Send to

Choose Destination
J Adolesc Health. 1997 May;20(5):384-95.

Preteenage drug use in Australia: the key predictors and school-based drug education.

Author information

Centre for Health Program Evaluation, University of Melbourne, Australia.



The promise that early intervention school-based drug education is it will have public health benefits. This argument was explored through identifying the key predictors of early adolescent social drug use.


A cross-sectional survey involving 3,019 6th year students, aged 11-12 years (participation rate: 99%), enrolled in 86 Melbourne primary schools was carried out to determine students' social drug use. In addition data were collected on known key social, personal, and education predictors. Logistic regression was used to identify significant predictors of drug experimentation and use.


The key predictors of girls' tobacco use were friends' smoking (OR: 6.7), low literacy (OR: 4.4), and alcohol use (OR: 3.9). For boys they were friends' smoking (OR: 8.6), low literacy (OR: 4.2), and alcohol use (OR: 3.1). For alcohol use, the key predictors for girls were smoking (OR: 4.2), parents' drinking (OR: 3.9), and friends' drinking (OR: 3.8). For boys they were friends' drinking (OR: 3.3), smoking (O: 2.8), and poor literacy (OR: 2.6). Regarding analgesic use, for girls the key predictors were alcohol use (OR: 3.3), analgesic self-administration (OR: 2.4), and parents' drinking or working as tradespersons/laborers (OR: 1.7, respectively). For boys they were analgesic self-administration (OR: 2.5), drinking (OR: 1.9), smoking (OR: 1.7).


The key predictors of social drug use--which suggests multiple recruitment pathways--were all outside the ambit of drug education programs. The impact of education on recruitment to drug use or experimentation among the young is therefore likely to be slight.

[Indexed for MEDLINE]

Supplemental Content

Full text links

Icon for Elsevier Science
Loading ...
Support Center