Table 4 Overview of problems encountered for drug prevention evaluations

   - An intervention can induce different effects when it is implemented in different situations (e.g. different schools), because the quality and context of the implementation determines its effectiveness. As a consequence, it is very difficult to estimate the overall effectiveness of preventive interventions.

   - Lack of financial sources

   - Lack of expertise

   - Lack of mandates and orders that encourage effect evaluations

   - ...

Judicial problem

Cannabis is an illegal drug in most countries, and self-reporting is often the only way to gather information on frequency of (ab)use. The validity of self-reported data may be questionable.

Ethical problem

Some methods (e.g. the Randomised Control Trial, RCT), require allocation of subjects to different groups. Some persons will be assigned to an experimental group that is exposed to a preventive intervention, whereas others will be assigned to the control group that is not exposed to that intervention. Yet, it seems unethical to deny some the right of being exposed to an intervention that might be effective in reducing drug use.

Methodological problems

An overview of methodological problems encountered in evaluating primary drug prevention, is presented by Uhl (1998, 2000). It goes beyond the scope of this article to give an extensive summary. Some of the main aspects are:

   - The interest goes primary to long term effects of prevention. However, this kind of research is difficult to accomplish (amongst other things, there are problems to follow up participants over many years).

   - Simultaneous influences: people are confronted with the drug topic in many different ways (e.g. media, courses, family, friends,...). Because of this, it is nearly impossible to separate intervention induced effects from effects caused through other influences.

   - Dependence of observational units: a central, implicit assumption behind many statistical procedures is independency of observations. In drug prevention evaluation research it is common that the observational units are sampled in clusters or groups (e.g. school classes) and treated as independent units. The violation of this assumption may result in an increased rate of significance by mere chance.

   - ...