EMCDDA Home
  • EN
Search

Best practice portal:
Prevention interventions for school students

This page refers to the current evidence on the effectiveness of the available prevention interventions for school students. Information on the methodology used and the definition of terms can be found on the methodology page.

Date of last update: 06.2016    Next update: 12.2016

Available evidence for prevention interventions for school children

Summary:  Programmes combining interventions based on social influence and on skill-based interventions have been proven to be effective in reducing licit and illicit drug use; interventions aimed at disadvantaged students and interventions peer-lead have shown promising results.

Beneficial

Social influence–based interventions for alcohol use

Social influence approaches use normative education methods and anti-drugs resistance skills training. School-based programmes that implement the concepts of social influence and life skills were found in large trial (Faggiano et al., 2010) to be effective in:

  • reducing overall drunkenness (OR 0.80, 95 % CI 0.67 to 0.97, 1 RCT, N= 7079);
  • reducing drunkenness in 3 or more episodes (OR 0.62, 95 % CI 0.47 to 0.81, 1 RCT, N= 7079) 

Life skill-based interventions to reduce any drug use measured at <12 months follow-up

Life skills programmes were found in a systematic review (Faggiano et al., 2014) to be more effective than usual curricula in:

  • reducing generic drug use at <12 months follow-up (RR 0.27, 95 % CI 0.14 to 0.51, 2 studies, N=2512)

Combined approaches of skill-based and social influence–based interventions to reduce cannabis use at 12+ months follow-up and any drug use at <12 months follow-up

School programmes based on a combination of social competence and social influence approaches were found in a systematic review (Faggiano et al., 2014) to be more effective than usual curricula in:

  • reducing cannabis use at 12+months follow-up (RR 0.83, 95 % CI 0.69 to 0.99, 6 studies, N=26 910);
  • reducing generic drug use at <12 months (RR 0.76, 95 % CI 0.64 to 0.89, 1 study, N=6362)

Likely to be beneficial

Interactive programmes vs non interactive ones for problematic students use of licit and illicit drugs

School-based interactive programmes (i.e. those involving discussion) implemented in schools with predominantly special population, were found in a meta-analysis of 207 studies (Tobler et al., 2000), in a re-analyses of the same set of studies (Roona et al., 2003) and in a further analysis of 15 studies (N = 15.571 sampling units) (Porath-Waller et al., 2010), to be more effective than non-interactive programmes (i.e. a lecture) in:

  • reducing smoking or non-smoking tobacco use, alcohol use or drinking/driving, “cannabis” or other illicit drugs’ use (SMD 0.21 vs. -0.05, p < 0.000);
  • statistically significant delay in the onset of substance use;
  • reducing youth cannabis use.

Multi-component prevention programs for alcohol misuse in young people

Multi-component prevention programs (intervention delivered in more than one setting) have shown in a systematic review (Foxcroft et al., 2011b) of 20 RCTs to be effective in:

  • reducing alcohol misuse in adolescents (12 of the 20 trials showed some evidence of effectiveness).

Peer-led approaches

A review of 29 reviews (McGrath et al., 2006) found evidence in favour of the effectiveness of peer educators in school-based drug prevention programmes in:

  • reducing all substances use at post-test (SMD = 0.24, 95 % CI 0.06–0.41, Z-test p < 0.01);
  • reducing tobacco smoking at post-test (SMD = 0.17, 95 % CI 0.05–0.21, Z-test p < 0.01).

However, this relative effectiveness did not extend to 1 or 2 year follow-ups (McGrath et al., 2006).

School-based alcohol-specific prevention programs in preventing alcohol misuse in school-aged children up to 18 years of age

School-based prevention programs have shown to be effective in a systematic review (Foxcroft et al., 2011a) of 53 RCTs in:

  • reducing alcohol misuse in adolescents (statistically significant outcomes in 6 out of 11 alcohol-specific trials).

Life skill-based interventions to reduce cannabis use

Life skills programmes were found in a systematic review (Faggiano et al., 2014) to have a more positive trend than usual curricula in::

  • reducing use of cannabis at >12 months follow-up (RR 0.90, 95 % CI 0.81 to 1.01, 4 studies, N=9456);
  • reducing use of cannabis at 12+ months follow-up (RR 0.86, 95 % CI 0.74 to1.00, 1 study, N=2678)

Combined approaches of skill-based and social influence–based interventions to reduce cannabis use at <12 months follow-up

School programmes based on a combination of social competence (life skills-based) and social influence approaches were found in a systematic review (Faggiano et al., 2014) to have a more positive trend than usual curricula in:

  • reducing cannabis use at <12 months (RR 0.79,  95 % CI 0.59 to1.05, 3 studies, N=8701)

Social influence–based interventions to reduce cannabis use at <12 months follow-up

Social influence approaches use normative education methods and anti-drugs resistance skills training. ing School-based programmes that implement the concepts of social influence were found in a systematic review (Faggiano et al., 2014) to have a more positive trend than usual school curricula or no intervention in:

  • reducing use of cannabis at <12 months (RR 0.88, 95 % CI 0.72 to 1.07, 3 studies, N=10 716) 

Trade-off between benefits and harms

No interventions met the criteria for this category.

Unknown effectiveness

Booster sessions

A review of 29 reviews  (McGrath et al., 2006) found evidence that booster sessions or similar extra components that aimed to reinforce the effects of a programme have a positive impact on the pre-specified outcomes. However, since the relationship between booster sessions and programme outcomes was not statistically examined, the link should be treated as hypothetical.

Combined approaches of skill-based and social influence–based interventions to reduce hard drug use

School programmes based on a combination of social competence (life skills-based) and social influence approaches were found in a systematic review (Faggiano et al., 2014) to have no different effect than usual school curricula or no intervention in:

  • reducing hard drug use at <12 months follow-up (conflicting results between dichotomous and continuous outcomes within 1 study, N=693)
  • reducing hard drug use at 12+ months follow-up (RR 0.86, 95 % CI 0.39 to 1.90, 2 studies, N=1066)

Life skill–based interventions to reduce hard drug use

Life skills programmes were found to be no more effective than usual curricula (Faggiano et al., 2014) in:

  • reducing use of hard drugs at >12 months follow-up (RR 0.69, 95 % CI 0.40 to 1.18, 1 study, N=2090)
  • reducing use of hard drugs at 12+ months follow-up (MD -0.01, 95 % CI -0.06 to 0.04, 1 study, N=1075)

Multi-component versus single components programs for alcohol misuse in young people

There is little evidence that interventions with multiple components are more effective than interventions with single components as shown in a systematic review (Foxcroft et al., 2011b):

  • only 1 of the 7 trials where assessment of the additional benefit of multiple versus single component interventions was possible, clearly showed a benefit of components delivered in more than one setting.

School-based generic prevention programs in preventing alcohol misuse in school-aged children up to 18 years of age

School-based generic prevention programs (not including life/social skills approaches, see above in the Beneficial category) have not shown enough evidence to be effective in a systematic review (Foxcroft et al., 2011a) of 53 RCTs in:

  • reducing alcohol misuse in adolescents (no statistically significant outcomes in 24 out of 39 generic trials).

School-based brief interventions to reduce substance use and delinquent-type behaviours

School-based brief interventions were found in a systematic review (Carney et al., 2016, 6 RCT, N=1 176) to have no different effect than information-only interventions (eg. general health promotion materials and harm reduction information) in:

  • reducing alcohol and cannabis use
  • reducing delinquent-type behaviours

There was very low-quality evidence that brief school-based interventions may be more effective in reducing alcohol and cannabis use than no-intervention (i.e when compared to assessment-only) and that these reductions were sustained at long-term follow-up, however it is premature to make a definitive statement.

Social influence–based interventions to reduce hard drug use and cannabis use at 12+ months follow-up

Social influence approaches use normative education methods and anti-drugs resistance skills training.  School-based programmes that implement the concepts of social influence were found in a systematic review (Faggiano et al., 2014) to have no different effect than usual school curricula or no intervention in:

  • reducing use of cannabis at 12+ months follow-up (RR 0.95, 95 % CI 0.81 to 1.13, 3 studies, N=10 716)
  • reducing hard drug use (just one study not providing data for meta-analysis found a significant protective effect)

Standalone knowledge-focused interventions to improve knowledge and reduce use

Knowledge focused programs were found in a systematic review (Faggiano et al, 2014) to have no different effect than usual school curricula or no intervention in:

  • improvinge participants’ knowledge of illicit drugs (SMD 0.91, 95 % CI 0.42–1.39, 1 study, N=165);
  • reducing illicit drug use.

Standalone anti-alcohol/tobacco peer programmes

There is evidence from a narrative review including results of one RCT (N=326) (Velleman 2009) that specific training given to young people in order to make them influence each other is not effective in:

  • reducing alcohol and tobacco use;
  • increasing knowledge and attitudes towards alcohol and tobacco

Evidence of ineffectiveness

No interventions met the criteria for this category.

References and definitions

References

Explanation of terms used

Below you can find definitions and further explanation for some of the terms used in this section of the Best practice portal. A more general glossary for the best practice portal is also available.

Affective-focused prevention intervention

A type of prevention intervention which aims to they aim to modify inner qualities (personality traits such as self-esteem and self-efficacy, and motivational aspects such as the intention to use drugs).

BA

Before-after (BA) study design

BAL

Blood alcohol level (BAL)

Beneficial

Interventions for which precise measures of the effects in favour of the type of intervention were found in systematic reviews of relevant studies. An intervention ranked as ‘beneficial’ is suitable for most patients/contexts. See the relevant module methodology page for further information.

CBA

Controlled before-after (CBA) study design. UCBA stands for Uncontrolled before-after study design.

CBT

Cognitive behavioral therapy is an individual based intervention occurring in three stages. Phase 1 is aimed at determining and prioritizing the patient’s problems and constructing the treatment contract. Phase 2 is aimed at increasing coping competence and reducing risky behaviors. Phase 3 focuses on relapse prevention. Each session is administered once per week over a period of 4-6 months with 60- to 90-minute sessions (Beck AT, Wright FW, Newman CF, Liese B. Cognitive Therapy of substance abuse. New York: Guilford Press, 1993).

CCT

Controlled clinical trials (CCT)

Cohort study

A cohort study is a type of observational study that follows a group of people (i.e. a cohort) over time. In a prospective cohort study, the cohort is formed and then followed over time. In a retrospective cohort study, data is gathered for a cohort that was formed sometime in the past.

Confidence Interval (CI)

The Confidence Interval (CI) is a measure of the precision (or uncertainty) of study results. It is the interval that most likely includes the true value of the parameter we are calculating, where 'most likely' is taken by common usage to be a 95% probability. Thus the current expression of '95 % CI'. A wide CI indicates less precise estimates of effect and vice versa.

Practical interpretation

  • If the RR (the relative risk) = 1, or the CI (the confidence interval) = 1, then there is no significant difference between treatment and control groups
  • If the RR > 1, and the CI does not include 1, events are significantly more likely in the treatment than the control group
  • If the RR < 1, and the CI does not include 1, events are significantly less likely in the treatment than the control group
CPS

Current population survey (CPS)

Cross-sectional study

A cross-sectional study is a study employing a single point of data collection for each participant or system being studied.They are usually conducted to estimate the prevalence of the outcome of interest for a given population at a given point in time.

Evidence of ineffectiveness

Interventions that gave negative results if compared with a standard intervention or no intervention, for example. See the relevant module methodology page for further information.

Additional information for prevention
For ethical reasons this category in prevention should be considered as interventions with negative and undesired (iatrogenic) effect.

IP

Individual psychotherapy is a standard individual treatment based on counseling and motivational interviewing and focused on substance use triggers and strategies for relapse prevention. It includes elements of cognitive-behavioral therapy (CBT).

IQR

Interquartile range (IQR) - also called the midspread or middle fifty - is a measure of statistical dispersion. It is a trimmed estimator, defined as the 25% trimmed mid-range, and is the most significant basic robust measure of scale.

ITS

Intermittent time series design (ITS)

Knowledge-focused prevention intervention

A type of prevention intervention which aims to to enhance knowledge of drugs, and drug effects, and consequences.

Likely to be beneficial

Interventions that were shown to have limited measures of effect, that are likely to be effective but for which evidence is limited. An intervention ranked as ‘likely to be beneficial’ is suitable for most contexts/patients, with some discretion. See the relevant module methodology page for further information.

Number Needed to Treat (NNT)

The Number Needed to Treat (NNT)indicates the number of patients that needs to be treated to obtain one respondent patient. Numerically the NNT is the reciprocal of the difference between the proportion of events in the experimental and the comparison group (absolute risk reduction). Taking into consideration that the ideal NNT would be 1 (the unreal situation in which every single patient succeeded) it is easily understood that a NNT value close to 3 or 4 would be very good.

Adjusted Odds Ratio (AOR)

The Adjusted Odds Ratio is a way of comparing whether the probability of a certain event is the same between two groups, yet they are calculated adjusting for or controlling for other possible contributions from other variables (tipically demographic variables) in the model. An AOR equal to 1 implies that the the event is equally probable in both groups. An AOR greater than 1 implies that the event is more likely in the first group. An AOR less than 1 implies that the event is less likely in the first group.

Odds Ratio (OR)

The Odds Ratio is a way of comparing whether the probability of a certain event is the same between two groups. Like the Relative Risk, an OR equal to 1 implies that the the event is equally probable in both groups. A OR greater than 1 implies that the event is more likely in the first group. A OR less than 1 implies that the event is less likely in the first group. In medical research, the odds ratio is commonly used for case-control studies, as odds, but not probabilities, are usually estimated. Relative risk is used in randomized controlled trials and cohort studies.

p value

A p-value is a measure of how much evidence we have against the null hypothesis. The null hypothesis represents the hypothesis of no change or no effect. The smaller the p-value, the more evidence we have against the null hypothesis thus it is more likely that our sample result is true. Traditionally, researchers will reject a null hypothesis if the p-value is less than 0.05.

RBS

Responsible beverage service (RBS)

RCT

Randomised controlled trial (RCT)

Relative Risk (RR)

The Relative Risk (RR) is used to compare the risk in the two different groups of people, i.e. treated and control groups to see if belonging to one group or another increases or decreases the risk of developing certain outcomes. This measure of effect will tell us the number of times an outcome is more likely (RR > 1) or less likely (RR < 1) to happen in the treatment group compared with the control group.

Practical interpretation

  • If the RR (the relative risk) = 1, or the CI (the confidence interval) = 1, then there is no significant difference between treatment and control groups
  • If the RR > 1, and the CI does not include 1, events are significantly more likely in the treatment than the control group
  • If the RR < 1, and the CI does not include 1, events are significantly less likely in the treatment than the control group
Trade-off between benefits and harms

Interventions that obtained measures of effects in favour of the intervention, but that showed some limitations or unintended effects that need to be assessed before providing them. See the relevant module methodology page for further information.

 
Unknown effectiveness

Interventions for which there are not enough studies or where available studies are of low quality (with few patients or with uncertain methodological rigour), making it difficult to assess if they are effective or not. Interventions for which more research should be undertaken are also grouped in this category.

Additional information for prevention
For prevention interventions,  this  is also known as 'zero effect'.

Skill-focused prevention intervention

A type of prevention intervention which aims to enhance students’ abilities in generic skills, refusal skills and safety skills.

Standardised Mean Difference (SMD)

The Standardised Mean Difference (SMD) is the difference in means divided by a standard deviation. Note that it is not the standard error of the difference in means (a common confusion). The standardized mean difference has the important property that its value does not depend on the measurement scale. It may be useful if there are several trials assessing the same outcome, but using different scales.

z score (Standard Score)

The z-score (aka, a standard score) indicates how many standard deviations an element is from the mean of the population.

 

Case definition

According to recent European surveys (1), on average, 23 % of boys and 17 % of girls have tried illicit drugs at least once during their lifetime . The term ‘any illicit drug’ includes cannabis, amphetamines, cocaine, crack, ecstasy, LSD and heroin. Reported use of illicit drugs varies considerably across the countries (ESPAD 2007). Prevention targeting schools offer a systematic and efficient way of reaching large numbers of young people.

(1) The European school survey project on alcohol and other drugs (ESPAD) uses standardised methods and instruments to measure drug and alcohol use among representative samples of 15- to 16-year-old school students. Surveys have been conducted in 1995, 1999, 2003 and 2007.
In 2007, data were collected in 35 countries, including 25 EU Member States, Norway and Croatia (AR 2010).

Risk

Since adolescence is per se a risk factor, and the majority of illicit drug users start to use at this age, school is an efficient setting for universal prevention. Moreover, within the larger target group of school youth there are particularly vulnerable sub-groups including early school leavers and students with social or academic problems (as well as young offenders, young people in institutional care, young people living in neighbourhoods where multiple risk factors and problems associated with drug use are concentrated). Overlaps between these groups may exist. For example, children with poor school attendance or poor academic performance may also be experiencing other problems such as problems in the family or juvenile crime. Furthermore, there may be cumulative effects of belonging to more than one vulnerability category, which may be associated with an even greater likelihood of drug use.

Interventions

School programs can be designed to provide knowledge about the effects of drugs on the body and psychological effects, as a way of building negative attitudes toward drugs; to build individual self-esteem and self-awareness, working on psychological factors that may place people at risk of use; to teach refusal and social life skills; to change perception of prevalence (or informal norms); and to encourage alternative activities to drug use, which instil control abilities.

Interventions for school students can be classified into the following groups:

  • skill focused interventions, aimed to enhance students’ abilities in generic skills, refusal skills, safety skills
  • affective focused interventions, aimed to modify inner qualities (personality traits such as self-esteem and self-efficacy, and motivational aspects such as the intention to use drugs)
  • knowledge focused interventions, aimed to enhance knowledge of drugs, and drug effects, and consequences

Also interventions with family- or community-based components were included in this section, provided the school-based component was predominant.

Outcomes

The main aim of prevention interventions delivered to school children is to deter or to delay the onset of substance use by providing all individuals with the information and skills necessary to prevent the problem.
Studies usually consider also the outcomes related to knowledge and /or awareness of drugs risks, however these outcomes are surrogate ones which means that there are no proofs that awareness/knowledge has impact on use. In absence of a clear relation between awareness/knowledge and outcomes on use, we decided NOT to consider them.

Primary outcomes

  • Reduction of substance use
  • Reduction of risky behaviour
  • Reduction of intention to use

References

 

Page last updated: Thursday, 23 June 2016