-eight samples were missing because procedures were not yet in place or equipment and supplies were unavailable.

8. Some subjects who were included in the analysis did not complete certain portions of the DISC interview. One subject refused to answer items about heroin and opiate use. Of greater concern for purposes of this analysis were the subjects who reported lifetime use but refused to answer questions about use in the past 6 months (1 for marijuana, 16 for cocaine, 17 for heroin/opiates, 1 for downers, 46 for hallucinogens, and 3 for tranquilizers).

9. Barbiturates, methaqualone, and propoxyphene should be combined for comparison to self-reported use of “downers”; however, there were no positive EMIT results for either methaqualone (which is no longer produced in the United States) or propoxyphene. EMIT tests for benzodiazapenes were compared to self-reported use of “other tranquilizers, Valium, Librium, Ativan.” Phencyclidine is the only hallucinogen test in the EMIT–10 panel; self-reported hallucinogen use (“hallucinogens, LSD, peyote, mescaline, PCP, mushrooms”) was compared only to the EMIT phencyclidine test.

Detection and Prevalence of Substance Use Among Juvenile Detainees	OJJDP Bulletin • June 2004