Study: Are Drug Screens Sufficient for Adolescent Treatment?

Could it be that we are missing something when it comes to treating adolescents with a substance use disorder? According to a study in the most recent edition of the Journal of Studies on Alcohol Drugs, if you aren't using drug screens when treating this population, the answer is likely "Yes".
In the article, Schuler, et al. looked at data from SAMHSA's CSAT 2007 adolescent treatment database, which tracks outcomes for CSAT-sponsored providers. The total sample consisted of 5,186 adolescents who received either Motivational Enhancement Therapy/Cognitive Behavioral Therapy5 (MET/CBT5) - with or without biological drug screen (BDS) - or were part of the BDS-only or No-Treatment groups within another study. Below is a breakdown of the subjects:

All participants responded to the GAIN structured clinical interview, so scores on the Substance Use Frequency Scale and Substance Problem Scale were the primary outcomes measured. Propensity score methods were used to adjust for baseline differences among youth in the four groups, given the non-randomized nature of the data. The results are striking:

The BDS-only condition seemed to outperform all groups at baseline, 3-, 6- and 12-months on Substance Problem Scale scores and at 6- and 12-months on Substance Frequency Scale scores.
What could account for the significant differences? The authors point out that many adolescents who are involved in treatment and/or the criminal justice system may earn rewards for negative drug screens - or face significant consequences for positive screens. Therefore, the self-report nature of the data could skew the results. That would not necessarily explain the consistent differences across groups, however.

What do you think about these results? If you work with adolescents, does this surprise you? Could evidence like this impact the interventions you use?