Artigos

Conteúdo (article):

Title: Physical violence and associated factors in participants of the National Student Health Survey (NSHS)

RESUMO

Objective: To analyze the association between physical violence in students with socioeconomic factors, family context, mental health, individual risky behaviors, and unsafe environment. Method: Study with data from the National School Health Survey (NSHS) in 2015, with 9th graders. The outcome variable was the report of involvement in a physical struggle and the associated exposure variables used were related to socioeconomic and demographic conditions, family supervision and support, mental health, risky behaviors, and unsafe environment. Multivariate logistic regression with a hierarchical approach was used in the analyzes. Results: The prevalence of involvement in a fight was higher in boys (30.2%; CI 29.3 - 31.0) than in girls (16.7%; CI 16.0 - 17.4). In both genders, there was a greater chance of involvement with physical violence when using drugs, missing classes, sedentary lifestyle, insomnia, loneliness and insecurity at school or in the community and, especially, when victimized by family aggression, OR 2.59 (CI 2.31-2.90) in boys and girls OR 2.42 (CI 2.17-2.71). There was a reduction in the chance of involvement in physical violence in boys because they were working and, in girls, when they study in a private school, having their problems and concerns welcomed by their parents or their participation in school activities.

Keywords: Violence, Adolescent; Adolescent behavior; School health


INTRODUCTION

The search for challenges, independence, and experimentation in adolescence is frequent and violence is one of the risk behaviors most addressed by the literature and health programs in this stage of life1.
According to the WHO, accidents and violence represent the biggest public health problem in children and adolescents in developing countries2. In 2009, the notification system for extra-family and community domestic violence was added to the Ministry of Health Information System for Notifiable Diseases, according to the Child and Adolescent Statute (ECA in Portuguese) and the data showed that in 2011, the age group between 15 and 19 years old was the second in which there were more calls for violence in the country3.
The Global Burden of Disease Study4 carried out annual assessments of diseases, sequelae, and risk factors between 1990 to 2013 in 188 countries and interpersonal violence was the fourth leading cause of death in adolescents aged 15 to 19 years old. Aggressive behavior can lead to several losses for the adolescent, his family, and community and among them, medical care costs, morbidities with permanent sequelae, learning disorders, school absenteeism, feelings of fear, and death stand out5. Also, acts of violence directly interfere with the quality of life of an entire society6.
Violence in adolescence has been studied from different angles with different types of variable outcomes and in this study, we considered the physical fight or combat, more socially acceptable because it does not involve illegal elements, such as the use of weapons.
As for the prevalence of involvement in fights and physical combat, a study with 161,082 students from 35 countries in North America and Europe showed values that varied from 37% to 69% in boys and 17% to 32% in girls7. In Brazil, the prevalence ranges from 16.2% to 20.9%8,9.
Studies carried out from the last three editions of the National School Health Survey (NSHS) demonstrated the growing temporal trend of situations of violence, as the prevalence rates increased from 2009 to 2015 in all the violence indicators evaluated.10 There was growth from 6.4% to 12.8% (average variation of 1.1) of the number of absences from classes due to insecurity on the school path, the prevalence of absences due to violence in the school environment increased from 5.5 to 9, 3, (average variation of 0.7), fights over firearms ranged from 4.0 to 5.6 (average variation of 0.3) while fighting with cold weapons from 6.1 to 8.2 (variation average of 0.4) and physical aggression in the family environment from 9.5 to 16.2 (average variation of 1.1). We found different patterns of growth of violence and factors of greater vulnerability in the studies, such as studying in public institutions and being male10.
The interaction of factors in the individual, relational, community, and social spheres can influence the development of youth violence. Some general factors are the difficulty in solving personal problems, young people who have suffered abuse or aggression, lack of parental supervision and monitoring, dating and friendship relationships, inserted in a hostile and violent community, educational opportunities, financial support, and public policies5.
International cross-sectional11 and longitudinal12 studies show the association between physical violence in youth and risky behaviors and emotional state, highlighting the use of illicit drugs, alcohol, tobacco, early sexual initiation, poverty, high-crime neighborhood, delinquency of peers, low supervision and family support, parental conflict, missing classes, low school performance, hyperactivity, bullying, antisocial behavior, insomnia, depressive symptoms, suicidal ideation, and sedentary lifestyle.
Thus, this study aimed to analyze the association between the physical struggle in schoolchildren with socioeconomic, family, mental health, risky behaviors, and unsafe environments in a national context.

METHOD

This is a cross-sectional study that used secondary data from the 2015 National Student Health Survey (NSHS )13.
NSHS of 2015 is a school-based study and the population analyzed was students from the 9th grade of high school. The sampling took place by clusters, in two stages - schools as primary units and classes, as secondary units, selected at random. Proportionally, students from public schools (municipal, state, and federal) and private schools participated. We excluded participants who did not answer questions about gender and age, totaling 102,072 valid questionnaires13. Data collection was carried out in 2015, through a structured and self-administered questionnaire presented to the student, applied by IBGE technicians on smartphones13. The organization, planning, and resources for the development of the research came from the Brazilian Institute of Geography and Statistics (IBGE), in partnership with the Ministry of Health and support from the Ministry of Education13.
The dependent variable in this study was the report of involvement in a physical fight, whether as aggressor or victim, at least once in the last twelve months. The exposure variables evaluated were grouped into five blocks, with socioeconomic, demographic, and school-related conditions allocated in the distal block of the model; the variables of risk behaviors, mental health, and family supervision/support allocated in the intermediate blocks, and the characteristics of the presence of an unsafe environment, proximal to the outcome.
As for the socioeconomic block, the variables were age, gender, racial color, maternal education, having monthly housekeeping and the number of bathrooms with shower at home, having paid work or not, number of residents in the house, and type of school, whether public or private.
The variables of context and family support were living with the father and/or the mother if in the last 30 days the parents or guardians knew what the adolescent did in their free time, if they understood their problems and concerns, if there were duties or tasks and if the students ate meals with their families.
Regarding mental health, the variables were the feeling of loneliness, insomnia in the last 12 months, and the existence of close friends.
As for the behavioral aspects, we analyzed drug consumption - alcohol, tobacco, and illicit drugs - in the last 30 days, physical activity in the last seven days, and skipping classes at least once in the last 30 days without the parents\' knowledge.
Regarding physical activity, we used the International Physical Activity Questionnaire (IPAQ)13 and we considered active the adolescent who practiced 300 minutes or more of activity per week.
Regarding the variables on insecurity, family aggression, community insecurity on the way home from school, and insecurity at school were evaluated in the last 30 days.
The association between exposure variables and physical violence was estimated by Pearson\'s χ2 test, with a significance level of p≤0.05, the magnitudes were measured by Odds Ratio with 95% CI obtained through multivariate logistic regression, having the category “never fought” as a reference.
The selection of exposure variables associated with physical violence with p ≤ 0.20 obtained in univariate logistic regression analyzes was included in the multivariate analysis, and they were placed in blocks in the model, where each of the five blocks corresponded to a hierarchical level.
For the final model, the backward method was used and the variables associated with the outcome remained with p≤ 0.05.
To detect the existence of multicollinearity (perfect or approximate linear dependence between at least two explanatory variables), in each of the adjusted multivariate models, we analyzed the correlation matrix between the pairs of estimated coefficients. In these matrices, high correlations (usually greater than 0.80-0.85) indicate a possible problem of collinearity. For each estimated model, the respective matrices were analyzed, and no important signs of collinearity were detected.
We used the Stata statistical package (version 13.1) and the “svy” command suitable for data analysis obtained by a complex sampling plan.
Due to a large number of information losses on maternal education (around 25% of responses not obtained), this variable was not considered in the multivariate analysis.
NSHS was approved by the National Research Ethics Commission of the Ministry of Health, opinion No. 1,006,467, of March 30, 2015 (IBGE, 2016). To participate in the research, students should agree to the Informed Consent Term, located on the first page of the smartphone.

RESULTS

In the sample, young people were predominant between 14 and 16 years old (78%), absence of a housekeeper (90.5%), with only one bathroom with shower (61.7%), from a public school (85.6%), whose mothers had completed high school (30.9%). Family aggression was present in 14.4% of the assessed population, while the aggression at school or on the way home/school was found in 9.5% and 11.4% of students. Most never used legal drugs (76% for alcohol and 93% for smoking) or illicit drugs (95.9%) and more than 79% of students were classified as sedentary. Among the evaluated students, 89.9% lived with their mothers and 43.8% reported welcoming their parents when there were problems and concerns. Also, 80.7% of adolescents ate meals with parents or guardians, and 31.8% of parents or guardians checked their homework - data not shown in the table.
The prevalence of involvement in fights, one or more times in the last 12 months (table 1), was 23.3% (CI 22.6 - 23.8), higher in boys 30.2% (CI 29.3 - 31.0) than in girls 16.7% (CI 16.0 - 17.4) (Table 1).

In the bivariate analysis, there is a higher proportion of boys with reports of physical violence when they lived alone, having a housekeeper and with two or more bathrooms, a mother with higher education, young people who already worked, studied in a private school, history of family aggression at home and insecurity in the surroundings and at school, drug use behaviors and missing classes, not living with the mother or father, not perceiving support and supervision from the parents and feeling alone. The results in the girls were similar, except reports of a fight when they were black or of other races, being the daughter of an illiterate mother or with incomplete elementary school, while in boys the violence was greater when mothers with higher education, studying in public school, instead of private like the boys, and not having friends (Tables 2 and 3).
There were eight models of logistic regression based on the hierarchical model. In boys, there was an adjustment for the variables age, several people living in the house, having a housekeeper, bathroom, working adolescent, and type of school (model 1); the variable resident/house was excluded (model 2); the variables related to family supervision/support were inserted in model 3, all maintained a positive association according to the bivariate analysis; insomnia, feeling of loneliness and number of close friends, related to mental health (model 4); the variable about friends was excluded (model 5); the consumption of alcohol, drugs, and tobacco was included, if classes and physical activity were missed (model 6); if he lives with his mother was excluded (model 7). The final model included the variables family aggression, insecurity at school, and on the way home/school (Table 4).
The models in the girls were adjusted for the variables age, race/skin color, number of people living in the house, bathroom, working adolescent, if paid work, and type of school (model 1); the variable living in the house was excluded (model 2); the family supervision/support variables (model 3); the variable if he lives with his mother was excluded (model 4) because it lost the strength of association; insomnia, feeling of loneliness and number of close friends were inserted in model 5, the variable about friends in model 6 was excluded; the variables alcohol consumption was inserted in model 7, drugs and smoking if classes and physical activity were missed. All variables included in the model maintain the strength of association. The final model included the variables family aggression, insecurity at school, and on the way home/school (Table 5).
In the adjusted analysis, both boys and girls increased the chance of involvement in physical violence in the last 12 months when consumption of illicit drugs, 1.55 (1.26 - 1.90) in boys and 1.70 (1, 32 - 2.19) in girls, alcohol consumption 2.49 (2.23 - 2.78) and 2.82 (2.44 - 3.26), smoking 1.43 (1.16 - 1, 72) and 2.15 (1.75 - 2.65), absence from classes without parental permission 1.31 (1.18 - 1.46) and 1.31 (1.16 - 1.49), physical inactivity 1.35 (1.24 - 1.48) and 1.33 (1.17 - 1.50), insomnia 1.27 (1.15 - 1.35) and 1.42 (1.27 - 1, 59), feeling of loneliness 1.24 (1.04 - 1.38) and 1.13 (1.01 - 1.26), insecurity at school 1.68 (1.41 - 2.00) and 1, 29 (1.09 - 1.54), from home/school 1.24 (1.07 - 1.43) and 1.48 (1.27 - 1.72) and family aggression 2.59 (2 , 31 - 2.90) and 2.42 (2.17 - 2.71) in girls, a condition that contributed most to the physical violence of young people, in addition to alcohol consumption.
Also, boys were associated with the involvement of fights with studying at a private school 1.52 (1.35 - 1.72) and living in a residence with two or more bathrooms 1.56 (1.29 - 1.90), suggesting being of a higher social class and not living with their father 1.12 (1.04 - 1.23) (Tables 4 and 5).
On the other hand, there was a reduction in the chance of involvement in a fight in boys, the fact that they were working 0.76 (0.69 - 0.85) and, in girls, the acceptance of problems and concerns by parents 0.89 (0, 71 - 0.93), parents\' knowledge about their daughter\'s activities when in free time 0.81 (0.71 - 0.93) and the fact that they studied in a private school 0.78 (0.69 - 0.89 ) (Tables 4 and 5).



DISCUSSION
The main factors associated with physical violence in 9th-grade schoolchildren found in this study constituted family aggression more than once in 30 days, understanding it as the youth\'s coexistence in an unsafe family environment and those of a behavioral nature evidenced by licit and illicit drugs, physical inactivity and absence from classes.
Family aggression can show a hostile environment at home, favoring the youth\'s learning of models of aggressive behavior developed both within the family and in society. The experience of children and adolescents in situations of violence is related to a greater propensity to have behaviors that reproduce aggressiveness in current or future relationships17.
Both national and international studies corroborate the relationship between aggression in the intrafamily environment and violent behavior in childhood and adolescence. In Fortaleza, CE, we identified that being beaten at home increased youth involvement in interpersonal violence as an aggressor by 13%17. In São Gonçalo, RJ, a survey showed that students who reported severe psychological, sexual or physical violence by family members or people in significant relationships increased the chance of being involved in violence in the community and in the school environment more than three times than in students who did not report domestic violence19. In Minnesota, USA the results were similar even when assessing physical or sexual abuse by a family member or not, or simply witnessing physical abuse in the family20.
Several forms of manifestations of violence can contribute to the feeling of insecurity in schools, such as bullying, use of firearms or cold weapons, physical fighting, robberies, damage to property, drug use, and violence in the surrounding neighborhood or community that the school is inserted10,22. According to studies, the exposure to violence at school and in the community can compromise academic performance, school attendance, causing physical and emotional damage22 and the feeling of insecurity is commonly seen more frequently in public schools, for reasons of location or aspects related to the school organization 10.
The consumption of alcohol and other drugs has a high prevalence in adolescents. Although there has been a reduction in the consumption of cigarettes at this stage of life, in recent years, there has been an increase in the consumption of alcohol and illicit drugs, which has shown an association with violent behaviors23, as in this study.
A survey in the USA24 showed that students who did not use marijuana and alcohol were the least involved in fights (77.3%). In Thailand, there was an increase of 200% in involvement in fights without injuries and 40% in fights with serious injuries in students who drank25.

In the school-based survey carried out in Argentina, the Global School-Based Health Survey (GSHS), investigated through self-report about high alcohol consumption in the last 30 days in 1,328 high school students aged 13 to 15 years old. About 55 % of boys and 49.2% of girls were almost twice as likely to be involved in physical violence than students who reported not consuming alcohol40.

Risky factors for physical struggle in adolescence can occur simultaneously, which can lead to an increase in fighting at this stage. In a cross-sectional study, using secondary data from the National Youth Risk Survey, in a sample of 12,404 high school students in the United States, the use of marijuana, alcohol, and their consumption and their relationship with the physical struggle was analyzed. Students who did not use these substances were the least involved in fights (77.3%). Physical struggles were more prevalent among adolescents who reported alcohol and marijuana use (59.62%), among students who used marijuana (49.85%) and alcohol (40.97%)24.
A survey carried out in 10 cities in Latin America evaluated patients in emergencies due to intentional injuries who drank alcohol before the injuries. The relative risk was 5.6 times higher in people who drank 0 to 2 doses of alcohol, and in about 32.4% of injuries were reported to cause alcohol consumption26.
National studies that used the 2009 NSHS sample also point to a positive association in male students who consumed alcohol OR adjusted = 2.21, 95% CI and drugs OR adjusted = 2.75, 95% CI with involvement in physical violence27.
In Brazil, urban violence was evaluated either as an aggressor or as a victim in people between 15 and 64 years old, and the prevalence was higher in cocaine users 19.7% and in people who frequently consumed alcohol 18.1%28.
Predictors of alcohol consumption, illicit drugs, and physical violence in youth have been investigated. However, there has been difficulty in establishing causal relationships between these behaviors, due to the complexity in defining the order of these behaviors28,29.
The compromise of mental health with the presence of insomnia and feeling of loneliness also increased the chance of interpersonal violence, which can be indicative of anxiety disorder, depression, or a Common Mental Disorder. Research shows that adolescents with depressive symptoms, feelings of loneliness, and anxiety are more likely to be involved in aggressive behaviors either as victims or as aggressors30,31,32.
A cross-sectional survey that used school-based data in Saudi Arabia investigated the relationship between physical violence in the last year and the mental health of 9073 students through self-report research. A positive association was found between symptoms of depression and anxiety and interpersonal violence. Schoolchildren who were involved in a fight were 1.7 times more likely to have depressive symptoms, while schoolchildren who reported signs of anxiety were 1.48 times more likely to fight and 1.84 more likely to have mental health impairment when compared to students who did not mention physical struggle32.
In Brazil, data from the 2015 National School Health Survey (NSHS) showed an association between depressive symptoms, feelings of loneliness and insomnia with the consumption of legal and illegal drugs, showing the tendency for the group of risk behaviors in adolescence33
In this study, the absence of a father figure at home in the boys was associated with physical violence, which may show the lack of a nuclear family model. A study in São Paulo showed a higher prevalence of actions involving crime and violence, risky sexual behavior, and consumption of drugs and alcohol in young people who do not live with either parent and, then, young people who lived with only one parent, justifying the importance of family structure for the prevention of risky behavior in young people.34
In the girls, those who perceived receiving family support daily were less involved in a physical struggle, which is similar to other investigative works that emphasize the importance of family support as a factor that protects young people regarding risk behaviors.31, 35
A better economic condition in boys evidenced by the proxy variables studying in a private school and having two or more bathrooms at home increased the possibility of involvement in fights. In the bivariate analysis, a higher prevalence of mothers with higher education was observed in boys who reported a fight, and this context indicates that a higher social class has been controversial in studies evaluating any type of violence in young people. In a study conducted with data from NSHS 2009, there was no difference regarding involvement in fights between public and private schools21.
Lower social class markers have been linked to violence in adolescents, as shown by Kipping et al. (2014), who observed in a cohort study in the United Kingdom that the socioeconomic status analyzed through maternal school, family income, and parents\' social class, the lowest social level was associated with violent behavior and several other harmful behaviors to adolescent health36.
The private school space in general offers more security, disciplinary control, and a better structure for the learning and development of young people37. Although it seems contradictory, some studies have also shown a higher prevalence of risky behavior in students from private schools than in public schools, and the most observed risk behaviors were reckless driving, involvement in fights, alcohol, cigarette, and illicit drugs38,39.
Albuhairanl et al. (2017)32 found an association between higher maternal school level and students\' reports of involvement in fights in the last year, in Saudi Arabia. The researchers report the unusual nature of these findings and suggest that it could be because the mothers with a higher level of education represent a small part of the sample (23.3%)32.
Dropout rates are significantly higher in public school students, especially at the end of elementary school and in high school, and part of the students involved in fights may be out of public school, leading to a sample of this study with disproportionate loss of students in the variable physical struggle40.
We identified the variable of the work of adolescents in this study as a protective factor for the involvement of interpersonal violence in boys, which is corroborated by another national study41, but it has been a variable little evaluated in studies of this nature.
There was also an increase in physical violence behaviors in students who missed classes without parental permission in six countries in the Western Pacific and in students who missed classes without parental permission. They were more likely to fight, OR = 1, 72 (1.51 - 1.95) than students who did not miss classes11. Students who miss classes are more likely to have other health risk behaviors such as the use of alcohol and drugs, the simultaneity of health risk factors increases the chance of involvement in situations of violence8,11.
Research on physical struggle involving schoolchildren oscillates widely in the variables associated with this outcome. However, some behaviors and profiles presented in the literature have a high frequency of association with physical violence. They are male gender, consumption of alcohol, cigarettes, and drugs, skipping classes, living in places with high levels of violence, having depressive symptoms, conflicts in the family system, and little parental supervision11,15,32,42. These findings are similar to the variables found in this study that were associated with physical violence.
The literature has shown that contextual and individual factors influence aggressive behavior, and interventions to contain violence must be based on scientific evidence that shows violence in young people who are also under school supervision, a space with great potential for observation and intervention, reducing conflicts. Understanding the vulnerabilities in which adolescents are inserted is essential to formulate strategies for health promotion according to their needs and experiences to provide subsidies for directing public policies, promoting the improvement of health indicators, and preventing risk behaviors at this stage.
In addition to the cross-sectional design that does not propose causal inference, a study limitation was the greater possibility of information bias, considering the age group involved, especially when answering subjective questions or who need knowledge and retention, such as the mother\'s education level.
NSHS only evaluates children who are enrolled and present at school but school dropout in young people is high in the country, especially those from public schools and with older age, at the end of elementary school or in high school. Despite the limitations, this study has a good representation of adolescents in Brazil and provides contributions for understanding the factors associated with involvement in schoolchildren\'s fights.
Most of the studies evaluated on the involvement in physical violence in students nationwide are analysis of aspects related to the consumption of alcohol, drugs, bullying, and socioeconomic conditions, and there are still few types of research that gather factors associated with mental health, social interaction and family support and domestic violence, as in this study. The importance of NSHS for the investigation and planning of actions to prevent violent behavior in schoolchildren is highlighted, as it allows the analysis of several contextual and individual variables.
The main factors associated with violence, such as physical struggle, reported by boys and girls, 9th graders are due to the feeling of insecurity, mainly the intrafamily resulting from family aggression, but also that perceived in the community and at school and the factors of natural behavior such as the consumption of legal drugs (tobacco and alcohol), illicit drugs, physical inactivity and skipping classes. The impairment of mental health was also associated with involvement in fights, and in boys, the absence of a father at home.
The factors related to involvement in fights in girls and boys were similar, but the conditions that reduce them in boys were working and in girls the perception of receiving support from parents and studying in a private school.
Multiple factors are associated with involvement in physical violence in adolescents, either as a victim or aggressor, showing that an unhealthy lifestyle for young people and greater social vulnerability, reinforcing the need for programs that address the complexity and coexistence of related causes.

References

1. Cruzeiro AS, Silva RAH, Horta BL, Souza LDM, Faria AD, Pinheiro RT, et al. Prevalência e fatores associados ao transtorno da conduta entre adolescentes: um estudo de base populacional. Cad. Saúde Pública 2008; 24 (9): 2013-2020. http://dx.doi.org/10.1590/S0102-311X2008000900007.
2. WHO. World Health Organization Improving Health trough schools: national and international strategies; 1999.
3. Waiselfisz JJ. Mapa da Violência 2013: Homicídios e Juventude no Brasil. Brasília: Secretaria-Geral da Presidência da República, Secretaria Nacional de Juventude, Secretaria de Políticas de Promoção da Igualdade Racial, 2013.
4. Mokdad AH., Forouzanfar MH, Daoud, F, Mokdad, AA, Bchearaoui CE, Moradi-Lakeh M, et al. Global burden of diseases, injuries, and risk factors for young people’s health during 1990–2013: A systematic analysis for the global burden of disease study 2013. Lancet 2016; 387:2383–2401. https://doi.org/10.1016/S0140-6736(16)00648-6.
5. Ferdon DC, Simon TR. Preventing Youth Violence: Opportunities for Action. Atlanta, GA: National Center for Injury Prevention and Control, Centers for Disease Control and Prevention; 2014.
6. Organização Panamericana de Saúde, Relatório mundial de Violência e Saúde, Washington; 2003.
7. Pickett W, Craig W, Harel Y, Cunningham J, Simpsom K, Molcho M, et al. Cross-national study of fighting and weapon carrying as determinants of adolescent injury. Pediatrics 2005; 116(6):855-63.https://doi.org/10.1542/peds.2005-0607.
8. Leão AS, Moura SNM, Gonçalves ECA, Silva DAS, Silva RJDS, Thomazzi SM, et al. Simultaneous Health Risk Behaviors in Adolescents Associated with Higher Economic Class in the Northeast of Brazil. The Scientific World Journal 2017; 2017: 3587567. https://doi.org/10.1155/2017/3587567.
9. Malta DC, Mascarenhas, MDM, Dias, AR, Prado, RR, Lima CM, Silva, MMA et al. Tendências para situações de violência vivenciadas por estudantes nas capitais brasileiras e no Distrito Federal: resultados da Pesquisa Nacional de Saúde do Escola (PeNSE 2012). Rev. bras. epidemiol. 2014; 17( Suppl 1 ):158-171.http://dx.doi.org/10.1590/1809-4503201400050013.
10. Pinto IV, Barufaldi LA, Campos MO, Malta DC, Souto Rayone MCV, Freitas MG, de et al. Tendências de situações de violência vivenciadas por adolescentes brasileiros: Pesquisa Nacional de Saúde do Escolar 2009, 2012 e 2015. Rev. bras. epidemiol. 2018; 21:e180014. http://dx.doi.org/10.1590/1980-549720180014.supl.1.
11. Yang L, Zhang Y, Bovet P. Physical Fighting and Associated Factors among Adolescents Aged 13–15 Years in Six Western Pacific Countries. International Journal of Environmental Research and Public Health 2017; 14(11):1427. https://doi.org/10.3390/ijerph14111427.
12. Caledonia, KL, Wilson ML, El Grammal HA, Hagras AM. Physical fighting among Egyptian adolescents: social and demographic correlates among a nationally representative sample. Peer J 2013;1: e125. http://dx.doi.org/10.7717/peerj.125.
13. Instituto Brasileiro de Geografia e Estatística. Pesquisa Nacional por Amostra de Domicílios 2012-2015. Rio de janeiro; 2016.
14. WHO. World Health Organization Noncommunicable diseases: country profiles 2011. Genebra: WHO Press 2011;209.
15. Bala MO, Chehab MA, Al-Dashan A, Saadeh S, Al Khenji A, et al. Violence among Adolescents in Qatar: Results from the Global School-based Student Health Survey, 2011. Cureus 2018; 10(7): e2913. http://dx.doi.org/10.7759/cureus.2913.
16. Senanayake SJ, Gunawardena S, Wickramasinghe S, Wickramasinghe C, Gunawardena NS, Lokubalasooriya A, et al. Prevalence and Correlates of Interpersonal Violence Among In-School Adolescents in Sri Lanka: Results From the 2016 Sri Lankan Global School-Based Health Survey. Asia Pac J PublicHealth2019;31(2):147-156. http://dx.doi.org/10.1177/1010539519825600.
17. Brook JS, Brook DW, Whiteman M. Growing up in a violent society: longitudinal predictors of violence in Colombian adolescents. Am J Community Psychol 2007; 40(1-2):82-95. http://dx.doi.org/10.1007/s10464-007-9126-z.
18. Nobre CS, Vieira LJES, Noronha, CV, Frota MAN. Fatores associados à violência interpessoal entre crianças de escolas públicas de Fortaleza, Ceará, Brasil. Ciênc. saúde coletiva 2018; 23 (12):4299-4309. http://dx.doi.org/10.1590/1413-812320182312.29222016.
19. Assis SG, Avanci JQ, Santos NC, Malaquias JV, Oliveira RVC.Violência e representação social na adolescência no Brasil. Rev Panam Salud Publica2004; 16(1):43–51. http://dx.doi.org/10.1590/S1020-49892004000700006.
20. Duke NN, Pettingell SL, McMorris BJ, Borowsky IW. Pediatrics. 2010;125. Adolescent violence perpetration: Associations with multiple types of adverse childhood experiences. Pediatrics 2015; 125(4):e78–786. http://dx.doi.org/10.1542/peds.2009-0597.
21. Malta DC, Souza ER, Silva MMA, Silva CS, Andreazzi MAR, Crespo C et al . Vivência de violência entre escolares brasileiros: resultados da Pesquisa Nacional de Saúde do Escolar (PeNSE). Ciênc. saúde coletiva 2010; 15( Suppl 2 ): 3053-63. http://dx.doi.org/10.1590/S1413-81232010000800010.
22. Borofsky LA, Kellerman I, Baucom B, Oliver PH, Margolin G. Community violence exposure and adolescents school engagement and academic achievement over time. Psychology of Violence 2013; 3(4):381–95. http://dx.doi.org/10.1037/a0034121.
23. LIM SS, Vos T, Flaxman AD, Danaei G, Shibuya K, Adair-Rohani et al. A comparative risk assessment of burden of disease and injury attributable to 67 risk factors and risk factor clusters in 21 regions, 1990-2010: a systematic analysis for the Global Burden of Disease Study 2010. Lancet 2012; 380(9859): 2224-60. http://dx.doi.org/10.1016/S0140-6736(12)61766-8.
24. Mercado-Crespo MC, Mbah, AK. Race and ethnicity, substance use, and physical aggression among U.S. high school students. Journal of Interpersonal Violence 2013; 28(7):1367–84. http://dx.doi.org/10.1177/0886260512468234.
25. Chaveepojnkamjorn W, Pichainarong N. Current drinking and health-risk behaviors among male high school students in Central Thailand. BMC Public Health 2011; 11: 233. http://dx.doi.org/10.1186/1471-2458-11-233.
26. Abdalla RR, Massaro L, Queiroz CMA, Laranjeira R, Caetano R, Madruga CS. Association between drug use and urban violence: Data from the II Brazilian National Alcohol and Drugs Survey (BNADS). Addictive Behaviors Reports 2018; 7: 8-13. http://dx.doi.org/doi:10.1016/j.abrep.2017.11.003.
27. Carvalho AP, Silva TC, Valença PAM, Ferreira SCFB, Colares V, Menezes VA. Consumo de álcool e violência física entre adolescentes: quem é o preditor? Ciência & Saúde Coletiva 2017; 22(12):4013-4020. http://dx.doi.org/10.1590/1413-812320172212.06172016.
28. Pardini DA, Loeber R, Farrington DP, Stouthamer–Loeber M.Direct protective factors for nonviolence. American journal of preventive medicine 2012; 43(2): S28-S40. https://doi.org/10.1016/j.amepre.2012.04.024.
29. Caledonia KL, Wilson ML, El Gralmal HA, Hagrass AM. Physical fighting among Egyptian adolescents: social and demographic correlates among a nationally representative sample. Peer J 2013; 1: e125. https://doi.org/10.7717/peerj.125.
30. Albuhairan F, Abou O, Sayed DA, Badri M, Alshari S, Vries N. The relationship of bullying and physical violence to mental health and academic performance: A cross-sectional study among adolescents in Kingdom of Saudi Arabia, Int J PediatrAdolesc Med 2017; 4(2):61–65.https://doi.org/10.1016/j.ijpam.2016.12.005.
31. Malta DC, Machado IE, Felisbino-Mendes MS, Prado RR, Pinto MAS, Oliveira CM, et al. Uso de substâncias psicoativas em adolescentes brasileiros e fatores associados: Pesquisa Nacional de Saúde dos Escolares, 2015. Rev. Bras. de Epidem 2018; 21(suppl 1):e180004. https://doi.org/10.1590/1980-549720180004.supl.1.
32. Peres CA, Rutherford G, Borges G, Galano E, Hudes ES, Hearst N: Family structure and adolescent sexual behavior in a poor area of Sao Paulo, Brazil. JAdolescHealth2008;42(2):177-183.https://doi.org/10.1016/j.jadohealth .2007.08.007.
33. Elsaesser C, Gorman-Smith D, Henry D, Schoeny M. The Longitudinal Relation Between Community Violence Exposure and Academic Engagement During Adolescence: Exploring Families’ Protective Role. Journal of Interpersonal Violence 2017; 0886260517708404;1–22. https://doi.org/10.1177/0886260517708404
34. Kipping RR, Smith M, Heron J, Hickman M, Campbell R. Multiple risk behavior in adolescence and socio-economic status: findings from a UK birth cohort. Eur J Public Health 2014; 2(1):44-49.https://10.1093/eurpub/cku078.
35. Shakeel MD, DeAngelis, CA. Can private schools improve school climate? Evidence from national representative sample. Journal of School Choice 2018; 12(3):426-445. https://10.1080/15582159.2018.1490383.
36. Cotrim BC, Carvalho CG, Gouveia N. Comportamentos de saúde entre jovens estudantes das redes pública e privada da área metropolitana do Estado de São Paulo. Rev Saúde Pública 2000; 34(6):636-645. http://dx.doi.org/10.1590/S0034-89102000000600012.
37. Andrade FH. Can the school context moderate the protective effect of parental support on adolescents\' alcohol trajectories in urban Chicago? Drug and alcohol dependence 2013; 133 (2):330-337. http://dx.doi.org/10.1016/j.drugalcdep.2013.06.010
38. Instituto Nacional de Estudos e Pesquisas Educacionais Anísio Teixeira. Sinopse Estatística da Educação Básica 2015. Brasília-DF: Inep, 2016.
39. Brito ALS, Hardmam CM, Barros MV. Prevalence and factors associated with the co-occurrence of health risk behaviors in adolescents. Rev Paul Pediatr. 2015; 33(4): 423-430. http://dx.doi.org/10.1016/j.rpped.2015.02.002.
40. Pierobon M, Barakb M, Hazratib S e JacobseKHAlcohol consumption and violence among Argentine adolescents. 2013 89(1):100-107. http://dx.doi.org/10.1016/ j.jped.2013.02.015.