Violence in intimate relationships in adolescents: effectiveness of an intervention by peers through forum theater

1 Center for Innovative Care and Health Technology (ciTechCare), School of Health Sciences, Polytechnic Institute of Leiria, Leiria, Portugal. 2 Center for Health Technology and Services Research – NursID (CINTESISNursID), Porto, Portugal. ABSTRACT Objective: The main goal was the peer evaluation of the effectiveness of a health education intervention, also seeking the characterization of violent behaviors experienced by teens in intimate relationships. Method: A pretest-posttest quasi-experimental study was performed with a non-probabilistic sample of 197 teens of both genders without control group. The intervention performed consisted of one peer-run forum theater session about intimate violence in dating relationships of teens. Results: Boys reported being both victims and perpetrators of violent behaviors more frequently than girls. The sample reported more behaviors associated with victimization than perpetration. Girls showed a higher level of knowledge on the topic and this knowledge increased after the intervention. Conclusion: Psychological violence is highly prevalent in dating relationships, particularly in cases of victimization. It seems essential to understand and act upon adolescents’ perceptions about what violent behavior is. This study confirmed the effectiveness of health education in acquiring knowledge on this topic.


INTRODUCTION
Violence in intimate relationships is defined as the behavior by an intimate partner that causes physical, sexual or psychological damage, including acts of physical aggression, sexual coercion, psychological abuse and controlling behaviors (1) , which are more likely to occur in long relationships involving frequent contact (2) .
In the last ten years, knowledge acquired in this area has brought to light one of the most important challenges for public health (3) . Stalking, physical, emotional, and sexual violence in intimate relationships began to be considered a serious public health problem (4) . Although there is a lot of research about this problem, its extent and severity among adolescents has only been studied during the last decade (5)(6) .
Research on this topic and its prevention is important because adolescence is a crucial stage of relational development. Based on violence in intimate relationships at this stage of the developmental cycle, it is possible to predict the quality of intimate relationships in adulthood (7) . Indeed, violence at this developmental stage can be a precursor of domestic violence (8) .
The great variability in prevalence data is due to several factors, namely the type of evaluation instrument, adolescents' age, the data collection method used, among others (9) . These figures vary between 3.8% and 41.9% in physical violence suffered (i.e. victimization), and between 0.4% and 57.3% in physical violence perpetrated; between 4.2% and 97% in psychological violence suffered (victimization), and between 8.5% and 95.5% in psychological violence perpetrated; between 1.2% and 58.8% in sexual violence suffered (victimization), and between 0.1% and 64.6% in sexual violence perpetrated (10) . Despite the bidirectionality of violence, several studies report a higher prevalence of psychological aggressions by women and sexual aggressions by men, and higher psychological and sexual victimization in women (2)(3)5,9) .
In the United States of America, the National Intimate Partner and Sexual Violence Survey reported that 22% of adult women and 15% of adult men had experienced some type of violence from their partners for the first time between the ages of 11 and 17 (11) .
Violence in intimate relationships has physical and mental consequences (12) and is associated with injuries and deaths resulting from aggression, sexual and reproductive health disorders, depression, anxiety, suicidal thoughts, and misuse of alcohol, drugs, and tobacco (13) . In a longitudinal study, was found that female (girls) victims of intimate violence exhibited increased alcohol consumption, depressive symptomatology, suicidal thoughts, and tobacco consumption. On the other hand, male (boys) victims of intimate violence showed an increase in antisocial behavior, suicidal ideation, and cannabis consumption (14) .
Violence in dating relationships is a multi-causal phenomenon involving cultural, family and personal risk factors, namely deficits in negotiation skills in conflict situations (15) and emotion regulation skills (16) , exposure to violence within the family and conflicts between parents (8,17) , use of alcohol and drugs (6,17) and coexistence with other couples with violent behaviors (17) .
Programs implemented for primary prevention of violence in intimate relationships differ in duration, number of sessions, content, target population, instruments used, and the number and type of evaluations, namely the use of follow-ups and control groups (18) .
The duration of interventions ranges from a 50-minute session to 18 sessions over four months (19) . In most studies, the implementation of intervention programs has had a positive influence on improving knowledge and attitudes toward violence in intimate relationships (12) . However, in some studies these interventions have produced the opposite effect (12) .
Programs aimed only at acquiring knowledge and attitudes appear to be less effective than programs working on conflict resolution strategies (20) .
Interventions can be carried out in different ways (5) and classified into structured videos, role-playing or facilitated discussions, courses, and games -and these activities can be carried out individually, as a group, or with couples of adolescents.
Developers of intervention programs should consider information and knowledge, as well as interpersonal conflict resolution strategies, communication, changes in stereotypes about gender roles and attitudes toward violence, the development of empathy, and self-esteem promotion (21) .
Considering current knowledge on the topic, the aim of this study was to identify the sociodemographic characteristics of adolescents attending secondary school, the characteristics of their dating relationships, and violent behaviors in these relationships; and evaluate the effectiveness of a health education intervention on dating violence using a peer-run forum theater.
This study aimed at addressing the following research question: What is the effectiveness of a health education intervention using a peer-run forum theater on the knowledge of intimate and dating relationship violence?

Study deSign
In this pretest-posttest quasi-experimental study without a control group, was developed and implemented a health education intervention structured in three phases. The 120-minute single-session intervention was conducted by four students in their final year of the nursing undergraduate course by using a forum theater play with mediation of three nursing teachers. The phases of the intervention (using the forum theater methodology) were: Presentation of the intervention group and icebreaker activities (social interaction games between participants organized by nursing students) were developed for all participants to foster a closer relationship between the peer educators and adolescents targeted by the intervention; Presentation of the forum theater play by peer educators. This play portrays the experience of a dating relationship with various types of violent behavior. It was created by the "White Ribbon" project research team from the Nursing School of Coimbra. The play lasts 10 minutes and is performed by two nursing students on a stage in front of students attending the school, who are involved in the study. The aim of the play is to represent the dynamic between young people and the spaces frequented by them, namely the school, coffee shops/bars, and home. Two nursing students present the case of two adolescents who start dating and multiple situations of intimate violence are shown.
The nursing teacher played the role of facilitator, fostering discussion and debate among participants about various subjects relating to violence in intimate relationships based on the play. At this stage of the intervention, participants were allowed and encouraged to present appropriate strategies to deal with the several violence situations shown in the play, thereby promoting positive strategies for conflict resolution. At this time, adolescents were invited to replace the actress and play her role, but display healthy behavior or solutions to the problem being discussed (22) . This technique allowed the exploration of stories about power relationships between oppressors and the oppressed by giving participants the opportunity to take part in the play and present positive strategies for conflict resolution.
The facilitator also managed and fomented the discussion on the theme dramatized in the play, highlighting the most pertinent comments from analyzes made by adolescents based on the story presented (23) .
All students participated in the different stages of the procedure. Four adolescents were invited to replace the actors and play their roles but were asked to display healthy behaviors or solutions to the problem under discussion.

PoPulation
This study was developed in a high school located in a central city of Portugal with adolescents studying in the 10 th , 11 th and 12 th grades. The population of this study comprises teenagers of both genders in a relationship or not.
A formal invitation was sent to the school. After the school's acceptance to participate in the study, class directors presented the project and invited the students to participate in the intervention. During a parents' meeting, was asked the parents' consent for the participation of their children. Only the students who signed the informed consent form and were authorized by their parents (who also signed an informed consent form) participated in the project.
Two interventions were implemented, one for each group of students (a group of 128 students and another group of 64 students).

Selection criteria
In this study, inclusion criteria considered for sample definition were the following: high school students at schools selected for the intervention, who were available, agreed to participate in the study, and received parental/guardian permission to participate in the study.
The following exclusion criteria were applied: students whose parents did not sign the informed consent form and married students.
The study considered 197 students of both genders in the 10 th , 11 th and 12 th grades with a mean age of 16.6 ± 1.1 years.
Ten (6.6%) students participated in the pretest but were not included in the posttest because they were not present during the intervention.

data collection
Considering the study design, two questionnaires were administered in two stages of the evaluation.
In the first stage, one week before the health education intervention with forum theater, a questionnaire was administered with three groups of variables.
Sociodemographic and dating relationship characteristics (age, gender, grade, existence and duration of dating relationship, experience of situations of intimate violence, intention to act in situations of intimate violence).
Knowledge of intimate and dating relationship violence (24) , including causes, consequences, predisposing factors, types of violence, myths and issues associated with jealousy, with 47 true-false propositions scored 1 for each right answer and 0 for each wrong answer. The maximum possible score was 47, and the higher the score obtained the higher the student's level of knowledge.
Victimization and perpetration behavior carried out by adolescents during dating and intimate relationships (24) . This instrument consists of 18 statements with dichotomous responses that relate to behaviors reflecting the different types of physical, psychological, and sexual violence from the perspective of the aggressor ("you did this to your boyfriend/ girlfriend") and the victim ("your boyfriend/girlfriend did this to you").
In the second stage of the evaluation, which took place after the forum theater intervention, the questionnaire enabled a reevaluation of knowledge about violence in dating and intimate relationships.
The pretest was implemented in the week prior to the intervention, and the posttest was conducted on the same day, immediately after the intervention.
The intervention and data collection were conducted in February 2016 (with the group of 128 students) and in November 2017 (with the group of 64 students). The sample was composed of students recruited from sequential years at the selected school.

data analySiS and ProceSSing
Data were processed using the IBM SPSS Statistics, Version 20.0.
Descriptive and inferential statistical measures were used to verify the hypothesis. Although the variables under study did not display a normal distribution in the Kolmogorov-Smirnov test (<0.05), the central limit theorem was used (sample>30) with parametric statistical tests, namely paired and independent Student's t and Pearson's correlation tests.
www.scielo.br/reeusp Violence in intimate relationships in adolescents: effectiveness of an intervention by peers through forum theater Rev Esc Enferm USP · 2020;54:e03539 ethical aSPectS The study was approved by the Research Ethics Committee of the Health Sciences Unit: Nursing, Nursing School of Coimbra (no. 296/08/2015) and permission from the General Direction of Innovation and Curriculum Development in the Ministry of Education and the school directors.
All participants and their parents were informed of the study objectives, assured of data confidentiality, their rights, and signed the Informed Consent Form.

SociodemograPhic and dating relationShiPS characterization
The 197 students in the sample had a mean age of 16.6 ± 1.1 years, and 127 were males. As shown in Table 1, 159 participants reported a current or past dating relationship (five participants reported having both a past and current dating relationship). The duration of reported relationships was of 10 ± 12.0 months on average. Most participants (98.2%) were heterosexual and 1.8% were homosexual; 61.2% were not sexually active.
Regarding the existence of violence episodes in dating and intimate relationships, five adolescents (three boys and two girls) reported experiencing or having experienced some type of violence in their relationship.

Violent behaViorS (Victimization and PerPetration) in dating relationShiPS
Regarding violent behaviors during dating and intimate relationships (Table 2), psychological violent behaviors were more prevalent among both boys and girls. Of adolescents who reported having been victims of some kind of violence, boys reported being victims of "judgement, criticism and correction" (31.0%), "blackmail behaviors" (17.2%), "negative comments about their appearance" (17.2%), and "invasion of privacy" (17.2%), while girls reported being victims of "judgement, criticism and correction" (32.8%) and "blackmail behaviors" (19.0%). Psychological violent behaviors were also more prevalent among both boys and girls who reported having carried out some sort of violent behavior, particularly "judging, criticizing and correcting" (26.7% in boys and 28.1% in girls).
Adolescents who reported not having been victims of any of the 18 listed behaviors accounted for 48.9% of the sample, and 54.7% were not perpetrators (aggressors) toward their girlfriends or boyfriends.
By applying the independent Student's t-test, boys reported, on average, that they had been victims of more violence (1.7 ± 2.5) and were more often perpetrators (1.3 ± 0.7) than girls (1.2 ± 2.2 for victimization and 0.73 ± 1.2 for perpetration). Those differences were not statistically significant (p> 0.05).
By applying the paired Student's t-test, were identified the differences between the number of victimization and perpetration behaviors, and both boys and girls reported more behaviors associated with victimization than perpetration (p<0.01).
We sought to determine whether there was a statistically significant correlation between an increase in the number of victimization and perpetration behaviors and the level of knowledge about violence in dating relationships before the intervention. In the group of boys, the correlation between the level of knowledge and the number of perpetration behaviors was weak, negative, and significant (r=-0.288; p<0.01), and there was no significant correlation between knowledge and behavior of victimization (p>0.05). In girls, the correlations between knowledge and both types of behavior were weak and non-significant (p>0.050).

adoleScentS' knowledge of dating and intimate
Violence before and after the interVention From the analysis in Table 3, on average, girls exhibited a higher level of knowledge about violence in dating and intimate relationships than boys (p<0.01) both before and after the intervention.
The results of the paired Student's t-test on the level of knowledge before and after the intervention showed statistically significant differences for girls (t =-4.480; p<0.01), but not for boys (t=1.039, p>0.05). This means that the intervention performed through forum theater by peers was effective in the knowledge promotion in girls, but not in boys.

DISCUSSION
This study has allowed an analysis of knowledge and violent behaviors in dating and intimate relationships (as a victim or perpetrator), and of the effectiveness of a health education intervention to improve this knowledge and deconstruct myths about healthy relationships maintained by adolescents.
The reported prevalence of violence in intimate relationships varies widely as methods of determination differ (9) , which may also be a finding in this study. Of the adolescents questioned, 2.5% reported having already been victims of this type of violence, a very different result from when they were asked about violent behaviors beyond physical violence. Based on the 18 behaviors listed, the prevalence of violence was 51.1% for victimization behaviors, and 45.3% of adolescents reported they were perpetrators (aggressors) toward their girlfriend or boyfriend of at least one of the behaviors reflecting a form of violence. These values are higher than those indicated by some studies (2) and lower than those found by others (25) , and confirm the reports by some authors (6) who stated that violence among adolescents is a major problem.
The differences found in the perception of what adolescents consider to be violent behaviors, both as victims and perpetrators, lead to the need for an in-depth analysis of these differences, namely by studying and analyzing a comparison of self-awareness of engaging in violent behaviors and understanding the definition of violent behavior. It is important to act in the direction of promoting a proper understanding of the definition of violence in intimate and dating relationships among adolescents. The fact that some behaviors are downplayed also becomes relevant in this context. Most participants reported they would seek support or help if they were victims of violence, and in some studies, most participants reported they did not seek formal help when they were victims of violence (26)(27) , and others sought informal help, namely from friends (28) , and women tend to seek support more often than men (29) .
Boys reported having been victims of violent behaviors such as blackmail; negative comments about appearance; invasion of privacy; and judgment, criticism and correction to a greater extent (17.2% to 31%) than the other behaviors. Girls reported having been mostly victims of blackmail and judgment, criticism and correction (19% to 32.8%) rather than other violent behaviors. The difference between the number of adolescents who reported having experienced dating or intimate violence and the number of violent behaviors that adolescents reported having experienced is striking. Furthermore, the number of violent behaviors experienced as perpetrators is lower than the number of violent behaviors experienced as victims. Boys mostly reported engaging in judgment, criticism and correction; negative comments about appearance; and blackmail (12.1% to 26.7%), whereas girls mostly reported acts of judgment, criticism and correction (28.1%) toward their partners.
In terms of numbers of victimization and perpetration behaviors, there are no statistically significant differences between boys and girls. This study reveals a bidirectionality in violence with psychological violence being most prevalent in both boys and girls, which is similar to other authors' findings (10) .
Although boys and girls report more behaviors associated with victimization than perpetration (p<0.01), we can see that violence is mutual in these adolescents' intimate relationships. This has already been shown in other studies (2) .
Upon analyzing knowledge about violence in intimate relationships, girls clearly have higher levels of knowledge than boys.
Girls generally have higher levels of knowledge associated with healthy behaviors, and the impact of health education sessions is higher in girls (24) . This was also the result found in this study, which proved the effectiveness of the intervention performed for knowledge promotion in female adolescents. The health education session conducted using peer-run forum theater is an effective strategy for promoting knowledge about dating and intimate violence in young females, and it is important to keep these differences in mind in future studies.
Other authors have acknowledged the difficulty of comparing results on adolescents' level of knowledge about violence in dating and intimate relationships (19) due to the fact that each study uses different scales to measure knowledge, although it is important to highlight the results of other papers (12) . In addition, they found that most studies reported improvement in knowledge with educational programs, and that programs oriented only toward knowledge acquisition seem to be less effective than programs addressing conflict resolution strategies (20) .
The fact that before the intervention, the level of knowledge of violence in dating and intimate relationships was not associated with the number of perpetration and victimization behaviors leads us to believe that violent behaviors do not depend only on the level of knowledge. Although the effectiveness of improving conflict resolution strategies was not evaluated, this area was favored by this intervention and should be assessed in future studies.
It is also important to analyze the impact of this knowledge improvement on the understanding of the type of behaviors found in intimate relationships and the reduction of the frequency of violent behaviors in dating relationships between adolescents. A subsequent evaluation to analyze these aspects after the intervention seems appropriate in order to ascertain if the increase in knowledge translates into behavioral improvement and a decrease in violent situations in young people's intimate relationships.
In addition, it will be necessary to develop interventions with adolescents before they begin their first intimate relationships, since violence in these relationships has implications for future relationships and may be passed on from one generation to the next.
This study had some limitations, namely the use of a small sample from a single Portuguese school, and the absence of a control group. Further studies are necessary using larger and randomized samples and a control group. It is difficult to associate the improvement in their knowledge with the intervention without using a control group to confirm this. Extended follow up is recommended to confirm that participants used the acquired knowledge in their daily lives.

CONCLUSION
In conclusion, the sample studied showed similar levels of dating violence to those found in the bibliography, and it seems essential to understand and act on adolescents' perception of what constitutes violent behavior.
A need to invest in multiple intervention programs was observed in light of the perceived effectiveness of this intervention with a single session linked to training in communication skills and conflict resolution through the forum theater technique. Violence prevention programs should focus not only on intimate relationships, but also on behavior toward others (i.e., the development of interpersonal skills).
By considering that nowadays young people start intimate relationships at younger ages, it is crucial to implement interventions among adolescents before they begin having these kinds of relationships.
Nurses' competences include working in the area of health education and promotion among young people. Therefore, this project provides an encouraging and appealing intervention to be developed with young people. In addition, the positive impact of such an intervention during adolescence can lead to a change in adolescents' behaviors in their future intimate relationships.
The strategy under study, focused on violent behaviors (victimization and perpetration) in dating relationships, provides a quick way of carrying out the initial triage for both victims and perpetrators, thus allowing for early intervention.