Objective: The aims of this study are to compare DSM-IV criteria for alcoho
l and cannabis use disorders with its predecessor, DSM-III-R, and to examin
e the validity of the new criteria in an adolescent drug clinic sample. Met
hod: During evaluation, a sample of 772 adolescents (63% boys, 77% white) w
ere administered a structured interview of diagnostic symptoms and addition
al problem severity measures. Independent staff ratings of problem severity
and treatment referral were collected as well. Results: Compared to its pr
edecessor, DSM-III-R, application of the DSM-IV criteria for alcohol and ca
nnabis users resulted in more abuse assignments and fewer dependence assign
ments. The shift in assignments appeared to be largely due to a lowering of
the abuse threshold, rather than to a tightening of the dependence criteri
a. The external validity data generally supported the DSM-IV abuse and depe
ndence distinction in adolescents, and the newer criteria were as valid as
the older criteria. Conclusions: In contrast to DSM-III-R, the DSM-IV syste
m yields more abuse cases and fewer dependence cases among adolescent alcoh
ol and cannabis abusers. Validity evidence for the new criteria are defensi
ble, yet the findings are seen as a starting point for discussing the need
for tailoring substance use disorder criteria for adolescents.