This study aimed to prospectively investigate the potential relationships between scholastic factors with IDM in older adolescents. The analyses revealed some important findings that should be highlighted. First, scholastic factors were systematically associated with IDM, with poorer scholastic achievement in adolescents who reported illicit drugs use at baseline and follow-up. Second, behavioral grade, and absences from school at baseline were predictors of initiation of IDM in the following period. Therefore, the initial study hypothesis was confirmed.
4.1. Scholastic Variables and the Consumption of Illicit Drugs
The problem of the education-drug connection has been within the scope of the scientific and professional debates for more than five decades. While the first studies were mostly oriented towards identifying patterns of drug use in children and adolescents at different levels of schooling [33
] and young people’s knowledge of drug use [35
], later studies were more oriented towards identifying factors related to drug use [36
], as well as those variables specifically explaining school success/failure (i.e., school performance, educational failure) [37
]. Currently, interest in the relationship between educational success/failure and the consumption of illicit drugs is evident [38
Our results showed specific relationships between scholastic variables (educational achievement) and the use of illicit drugs. In brief, in both the cross-sectional analyses (i.e., at baseline and follow-up), those adolescents who performed poorly in school were more likely to consume drugs; in terms of this matter, our results are consistent with previous studies on this problem [15
]. However, the authors share the opinion that two findings on the association between scholastic variables and the consumption of drugs deserve special attention. First, the association between scholastic variables and drug consumption were much stronger for follow-up than for baseline testing; and second, the number of school absences was the strongest predictor of drug use both at baseline and follow-up-testing.
Although studies have regularly confirmed a negative association between these variables [15
], it seems that the strength of the relationship between scholastic variables and IDM in adolescents increased between the testing waves and was stronger at the end of the period of high school education. We were not able to find any study that examined this problem using the same experimental approach (i.e., calculating associations between educational variables and drug use at two-time points), and consequently, we were not able to compare our results with previous studies. However, based on the prevalence and dynamics of changes in studied variables, we can offer some specific explanations for our findings.
Briefly, the prevalence of drug use over the course of the study evidently increased (from 4% to 7%). Consequently, for simple statistical reasons (i.e., a higher prevalence and, therefore, a more balanced number of participants in the “affected” and “non-affected” groups), it was more likely that logistic regression would reach statistical significance at follow-up than at baseline measurements. Additionally, the higher variability in educational achievements at the follow-up measurement could have influenced the statistical calculation similar to what was previously explained for the prevalence of drug consumption.
Previous studies regularly used GPA as a general measure of students’ academic achievement, but our approach using some “non-standard” variables of scholastic achievement (e.g., behavioral grade, absences from school, unexcused absences from school) seems to be appropriate [39
]. Indeed, the strongest predictor of drug consumption was “school absence”, and those students who reported more school absences (irrespective of whether those absences were formally recorded as “excused” or “unexcused”) were at risk of consuming illicit drugs at baseline and follow-up. These results additionally confirm our previous discussion on the possible background of the stronger correlation between the variables at the follow-up measurement. In brief, school absences must influence other educational measures (GPA mostly), but this takes time. Therefore, correlations between GPA and drug consumption were not identified until follow-up.
Although studies have frequently examined the association between scholastic variables and the consumption of illicit drugs, the causality of this relationship is still relatively unknown [38
]. Theoretically, it is possible that the consumption of drugs resulted in low academic achievement because of the negative effects of drugs on cognitive function, which consequently resulted in low learning capacity [41
]. For example, smoking marijuana, which was the most frequently reported type of drug consumed in this sample (Figure S1
), impairs short-term memory and attention. Consequently, one of the possible side-effects of marijuana usage is related to impaired educational attainment in adolescents [41
Indeed, according to physiological influences of drugs on learning capacity, it seems reasonable to identify illicit drug consumption as the cause of low academic achievement in adolescents. However, it is also possible that causality should be interpreted based on the theory of social influence [44
]. It is known that adolescents who are not well prepared for school duties often skip classes [45
]. This puts them into “out-of-school” situations, where they are more likely to communicate with peers who consume substances, including those who use illicit drugs. Consequently, this increases the likelihood that they will start consuming drugs, as an effect of “low academic achievement” that results in out-of-school situations. Both explanations on the cause-effect relationship between the consumption of illicit drugs and low scholastic achievement are reasonable. Not surprisingly, the authors who examined these issues in cross-sectional studies repeatedly stated that for a more profound interpretation, prospective analyses are needed [6
This is one of the first studies in southeastern Europe, and it is probably the first one in the former Yugoslav territory to prospectively examine the problem of the initiation of IDM in adolescents [1
]. Therefore, our results on the significant influence of two educational variables (i.e., behavioral grade and school absence) on the initiation of drug use are novel to some extent. These findings have two important implications with respect to our study aims. First, results directly support our previous consideration of the applicability of “non-standard” educational variables for evaluating adolescents who are at a higher risk for consuming illicit drugs. Second, these findings clearly indicate that educational failure is the cause of the initiation of consumption of illicit drugs for adolescents over their last two years of high school education. While the first issue (i.e., the applicability of non-standard educational variables in evaluating at-risk adolescents) has already been discussed, in the following text, we will focus on the causality between scholastic variables and the initiation of drug use.
The variables of scholastic achievement that were found to be predictors of the initiation of drug use are specific. While the number of absences from school is a relatively objective measure, behavioral grade is a result of teachers’ personal evaluations of students’ behavior during a period of time (i.e., semester) [46
]. From our standpoint, the fact that behavior grade was a significant predictor of the initiation of drug use is highly encouraging because it points to teachers’ high pedagogical competences. In brief, this type of evaluation of scholastic success/failure includes different non-formal indices of the students’ behavior, such as their attitude towards their peers and school staff, their approach to formal and non-formal school duties, and their involvement in extracurricular activities [46
]. Behavioral grade is scored only by the class principal, who is the teacher who is responsible for one class per academic year, and in most cases, who follows the same class during all four years of high school. Throughout this period, class principals become familiarized with their classes, and obviously are capable of objectively recognizing different kinds of behavioral failures of their students.
4.2. Limitations and Strengths
The main limitation of this study is that it was based on self-reported data. Therefore, participants may have tended to provide socially acceptable answers. However, we believe that the study design and our experience from previous studies in the protection of the anonymity of the participants reduced this possibility. Additionally, this study did not consider some important educational variables, such as college plans, which could be useful in the discussion and interpretation of the relationships between educational success/failure and drug consumption/initiation. Additionally, we should not ignore the possible effects of some confounders that were not observed in the investigation (e.g., parental educational status, parental employment, parental IDM). However, we believe that the variables included in this study as possible confounders (parental monitoring, SES) at least partially covered the aspects of the confounding indices that were not evaluated.
This is one of the first studies conducted in southeastern Europe and probably the first one in the territory of former Yugoslavia to prospectively examine the association between educational and sport variables and drug consumption and initiation of drug use in adolescents. The prospective study design allowed not only the identification of this relationship but also the interpretation of the causality between the studied variables, which is an important strength of this investigation.