Navigation path

Additional tools

  • Print version
  • Decrease text
  • Increase text

European Drug Abuse Prevention Trial (EU-Dap)

Evidence level:
Best PracticePromising PracticeEmergent Practice
Evidence of Effectiveness:
Enduring Impact:
Review criteria and process

Policy category

Helping Vulnerable Children, Facilitating Positive Transitions to Adulthood

Recommendation Pillars

Put in place mechanisms that promote children’s participation in decision making that affect their lives

Countries that have implemented practice

Austria, Germany, Sweden, Spain, Italy, Greece, Belgium

Age Groups

Teenagers (age 13 to 19)

Years in Operation

2003  - still operating

Type of Organization Implementing Practice

International government

Practice Overview

The European Drug Abuse Prevention Trial (EU-Dap) is an experimental evaluation of a school-based drug abuse prevention programme conducted in seven EU countries (Austria, Belgium, Germany, Greece, Italy, Spain, and Sweden).

The programme, Unplugged, uses a comprehensive social-influence approach to reduce use of alcohol, tobacco, and illegal drugs among 12-14 year old students. In the literature, this approach has been repeatedly shown to “reduce onset of use or significantly reduce cigarette, alcohol, and marijuana consumption” among young people (Sussman 2004). Specifically, the Unplugged programme consists of 12 one-hour units taught by classroom teachers who have previously received a 2.5 day training course on the programme material.

Evidence of Effectiveness


Evaluation 1

Faggiano, F et al. The effectiveness of a school-based substance abuse prevention program: EU-Dap cluster randomized controlled trial. Preventative Medicine. 2008. 47: 537-543.


Summary of Results for Evaluation 1


Mean Difference Between Treatment and Control Groups

Outcomes improved (statistically significant)



Daily smoking


Any drunkenness


Frequent drunkenness


Any Cannabis*


Outcomes with no effect



Any smoking



Frequent smoking



Frequent Cannabis



Any drug



*only marginally significant (p<0.1)




Evaluation 2  (18-month Follow-Up)


Faggiano, F. et al. The effectiveness of a school-based substance abuse prevention program: 18-month follow-up of the EU-Dap cluster randomized controlled trial. Drug and Alcohol Dependence. 2010. 108: 56-64.


Summary of Results for Evaluation 2 (18-month Follow-Up)


Mean Difference Between Treatment and Control Groups

Outcomes improved (statistically significant)



Any drunkenness


Frequent drunkenness


Outcomes with no effect



Any smoking



Frequent smoking



Daily smoking



Frequent Cannabis



Any Cannabis



Any drug




The programme was validated in 7 EU member states simultaneously, and sufficient programme materials exist for further validation.

Enduring Impact

Follow-up only occurred 6 months and 18 months after the initial survey (3 months and 15 months after the treatment was applied). Further follow-ups are possible, although the study participants would now be adults (12-14 years old in 2005, now 19-23 years old).

Issues to Consider

The authors admit that the study has a major drawback: there is selective attrition between the treatment and control groups. This drawback is serious because it casts into doubt the validity of the study. In the follow-up, authors state that most of these differences can be attributed to one Italian school with a large student body and high substance use. Authors also state that when they removed this school from the sample, most of the differences between groups disappeared. This data was not shown in either evaluation.

In a letter to the editor, Dennis Gorman (2009) brings up this particular criticism, among others pertaining to the study design, outcome significance, outcomes chosen for analysis, sensitivity analysis, and even Faggiano’s adherence to the scientific method. Specifically, Gorman criticized the change of a 4-arm study design in the data collection phase to a 2-arm study design in the data analysis phase. He also criticized the difference in group baseline characteristics (as discussed above), the vagueness of some sensitivity analyses, and the inconsistent labeling of outcomes as significant or insignificant. Finally, Gorman criticized Faggiano for designing a study intended to “verify [a] hypothesis” rather than “a rigorous test designed to falsify it.”

Faggiano replied to these criticisms in another letter to the editor (2009). In this he claims that the study design was changed to increase statistical power. He also reiterates arguments made in the original papers for sensitivity analysis, outcome choice, and outcome significance. Finally, Faggiano makes the point that large scale public health interventions are difficult to carry out and that he essentially disagrees with Gorman’s interpretation of the scientific method. He ends by saying that “the overall tendency of the results and the lack of major bias should be noted,” and begs the question that these results should be dismissed “as ‘negative evidence.’”

Additionally, the evaluation of the study can be difficult to interpret because no standard errors or p-values are provided. They do provide the reduction in outcome risk after treatment and model adjustment, but only confirm in the text of the evaluation (not visually) which outcomes were statistically significantly different after treatment.

Contact Information


Peer van der Kreeft




Faculty of Education, Health and Social Work, University College Ghent


Kortrijksesteenweg 14, BE-9000 Gent


+32 (9) 243 34 84


Available Resources

The best available information about EU-Dap can be found on the study website.

Evaluation Details

A sample of 6370 12-14 year old students was constructed from 141 junior high schools in 9 cities within 7 EU countries. All of these students completed pre-test surveys in the fall of 2004 and follow-up surveys in early 2005. The control group consisted of 3174 students, and the treatment group consisted of 1084 students in the basic arm, 1068 students in the parent arm, and 1044 students in the peer arm. Attrition among treatment schools was higher than among control schools, and the baseline prevalence for all substance abuse behaviours was different among treatment and control groups.

The authors also conducted a 15-month follow-up survey. In this survey, 5541 students remained in the sample, with a control group of 2730 students, and a treatment group consisting of 956 students in the basic arm, 972 students in the parent arm, and 883 students in the peer arm. Attrition was not differential between treatment and control groups.

The authors adjusted naïve treatment effects with multilevel models based on student, school, and city levels. They also controlled for daily smoking prevalence at baseline to account for city-level differences.


Last updated

December 2012