Trilhos - Evaluation of a developmental prevention program

BACKGROUND
Trilhos is a programme aimed to promote personal and social skills for students between the 7th and 9th grades, evidence-based, build following the guidelines for prevention in the school setting. It is a manualised programme, using interactive methods to promote decision-making and problem-solving skills, emotional regulation and communication. Trilhos is delivered by teachers and, in order to ensure the quality of the implementation, teachers receive training and technical support. 
METHODS
Data was collected in a convenience sample of 703 students between 7th and 9th grades with an average age of 13.5 (SD=1.07).Research design is quasi-experimental, with no control group, and pre and post-test data collection (beginning and end of the school year). The mediating variables are assessed using the Bar-On Emotional Intelligence Scale (youth version, adapted by Candeias et al., 2011), the Interpersonal Competence Questionnaire (adapted by Assunção et al., 2010), and the Risk Perception Scale. Behavioural variables are assessed through self-reported measures of psychoactive substance use (throughout life, in the last year and in the last month).The data was analysed using SPSS (version 24). Non-parametric tests for paired samples were used: Wilcoxon test (mediating variables) and McNemar's test of differences in proportions (behavioural variables).
RESULTS
Regarding the mediating variables, there were no significant differences between the pre- and post-test in risk perception. There were no significant differences in the overall score of the emotional intelligence scale. However, there were significant differences in the subscales of adaptability (p=0.003), general humour (p= 0.007), interpersonal (p=0.020) and stress management (p.<0.001). In interpersonal competence, only the deepening relationships subscale did not show significant differences, while the differences in the overall average (p=0.016) and in the subscales initiating relationships (p< 0.001), emotional support (p< 0.001), conflict management (p=0.039) and negative assertion (p=0.014) were significant. Regarding the behavioural variables, there were only significant differences in alcohol experimentation (p=0.007).
CONCLUSIONS
The absence of a control group is an important constraint in this analysis. However, it can be considered that the programme contributes to the development of the mediating variables. Regarding behavioural variables, it is difficult to draw conclusions without a comparison group, however, the fact that there was an increase only in alcohol experimentation and not in all the other prevalences for alcohol, tobacco or cannabis use, although with due reservations, points to the achievement of the behavioural objective. These results are in line with prevention research that emphasises the importance of developmental interventions as a preventive strategy, especially when combined with environmental interventions.