Prevalence of sexual violence and associated factors among primary school students – Brazil , 2015

The objective of this study was to describe and analyze factors associated with sexual violence (SV) among primary school students in Brazil. Data from the National School Health Survey (PeNSE in Portuguese) in 2015 was analyzed. The prevalence of total and disaggregated SV was calculated according to variables such as sociodemographic data, family context, mental health, risk behaviors, safety, and physical activity. The Odds Ratios of suffering SV were estimated according to variables that were statistically associated (p < 0.05) by means of multivariate analysis. The prevalence of SV was 4.0%. SV among school-age adolescents was associated with characteristics such as: age of < 13 years old; female; black skin color; working; being assaulted by family members; having insomnia; feeling lonely; not having friends; consuming tobacco / alcohol regularly; having tried drugs; having started sexual activity; feeling insecure on the way to or at school; and having suffered bullying. Studying in a private school, having a mother with higher education, living with parents, and supervision by relatives were protective factors to SV. It was possible to identify students’ vulnerabilities to SV, which can support researchers, professionals, and families in the prevention of this type of violence.


Introduction
Sexual violence against adolescents is very common and threatens the health and well-being of millions of young people throughout the world, which represent a significant portion of the population, and should be prioritized in public policies to prevent such harm [1][2][3] .Owing to the importance of the subject, it is paramount to cast an attentive eye on the situations of sexual violence experienced by adolescents in their everyday life, which occur as much in the inter-family sphere as outside of it, such as in school or in the school environment 1,3,4 .
Among different types of violence 5,6 , sexual abuse is defined as any unwanted sexual act or attempt to carry one out, undesired sexual commentaries or advances, acts aimed at sexual trafficking or in some way against the sexuality of a person using coercion, practiced by any person independent of their relation to the victim, in any scenario including at home or at work, but not limited to these 6 .Not all physical, psychological, and social harm caused by sexual violence against adolescents results in injuries, disability, or death.Thus, its consequences can be immediate or can manifest years after the occurence 6,7 .Many victims are children or young people who do not know how to protect themselves, while others -in spite of being older -are forced to remain silent and do not seek help due to social pressures or conventions, causing even greater suffering 7,8 .
For a long time, sexual violence as a research topic was neglected nearly everywhere in the world 6 .However, since its effects disrupted the life of people across a long time period, the topic came to be investigated in recent times as a public health problem on a grand scale.It is known that in addition to physical harm, it can lead to inappropriate consumption of alcohol and other drugs, depression, suicide, dropping out of school, unemployment, and recurring difficulties with relationships [5][6][7][8][9] .
Studies that evaluate the prevalence of sexual abuse among school-aged adolescents are still in short supply.One study recently carried out in the United States indicated that 6.7% of students reported being forced to have sexual relations against their will, with greater frequency among girls (10.3%) compared to boys (3.1%) 9 .In a study done with 6,709 public school students in ten Brazilian capital cities, 1.6% of adolescents interviewed affirmed they have suffered sexual violence inside of school, and 5.6% stated knowledge of sexual violence that occurred in the school environment 10 .It is known that exposure to health risk factors is much greater in the adolescent phase, a period of life that shows an association of multiple factors that strengthen and mutually reinforce the phenomenon of sexual violence 6,11,12 .
Faced with the absence of statistics on a national scale regarding sexual abuse among school-aged adolescents, the National School Health Survey ('Pesquisa Nacional de Saúde do Escolar' or PeNSE) included the theme of sexual violence in their 2015 publication.with the intention of better understanding this serious problem, as well as to furnish data for the planning of strategies to confront it.Thus, this article has the aim of describing the prevalence of sexual violence among school-aged adolescents, and of identifying factors associated with this phenomenon.

Methodology
This was a cross-sectional study, with data from the National School Health Survey (PeNSE) conducted in 2015.PeNSE investigated issues concerning socioeconomic aspects; family situation; physical activity; experimentation with and consumption of alcohol, tobacco, and other drugs; mental and sexual health; and safety, among other topics.The study's population encompassed students between 13 and 17 years of age attending the 9 th year of primary schooling, as the World Health Organization (WHO) considers this age range as a reference for carrying out surveys with students, and because in Brazil this grade concentrates more than 80% of students from 13 to 15 years old 13 .
The sample was comprised of students enrolled in the 9 th grade of primary education in the school year of 2015, and regularly attending public and private schools in the urban and rural areas, designed to estimate the population parameters (proportions or prevalence) in different geographic domains.These were the 26 municipalities of the capital cities and the Federal District (DF), the 26 federative units or states, and the five greater regions of Brazil, totaling 53 strata.
In each of the 53 strata, a sample of schools was identified and chosen.Following this, a sample of the classes of each school was selected, obtaining an independent sample of students in each of the strata.In these geographic strata, 207 allocation strata were created with probabilities proportional to the size of the school.The size of the sample of each geographic stratum was calculated to furnish estimates of the proportions of some characteristics of interest, with a maximum sample error of approximately 3% and confidence level of 95%.The sample was comprised of 675 municipalities, 3160 schools, and 4159 classes.The researchers administered 102,301 questionnaires, and 102,072 of these were validated and analyzed.
Data collection was carried out between April and September of 2015, using smartphones in which the structured questionnaires were inserted and self-administered.The data was collected in the schools during the students' class time.
Data on the prevalence of sexual violence was obtained via the question: "Have you been forced to have sexual relations at some time in your life?" (Analyzed categories: Yes.No).In addition to estimating the prevalence of sexual violence in the studied population, associations of this phenomenon with other variables were identified, such as sociodemographic aspects, family situation, mental health, risk behavior, safety, and the practice of physical activity.
The prevalence of sexual violence was initially calculated with its respective confidence intervals of 95% (CI95%).To verify these associated factors, bivariate analyses were carried out with estimates of the Odds Ratio (OR) and its respective CI95% to the significance level of 0.05.Following this, a multivariate analysis was conducted for an outcome inserted into the model of independent variables that presented association with the outcomes at a significance level inferior to 0.20, calculating the adjusted ORs (ORa) and its respective CI95%.All of the analyses were done in the program SPSS version 2.0, utilizing procedures of the Complex Samples Modules, suitable for analysis of data obtained by a complex sampling plan.
The study was approved by the National Committee on Research Ethics ('Comissão Nacional de Ética em Pesquisa' , or CONEP), and by the National Health Council ('Conselho Nacional de Saúde' , or CNS), which oversees and approves health research involving human beings, by Decision Number 1.006.467 of March 30.2015.

Findings
As seen in Table 1, it is possible to identify the prevalence of sexual violence among students of the 9th grade of primary education in Brazil in 2015, and its association with the aspects explored in this analysis.The total prevalence of sexual violence was 4.0%, and was higher among students < 13 years old and 16 or older.The significantly highest rates of sexual violence were observed among female students of black skin color in public schools, who were children of mothers without schooling.Sexual violence was reported in greater proportions among students that did not live with their mother and/or father, and among those that worked and received remuneration for their work (Table 1).
Regarding the family context, reporting of sexual violence was greater among students that missed classes, were unsupervised by family members, and among those physically assaulted by a relative.Sexual violence was most frequent among students that reported insomnia, feeling alone, and not having friends.This type of violence was most reported among students with risk behaviors such as smoking, alcohol consumption, experimentation with drugs, and having begun their sexual life.The chances of suffering sexual violence were greater for students that felt unsafe in the route between school and their home and in the school itself, as well as those that reported having suffered bullying.There was no difference according to amount of physical exercise or activity (Table 1).
Drawing on the calculation of the raw OR (Table 1) and the multivariate analysis with adjusted OR for all the variables in the model (Table 2), we verified that the chance of suffering sexual violence was greatest among female students with < 13 years of age.Other risk factors for sexual violence were also identified, such as having black skin, being a child of a mother without education, working, and being assaulted by family members.In the mental health context, the chance to suffer sexual violence was greater for students that reported insomnia, and among those that related feeling alone or not having friends.Regular consumption of tobacco and alcohol, experimentation with drugs, and the beginning of sexual activity were confirmed as risk factors for sexual violence.The chances of suffering sexual violence were greater for students that felt unsafe on the route between home and school and in the school itself, as well as those that reported having suffered bullying (Table 2).
Associated and protective factors in relation to the occurrence of sexual violence included studying in a private school, being the child of a mother with a high education level, living with

Discussion
Any person can experience sexual violence, independent of socioeconomic condition, skin color, or culture 2,14 .Nevertheless, some factors can be associated with the vulnerability of ado- The relation between suffering violence and age demonstrates that children under 13 are more vulnerable to victimization.This can be related to the issue of not knowing how to react during such a situation, lack of maturity, not really understanding what is happening, shame, and fear of the aggressor, aspects that are confirmed by other studies 6, 15 .
In accord with other research 12,[15][16][17][18] , the greater prevalence of sexual violence in females can be explained by cultural factors that, across time, place women in situations of abuse and degradation, a condition perpetuated in society.On the other hand, records of sexual violence among male adolescents can appear lesser in proportion to females owing to embarrassment, fear, stereotypes, and shame of the parents and of society.This was demonstrated in a study on the vulnerability of adolescence in which 3% of boys reported having suffered sexual violence, but only 33% of those searched for assistance in the health sector 3 .
Adolescents that work are more vulnerable to risk factors such as using alcohol and other drugs, having sexual relations 19 , as well as suffering sexual violence, as indicated in this study.The greater the intensity of the work, the more time young people spend in activities considered at risk for sexual violence 19 .
Suffering harm by family members and missing classes without parental permission is associated with suffering sexual violence, while living with a mother or father and receiving family supervision are considered protective factors.In this sense, some studies 14,20 emphasize the fact the children accompanied by their parents are less  susceptible to suffering various types of violence.Other studies 2,21,22 demonstrate that sexual violence against adolescents is more likely to occur in a family context of negligence, with a prevalence of authoritarian parental styles, relations of subordination between family members, and carried out by persons close to the victim.The results presented here confirm this perspective and show the importance of healthy familial relationships.Among the mental health variables, adolescents that feel alone, have insomnia, and do not have friends present the greatest chance of suffering sexual violence.Studies 2,21,22 report that a good mood, joy, happiness and satisfaction with life are characteristics with a positive effect for mental health and, therefore, allow for less exposure to situations of tension and violence, diminishing risks of involvement in circumstances favorable to violence.Hence, if the family protects the adolescent, their development will be complete and healthy.Family relations have importance in the mental health of students, and the influence on the psychological, emotional, and interpersonal relations of adolescents.
In relation to risk behaviors, the regular use of alcohol, drug experimentation, and the practice of sexual relations are also associated with sexual violence.The literature 7 indicates a strong association between sexual abuse and episodes of depression.Various psycho-social consequences can be related to sexual abuse in the long turn, such as psychological disorders including depression, suicidal ideation, anxiety disorders, and post-traumatic stress disorder (PTSD), as well as problems of physical health and sexual risk 7,8,23 .Maternal and parental support contributed to the diminishing of PTSD 8 .In one study 24 on the consumption of alcohol among adolescents, it was proven that its regular use makes the individual vulnerable to being a victim of various types of violence, among them sexual violence 2,21,23 , corroborating the findings of this study.
The variables of feeling unsafe on the route between the home and school, feeling unsafe in the school itself, and having suffered bullying demonstrated association with sexual violence.Bullying that commonly occurs in the school and its environment can push away students, provoking skipping or dropping out of school, or creating fragile links with the school.This increases the risk of involvement in situations of sexual violence among already persecuted or emotionally fragile young people 10,21 .
Available evidence shows that children and adolescents who are victims of sexual violence face more health problems, incur significantly higher costs in healthcare, visit health services for consultations more times during their life, and are hospitalized more frequently and for longer durations than those that have not suffered sexual violence 4 .
Economic inequality, abuse of alcohol, and inappropriate parental practices increase the probability of interpersonal violence, which includes sexual violence.Adolescents that suffer this type of violence are exposed to greater risk of involvement in aggressive and antisocial behavior in their adult life 4 .
The offering of high quality services for the attendance and support of victims is important to reduce trauma, aid in recovering, and prevent new acts of sexual violence.Any broad strategy of violence prevention must identify means to attenuate these risks, or offer protection against them, including via public policies and specific services 25 .
To reduce the vulnerability of young people in relation to sexual violence, strategies must be developed, such as fostering safe, stable, and protective relationships between children and young people and their parents and caretakers.Strategies can also be pursued that develop life skills in children and adolescents; reduce the availability and the harmful use of alcohol; promote gender equality; change cultural and societal norms that support violence; and create programs for serving victims, including their identification and care 4 .

Conclusion
The results of this study indicate that sexual violence among school-aged adolescents is associated with individual characteristics such as: age <13 and ≥16 years, female sex, black skin, working, being harmed by relatives, having insomnia, feeling alone, not having friends, regular tobacco and alcohol consumption, having experimented with drugs and sexual relations, feeling unsafe on the route between school and home as well as in the school itself, and to have suffered bullying.Such associations can help professionals in health, safety, and education, as well as the parents and community as they seek measures to confront the prevention of this type of violence.
It is hoped that the data can serve as a base of support for the strengthening of public policies regarding sexual violence, in light of the fact that they offer indicators that can contrib-ute for the development of inter-sectorial and interdisciplinary actions.We emphasize the need for further studies for a better understanding of the complexity of the phenomenon of sexual violence.

Collaborators
MJ Santos, MDM Mascarenhas and DC Malta worked on the conceptualization of the study, in the analysis and interpretation of data, literature review, critical revision, and approval of the draft of the manuscript.CM Lima and MMA Silva contributed to the critical analysis and the final revision of the text.All of the authors approved the final version of the manuscript.

Table 1 .
Prevalence of sexual violence among students of the 9th year of primary school and raw OR, according to sociodemographic aspects and variables of family context, mental health, risk behaviors, safety, and physical activity.Brazil, 2015.
it continues their mother and/or father, and having family supervision (Table2).

Table 1 .
Prevalence of sexual violence among students of the 9th year of primary school and raw OR, according to sociodemographic aspects and variables of family context, mental health, risk behaviors, safety, and physical activity.Brazil, 2015.

Table 2 .
Risk factors associated with sexual violence among students in the 9th year of primary school.Brazil, 2015.

Table 2 .
Risk factors associated with sexual violence among students in the 9th year of primary school.Brazil, 2015.