The sample for the analysis consisted of all eligible drug felony defendants randomly assigned to one of the two dockets between September 1, 1994, and January 31, 1996. The 311 eligible defendants on the standard docket served as the comparison group. Two treatment samples were delineated: the 365 eligible defendants on the sanctions docket and the 240 defendants who agreed to participate in the program (66 percent of the eligible defendants). Data consisted of demographic characteristics, court actions, and drug-test results recorded by the court and Uniform Crime Report arrest records. Other data from the evaluation included follow-up surveys with subsamples (160 graduated-sanctions participants and 156 controls) conducted 1 year after sentencing. Findings showed that program participants were significantly less likely to use drugs prior to sentencing, and in the year after sentencing, they were significantly less likely to be arrested or had significantly fewer arrests. Suggestions are offered for future research in this area. 5 tables, 3 figures, 4 notes, and 43 references