At the time of writing, trends in adolescent substance use in Scotland show a general decline over time. The recently published Scottish Schools Adolescent Lifestyle and Substance Use Survey ( SALSUS) showed that the proportion of pupils who reported that they had used drugs in the last month has been gradually decreasing since 2002, with the exception of 15 year old boys, for whom there has been a small increase between 2013 and 2015  . The focus of this literature review is on drugs, but tobacco and alcohol are also considered - across all 3 substances prevalence has remained largely stable since 2013, against a backdrop of considerable decline over the last two decades  . While this general downward trend in reported substance use is welcome, there is little understanding currently as to why this is the case, why reported use of certain substances may be on the rise for particular groups and what role drug prevention delivered in Scotland has played in this.
The aim of this review is to explore the evidence of effectiveness of different types of drug prevention and education for children and young people, principally that which is delivered in schools. There is a need for clarity around 'what works' and what does not, to inform approaches taken towards drug prevention and education for young people in Scotland. This is particularly important because currently there is a poor understanding of what prevention activity is being delivered in Scotland, both in schools and more widely. The lack of a national picture of prevention activity in Scotland was identified as a gap and highlighted as a priority for research on prevention in Scotland in both the Report of the Special Working Group on Prevention (2012) and in the Scottish National Research Framework for Problem Drug Use and Recovery (2015).
It is recognised that drug education and prevention for children and young people in schools constitutes only one small aspect of drug prevention. Dickie's extensive logic model on Prevention in her Outcomes Framework for Problem Drug Use (2014) exemplifies the vast scope of prevention activities beyond schools-based prevention that feed into achieving prevention outcomes (attached at Annex A). For the purposes of this paper the evidence on the effectiveness of school-based prevention programmes is prioritised. However, the importance of prevention systems has also been emphasised, and ideally schools-based prevention should be considered as part of a bigger prevention system which encompasses relevant policy, supporting structures, organisations, workforce, prevention ethos and culture etc. (Sumnall, 2016).
There are already several summaries of the evidence of effectiveness of drug prevention activities. This review of the evidence was completed in a short timescale, and is not intended as an exhaustive critical appraisal of the literature. The paper has sought not to repeat previous work but instead to draw together the evidence and findings, to help inform responses to prevention and education in Scotland.
The literature search was conducted by the Scottish Government Library and covered a wide range of resources, including: IDOX, EBSCOHOST (Academic Search, SocIndex), PROQUEST (Applied Social Sciences Index and Abstracts ( ASSIA), ERIC, PAIS International, International Bibliography of the Social Sciences ( IBSS), ProQuest Sociology, Social Services Abstracts, Sociological Abstracts) and Web of Science. The majority of the literature was published within the last five years, although some sources are older, including the evaluation of the effectiveness of drug prevention and education in Scotland (Stead et al., 2007) and the accompanying literature review (Stead and Angus, 2004), which were included because of their relevance to Scotland. The library search included the international literature, but because of the volume of published material, the Americas and Africa were excluded.
However, given the vast quantity of relevant literature on this topic and the time constraints that this literature review was conducted in, the focus has been on the most robust and current systematic reviews on drug prevention and education, and many of the articles on smaller studies have not been cited here. Cochrane Reviews are systematic reviews of primary research in human health care and health policy, and investigate the effects of interventions for prevention, treatment, and rehabilitation. The Cochrane Drugs and Alcohol review group have published several systematic reviews on specific substance use prevention. These are internationally recognized as the highest standard in evidence-based health care resources and provide the most robust evidence available regarding the effectiveness of school-based drug education. For this reason, "Universal school-based prevention for illicit drug use" (Faggiano, Minozzi, Versino and Buscemi, 2014) is key to understanding 'what works' in drug prevention in schools, and is considered with reference to the companion reviews on alcohol and tobacco.
Robust reviews of systematic reviews, from United Nations Office of Drug Control ( UNODC) (2015) and Brotherhood, Atkinson, Bates and Sumnall (2013) are other key sources used in this literature review. The UNODC 'International Standards on Drug Use Prevention' (2015) summarises the scientific evidence on effective drug prevention interventions, through a review of systematic reviews and meta-analysis and assessment of primary studies. The research by Brotherhood et. al. (2013) was commissioned as part of the ALICE RAP, EC funded project, and reviewed high quality systematic reviews of primary studies which evaluated the effectiveness of policies and interventions which target substance use (alongside other addictive behaviours). Lastly, Stead and Angus (2004) is drawn on to provide findings on the effectiveness of drug prevention and education in schools. While this review is not systematic, it is comprehensive and complements the other systematic reviews cited in this document.
Defining drug prevention and education
There is no commonly accepted definition of 'drug prevention' in Europe. The European Monitoring Centre for Drugs and Drug Addiction ( EMCDDA) define this as: any policy, programme, or activity that is (at least partially) directly or indirectly aimed at preventing, delaying or reducing drug use, and/or its negative consequences such as health and social harm, or the development of problematic drug use ( EMCDDA, 2011). This applies to all psychoactive substances, both legal and illegal. Drug prevention activities can target whole populations, subpopulations, or individuals and may also address common factors that reduce vulnerability to drug use or which promote healthy development in general.
Drug prevention and education are often discussed interchangeably but there is a difference between the two. While drug education aims to provide information, facts, consequences and advice about drugs, upon which individuals can base decisions and make informed choices, its primary objective is not to change behaviour, as is the aim of prevention. However, prevention activities may include "prominent educational components" ( ACMD, 2015). Likewise, while the outcomes of drug education are more limited than some of the more comprehensive prevention programmes, drug education can also contribute to preventive outcomes (Thurman and Boughelaf, 2015). Lastly, what is delivered and termed as 'prevention' in schools may in reality be more akin to education.
Prevention science is a relatively new, multi-disciplinary field which has developed rapidly over the last forty years  . Its main aim is "to improve public health by identifying malleable risk and protective factors, assessing the efficacy and effectiveness of preventive interventions and identifying optimal means for dissemination and diffusion" (Society for Prevention Research, 2011). There is now a much better understanding of 'risk factors', those which put individuals at a greater risk of initiating drug use, and 'protective factors', those which contribute to making individuals less vulnerable to this occurring. Amongst the many factors associated with developing drug use (alongside other risky behaviours) are: biological processes, personality traits, mental health disorders, family neglect and abuse, poor attachment to school and the community, favourable social norms and conducive environments, and growing up in marginalised and deprived communities ( UNODC, 2015). Known protective factors to drug use and other negative behaviours include: psychological and emotional well-being, personal and social competence, a strong attachment to caring and effective parents and to school and communities that are well resourced and organised ( UNODC, 2015). Drug prevention can tackle the risk factors that increase a person's vulnerability to developing drug use, and build protective factors, building resilience, offering opportunities for alternative and healthier life choices and developing better skills and decision making abilities (Public Health England, 2015).
The EMCDDA classifies prevention types according to a scheme developed by Mrazek and Haggerty (1994). The categories are complementary to one another and replace the previously used categorisation of primary, secondary, and tertiary prevention (although this latter categorisation is still used in public health and is still relevant). This categorisation is based on the overall vulnerability of the people addressed - the known level of vulnerability for developing substance use problems distinguishes between the categories, rather than how much or whether people are actually using substances:
- universal prevention addresses a population at large and targets the development of skills and values, norm perception and interaction with peers and social life;
- selective prevention addresses vulnerable groups where substance use is often concentrated and focuses on improving their opportunities in difficult living and social conditions;
- indicated prevention addresses vulnerable individuals and helps them in dealing and coping with the individual personality traits which make them more vulnerable to escalating drug use.
There is also interest more recently in environmental prevention, interventions that do not use persuasion to change people's attitudes and behaviour, but instead use interventions that try to limit the availability of opportunities to use drugs, through national policies, restrictions and actions that affect social and cultural norms, e.g. drug driving policies ( EMCDDA, 2011; ACMD, 2015).
- environmental prevention addresses societies or social environments and targets social norms including market regulations.
There is support for the use of the US Institute of Medicine ( IoM) prevention classification system as a means of describing the form of prevention available, from EDPQS (2015) and the ACMD (2015). This classification system illustrates the continuum of services/interventions and provides a common language to describe prevention and assists in the planning, delivery and evaluation of activities. It contains the Universal, Selective and Indicated categories used by the EMCDDA above.
Figure 1. The Institute of Medicine model of prevention (1994; 2009)
Drug prevention is relevant across the lifespan, despite often being considered as most relevant to young people. As stated in the introduction, the main focus of this literature review is on universal approaches to drug education and prevention amongst young people, delivered through school curricula, although approaches beyond schools are also considered. Schools represent the most efficient way of reaching large numbers of young people, so represent the best setting for universal preventive interventions (Faggiano et. al., 2014).
The evidence base
The EDPQS (2015) state that whilst few people would argue with the view that prevention is better (and cheaper) than cure, much of what is done in the name of drug prevention is still not based on what 'works' or on what constitutes 'quality', and scarce resources are still being spent on ineffective approaches.
There is little clear evidence of 'what works' in drug prevention and the UK prevention evidence base is particularly poor ( ACMD, 2015). Evaluating prevention is difficult, in particular, measuring something that has not yet happened, and unpicking which intervention made the difference in the long term (Evaluation Support Scotland, 2016). Sumnall points out that evidence on what 'works' will be contingent upon; how prevention is defined, geography, the type of activities included, the outcomes specified etc. (Sumnall, 2016). In terms of geography, the international nature of most of the evidence (particularly from the USA), raises questions around the transferability and adaptability of programmes to the British context. The advantages and barriers to introducing North American prevention programmes to Europe is explored in depth in the EMCDDA's paper 'North American drug prevention programmes: are they feasible in European cultures and contexts?' ( EMCDDA, 2013). Long term behaviour change is difficult and expensive to measure, and so very few evaluations track participants for long follow up times. This report concludes that it is possible to transfer programmes but careful adaptation and evaluation is required, and success is not guaranteed. Scarce resources and opportunities mean that rigorous evaluations are often not conducted, especially in low to middle income countries ( UNODC, 2015). Many evaluations therefore focus on 'surrogate' indicators of substance use - short term outcomes, and intermediate measures such as knowledge and attitudes (see chapter 2).
Midford and Munro (eds., 2006) write that much of the robust evidence on 'what works' in drug education from the USA comes from studies that have evaluated the rather narrow goals of abstinence and delayed onset of drug use. For this reason, any reported drug use equates to programme failure, even though programmes may have had an influence on patterns of use or associated harm. They write "Most contemporary drug education research is simply not designed to explore if broader prevention benefits can be achieved" (Midford and Munro eds., 2006, p215). This is a sentiment echoed by Strang et. al. (2012) in their review of the evidence for effective interventions for a Lancet Addiction Series.
The quality of prevention studies and whether they contain biases is also important when considering the evidence of 'what works'. In certain studies that have shown statistically significant findings, often the effect is meaningless, e.g. a reduction of drug use frequency of 0.5 episodes in a month (Sumnall, 2016). The UNODC also highlight publication bias as an issue, whereby publications reporting positive results are more likely to be published than those reporting negative findings, which risks an overestimation of the effectiveness of drug prevention programmes and policies ( UNODC, 2015).
In contrast to the weak (although improving) evidence base on 'what works' in drug prevention, there is much stronger evidence on which prevention approaches are ineffective in improving drug use outcomes ( ACMD, 2015).
Structure of the report
Chapter 2 of this review focusses on schools-based drug prevention and education, how success is measured, the evidence of effectiveness for different approaches used in schools and other components necessary for effective drug prevention and education in schools. Chapter 3 explores effectiveness of drug education and prevention beyond the school setting, and considers the evidence to support peer led interventions and specific programmes for vulnerable young people. Chapter 4 discusses specific manualised and licensed prevention programmes and considers some of the issues and challenges involved in implementing these programmes in different contexts. Chapter 5 highlights the evidence for ineffective approaches to drug prevention and chapter 6 draws on recommendations from the literature for policy makers. Lastly, thoughts are presented on the implications of this review for drug education and prevention in Scotland, in particular the need to map prevention activity for young people being delivered in Scotland. This mapping can then inform an assessment of whether prevention and education being delivered to children and young people in Scotland resonates with the evidence on what is most likely to be effective, highlighted in this review.