Violência nas escolas brasileiras: fatores associados e avaliação de um programa de prevenção
Data
2018
Autores
Gusmões, Júlia Dell Sol Passos 
Orientadores
Sanchez, Zila van der Meer
Tipo
Dissertação de mestrado
Título da Revista
ISSN da Revista
Título de Volume
Resumo
Objetivo: Avaliação do programa de prevenção ao uso de drogas #Tamojunto em relação aos padrões de violência encontrados na escola (artigo 1) e responder se o uso de drogas prediz o envolvimento em eventos violentos após 9 meses entre estudantes que participaram da avaliação do programa (artigo 2). Método: Um ensaio controlado randomizado entre 6.637 alunos de 72 escolas em 6 cidades brasileiras foi conduzido com coletas no baseline e 9 e 21 meses após o baseline. Análises longitudinais usando GEE (Generalized Estimating Equations) foram utilizadas para avaliar o efeito do programa nos 9 meses e nos 21 meses para bullying e violência física, tanto no total quanto para análises estratificadas por idade e sexo. Para as análises de predição de violência escolar utilizou-se uma CFA (Confirmatory Factor analysis) para validar os componentes de violência do questionário e regressão linear para avaliar se as variáveis do baseline referentes ao uso de drogas no último mês (álcool, tabaco, maconha, inalantes, cocaína e binge), envolvimento em episódios de violência (bullying e agressões verbal, física e sexual) e fatores sociodemográficos (sexo, classe socioeconômica, idade e grupo) impactam a violência após 9 meses. Resultados: Foi encontrado que o programa #Tamojunto reduziu a chance de receber bullying nos 9 meses de acompanhamento, principalmente para meninas de 13 a 15 anos (OR = 0,59, 95% IC [0,42, 0,84] e p = 0,003). O efeito não se manteve no tempo, perdendo a significância no acompanhamento de 21 meses. Não foi encontrado efeito para a prática de bullying nem para sofrer ou praticar violência física. Em relação às análises de possíveis fatores de risco, foi encontrado que o envolvimento anterior em episódios violentos (?=0,409, p<0,001), uso de álcool (?=0,076, p=0,038) e inalantes (?=0,036, p=0,035) predisseram maior envolvimento em eventos de violência depois de 9 meses. Ao contrário, ser do sexo feminino (?=-0,044, p=0,001) aparece como possível fator de proteção. Conclusões: Conclui-se que o programa #Tamojunto pode ter efeito de curto prazo para bullying entre meninas, porém adaptações no programa podem ser necessárias para que tal efeito se sustente ao longo do tempo. Além disso, é importante que programas de prevenção de uso de drogas abordem componentes para a prevenção tanto da violência quanto do uso de drogas simultaneamente.
Objective: To evaluate the #Tamojunto drug prevention program in relation to the violence patterns found at school (article 1) and to respond if drug use predicts involvement in violent events after 9 months among students who participated in the evaluation of the program. Method: A randomized controlled trial among 6637 students from 72 schools in 6 Brazilian cities was conducted with collections at the baseline and 9 and 21 months after baseline. Generalized Estimating Equations (GEE) were used to evaluate the effect of the program at 9 months and 21 months for bullying and physical violence, both in total and for stratified analyzes by age and sex. For the school violence prediction, a CFA (Confirmatory Factor Analysis) was used a to validate the violence components in the questionnaire and a linear regression was performed to assess whether the baseline variables related to drug use in the last month (alcohol, tobacco, marijuana, inhalants, (bullying and verbal, physical and sexual aggression) and sociodemographic factors (gender, socioeconomic class, age and group) have an impact on violence after 9 months. Results: It was found that the #Tamojunto program reduced the chance of receiving bullying in the 9 months of follow-up, especially for girls aged 13 to 15 years (OR = 0.59, 95% CI [0.42, 0.84], and p = 0.003). The effect was not sustained through time, losing its significance in the 21 months follow-up. No effect was found for the practice of bullying nor for suffering or practicing physical violence. Regarding the possible risk factors analyzes, it was found that previous involvement in violent episodes (? = 0.040, p = 0.038), using alcohol (?=0.076, p=0.038) and inhalants (? = 0.036, p = 0.035) predicted greater involvement in violence events after 9 months. In contrast, being female (? = -0.044, p = 0.001) appears as a possible protective factor. Conclusions: It is concluded that the #Tamojunto program may have a short-term effect on bullying among girls, but adaptations in the program may be necessary for this effect to be sustained over time. In addition, it is important that drug use prevention programs address components for the prevention of both violence and drug use simultaneously.
Objective: To evaluate the #Tamojunto drug prevention program in relation to the violence patterns found at school (article 1) and to respond if drug use predicts involvement in violent events after 9 months among students who participated in the evaluation of the program. Method: A randomized controlled trial among 6637 students from 72 schools in 6 Brazilian cities was conducted with collections at the baseline and 9 and 21 months after baseline. Generalized Estimating Equations (GEE) were used to evaluate the effect of the program at 9 months and 21 months for bullying and physical violence, both in total and for stratified analyzes by age and sex. For the school violence prediction, a CFA (Confirmatory Factor Analysis) was used a to validate the violence components in the questionnaire and a linear regression was performed to assess whether the baseline variables related to drug use in the last month (alcohol, tobacco, marijuana, inhalants, (bullying and verbal, physical and sexual aggression) and sociodemographic factors (gender, socioeconomic class, age and group) have an impact on violence after 9 months. Results: It was found that the #Tamojunto program reduced the chance of receiving bullying in the 9 months of follow-up, especially for girls aged 13 to 15 years (OR = 0.59, 95% CI [0.42, 0.84], and p = 0.003). The effect was not sustained through time, losing its significance in the 21 months follow-up. No effect was found for the practice of bullying nor for suffering or practicing physical violence. Regarding the possible risk factors analyzes, it was found that previous involvement in violent episodes (? = 0.040, p = 0.038), using alcohol (?=0.076, p=0.038) and inhalants (? = 0.036, p = 0.035) predicted greater involvement in violence events after 9 months. In contrast, being female (? = -0.044, p = 0.001) appears as a possible protective factor. Conclusions: It is concluded that the #Tamojunto program may have a short-term effect on bullying among girls, but adaptations in the program may be necessary for this effect to be sustained over time. In addition, it is important that drug use prevention programs address components for the prevention of both violence and drug use simultaneously.
Descrição
Citação
GUSMÕES, Júlia Dell Sol Passos. Violência nas escolas brasileiras: fatores associados e avaliação de um programa de prevenção. 2018. 121 f. Dissertação (Mestrado) - Escola Paulista de Medicina. Universidade Federal de São Paulo, São Paulo, 2018.