LifeSkills Training (Middle school substance abuse prevention curriculum)

Randomized controlled trial shows sizable decrease in students' substance abuse.

Description of the intervention:  LifeSkills Training is a middle school substance abuse prevention curriculum designed to: 1) teach students social and self-management skills, including skills in resisting peer and media pressure to smoke, drink, or use drugs; and 2) inform students of the immediate consequences of substance abuse.  Teachers first explain a life skill (e.g. giving assertive responses) and then demonstrate how to use it.  Students then practice using this skill in role plays in front of the class, and receive feedback from the instructor and their peers on their use of the skill.  Finally, students are given behavioral homework assignments, which require them to use a life skill outside of school (e.g., saying hello to one new person everyday).

The LifeSkills Training organization provides brief training to the regular classroom teachers who ultimately teach the curriculum.  The teacher training is usually carried out through a one-day workshop accompanied by written instructions and curriculum materials (but for about half of the teachers in study 2 below, a two-hour training video was used in lieu of the workshop).  The teachers then provide LifeSkills Training to students in 15 classroom sessions, each approximately 40-45 minutes in length, during students' first year of middle school (either 6th or 7th grade).  Additionally, over the next two years, the teachers provide students with a total of 5-15 review sessions to reinforce what they have learned.

Curriculum materials cost approximately $7 per student per year.  The teacher training workshops cost approximately $100 per teacher.  (The costs shown are in 2008 dollars.)

Click here to go to LifeSkills Training's website.

EVIDENCE OF EFFECTIVENESS

^NEWStudy 1

Randomized controlled trial of 36 rural Midwestern schools with at least 20 percent of students from low-income households.  Schools were randomly assigned to either:  (i) a group whose 7th graders received LifeSkills Training, (ii) a group whose 7th graders received LifeSkills Training plus a family skills training program, or (iii) a control group that received neither program.  The schools contained a total of approximately 1800 7th grade students, 96% of whom were white.

The effects of LifeSkills Training plus family skills training are not reported here because there were no differences in effects between LifeSkills Training alone (group(i)) and LifeSkills Training plus family skills training (group(ii)).

• 21% less likely to have ever smoked cigarettes (53% of LifeSkills Training students reported ever smoking vs. 67% of control group students).
• 23% less likely to have ever used marijuana (29% vs. 38%).
• 11% less likely to have ever been drunk (61% vs. 68%).  This finding was significant at the .10 level but not the .05 level.
• Equally likely to have ever drunk alcohol (i.e., there was no effect on this outcome).

For the full sample, the study found no statistically significant effect on the frequency of substance use (as opposed to substance use initiation, discussed above).  However, for the subgroup of students who already were using two or more substances at the beginning of the study (i.e., the “high-risk“ subgroup), LifeSkills Training produced significant reductions in the frequency of alcohol, cigarette, and marijuana use, drunkenness, and monthly multi-substance use.  The study did not report the magnitude of these effects.

Discussion of study quality (click here for glossary of terms)

• The study had a large, multi-site sample and a long-term follow-up (5.5 years after random assignment).
• Prior to the intervention, the LifeSkills Training and control groups were equivalent in previous levels of substance use and a preponderance of demographic characteristics.
• The study measured the intervention's effects using outcome data for all students in the LifeSkills Training group -- including those whose teachers did not faithfully implement all aspects of the intervention (i.e., the study used an “intention-to-treat“ analysis).
• The study evaluated LifeSkills Training as it is normally implemented in typical classrooms by public school teachers, thus providing evidence of its effectiveness in real-world conditions.
• Alcohol, tobacco, and marijuana use were measured through student self-reports on a written questionnaire, but confidentiality and other procedures were implemented to strengthen the validity of these reports.
• The study's statistical analysis accounted for the fact that schools, rather than individual students, were randomly assigned to the LifeSkills Training and control groups.
• Study Limitations (1) The study had no school attrition (i.e., no schools dropped out of the study after random assignment), but had a moderate to high level of student attrition due to absenteeism, dropouts, and transfers.  Specifically, outcome data were collected for about 67% of the original sample at the 5.5 year follow-up.  Statistical tests suggest that the attrition did not result in any observable differences between the LifeSkills Training and control groups that might undermine their equivalence, and the rate of attrition was about the same for the two groups.  However, the study's results should be treated with caution because it is possible that the attrition may have caused unobservable differences between the two groups, leading to inaccurate estimates of LifeSkills Training's effects.  (2) In estimating the intervention's effects in 12th grade, the study primarily used data from students who completed the 12th grade survey.  However, for students who did not participate in the 12th grade survey, the study used their survey data from earlier grades (as part of an analysis method known as Full Information Maximum Likelihood estimation).  Consequently, the study results described above may reflect some shorter-term effects of the intervention and not just the effects at 12th grade.

Source (click on linked authors' names for their contact information)

Spoth, Richard, G. Kevin Randall, Linda Trudeau, Chungyeol Shin, and Cleve Redmond.  “Substance use outcomes 5 1/2 years past baseline for partnership-based, family-school preventive interventions.”  Drug and Alcohol Dependence 96 (2008) 57-68.

Randomized controlled trial of 57 middle schools, with a sample of approximately 6,000 7th grade students, in three rural or suburban areas in the state of New York.  The students were mostly white (91%) and from predominantly middle-class families.  Schools were matched based on geographic location and smoking rates, and then randomly assigned either to a group that provided LifeSkills Training to its 7th grade students or to a control group that did not. 

Effects of the LifeSkills Training program 6 years after random assignment--i.e., the end of 12th grade

Compared to the control group students in the LifeSkills Training group were:

Discussion of study quality (click here for glossary of terms)

• The study had a large, multi-site sample and a long-term follow-up (6 years).
• Prior to the intervention, there were no significant differences between students in LifeSkills Training schools and those in control schools in substance use or demographic characteristics.
• The study measured the intervention's effects using outcome data for all students in the LifeSkills Training group -- including those whose teachers did not faithfully implement all aspects of the intervention (i.e., the study used an “intention-to-treat“ analysis).
• The study evaluated LifeSkills Training as it is normally implemented in typical classrooms by public school teachers, thus providing evidence of its effectiveness in real-world conditions.
• Self-reported levels of tobacco use were corroborated with breath samples.  Alcohol and marijuana use were measured through student self-reports on a written questionnaire, but confidentiality and other procedures were implemented to strengthen the validity of these self reports.
• The study's statistical analysis accounted for the fact that schools, rather than individual students, were randomly assigned to the LifeSkills Training and control groups.
• Study Limitation:  The study had low school attrition (only one school dropped out of the study after random assignment), but a fairly high level of student attrition due to absenteeism, dropouts, and transfers.  Specifically, outcome data were collected for slightly less than 60% of the original sample of students at the six-year follow-up.  Statistical tests suggest that the attrition did not result in any observable differences between the LifeSkills Training and control groups that might undermine their equivalence.  However, the study results should be treated with caution because it is possible that the attrition may have resulted in unobservable differences between the two groups, leading to inaccurate estimates of LifeSkills Training's effects.

Sources  (click on linked authors' names for their contact information)

Botvin, Gilbert J., Eli Baker, Linda Dusenbury, E.M. Botvin and Tracy Diaz.  "Long-term Follow-up Results of a Randomized Drug Abuse Prevention Trial in a White Middle-class Population."  JAMA, April 12, 1995-- Vol. 273, No. 1; pp. 1106-1112.

Botvin, Gilbert J., Sharon Mihalic, and Jennifer Grotpeter (1998).  Blueprints for Violence Prevention, Book Five: LifeSkills Training.  Boulder, CO: Center for the Study and Prevention of Violence.  http://www.colorado.edu/cspv/blueprints/modelprograms/LST.html  

Other Studies

Other randomized controlled trials of LifeSkills Training have been conducted, but we do not describe them here because they fall outside this website's criteria (e.g., because of sample attrition, only short-term follow-up, or other issues).  The results of these other studies are broadly consistent with the study results described above.