, 2012, who found significant results with prizeCM in cocaine-dependent outpatients conducted their trials in community-based clinics and compared CM to treatment as usual, and not to CBT. A prizeCM only condition might have been helpful in order to draw conclusions regarding independent

effects of the prizeCM component. However, for ethical reasons a prizeCM only condition was not feasible. Another explanation could be that the value of our prizes may have been too low to produce a rewarding effect in our participants and greater incentives might have provided more favorable outcomes. Finally, the sample size was too small as indicated by the power analysis. The overall retention rate of 63.3% at week 24 in the present trial was high. Previous studies over 12 weeks (Epstein et al., 2003 and Rawson et al., 2006) found comparable retention rates. In accordance with the findings of Epstein et al.

(2003) (CBT = 79%; combination = 69%), the two Selleck PD0332991 groups in the present study did not differ in retention, while Rawson et al. (2006) found a significant group difference (CBT = 40%; combination = 59%). Furthermore, both interventions were effective in a clinical sample of patients with high rates of poly drug use and comorbid psychiatric conditions, which supports previous findings (Magill and Ray, 2009, Petry et al., 2013 and Rash et al., 2008). One explanation for this is that the Bortezomib mouse effects of prizeCM and CBT in reducing psychiatric symptoms are mediated by reductions in drug use. Interestingly, trends in favoring prizeCM plus CBT were found at 6-month follow-up.

This found finding suggests an additive effect of the two psychosocial approaches. Furthermore, prizeCM produced a sustained effect beyond the cessation of incentives, which is in support of the findings of McKay et al. (2010) and Epstein et al. (2003). Those patients who completed the trial showed highly significant reductions in SDS, BDI, and cocaine craving scores as well as in 4 ASI composite scores (drug, alcohol, employment, psychiatric problems), indicating clinical relevance of both treatment interventions. Furthermore, participants in both intervention groups were very satisfied and reported high acceptability of CBT and prizeCM. Finally, those patients with higher frequency of cocaine use at baseline had a higher risk of dropping out of the study. Therefore, we suggest that these patients should have a detoxification treatment before beginning CBT or prizeCM plus CBT. The generalizability of the present findings is limited by the sample size. The original targeted sample size could not be achieved, due to difficulties in recruiting participants. First, the recruiting process was initially planned at two sites, but data from one center was not included due to poor adherence to the study protocol and incomplete data. Second, the recruitment had to be stopped after 3 years due to running out of funds.