Project ALERT is a school-based curriculum designed to reduce substance use among middle school students. It is delivered to seventh grade students, with booster sessions delivered in the eighth grade.
Project ALERT aims to teach children to establish no-drug-use norms, to develop reasons not to use drugs, and to resist pro-drug pressures. The program consists of a 14-lesson curriculum, participatory activities, and videos. It includes guided classroom discussions and small group activities that are designed to stimulate peer interaction and challenge students. It also includes intensive role-playing for students to practice resistance skills. Parent-involved homework assignments extend the learning process for participants. In the original curriculum, students in the seventh grade received eight lessons; under the revised curriculum, they receive an additional three sessions. The revised curriculum introduces material on smoking cessation and alcohol use. Parental involvement activities are included as well, such as adolescent interviews with parents and parent/child drug IQ tests to assess knowledge and social influences. Students participate in three lessons in eighth grade.
The program is built using strategies from the health belief model, the social learning model, and the self-efficacy theory of behavior. The health belief model dictates an emphasis on cognitive factors that can promote healthy behaviors. For Project ALERT, this means increasing awareness of the negative personal consequences of drug use and the benefits of nonuse, as well as promoting resistance by increasing recognition of pro-drug pressures and beliefs. The social learning model suggests an emphasis on social norms and a recognition of significant others as key determinants of behavior. The self-efficacy theory of behavior recognizes that, for youths to take effective action, they must first believe they have the ability to do so. Strategies include building resistance self-efficacy and modeling and practicing desired behaviors (Ellickson, Bell, and McGuigan 1993; Ellickson et al. 2003).
11 to 18
Studies evaluating this program returned some mixed results; however, the preponderance of evidence in these studies shows that the program had no sustained effects in changing behaviors. While study 2 demonstrated positive impacts of the program on substance use behavior and beliefs, the study also had a shorter follow-up period. Study 3 had a similar follow-up period but found no positive program impacts. Study 3, which had the longest follow-up period, showed that program impacts largely decayed over time, and impacts on behavior decayed more quickly than the impacts on beliefs.
Ellickson, Bell, and McGuigan (1993) found that in middle school the program produced generally positive effects for all risk categories for use of cigarettes, alcohol, and marijuana. For instance, the program significantly reduced current, weekly, and daily smoking during eighth grade among previous experimenters, although it had a negative effect on smoking behavior of early users. The program had limited effects on alcohol use and beliefs.
While the program demonstrated some (negative) effects on alcohol use for some youths in 10th grade, by the end of high school the program no longer had any significant effects on behavior. The positive and boomerang effect for cigarette and marijuana use disappeared, as had the negative effects on alcohol use in grade 10. Generally, most impacts of the program on behavior had disappeared by ninth grade.
Beliefs About Substance Use
In middle school, the program curbed prosmoking beliefs among baseline nonsmokers and more committed users. As late as grade 10, participants in the two experimental groups demonstrated beliefs that drug use would have negative personal consequences and that resistance would earn friends’ respect. By 12th grade, most of the program’s impact on beliefs had disappeared.
Project ALERT had a statistically significant impact on cigarette use. Cigarette initiation rose in each group, such that by the end of eighth grade rates in the control group rose to 31.6 percent and in the treatment group to 25.5 percent. Thus, Project ALERT lowered cigarette initiation in the treatment group, reducing the proportion of new smokers by 19 percent. It also produced 23 percent reductions in current (past month) and regular (weekly) smoking in the treatment group.
Ellickson and colleagues (2003) found that Project ALERT had a statistically significant effect on marijuana use. Marijuana initiation rose in each group, such that, by the end of eighth grade, rates in the control group rose to 17 percent and in the treatment group to 13 percent. Thus, Project ALERT lowered marijuana initiation in the treatment group, reducing the proportion of new users by 24 percent. It also produced a 15 percent reduction in current (past month) and an 18 percent reduction in regular (weekly) marijuana use in the treatment group, but neither of these reductions was statistically significant.
Project ALERT produced significantly lower overall alcohol misuse in students in the treatment group compared with students in the control group. Treatment group students were also significantly less likely to engage in drinking that resulted in negative consequences and marginally less likely to engage in multiple forms of high-risk drinking. The program did not reduce alcohol initiation or current use.
St. Pierre and colleagues (2005) found no statistically significant difference between the treatment and control groups for use of alcohol, cigarettes, and marijuana. The only effects that reached statistical significance were harmful effects of the teen-led program on past year marijuana use, although this result could be due to chance.
Change in Mediator Variables
There were no significant differences between the treatment and control groups on mediator variables (e.g., normative perceptions of friends’ approval of substance use). Only one contrast reached statistical significance: expectations of future marijuana users were higher among treatment group students than among control group students, but this result may be due to chance alone.