A series of studies used the same basic research design in randomly assigning consenting drug-court clients either to attendance at biweekly judicial status hearings throughout their participation in drug court or to be monitored by their treatment case managers, who would petition the court for a status hearing if infractions occurred. Baseline measures focused on problems related to drug and alcohol use, legal status, medical conditions, family and peers, and employment. Measures of performance during treatment included attendance at counseling sessions, drug-test results, self-reported substance use and criminal activity, and graduation rates. Studies were conducted at three sites in Delaware, with both misdemeanor and felony clients represented. The studies found that high-risk clients, who had antisocial personality disorder or had previously failed in drug treatment, performed significantly better under the drug court when they were required to attend frequent status hearings before the judge. In contrast, low-risk offenders, who did not have the aforementioned characteristics, performed better when they were monitored by their treatment case managers instead of being required to attend regular judicial status hearings. These findings held across all study sites, which included both rural and urban communities. The implications of these findings are discussed, and directions for future research on drug courts are recommended. 4 figures, 1 table, and 40 references