Table 3.
Criterion validity of cannabis screening instruments
| Scale | Study | Criterion | Cut-off | Sensitivity | Specificity | Positive predictive value | Negative predictive value | Area under the curve |
| General population samples | ||||||||
| SDS | Steiner et al. (2008) | DSM-IV dependence | 2 [men: 4] |
93.6% [81.0%] | 74.0% [86.9%] | 28.8% | 99.0% | .92 |
| SDS | Martin et al. (2006) | DSM-IV dependence | 4 | 65.1% | 94.3% | .85 | ||
| SDS | Kedzior et al. (2007) | Any DSM-IV or ICD-10 diagnosis | 72.7%a | 96.4%a | 94.1%a | 81.8%a | ||
| PUM | Okulicz-Kozaryn (2007) | ICD-10 dependence | 2 | 80.9% | 87.5% | 86.6% | 82.2% | .91 |
| CAST | Legleye et al. (2007) | POSITc | 4 | 92.9% | 81.4% | 45.8%b | 96.5%b | .92 |
| CUDIT | Annaheim et al. (2008) | Use at school | 7 | 71.1% | 64.2% | 14.1%b | 96.4%b | .74 |
| Use and drive | 5 | 66.7% | 58.9% | 29.6%b | 83.3%b | .63 | ||
| Depression | 5 | 57.8% | 50.4% | 43.8%b | 64.2%b | .57 | ||
| Use to cope | 5 | 75.0% | 58.9% | 49.2%b | 81.6%b | .76 | ||
| Self-evaluation | 7 | 71.3% | 69.7% | 36.6%b | 91.0%b | .77 | ||
| Specific user samples | ||||||||
| SDS | Swift, Copeland & Hall (1998) | DSM-III-R dependence | 3 | 64.0% | 82.0% | .77 | ||
| SDS | Hides et al. (2007) | DSM-IV dependence | 2 | 86.0% | 83.0% | 83.0% | 86.0% | |
| CUDIT | Adamson & Sellman (2003) | DSM-IV dependence | 8 | 73.3% | 94.7% | 84.6% | 90.0% | .87 |
a Computed by the authors. b Received on request
c Problem Oriented Screening Instrument for Teenagers [28] measures psychological, physical and social problems.