Citation

Johnson KW, Shamblen SR, Ogilvie KA, Collins D, Saylor B. Prev. Sci. 2009; 10(4): 298-312.

Affiliation

Pacific Institute for Research and Evaluation, Louisville Center, 1300 S. Fourth Street, Ste. 300, Louisville, KY, 40208, USA, kwjohnson@pire.org.

Copyright

(Copyright © 2009, Holtzbrinck Springer Nature Publishing Group)

DOI

10.1007/s11121-009-0132-2

PMID

19440837

Abstract

This study tests for the efficacy of a school-based drug prevention curriculum (Think Smart) that was designed to reduce use of Harmful Legal Products (HLPs, such as inhalants and over-the-counter drugs), alcohol, tobacco, and other drugs among fifth- and sixth-grade students in frontier Alaska. The curriculum consisted of 12 core sessions and 3 booster sessions administered 2 to 3 months later, and was an adaptation of the Schinke life skills training curriculum for Native Americans. Fourteen communities, which represented a mixture of Caucasian and Alaska Native populations in various regions of the state, were randomly assigned to intervention or control conditions. Single items measuring 30-day substance use and multi-item scales measuring the mediators under study were taken from prior studies. Scales for the mediators demonstrated satisfactory construct validity and internal reliability. A pre-intervention survey was administered in classrooms in each school in the fall semester of the fifth and sixth grades prior to implementing the Think Smart curriculum, and again in the spring semester immediately following the booster session. A follow-up survey was administered 6 months later in the fall semester of the sixth and seventh grades. A multi-level analysis found that the Think Smart curriculum produced a decrease (medium size effect) in the proportion of students who used HLPs over a 30-day period at the 6 month follow-up assessment. There were no effects on other drug use. Further, the direct effect of HLPs use was not mediated by the measured risk and protective factors that have been promoted in the prevention field. Alternative explanations and implications of these results are discussed.

Language: en