Prevalence of school violence and use of alcohol and other drugs in adolescents

ABSTRACT Objective: to analyze the school violence suffered and practiced and its association with the use of alcohol and other drugs in adolescents between 12 and 18 years old. Method: the study sample consisted of 643 adolescents enrolled in six schools, who answered two self-administered questionnaires: “Global School-based Student Health Survey” and “Violence in School”. Statistical analysis was performed using the chi-square test and the degree of association between the variables was analyzed using the prevalence ratio. Results: the prevalence of school violence suffered and practiced was 62.2% and 51.9%, respectively. About 44.6% of the aggressors said they did not want to change their behavior. There was an expressive prevalence of alcohol use (16.5%), tobacco (15.7%) and illicit drugs (6.8%), and drunkenness (12.6%). There was a significant association between the violence suffered and the age group of 12 to 14 years old (p=0.001); (p=0.011) and education level in elementary school (p<0.001). In mothers with less than eight years of studies, the association was significant for the violence practiced (p=0.002). Conclusion: the study contributes to the aspects involved in school violence, which can subsidize actions and policies in this area.


Introduction
Violence is a problem that infiltrates society, threatening the healthy development of people. For these issues, it needs to be studied transversally. It is responsible for a large proportion of deaths in several countries, particularly among children, adolescents, and young adults. The exposure to violence can cause immediate physical injury that health professionals can handle. However, it can result in physical and mental health problems that are not often apparent to these professionals. Also, violence directly affects expenditures on health care; and indirectly, it generates stagnation of economic development, and increases inequities and deteriorates human capital (1) .
Violence is materialized in several social spaces and, in recent years, it has been frequent in the school environment. This contradicts the role of the school. School violence is expressed in a more explicit perspective of violence, such as aggression between individuals, and in the symbolic violence that occurs through the rules, norms and cultural habits of an unequal society (2) . drugs (3)(4) . The family emerges as essential in this debate by its implication in these behaviors presented by the adolescents, besides the consideration of the ecological gaze for the understanding of the violence (5)(6)(7) . Recent literature review has brought the factors associated with the perpetuation of violence, with poor impulse control; use of alcohol or other drugs; unhealthy relationships between caregivers and children; family economic stress; exposure to community violence; increase in social inequities were factors strongly correlated with the occurrence of violence among adolescents and young people (1) .
The present study lists as risk behaviors factors related to adolescents' lifestyle, such as alcohol and drug use, as well as to violent behaviors considered as predictors of the violent act. In Brazil, the analysis of these variables by the National Survey of School Health (PeNSE) conducted in 2009, using data from the Brazilian Institute of Geography and Statistics (8) , revealed that 71.4% of schoolchildren have ever tried alcoholic beverages. The use of alcohol in adolescence is a factor of exposure to health problems in adulthood, as well as significantly increasing the risk of the individual becoming an abusive consumer throughout life (9) .
According to data from PeNSE conducted in 2012, 15.9% of students suffered physical aggression in the last twelve months before conducting the research; in the PeNSE of 2015, there was an increase in the percentage of aggression to 17.3% (3)(4)10) . Comparing the PeNSE data for 2009, 2012 and 2015, there is an increase in prevalence in almost all variables related to situations of violence involving students, as well as the severity of violence experienced by adolescents (11) .
On the other hand, there is an association between the adoption of a healthy lifestyle by young people and the decrease in the relative risk of being frequently intimidated and victimized (12) .
In this dimension, it is fundamental to work the behavior of adolescents in the school environment; aiming to identify the behavioral risk factors and protect these individuals are exposed. Therefore, the present study aims to identify the prevalence of violence suffered and practiced by adolescents in the school context and to analyze its association with the use of alcohol, tobacco, and other drugs. The proposed approach is shown as an emerging health issue, considering the importance that the school institution occupies in adolescents' lives, as well as the seriousness of the violent incidents that occur in this environment involving these actors. It is emphasized that adolescence constitutes a "key period" of human development, in which actions that promote a healthy adult life can be implemented (13) .

Methods
This study is part of a study carried out with the objective of studying the violence suffered and/ or practiced by adolescents in the school context and the associated individual and environmental factors, in public schools in the city of Recife, state of Pernambuco, School, in 2013, and who accepted to participate in the research. Adolescents who were separated on medical leave, maternity leave or suspended; and those who did not meet the inclusion criteria were excluded. The probabilistic sampling method stratified by school was used to determine the sample size and to select the individuals, considering the estimated population size of 4,905 students enrolled in the six state schools of the studied neighborhood; the margin of error of 5.0%; the 95% confidence that the margin of error is not exceeded; the expected proportion of 50.0% for each response category, which maximizes sample size.
The total number of students in the sample (643) was distributed proportionally between the six selected schools and the three school shifts. All calculations for sample determination were performed through the Epi-Info program version 6.04d for DOS.
The data were collected from 02/07/2013 to 06/05/2013 by a nurse and author of this article, using two self-administered questionnaires already validated in Brazil. The first questionnaire, containing sociodemographic and behavioral questions, aims to evaluate the exposure to health risk behaviors in adolescents, being denominated Global School-Based Student Health Survey (14)(15) . The second questionnaire has school violence as its central theme and its objective is to evaluate the violence suffered, practiced and/or witnessed by adolescents within the school context (16)(17) .
The data collected were scanned in spreadsheets with double entry and the data bank was validated by a third person.
The variable outcome of the study was the occurrence of violence suffered and/or practiced in the school context, in the last two weeks that preceded the study. The type of violence was explored, who was the aggressor, who was the victim, the place where the violence occurred, how many times they were an aggressor or victim, whether they continue to attack or be attacked. The explanatory variables were selected from the Global School-Based Student Health Survey (14)(15) , being age, gender, education level, residing with the mother, residing with the father, length of study of the mother, alcohol consumption and other illicit drugs (age at the beginning of consumption, number of doses of alcohol, drunkenness, where the drug was obtained).
The results were analyzed through absolute and percentage distributions. To evaluate the existence of an association between two categorical variables, the Chisquare statistical test was used and when the conditions for its use were not met, the Fisher's exact test was used. To evaluate the degree of association between the variables, the prevalence ratio (PR) and the respective confidence interval (CI) were obtained. The margin of error was 5%, and the degree of association between the variables was assessed by the prevalence ratio and the respective confidence interval (CI 95.0%). The Statistical Package for the Social Sciences (SPSS), version 21, was used to carry out all the statistical analyzes of the results.

Results
The main sociodemographic characteristics of the sample analyzed were: slightly more than half (56.5%) of the adolescents were between 15 and 18 years oldthe others (43.5%) were between 12 and 14 years old; most of them were female (64.2%), single (93.6%), did not work (87.9%), and were self-considered as non-white  and most who answered affirmatively, 5.3% used drugs of this type one to two times.

Discussion
The prevalence of violence in the school context, in the condition in which the adolescents were victims of the aggression, was 62.2%. In Brazil, the estimated prevalence is 10% to 70% and, in the international scenario, from 8% to 60% (18)(19) . In spite of the diversification of the prevalence of violence in the school context, these studies highlight common central issues: school space is not immune to the presence of violence, and adolescents' exposure to school violence is a worldwide concern that has led researchers from several areas of knowledge to investigate the occurrence of this phenomenon, such as health and education. A recent study addressing this theme identified the fragility or absence of institutional links to address school violence, given the need to think about this problem in an expanded and articulated network (20) .

Recent research indicates that 70-80% of victims
of violence and their bullies are in the same class at school (23) . Also, school violence is still associated with males, corroborating studies in the area about bullying (24)(25) . However, the literature has shown that the girls have attacked as much as the boys, and in a direct way (17) . Students enter the school setting with efforts should take a multi-component approach to effectively addressing behavioral concerns. Programs focused on rephrasing normative beliefs while using skill-building techniques, may be best suited for improving behavior in the classroom and throughout the school (26) .
Concerning the violence among adolescents in the school context, two worrying facts revealed by the present study were the high incidence (44.6%) of aggressive adolescents who stated that they did not a climate of peace culture can act as a significant protective factor for violence among adolescents (27)(28)(29) .
The family context emerges as preponderant in the discussion of school violence; the family is shown as the first environment of socialization and internalization of emotions and behaviors, which will be experienced in other spaces (6)(7) . Although the instruments used in this study did not allow determining the number of nuclear families among the adolescents, it was identified that most of the adolescents live with the mother. This may indicate single-parent families and/or the existence of family conflicts. Literature review has brought about that family characteristics are determinant for the involvement of adolescents with bullying, especially single parenting; this fact is related to a possibility of less time for parent-child interaction and greater family stress (5) .
Another aspect related to family relationships was the significant association between violence practiced and low level of education of the mother; this issue corroborates literature in the area, "pointing out that education enhances a culture of tolerance and respect for human rights" (30) . In a study about the meanings attributed to violence in the school context by teachers, they emphasized the importance of work and a directed look at the family, given the relevance of this institution to the formation and behavior of children and adolescents in other social environments (31) . Despite this importance, the literature corroborates that family-school integration is still a challenge (20,31) .
The violence suffered by the adolescents showed association between 12 and 14 years old. The literature corroborates such findings regarding the various types of violence to which adolescents are exposed and warns that these adolescents may live and mean violence in different ways (27,32) . Therefore, it suggests the development of instruments that identify and allow more targeted interventions for these age groups.
Regarding health risk behaviors in the present study, 5.3% of adolescents used drugs and 16.5% consumed alcoholic beverages; in this last group, 12.6% of the individuals already got drunk. The PeNSE of 2012 revealed that 26.1% of 9 th grade students consumed alcoholic beverages in the last 30 days before the survey, with drunkenness reported by 21.8% of respondents (33) .
The PeNSE of 2015 showed that 55.5% of the students in the 9 th grade of elementary education responded positively, experimenting with alcoholic beverages (10) . alcohol consumption in adults as part of a comprehensive approach to preventing alcohol-related harm (35) .
Another health risk behavior analyzed in this study was tobacco consumption. About 15% of adolescents already smoked some form of tobacco, and 61. Rev. Latino-Am. Enfermagem 2019;27:e3110.
Tobacco is one of the most important determinants for triggering chronic diseases. According to the World Health Organization, tobacco leads the ranking of causes of preventable deaths in the world. Also, the early onset of smoking is associated with an increased chance of using other substances harmful to health, such as alcohol and illicit drugs (36) . In the city of Belo Horizonte, Minas Gerais, Brazil, the prevalence of smoking among adolescents and young adults was 11.7%, and the main factor associated with tobacco use was excessive alcohol consumption (37) .
The World Health Organization also highlights that the use of alcohol and tobacco by individuals under 14 is associated with increased risk of school dropout, aggression, suicide and alcohol intoxication, and the mental health, with a number of negative consequences in the short and long-term (36) . Systematic review of literature on the influence of the social network on the smoking behavior of adolescents has brought about that social isolation is related to smoking among adolescents.
Also, peer selection and influence contribute to the initiation and maintenance of smoking in adolescents (38) .
Thus, the importance of actions among peers is observed, essentially in the privileged school context.
In addition, the look at family support is relevant. A study that analyzed predictors of smoking in young adults from smokers and non-smokers showed that low perceived levels of family social support were a critical factor for smoking among smokers and non-smokers.
The association between alcohol use and depressive symptoms was also relevant (39) .
Finally, some limitations of the study are highlighted. Despite the reliability and validity of the research, the transversal design does not allow specific temporal analyzes for school violence. The study site was a specific neighborhood, which preserves particularities, limiting the generalization of the findings.
The instruments used also do not allow bringing some familiar characteristics, like the number of nuclear families, preventing associations in this scope.

Conclusion
The results obtained in this study showed the high