Mediadores e fatores associados ao consumo de drogas em adolescentes brasileiros: um ensaio controlado randomizado do programa de prevenção #Tamojunto2.0
Data
2023-02-28
Autores
Garcia Cerde, Rodrigo de Jesus 
Orientadores
Sanchez, Zila Van der Meer
Tipo
Tese de doutorado
Título da Revista
ISSN da Revista
Título de Volume
Resumo
Objetivo: Investigar sobre os mediadores do efeito do programa de prevenção ao uso de drogas #Tamojunto2.0 e outros fatores associados ao consumo de substâncias em uma amostra de adolescentes de escolas públicas brasileiras. Métodos: Este trabalho utilizou dados do ensaio controlado randomizado que avaliou a efetividade do programa de prevenção ao uso de drogas de base escolar #Tamojunto2.0 em 2019. Os participantes (N = 5208) foram alunos de oitavo ano do ensino fundamental II de escolas públicas brasileiras de Eusébio, Fortaleza e São Paulo. O instrumento utilizado para a coleta de dados foi desenhado a partir de outros instrumentos previamente utilizados em estudos de avaliação do efeito de programas de prevenção ao uso de drogas tanto no Brasil como no exterior. O questionário foi autoaplicável na sala de aula, sem professores presentes e com a coordenação de uma equipe de pesquisadores da Universidade Federal de São Paulo (UNIFESP). Foram coletados em três tempos (baseline, aos 9 meses e aos 26 meses) dados sociodemográficos, uso de drogas, atitudes sobre o consumo de drogas, crenças comportamentais sobre álcool e maconha, habilidades de tomada de decisão, habilidades de recusa, conhecimentos sobre drogas, sintomatologia psiquiátrica, e outras questões sobre bullying, satisfação com a imagem corporal e transtornos alimentares. Devido ao isolamento social produzido pela pandêmica de Covid-19, a terceira onda de coleta de dados foi realizada através de um questionário online. Para a análise dos dados, utilizou-se métodos de equações estruturais (análise confirmatória fatorial, classes latentes e análise de mediação), modelos de diferenças em diferenças (modelos de efeitos mistos multinível) e regressões lineares, logísticas e multinomiais multivariáveis. Resultados: Observou-se a capacidade preditiva da escala de atitude sobre o consumo de drogas, através da (1) associação direta entre as atitudes positivas e a chance de ser poli usuário de drogas (aOR= 8.07, 95%CI = 7.59;8.55), e da (2) associação inversa entre as atitudes negativas e a chance de ser poli usuário (aOR= 0.70, 95%CI = 0.57;0.83). A análise dos modelos multiníveis de efeitos mistos mostrou que o programa #Tamojunto2.0 incrementou os conhecimentos sobre drogas (Coef.= 0.26, 95%CI= 0.17; 0.34) e as crenças negativas/não-positivas sobre álcool (Coef.= 0.24, 95%CI= 0.05; 0.42). A análise de mediação sugere que o programa #Tamojunto2.0 foi efetivo em retardar o consumo de álcool via o aumento das crenças negativas e não-positivas sobre o álcool (Coef.= -0.024, 95%CI= -0.043; -0.006). Por fim, a análise das mudanças produzidas pela pandemia de COVID-19 no consumo de sustâncias mostrou uma diminuição generalizada no uso, mas aqueles que iniciaram, mantiveram ou aumentaram seu consumo, tiveram mais chances de apresentar problemas de conduta, desatenção-hiperatividade, problemas como seus pares e problemas emocionais. Conclusões: Os resultados sugerem que as variáveis mediadoras dos programas de prevenção têm papel preditor na redução no uso de drogas entre adolescentes. Ao mesmo tempo, mostrou-se que o programa foi efetivo em incrementar os conhecimentos sobre drogas e as crenças negativas/não-positivas sobre álcool. No entanto, foi demonstrado que #Tamojunto2.0 efetivamente previu a iniciação do uso de álcool apenas através do incremento nas crenças negativas/não-positivas sobre álcool. Por fim, foi observado uma diminuição generalizada no consumo de sustâncias durante a pandemia de COVID-19, mas também foram identificados adolescentes que iniciaram, mantiveram e incrementaram seu consumo, sendo aqueles que tiveram mais chances de apresentar sintomatologia de problemas psiquiátricos. Estes achados podem ser utilizados para aprimorar os programas preventivos, já que evidenciou-se que desconstruir as crenças positivas e incrementar as negativas sobre o uso de drogas, pode ser uma estratégia com maior efetividade para prevenir o consumo na população adolescente. Futuras pesquisas podem dar maior claridade ao longo prazo sobre o efeito da pandemia no consumo de drogas entre os adolescentes e sua saúde mental, além de pesquisar sobre o efeito de programa #Tamojunto2.0, no período de pós-pandemia.
Objective: To investigate the mediators of the drug use prevention program #Tamojunto2.0 and other factors associated with substance use in a sample of adolescents from Brazilian public schools. Methods: This paper used data from the randomized controlled trial that evaluated the effectiveness of the schoolbased drug use prevention program #Tamojunto2.0 in 2019. Participants (N = 5208) were eighthgrade elementary II students from Brazilian public schools in Eusébio, Fortaleza, and São Paulo. The instrument used for data collection was designed from other instruments previously used in studies evaluating the effect of drug use prevention programs both in Brazil and abroad. The questionnaire was selfadministered in the classroom, without teachers present, and coordinated by a research team from the Universidade Federal de São Paulo (UNIFESP). Sociodemographic data, drug use, attitudes about drug use, behavioral beliefs about alcohol and marijuana, decisionmaking skills, refusal skills, knowledge about drugs, psychiatric symptoms, and other questions about bullying, body image satisfaction, and eating disorders were collected at three times (baseline, 9months followup, and 26months followup). Due to the COVID19 pandemic social isolation, the third wave of data collection was performed via an online questionnaire. To carry out the data analysis, structural equation methods (confirmatory factor analysis, latent classes, and mediation analysis), differenceindifferences models (multilevel mixed effects models), and multivariable linear, logistic and multinomial regressions were used. Results: The predictive capacity of the attitude about drug use scale was observed through the (1) direct association between positive attitudes and the chance of being poly drug user (aOR= 8.07, 95%CI = 7.59;8.55), and the (2) inverse association between negative attitudes and the chance of being polydrug user (aOR= 0.70, 95%CI = 0.57;0.83). Multilevel mixed effects model analysis showed that the #Tamojunto2.0 program increased drug knowledge (Coef.= 0.26, 95%CI= 0.17; 0.34) and negative/nonpositive beliefs about alcohol (Coef.= 0.24, 95%CI= 0.05; 0.42). Mediation analysis suggests that the intervention was effective in delaying alcohol consumption via increasing negative and nonpositive beliefs about alcohol (Coef.= 0.024, 95%CI= 0.043; 0.006). Finally, analysis of the changes produced by pandemic COVID19 in substance use showed an overall decrease in use, but those who initiated, maintained, or increased their consumption were more likely to have conduct problems, inattentionhyperactivity, peer problems, and emotional problems. Conclusions: The results suggest that mediating variables of prevention programs have predictive capacity for drug use prevention among adolescents. At the same time, it was shown that the program was effective in increasing drug knowledge and negative/nonpositive beliefs about alcohol. However, #Tamojunto2.0 was shown to effectively predict alcohol initiation only by increasing negative/nonpositive beliefs about alcohol. Finally, a generalized decrease in substance use was observed during the COVID19 pandemic, but also had adolescents who initiated, maintained, and increased their consumption, those who had more chances to presenting psychiatric symptomatology problems. These findings can be used to improve preventive programs, since it showed that deconstructing positive beliefs and increasing negative ones about drug use may be more effective strategies to prevent consumption in the adolescent population. On the other hand, future research can provide more longterm clarity on the effect of the pandemic on drug use among adolescents and their mental health, as well as research on the effect of the #Tamojunto2.0 program in the postpandemic period.
Objective: To investigate the mediators of the drug use prevention program #Tamojunto2.0 and other factors associated with substance use in a sample of adolescents from Brazilian public schools. Methods: This paper used data from the randomized controlled trial that evaluated the effectiveness of the schoolbased drug use prevention program #Tamojunto2.0 in 2019. Participants (N = 5208) were eighthgrade elementary II students from Brazilian public schools in Eusébio, Fortaleza, and São Paulo. The instrument used for data collection was designed from other instruments previously used in studies evaluating the effect of drug use prevention programs both in Brazil and abroad. The questionnaire was selfadministered in the classroom, without teachers present, and coordinated by a research team from the Universidade Federal de São Paulo (UNIFESP). Sociodemographic data, drug use, attitudes about drug use, behavioral beliefs about alcohol and marijuana, decisionmaking skills, refusal skills, knowledge about drugs, psychiatric symptoms, and other questions about bullying, body image satisfaction, and eating disorders were collected at three times (baseline, 9months followup, and 26months followup). Due to the COVID19 pandemic social isolation, the third wave of data collection was performed via an online questionnaire. To carry out the data analysis, structural equation methods (confirmatory factor analysis, latent classes, and mediation analysis), differenceindifferences models (multilevel mixed effects models), and multivariable linear, logistic and multinomial regressions were used. Results: The predictive capacity of the attitude about drug use scale was observed through the (1) direct association between positive attitudes and the chance of being poly drug user (aOR= 8.07, 95%CI = 7.59;8.55), and the (2) inverse association between negative attitudes and the chance of being polydrug user (aOR= 0.70, 95%CI = 0.57;0.83). Multilevel mixed effects model analysis showed that the #Tamojunto2.0 program increased drug knowledge (Coef.= 0.26, 95%CI= 0.17; 0.34) and negative/nonpositive beliefs about alcohol (Coef.= 0.24, 95%CI= 0.05; 0.42). Mediation analysis suggests that the intervention was effective in delaying alcohol consumption via increasing negative and nonpositive beliefs about alcohol (Coef.= 0.024, 95%CI= 0.043; 0.006). Finally, analysis of the changes produced by pandemic COVID19 in substance use showed an overall decrease in use, but those who initiated, maintained, or increased their consumption were more likely to have conduct problems, inattentionhyperactivity, peer problems, and emotional problems. Conclusions: The results suggest that mediating variables of prevention programs have predictive capacity for drug use prevention among adolescents. At the same time, it was shown that the program was effective in increasing drug knowledge and negative/nonpositive beliefs about alcohol. However, #Tamojunto2.0 was shown to effectively predict alcohol initiation only by increasing negative/nonpositive beliefs about alcohol. Finally, a generalized decrease in substance use was observed during the COVID19 pandemic, but also had adolescents who initiated, maintained, and increased their consumption, those who had more chances to presenting psychiatric symptomatology problems. These findings can be used to improve preventive programs, since it showed that deconstructing positive beliefs and increasing negative ones about drug use may be more effective strategies to prevent consumption in the adolescent population. On the other hand, future research can provide more longterm clarity on the effect of the pandemic on drug use among adolescents and their mental health, as well as research on the effect of the #Tamojunto2.0 program in the postpandemic period.