Previous research by the NSW Bureau of Crime Statistics and Research has shown that the cost-effectiveness of the NSW Drug Court could be increased through the early identification of those Drug Court participants at risk of non-compliance with program requirements. There are two reasons for this. Firstly, early identification would permit appropriate adjustments to be made to treatment in order to reduce the risk of program failure. Secondly, those who are at serious risk of non-compliance with the program could be removed from the program early on to reduce the cost they impose on the program. This bulletin describes an investigation designed to identify early indicators of future program compliance. The cohort selected for investigation consisted of those participants who commenced on the NSW Drug Court program between January 1, 2001 and June 30, 2002, and who had remained on the program for at least three months. Of the 217 participants who had been on the Drug Court program for at least three months, 79 per cent were still on the program at six months. The study finds that several indicators of program compliance during the first three months were predictive of subsequent program retention at six months. Whether or not sanctions had been imposed during the first three months (including custodial episodes and suspended sanctions) was predictive of a participant having been subsequently terminated prematurely from the program at six months. Interestingly, those participants who had accrued suspended sanctions during the baseline period but who had then had some of them waived (for improved program compliance), were less likely to be subsequently terminated from the program than those who did not have any of their sanctions waived. The number of bench warrants issued for having absconded from the program was also predictive of subsequent program termination. In terms of the utility of baseline urinalysis testing, low-test provision during the first three months was highly predictive of subsequent program termination. In terms of offending, missed program appointments and having tested positive to both stimulants and opiates during the baseline period were identified as being independently predictive of subsequent offending. A range of baseline compliance indicators predicted later concurrent use of stimulants and opiates, including custodial sanctions, suspended sanctions and missed program appointments. These findings have significant implications for the management of the NSW Drug Court program.