Two types of data files were constructed. First, a set of cross sectional files composed of 57,944 cases in the Prosecutor's Management Information System (PROMIS) and matching urinalysis test records were located for 1973 through 1977. These files were case based, and persons arrested several times had multiple cases included in the files. The second was a defendant-based, longitudinal file containing the arrest records for 7,087 persons from 1973 through 1978; each panel member's prepanel and postpanel cases were extracted from PROMIS and added to the file, yielding a total of 19,277 cases. Information on drug use and time incarcerated was also included. The results of the cross-sectional file study showed that arrestees below age 21 were relatively unlikely to have been found to be using drugs. Those between 21 and 45 had the greatest risk of detection, while a marked decline was noted for those over 30. Drug positive urinalysis test specimens were found for 24 percent of the female arrestees and for 20 percent of the males. Persons charged with drug-related offenses were more likely to be drug positive, and those charged with violating bail were even more likely to be so. The study of the longitudinal file revealed that persons detected to be drug positive at the time of their panel were more likely to be rearrested during the postpanel period than were persons who were drug negative. In addition, 30 percent of the drug positive arrestees had 3 or more subsequent arrests, compared to 18 percent of the drug negative arrestees. Finally, persons who were drug positive at their panel arrest had about a 50 percent likelihood of being found positive at the time of an immediately prior arrest of at their next arrest, compared to 15 to 21 percent of those who were drug negative. Related studies are reviewed. Data tables and a 5-item reference list are included.