Risultati della revisione Cochrane sull’efficacia degli interventi di prevenzione primaria dell’uso di sostanze illecite effettuati nelle scuole

[Machine translation] Intro Substance addiction is commonly described as a chronic relapsing condition. The experimental use of illicit substances generally begins during adolescence. Primary prevention interventions should be focused on reducing the frequency of first use, or blocking the transition from experimental use to addiction. In past years, many prevention programs have been carried out, but they have rarely been evaluated through appropriate study designs (randomized, controlled trials), and just as rarely have they been analyzed with adequate statistical methods. Even if many literature reviews have been published, none of them has so far taken into account all the characteristics of the studies (in particular the appropriateness of the study design and the peculiar characteristics of the interventions); for this reason, the effectiveness of the different programs remains to be demonstrated. Objectives of the review Evaluate the effectiveness of drug use prevention interventions carried out in schools compared with usual curricular activities or with different interventions, in improving knowledge, developing skills, promoting changes in attitudes, and preventing or reducing substance use among primary and secondary school students. Research strategy The research strategy developed according to the methodology suggested by the Cochrane Collaboration, was applied to the following databases: MEDLINE, EMBASE, ERIC, PSYCHINFO, Cochrane Central Register of Controlled Trials, ACP Journal Club, Cochrane Database of Systematic Reviews, Database of Abstracts of Reviews of Effects, Register of the Cochrane Drug and alcohol group, updated in February 2004, and to the SOCIOLOGICAL ABSTRACTS database updated to 2000. The bibliographic references cited in the articles have been checked and personal contacts with the authors have been made to identify other potentially relevant (even unpublished) studies. Selection Criteria All studies (RCT, CCT or prospective controlled studies) evaluating substance use prevention interventions carried out in schools were considered. Data extraction and analysis All abstracts, and then the recovered articles, were read and evaluated by two reviewers. The extraction of data from the articles was carried out separately by the two reviewers. Discrepancies in evaluation or data extraction were resolved by a third reviewer. The extracted data was entered and analyzed with RevMan software. The evaluation of the methodological quality of the studies was conducted using the checklist developed by the Alcohol and Drugs Group of the Cochrane Collaboration. The interventions were classified according to the educational objectives of the program (focused on skills, socio-affective and focused on knowledge) and according to other characteristics (method of administration of the intervention, type of administrators, length of follow-up, activation of the context). Results 31 studies were included, including 28 RCTs and 3 CPS (prospective controlled studies). 48 studies were excluded. Most of the included studies come from the United States (27 out of 31); in addition, most studies evaluate programs carried out on 6th-7th grade students, and the effect of the program is measured on variables at the post-test (immediately following the program itself). Results from randomized controlled trials (RCT) The results are stratified according to the educational objectives of the program. 1) Knowledge vs usual curriculum Programs focused on knowledge seem to significantly improve knowledge at the post test (SMD=0.91; 95% CI: 0.42, 1.39 - heterogeneity test: p=0.17) but show no significant effect on decision-making abilities, self-determination, attitudes towards cocaine and intention to use cocaine, when compared to normal curricular programs. 2) Skill vs usual curriculum Skill-focused programs, when compared to normal curricular programs, seem to have an effect on increasing knowledge (WMD=2.60; 95% CI: 1.17-4.03), decision-making abilities (SMD=0.78; 95% CI: 0.46-1.09 - heterogeneity test: p=0.09), self-esteem (SMD=0.22; 95% CI: 0.03-0.40 - heterogeneity test: p=0.32), resistance to peer pressure (RR=2.05; 95% CI: 1.24-3.42), on generic drug use (partially significant result: RR=0.81; 95% CI: 0.64, 1.02- heterogeneity test: p=0.30), on the use of marijuana (RR=0.81; 95% CI: 0.70, 0.94 - heterogeneity test: p=0.25) and on the use of hard drugs (RR=0.36; 95% CI: 0.13-0.98). No effect is evident for the other measured outcomes. 3) Socio-affective vs usual curricula Interventions with a socio-affective approach have a greater effect than normal curricular activities, in improving knowledge (SMD=1.88; 95% CI: 1.27, 2.50 - heterogeneity test: p=0.36) and decision-making skills (SMD=1.35; 95% CI: 0.79, 1.9 - heterogeneity test p=0.82). 4) Skills vs knowledge There is no significant difference between the two types of programs. 5) Skill vs socio-affective Programs focused on abilities seem to have a greater effect than socio-affective programs only with regard to self-esteem (WMD=1.90; 95% CI: 0.25, 3.55). 6) Socio-affectives vs knowledge Interventions with a socio-affective approach, when compared with interventions focused on knowledge, seem to be slightly more effective in improving knowledge (SMD=0.60; 95% CI: 0.18.1.03 - heterogeneity test: p=0.94), and decision-making skills (SMD=1.22; 95% CI: 0.33, 2.12 - heterogeneity test: p=0.11). Results from prospective controlled studies Controlled prospective studies fail to produce statistically significant results. Conclusions The results of this systematic review highlight the value of primary prevention interventions carried out in schools, based on the development of individual skills and competencies. In fact, they are effective in reducing substance use and promoting intermediate outcomes. However, the evidence of the effectiveness of such interventions is still weak and seems necessary: - replicate well-designed studies, with long-term follow-up, possibly randomized - conduct new studies to evaluate the effect of the individual components when added to the basic intervention (for example: effect of the administration of the intervention by peers, effect of reinforcement sessions, involvement of relatives or the social context). New studies should be evaluated with appropriate statistical analysis techniques.